Cell autonomous role of border associated macrophages in ApoE4 neurovascular dysfunction and susceptibility to white matter injury

doi:10.21203/rs.3.rs-3222611/v1

Download PDF

Article

Cell autonomous role of border associated macrophages in ApoE4 neurovascular dysfunction and susceptibility to white matter injury

https://doi.org/10.21203/rs.3.rs-3222611/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 18 Sep, 2024

Read the published version in Nature Neuroscience →

Version 1

posted

You are reading this latest preprint version

Apolipoprotein-E4 (ApoE4), the strongest genetic risk factor for sporadic Alzheimer’s disease, is also a risk factor for microvascular pathologies leading to cognitive impairment, particularly subcortical white matter injury. These effects have been attributed to alterations in the regulation of the brain blood supply, but the cellular source of ApoE4 and the underlying mechanisms remain unclear. In mice expressing human ApoE3 or ApoE4 we report that border associated macrophages (BAM), myeloid cells closely apposed to neocortical microvessels, are both the source and the target of the ApoE4 mediating the neurovascular dysfunction through reactive oxygen species. ApoE4 in BAM is solely responsible for the increased susceptibility to oligemic white matter damage in ApoE4 mice and is sufficient to enhance damage in ApoE3 mice. The data unveil a new aspect of BAM pathobiology and highlight a previously unrecognized cell autonomous role of BAM in the neurovascular dysfunction of ApoE4 with potential therapeutic implications.

Biological sciences/Neuroscience

Health sciences/Diseases/Neurological disorders/Cerebrovascular disorders

The integrity of the neurovasculature is critically important for cognitive health, and neurovascular dysfunction and damage have been implicated in a wide variety of conditions associated with cognitive impairment^{1, 2}. In particular, microvascular damage to the subcortical white matter is a major contributor to age-related dementia ^{3, 4} and may also play a role in Alzheimer’s disease (AD) ^5-8. Supplied by terminal arterioles with limited collateral flow from adjacent vascular territories ^{9, 10}, the subcortical white matter is particularly vulnerable to microvascular injury ^{11, 12}. While vascular risk factors, such as hypertension and diabetes, are strongly linked to white matter lesions, accumulating evidence indicates that ApoE4, the leading genetic risk factor in sporadic AD, also increases the risk for cognitive impairment produced by vascular factors ^13-15. Thus, ApoE4 carriers exhibit vascular pathology, microvascular alterations, and more white matter lesions linked to cognitive impairment^16-22. Furthermore, in Aβ immunotherapy ApoE4 positivity increases the incidence of amyloid related imaging abnormalities (ARIA), a treatment-limiting adverse event related to cerebral amyloid angiopathy and microvascular dysfunction ^23-25, highlighting the negative neurovascular impact of ApoE4. Thus, ApoE4 promotes white matter injury as well as other neurovascular pathologies.

ApoE4 is well-known to be associated with alterations of the neurovasculature. ApoE4-positive individuals have dysregulated cerebral blood flow (CBF) ^26-30, as well as increased permeability of the blood brain barrier (BBB) in the setting of AD ^{31, 32}. Mice expressing human ApoE4 under the control of the mouse ApoE promoter (ApoE4-TR mice) exhibit a profound disruption in the ability of neural activity to increase cerebral blood flow (CBF) and of cerebral endothelial cells to regulate microvascular perfusion ³³, the integrity which is required for normal cognition ^{34, 35}. In a model of white matter injury produced by bilateral common carotid artery stenosis (BCAS) ApoE4-induced neurovascular dysfunction reduces white matter CBF and enhances white matter lesions and cognitive deficits ³³. These alterations occur at an early age (2-3 months) and precede the BBB dysfunction ^{33, 36}. However, the cellular source(s) of ApoE4, the effector cells in the cerebral microvasculature and the signaling mechanisms driving the dysfunction remain unclear.

The brain harbors myeloid cells that arise from the yolk sac early in development and populate different sites in the brain and its coverings in the post-natal period ³⁷. While microglial cells reside in the brain parenchyma, border associated macrophages (BAM) seed the meninges, the choroid plexus, and the space surrounding microvessels as they dive into the brain (perivascular space) ^{38, 39}. BAM are enriched with free radical producing enzymes and, owing to their proximity to pial arterioles in the leptomeninges and to penetrating arterioles in the perivascular space, have recently emerged as a major cause of neurovascular dysfunction in models of hypertension, vascular amyloid pathology, and sub-arachnoid hemorrhage ^40-44.

In the present study we investigated the sources and targets of ApoE responsible for neurovascular dysfunction and the mechanisms of the effect. We found that ApoE4 acts on BAM to alter critical cerebrovascular regulatory mechanisms through NADPH oxidase dependent production of reactive oxygen species (ROS). The dysfunction is abolished by BAM depletion or by genetic deletion of ApoE4 selectively in BAM, indicating that these cells are the sole source of the ApoE4 mediating the deleterious vascular effects. Using a bone marrow transplantation strategy, we found that ApoE4-positive BAM induce neurovascular dysfunction in ApoE3-TR mice, whereas ApoE3-positive BAM rescue neurovascular dysfunction in ApoE4-TR mice, indicating that BAM are also the target of ApoE4. Attesting to the pathogenic effect of BAM ApoE4 on white matter injury, ApoE4-positive BAM enhance white matter damage and cognitive impairment in ApoE3-TR mice, while ApoE3-positive BAM rescue this phenotype in ApoE4-TR mice. The findings establish BAM as both the source and the target of the ApoE4 acting on the cerebral microvasculature, unveiling a previously unappreciated cell autonomous role of brain-resident macrophages in the neurovascular dysfunction and propensity of white matter injury associated with ApoE4.

1. Neocortical application of ApoE4, but not ApoE3, attenuates neurovascular responses initiated by neural activity and the endothelium

ApoE4-TR mice exhibit profound alterations in the increases in CBF induced by neural activity and by vasodilating agents acting through the endothelium ³³. To determine if ApoE4 is directly responsible for these neurovascular effects we examined if topical application of recombinant (r) ApoE4 to the neocortex reproduces the neurovascular dysfunction observed in ApoE4-TR mice ³³. CBF was monitored in the whisker barrel cortex by laser-Doppler flowmetry in anesthetized WT mice (age 3-4 months) with controlled blood pressure and blood gasses ⁴⁵. Bathing the exposed neocortex with rApoE4 (0.2-10 µg/ml) attenuated the increase in CBF induced by mechanical stimulation of the facial whisker in a concentration related manner (Fig. 1A). Similarly, rApoE4 attenuated the rise in CBF induced by neocortical application of acetylcholine, a vasodilator that acts by releasing nitric oxide from the cerebral endothelium (Fig. 1A) ⁴⁶. The CBF response to application of the smooth muscle relaxant adenosine was not attenuated (Fig. 1A), attesting to the selectivity of the effect of ApoE4 on functional hyperemia and endothelial vasoactivity. Lipidation influences ApoE biological activity ^{47, 48}, but the neurovascular effects of lipidated rApoE3 and rApoE4 were comparable to those of the non-lipidated ApoE (Fig. 1A). The reduction of functional hyperemia tended to be more marked with lipidated rApoE4 (Fig. 1A), possibly due to lipidation enabling deeper penetration into the whisker barrel cortex. rApoE3 was devoid of cerebrovascular effects (Fig. 1B). To test if the neurovascular effects of rApoE4 were receptor mediated we applied rApoE4 after pre-treatment of the neocortex with receptor associated protein (RAP), an inhibitor of ApoE receptors ⁴⁹. We found that RAP by itself did not affect cerebrovascular responses, but completely reversed the rApoE4-induced dysfunction (Fig. 1C). These observations indicate that exogenous ApoE4 reproduces the neurovascular dysfunction observed in ApoE4-TR mice and that the effect is receptor mediated.

2. ApoE4 induces neurovascular dysfunction through ROS production in BAM

The CBF dysfunction in ApoE4-TR mice is attenuated by a ROS scavenger ³³. To provide insight into the enzymatic source of the radicals we examined the effect of a highly selective peptide inhibitor of NADPH oxidase (gp91ds-tat) ⁵⁰, a major source of cerebrovascular oxidative stress ^51-53. Pre-treatment of the cortex with gp91ds-tat did not influence baseline CBF responses but prevented the attenuation in functional hyperemia and endothelium-dependent vasoactivity in ApoE4-TR mice, while the scrambled control peptide had no effect (Fig. 2A). Similarly, gp91ds-tat counteracted the neurovascular dysfunction induced by neocortical application of rApoE4 in WT mice (Fig. 2B). CBF responses to adenosine were not affected in ApoE4-TR mice or with rApoE4 application in WT mice (Extended Data Fig. 1A). No changes in CBF responses were observed in ApoE3-TR mice or with rApoE3 application in WT mice (Fig. 2A,B; Extended Data Fig. 1A). These findings implicate NADPH oxidase-derived ROS in the mechanisms of the neurovascular dysfunction of ApoE4.

