The Spatial Context – through the lens of method development
Time: Fri 2021-12-17 10.00
Location: The Air & Fire auditorium, Science for Life Laboratory, Tomtebodavägen 23, Solna
Subject area: Biotechnology
Doctoral student: Linda Kvastad , Genteknologi, Department of Gene Technology, KTH Royal Institute of Technology, Science for Life Laboratory, Solna, Sweden.
Opponent: Professor Lars Feuk, Dept. of Immunology, Genetics and Pathology, Uppsala University, Sweden.
Supervisor: Professor Joakim Lundeberg, Genteknologi, Science for Life Laboratory, SciLifeLab
In the present moment of time, we find ourselves in a period where the advancement of genomic tools is progressing at a fast pace. Of particular interest for this thesis is the study of gene activity. What patterns of genes are expressed? Where are they expressed? How can we use this knowledge to improve our quality of life? The research presented in this thesis focuses on developing and applying new tools for interrogating cells and tissues. In Paper I, we describe a protocol for transcript profiling of single cells, capable of measuring the relative expression levels for genes of interest. We successfully applied our method to cancer cells from metastatic breast cancer patients. Profiling 2 to 4 single cells per patient and measuring gene-specific expression from targets previously associated with metastatic breast cancer supports the use of our protocol as a diagnostic tool. In Paper II, we present an assay for spatial RNA quality evaluation, used to estimate the success for tissue specimens before proceeding with more expensive spatial sequencing methods. We showed that the method is capable of measuring high RNA quality in tissue areas of both high and low cell density and that the spatial RNA integrity patterns are reflected in spatial transcriptomics data. In Paper III, we present a protocol for performing spatial mRNA genome-wide expression profiling of FFPE tissue specimens. Thus, we bridge a gap between traditional tissue preservation methods and novel gene technologytools. We found a high Pearson correlation of 0.95 between formalin-fixation paraffin embedding (FFPE) and Fresh Frozen (FF) mouse brain datasets. Although the FPPE samples yielded fewer transcripts and genes compared to FF, there was a high agreement in gene expression between paired anatomical areas for FFPE and FF samples. In Paper IV, we present an approach to investigate in situ transcript derivedinferred copy number variation (iCNV) profiles based on spatial transcriptomics data. In a normal lymph node that displays both distinct gene expression patterns and histological landmarks, we observed a neutral iCNV profile. In contrast, we found huge variabilities investigating several malign tissue types ranging from homogenous (pediatric medulloblastoma) to highly variable genomes (ductal breast cancer and glioblastoma). Strikingly, we also observed similar iCNV profiles in both tumor and benign tissue areas from prostate and skin cancer. In Paper V, we explore the transcriptional and genomic landscape in pediatric tumors from 14 patients. Microglia cells have been implicated to play an important role in the tumor microenvironment, and we found spatial co-localization of microglia and epithelial-to-mesenchymal transition (EMT) signatures in our patient cohort. Furthermore, we found homogenous and recurrent iCNV profiles in the high-grade tumors of relapse patients and identified expression of gene SPP1 in the tumor stroma as a potential prognostic mRNA marker in pediatric brain tumor relapse patients.