Cancer-germline gene expression in pediatric solid tumors using quantitative real-time PCR
Article first published online: 3 OCT 2006
Copyright © 2006 Wiley-Liss, Inc.
International Journal of Cancer
Volume 120, Issue 1, pages 67–74, 1 January 2007
How to Cite
Jacobs, J. F.M., Brasseur, F., Hulsbergen-van de Kaa, C. A., van de Rakt, M. W.M.M., Figdor, C. G., Adema, G. J., Hoogerbrugge, P. M., Coulie, P. G. and de Vries, I. J. M. (2007), Cancer-germline gene expression in pediatric solid tumors using quantitative real-time PCR. Int. J. Cancer, 120: 67–74. doi: 10.1002/ijc.22118
- Issue published online: 30 OCT 2006
- Article first published online: 3 OCT 2006
- Manuscript Accepted: 27 APR 2006
- Manuscript Received: 17 MAR 2006
- The Quality of Life Gala
- Stichting Vrienden van het Kinderoncologisch Centrum Zuid-Oost Nederland
Cancer-germline genes (CGGs) code for immunogenic antigens that are present on various human tumors but not on normal tissues. The importance of CGGs in cancer immunotherapy has led to detailed studies of their expression in a range of human tumors. We measured the levels of expression of 12 CGGs in various pediatric solid tumors to identify targets for therapeutic cancer vaccines. Quantitative real-time PCR (qPCR) was used to measure the expression of 8 MAGE genes and of genes LAGE-2/NY-ESO-1 and GAGE-1, 2, 8 in 9 osteosarcomas, 10 neuroblastomas, 12 rhabdomyosarcomas and 18 Ewing's sarcomas. Nine tumors were also examined by immunohistochemistry with monoclonal antibodies specific for the MAGE-A1, MAGE-A4 and NY-ESO-1 proteins. All osteosarcoma and 80% of neuroblastoma samples expressed several CGGs at high levels. Six of 12 rhabdomyosarcomas and 11 of 18 Ewing's sarcomas expressed at least one CGG. Immunohistochemistry data correlated well with qPCR results and showed a homogeneous protein distribution pattern in most positive tumors. No correlation was found between the levels of CGG expression in the tumors and clinicopathological parameters of the patients. Pediatric solid tumors express several CGGs, which encode antigens that could be targeted in therapeutic vaccination trials. Several CGGs of the MAGE, GAGE and LAGE families are coexpressed in a large proportion of osteosarcoma and neuroblastoma samples. Some rhabdomyosarcomas express several of these genes at high levels. Ewing's sarcomas have an overall low CGG expression. © 2006 Wiley-Liss, Inc.
Human tumors bear antigens that can be specifically recognized by autologous T lymphocytes and antibodies.1, 2, 3 It has been shown in animal models and in clinical trials that vaccination with these tumor-specific antigens can lead to the eradication of tumors.4, 5 Over the last decade, promising new treatment modalities of induction of tumor-specific T cells have been proposed and applied.6, 7
Cancer-germline genes (CGGs) are expressed in male germline cells, not in other normal adult tissues, and in many tumors. Because male germline cells do not express HLA molecules on their surface, they do not express antigens that can be recognized by T cells. As a result, the antigens encoded by CGGs are strictly tumor-specific T cell or antibody targets. These antigens are shared by many tumors. CGGs include the MAGE,8, 9, 10 GAGE11, 12 and LAGE/NY-ESO-113, 14 families. The MAGE gene family comprises 24 genes ranged in 3 subfamilies, MAGE-A, -B and -C. A large number of antigenic peptides encoded by the MAGE, GAGE and LAGE/NY-ESO-1 genes have been found to be presented to CD4 and CD8 lymphocytes by HLA molecules expressed at the surface of tumor cells.15 Several clinical trials in which antigens encoded by CGGs were used as vaccines have shown induction of immunological responses and, in a small number of patients, clinical responses.16, 17, 18, 19, 20, 21, 22
Immunotherapy is an attractive therapeutic modality for pediatric cancer patients because it has a very mild toxicity, and the immune system of these patients is more potent and flexible compared with that of adults.23, 24 However, the development of appropriate vaccines has been hampered by the lack of knowledge regarding expression of tumor-specific antigens on pediatric tumors.
The aim of this study was to analyze the expression of CGGs in pediatric extracranial solid tumors. We report the results of a quantitative real-time PCR analysis of the expression of 12 CGGs in a panel of pediatric neuroblastomas, Ewing's sarcomas, rhabdomyosarcomas and osteosarcomas. For a subset of tumors, immunohistochemistry was used to analyze also the expression and distribution of proteins encoded by 3 CGGs.
