Broadcasters, receivers, functional groups of metabolites and the link to heart failure progression using polygenic factors

Azam Yazdani, Raúl Méndez-Giráldez, Akram Yazdani, Daniel J. Schaid, Sek Won Kong, M. Wahdian Rijalil Hadi, Ahmad Samiei, Clemens Wittenbecher, Jessica Lasky‐Su, Clary B. Clish, Francesco Marotta, Michael R. Kosorok, Samia Mora, Jochen D. Muehlschlegel, Daniel I. Chasman, Martin G. Larson, Sarah H. ElseaResearch Square 2023 · DOI

TL;DR

This study uses a systems biology approach to map the complex network of metabolites in the human body and their connections to heart failure (HF) risk. By building data-driven and causal networks, researchers identified specific metabolites that act as "broadcasters" or "receivers" and found that certain amino acids, like glycine and asparagine, are inversely associated with HF risk, potentially linking diet and genetics to cardiovascular health.

What Problem Does This Paper Address?

Understanding heart failure (HF) progression is a major challenge in medicine. While we know that genetics, lifestyle, and environmental factors play a role, the precise molecular mechanisms linking these factors to HF are not fully understood. Traditional association studies often look at one or a few metabolites at a time, which can miss crucial interactions and lead to confounding results because they don't account for the complex web of metabolic pathways. This paper aims to address this by using a systems-level approach to map the "metabolomic architecture" and identify how metabolites interact with each other and with HF risk, while also accounting for genetic influences.

Methods

The researchers employed a sophisticated systems biology approach, combining data-driven and causal network analyses.

  • Data-driven Networks: They used partial correlation to build networks where connections represent relationships between metabolites that are not explained by other metabolites in the network. This was done separately for amino acids and amines (AAA), bile acids, organic acids, nucleotides, sugars (BONS), and lipids.
  • Causal Networks: To infer causality, they augmented these networks with polygenic factors derived from genetic variation data, leveraging principles of Mendelian randomization (MR). This helps to identify the direction of influence between metabolites and their association with HF.
  • Confounding Metabolite Identification: A key innovation is using the network structure to identify and control for confounding metabolites – those that influence both a metabolite of interest and HF risk. This allows for a more robust identification of metabolites directly associated with HF.
  • Cox Proportional Hazards Models: These statistical models were used to assess the association of individual metabolites with future HF incidence, while controlling for confounding metabolites and other covariates like age, sex, and BMI.

Datasets

The study utilized data from the Framingham Heart Study (FHS) offspring cohort.

  • Participants: 2,526 individuals free of overt HF at baseline.
  • Metabolite Data: Plasma samples were analyzed using three platforms, measuring 42 amino acids and amines (AAA), 54 BONS, and 104 lipids.
  • Genetic Data: Genotyping data for over 550,000 single nucleotide polymorphisms (SNPs) were available.
  • Outcome Data: Participants were followed for incident HF events over a period of 20-25 years.

Key Insights and Ideas

  • Metabolite "Personalities": The study identified metabolites acting as "broadcasters" (influencing many others) and "receivers" (being influenced by many others), revealing a functional hierarchy within metabolic pathways.
  • Diet and Genetics Link to HF: Certain essential amino acids, like glycine and asparagine, were found to be inversely associated with HF risk. Since these are diet-derived, this highlights a direct link between dietary intake, genetic predisposition (via polygenic factors), and cardiovascular health.
  • Network Structure Matters: The research strongly emphasizes that understanding the interconnectedness of metabolites is crucial for accurate association studies and for uncovering mechanistic insights into disease progression. Ignoring these connections can lead to confounding and missed signals.
  • Beyond Single-Metabolite Associations: The approach moves beyond simply identifying metabolites associated with HF to understanding how they are associated, by considering their role within the broader metabolic network and their potential to confound other associations.

Key Results

  • The study successfully constructed data-driven and causal networks for AAA, BONS, and lipid metabolites.
  • Several metabolites were identified as significantly associated with HF risk after controlling for confounders.
  • Inverse Association (Lower HF Risk): Glycine, asparagine, and serotonin were associated with a decrease in HF risk. Asparagine was directly influenced by threonine and directly affected glycine.
  • Increased HF Risk: Isoleucine, ureidopropionic acid, glycocholic acid, inosine, 2-hydroxybutyrate, isocitric acid, and hexose metabolites were associated with increased HF risk.
  • Lipids: LPC 18:2 showed a negative effect on HF risk.
  • The study identified specific "broadcaster" and "receiver" metabolites within the AAA and BONS networks, and subnetworks within the lipid network.

Limitations and Future Work

  • Platform-Specific Analysis: Metabolite-HF associations were analyzed separately for each platform (AAA, BONS, lipids). Future work could integrate these analyses to account for cross-platform confounding.
  • Polygenic Factors for BONS/Lipids: No suitable polygenic factors were identified for the BONS and lipid networks to establish causal directions using MR, limiting causal inference in these areas.
  • Future Directions: The authors suggest that the prioritized list of HF-associated metabolites identified in this study can guide the design of more focused experimental studies to further elucidate the pathways involved in HF progression. They also highlight the potential for this systems approach to be applied to other complex diseases.

Summary updated

Abstract

In a prospective study with records of heart failure (HF) incidence, we present metabolite profiling data from individuals without HF at baseline. We uncovered the interconnectivity of metabolites using data-driven and causal networks augmented with polygenic factors. Exploring the networks, we identified metabolite broadcasters, receivers, mediators, and subnetworks corresponding to functional classes of metabolites, and provided insights into the link between metabolomic architecture and regulation in health. We incorporated the network structure into the identification of metabolites associated with HF to control the effect of confounding metabolites. We identified metabolites associated with higher or lower risk of HF incidence, the associations that were not confounded by the other metabolites, such as glycine, ureidopropionic and glycocholic acids, and LPC 18:2. We revealed the underlying relationships of the findings. For example, asparagine directly influenced glycine, and both were inversely associated with HF. These two metabolites were influenced by polygenic factors and only essential amino acids which are not synthesized in the human body and come directly from the diet. Metabolites may play a critical role in linking genetic background and lifestyle factors to HF progression. Revealing the underlying connectivity of metabolites associated with HF strengthens the findings and facilitates a mechanistic understanding of HF progression.
BibTeX
@inproceedings{yazdani2023broadcasters,
  author = {Azam Yazdani and Raúl Méndez-Giráldez and Akram Yazdani and Daniel J. Schaid and Sek Won Kong and M. Wahdian Rijalil Hadi and Ahmad Samiei and Clemens Wittenbecher and Jessica Lasky‐Su and Clary B. Clish and Francesco Marotta and Michael R. Kosorok and Samia Mora and Jochen D. Muehlschlegel and Daniel I. Chasman and Martin G. Larson and Sarah H. Elsea},
  title = {{Broadcasters, receivers, functional groups of metabolites and the link to heart failure progression using polygenic factors}},
  booktitle = {Research Square},
  year = {2023},
  doi = {10.21203/rs.3.rs-3246406/v1},
}
Figure