UHPLC-MS Reveals Metabolic Signatures in Epilepsy

Scientists are increasingly finding that problems with how the body handles energy and metabolism might be tied to epilepsy, but exactly how the body's chemistry changes during long-term, repeated seizures still is not fully understood. Researchers at the China-Japan Union Hospital of Jilin University (China) set out to figure out what these body-wide chemical changes look like in mice that had been given repeated small doses of the drug pentylenetetrazol (PTZ) to trigger seizures over time, mimicking chronic epilepsy. They also wanted to see whether the patterns they found matched up with data from human epilepsy studies. To do this, they analyzed blood plasma samples using ultrahigh-performance liquid chromatography-mass spectrometry (UHPLC-MS)-based untargeted metabolomics to detect and measure a wide range of small molecules in the body. Their findings were published in the journal Frontiers in Neuroscience.¹

Why Is Epilepsy Still Hard to Treat?

Epilepsy is a common brain condition where people have repeated seizures that happen on their own, caused by unusual electrical activity in the brain.²Even though there are many anti-seizure medications on the market, about one in three patients keep having seizures that do not respond to treatment. This suggests that scientists still do not fully understand the biological processes behind how epilepsy develops in the first place.³

Why Are Researchers Now Looking at Metabolism as a Possible Piece of the Puzzle?

In recent years, scientists have started paying more attention to the idea that problems with the body's metabolism, how it produces and uses energy, may play an important role in epilepsy. Since brain cells rely on a steady, carefully controlled energy supply to work properly, any disruption to that balance could affect how easily brain cells fire, how they communicate with each other, and how much inflammation occurs in the brain, all of which can contribute to seizures.^2,3

What Metabolic Changes Show Up in the Blood of Seizure-prone Mice, and Do They Mirror Patterns Seen in Human Epilepsy Brain Tissue?

For this study, male mice were given repeated injections of PTZ to develop chronic epilepsy, while a comparison group of mice received saltwater injections instead. The researchers then compared the blood chemistry of the two groups to figure out which metabolites were different between them, using a couple of different statistical approaches. They also used Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment to see which biological processes and pathways these different molecules were linked to. In addition, they used least absolute shrinkage and selection operator (LASSO) regression as a way of narrowing down a large list of possible markers to the ones that matter most, as an exploratory step.¹

To help connect their mouse findings to humans, the researchers also looked at existing human data (specifically, hippocampal tissue from the public dataset GSE256068) and re-analyzed it to see the activity of the genes analyzed, then ran the same pathway analysis to see if similar biological processes occurred.¹

The mice with drug-induced chronic seizures showed a noticeably different chemical fingerprint in their blood compared to the healthy control mice. In total, the researchers spotted 349 molecules that differed between the two groups, and most of them were tied to amino acid processing, tryptophan, and the body's core energy-processing pathways.¹

Using LASSO regression, the researchers zeroed in on six molecules that stood out as the most promising candidates, all of which are listed in the paper's supplementary materials. Of those six, two were singled out for a closer look: one called indolelactic acid (ILA, which is related to tryptophan) and a fat molecule called DG (20:3n6/0:0/20:3n6), because they seemed most relevant to understanding the tryptophan and fat-related changes seen in the seizure mice.¹

When the researchers re-examined human brain tissue data, they found changes in some of the same general categories, things related to brain-signaling chemicals (specifically the GABA system), fat-related molecule production, inflammation-linked pathways in the brain, and general inflammatory activity. This gave them a helpful human-based comparison point, supporting the idea that the neurotransmitter, fat, and inflammation-related changes seen in the seizure mice may also be relevant in human epilepsy.¹

“PTZ-induced kindling,” write the authors of the paper,¹ “was associated with marked plasma metabolic alterations involving amino acid-, tryptophan-, and lipid-related pathways. ILA and DG (20:3n6/0:0/20:3n6) emerged as candidate plasma metabolic features associated with the PTZ-kindled seizure phenotype. Complementary analysis of human temporal lobe epilepsy with hippocampal sclerosis (TLE-HS) hippocampal transcriptomic data identified related alterations in neurotransmission-, lipid-, and inflammatory-associated pathways. This pathway-level convergence supports the broader relevance of metabolic dysregulation to epilepsy, while the mechanistic roles and biomarker potential of individual metabolites require independent validation.”

References

1. Ren, J.; Wang, F.; Li, Z. et al. Amino acid- and Lipid-related Metabolic Remodeling in PTZ-kindled Mice Reveals Candidate Plasma Signatures of Chronic Epilepsy. Front Neurosci. 2026, 20, 1889410. DOI: 10.3389/fnins.2026.1889410
2. Krishnamurthy, K. B. Epilepsy. Ann Intern Med. 2025, 178 (4), ITC49-ITC64. DOI: 10.7326/ANNALS-25-00494
3. Löscher, W.; Klein, P. The Pharmacology and Clinical Efficacy of Antiseizure Medications: From Bromide Salts to Cenobamate and Beyond. CNS Drugs2021, 35 (9), 935-963. DOI: 10.1007/s40263-021-00827-8