Next, we sought to define the cellular sources of the ROS. BAM have emerged as a leading cause of vascular oxidative stress via NADPH oxidase derived ROS ^{40, 42, 43}. To investigate if these cells were the source of ROS, we first examined if rApoE4 induces ROS production in BAM. To this end, WT mice were injected intracerebroventricularly (icv) with cyanine-5 (Cy5)-dextran, which is phagocytosed by BAM rendering them fluorescent (Fig. 2C) ^{40, 42, 44}. The following day, mice were anesthetized, equipped with a cranial window, injected i.v. with the ROS marker dihydroethidine (DHE) and with FITC-dextran to label cerebral microvessels ^{42, 44}, and imaged with a 2-photon microscope ^{43, 45}. Vehicle, rApoE3 or rApoE4 was applied to the exposed neocortex and ROS-dependent fluorescence was monitored in Cy-5-positive macrophages surrounding blood vessels at a depth of 150 µm from the pial surface. rApoE4 increased the ROS signal in BAM, but rApoE3 failed to do so (Fig. 2C). Furthermore, in WT mice rApoE4, but not rApoE3, increased ROS production, assessed by flow cytometry, in BAM (CD36⁺, CD45^high) and microglia (CD45^Int, CD11b⁺), but not in endothelial cells (Fig. 2D). In ApoE4-TR mice, the ROS signal was increased in ApoE4 BAM but not in ApoE4 microglia or endothelial cells (Fig.2E). ROS were not elevated in ApoE3-TR mice (Fig. 2E). Of note, ROS increases in microglia were not observed in ApoE4-TR mice (Fig. 2E). These findings, collectively, demonstrate that ApoE4 triggers ROS production in BAM and that BAM are a major source of vascular oxidative stress in ApoE4-TR mice.

3. BAM are required for neurovascular dysfunction induced by ApoE4

Having established that the deleterious neurovascular effects of ApoE4 are mediated by NADPH oxidase-derived ROS and that ApoE4 induces ROS production in brain macrophages, we investigated if ApoE4 acts on BAM to mediate the neurovascular dysfunction. To this end, we depleted BAM using liposome-encapsulated clodronate (CLO) ^{40, 42, 43}. CLO or liposome controls were injected icv in WT, ApoE3-TR and ApoE4-TR mice and, 5-7 days later mice were equipped for monitoring of CBF by laser-Doppler flowmetry. CLO depleted BAM by 80-90% in WT, ApoE3-TR and ApoE4-TR mice (Fig. 3A) without affecting microglia (Extended Data Fig. 1B), as shown before ^{40, 42, 43}. In WT and ApoE3-TR mice, CLO did not alter functional hyperemia, endothelial vasoactivity to acetylcholine or CBF responses to the smooth muscle relaxant adenosine (Fig. 3B; Extended Data Fig. 1C). Remarkably, CLO completely reversed the attenuation of functional hyperemia and endothelium-dependent response in ApoE4-TR mice (Fig. 3B). Since application of rApoE4 to the exposed neocortex reaches leptomeningeal and perivascular spaces and increases ROS in BAM (Fig. 2C), we asked if CLO would also ameliorate the neurovascular dysfunction induced by neocortical application of rApoE4. CLO treatment did not affect neurovascular responses when vehicle or rApoE3 was applied to the neocortex, but completely prevented the attenuation in functional hyperemia and endothelial vasoactive function induced by rApoE4 application (Fig. 3C). CBF responses to adenosine were not affected (Extended Data Fig. 1C). These observations indicate that BAM are required for the neurovascular dysfunction induced by ApoE4.

4. BAM are the source of the ApoE mediating the neurovascular dysfunction

Next, we asked if BAM are the source of the ApoE4 mediating the neurovascular dysfunction. In addition to astrocytes and disease-associated microglia ⁴⁷, macrophages also express and secrete ApoE ⁵⁴. To document that ApoE is also expressed in BAM we mined publicly available single-cell RNAseq databases of mouse brain for ApoE transcripts in astrocytes, myeloid and vascular cells (see methods). We found that ApoE is expressed in BAM at levels comparable to those of astrocytes, higher than microglia, endothelial cells, and cells of the vascular wall (Fig. 4A). ApoE expression in perivascular macrophages has also been observed in the human brain ²². Therefore, BAM could very well be the source of the ApoE4 mediating the neurovascular dysfunction.

To provide evidence that ApoE4 derived from BAM is responsible for the neurovascular dysfunction observed in ApoE4-TR mice we developed a mouse model to genetically target BAM. For tamoxifen-inducible Cre recombinase expression in BAM, we generated a new mouse line in which a CreERT2 cassette was inserted into the Mrc1 locus (Mrc1^CreERT2; here designated Mrc1^Cre) (Extended Data Fig. 2A). To test the cellular selectivity of the Cre recombinase expression Mrc1^Cre mice were crossed with tdTomato mice (R26^tdT: Jackson Laboratory Strain #: 007914). After tamoxifen treatment of Mrc1^Cre+/R26^tdT crosses, recombinase activity was observed in over 80-90% of cells expressing the BAM marker CD206 (Extended Data Fig. 2A). No recombinase activity was seen in microglia or endothelial cells (Extended Data Fig. 2A).

To selectively knock-down ApoE4 in BAM we crossed Mrc1^Cre with ApoE4^fl/fl mice. The efficiency of the selectivity and downregulation for ApoE in BAM was verified with RNAscope in situ hybridization. In corn-oil (vehicle)-treated Mrc1^Cre+/ApoE4^fl/fl or Mrc1^Cre+/ApoE3^fl/fl mice, ApoE was observed both in BAM and in astrocytes, a major ApoE source (Fig. 4B,C; Extended Data Fig. 2B-C,F-G; Extended Data Fig. 3A,C,E-F). Then we went on to study the effect of ApoE deletion in BAM. Tamoxifen treatment of Mrc1^Cre+/ApoE4^fl/fl or Mrc1^Cre+/ApoE3^fl/fl mice deleted ApoE in BAM (Fig. 4B,C; Extended Data Fig. 2D-E), without altering ApoE in astrocytes (Extended Data Fig. 3B-F), or ApoE levels in brain, CSF or blood (Fig. 4D; Extended Data Fig. 3G,H). With tamoxifen treatment, neurovascular responses in Mrc1^Cre+ mice were not different from those in WT mice, while in Mrc1^Cre-/ApoE4^fl/fl mice functional hyperemia and endothelium-initiated CBF response were attenuated as much as in ApoE4-TR mice (Fig. 4E), as anticipated since ApoE4^fl/fl mice are equivalent to ApoE4-TR mice ⁵⁵. Conversely, in tamoxifen treated Mrc1^Cre+/^/ApoE4^fl/fl mice neurovascular and endothelial responses were completely normal (Fig. 4F), but importantly application of rApoE4 to the neocortex reproduced in full the neurovascular dysfunction (Fig. 4E). This observation rules out the possibility that CBF responses were normalized because of deficits in ApoE receptors and/or downstream signaling pathways needed for ROS production. CBF responses to adenosine were not affected in all groups (Extended Data Fig. 1D). Similarly, rApoE4 attenuated neurovascular responses in Mrc1^Cre+ mice, attesting to their susceptibility to the neurovascular effects of ApoE4 (Extended Data Fig. 4A). Of note, in the absence of tamoxifen, Mrc1^Cre+/ApoE4^fl/fl exhibited attenuated CBF responses (Extended Data Fig. 4B), as did tamoxifen treated Mrc1^Cre-/ApoE4^fl/fl mice (Fig. 4E) or ApoE4-TR mice (Fig. 3B), ruling out effects of tamoxifen. Confirming the link between ApoE4 and ROS production in BAM, the baseline ROS signal, assessed by DHE in cortical brain slices, was increased in BAM of Mrc1^Cre+/ApoE4^fl/fl mice compared to Mrc1^Cre+ mice and was reduced by treatment with tamoxifen (Extended Data Fig. 4C,D). As anticipated, neurovascular responses were not altered in tamoxifen treated Mrc1^Cre+/ApoE3^fl/fl mice (Fig. 4F; Extended Data Fig. 1D). Therefore, downregulation of ApoE4 in BAM did not alter ApoE levels in the brain, but rescued the neurovascular dysfunction in full, consistent with BAM being the sole source of ApoE4 mediating the vascular effects.

