Mitochondrial capacity limits neuronal energetic feasibility under chronic behavioural stress associated with AI-intensive environments: a neuron–astrocyte genome-scale model of human grey matter

Abstract Sustained cognitive engagement, short and mistimed sleep, chronic stress and reduced recovery co-occur in AI-intensive environments, yet how this profile acts on brain energy metabolism is unknown. We built a neuron–astrocyte genome-scale model of human grey matter from Human-GEM v2.1.1, contextualised each cell with human single-nucleus expression, made transport, sodium-pumping and glutathione-recycling costs explicit, and calibrated one parameter to measured cerebral glucose use. Without further fitting the model reproduced the astrocytic share of oxidative metabolism measured by ¹³C spectroscopy (15.7% vs 14%), both the oxidative and the glycolytic regimes of human brain activation, and an independent reconstruction; deletion of genes causing inherited brain-energy disorders compromised its energetic margin far more often than deletion of other genes (odds ratio 79, p = 6 × 10⁻²⁰). Translating the five stressors into constraints, sleep loss and cognitive load each exhausted the margin within the protocol range, at a 25% loss of mitochondrial capacity and a 39% rise in signalling demand, whereas circadian misalignment, chronic stress and reduced recovery acted on arms with reserve. Combined, the stressors exhausted the margin at 0.72 times the moderate profile; across 400 parameter draws demand was met in 99.2% at low and 28.5% at moderate exposure. Substrate oxidation, not substrate supply, set the margin: restoring sleep or reducing load restored it, whereas ketone bodies, lactate or fibre-derived acetate helped only when glucose delivery itself was limiting. Results held across thresholds, releases, solvers and enzyme-capacity limits, and the thresholds are testable with ³¹P and ¹³C spectroscopy.

Authors

Institutions

Publication Details

Journal
Research Square
Published
2026-11-16
DOI
https://doi.org/10.21203/rs.3.rs-11551364/v1
Primary Topic
Mitochondrial Function and Pathology
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Mitochondrial capacity limits neuronal energetic feasibility under chronic behavioural stress associated with AI-intensive environments: a neuron–astrocyte genome-scale model of human grey matter

Jofrey Raymond, Christina Charles, Musa Chacha, Shangwe Leonard Kigoda
Research Square
Mitochondrial Function and Pathology
preprint

Mitochondrial capacity limits neuronal energetic feasibility under chronic behavioural stress associated with AI-intensive environments: a neuron–astrocyte genome-scale model of human grey matter

Jofrey Raymond, Christina Charles, Musa Chacha, Shangwe Leonard Kigoda
preprint en

Abstract

Abstract Sustained cognitive engagement, short and mistimed sleep, chronic stress and reduced recovery co-occur in AI-intensive environments, yet how this profile acts on brain energy metabolism is unknown. We built a neuron–astrocyte genome-scale model of human grey matter from Human-GEM v2.1.1, contextualised each cell with human single-nucleus expression, made transport, sodium-pumping and glutathione-recycling costs explicit, and calibrated one parameter to measured cerebral glucose use. Without further fitting the model reproduced the astrocytic share of oxidative metabolism measured by ¹³C spectroscopy (15.7% vs 14%), both the oxidative and the glycolytic regimes of human brain activation, and an independent reconstruction; deletion of genes causing inherited brain-energy disorders compromised its energetic margin far more often than deletion of other genes (odds ratio 79, p = 6 × 10⁻²⁰). Translating the five stressors into constraints, sleep loss and cognitive load each exhausted the margin within the protocol range, at a 25% loss of mitochondrial capacity and a 39% rise in signalling demand, whereas circadian misalignment, chronic stress and reduced recovery acted on arms with reserve. Combined, the stressors exhausted the margin at 0.72 times the moderate profile; across 400 parameter draws demand was met in 99.2% at low and 28.5% at moderate exposure. Substrate oxidation, not substrate supply, set the margin: restoring sleep or reducing load restored it, whereas ketone bodies, lactate or fibre-derived acetate helped only when glucose delivery itself was limiting. Results held across thresholds, releases, solvers and enzyme-capacity limits, and the thresholds are testable with ³¹P and ¹³C spectroscopy.

Research Square
Nelson Mandela African Institution of Science and Technology (TZ)
Good health and well-being
Mitochondrial Function and Pathology
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon