Peripheral blood metabolome predicts mood change-related activity in mouse model of bipolar disorder

Hideo Hagihara, Tomoyasu Horikawa, Yasuhiro Irino, Hironori Nakamura, Juzoh Umemori, Hirotaka Shoji, Masaru Yoshida, Yukiyasu Kamitani, Tsuyoshi MiyakawaMolecular 2019 · DOI

TL;DR

This study demonstrates that patterns of metabolites in peripheral blood can retrospectively predict mood-related activity levels in a mouse model of bipolar disorder. This finding suggests a potential for non-invasive, blood-based biomarkers to track mood states and offers insights into the brain-periphery axis's role in mood regulation.

What Problem Does This Paper Address?

Bipolar disorder is characterized by extreme mood swings and associated changes in activity levels. Current diagnosis relies heavily on subjective patient reports, which can be biased. There's a significant need for objective, quantitative measures to track mood states, especially in preclinical models that can inform human studies. While previous research has explored brain-based markers, the authors aim to investigate if peripheral blood metabolomics can serve as a more accessible proxy for these mood-related changes. They leverage a mouse model exhibiting infradian (longer than 24-hour) cycles of locomotor activity, mimicking aspects of bipolar disorder, to explore this connection.

Methods

The researchers employed a multi-pronged approach:

  • Locomotor Activity Monitoring: They continuously monitored the locomotor activity (LA) of mice in their home cages over extended periods.
  • Peripheral Blood Sampling: Blood was collected from the mice at specific time points.
  • Metabolomic Analysis: Gas chromatography-mass spectrometry (GC/MS) was used to analyze the metabolite profiles in the collected blood samples.
  • Statistical Modeling:
    • Linear Regression: Cross-validated linear regression models were constructed to predict LA levels based on the identified blood metabolome patterns.
  • Nested Cross-Validation: This technique was used for both feature selection (identifying which metabolites were most predictive) and model fitting, ensuring robust generalization to unseen data.
  • Leave-One-Out Cross-Validation: Used to evaluate prediction accuracy.
  • Pathway Enrichment Analysis: To understand the biological significance of the predictive metabolites, they used MetaboAnalyst to identify enriched metabolic pathways.
  • Gene Expression Analysis: They also investigated if peripheral blood metabolome patterns could predict the expression of specific genes (circadian-related genes, including Arntl) in the hippocampus, a brain region crucial for mood regulation.

Datasets

The study utilized data from 35 Camk2a+/- mutant mice. The locomotor activity data was previously collected and analyzed in a prior study by the same research group. Hippocampal gene expression microarray data from the same mice was also used, which had been deposited in the Gene Expression Omnibus (GEO) under accession number GSE68293. No new benchmarks were introduced; rather, existing data was re-analyzed with a focus on peripheral metabolomics.

Key Insights and Ideas

  • Peripheral Blood as a Mood Mirror: The most striking finding is that the complex chemical fingerprint of peripheral blood can, in fact, reflect and predict an animal's mood-related activity state, even for activity that occurred days prior.
  • Mitochondrial Metabolism Underpins Mood: The predictive metabolites are heavily enriched in pathways related to mitochondrial function, specifically the "Warburg effect" and the "citric acid cycle." This suggests that disruptions in cellular energy production and metabolism might be a core mechanism linking peripheral changes to mood states.
  • Brain-Periphery Axis Connection: The study provides evidence that peripheral metabolic signals can predict gene expression in the brain (hippocampus), specifically for genes like Arntl implicated in bipolar disorder. This highlights a potential communication pathway between the body and the brain in the context of mood regulation.
  • Temporal Dynamics of Prediction: The ability to predict LA varied depending on the time window before sampling, suggesting that different metabolic signatures might be relevant for different temporal scales of mood changes.

Key Results

  • Predictive Power: Linear regression models successfully predicted 24-hour locomotor activity (24h LA) from peripheral blood metabolome patterns with a significant correlation (r = 0.499, P = 0.0023). Thirty-three out of 106 metabolites were selected for the predictive model.
  • Temporal Prediction: Peripheral blood metabolome patterns could also predict 3-hour locomotor activity (3h LA) up to 3 days prior to sampling, with significant correlations found in five specific time windows (e.g., 63-66 hours before sampling).
  • Gene Expression Prediction: Peripheral blood metabolome patterns predicted the expression of the Arntl gene in the hippocampus with a strong correlation (r = 0.633, P = 4.51 x 10⁻⁵).
  • Pathway Enrichment: Key enriched pathways for predictive metabolites included the Warburg Effect, Citric Acid Cycle, and Transfer of Acetyl Groups into Mitochondria.

Limitations and Future Work

  • Confounding Factors: The authors acknowledge that physical activity itself, independent of mood state, could influence peripheral metabolome patterns. Intense exercise, for instance, can affect mitochondrial pathways.
  • Causality: While the study shows correlations, it doesn't definitively establish causality between the observed metabolic alterations and mood changes.
  • Translational Potential: The study is conducted in a mouse model, and further research is needed to validate these findings in human subjects.
  • Future Directions: The authors suggest exploring the most appropriate sample types (blood, saliva, urine, skin), processing methods (metabolomics, proteomics, transcriptomics), and optimizing prediction algorithms for human applications. They also emphasize the need to determine the causal relationship between peripheral metabolic alterations and mood states.

Summary updated

Abstract

Bipolar disorder is a major mental illness characterized by severe swings in mood and activity levels which occur with variable amplitude and frequency. Attempts have been made to identify mood states and biological features associated with mood changes to compensate for current clinical diagnosis, which is mainly based on patients' subjective reports. Here, we used infradian (a cycle > 24 h) cyclic locomotor activity in a mouse model useful for the study of bipolar disorder as a proxy for mood changes. We show that metabolome patterns in peripheral blood could retrospectively predict the locomotor activity levels. We longitudinally monitored locomotor activity in the home cage, and subsequently collected peripheral blood and performed metabolomic analyses. We then constructed cross-validated linear regression models based on blood metabolome patterns to predict locomotor activity levels of individual mice. Our analysis revealed a significant correlation between actual and predicted activity levels, indicative of successful predictions. Pathway analysis of metabolites used for successful predictions showed enrichment in mitochondria metabolism-related terms, such as "Warburg effect" and "citric acid cycle." In addition, we found that peripheral blood metabolome patterns predicted expression levels of genes implicated in bipolar disorder in the hippocampus, a brain region responsible for mood regulation, suggesting that the brain-periphery axis is related to mood-change-associated behaviors. Our results may serve as a basis for predicting individual mood states through blood metabolomics in bipolar disorder and other mood disorders and may provide potential insight into systemic metabolic activity in relation to mood changes.
BibTeX
@inproceedings{hagihara2019peripheral,
  author = {Hideo Hagihara and Tomoyasu Horikawa and Yasuhiro Irino and Hironori Nakamura and Juzoh Umemori and Hirotaka Shoji and Masaru Yoshida and Yukiyasu Kamitani and Tsuyoshi Miyakawa},
  title = {{Peripheral blood metabolome predicts mood change-related activity in mouse model of bipolar disorder}},
  booktitle = {Molecular},
  year = {2019},
  doi = {10.1186/s13041-019-0527-3},
}
Figure