Development of spatial proteomics of cancer tissues using pixel-by-pixel laser capture microdissection (LCM) and MALDI mass spectrometry imaging (MALDI-MSI)

Abstract

Some diseases are notably challenging to treat due to their cellular heterogeneity. Proteins, as essential functional components of cells, are of significant interest in studying this diversity. A notable example of a disease characterized by this heterogeneity is triple-negative breast cancer, which is the focus of this study. The objective of this project is to explore, both spatially and functionally, tissue heterogeneity through the proteome of cells present in this tissue. This could allow detailed characterization and mapping of the cancer proteomic heterogeneity, subsequently identifying key regulators of biological processes involved in its progression and aggressiveness. Additionally, this exploration could aid in discovering new targets for drug development and identifying predictive signatures of treatment response. To achieve this, we plan to employ spatial proteomics through (1) MALDI mass spectrometry imaging, a method commonly used in the literature, as well as (2) pixel-by-pixel laser microdissection followed by rapid shotgun mass spectrometry proteomics and (3) single-cell proteomics technologies to determine the spatial distribution of proteins and their related biological functions. Utilizing the Matrix-Assisted Laser Desorption/Ionization (MALDI) imaging technique, we enzymatically digested proteins into peptides using trypsin and cleaved glycans with Peptide-N-Glycosidase F (PNGase) on tissue sections. Hierarchical clustering was employed to display the findings. This approach successfully mapped areas correlating with the tissue's histological characteristics, providing primarily spatial information at a molecular level. Subsequently, we developed an advanced method to gain additional functional insights. This novel approach, termed pixel-by-pixel spatial and functional proteomics, enables a comprehensive analysis of tissue heterogeneity. It achieves this with a dual emphasis on high spatial resolution, at a fine scale of 25μm, and a good proteome coverage (around 1000 proteins per pixel). This methodology allows for a detailed and nuanced exploration of the spatial distribution of proteins within tissues, offering insights into their functional aspects with remarkable precision. To optimize this method, it was crucial to precisely define the shape, size, and pixel sampling strategy to enhance resolution while ensuring high performance and efficiency. By comparing treated and untreated tissue using Perseus and the String Protein platform, we obtained functional information regarding the treatment's effect. Consequently, this second methodology yields more comprehensive data and functional information compared to the MALDI MSI approach. Nevertheless, conducting an unsupervised analysis encompassing all pixels could enable the comparison and integration of the two methods. Although the single- cell proteomic technique is not explored within the scope of this current study, its integration remains a crucial consideration for the broader objectives of the project.