High-dimensionality reduction clustering of complex carbohydrates to study lung cancer metabolic heterogeneity

Abstract

The tumor microenvironment contains a heterogeneous population of stromal and cancer cells that engage in metabolic crosstalk to ultimately promote tumor growth and contribute to progression. Due to heterogeneity within solid tumors, pooled mass spectrometry workflows are less sensitive at delineating unique metabolic perturbations between stromal and immune cell populations. Two critical, but understudied, facets of glucose metabolism are anabolic pathways for glycogen and N-linked glycan biosynthesis. Together, these complex carbohydrates modulate bioenergetics and protein-structure function, and create functional microanatomy in distinct cell populations within the tumor heterogeneity. Herein, we combine high-dimensionality reduction and clustering (HDRC) analysis with matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI) and demonstrate its ability for the comprehensive assessment of tissue histopathology and metabolic heterogeneity in human FFPE sections. In human lung adenocarcinoma (LUAD) tumor tissues, HDRC accurately clusters distinct regions and cell populations within the tumor microenvironment, including tumor cells, tumor-infiltrating lymphocytes, cancer-associated fibroblasts, and necrotic regions. In-depth pathway enrichment analyses revealed unique metabolic pathways are associated with each distinct pathological region. Further, we highlight the potential of HDRC analysis to study complex carbohydrate metabolism in a case study of lung cancer disparity. Collectively, our results demonstrate the promising potentials of HDRC of pixel-based carbohydrate analysis to study cell-type and regional-specific stromal signaling within the tumor microenvironment.

Keywords: Clustering analysis; Glycogen; Lung cancer; MALDI imaging; N-linked glycans.