Material and methods
Fresh-frozen tumor samples were available at the Department of Pathology at the Radboud University Nijmegen Medical Centre. All samples were from pediatric patients (0 to 20-year-old) with a malignant solid extracranial tumor diagnosed at the Department of Pediatric Hemato-Oncology between 2000 and 2004. Sections of the frozen samples were stained with hematoxylin-eosin and reviewed by the pathologist to verify tumor histology and to evaluate the percentage of tumor cells. Samples were only considered for study if the contents of tumor cells was ≥80% and the yield of DNase-treated RNA was >1 μg.
RNA isolation and cDNA synthesis
Total RNA was isolated with TriZol reagent (Invitrogen, Carlsbad, CA) and samples were treated with deoxyribonuclease I (Invitrogen) according to the manufacturer's protocol. To generate cDNA, 1 μg DNase-treated RNA was reverse-transcribed with the SuperScript First-Strand Synthesis System for RT-PCR (Invitrogen) using oligo(dT) primer and 50 units SuperScript II, according to the manufacturer's protocol. After first-strand synthesis, samples were diluted to a final volume of 100 μl with water.
Duplex PCR amplification of β-actin and GAPDH transcripts was carried out in a 25-μl reaction volume containing 2.5 μl of cDNA, 1× PCR Buffer (10 mM Tris-HCl pH 8.3, 50 mM KCl, 1.5 mM MgCl2), 100 μM each dNTP, 0.4 μM each primer and 0.625 units Taq DNA polymerase (TaKaRa, Shiga, Japan). β-actin primers were as described.25GAPDH primers (originally available from Clontech, Palo Alto, CA; kindly provided by Dr. B. Lethé) were 5′-TgAAggTCggAgTCAACggATTTggT-3′ (sense) and 5′-CATgTgggCCATgAggTCCACCAC-3′ (antisense). Cycling was performed in a TRIO-Thermoblock thermocycler (Biometra, Göttingen, Germany) as follows: 94°C for 4 min, followed by 22 cycles of 1 min at 94°C, 1 min at 65°C, and 1 min at 72°C. Cycling was concluded with a final extension step at 72°C for 15 min. PCR products were fractionated in 1.3% agarose gel and visualized by ethidium bromide fluorescence (β-actin, 626 bp; GAPDH, 983 bp). PCR amplification of MAGE-A transcripts was carried out with the primer pair designed by Zammatteo et al.26 These primers derive from a consensus nucleotide sequence for the last exon of the 12 MAGE-A genes and give amplicons of (539 bp. PCR conditions were as described earlier, except that PCR was performed for 30 cycles.
Quantitative real-time PCR
Expression of CGGs and of the reference gene β-actin, was measured by quantitative PCR, based on TaqMan methodology, using the ABI PRISM 7700 Sequence Detection System (Applied Biosystems, Warrington, UK). PCR reactions were prepared with the qPCR Core Kit w/o dUTP reagents (Eurogentec, Seraing, Belgium). Each reaction (25 μl) contained 2.5 μl of cDNA, 1× PCR buffer containing the passive reference dye ROX, 5 mM MgCl2, 200 μM each dNTP, 200 nM each primer, 100 nM probe and 0.625 units DNA polymerase. Sequences of primers and probes, and thermal cycling conditions are available from the authors (Brasseur et al., manuscript in preparation). Probes with 6FAM and TAMRA labels were from Eurogentec. Probes with 6FAM and MGB-NFQ labels (for MAGE-A3 and MAGE-A6) were from Applied Biosystems. Quantification of the samples was achieved by extrapolation from a standard curve of 4 serial dilution points of cDNA of the relevant gene (from 105 to 102 copies per reaction for CGGs, and from 106 to 102 copies per reaction for β-actin). Samples and standard dilution points were assayed in duplicate or triplicate. Standard calibration curves for β-actin and all CGGs are shown in Figure 1. The assays were linear over 4 (CGGs) or 5 (β-actin) orders of magnitude and had similar PCR efficiencies (slope from −3.48 to −3.89). Differences in sensitivity between the assays for the various genes (y-intercept from 38.2 to 42.4) were due in part to differences in the actual cDNA copy number in the standard dilutions. cDNA copy numbers in the standards were verified by testing minimally 12 replicates of the 1-copy dilution in each qPCR run. If needed, copy numbers of the test samples were corrected by a factor calculated on the basis of the results for the 1-copy dilution. Normalization of samples was achieved by dividing the copy number of CGG by that of the reference gene, β-actin.
