Javier S. Castresana

RASSF1A, BLU, NORE1A, PTEN and MGMT Expression and Promoter Methylation in Gliomas and Glioma Cell Lines and Evidence of Deregulated Expression of de novo DNMTs

Methylation of CpG islands in gene promoters can lead to gene silencing. Together with deletion or mutation, it may cause a loss of function of tumor suppressor genes. RASSF1A (3p21.3), NORE1A (1q32.1) and BLU (3p21.3) have been shown to be downregulated by methylation in cancer, and PTEN (10q23.3) and MGMT (10q26.1) are located in areas commonly deleted in astrocytomas. MGMT methylation predicts a better response and a longer overall survival in patients with glioblastomas treated with temozolomide. We analyzed 53 astrocytoma samples and 10 high‐grade glioma cell lines. Gene expression was assessed by RT‐PCR. Bisulfite sequencing, MSP and a melting curve analysis‐based real‐time PCR were performed to detect promoter methylation. Treatments with 5′‐aza‐2′‐deoxicitidine were applied to restore gene expression in cell lines. Ninety‐two percent of tumor samples were methylated for RASSF1A, 30%–57% for BLU and 47% for MGMT, suggesting promoter methylation of these genes to be a common event in glioma tumorigenesis. Only 4% of the tumors revealed a methylated promoter for NORE1A. No association between methylation and loss of expression could be established for PTEN. We identified de novo DNMTs overexpression in a subset of tumors which may explain the methylation phenotype of individual gliomas.

Detection of methylation in promoter sequences by melting curve analysis-based semiquantitative real time PCR.

BackgroundWe present two melting curve analysis (MCA)-based semiquantitative real time PCR techniques to detect the promoter methylation status of genes. The first, MCA-MSP, follows the same principle as standard MSP but it is performed in a real time thermalcycler with results being visualized in a melting curve. The second, MCA-Meth, uses a single pair of primers designed with no CpGs in its sequence. These primers amplify both unmethylated and methylated sequences. In clinical applications the MSP technique has revolutionized methylation detection by simplifying the analysis to a PCR-based protocol. MCA-analysis based techniques may be able to further improve and simplify methylation analyses by reducing starting DNA amounts, by introducing an all-in-one tube reaction and by eliminating a final gel stage for visualization of the result. The current study aimed at investigating the feasibility of both MCA-MSP and MCA-Meth in the analysis of promoter methylation, and at defining potential advantages and shortcomings in comparison to currently implemented techniques, i.e. bisulfite sequencing and standard MSP.MethodsThe promoters of the RASSF1A (3p21.3), BLU (3p21.3) and MGMT (10q26) genes were analyzed by MCA-MSP and MCA-Meth in 13 astrocytoma samples, 6 high grade glioma cell lines and 4 neuroblastoma cell lines. The data were compared with standard MSP and validated by bisulfite sequencing.ResultsBoth, MCA-MSP and MCA-Meth, successfully determined promoter methylation. MCA-MSP provided information similar to standard MSP analyses. However the analysis was possible in a single tube and avoided the gel stage. MCA-Meth proved to be useful in samples with intermediate methylation status, reflected by a melting curve position shift in dependence on methylation extent.ConclusionWe propose MCA-MSP and MCA-Meth as alternative or supplementary techniques to MSP or bisulfite sequencing.

Stromelysin‐1/matrix metalloproteinase‐3 (MMP‐3) expression accounts for invasive properties of human astrocytoma cell lines