5. BAM as source and target of the ApoE4 mediating neurovascular dysfunction

The results in Mrc1^Cre+/ApoE4^fl/fl mice indicate that BAM are the source of ApoE4 mediating the vascular impairment. Furthermore, the observation that application of rApoE4 to pial and perivascular spaces in WT mice induces neurovascular dysfunction through BAM (Fig. 3C) suggests that BAM could also be target of ApoE4. If so, we hypothesized that repopulation of perivascular spaces of ApoE4-TR mice with ApoE3 macrophages would rescue the neurovascular dysfunction and, conversely, ApoE4 macrophages would induce the dysfunction in ApoE3-TR mice. To test this hypothesis, we used a bone marrow chimera-based approach. Bone marrow chimeras have been successfully used to study BAM ^{40, 42, 43, 56, 57}. Brain irradiation depletes BAM and makes their niche accessible to bone marrow-derived macrophages, which repopulate the perivascular space and meninges, retain the ability to mediate vascular oxidative stress and neurovascular dysfunction ^{40, 42, 43} and are transcriptomically like BAM ⁵⁸. Microglia are not replaced in this time frame ^{40, 42, 44, 58, 59}. Using this strategy, we tested whether ApoE4 macrophages induce neurovascular dysfunction in an ApoE3-TR host, and, conversely, whether ApoE3 macrophages rescue the dysfunction in an ApoE4-TR host. Bone marrow from ApoE4-TR or ApoE3-TR mice was transplanted in ApoE3-TR (E4→E3TR) or ApoE4-TR mice (E3→E4TR), respectively. Experiments were performed 3 months after bone marrow transplantation, when the acute brain changes induced by irradiation have subsided and mice have an intact BBB, as well as neurovascular and cognitive function ^{40, 42, 43, 60}. WT→WT or E4→E4TR bone marrow chimeras served as control to confirm lack of untoward effects of the transplantation procedure. As in previous studies ^{40, 42}, neural, endothelial, or smooth muscle evoked CBF responses in WT→WT chimeras were not different from those of naïve WT mice (Fig. 5; Extended Data Fig. 1E), confirming that transplantation did not impair neurovascular regulation. Furthermore, in E4→E4TR chimeras CBF responses were attenuated as much as in ApoE4-TR mice (Fig. 5), indicating that bone marrow-derived macrophages repopulating perivascular and leptomeningeal spaces induce neurovascular dysfunction just like yolk sack-derived BAM. Remarkably, however, ApoE4-positive macrophages were able to induce neurovascular dysfunction in ApoE3-TR mice and, conversely, ApoE3-positive macrophages completely rescued the dysfunction in ApoE4-TR mice (Fig. 5). Therefore, ApoE3 in BAM normalize neurovascular function in ApoE4-positive host brains, whereas ApoE4 in BAM are sufficient to induce neurovascular dysfunction in ApoE3-positive host brains. These data provide evidence that the ApoE status of BAM is a key determinant of neurovascular function independently of the ApoE status of the host.

6. ApoE4 in BAM enhances white matter injury

Finally, we wondered if ApoE4 in BAM is also involved in the exacerbation of WM damage observed in ApoE4-TR mice ³³. To this end, we induced chronic white matter damage to the corpus callosum by reduction in CBF using BCAS, a well-established model of oligemic white matter injury ⁶¹, in ApoE bone marrow chimeras. Stainless-steel coils, 0.18mm in diameter ^{33, 62}, were placed around both common carotid arteries to reduce the vessel’s lumen and lower forebrain CBF in WT→WT, E3→E3TR, E4→E3TR, E3→E4TR, and E4→E4TR chimeras (Fig. 6A). In agreement with previous data in ApoE3-TR and ApoE4-TR mice ³³ the CBF reduction, assessed by laser-speckle imaging, was comparable in WT→WT, E3→E3TR, but was more pronounced in E4→E4TR chimeras (Fig. 6B). However, in E4→E3TR the CBF reduction was comparable to that of E4→E4TR chimeras, and in E3→E4TR comparable to that of E3→E3TR chimeras (Fig. 6B). In all groups, BCAS induced white matter injury, as shown by the Kluver-Barrera stain, and by loss of myelin basic protein and oligodendrocytes (Fig. 6C-E). We also noticed loss of the integrity of the nodes of Ranvier, assessed by the spatial relationship between the paranodal contactin-associated protein Caspr and the nodal sodium channel Nav1.6 ³³ (Fig. 6F). Cognitive testing by the novel object recognition and Y-maze tests also showed impairment (Fig. 7A,B), as reported previously ³³. In agreement with the CBF data, E4→E3TR chimeras had worse white matter damage and cognitive deficits, which did not differ from those in E4→E4TR chimeras, while the outcome in E3→E4TR chimeras was comparable to that of E3→E3TR chimeras (Fig. 7A,B).

We investigated the role of BAM in the neurovascular dysfunction and increased susceptibility to white matter injury associated with the ApoE4 genotype. We demonstrated that ApoE4 alters critical homeostatic neurovascular regulatory mechanisms through NADPH oxidase-derived ROS, and that BAM are a pathogenic source of ROS in ApoE4-TR mice and in WT mice exposed to rApoE4. We also showed that BAM depletion normalizes neurovascular function in ApoE4-TR mice and prevents the dysfunction in rApoE4-treated WT mice. Selective deletion of ApoE4 in BAM using a novel Mrc1^Cre+/ApoE4^fl/fl mouse did not alter ApoE levels in brain, CSF or blood, but completely rescued the neurovascular alterations, consistent with BAM being the sole source of ApoE targeting the microvessels. Using a bone marrow chimera-based strategy we found that ApoE4-positive BAM cause neurovascular dysfunction in ApoE3-TR mice, while ApoE3-positive BAM rescue in full the dysfunction in ApoE4-TR mice. Consistent with these vascular findings, in an oligemic model of chronic white matter injury, ApoE4-positive BAM worsened the flow reduction, white matter damage, and cognitive dysfunction in ApoE3-TR mice, while ApoE3-positive BAM fully reversed these effects in ApoE4-TR mice. These findings, collectively, unveil a critical role for BAM in the vascular effects of ApoE4 and identify these cells as both the source and the target of the ApoE4 mediating the dysfunction (Extended Data Fig. 5).

ApoE is produced by different cell types in brain, major sources including astrocytes and disease-associated microglia ⁴⁷, and, as also shown here, BAM. However, our data demonstrate that cellular sources of ApoE4 other than BAM do not contribute to neurovascular dysfunction in ApoE4-TR mice. First, BAM depletion rescues vascular function in ApoE4-TR mice, in which other cell types producing ApoE4 are still present. Second, selective ApoE4 deletion in BAM does not reduce brain or blood levels of ApoE but rescues neurovascular function in full, indicating that ApoE4 from other sources does not play a role. Third, ApoE3-positive BAM are sufficient to normalize neurovascular function in an ApoE4 host, and, conversely, ApoE4-positive BAM are sufficient to induce neurovascular dysfunction in an ApoE3 host. Therefore, the data reveal a previously unappreciated perivascular compartmentation of the ApoE4 acting on the vasculature, which is probably related by the proximity of the ApoE4-ROS source to the vasculature, i.e., BAM abutting resistance arterioles regulating cerebral perfusion ³⁵. Whether such compartmentation also applies to sources of ApoE4 involved in other aspects of ApoE4 pathobiology, for example exacerbation of amyloid, tau or synuclein pathology ^63-65, remains to be established.

Previous studies have shown that ApoE4 expression in endothelial and smooth muscle cells or in the periphery can compromise neurovascular integrity ^66-68. These elegant studies have provided further evidence of the powerful vascular effects of ApoE4 but could not determine the targets of ApoE4, the cellular mechanisms of the vascular dysfunction nor the pathogenic consequence for the subcortical white matter when brain perfusion was threatened. Our data establish that in ApoE4-TR mice BAM are the unique effectors of the neurovascular dysfunction and susceptibility to white matter damage, in a cell-autonomous manner. However, it remains unclear whether in pathological conditions other vascular or systemic sources of ApoE4 may also play a role.

The findings have significant implications for the understanding of the increased propensity to microvascular alterations observed in ApoE4-positive individuals, including white matter damage and the ARIA syndrome in the setting of anti-Aβ immunotherapy ^{23-25, 69}. Our data raise the possibility that ApoE4 derived from BAM may contribute to these conditions and as such could be a target for therapeutic interventions. However, it is conceivable that in the presence of Alzheimer pathology or vascular risk factors, such as hypertension, ApoE4 originating from microglia or astrocytes, could also be involved. Studies addressing the role of ApoE4 in BAM and other cell types in pathological states may provide the opportunity to answer this question.

In conclusion, we demonstrated a new cell autonomous role of BAM in the neurovascular dysfunction of ApoE4 and provided evidence for a previously unrecognized perivascular compartmentation of the cellular source of the ApoE4 targeting the cerebral microvasculature and enhancing white matter injury. The findings expand our understanding of the role of BAM in neurovascular diseases leading to cognitive impairment, and provide a putative target to counteract the increased susceptibility to white matter injury and microvascular pathology associated with the ApoE4 genotype.

Acknowledgements

Supported by NIH grants R01-NS126467 (CI), R01-NS097805 (LP), K22-NS123507 (MMS), and the BrightFocus Foundation (AA). Support from the Feil Family Foundation is gratefully acknowledged.