Immunochemistry was performed on 4 μm tissue sections of formalin-fixed paraffin-embedded tissue blocks. Sections were boiled for 20 min in citrate buffer (10 mM, pH 6.0) for antigen retrieval. The following mouse IgG1 monoclonal antibodies (mAb) were used: E978 (anti-NY-ESO-1)27 (Zymed, San Francisco, CA), MA454 (anti-MAGE-A1)28 (Zymed) and 57B (anti-MAGE-A4)29, 30 (kindly provided by Dr. G.C. Spagnoli, University Hospital Basel, Switzerland). Testis tissue with intact spermatogenesis was used as positive control. Tissue sections were incubated with mAb diluted in PBA (2.5 μg/ml E978, 1 μg/ml MA454, 5 μg/ml 57B), or with IgG1 negative control antibody, at room temperature for 1 hr. Primary antibodies were then detected with a biotinylated horse-antimouse secondary reagent (Vector Laboratories, Burlingame, CA) followed by an avidin-biotin complex system (ABC Elite, Vector Laboratories). Diaminobenzidine tetrachloride served as a chromogen. Immunoreactivity was assessed by the pathologist, blindly with respect to the mAb used.
We analyzed the expression of cancer-germline genes (CGGs) in 49 freshly frozen tumors by reverse transcription and polymerase chain reaction (PCR) amplification. All samples were histologically proven malignant solid tumors of extracranial origin, resected from patients ≤20 years old at the time of surgery (Table I). The integrity of cDNA samples was verified by conventional, 22-cycles PCR amplification of 626 bp β-actin and 983 bp GAPDH products (data not shown). cDNA obtained from all 49 samples was then used as template in a conventional PCR amplification with consensus primers for the 12 genes of the MAGE-A family. Thirty-one samples scored positive, with 42–100% of positive samples depending on the type of tumor (Table I).
|Tumor type||Number of patients||Average age in years (range)||% of MAGE-A-positive Tumors1|
|Ewing's sarcoma||18||13 (2–20)||50|
CGG expression in pediatric tumors
The 31 MAGE-A-positive samples were tested by quantitative real-time PCR (qPCR), using TaqMan methodology, to measure the expression of MAGE-A1, A2, A3, A4, A6, A10 and A12. In addition, all 49 samples were tested by qPCR for expression of genes MAGE-C2, LAGE-2/NY-ESO-1 and GAGE. Quantification was obtained using standard curves, as explained in the Material and Methods section and Figure 1. Expression of GAGE was tested with a qPCR assay that detects GAGE-1, 2 and 8. CGG expression levels were calculated relative to β-actin expression (Fig. 2). For instance, the MAGE-A1/β-actin ratio for the first neuroblastoma sample was 1.1 × 10−2, indicating that 1.1 cDNA copies MAGE-A1 were present for every 100 cDNA copies of β-actin.
Eight of the 10 samples expressed at least 3 of the 7 MAGE-A genes that were tested. Expression of MAGE-C2, NY-ESO-1 or GAGE-1, 2, 8 was detected in 5 of the 8 MAGE-A-positive samples and in none of the negative samples. The pattern of expression of CGGs appeared to be highly clustered: all positive samples expressed at least 3 of the 10 genes tested (Table II). The levels of CGG expression in the neuroblastoma samples were usually high. For example, 3 samples expressed gene MAGE-A10 at a level that was only 100-fold lower than that of the expression of the β-actin gene (Fig. 2). Four out of the 10 samples expressed at least 1 CGG at a CGG/β-actin ratio >10−3.
|Sample||MAGEA1||MAGEA2||MAGEA3||MAGEA4||MAGEA6||MAGEA10||MAGEA12||MAGEC2||NY-ESO-1||GAGE 1, 2, 8|
Eleven of the 18 Ewing's sarcoma samples expressed at least 1 CGG. Expression of MAGE-A1, MAGE-A2 and NY-ESO-1 could not be detected in any sample. Although a trend toward coexpression of multiple CGGs was observed, the number of coexpressed genes was less than in the other tumor types. The level of CGG expression in Ewing's sarcomas was always low (CGG/β-actin ratios <1 × 10−3).
