Artikel
Multiomic mapping of spatiotemporal architecture identified decoupled lineage and reactive transcriptional states in glioblastoma
Multiomische Kartierung der räumlichen und zeitlichen Architektur identifizieren entkoppelte Abstammungslinien und reaktive Transkriptionszustände beim Glioblastom
Suche in Medline nach
Autoren
-
Vidhya Madapusi Ravi
- Medical Center, University of Freiburg, Translational NeuroOncology Research Group, Freiburg, Deutschland; Medical Center, University of Freiburg, Department of Neurosurgery, Freiburg, Deutschland; Medical Center, University of Freiburg, Neuroelectronic Systems, Freiburg, Deutschland; Medical Center, University of Freiburg, Freiburg, Deutschland
-
Paulina Will
- Medical Center, University of Freiburg, Translational NeuroOncology Research Group, Freiburg, Deutschland; Medical Center, University of Freiburg, Department of Neurosurgery, Freiburg, Deutschland; Medical Center, University of Freiburg, Freiburg, Deutschland
-
Jürgen Beck
- Medical Center, University of Freiburg, Translational NeuroOncology Research Group, Freiburg, Deutschland; Medical Center, University of Freiburg, Department of Neurosurgery, Freiburg, Deutschland; Medical Center, University of Freiburg, Freiburg, Deutschland
-
Ulrich Hofmann
- Medical Center, University of Freiburg, Department of Neurosurgery, Freiburg, Deutschland; Medical Center, University of Freiburg, Neuroelectronic Systems, Freiburg, Deutschland; Medical Center, University of Freiburg, Freiburg, Deutschland
-
Oliver Schnell
- Medical Center, University of Freiburg, Translational NeuroOncology Research Group, Freiburg, Deutschland; Medical Center, University of Freiburg, Department of Neurosurgery, Freiburg, Deutschland; Medical Center, University of Freiburg, Freiburg, Deutschland
-
Dieter Henrik Heiland
- Medical Center, University of Freiburg, Translational NeuroOncology Research Group, Freiburg, Deutschland; Medical Center, University of Freiburg, Department of Neurosurgery, Freiburg, Deutschland; Medical Center, University of Freiburg, Freiburg, Deutschland
| Veröffentlicht: | 4. Juni 2021 |
|---|
Gliederung
Text
Objective: Glioblastomas (GBM) are highly malignant tumors of the central nervous system (CNS). Evidence suggests that these tumors display a large intra- and inter-patient heterogeneity, hallmarked by subclonal diversity and dynamic adaptation within developmental hierarchies. However, the source for dynamic reorganization of cellular states within their spatial context remains elusive. Here we integrate spatial transcriptomics and matrix assisted laser desorption/ionization (MALDI) imaging to acquire a high-dimensional map of transcriptional/metabolomic interactions.
Methods: We performed spatial transcriptomics using Visium technology. The dataset consists of 10µm sections from 24 donors (tumor n=18, non-neoplastic n=6), placed on a glass slide with 5000 targeted capture areas (6x6mm). Each capture area contains oligo-dt nucleoids with unique barcodes coding for their spatial coordinates. After permeabilization mRNA binds to the oligo-dt and are then extracted for library preparation. Additionally, consecutive slices were also used to perform matrix assisted laser desorption/ionization (MALDI), with a spatial resolution of 50µm. We established a R-based software package for integrative analysis of our data (SPATA, https://github.com/heilandd/SPATA-Lab).
Results: We mapped over 90k spatially resolved transcriptomes from 22 malignant and non-neoplastic human tissue samples and were able to reconstruct exclusive as well as shared transcriptional programs with repetitive spatial patterns, in both malignant and non-neoplastic CNS tissue. We inferred that brain tumors differentiate along defined glial or neural developmental lineages and further adapt to inflammatory or metabolic stimuli. These findings were confirmed by a large overlap of both metabolomic and transcriptional programs, which were shared across the whole cohort. Analysis of copy number variations (CNV) revealed spatially coherent subclonal organization, associated with a reactive transcription program within regions of nutrient scarcity, which confirms that environmental stress leads to selection pressure.
Conclusion: Our data suggest that metabolic and immunological alterations of the microenvironment of tumors drive their heterogeneity. It still remains unclear to what extent the tumor itself is responsible for regulation of its own microenvironment.