Author contributions

A.A., L.P., S.S., M.S., N.C., and Y.H. conducted the experiments and performed the data analysis; G.W. performed the ROS measurement in BAM ex vivo; N.C. performed the RNAscope and histology experiments and assisted in mouse genotyping; P.Z. assisted in mouse breeding; M.S. and D.M.H. provided lipidated ApoE3 and ApoE4 and edited the manuscript; L.P., J.A. and C.I. supervised the research; L.P. and C.I. provided funding and wrote the manuscript.

Competing financial interest

D.M.H. is on the scientific advisory board of C2N diagnostics and has equity. D.M.H. is on the scientific advisory board of Denali Therapeutics, Genentech, and Cajal Therapeutics and consults for Asteroid. C.I. is on the scientific advisory board of Broadview Ventures.

Data availability

The authors declare that the data supporting the findings of this study are available within the article and Extended Data Figures.

Mice

All procedures were approved by the Institutional Animal Care and Use Committee of Weill Cornell Medicine and performed according to the ARRIVE guidelines ⁷⁰ and, as much as feasible, in a blinded fashion. Experiments were performed in homozygous ApoE3-TR and ApoE4-TR mice on a C57BL/6 genetic background ⁷¹. ApoE3-floxed (ApoE3^fl/fl) and ApoE4-floxed (ApoE4^fl/fl) homozygous mice on a C57BL/6 genetic background ⁵⁵ were crossed with Mrc1^CreERT2 mice (see “Generation of the Mrc1 targeting construct”). C57BL/6 mice were used as wild-type (WT) controls. We use male and female mice aged 3-6 months.

Generation of the Mrc1 targeting construct

Mrc1^CreERT2 mice (Mrc1^Cre+) were generated by Cyagen. The Mrc1 targeting construct (Extended Data Fig. 2A) was linearized by restriction digestion with NotI, followed by phenol/chloroform extraction and ethanol precipitation. The linearized vector was transfected into C57BL/6 ES cells according to Cyagen’s standard electroporation procedures. The transfected ES cells were subject to G418 selection (200 μg/mL) 24 hours post electroporation. Total 179 G418 resistant clones were picked and amplified in 96-well plates. Two copies of 96-well plates were made: one copy for storage at -80°C and the other copy for DNA isolation and subsequent PCR screening for homologous recombination. The PCR screening identified 18 potential targeted clones, from which 6 were expanded and further characterized by Southern blot analysis. Five of the six expanded clones were confirmed to be correctly targeted. The Mrc1 gene (NCBI Reference Sequence: NM_008625.2) is located on mouse chromosome 2. Thirty exons have been identified, with the ATG start codon in exon 1 and the TAG stop codon in exon 30. In the targeting vector, the coding region of exon 1 plus part of intron 1 was replaced with the CreERT2-polyA cassette. Correct integration of the CreERT2-polyA cassette was confirmed by diagnostic PCR. Sequencing of the genomic region of interest after amplification by PCR was further used to verify correct integration. After germline transmission, correctly targeted founders (Mrc1^Cre+ mice) were crossed with C57BL/6J mice to establish colonies for this study. To assess Mrc1 Cre recombinase in Mrc1^Cre+ mice, we crossed with tdTomato mice (R26^tdT: Jackson Laboratory Strain #: 007914), activated Cre recombinase with tamoxifen (TAM; Sigma, T-5648) dissolved in corn oil (Sigma, C-8267), and assessed the labeling efficacy. In 8–10-week-old mice, Cre recombination was induced by injecting 5 mg TAM subcutaneously for 5 consecutive days. Then, CD206⁺ BAMs were counted and analyzed (Extended Data Fig. 2A). At least three sections from a minimum of three mice were used for each analysis. Quantification on three-dimensionally reconstructed CD206⁺ cell images was obtained using Imaris software.

ApoE expression in brain myeloid and vascular cells

Mouse brain single cell mRNA data were extracted from the GSE174574 dataset deposited in the Gene Expression Omnibus repository (https://www.ncbi.nlm.nih.gov/geo/). The data used was derived from three male C57BL/6 mice (6-8 weels of age) that underwent sham surgery for middle cerebral artery occlusion ⁷². We chose this dataset because it contains brain myeloid, endothelial, mural cells, and astrocytes. After quality filtering that removed cells with <500 and >20,000 transcripts and more than 20% reads from mitochondrial genes, the count matrices were processed with Seurat (Vers. 4.1.0) ⁷³ in the R statistical environment. Log normalization, variable gene detection, scaling, and principal component analysis was performed with default settings. The R package Harmony ⁷⁴ was used to correct the matrix for batch effects. We used the Louvain algorithm as implemented in Seurat FindClusters to perform graph-based clustering on the neighbor graph that was constructed with the FindNeighbors function call on Harmony-derived embeddings. We performed unsupervised cell type annotation using the SingleR package ⁷⁵ with ImmGen ⁷⁶, BrainImmuneAtlas ⁷⁷, and Tabula Muris ⁷⁸ as reference datasets. Data were visualized in R using ggplot2 (V3.3.6).

Intracerebroventricular injection of clodronate or dextran

Liposomes containing clodronate or PBS were administered intracerebroventricularly as previously described ^{40, 42, 43}. Isoflurane-anesthetized mice were placed in a stereotaxic frame. Ten microliters of clodronate liposomes (7 mg/mL) or PBS liposomes (vehicle) were injected into the cerebral ventricles with a glass micropipette (rate <0.5 µl/min) through a burr hole drilled on the right parietal bone. Mice were used in the experiments 5 to 7 days later, when BAM depletion is well developed and stable ^{40, 42, 43}. In some experiments, BAM were identified by their ability to phagocytize dextran ^{40, 42}. For dextran injections, 10 µl of Alexa Fluorâ 680 dextran (10,000 MW, anionic, fixable, ThermoFisher Scientific, D34680; 2.5 mg/ml) in PBS or PBS alone were slowly injected into the cerebral ventricles with a glass micropipette through a burr hole drilled on the right parietal bone ⁴². BAM labeling was examined 24 hrs later.

Labeling cortical blood vessels with DiO

Cortical blood vessels were labeled with the lipophilic dye DiO [DiOC18(3) (3,3'-Dioctadecyloxacarbocyanine Perchlorate)], as described ^{42, 44}. Briefly, mice were anesthetized (5% isoflurane) and transcardially perfused with PBS (2 ml) followed by DiO (1:50, V-22886, Molecular Probes; 5ml/mouse) and then by 4% paraformaldehyde (PFA). Brains were harvested and post-fixed in 4% PFA overnight, then cut (thickness 150 µm) using a vibratome and examined under the confocal microscope (Leica SP8).

Lipidation of recombinant ApoE3 and ApoE4

Recombinant (r) ApoE3 (Cat# A128, Leinco Technologies) and rApoE4 (Cat# A129, Leinco Technologies) were purchased and lipidated as previously described ⁷⁹. Briefly, lyophilized rApoE3/4 was solubilized in DPBS buffer with 1mM DTT and 1mM EDTA at a final concentration of 25 µM. Separately, POPC (Avanti Polar Lipids) and cholesterol (Avanti Polar Lipids) dissolved in chloroform were combined in a glass vial for a final molar ratio of ApoE:POPC:Cholesterol at 1:50:10. The POPC/cholesterol mixture was dried under nitrogen until chloroform was completely evaporated. Then, DPBS was added to the dried POPC/cholesterol mixture and allowed to hydrate for 30 minutes. The mixture was added with sodium cholate (Sigma) at a 4:1 ratio (g/g) sodium cholate:POPC and incubated for 1 hour. Next, rApoE3/4 were added to sodium cholate:POPC:cholesterol mixture and incubated for 1hr at room temperature. The mixture was dialyzed in PBS at 4°C for 48 hrs using a 10,000 MWCO Slide-A-Lyzer Dialysis Cassette (Thermo Scientific) with three buffer changes. Samples were purified using a Superose 6 10/300 Increase GL column (Cytiva) with a flow rate of 0.5mL/min in PBS buffer. Samples were concentrated using an Amicon Ultra-15 10,000 MWCO concentrator (Sigma) at 4,000g for 20min at 4°C.

Monitoring of cerebral blood flow.

CBF was monitored using laser-Doppler flowmetry or laser-speckle flowmetry as previously described ^{33, 40, 42, 45} and briefly summarized below.

