Article
Analysis of the biochemical profile of low-grade glioma (LGG) with different Idh1 mutation status using vibrational spectroscopy
Search Medline for
Authors
-
Ortrud Uckermann
- Klinik und Poliklinik für Neurochirurgie, Universitätsklinikum Carl Gustav Carus der Technischen Universität Dresden
-
Tareq Juratli
- Klinik und Poliklinik für Neurochirurgie, Universitätsklinikum Carl Gustav Carus der Technischen Universität Dresden
-
Roberta Galli
- Klinisches Sensoring und Monitoring, Medizinische Fakultät, Technische Universität Dresden
-
Dietmar Krex
- Klinik und Poliklinik für Neurochirurgie, Universitätsklinikum Carl Gustav Carus der Technischen Universität Dresden
-
Kathrin Geiger
- Abteilung Neuropathologie, Universitätsklinikum Carl Gustav Carus der Technischen Universität Dresden
-
Gabriele Schackert
- Klinik und Poliklinik für Neurochirurgie, Universitätsklinikum Carl Gustav Carus der Technischen Universität Dresden
-
Matthias Kirsch
- Klinik und Poliklinik für Neurochirurgie, Universitätsklinikum Carl Gustav Carus der Technischen Universität Dresden
-
Gerald Steiner
- Klinisches Sensoring und Monitoring, Medizinische Fakultät, Technische Universität Dresden
| Published: | May 13, 2014 |
|---|
Outline
Text
Objective: Mutations in human cytosolic isocitrate dehydrogenases 1 (IDH1) are a common feature of primary brain tumors and cause profound changes of the metabolites of the Krebs cycle. Vibrational spectroscopy includes label-free techniques that probe a sample’s molecular composition. Therefore, we investigated the ability of Fourier-transform infrared (FT-IR) spectroscopy to analyze the IDH1-mutation status of low-grade gliomas (LGG).
Method: Brain tumor samples of human LGG were obtained during routine surgery and the IDH1 mutation status was determined by direct DNA sequencing. Conventional histology and IDH immunohistochemistry was performed on corresponding sections. Cryosections of tumor samples were prepared on non-fluorescent CaF2 slides (IDH1-wt: n=4; IDH1-mut: n=5). FT-IR spectroscopy was performed to obtain 256 spectra for each sample totaling 2314). The spectroscopic data sets were preprocessed and evaluated using multivariate data analysis (cluster analysis, principal component analysis (PCA) and classification).
Results: The analysis of difference spectra (all IDH1-mut vs. all IDH1-wt) showed changes in the region around 1100 cm-1 which is attributed to saccharides. In addition, amide II at 1650 cm-1 was reduced in the IDH1-mut spectral dataset, whereas the band at 1740 cm-1 assigned to C=O stretching vibrations was more pronounced. Cluster analysis was not able to group spectra according to the IDH1-mutation status, but mainly recognized differences between gray and white matter. However, PCA indicated differences between the two groups based on spectral regions assigned to saccharides, proteins and to C=O stretching vibrations, as identified previously in the difference spectra. Supervised classification identified relevant spectral regions at 1510, 1406, 1525, 1327, 1587 and 1601 cm-1 and was able to discriminate FT-IR spectra of IDH1-mutation and IDH1-wt with an accuracy of 88%.
Conclusions: On the basis of FT-IR spectroscopy, samples of LGG carrying IDH1 mutations can be differentiated from IDH1 wild type tumors. Relevant spectral regions assigned to saccharides and C=O stretching vibrations can be explained by changes in the Krebs cycle and in particular by the accumulation of 2-hydroxyglutarate (rich in C=O bonds).
