Acute systemic inflammation has long been suspected to trigger harmful processes within the brain, contributing to neurodegenerative disorders such as Alzheimer's and Parkinson's disease. A new study published online in the Journal of Proteome Research on November 30, 2025, led by Professor Kei Zaitsu from the Faculty of Biology-Oriented Science and Technology, Kindai University, Japan, now provides compelling evidence that such inflammation causes region-specific metabolic disruption, identifying biochemical pathways that fail during the early stages of neuroinflammation.
Using a high-dose lipopolysaccharide (LPS) mouse model to induce acute systemic inflammation, the researchers analyzed metabolic changes in four brain regions—the cerebrum, hippocampus, cerebellum, and hypothalamus. After intraperitoneal LPS administration, serum Interleukin-1 beta (IL-1β) levels rose significantly, confirming a robust inflammatory response. Prof. Zaitsu explained, "We applied PiTMaP-based brain metabolomics to investigate metabolic changes in different brain regions of LPS-induced acute inflammation model mice."
The research team, including Mr. Shinnosuke Sugiura (Kindai University), Dr. Masaru Taniguchi (Nagoya City Public Health Research Institute), Dr. Kazuaki Hisatsune (Forensic Science Laboratory, Aichi Prefectural Police Headquarters), and Dr. Tomomi Asano (Kinjo Gakuin University), employed PiTMaP—a rapid metabolomics platform capable of analyzing intact metabolites directly from brain tissue without complex preparation steps. This allowed high-throughput profiling of more than 70 metabolites from each brain region. Principal component and discriminant analyses revealed a striking pattern: only the cerebrum showed significant metabolic disruption, while other regions remained largely unchanged.
Among the cerebrum-altered metabolites, N-acetylaspartic acid (NAA), a well-known marker for neuronal damage, was markedly reduced. NAA decline is associated with neuronal injury and impaired neuroenergetics. In this study, NAA levels showed a strong negative correlation with serum IL-1β, suggesting that systemic inflammation may directly contribute to neuronal stress. The researchers further observed a significant decrease in aspartic acid and malic acid, both integral to the malate–aspartate shuttle (MAS), a mitochondrial pathway essential for transferring cytosolic NADH into the mitochondria. Disruption of MAS has been previously linked to inflammation-induced mitochondrial dysfunction, and the current findings reinforce this mechanistic connection. Correlation analysis revealed tight links between these metabolites, implying impaired mitochondrial metabolic flow during inflammatory stress.
A second major finding was the accumulation of urea, suggesting activation or dysregulation of the urea cycle within cerebrum astrocytes. The urea cycle is increasingly recognized as a contributor to neurodegenerative pathology, with urea buildup reported in conditions such as Alzheimer's and Huntington's disease. In this study, urea and NAA emerged as "hub metabolites" (i.e., central nodes identified through network analysis) highlighting their importance in inflammation-induced metabolic disruption.
The combined metabolomic pattern—reduced NAA, decreased MAS intermediates, and elevated urea—supports the hypothesis that acute systemic inflammation can impair neuronal metabolism even before structural damage is detectable. These early biochemical shifts may provide a valuable window for identifying neuroinflammatory vulnerability. "These findings suggest that aspartic acid metabolism, MAS, and urea accumulation are closely associated with LPS-induced inflammation," concluded Prof. Zaitsu.
In the long term, this work paves the way for developing metabolic biomarkers that could help identify individuals at risk of neurodegenerative diseases driven or exacerbated by chronic inflammation. If validated in future human studies, NAA and urea levels, which are possibly measured peripherally, may help clinicians detect neuroinflammation before symptoms arise, enabling earlier intervention. Their integrated approach combining metabolomics, bioinformatics, and machine learning (including Random Forest classification achieving an area under the receiver operating characteristic curve of 1.0) underscores the power of advanced analytical tools in uncovering hidden metabolic signatures of disease.
This study deepens our understanding of how the brain responds to acute systemic inflammation, highlighting vulnerable metabolic pathways and offering new leads toward early detection strategies for inflammation-associated neurological disorders.