CBF response to neural activity and to endothelium-dependent and independent agonists: As described in detail previously ^{42, 45, 80}, anesthesia was induced with isoflurane (1-2%) and maintained with urethane (750 mg/kg; i.p.) and a-chloralose (50 mg/kg; i.p.). A femoral artery was cannulated for recording of arterial pressure and collection of blood samples. A 2x2 mm opening was drilled in the parietal bone overlying the somatosensory cortex, the dura was removed, and the site was superfused with a modified Ringer solution (37°C; pH 7.3-7.4) ^{40, 45}. Relative CBF was continuously monitored at the site of superfusion with a laser-Doppler flowmeter (Perimed). Arterial blood pressure, blood gases and rectal temperature were monitored and controlled. CBF recordings were started after arterial pressure (MAP, 78-85 mmHg) and blood gases (pO₂, 120-140 mmHg; pO₂, 33-40 mmHg; pH, 7.3-7.4) were in a steady state ^{45, 80}. For functional hyperemia, the whiskers were mechanically stimulated for 30 sec and the associated increase in CBF was recorded. To test endothelium-dependent responses ACh (100 µM; Sigma) was superfused on the cranial window, and the resulting change in CBF recorded. The CBF response to superfusion with adenosine (400 µM; Sigma) was also tested. To test the effect of topical application of recombinant (r) ApoE3 or rApoE4 on functional hyperemia, ACh, and adenosine, the cranial window was superfused with a Ringer’s solution containing rApoE4 (Cat# A219, Leinco Technologies) or rApoE3 (Cat# A218, Leinco Technologies) and lipidated rApoE4 or rApoE3 (see lipidation section above). Lipidated or non-lipidated rApoE was reconstituted in PBS and then diluted in normal Ringer’s solution. The CBF response to whisker stimulation, ACh, or adenosine was tested 40 min after rApoE superfusion. In some experiments, the CBF response to whisker stimulation, ACh, or adenosine was tested before and 40 min after superfusion of the cranial window with the ApoE receptors inhibitor receptor-associated protein (RAP, 200 nM; Molecular Innovations), gp91ds-tat (gp91ds, 1 µM; Cat# AS-63818, AnaSpec), or scrambled gp91ds-tat (sgp91ds, 1 µM; Cat# AS-63821, AnaSpec).

Chronic CBF recordings after BCAS: CBF was monitored with laser-speckle imaging (Omegazone; Omegawave). Mice were anesthetized with 1-2% isoflurane, and the scalp was removed to expose the skull. The following day, mice were re-anesthetized and the exposed skull was illuminated by with laser light (780 nm). The scattered light was filtered and detected by a CCD camera positioned over the skull. The raw speckle images were used to compute speckle contrast, which in the mouse neocortex reflects the velocity of moving red blood cells up to a depth of ≈700µm. Color-coded blood flow images were obtained in high-resolution mode (639 x 480 pixels; 1 image/sec) and the sample frequency was 60 Hz. One CBF image was generated by averaging numbers obtained from 20 consecutive raw speckle images. The recordings were initiated after the CBF was stable, and five recordings of blood flow image were averaged. The CBF reduction induced by BCAS was calculated as a percentage of the pre-stenosis CBF value. CBF changes were recorded 2hrs, 24hrs, 2 weeks, and 4 weeks after BCAS.

ROS measurement

ROS production was assessed with 2-photon microscopy in vivo, flow cytometry or in brain slices using dihydroethidine (DHE) as a marker ^{40, 42, 44}.

2-photon microscopy: BAM were labeled with an intracerebroventricular injection of 10 μL of Alexa Fluor 647 dextran (10 000 MW, anionic, fixable, ThermoFisher Scientific, Cat# D22914; 2.5 mg/mL) in PBS as described above ^{40, 42, 44}. The next day, mice were briefly anesthetized with isoflurane (1.5–2%) and injected i.v. with dihydroethidium (10 mg/kg; ThermoFisher; Cat # D11347). One hour later, mice were re-anesthetized with isoflurane (1.5–2%) and equipped with a cranial window superfused with Ringer as described in the CBF experiments. Mice were injected retro-orbitally with fluorescein dextran-conjugated dye (2.5% w/v FITC 70 kDa) diluted in sterile saline (50 μl) to visualize the vasculature ^{43, 45} and imaged under a two-photon microscope (Fluoview FVMPE, Olympus) with a solid-state laser (InSight DS+; Spectra physics) set to an 820 nm wavelength. Image stacks were acquired through Fluoview software (FV31S-SW, v.2.3.1.163, Olympus). A map of the vasculature was taken through an x5 objective (MPlan N 5 x 0.1 NA, Olympus) to identify vessels branching from pial arteries at the cortical surface that feed the barrel area. Once the blood vessels to be imaged were identified, we switched to a 25x objective (XLPlan N 25 x 1.05 NA, Olympus) to identify BAM along the penetrating blood vessels. Then, z-stack images (509 × 509 µm²; 800 × 800 pixels) were acquired at <150 µm depth from the surface of the brain. The superfusion solution was switched from normal Ringer to Ringer containing rApoE3 (10 µg/ml; Cat# A218, Leinco Technologies) or rApoE4 (10 µg/ml; Cat# A219, Leinco Technologies). This concentration as chosen to assure sufficient penetration of ApoE into the neocortex. Forty minutes later, second images were acquired. Images were analyzed using ImageJ software. DHE fluorescence intensity in dextran-positive cells was compared before and after rApoE superfusion.

Flow cytometry: Isolation of brain cells was performed as described ^{43, 81}. Mice were anesthetized with pentobarbital (100 mg/kg, i.p.) and transcardially perfused with heparinized PBS. Brain cell isolation was performed by enzymatic digestion with Liberase DH (Roche Diagnostics) and Dispase (Worthington). Brain hemispheres were separated from the cerebellum and olfactory bulb and gently triturated in HEPES-HBSS buffer containing the following: 138mM NaCl, 5mM KCl, 0.4mM Na₂HPO₄, 0.4mM KH₂PO₄, 5mM d-glucose, and 10mM HEPES using a Gentle MACS dissociator (Miltenyi Biotec) following the manufacturer's instructions. The suspension was digested with 125 μg/ml Liberase, 0.8U/ml dispase, and 50 U/ml DNase I at 37°C for 45 min in an orbital shaker at 100 rpm. Brain cells isolated were washed and subjected to 30% Percoll (GE Healthcare) density gradient centrifugation at 500g for 15 min, and washed with 20 mL 2% FBS in PBS and centrifuged at 500g for 7 min. Following isolation of brain cells, cells were incubated with dihydroethidium (DHE, 2.5mM) in stimulation buffer (RPMI-1640, 10% (v/v) heat inactivated FBS, 100 units/mL penicillin, 100 μg/mL streptomycin) for 30 minutes at 37° and 5% CO₂. Some cells were pooled and separated for stimulation experiments, and were incubated with PBS, human rApoE3 or rApoE4 (10µg/mL) for 30 minutes prior to addition of DHE (as above). Cells were washed with flow cytometry buffer (1X PBS, 2% FBS, 0.05% NaN₃) and centrifuged at 500g for 7 min. For surface marker analysis, 1 × 10⁶ cells approximately were resuspended in 50 µL of flow cytometry buffer. Cells were blocked with anti-CD16/CD32 for 10 min at 4 °C and then stained with the appropriate antibodies for 15 minutes at 4 °C. Cells were washed with flow cytometry buffer, resuspended in 200 µL of flow cytometry buffer and acquired with NovoSampler Q (NovoCyte Quanteon), and absolute cell numbers and frequencies were recorded. Samples were analyzed with FlowJo (Vers.10, Tree Star). Appropriate isotype controls, “fluorescence minus one” staining, and staining of negative populations were used to establish sorting parameters. Endothelial cells were identified as CD45^-Ly6C⁺, microglia were identified as CD45^intCD11b⁺ ^{43, 81}, and BAM were identified as CD45^hiCD11b⁺CD36⁺ ^{43, 81}. Antibodies used were CD16/CD32 (clone 93, rat IgG2b, PRID: AΒ_312800; Biolegend), CD45 (label: BV510, clone 30F-11, rat IgG2b κ, PRID: AΒ_2561392; Biolegend), Ly6C (label: FITC, clone HK1.4, rat IgG2c κ, PRID: AΒ_1186134; Biolegend), CD11b (label: PECy7, clone M1/70, rat IgG2b κ, PRID: AΒ_830641; Biolegend), and CD36 (Label: FITC/PE, clone MF3, rat IgG2a, Cat # MA5-528168/MA5-16832; Invitrogen).