Tumor invasiveness is an intrinsic feature of most glial tumors that accounts for their malignant and locally destructive nature. We evaluated the subcutaneous (sc) tumorigenicity and in vivo invasiveness of 9 astrocytoma cell lines together with their respective metalloprotease activity in order to establish their biologic behavior and malignant potential. Invasiveness was assessed with an in vivo invasion assay using tracheal xenotransplants subcutaneously implanted into Scid mice. This assay permitted us to evaluate the penetration of tumor cells into the transplanted deepithelialized tracheas previously inoculated with either normal primary glial cells or with astrocytoma‐derived cell lines. Although only 2 cell lines were tumorigenic after sc inoculation, 5 out of 9 tumor cell lines were tumorigenic in the tracheal graft system. The astrocytoma cell lines showed varying levels of penetration into the tracheal wall. The tumor lines GOS3, M059K, CCFSTTG1 and A172, as well as primary normal astrocytes, were nontumorigenic and noninvasive in this experimental model. LN405, SW1088 and SW1783 cells that were not tumorigenic as sc xenotransplants, on the other hand, grew well in the tracheal graft system showing low levels of in vivo invasiveness. U87MG and U118MG cells were tumorigenic as sc xenotransplants and showed high levels of invasiveness. In parallel to these in vivo studies, the constitutive levels of secreted gelatinases and stromelysins (MMP‐3 and MMP‐11) were investigated using conditioned media submitted to gelatin or casein‐substrate zymography and Western blot analysis. Neither the gelatinases (MMP‐2 and MMP‐9) nor MMP‐11 showed a direct correlation with the levels of in vivo tumor cell invasiveness. Conversely, secretion of MMP‐3 correlated closely with tumorigenicity and invasiveness. In vitro tumor cell invasiveness was significantly reduced after incubation with the metalloproteinase inhibitor GM6001. This positive correlation between MMP‐3 and the depth of tracheal wall penetration led us to conclude that the invasive properties of brain tumor cells may be due to the direct or indirect proteolytic effects of MMP‐3 on extracellular matrix (ECM) macromolecules and that this enzyme might be a potential target for future therapies.

Frequent promoter hypermethylation of RASSF1A and CASP8 in neuroblastoma

BackgroundEpigenetic alterations and loss of heterozygosity are mechanisms of tumor suppressor gene inactivation. A new carcinogenic pathway, targeting the RAS effectors has recently been documented. RASSF1A, on 3p21.3, and NORE1A, on 1q32.1, are among the most important, representative RAS effectors.MethodsWe screened the 3p21 locus for the loss of heterozygosity and the hypermethylation status of RASSF1A, NORE1A and BLU (the latter located at 3p21.3) in 41 neuroblastic tumors. The statistical relationship of these data was correlated with CASP8 hypermethylation. The expression levels of these genes, in cell lines, were analyzed by RT-PCR.ResultsLoss of heterozygosity and microsatellite instability at 3p21 were detected in 14% of the analyzed tumors. Methylation was different for tumors and cell lines (tumors: 83% in RASSF1A, 3% in NORE1A, 8% in BLU and 60% in CASP8; cell lines: 100% in RASSF1A, 50% in NORE1A, 66% in BLU and 92% in CASP8). In cell lines, a correlation with lack of expression was evident for RASSF1A, but less clear for NORE1A, BLU and CASP8. We could only demonstrate a statistically significant association between hypermethylation of RASSF1A and hypermethylation of CASP8, while no association with MYCN amplification, 1p deletion, and/or aggressive histological pattern of the tumor was demonstrated.Conclusion1) LOH at 3p21 appears in a small percentage of neuroblastomas, indicating that a candidate tumor suppressor gene of neuroblastic tumors is not located in this region.2) Promoter hypermethylation of RASSF1A and CASP8 occurs at a high frequency in neuroblastomas.

Regulation of sonic hedgehog-GLI1 downstream target genes PTCH1, Cyclin D2, Plakoglobin, PAX6 and NKX2.2 and their epigenetic status in medulloblastoma and astrocytoma

