Back

Oxidative Stress, Neuroinflammation, and Neuronal Loss Start in Midlife: a 7 Tesla Magnetic Resonance Spectroscopy Healthy Adult Age-Span Study

Detcheverry, F. E.; Senthil, S.; Arnaoutelis, R.; Antel, S.; Assemlal, H.-E.; Karimaghaloo, Z.; Arnold, D. L.; Near, J.; Narayanan, S.; Badhwar, A.

2025-10-21 neurology
10.1101/2025.10.19.25338308 medRxiv
Show abstract

INTRODUCTIONWhile changes associated with age-related diseases, like oxidative stress, begin in midlife, most aging studies focused on older individuals. Our study assessed in vivo brain metabolites in healthy adults, including the understudied middle-age group. METHODS7 tesla magnetic resonance spectroscopy data were acquired from 95 healthy adults (48 women) aged 20-79 years. Eight metabolites were measured in posterior cingulate cortex (PCC) and centrum semiovale white matter (CSWM). RESULTSWith increasing age, we found (a) lower glutathione and glutamate, and higher myo-inositol in PCC, and (b) lower N-acetylaspartate and glutamate, and higher myo-inositol, total creatine, and N-acetylaspartyl-glutamate in CSWM. Notably, most changes started in midlife and were driven by age-related changes in women. DISCUSSIONOverall, we found evidence that oxidative stress, neuroinflammation, and neuronal vulnerability begin in midlife in healthy adults. Targeting these processes in midlife may slow brain aging and reduce age-related neurodegenerative diseases risk, including Alzheimers disease. GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=93 SRC="FIGDIR/small/25338308v2_ufig1.gif" ALT="Figure 1"> View larger version (35K): org.highwire.dtl.DTLVardef@938070org.highwire.dtl.DTLVardef@1d55614org.highwire.dtl.DTLVardef@1066a26org.highwire.dtl.DTLVardef@22ee8c_HPS_FORMAT_FIGEXP M_FIG C_FIG

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.