Generated editorial illustration of the transplantation and hippocampal mechanism. It is not study microscopy and does not depict a human treatment. Credit: CellXperience.
The donor cells matter most while they are present.
A bioRxiv preprint posted October 21 uses timed depletion to probe how young bone-marrow-derived microglia-like cells affect an aged mouse brain. The central observation is narrower than permanent cell replacement. Donor-derived macrophages were abundant in the hippocampus one month after marrow transplantation, and removing them weakened dendritic-spine measures and hippocampal behavior. Three months later, the same cell population was scarce in the hippocampus, and partial depletion did not erase the earlier memory gains.
The authors connect that early window to BDNF-TrkB signaling. The model is mechanistically useful and highly artificial: old female mice were irradiated before marrow transplantation, behavioral groups were small, and the depletion system left dead labeled cells in a brain that struggled to clear them.
The experiment begins with whole-body conditioning
Recipient C57BL/6J mice were female and 18 to 20 months old. They received 10-Gy irradiation, followed by two million Sca-1-positive marrow cells from young or old donors through the tail vein. The resulting chimeras allowed donor-derived macrophages to be distinguished in the brain.
This is a marrow-reconstitution model, not a direct injection of purified microglia into the hippocampus. Sca-1 enrichment includes primitive marrow populations rather than one committed macrophage precursor. Irradiation opens a path for peripheral cells and changes the immune environment throughout the body.
Those features are central to interpretation. The study asks what donor-derived brain macrophages do after a forceful reset of the hematopoietic system. It does not show that young cells would enter an intact aged brain after a simple infusion.
For selective depletion, donor marrow came from mice carrying a tamoxifen-inducible CX3CR1-linked diphtheria-toxin-receptor system. Tamoxifen marked the donor macrophage lineage, and diphtheria toxin was given later to remove labeled cells. The investigators triggered depletion either one or three months after transplantation, then examined behavior, immune repopulation, and brain tissue.
One month exposes the hippocampal contribution
At the one-month time point, donor-derived microglia-like cells were abundant in the hippocampus. The toxin regimen removed about 95 percent of the labeled population in the experiment highlighted by the authors.
Depletion reduced the number of dendritic protrusions on hippocampal pyramidal neurons. Performance also fell in novel-object recognition and Barnes-maze measures of spatial learning and memory. As the macrophage niche partly recovered, spine counts and some behavioral measures moved back toward baseline.
This gain-and-loss design provides stronger mechanistic leverage than a transplantation comparison alone. The early cognitive phenotype depended on the continued presence of the donor-derived cell population under the conditions tested.
The experiment still does not prove that each depleted macrophage was a mature, resident-like microglial cell. CX3CR1 is shared across several mononuclear-phagocyte populations, and marrow-derived cells in an irradiated brain can differ from embryonically established microglia. “Microglia-like” is the appropriate label.
BDNF-TrkB is the leading mechanism, not the final one
The team screened hippocampal gene and protein signals across the transplantation and depletion conditions. BDNF emerged as a candidate associated with the young donor-derived cells. BDNF protein was detected in isolated donor microglia-like cells, and its hippocampal signal was stronger while those cells were present.
Investigators then gave ANA-12, a TrkB antagonist, for five days. Blocking the BDNF receptor reduced dendritic-spine measures and impaired novel-object recognition and Barnes-maze performance in mice carrying young donor marrow. The combination resembles the effect of depleting the donor-derived macrophages.
The pharmacology supports a BDNF-TrkB route, though it does not identify the donor cells as the only relevant BDNF source. ANA-12 acts on TrkB signaling across the brain. A cell-specific BDNF knockout in the donor lineage would test necessity more directly, while controlled BDNF rescue after depletion would test sufficiency.
The study also did not measure synaptic physiology. Spine counts and behavioral assays suggest altered plasticity without showing that the newly maintained protrusions form stable, functional synapses.
Three months reveals a messy depletion model
At three months, donor-derived cells were less prominent in the hippocampus. Partial depletion reduced total labeled macrophages by roughly 72 percent, with different efficiencies across F4/80-high and F4/80-low populations. The hippocampal behavioral improvement did not reverse, while grip and motor measures declined.
The depleted niche then failed to repopulate in the usual way. Recruitment of circulating monocytes was markedly reduced, and neutrophils entered the brain. Labeled cells with damaged membranes remained visible weeks after toxin treatment, suggesting slow removal of dead cells in the aged animals.
This complicates a simple “cells removed, phenotype lost” interpretation. At the later time point, toxin exposure changed the niche, left debris, and altered the incoming immune population. The absence of a memory decline may mean the early donor cells produced a lasting effect. It may also reflect incomplete depletion, regional scarcity before toxin, or a behavioral assay that could not resolve smaller changes.
The authors’ transient-homing model is plausible because donor cells were present where the one-month phenotype was measured and largely absent from the hippocampus by three months. “Permanent modification” is too strong for the current data. The persistence of a behavioral score does not establish permanent neuronal rewiring.
The translational distance is long
Most experimental groups contained three to six animals, with behavioral comparisons commonly using six and imaging or molecular assays using fewer. Only aged female recipients were studied. The preprint has not completed peer review.
Whole-body irradiation carries major clinical risk and makes the brain permissive to blood-derived cells. A practical therapy would need a safer conditioning strategy, a defined donor-cell product, reproducible trafficking into relevant brain regions, and evidence that the cells do not create inflammatory or proliferative harm. Human BDNF biology also cannot be inferred from a small mouse transplantation study.
The work is most useful as a timing experiment. It suggests that a donor-derived macrophage population can influence aged hippocampal plasticity during a limited residence window, and that BDNF-TrkB signaling may carry part of that effect. It does not yet justify a claim of brain rejuvenation or a ready cell therapy.
Source and disclosures
This analysis uses the complete bioRxiv preprint, “Young bone marrow-derived microglia-like cells improve cognitive function in the aged brain through BDNF-TrkB signalling”, version 1 posted October 21, 2026. The authors report no competing interests; the work was supported by Canadian public, foundation, and philanthropic funding described in the manuscript. Data are available from the corresponding author on reasonable request. The hero is an original generated editorial illustration, not primary evidence.