Research projects in Cecilia Lindskog Bergströms group
Single cell mapping of reproductive organs
A large proportion of the genes with elevated expression in certain organs belong to male and female reproductive organs, and studies have shown that many of the corresponding proteins have an unknown function. Of particular interest is the testis, fallopian tube, ovary and endometrium. In testis, spermatogenesis is a complex interplay of thousands of proteins across multiple cellular states. Although the process where spermatogonial stem cells transform and mature haploid sperm cells has been extensively well-defined, still a substantial number of the proteins involved in the different stages have an unknown function and role. Similarly, there are numerous challenges associated with studying the ovary, including temporal changes in ovarian tissue, limited access to clinical samples and a small number of oocytes in each sample. To understand the underlying mechanisms of infertility, reduce the negative effects of menopause, and improve assisted reproductive technologies it is crucial to decode the complex ovarian aging processes and to create an expression map of ovarian cells over time and space. Several state-of-the-art technologies at the transcriptome and proteome level, including antibody-based multiplexed imaging are used to generate high-resolution spatial protein maps. The generated data will provide detailed insights at a single cell resolution, and possibilities to link spatial information with protein function and molecular mechanisms related to disease and various disorders, including infertility.
Global mRNA expression levels across 37 different normal tissues and organs has allowed for classification of all human genes into categories based on abundance and distribution. Genes with exclusive or elevated expression in certain organs provide interesting starting points for tissue-specific research in health and disease. Several ongoing projects focus on such genes, combined with strategies aiming at determining the cell type-specific location in the tissue using antibody-based techniques. The generated data provides novel insights on genes and proteins essential for tissue and cell type-specific functions unique to certain organs, and aids in assigning potential functions of previously uncharacterized proteins.
Machine learning for cancer diagnostics
The project's objective is to create an algorithm for detection of cancer cells in the prostate, that can aid pathologists in the diagnostic process. This would both lead to a quicker diagnosis, but also reduce issues related to interobserver variability. Immunofluorescence is used to mark specific structures in normal or cancer prostate tissue samples. The stained slides are scanned with a fluorescent scanner, then the tissues are stained with hematoxylin and eosin (H&E) and scanned again in brightfield. The immunofluorescence staining is overlaid with H&E and used as a mask to detect and outline cancer areas. The marked area serves as the ground truth for training of machine learning algorithms.
Cancer, biomarkers and tumor microenvironment
Several projects aim to further analyze the role of proteins identified as potential cancer biomarkers, either from a diagnostic or prognostic perspective. Tumor material from well-defined patient cohorts with tumors representing several major forms of human cancer are collected and assembled into tissue microarrays. In addition to tumor material, clinical data is also collected to create databases that allows for testing and validation of protein expression patterns of importance for diagnostics, prognosis and functional tumor biology studies. Special emphasis is put on ovarian cancer and lung cancer, but our research group also collaborates on projects related to several other cancer types.
Lung cancers demonstrate high expression of so-called cancer-testis antigens (often abbreviated as CTAs). Studies have established that CTAs are immunogenic and can induce significant immune responses. Under normal conditions, CTAs are mainly restricted to testis and the immune response is suppressed since testicular cells are devoid of MHC molecules. Due to these characteristics, CTAs may potentially serve as treatment options for immunotherapy or cancer vaccines. Based on our knowledge from studying the normal physiological aspects of testis biology and our unique access to antibodies against testis proteins we have set up a stringent platform for mapping CTA targets in lung cancer.
For highly heterogeneous tumors, such as those associated with high-grade serous ovarian carcinomas (HGSOC), mapping at a single cell resolution for both mRNA and proteins is crucial to capture the diversity. To study the HGSOC tumor ecosystem we combine multi-omics high-resolution techniques to discover novel gene and protein signatures specific for certain cell types and subsets thereof, including the tumor microenvironment. Further understanding of the molecular diversity in HGSOC tumors will aid in discovery of potential therapeutic targets as well as the discovery of biomarkers for diagnosis and stratification of patients.
Antibody validation, performance and characterization
The objective of these technical development projects is to develop assays and strategies for further validation of antibodies in tissues and cells. Strategies include refinement of guidelines for antibody validation using standardized immunohistochemistry, but also further development of new methods, including multiplexed immunofluorescence, proximity ligation assay and RNAScope. The techniques used have different advantages and disadvantages regarding sensitivity, specificity, resolution, and implementation in a high-throughput workflow.