DNA CpG methylation in sequential glioblastoma specimens

Abstract

Purpose Glioblastoma is the most aggressive form of brain tumors. A better understanding of the molecular mechanisms leading to its evolution is essential for the development of treatments more effective than the available modalities. Here, we aim to identify molecular drivers of glioblastoma development and recurrence by analyzing DNA CpG methylation patterns in sequential samples. Methods DNA was isolated from 22 pairs of primary and recurrent formalin-fixed, paraffin-embedded glioblastoma specimens, and subjected to reduced representation bisulfite sequencing. Bioinformatic analyses were conducted to identify differentially methylated sites and pathways, and biostatistics was used to test correlations among clinical and pathological parameters. Results Differentially methylated pathways likely involved in primary tumor development included those of neuronal differentiation, myelination, metabolic processes, synapse organization and endothelial cell proliferation, while pathways differentially active during glioblastoma recurrence involved those associated with cell processes and differentiation, immune response, Wnt regulation and catecholamine secretion and transport. Conclusion DNA CpG methylation analyses in sequential clinical specimens revealed hypomethylation in certain pathways such as neuronal tissue development and angiogenesis likely involved in early tumor development and growth, while suggested altered regulation in catecholamine secretion and transport, Wnt expression and immune response contributing to glioblastoma recurrence. These pathways merit further investigations and may represent novel therapeutic targets.

Involvement of the Catecholamine Pathway in Glioblastoma Development

“…Altogether, these data and our previous observations [ 2 ] confirm the involvement of catecholamine pathway molecules in GBM. Since our previous epigenomic analysis did not specifically focus on individual catecholamine markers in individual tumor samples and had no expression assessments, in the present study, we aimed to analyze further the involvement of the four selected markers (ADRA1D, ADRBK1/GRK2, DRD2 and SLC18A2) in GBM, by zooming into their promoter and gene CpG methylation data in comparison with their protein expression levels in sections dissected from the same blocks of sequential GBM specimens.…”

“…The formalin-fixed, paraffin-embedded (FFPE) GBM blocks were surgically removed sequential specimens, while the histological controls (HC) included postmortem FFPE normal brain specimens (obtaining surgically removed normal control brain specimens are not feasible; other neurological control brain tissue FFPE specimens are also not available at our center). The histopathological diagnosis of GBM was established according to the recent World Health Organization guidelines [ 49 ] and samples were subjected to several rounds of quality selection [ 2 ]. These GBM specimens were also included in our previous DNA CpG methylation study [ 2 ].…”

“…The histopathological diagnosis of GBM was established according to the recent World Health Organization guidelines [ 49 ] and samples were subjected to several rounds of quality selection [ 2 ]. These GBM specimens were also included in our previous DNA CpG methylation study [ 2 ]. However, of the 22 tumor pairs in the epigenomic analyses, only 21 pairs could be used here, because one pair of blocks had insufficient amount of tissue for the execution of immunohistochemistry (IHC).…”

“…Recently, we carried out a reduced representation bisulfite sequencing (RRBS) study to determine DNA CpG methylation genome-wide in 22 pairs of sequential GBM samples [ 2 ]. Gene ontology (GO) analysis revealed differential methylation in promoters and genes of several pathways, including hypomethylation in pathways of catecholamine secretion and transport in the primary compared to the recurrent tumor cohort.…”

“…This observation suggested that catecholamines may predominantly contribute to early stages of GBM development. Based on the information from our RRBS methylome, here we selected four catecholamine markers differentially methylated at the cohort level (alpha 1D adrenergic receptor—ADRA1D; adrenergic beta receptor kinase 1 or G protein-coupled receptor kinase 2—ADRBK1/GRK2; dopamine receptor D2—DRD2; and synaptic vesicle monoamine transporter—SLC18A2; Supplementary Table S1a ) [ 2 ], and analyzed further their involvement in GBM.…”