PURPOSE: Testisin, a serine protease abundantly expressed only in normal testes, is thought to be a tumor suppressor gene silenced by aberrant methylation in testicular germ cell tumors (TGCT). This study aimed at identifying the CpG sites relevant for testisin gene silencing when methylated and evaluating the potential of aberrant methylation as a biomarker in TGCT. METHODS: Bisulfite sequencing and subsequent real-time RT-PCR were carried out in germ cell tumor cell lines and a cervical carcinoma cell line to reveal CpG sites implicated in testisin gene silencing. The normalized index of methylation (NIM) and the relative gene expression (RGE) were calculated in 33 TGCT and 19 normal testicular tissue samples by quantitative methylation-specific PCR (QMSP) and real-time RT-PCR. RESULTS: CpG sites in the 5'untranslated region proved to be relevant for testisin gene silencing when methylated. Targeting these CpG sites by QMSP, we demonstrated that the median NIM was 8.6 times higher in the TGCT group than in normal testicular samples. This was associated with a median 23.6 times lower RGE in the TGCT probes. Non-seminomas had a significantly higher NIM than seminomas, but RGE was downregulated to comparable levels in both subgroups. QMSP was highly sensitive and distinguished TGCT from normal samples with excellent specificity. CONCLUSIONS: Our results demonstrate for the first time that aberrant methylation of specific CpG sites near the transcription initiation site is an important factor in testisin gene silencing in TGCT.
PURPOSE: Testisin, a serine protease abundantly expressed only in normal testes, is thought to be a tumor suppressor gene silenced by aberrant methylation in testicular germ cell tumors (TGCT). This study aimed at identifying the CpG sites relevant for testisin gene silencing when methylated and evaluating the potential of aberrant methylation as a biomarker in TGCT. METHODS: Bisulfite sequencing and subsequent real-time RT-PCR were carried out in germ cell tumor cell lines and a cervical carcinoma cell line to reveal CpG sites implicated in testisin gene silencing. The normalized index of methylation (NIM) and the relative gene expression (RGE) were calculated in 33 TGCT and 19 normal testicular tissue samples by quantitative methylation-specific PCR (QMSP) and real-time RT-PCR. RESULTS: CpG sites in the 5'untranslated region proved to be relevant for testisin gene silencing when methylated. Targeting these CpG sites by QMSP, we demonstrated that the median NIM was 8.6 times higher in the TGCT group than in normal testicular samples. This was associated with a median 23.6 times lower RGE in the TGCT probes. Non-seminomas had a significantly higher NIM than seminomas, but RGE was downregulated to comparable levels in both subgroups. QMSP was highly sensitive and distinguished TGCT from normal samples with excellent specificity. CONCLUSIONS: Our results demonstrate for the first time that aberrant methylation of specific CpG sites near the transcription initiation site is an important factor in testisin gene silencing in TGCT.