Transferrin restrains rather than drives neuronal ferroptosis after intracerebral haemorrhage
- 1Department of Neurology, The Affiliated Hospital of Yangzhou University, Yangzhou University
- 2School of Traditional Chinese Medicine, Faculty of Medicine, Yangzhou University
- 3Laboratory of Intensive Care, The Affiliated Hospital of Yangzhou University, Yangzhou University
Abstract
Introduction: The transferrin‐transferrin receptor axis is the principal route by which neurons acquire iron, and it is generally regarded as a driver of ferroptosis: transferrin receptor 1 is used as a ferroptosis marker, and blocking transferrin‐bound iron uptake is protective in several models of iron‐dependent injury. That framework predicts that transferrin should be harmful after intracerebral haemorrhage (ICH), a condition defined by catastrophic iron overload. Clinical observations point the other way, since patients with higher circulating transferrin have smaller haematomas and better outcomes
Methods: We resolved this contradiction using bidirectional, neuron‐targeted manipulation of transferrin in a rat autologous‐blood model of ICH. Transferrin was silenced or overexpressed in neurons with adeno‐associated virus given 3 weeks beforehand, the same manipulations were applied to hemin-exposed differentiated PC‐12 neurons, transcriptomes were sequenced, and recombinant apo‐transferrin was given intraperitoneally 1 h after ICH at 4, 10 or 20 mg/kg.
Results: Endogenous transferrin increased two‐ to threefold in perihaematomal neurons and microglia, peaking on day 3, the same time point at which ferroptotic injury was maximal. Silencing neuronal transferrin did not protect; it increased perihaematomal iron, hydrogen peroxide, and malondialdehyde; shrank neuronal mitochondria further; and enlarged the lesion. Overexpression produced the mirror image, driving iron below sham levels and halving neurodegeneration. Inversion was reproduced in hemin‐exposed PC‐12 neurons. Transcriptionally, transferrin did not reinforce antioxidant defence. It abolished the need for it. Hemin moved 6,976 genes in wild‐type neurons but only 343 in transferrin‐overexpressing ones, a 95% collapse of the injury programme. Across the 6,974 hemin‐responsive genes, the effect of transferrin ran opposite to the effect of hemin (Pearson r = −0.869), reversing 99% of the genes that both perturbations moved significantly. The hemin‐induced oxidative stress programme, including the HO‐1/SLC7A11/TXNRD1 module, returned towards baseline. Recombinant apo‐transferrin given 1 h after ICH reproduced these effects with a sharp threshold at 20 mg/kg, halved lesion volume, and restored sensorimotor and cognitive function at 14 days.
Discussion: We propose that ICH inverts transferrin's function: confronted with heme‐derived iron that bypasses the transferrin cycle, transferrin is repurposed from an iron donor towards a high‐affinity extracellular iron sink. Its buffering capacity, rather than iron load itself, sets the threshold for neuronal ferroptosis. That capacity is exhaustible, saturating well below what the injured brain requires.
Related
- Predominantly mild thrombocytopenia, haematoma expansion and clinical outcomes in intracerebral haemorrhage
- C1qc Mediates Blood–Brain Barrier Disruption in a Mouse Model of Intracerebral Haemorrhage Through Microglia‐Derived Migrasomes
- Refining Clinical Management Advice From Radiology for Intracerebral Haemorrhage: A Two-Cycle Clinical Audit With Intervening Service Improvement.
- Blood pressure management in acute intracerebral haemorrhage: a Time–Trajectory–Phenotype framework
- Vagus Nerve Stimulation in Intracerebral and Subarachnoid Haemorrhage: A Systematic Review, Narrative Synthesis and Exploratory Meta-Analysis
BibTeX
@article{liu2026transferrin,
title = {{Transferrin restrains rather than drives neuronal ferroptosis after intracerebral haemorrhage}},
author = {Zhen Liu and Xuehan Zhuang and Bo Feng and Geyan Tian},
journal = {Frontiers in Pharmacology},
year = {2026},
month = oct,
doi = {10.3389/fphar.2026.1945504},
url = {https://www.frontiersin.org/articles/10.3389/fphar.2026.1945504/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Transferrin restrains rather than drives neuronal ferroptosis after intracerebral haemorrhage
AU - Zhen Liu
AU - Xuehan Zhuang
AU - Bo Feng
AU - Geyan Tian
PY - 2026
DA - 2026/10/28
JO - Frontiers in Pharmacology
DO - 10.3389/fphar.2026.1945504
UR - https://www.frontiersin.org/articles/10.3389/fphar.2026.1945504/full
AB - Introduction: The transferrin‐transferrin receptor axis is the principal route by which neurons acquire iron, and it is generally regarded as a driver of ferroptosis: transferrin receptor 1 is used as a ferroptosis marker, and blocking transferrin‐bound iron uptake is protective in several models of iron‐dependent injury. That framework predicts that transferrin should be harmful after intracerebral haemorrhage (ICH), a condition defined by catastrophic iron overload. Clinical observations point the other way, since patients with higher circulating transferrin have smaller haematomas and better outcomes Methods: We resolved this contradiction using bidirectional, neuron‐targeted manipulation of transferrin in a rat autologous‐blood model of ICH. Transferrin was silenced or overexpressed in neurons with adeno‐associated virus given 3 weeks beforehand, the same manipulations were applied to hemin-exposed differentiated PC‐12 neurons, transcriptomes were sequenced, and recombinant apo‐transferrin was given intraperitoneally 1 h after ICH at 4, 10 or 20 mg/kg. Results: Endogenous transferrin increased two‐ to threefold in perihaematomal neurons and microglia, peaking on day 3, the same time point at which ferroptotic injury was maximal. Silencing neuronal transferrin did not protect; it increased perihaematomal iron, hydrogen peroxide, and malondialdehyde; shrank neuronal mitochondria further; and enlarged the lesion. Overexpression produced the mirror image, driving iron below sham levels and halving neurodegeneration. Inversion was reproduced in hemin‐exposed PC‐12 neurons. Transcriptionally, transferrin did not reinforce antioxidant defence. It abolished the need for it. Hemin moved 6,976 genes in wild‐type neurons but only 343 in transferrin‐overexpressing ones, a 95% collapse of the injury programme. Across the 6,974 hemin‐responsive genes, the effect of transferrin ran opposite to the effect of hemin (Pearson r = −0.869), reversing 99% of the genes that both perturbations moved significantly. The hemin‐induced oxidative stress programme, including the HO‐1/SLC7A11/TXNRD1 module, returned towards baseline. Recombinant apo‐transferrin given 1 h after ICH reproduced these effects with a sharp threshold at 20 mg/kg, halved lesion volume, and restored sensorimotor and cognitive function at 14 days. Discussion: We propose that ICH inverts transferrin's function: confronted with heme‐derived iron that bypasses the transferrin cycle, transferrin is repurposed from an iron donor towards a high‐affinity extracellular iron sink. Its buffering capacity, rather than iron load itself, sets the threshold for neuronal ferroptosis. That capacity is exhaustible, saturating well below what the injured brain requires.
SN - 1663-9812
ER -