1H NMR detects different metabolic profiles in glioblastoma stem-like cells

Laura Guidoni, Lucia Ricci-Vitiani, Antonella Rosi, Alessandra Palma, Sveva Grande, Anna Maria Luciani, Federica Pelacchi, Simona di Martino, Cesare Colosimo, Mauro Biffoni, Ruggero De Maria, Roberto Pallini, Vincenza Viti

Research output: Contribution to journalArticlepeer-review

Abstract

The metabolic profiles of glioblastoma stem-like cells (GSCs) growing in neurospheres were examined by 1H NMR spectroscopy. Spectra of two GSC lines, labelled 1 and 83, from tumours close to the subventricular zone of the temporal lobe were studied in detail and compared with those of neural stem/progenitor cells from the adult olfactory bulb (OB-NPCs) and of the T98G glioblastoma cell line. In both GSCs, signals from myoinositol (Myo-I), UDP-hexosamines (UDP-Hex) and glycine indicated an astrocyte/glioma metabolism. For line 1, the presence of signals from N-acetyl aspartate, GABA and creatine pointed to a neuronal fingerprint. These metabolites were almost absent from line 83 spectra, whereas lipid signals, absent from normal neural lineages, were intense in line 83 spectra and remained low in those of line 1, irrespective of apoptotic fate. Spectra of OB-NPC cells displayed strong similarities with those from line 1, with low lipid signals and clearly detectable neuronal signals. In contrast, the spectral profile of line 83 was more similar to that of T98G, displaying high lipids and nearly complete absence of the neuronal markers. A mixed neural-astrocyte metabolic phenotype with a strong neuronal fingerprint was therefore found in line 1, while an astrocytic/glioma-like metabolism prevailed in line 83. We found a signal assigned to the amide proton of N-acetyl galactosamine in GSC lines and in OB-NPC spectra, whereas it was absent from those of T98G cells. This signal may be related to a stem-cell-specific protein glycosylation pattern and is therefore suggested as a marker of cell multipotency. Other GSC lines from patients with different clinical outcomes were then examined. Unsupervised analysis of spectral data from 13 lines yielded two clusters, with six lines resembling spectral features of line 1 and seven resembling those of line 83, suggesting that distinct metabolic phenotypes may be present in GSC lines.

Original languageEnglish
Pages (from-to)129-145
Number of pages17
JournalNMR in Biomedicine
Volume27
Issue number2
DOIs
Publication statusPublished - Feb 2014

Keywords

  • H NMR spectroscopy
  • Cell metabolism
  • GABA
  • Glioblastoma stem-like cells
  • Lipids
  • Metabolic profiles
  • N-acetyl aspartate

ASJC Scopus subject areas

  • Spectroscopy
  • Molecular Medicine
  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of '1H NMR detects different metabolic profiles in glioblastoma stem-like cells'. Together they form a unique fingerprint.

Cite this