Of the 12 samples, 6 expressed at least 1 CGG. The clustered pattern of expression of CGGs was most obvious in this tumor type: although 6 samples expressed no CGGs, 3 samples expressed at least 7 of the 10 genes. The level of expression in the positive samples was high. Three samples exhibited expression ratios >10−3 for 1 or more CGGs. These 3 samples also expressed NY-ESO-1 at levels >10−4.
The highest incidence of CGG expression was observed in these tumors: all 9 samples expressed at least 4 of the 7 MAGE-A genes that were tested. All samples expressed MAGE-A3 and A6. Seven samples expressed MAGE-C2, 8 expressed NY-ESO-1, and 9 expressed GAGE-1, 2, 8. Each sample expressed at least 6 of the 10 CGGs. MAGE-A4 was expressed less frequently (4 positive samples) than the other CGGs (6–9 positive samples). Osteosarcoma samples also expressed CGGs levels: in 8 of the 9 samples the CGG/β-actin ratio of at least 1 CGG was higher than 10−3. Expression ratios as high as 10−3 were observed for all CGGs except for MAGE-A4.
Cancer-germline protein distribution
Immunochemistry with monoclonal antibodies E978 (anti-NY-ESO-1), MA454 (anti-MAGE-A1) and 57B (anti-MAGE-A4) was performed on tissue sections from 9 tumors (identified with a circle in Fig. 2). Sections from normal testis tissue were used as positive controls. Representative results are shown in Figure 3. In testis, the 3 antibodies labeled spermatogonia and spermatocytes; spermatozoa were not stained. The staining with antibody E978 was less intense compared to antibodies MA454 and 57B, both in testis and tumor tissues. In tumors, the staining with MA454 or 57B was homogeneously distributed throughout the tissue, except for 1 neuroblastoma sample, which displayed a heterogeneous staining pattern with all 3 antibodies. This is illustrated in Figure 3, with high magnification inserts showing positive tumor cells positioned next to negative tumor cells in sample neuroblastoma 1. In the other positive tumors, the staining of the tumor cells with E978 had a homogeneous (1 tumor), heterogeneous (1 tumor) or focal (1 tumor) distribution. Immunohistochemistry and qPCR data of the 9 tested tumors correlated well. Staining of the tumor cells with each of the 3 antibodies was observed only in sections from tumors that tested positive for mRNA expression. No false-negative or false-positive immunohistochemistry result was observed in any of the 27 sections that were analyzed (chi-square, p < 0.001). No staining was observed with the IgG isotype negative control antibody.
The immunostaining of MAGE-positive cells in sample neuroblastoma 2, a metastatic lymph node from a neuroblastoma patient, clearly showed that the tumor cells invaded the parafollicular areas of the node. The B-cell areas, namely the follicle and the mantle zone, were not infiltrated by tumor cells (Fig. 4a). Hematoxylin-eosin staining confirmed that no tumor cells were present in the follicle and mantle zone (Fig. 4b).
CGG expression and tumor stage
Although the number of samples in each group of tumors was limited, the correlation between CGG expression and clinicopathological parameters was statistically analyzed. In none of the 4 groups a significant correlation could be observed between CGG mRNA expression by the tumors and parameters such as tumor stage (Fig. 4c shows the results for neuroblastomas), age of the patient and clinical outcome (data not shown).
The identification of tumor-specific antigens is an essential step in the development of therapeutic cancer vaccines. The importance of antigens encoded by CGGs as vaccine targets has led to detailed studies of their expression in various tumors.1, 32 Little information is available regarding expression of CGGs in pediatric tumors. Here we report CGG expression in pediatric solid tumors using quantitative real-time PCR. We observed that all osteosarcoma, most neuroblastoma and some rhabdomyosarcoma tumors expressed several CGGs at high levels. Ewing's sarcoma samples had an overall low CGG expression. CGG expression was not correlated with tumor stage or clinical parameters of the patients. CGG protein expression was confirmed by immunohistochemistry, with staining patterns that were frequently homogeneous.
In immunotherapeutic trials with defined tumor antigens, the level of expression of the target antigens may be of great importance for the success of the vaccine. Real-time quantitative PCR is a sensitive method for quantifying gene transcripts and is as such preferable to semiquantitative mRNA detection methods such as Northern blot and conventional PCR. Quantification of mRNA with real-time PCR needs appropriate standardization of the procedures for RNA extraction from tissues, cDNA synthesis and the real-time PCR itself.33 We had every tumor sample analyzed by the pathologist and selected only samples with more than 80% tumor cells. In addition, we calculated the mRNA expression levels of CGGs relative to those of the reference gene β-actin.
