The normal aging of the brain entails a web of interconnected mechanisms that affect both neurons and their supporting cells. Progressive loss of the cerebral white matter is part of the picture and contributes to cognitive decline1. In addition, deficits of the cerebrovascular system are associated with normal and pathological aging2,3,4. However, the causal links between vascular, neuronal and glial aging are difficult to untangle, because of the circular and reciprocal relationships that tie the maintenance of the myelin sheet with variations in neuronal activity and blood perfusion levels. Given the important therapeutic potential of targeting the cerebrovascular system, and its implication in neurodegenerative diseases, understanding these links is of utmost importance. In a recent study5, Stamenkovic and colleagues identify a structural weakness in the deep cortical blood circulation that may hold one of the keys to understanding the frailty of the aging myelin sheet.

Neurovascular research has seen a recent surge in efforts to describe the cellular and topological evolution of the cerebral vasculature during aging. A comprehensive mapping study3 by Bennett et al. used 3D whole-brain vascular reconstructions to identify multiple impairments in the brain vascular network during normal aging. Compared to young brains, aged mouse brains displayed reductions in vascular density and length, pericyte density and oxygen-carrying capacity, as well as increases in vessel radii, arteriole tortuosity and blood–brain barrier permeability. However, the impact of this natural evolution is unclear, as its effects tend to wash away when the vasculature is studied at in bulk. The study from Stamenkovic avoided this pitfall: the striking insight of this work was to focus on principal cortical veins (PCVs), the sparse, hulking, tree-like vessels that plunge from the surface to the bottom of the cortex and collect blood from a large network of tributary capillaries.

Reconstructing this deep circulation in vivo is no small feat: the authors used two-photon imaging with far-red dyes and objectives capable of very deep in vivo scanning through large cranial windows. They focused their efforts on painstakingly reconstructing, in young and aged mice, the functional topology of the capillary network that contributes to the drainage of the circulation at the interface between gray and white matter. A notable innovation was the study’s specific focus on pre-convergence capillaries — capillaries uniquely positioned at points of lowest blood flow, just before merging into venules — highlighting them as critical yet vulnerable components of cerebral circulation.

A crucial finding was the selective reduction and constriction of capillaries in the tributaries of PCVs during aging. These microvascular alterations caused a mild but significant hypoperfusion (about 20% reduction) predominantly localized in the deeper cortical layers and corpus callosum. This reduction in flow was associated with astrogliosis, microgliosis and a slight reduction in myelination in the deep cortical layers and associated white matter tracts in aged brains. Interestingly, the study also found that these pathological changes occur without causing hypoxia, suggesting that even mild chronic hypoperfusion might critically impair the metabolic support essential for maintaining white matter integrity.

This comprehensive description of alterations in deep cerebral blood flow at major drainage sites in the cortex seems to position this portion of the vascular network as particularly vulnerable to minute changes. Indeed, the density of capillaries decreases sharply in deep cortical layers6, and the PCVs are the only available outlets that reach deep enough to collect outbound flow. Further highlighting this vulnerability, the authors show that focused ablation of pre-convergence capillaries has an outsized effect on deep circulation. Together, these findings suggest a more vulnerable environment during aging: the complementary use of computational modeling further underscored that mild microvascular changes in the deep parts of the cortex profoundly alter cerebral perfusion patterns.

To test whether the reduction in deep cerebral blood flow is responsible for the alterations in white matter in aged mice, the authors fine-tuned a strategy to induce mild hypoperfusion in 6–9-month-old mice. They used unilateral common carotid stenosis, implanting micro-coils in the left carotid artery. The careful calibration and unilateral nature of this surgery resulted in a 20% reduction in cortical blood flow in deep layers, similar to that seen in deep layers during normal aging. This reduction in blood flow was sufficient to replicate the cellular alterations seen in the white matter and deep cortical layers during aging, also seemingly without triggering hypoxia.

Of note, the demyelination measured after induced hypoperfusion was milder than in normal aged mice. Combined with the fact that demyelination was not seen in other studies using similar, albeit more aggressive, hypoperfusion schemes7, this suggests that additional mechanisms drive the loss of myelin during aging, acting in parallel to vascular alterations. Changes in neuronal activity patterns or increased inflammation during aging are other avenues to consider here1.

A possible model could integrate mechanisms that independently regulate the maintenance of the capillary-venous network and myelination on top of the direct link between hypoperfusion, and myelin maintenance proposed here. For instance, both the vascular network and myelin maintenance can be directly affected by aging-related inflammation, which affects pericytes and perivascular macrophages. VEGF-A could also be positioned as a master regulator, as alteration in its expression, or the activity of its receptors FLT1 and KDR, can also alter both the vascular system and myelinating precursor cells8.

The mechanisms for the observed reduction in blood flow in the deep cortical circulation and the vulnerability of pre-convergence capillaries are still not well understood. Recent literature has increasingly recognized the role of pericytes in regulating capillary blood flow, as well as their vulnerability during aging9,10. The authors observed age-related pericyte loss specifically in the deep white matter vasculature. They speculate that alterations in pericytes during aging could influence blood flow and thereby precipitate the homeostatic elimination of underperfused capillaries (Fig. 1). As the authors suggest, establishing mechanistic links between aging and the loss of pericytes and their contractility is urgent.

Fig. 1: Convergent bifurcation as an Achilles’ heel of cortical cerebral blood flow.
figure 1

Aging affects the density of deep capillaries that collect blood in cortical layer 6 and white matter tracts, probably by reducing or stalling blood flow. This capillary loss has an outsized influence in deep cortical regions, where the vascular density is lower than in upper layers and drainage sites are sparse.

Full size image

The key contribution of this study is to emphasize the vulnerability of pre-convergence capillaries — the capillaries that experience the lowest flow rates and are therefore more prone to transient occlusions and subsequent regression. This notion aligns with emerging research suggesting that capillary stalling, often driven by pericyte dysfunction, endothelial inflammation or leukocyte adherence, might be an initiating event in broader network impairments11. However, the mouse cerebral vascular network spans several hundred meters12,13, and even with the help of recent large-scale 3D imaging, the hunt for causal links is a needle in a haystack problem. Investigating how network vulnerabilities are situated in different brain regions with varied arterio-venous and capillary structures would be an exciting follow up. By emphasizing the deep network of pre-convergence capillaries in the neocortex, this study provides a strong framework for a better understanding of aging throughout the brain.