Alzheimer’s Disease Sabotages the Bone Marrow, Starving the Brain of Healing Immune Cells

For decades, Alzheimer’s disease has been studied almost exclusively as a disorder of the brain: misfolded amyloid-beta plaques, tangled tau proteins, and the slow death of neurons. But a new study published in Nature Neuroscience suggests that some of the most consequential damage may begin far outside the skull, in the spongy tissue of the bone marrow where the immune system manufactures its front-line soldiers. A team led by Michal Schwartz of the Weizmann Institute of Science, together with Aleksandra Deczkowska of the Institut Pasteur, reports that Alzheimer’s disease corrupts the very production line that generates monocytes, the blood cells capable of migrating into the diseased brain and helping to clear pathology. The culprit, they found, is a maladaptive surge of type I interferon signaling, an immune alarm signal that, when chronically activated, warps the bone marrow’s output and ultimately starves the brain of the reinforcements it needs.

The idea that bone marrow-derived macrophages can help fight Alzheimer’s pathology is not new. Previous work from the same group and others showed that these cells, when they do manage to enter the brain, act as reinforcements for microglia, the brain’s resident immune cells. Microglia in the Alzheimer’s brain gradually become exhausted and dysfunctional, losing their ability to engulf amyloid plaques efficiently. Monocyte-derived macrophages arriving from the circulation can compensate, reducing inflammation and slowing cognitive decline in mouse models of amyloidosis. Boosting their spontaneous recruitment has been shown to ameliorate disease in several experimental settings. The puzzling question was why, if these cells are beneficial, the body does not send more of them to the battlefield. The new study provides a striking answer: the supply line itself is broken.

Using the 5xFAD mouse model of familial Alzheimer’s disease, the researchers traced the problem back to hematopoietic stem cells, the master progenitors residing in the bone marrow that give rise to all blood cell types. In healthy animals, a large fraction of these stem cells remain in a state of quiescence, a dormant mode that preserves their long-term capacity to replenish the immune system. In the Alzheimer’s mice, this quiescence was lost. The stem cells were being pushed out of dormancy prematurely, and the downstream assembly line, a process called myelopoiesis that generates monocytes and other myeloid cells, was visibly impaired. Single-cell RNA sequencing of the femur and skull bone marrow revealed that multipotent progenitors in the diseased animals carried a strong transcriptional signature of type I interferon activation, along with defects in their commitment to the monocyte lineage.

Type I interferons are best known as antiviral cytokines, the body’s first response to infection. But when their signaling persists without resolution, they can be deeply harmful to stem cells. Prior studies have shown that chronic interferon exposure drives hematopoietic stem cells out of quiescence, exhausts their self-renewal capacity, and accelerates aging of the blood system. The new work extends this concept to neurodegeneration. The researchers detected elevated levels of interferon-beta in the bone marrow of the Alzheimer’s model mice, and when they cultured stem cells in a dish with interferon-beta, monocyte differentiation was suppressed in a dose-dependent manner. The lymphocyte compartment, by contrast, remained largely unaffected, indicating a selective attack on the myeloid branch of the immune system.

Crucially, the damage was not confined to mice. By reanalyzing single-cell transcriptomic datasets from human patients, the team found that circulating classical monocytes in the blood of Alzheimer’s patients, and monocyte populations in their cerebrospinal fluid, displayed phenotypic abnormalities that closely mirrored those seen in the mouse bone marrow. A shared set of disease-relevant functions was dysregulated across species, and the patients’ monocytes carried a molecular score that distinguished them not only from healthy controls but also, progressively, from people with mild cognitive impairment. This suggests that the bone marrow dysfunction is not a rodent artifact but a feature of the human disease, potentially tracking its progression.

The mechanistic link to failed brain homing came down to a single receptor. The researchers found that classical monocytes in the Alzheimer’s mice expressed abnormally high levels of CXCR4, a receptor that keeps cells anchored in the bone marrow by responding to its ligand CXCL12. Monocytes from the diseased marrow migrated poorly toward CXCL12 gradients in Transwell assays, and their ability to leave the marrow and reach the brain was compromised. When the team treated mice with AMD3100, a drug that blocks CXCR4, circulating monocyte numbers rose and more monocytes infiltrated the brain. In other words, the interferon-driven marrow environment was effectively imprisoning the very cells the brain needed, trapping them at their site of production rather than releasing them to patrol the diseased nervous system.