Our results indicate that CGG/β-actin ratios as high as 0.1 can be observed in some pediatric tumor samples. High CGG expression (CGG/β-actin ratio >1 × 10−3) was observed in most neuroblastoma samples and in 100% of the osteosarcomas. In the Ewing's sarcomas and rhabdomyosarcomas, we observed expression of at least 1 CGG in 61% and 50% of the samples, respectively. The CGG/β-actin ratio in Ewing's sarcoma was lower than 1 × 10−3 in all samples. Nonquantitative data on the expression of MAGE-A1, MAGE-A2, MAGE-A3 and NY-ESO-1 in pediatric neuroblastoma and rhabdomyosarcoma have been reported by others.34, 35, 36, 37, 38, 39 These data and ours correlate well.
We frequently observed coexpression of multiple CGGs in our tumor samples. This phenomenon has been described also for other tumors, and is most likely a consequence of a global demethylation of the genome in the tumor.40, 41 It is also possible that activation of a single CGG leads to the activation of other CGGs (reviewed by Simpson et al.32). The pattern of expression of CGGs that we observe in neuroblastoma is similar to that observed in melanomas,42, 43 which also develop from progenitor cells originating from the neural crest.
Numerous studies indicate that CGG expression is correlated with advanced pathologic stage and worse prognosis in various tumor types, including gastric carcinoma,44, 45 colorectal cancer,46 breast cancer,47, 48 ovarian neoplasm,49 bladder cancer,50 nonsmall-cell lung cancer51, 52 and multiple myeloma.53, 54 However, CGG expression is not correlated with disease progression in other malignancies such as neuroblastoma,34, 35, 36, 37 melanoma55 and esophageal cancer.56 We observed CGG expression in newly diagnosed tumors, in tumors collected after treatment with chemotherapy (osteosarcoma samples 5, 6 and 8 in Fig. 2) and in metastases. We could not detect a correlation between CGG expression in the tumor and clinicopathological parameters of the patients. Therefore, vaccination of these patients with CGG antigens is conceivable at all disease stages, including when a large tumor burden is present. And considering that pediatric tumors tend to coexpress multiple CGGs, one may target several antigens in the same patient, which should reduce the risk of resistance through antigen-loss tumor variants.57
Proteins MAGE-A1, MAGE-A4 and NY-ESO-1 were visualized with immunohistochemistry. MAGE-A1 and MAGE-A4 were homogeneously present in most tumors. NY-ESO-1, in contrast, had homogeneous, heterogeneous or focal distribution patterns. The 27 immunohistochemical stainings of the 9 tested tumors correlated well with the qPCR data. For a complete overview of CGG expression in a tumor, PCR data are essential because for most CGGs no specific antibodies are available. It is of note that in neuroblastoma sample 2, no MAGE-A3 mRNA expression was detected (Fig. 2) but staining of tumor cells by monoclonal antibody 57B was observed (Fig. 4a). This is in line with the observation that 57B, which was raised against a recombinant MAGE-A3 protein, does not detect MAGE-A3 but reliably detects MAGE-A4.30 As shown in Figures 4a and 4b, the distribution of neuroblastoma cells in the lymph node of neuroblastoma sample 2 had a unique pattern: the B-cell areas were free of tumor cells throughout the lymph node. We hypothesize that the microenvironment in the mantle zones of this lymph node prevents the tumor cells from entering the B-cell areas.
In conclusion, we report high levels of CGG expression in a large fraction of pediatric solid tumors. These data indicate that antigens encoded by these CGGs can be used to vaccinate children with these tumors.
The authors thank Riki Willems for assistance with the pathology database, Kim Vermeulen, Maaike Looman and Madeleine Swinarska for technical assistance.
- 33Pitfalls of quantitative real-time reverse-transcription polymerase chain reaction. J Biomol Tech 2004; 15: 155–66., .
- 42High homogeneity of MAGE, BAGE, GAGE, tyrosinase and Melan-A/MART-1 gene expression in clusters of multiple simultaneous metastases of human melanoma: implications for protocol design of therapeutic antigen-specific vaccination strategies. Int J Cancer 1998; 77: 200–4., , , , , , , , .
- 54Expression of cancer/testis (CT) antigens MAGE-A1, MAGE-A3, MAGE-A4, CT-7 and NY-ESO-1 in malignant gammopathies is heterogeneous and correlates with site, stage and risk status of disease. Cancer Immun 2003; 3: 9., , , , , , , , , , , , et al.