Brain slices: Following euthanasia, the whole brain was quickly removed and submerged into ice-cold (4°C) oxygenated (95% O₂-5% CO₂) sucrose-aCSF (s-aCSF) and sectioned in 200 µm thickness using a vibratome (Leica VT1000). The s-aCSF was composed of the following (in mM): 248 sucrose, 5 KCl, 26 NaHCO₃, 5 MgSO₄, 1 NaH₂PO₄, 0.5 CaCl₂, and 10 glucose, pH=7.35. The lactic acid (l)-aCSF was composed of the following (in mM): 124 NaCl, 5 KCl, 26 NaHCO₃, 2 MgSO₄, 2 CaCl₂, 1 NaH₂PO₄, 10 glucose and lactic acid 4.5, pH=7.35. Brain slices were started to cut from the bregma. Once the slice free-floated, it was transferred to an incubation chamber filled with l-aCSF oxygenated with 95% O₂-5% CO₂. Brain slices were then allowed at least 60 minutes to recover from slicing followed by loading for 45 minutes with the ROS indicator DHE (2 μM) in oxygenated l-aCSF buffer. DHE fluorescent intensity was measured using a Bromide HE filter in Cy-5 labeled BAM along the penetrating vessels in neocortices. Using IPLab software (Scanalytics, Fairfax, VA), time-resolved fluorescent intensity was acquired at 5-min intervals with an exposure time of 100-msec for 20-min using a Nikon Diaphot 300 inverted microscope equipped with a CCD digital camera (Princeton Instruments, Trenton, NJ). BAM ROS (DHE intensity) production was assessed before and after ApoE superfusion and the change was expressed as relative fluorescence units. Data were acquired from N=3-4 mice per group, 1-2 brain slices/mouse, and 3-7 cells/slice.

in situ hybridization

Tissue was prepared for the RNAScope Multiplex Fluorescent Assay according to the manufacturer's instructions (Cat # 323100, ACD, Hayward, CA). The tissue was sectioned (20 µm) in a cryostat and mounted on charged slides. Slides were pretreated with target retrieval buffer (Cat # 322000) boiled for 5 min and treated with Protease III at 40 °C for 30 min (Cat # 322381), followed by incubation with mRNA probes for 2 h at 40 °C. The probes used were Mrc1 (Cat # 437511-C1), ApoE (Cat # 433091-C3), and GFAP (Cat # 486191-C2). A three-plex positive control probe (Cat # 320881) and three-plex negative control (Cat # 320891) were used to ensure the RNAScope positivity. After incubating with the probes, sections were counterstained with the blood vessel marker laminin (anti-rabbit; Cat # ab11575, abcam) followed by incubation with the nuclei marker DAPI (Cat # D1306, Thermo Fisher) and Cy3-conjugated secondary antibody. Then, slides were coverslipped in Prolong Gold Antifade Mountant (Thermo Fisher, Cat # P36930) and, after drying, images were acquired with Leica confocal microscope and analyzed with NIH ImageJ. Some sections were further processed and visualized using Imaris software.

ApoE measurement

ApoE was measured with an electrochemiluminescence- based multi-array method through the Quickplex SQ 120 system (Meso Scale Diagnostics LLC). After the CBF experiment, CSF samples were collected via a cisterna magna puncture. Blood samples were collected in a tube containing sodium citrate (3.8%) through the femoral artery catheter. Sodium citrate was added (1:9 ratio) to the blood samples. Plasma was collected following centrifugation at 2,000 x g for 15 minutes. Brain tissues were harvested following transcardiac perfusion of saline. They were homogenized with a sonicator in cold RAB buffer at 10 µl/mg and centrifuged at 50,000 x g for 20 minutes at 4°C. The supernatant was saved, and the pellet was homogenized using a sonicator in cold RIPA buffer at 10 µl/mg and centrifuged at 50,000 x g for 20 minutes at 4°C. The supernatant was then collected. According to the manufacturer's protocol, human ApoE levels were quantified using the MSD R-PLEX human ApoE assay (Cat# K1512IR-2).

Bone marrow transplantation

Procedures for BM transplantation have been previously described ^{40, 42, 44} and are only summarized. Whole-body irradiation was performed in 10-weeks-old mice (Nordion Gammacell 40 Exactor). Eighteen hours later, mice were transplanted with BM cells (2x10⁶, i.v.) isolated from ApoE3-TR, ApoE4-TR, and WT controls. Mice were housed in cages with sulfamethoxazole (0.12%; w/v) and trimethoprim (0.024%) added to drinking water for the first two weeks. Reconstitution of BM cells was verified 12 weeks after irradiation by testing the positive human ApoE3 and ApoE4 genomic DNA percentage in isolated blood leukocytes. Reference primers sequences were as follows: m_ICAM1_prom.3, 5′-GGACTCACCTGCTGGTCTCT-3′ and m_ICAM1_prom.4, 5′-GAACGAGGGCTTCGGTATTT-3′; target primers sequences were as follows: CD36_1, 5'- -3' and CD36_2, 5'- -3', m_Cybb_gt_1, 5'-CTGCTCACCAGCCTCTCTCTA-3' and m_Cybb_gt_2, 5'-CTGGAACCCCTGAGAAAGGAG-3' (Invitrogen). qRT-PCR was conducted with 20 ng of DNA, in duplicate 15 μl reactions using the Maxima SYBR Green/ROX qPCR Master Mix (2×) (Thermo Scientific). Chimerism was >95% for BM chimeras of ApoE3-TR, ApoE4-TR, and WT controls. A PCR cycling protocol consisting of 15 s at 95°C and 1 min at 60°C for 45 cycles was used for quantification. Human ApoE3 and ApoE4 relative expression levels were calculated by the 2 (−ΔΔ CT) method. To study BAM number and distribution after BM transplant in mice, BM from mice expressing GFP (GFP BM) was transplanted into ApoE3/4-TR mice or WT littermates at 10 weeks of age and the brain distribution of GFP expressing cells was examined at 22 weeks of age.

Bilateral common carotid artery stenosis

As described previously ³³, mice were anesthetized with isoflurane (1-2%) in a mixture of oxygen-nitrogen with rectal temperature maintained at 37°C. Both common carotid arteries were dissected thorough a midline incision and microcoils (internal diameter: 0.18 mm; Sawane Spring, Japan) were placed around the arteries ³³. Sham-treated mice underwent the same surgical procedure with no placement of microcoils.

Evaluation of white matter injury

Four weeks after BCAS, anesthetized mice were perfused transcardially ^{33, 45, 82}. Brains were removed, postfixed overnight, and sectioned with cryostat (thickness of 12 μm) or vibratome (40 µm). For immunohistochemical evaluation of BCAS-induced WM injury the following antibodies were used with the appropriate secondary antibodies: anti-myelin basic protein (MBP) (rat; 1:500; Millipore Sigma), the neurofilament anti-SMI312 (mouse;1:500, Covance), anti-contactin-associated protein (Caspr) (mouse; 1:300, Millipore Sigma) ³³, anti-voltage-gated sodium channel (Na_v) 1.6 (Na_v 1.6) (rabbit; 1:200, Alomone) ³³, or anti-oligodendrocyte transcription factor 2 (Olig2) (rabbit; 1:200; Millipore Sigma). Images were obtained with a confocal laser-scanning microscope (Leica SP5) and analyzed using ImageJ. The Klüver-Barrera stain was performed using the Luxol Fast Blue Stain Kit (ScyTek Laboratory Inc.). Brains were harvested after transcardiac perfusion with PBS and 4% PFA, sectioned with a vibratome (thickness 40 µm), and the positive (blue stained) area in the CC was quantified by ImageJ.

Cognitive testing

Methods for cognitive testing have been described previously ^{33, 40, 45, 82}. We elected to use the Y-maze and novel object recognition because these tests: (a) are sensitive to the cognitive effects of BCAS ^{33, 83-85}, (b) rely on white matter tracts connecting the hippocampus to the cortex and other brain regions ⁸⁶ and, as such, are appropriate in a model of white matter damage, (c) rely on the spontaneous behavior of mice and (c) do not required aversive environments or starving of the mice ^{87, 88}.

Y-maze spontaneous alternation behavior: Mice were placed into one of the arms of the maze (start arm) and allowed to explore only two of the three arms for 5 min (training trial). The closed arm was opened in the test trial, serving as the novel arm. After a 30-min interval between trials, the mice were returned to the same start arm and were allowed to explore all three arms for 5 min (test trial). Sessions were video recorded and analyzed using AnyMaze (San Diego Instruments) in a double-blinded fashion. Spontaneous alternation was evaluated by scoring the order of entries into each arm during the 5 min of the test trial. Spontaneous arm alternation (%) was defined as: number of arm alternations/(total number of arm visits-2) x 100 Spontaneous arm alternation (%) was defined as: number of arm alternations/(total number of arm visits-2) x 100

Novel object recognition: The test was performed in two consecutive days. On day one, mice were placed in the center of an empty open box and allowed to explore for 5 minutes. The box was cleaned with 70% ethanol between trials. On day 2, the mice were placed back to an open box with two identical objects in the center and allowed to explore for 5 minutes. Thirty minutes later, mice were exposed again to a familiar and a novel object and allowed to explore for 5 minutes. The exploring activity (facing, touching or sniffing the object) was monitored and analyzed using AnyMaze in a double-blinded manner, and the percent of the time spent exploring the novel vs. familiar objects was calculated.

Data analysis

Statistical analysis was performed using GraphPad Prism 10 (GraphPad Software, Inc). Histological, cerebrovascular, and behavioral analyses were conducted in a blinded fashion. Animals were randomly assigned to experimental groups. No animals were excluded. The number of mice required for assessing statistical significance of pre-specified effects was estimated by power analysis based on preliminary results and previous experience with the models used in the lab. Two-group comparisons were analyzed by the two-tailed t-test. Multiple comparisons were evaluated by one-way or two-way analysis of variance and Tukey’s test. Differences were considered statistically significant for probability values less than 0.05. Data are expressed as means ± SEM.