BackgroundThe Sonic hedgehog (Shh) signaling pathway is critical for cell growth and differentiation. Impairment of this pathway can result in both birth defects and cancer. Despite its importance in cancer development, the Shh pathway has not been thoroughly investigated in tumorigenesis of brain tumors. In this study, we sought to understand the regulatory roles of GLI1, the immediate downstream activator of the Shh signaling pathway on its downstream target genes PTCH1, Cyclin D2, Plakoglobin, NKX2.2 and PAX6 in medulloblastoma and astrocytic tumors.MethodsWe silenced GLI1 expression in medulloblastoma and astrocytic cell lines by transfection of siRNA against GLI1. Subsequently, we performed RT-PCR and quantitative real time RT-PCR (qRT-PCR) to assay the expression of downstream target genes PTCH1, Cyclin D2, Plakoglobin, NKX2.2 and PAX6. We also attempted to correlate the pattern of expression of GLI1 and its regulated genes in 14 cell lines and 41 primary medulloblastoma and astrocytoma tumor samples. We also assessed the methylation status of the Cyclin D2 and PTCH1 promoters in these 14 cell lines and 58 primary tumor samples.ResultsSilencing expression of GLI1 resulted up-regulation of all target genes in the medulloblastoma cell line, while only PTCH1 was up-regulated in astrocytoma. We also observed methylation of the cyclin D2 promoter in a significant number of astrocytoma cell lines (63%) and primary astrocytoma tumor samples (32%), but not at all in any medulloblastoma samples. PTCH1 promoter methylation was less frequently observed than Cyclin D2 promoter methylation in astrocytomas, and not at all in medulloblastomas.ConclusionsOur results demonstrate different regulatory mechanisms of Shh-GLI1 signaling. These differences vary according to the downstream target gene affected, the origin of the tissue, as well as epigenetic regulation of some of these genes.

Genetic heterogeneity in supratentorial and infratentorial primitive neuroectodermal tumours of the central nervous system.

Aims : Medulloblastoma (MB), a kind of infratentorial primitive neuroectodermal tumour (PNET), is the most frequent malignant brain tumour in childhood. In contrast, supratentorial PNET (sPNET) are very infrequent tumours, but they are histologically similar to MB, although they present a worse clinical outcome. We investigated the differences in genetic abnormalities between sPNET and MB.

Clinicopathologic correlations and short-term prognosis in musculoskeletal sarcoma with c-myc oncogene amplification.

The prognostic significance of c-myc oncogens amplification was investigated in a series of 44 patients with musculoskeletal malignancy. There were 21 bone tumors and 23 soft tissue sarcomas. Amplification of the c-myc oncogene was detected in 11 tumors: 4 bone tumors and 7 soft tissue sarcomas. In 9 of the 11 myc-amplified tumors a high grade of malignancy was revealed histologically. The relative frequency of c-myc amplification was higher in grade IV tumors. After a mean follow-up of 18 months, 6 of 11 patients with c-myc-amplified tumors and 12 of 33 patients with nonamplified tumors either exhibited recurrence or died of tumor disease. Both 3-year estimated overall survival and disease-free survival probability were lower for patients with myc-amplified tumors. Amplification of the c-myc oncogene was associated with poor short-term prognosis. Although gene amplification was encountered sporadically in musculoskeletal sarcomas, the reported data suggest that analysis of c-myc may provide valid information on prognosis.

Homozygous deletion and expression of PTEN and DMBT1 in human primary neuroblastoma and cell lines.

Neuroblastoma is the most common pediatric solid tumor. Although many allelic imbalances have been described, a bona fide tumor suppressor gene for this disease has not been found yet. In our study, we analyzed 2 genes, PTEN and DMBT1, mapping 10q23.31 and 10q25.3‐26.1, respectively, which have been found frequently altered in other kinds of neoplasms. We screened both genes for homozygous deletions in 45 primary neuroblastic tumors and 12 neuroblastoma cell lines. Expression of these genes in cell lines was assessed by RT‐PCR analysis. We could detect 2 of 41 (5%) primary tumors harboring PTEN homozygous deletions. Three of 41 (7%) primary tumors and 2 of 12 cell lines presented homozygous losses at the g14 STS on the DMBT1 locus. All cell lines analyzed expressed PTEN, but lack of DMBT1 mRNA expression was detected in 2 of them. We tried to see whether epigenetic mechanisms, such as aberrant promoter hypermethylation, had any role in DMBT1 silencing. The 2 cell lines lacking DMBT1 expression were treated with 5‐aza‐2′‐deoxycytidine; DMBT1 expression was restored in only one of them (MC‐IXC). From our work, we can conclude that PTEN and DMBT1 seem to contribute to the development of a small fraction of neuroblastomas, and that promoter hypermethylation might have a role in DMBT1 gene silencing.

