Prognosis in human glioblastoma based on expression of ligand growth hormone-releasing hormone, pituitary-type growth hormone-releasing hormone receptor, its splicing variant receptors, EGF receptor and PTEN genes.

PURPOSE: Glioblastoma (GB) is the most frequent brain tumor. Despite recent improvement in therapeutic strategies, the prognosis of GB remains poor. Growth hormone-releasing hormone (GHRH) may act as a growth factor; antagonists of GHRH have been successfully applied for experimental treatment of different types of tumors. The expression profile of GHRH receptor, its main splice variant SV1 and GHRH have not been investigated in human GB tissue samples. METHODS: We examined the expression of GHRH, full-length pituitary-type GHRH receptor (pGHRHR), its functional splice variant SV1 and non-functional SV2 by RT-PCR in 23 human GB specimens. Epidermal growth factor receptor (EGFR) and phosphatase and tensin homolog gene (PTEN) expression levels were also evaluated by quantitative RT-PCR. Correlations between clinico-pathological parameters and gene expressions were analyzed. RESULTS: Expression of GHRH was found to be positive in 61.9 % of samples. pGHRH receptor was not expressed in our sample set, while SV1 could be detected in 17.4 % and SV2 in 8.6 % of the GB tissues. In 65.2 and 78.3 % of samples, significant EGFR over-expression or PTEN under-representation could be detected, respectively. In 47.8 % of cases, EGFR up-regulation and PTEN down-regulation occurred together. Survival was significantly poorer in tumors lacking GHRH expression. This worse prognosis in GHRH negative group remained significant even if SV1 was also expressed. CONCLUSION: Our study shows that GHRH and SV1 genes expressed in human GB samples and their expression patterns are associated with poorer prognosis.

Expression of mRNA for human type-I LHRH receptor transcript forms in human benign prostatic hyperplasia.

The presence of four different isoforms of luteinizing hormone-releasing hormones (LHRH) and one LHRH receptor (LHRH-R) has been reported in vertebrates. In the human genome only LHRH-I and LHRH-II genes have been identified. The human LHRH-I gene is composed of four exons separated by three introns. Three LHRH receptor or receptor-like genes have been demonstrated. The well-established type-I LHRH receptor (LHRH-R-I) gene is composed of three exons separated by two introns. In this study we investigated the expression of transcript forms of LHRH-R-I in human benign prostatic hyperplasia (BPH) with reverse transcriptase-polymerase chain reaction (RT-PCR) using gene specific primers. Thirty-five human BPH specimens were obtained at surgery. Normal human pituitaries collected at autopsy served as control. RNA extraction and RT-PCR with gene-specific primers for LHRH-R-I forward (F1)/reverse (R1), LHRH-R-I F2/R3, LHRH-R-I F1'/R2' were carried out to determine the mRNA expression for LHRH-R-I transcript forms. The expected PCR products amplified with gene specific primers were LHRH-R-I F1/R1 with 319 bp, LHRH-R-I F2/R3 with 309 bp and LHRH-R-I F1'/R2' with 219 bp. PCR products for LHRH-R-I F1/R1 were detected in 21 (60%) and for LHRH-R-I F2/R3 in 5 of 35 (14%) BPH samples. No PCR products for LHRH-R-I F1'/R2' were found. In conclusion, we detected mRNA for LHRH-R-I in human BPH specimens. Our results suggest that LHRH-R-I gene may have more than two splice variants or uncharacterised transcript forms of LHRH-R-I. Our findings support the merit of further investigation of the expression of LHRH-R-I and its transcript forms in human BPH.

RAR beta2 suppression in head and neck squamous cell carcinoma correlates with site, histology and age.

Retinoids as important growth and differentiation regulating agents have a potential role in the chemoprevention of head and neck squamous cell carcinoma (HNSCC). Despite the promising preclinical and early clinical findings, limitations of application are raised by intrinsic resistance acquired during carcinogenesis. Retinoic acid receptor beta2 (RAR beta2) is one of the proximate mediators of retinoid signalling and its expression is often diminished in early stages of head and neck carcinogenesis. One form of retinoid resistance has been associated with the methylation-induced silencing of the RAR beta gene. We studied primary HNSCC samples of different anatomical sites in respect of methylation, expression and allelic loss of RAR beta gene. A strong correlation (p<0.01) was found between hyper-methylation and reduced expression of RAR beta2, however the allelic loss at 3p24, the locus of RAR beta, did not considerably influence its mRNA level. Hypopharynx tumors showed significantly lower hypermethylation (p<0.05) and higher mRNA expression levels of RAR beta2 compared to the tumors located at other sites of the head and neck. We could also provide evidence that poorly differentiated grade 3 tumors had significantly higher RAR beta2 expression and lower methylation levels (p<0.05) than better differentiated grade 1 and grade 2 tumors. In addition, we found a good correlation between the methylation degree of the RAR beta2 promoter and the ages of patients. Collectively, our results suggest that evaluation of several factors such as tumor location, age, histology and methylation state of the RAR beta gene might contribute to the selection of patients for retinoid-based chemoprevention.