The therapeutic implications were tested in two complementary ways. First, the researchers blocked type I interferon signaling with neutralizing antibodies against IFNAR1, the interferon receptor chain. Second, they transplanted bone marrow from IFNAR1-deficient donor mice into 5xFAD recipients, creating chimeric animals whose blood system could no longer respond to the cytokine. In both approaches, myelopoiesis was restored, monocyte phenotypes normalized, and, importantly, the number of monocyte-derived macrophages homing to the brain increased. The brains of the treated animals showed ameliorated disease pathology, including reduced synaptophysin-positive dystrophic neurites in the hippocampal CA1 region and decreased phospho-tau signal, alongside shifts in the composition and function of brain-resident myeloid cells. Cell-cell communication analysis revealed that the infiltrating cells re-engaged in productive interactions with microglia and other brain cells.

The study fits into a growing body of evidence that Alzheimer’s disease is not solely a brain disease but a systemic condition involving the entire neuroimmune axis. Recent work has shown that the skull and vertebral bone marrow serve as dedicated myeloid cell reservoirs for the brain’s protective membranes, communicating with the central nervous system through specialized vascular channels and cerebrospinal fluid routes. Other studies have linked bone marrow hematopoiesis to the progression of multiple sclerosis, and epidemiological analyses have connected systemic infections to subsequent dementia risk. The new findings add a critical layer to this picture: it is not just that peripheral immune cells influence the brain, but that the disease process reaches into the immune system’s factory and corrupts its output, creating a vicious cycle in which a failing brain fails to summon its own repair crew.

There are important caveats. The work relies heavily on the 5xFAD and APP knock-in mouse models, which recapitulate amyloid pathology but not the full complexity of human Alzheimer’s disease, and the human data come from reanalysis of existing datasets rather than from newly collected patient cohorts. Bone marrow transplantation and systemic interferon blockade are aggressive interventions with substantial clinical risks, and the researchers themselves note that translating these findings into therapy will require far more targeted approaches. Still, the identification of a druggable pathway, type I interferon signaling in the bone marrow, that controls both immune cell production and brain homing opens a genuinely new therapeutic frontier. It suggests that future treatments for Alzheimer’s might one day aim not only at plaques and tangles within the brain, but at restoring the health of the immune system that stands ready to defend it.

For a field that has endured repeated disappointments in clinical trials targeting amyloid and tau, the message of this study is both sobering and invigorating. It reframes Alzheimer’s disease as a failure of communication between the brain and the body’s immune headquarters, and it pinpoints a specific molecular saboteur, chronic type I interferon, that can be measured, modeled, and potentially neutralized. If the bone marrow dysfunction observed in mice and echoed in human patients proves to be a driver rather than merely a bystander of disease progression, then the road to effective treatment may run not through the brain itself, but through the quiet factories of blood cells hidden in our bones.

Subject of Research: Bone marrow myelopoiesis dysfunction and type I interferon signaling in Alzheimer's disease

Article Title: Bone marrow myelopoiesis dysfunction in Alzheimer’s disease limits monocyte homing to the brain and drives disease progression

Article References: Abellanas, M. A., Basurco, L., Purnapatre, M., Burgaletto, C., Castellani, G., Colaiuta, S. P., Peralta-Ramos, J. M., Ibraheem, A., Murad, S., Antonello, P., Kovacs, M., Androsova, Y., Nathansohn, B., Partney, H., Cahalon, L., Valdes-Mas, R., Josephides, J. M., Salame, T. M., Espelosin, M., … Schwartz, M. (2026). Bone marrow myelopoiesis dysfunction in Alzheimer’s disease limits monocyte homing to the brain and drives disease progression. Nature Neuroscience. https://doi.org/10.1038/s41593-026-02417-7

Image Credits: AI Generated

DOI: 10.1038/s41593-026-02417-7

Keywords: Alzheimer's disease, bone marrow, myelopoiesis, monocytes, type I interferon, macrophages, microglia, neuroimmunology, hematopoietic stem cells, CXCR4, 5xFAD mouse model, brain homing