Iadecola, C., et al. The Neurovasculome: Key Roles in Brain Health and Cognitive Impairment: A Scientific Statement From the American Heart Association/American Stroke Association. Stroke 54, e251-e271 (2023).
Zlokovic, B.V., et al. Vascular contributions to cognitive impairment and dementia (VCID): A report from the 2018 National Heart, Lung, and Blood Institute and National Institute of Neurological Disorders and Stroke Workshop. Alzheimer's & dementia : the journal of the Alzheimer's Association 16, 1714-1733 (2020).
Iadecola, C., et al. Vascular Cognitive Impairment and Dementia: JACC Scientific Expert Panel. J. Am. Coll. Cardiol. 73, 3326-3344 (2019).
Vemuri, P., et al. White matter abnormalities are key components of cerebrovascular disease impacting cognitive decline. Brain Commun 3, fcab076 (2021).
Gouw, A.A., et al. Heterogeneity of white matter hyperintensities in Alzheimer's disease: post-mortem quantitative MRI and neuropathology. Brain 131, 3286-3298 (2008).
McAleese, K.E., et al. Frontal white matter lesions in Alzheimer's disease are associated with both small vessel disease and AD-associated cortical pathology. Acta Neuropathol. 142, 937-950 (2021).
Phuah, C.L., et al. Association of Data-Driven White Matter Hyperintensity Spatial Signatures With Distinct Cerebral Small Vessel Disease Etiologies. Neurology 99, e2535-2547 (2022).
Soldan, A., et al. White matter hyperintensities and CSF Alzheimer disease biomarkers in preclinical Alzheimer disease. Neurology 94, e950-e960 (2020).
Feekes, J.A., Hsu, S.W., Chaloupka, J.C. & Cassell, M.D. Tertiary microvascular territories define lacunar infarcts in the basal ganglia. Ann. Neurol. 58, 18-30 (2005).
Vogels, V., Dammers, R., van Bilsen, M. & Volovici, V. Deep Cerebral Perforators: Anatomical Distribution and Clinical Symptoms: An Overview. Stroke 52, e660-e674 (2021).
Horsburgh, K., et al. Small vessels, dementia and chronic diseases - molecular mechanisms and pathophysiology. Clin. Sci. (Lond.) 132, 851-868 (2018).
Wardlaw, J.M., Smith, C. & Dichgans, M. Small vessel disease: mechanisms and clinical implications. Lancet Neurol. 18, 684-696 (2019).
Bordes, C., Sargurupremraj, M., Mishra, A. & Debette, S. Genetics of common cerebral small vessel disease. Nat. Rev. Neurol. 18, 84-101 (2022).
El Husseini, N., et al. Cognitive Impairment After Ischemic and Hemorrhagic Stroke: A Scientific Statement From the American Heart Association/American Stroke Association. Stroke 54, e272-e291 (2023).
Pendlebury, S.T., et al. APOE-epsilon4 Genotype and Dementia Before and After Transient Ischemic Attack and Stroke: Population-Based Cohort Study. Stroke 51, 751-758 (2020).
Grinberg, L.T. & Thal, D.R. Vascular pathology in the aged human brain. Acta Neuropathol. 119, 277-290 (2010).
Lamar, M., et al. APOE genotypes as a risk factor for age-dependent accumulation of cerebrovascular disease in older adults. Alzheimer's & dementia : the journal of the Alzheimer's Association 15, 258-266 (2019).
Mirza, S.S., et al. APOE epsilon4, white matter hyperintensities, and cognition in Alzheimer and Lewy body dementia. Neurology 93, e1807-e1819 (2019).
Pinheiro, A., et al. Association of Apolipoprotein E varepsilon4 Allele with Enlarged Perivascular Spaces. Ann. Neurol. 92, 23-31 (2022).
Rojas, S., et al. Higher prevalence of cerebral white matter hyperintensities in homozygous APOE-varepsilon4 allele carriers aged 45-75: Results from the ALFA study. J. Cereb. Blood Flow Metab. 38, 250-261 (2018).
Sudre, C.H., et al. APOE epsilon4 status is associated with white matter hyperintensities volume accumulation rate independent of AD diagnosis. Neurobiol. Aging 53, 67-75 (2017).
Utter, S., et al. Cerebral small vessel disease-induced apolipoprotein E leakage is associated with Alzheimer disease and the accumulation of amyloid beta-protein in perivascular astrocytes. J. Neuropathol. Exp. Neurol. 67, 842-856 (2008).
Salloway, S., et al. Amyloid-Related Imaging Abnormalities in 2 Phase 3 Studies Evaluating Aducanumab in Patients With Early Alzheimer Disease. JAMA neurology 79, 13-21 (2022).
Sperling, R.A., et al. Toward defining the preclinical stages of Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimer's & dementia : the journal of the Alzheimer's Association 7, 280-292 (2011).
van Dyck, C.H., et al. Lecanemab in Early Alzheimer's Disease. N. Engl. J. Med. 388, 9-21 (2023).
Hays, C.C., et al. APOE modifies the interaction of entorhinal cerebral blood flow and cortical thickness on memory function in cognitively normal older adults. Neuroimage 202, 116162 (2019).
Scarmeas, N., et al. APOE related alterations in cerebral activation even at college age. J. Neurol. Neurosurg. Psychiatry 76, 1440-1444 (2005).
Tai, L.M., et al. The role of APOE in cerebrovascular dysfunction. Acta Neuropathol. 131, 709-723 (2016).
Thambisetty, M., Beason-Held, L., An, Y., Kraut, M.A. & Resnick, S.M. APOE epsilon4 genotype and longitudinal changes in cerebral blood flow in normal aging. Arch. Neurol. 67, 93-98 (2010).
Zlokovic, B.V. Cerebrovascular effects of apolipoprotein E: implications for Alzheimer disease. JAMA neurology 70, 440-444 (2013).
Montagne, A., et al. APOE4 leads to blood-brain barrier dysfunction predicting cognitive decline. Nature 581, 71-76 (2020).
Yamazaki, Y., Zhao, N., Caulfield, T.R., Liu, C.C. & Bu, G. Apolipoprotein E and Alzheimer disease: pathobiology and targeting strategies. Nat. Rev. Neurol. 15, 501-518 (2019).
Koizumi, K., et al. Apoepsilon4 disrupts neurovascular regulation and undermines white matter integrity and cognitive function. Nature communications 9, 3816 (2018).
Katusic, Z.S., d'Uscio, L.V. & He, T. Emerging Roles of Endothelial Nitric Oxide in Preservation of Cognitive Health. Stroke 54, 686-696 (2023).
Schaeffer, S. & Iadecola, C. Revisiting the neurovascular unit. Nat. Neurosci. 24, 1198-1209 (2021).
Bell, R.D., et al. Apolipoprotein E controls cerebrovascular integrity via cyclophilin A. Nature 485, 512-516 (2012).
Masuda, T., Amann, L. & Prinz, M. Novel insights into the origin and development of CNS macrophage subsets. Clin Transl Med 12, e1096 (2022).
Masuda, T., et al. Specification of CNS macrophage subsets occurs postnatally in defined niches. Nature 604, 740-748 (2022).
Prinz, M., Masuda, T., Wheeler, M.A. & Quintana, F.J. Microglia and Central Nervous System-Associated Macrophages-From Origin to Disease Modulation. Annu. Rev. Immunol. 39, 251-277 (2021).
Faraco, G., et al. Perivascular macrophages mediate the neurovascular and cognitive dysfunction associated with hypertension. J. Clin. Invest. 126, 4674-4689 (2016).
Lin, X., et al. Perivascular Macrophages Mediate Microvasospasms After Experimental Subarachnoid Hemorrhage. Stroke 54, 2126-2134 (2023).
Park, L., et al. Brain Perivascular Macrophages Initiate the Neurovascular Dysfunction of Alzheimer Abeta Peptides. Circ. Res. 121, 258-269 (2017).
Santisteban, M.M., et al. Endothelium-Macrophage Crosstalk Mediates Blood-Brain Barrier Dysfunction in Hypertension. Hypertension 76, 795-807 (2020).
Uekawa, K., et al. Border-associated macrophages promote cerebral amyloid angiopathy and cognitive impairment through vascular oxidative stress. Res Sq, rs.3.rs-2719812 (2023).
Park, L., et al. Tau induces PSD95-neuronal NOS uncoupling and neurovascular dysfunction independent of neurodegeneration. Nat. Neurosci. 23, 1079-1089 (2020).
Toda, N., Ayajiki, K. & Okamura, T. Cerebral blood flow regulation by nitric oxide: recent advances. Pharmacol. Rev. 61, 62-97 (2009).
Chen, Y., Strickland, M.R., Soranno, A. & Holtzman, D.M. Apolipoprotein E: Structural Insights and Links to Alzheimer Disease Pathogenesis. Neuron 109, 205-221 (2021).
