Metabolic network analysis and its application in understanding the biology of aging

Abstract

A fundamental aim of life science is to understand the functioning of an organism and draw system-level connections between its genotype and behavior. Among the most significant types of interaction networks, cellular metabolism is a well-established descriptor of the phenotype of an organism. It relies on thousands of enzymatic reactions that are often represented as a dense and enmeshed web of biochemical conversions. Inspecting the metabolic network in its entirety is crucial in attaining a holistic understanding of the underlying biological mechanism. Colossal research efforts in the post-genomic era have enabled the curation of metabolic networks of entire organisms. Concurrently, advancements in computational strategies and algorithms have led to the inception of countless tools that utilize the essential information in genome-scale metabolic models to attain valuable insights into the physiology of an organism. Constraint-based modeling is a widely used and tested technique to model the metabolic network. Flux balance analysis (FBA) is the most popular constraint-based approach to predict intracellular metabolic flux distributions and network capabilities in genome-scale models. They are modeled on two fundamental assumptions. Firstly, that the cellular metabolism operates at a homeostatic condition, and secondly, that the cell typically organizes its metabolism to optimize a particular cellular objective. However, these assumptions significantly impair the applicability of the traditional flux balance analysis. Where the prediction accuracy is insufficient, additional constraints from omics data sources assist in obtaining biologically relevant inference. In this work, we showcase the addition of transcriptomic data to gain a profound understanding of the early events in the progression of Alzheimer’s disease model Caenorhabditis elegans. Transgenically expressing human amyloid-beta recapitulated the phenotypic disease response in the worms. Our analyses of the contextualized genome-scale metabolic network curated with the integration of experimentally derived gene expression data implicated metabolic alterations in Tricarboxylic Acid (TCA) cycle following low-level amyloid-beta expression. Along with metabolomic and enzymatic assays, we show repression in alpha-Ketoglutarate dehydrogenase. Identification of metabolic dysfunction as an early event is paramount in formulating mitigating efforts. Formulating a metabolic objective for use in constraint-based modeling is hazy, particularly for complex multicellular organisms. In addressing such concerns, we developed a computational algorithm, ∆FBA (delta–FBA), that focuses on the differences in the metabolic distribution between a pair of conditions. By formulating the mathematical problem to optimize for maximal consistency between the inferred flux alterations and integrated gene expression changes, ∆FBA predicts metabolic rewiring as an effect of genetic or environmental perturbations. We validated our strategy in a wide range of single-gene deletion knockouts and environmental modifications in Escherichia coli, where ∆FBA outperforms similar methods. Furthermore, our findings of metabolic changes in human diabetic subjects show the robustness of ∆FBA. Akin to metabolism, aging is a complex biological process that necessitates a system-level analysis in unraveling its etiology and progression. Despite numerous efforts, the aging process in humans is far from being completely understood. The Genotype-Tissue Expression (GTEx) project represents an invaluable repository of information that is ideal for studying human aging by examining the differences in the gene expression of over seven hundred individuals. Our bioinformatics and metabolic network analysis of the transcriptome associated human aging with conserved hallmarks of aging. Additionally, we show that the temporal changes in gene expression significantly contribute to the aging process. Our findings suggest that persistent moderators of cell fate and early repressors of cellular energetics could play a pivotal role in the progression of the aging process, one that culminates in accelerated decline.