Expression and epigenetic modulation of sonic hedgehog-GLI1 pathway genes in neuroblastoma cell lines and tumors

It is well known that sonic hedgehog signaling pathway plays a vital role during early embryonic development. It is also responsible for stem cell renewal and development of several cancers like colorectal and breast carcinoma and major brain tumors as medulloblastoma and glioblastoma. The role of sonic hedgehog signaling in the development of neuroblastoma has not been thoroughly investigated. In this study, we attempted to determine the expression of Bmi-1 stem cell marker and of Shh pathway downstream target genes glioma-associated oncogene homolog 1 (GLI1), protein patched homolog 1 (PTCH1), Cyclin D2, plakoglobin (γ catenin), NK2 homeobox 2 (NKX2.2), paired box gene 6 (PAX6), secreted frizzled-related protein 1 (SFRP1), and hedgehog interacting protein (HHIP) in 11 neuroblastoma cell lines and 41 neuroblastoma samples. Also, inhibition of the pathway was performed genetically by GLI1 knockdown siRNA or chemically by cyclopamine. After inhibition, low transcript expression was detected in downstream target genes like PTCH1, in the cell lines. We further preformed promoter methylation studies of Cyclin D2, PTCH1, HHIP, and SFRP1 genes by melting curve analysis-based methylation assay (MCA-Meth) and methylation-specific PCR (MSP). Results revealed no methylation in Cyclin D2 gene promoter in neuroblastoma samples or in cell lines; one cell line (MHH-NB-11) showed PTCH1 methylation; 3/11 (27%) cell lines and 9/41 (22%) neuroblastoma samples showed HHIP methylation; and 3/11 (27%) cell lines and 11/41 (27%) samples showed SFRP1 methylation. Taken together, our results suggest the possibility of two levels of control of the sonic hedgehog signaling pathway: transcriptional and epigenetic, which might offer new therapeutic possibilities to modulate the pathway and try to suppress tumor growth.

Microarray analysis of gene expression in vestibular schwannomas reveals SPP1/MET signaling pathway and androgen receptor deregulation

Vestibular schwannomas are benign neoplasms that arise from the vestibular nerve. The hallmark of these tumors is the biallelic inactivation of neurofibromin 2 (NF2). Transcriptomic alterations, such as the neuregulin 1 (NRG1)/ErbB2 pathway, have been described in schwannomas. In this study, we performed a whole transcriptome analysis in 31 vestibular schwannomas and 9 control nerves in the Affymetrix Gene 1.0 ST platform, validated by quantitative real-time PCR (qRT-PCR) using TaqMan Low Density arrays. We performed a mutational analysis of NF2 by PCR/denaturing high-performance liquid chromatography (dHPLC) and multiplex ligation-dependent probe amplification (MLPA), as well as a microsatellite marker analysis of the loss of heterozygosity (LOH) of chromosome 22q. The microarray analysis demonstrated that 1,516 genes were deregulated and 48 of the genes were validated by qRT-PCR. At least 2 genetic hits (allelic loss and/or gene mutation) in NF2 were found in 16 tumors, seven cases showed 1 hit and 8 tumors showed no NF2 alteration. MET and associated genes, such as integrin, alpha 4 (ITGA4)/B6, PLEXNB3/SEMA5 and caveolin-1 (CAV1) showed a clear deregulation in vestibular schwannomas. In addition, androgen receptor (AR) downregulation may denote a hormonal effect or cause in this tumor. Furthermore, the osteopontin gene (SPP1), which is involved in merlin protein degradation, was upregulated, which suggests that this mechanism may also exert a pivotal role in schwannoma merlin depletion. Finally, no major differences were observed among tumors of different size, histological type or NF2 status, which suggests that, at the mRNA level, all schwannomas, regardless of their molecular and clinical characteristics, may share common features that can be used in their treatment.