Lanfranco, M.F., Ng, C.A. & Rebeck, G.W. ApoE Lipidation as a Therapeutic Target in Alzheimer's Disease. International journal of molecular sciences 21, 6336 (2020).
LaDu, M.J., et al. Apolipoprotein E and apolipoprotein E receptors modulate A beta-induced glial neuroinflammatory responses. Neurochem. Int. 39, 427-434 (2001).
Rey, F.E., Cifuentes, M.E., Kiarash, A., Quinn, M.T. & Pagano, P.J. Novel competitive inhibitor of NAD(P)H oxidase assembly attenuates vascular O(2)(-) and systolic blood pressure in mice. Circ. Res. 89, 408-414 (2001).
Kazama, K., et al. Angiotensin II impairs neurovascular coupling in neocortex through NADPH oxidase-derived radicals. Circ. Res. 95, 1019-1026 (2004).
Knock, G.A. NADPH oxidase in the vasculature: Expression, regulation and signalling pathways; role in normal cardiovascular physiology and its dysregulation in hypertension. Free Radic. Biol. Med. 145, 385-427 (2019).
Park, L., et al. Nox2-derived radicals contribute to neurovascular and behavioral dysfunction in mice overexpressing the amyloid precursor protein. Proc. Natl. Acad. Sci. U. S. A. 105, 1347-1352 (2008).
Kockx, M.M., et al. Phagocytosis and macrophage activation associated with hemorrhagic microvessels in human atherosclerosis. Arterioscler. Thromb. Vasc. Biol. 23, 440-446 (2003).
Huynh, T.V., et al. Lack of hepatic apoE does not influence early Abeta deposition: observations from a new APOE knock-in model. Mol. Neurodegener. 14, 37 (2019).
Lambertsen, K.L., et al. Differences in origin of reactive microglia in bone marrow chimeric mouse and rat after transient global ischemia. J. Neuropathol. Exp. Neurol. 70, 481-494 (2011).
Morita-Takemura, S., et al. Responses of perivascular macrophages to circulating lipopolysaccharides in the subfornical organ with special reference to endotoxin tolerance. J. Neuroinflammation 16, 39 (2019).
Shemer, A., et al. Engrafted parenchymal brain macrophages differ from microglia in transcriptome, chromatin landscape and response to challenge. Nature communications 9, 5206 (2018).
Ajami, B., Bennett, J.L., Krieger, C., McNagny, K.M. & Rossi, F.M. Infiltrating monocytes trigger EAE progression, but do not contribute to the resident microglia pool. Nat. Neurosci. 14, 1142-1149 (2011).
Diserbo, M., et al. Blood-brain barrier permeability after gamma whole-body irradiation: an in vivo microdialysis study. Can. J. Physiol. Pharmacol. 80, 670-678 (2002).
Tuo, Q.Z., Zou, J.J. & Lei, P. Rodent Models of Vascular Cognitive Impairment. J. Mol. Neurosci. 71, 1-12 (2021).
Hattori, Y., et al. Substantial Reduction of Parenchymal Cerebral Blood Flow in Mice with Bilateral Common Carotid Artery Stenosis. Sci. Rep. 6, 32179 (2016).
Davis, A.A., et al. APOE genotype regulates pathology and disease progression in synucleinopathy. Sci. Transl. Med. 12 (2020).
Holtzman, D.M., et al. Apolipoprotein E isoform-dependent amyloid deposition and neuritic degeneration in a mouse model of Alzheimer's disease. Proc. Natl. Acad. Sci. U. S. A. 97, 2892-2897 (2000).
Wang, C., et al. Selective removal of astrocytic APOE4 strongly protects against tau-mediated neurodegeneration and decreases synaptic phagocytosis by microglia. Neuron 109, 1657-1674 e1657 (2021).
Liu, C.C., et al. Peripheral apoE4 enhances Alzheimer's pathology and impairs cognition by compromising cerebrovascular function. Nat. Neurosci. 25, 1020-1033 (2022).
Yamazaki, Y., et al. Vascular ApoE4 Impairs Behavior by Modulating Gliovascular Function. Neuron 109, 438-447 e436 (2021).
Yamazaki, Y., et al. ApoE (Apolipoprotein E) in Brain Pericytes Regulates Endothelial Function in an Isoform-Dependent Manner by Modulating Basement Membrane Components. Arterioscler. Thromb. Vasc. Biol. 40, 128-144 (2020).
Cummings, J., et al. Lecanemab: Appropriate Use Recommendations. The Journal of Prevention of Alzheimer's Disease (2023).
Percie du Sert, N., et al. The ARRIVE guidelines 2.0: Updated guidelines for reporting animal research. PLoS Biol. 18, e3000410 (2020).
Sullivan, P.M., et al. Targeted replacement of the mouse apolipoprotein E gene with the common human APOE3 allele enhances diet-induced hypercholesterolemia and atherosclerosis. J. Biol. Chem. 272, 17972-17980 (1997).
Zheng, K., et al. Single-cell RNA-seq reveals the transcriptional landscape in ischemic stroke. J. Cereb. Blood Flow Metab. 42, 56-73 (2022).
Stuart, T., et al. Comprehensive Integration of Single-Cell Data. Cell 177, 1888-1902 e1821 (2019).
Korsunsky, I., et al. Fast, sensitive and accurate integration of single-cell data with Harmony. Nature methods 16, 1289-1296 (2019).
Aran, D., et al. Reference-based analysis of lung single-cell sequencing reveals a transitional profibrotic macrophage. Nat. Immunol. 20, 163-172 (2019).
Heng, T.S., Painter, M.W. & Immunological Genome Project, C. The Immunological Genome Project: networks of gene expression in immune cells. Nat. Immunol. 9, 1091-1094 (2008).
Van Hove, H., et al. A single-cell atlas of mouse brain macrophages reveals unique transcriptional identities shaped by ontogeny and tissue environment. Nat. Neurosci. 22, 1021-1035 (2019).
Tabula Muris, C., et al. Single-cell transcriptomics of 20 mouse organs creates a Tabula Muris. Nature 562, 367-372 (2018).
Verghese, P.B., et al. ApoE influences amyloid-beta (Abeta) clearance despite minimal apoE/Abeta association in physiological conditions. Proc. Natl. Acad. Sci. U. S. A. 110, E1807-1816 (2013).
Uekawa, K., et al. Obligatory Role of EP1 Receptors in the Increase in Cerebral Blood Flow Produced by Hypercapnia in the Mice. PLoS One 11, e0163329 (2016).
Garcia-Bonilla, L., et al. Brain and blood single-cell transcriptomics in acute and subacute phases after experimental stroke. bioRxiv, 2023.2003.2031.535150 (2023).
Park, L., et al. tPA Deficiency Underlies Neurovascular Coupling Dysfunction by Amyloid-beta. J. Neurosci. 40, 8160-8173 (2020).
Lee, E.S., et al. A mouse model of subcortical vascular dementia reflecting degeneration of cerebral white matter and microcirculation. J. Cereb. Blood Flow Metab., 271678X17736963 (2017).
Patel, A., et al. Chronic cerebral hypoperfusion induced by bilateral carotid artery stenosis causes selective recognition impairment in adult mice. Neurol. Res. 39, 910-917 (2017).
Toyama, K., et al. Apoptosis signal-regulating kinase 1 is a novel target molecule for cognitive impairment induced by chronic cerebral hypoperfusion. Arterioscler. Thromb. Vasc. Biol. 34, 616-625 (2014).
Aggleton, J.P. & Pearce, J.M. Neural systems underlying episodic memory: insights from animal research. Philos. Trans. R. Soc. Lond. B Biol. Sci. 356, 1467-1482 (2001).
Dellu, F., Contarino, A., Simon, H., Koob, G.F. & Gold, L.H. Genetic differences in response to novelty and spatial memory using a two-trial recognition task in mice. Neurobiol. Learn. Mem. 73, 31-48 (2000).
Vorhees, C.V. & Williams, M.T. Assessing spatial learning and memory in rodents. ILAR journal / National Research Council, Institute of Laboratory Animal Resources 55, 310-332 (2014).

Yes there is potential Competing Interest. D.M.H. is on the scientific advisory board of C2N diagnostics and has equity. D.M.H. is on the scientific advisory board of Denali Therapeutics, Genentech, and Cajal Therapeutics and consults for Asteroid. C.I. is on the scientific advisory board of Broadview Ventures.

ExtendedDataFigures.docx

Download PDF

Journal Publication

published 18 Sep, 2024

Read the published version in Nature Neuroscience →

Version 1

posted

You are reading this latest preprint version

Cell autonomous role of border associated macrophages in ApoE4 neurovascular dysfunction and susceptibility to white matter injury

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Results

Discussion

Declarations

Materials and Methods

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1