Mechanisms of Cerebral Microbleeds

Erythrocyte-brain endothelial interactions induce microglial responses and cerebral microhemorrhages in vivo

BACKGROUND

Cerebral microhemorrhages (CMH) are associated with stroke, cognitive decline, and normal aging. Our previous study shows that the interaction between oxidatively stressed red blood cells (RBC) and cerebral endothelium may underlie CMH development. However, the real-time examination of altered RBC-brain endothelial interactions in vivo, and their relationship with clearance of stalled RBC, microglial responses, and CMH development, has not been reported.

METHODS

RBC were oxidatively stressed using tert-butylhydroperoxide (t-BHP), fluorescently labeled and injected into adult Tie2-GFP mice. In vivo two-photon imaging and ex vivo confocal microscopy were used to evaluate the temporal profile of RBC-brain endothelial interactions associated with oxidatively stressed RBC. Their relationship with microglial activation and CMH was examined with post-mortem histology.

RESULTS

Oxidatively stressed RBC stall significantly and rapidly in cerebral vessels in mice, accompanied by decreased blood flow velocity which recovers at 5 days. Post-mortem histology confirms significantly greater RBC-cerebral endothelial interactions and microglial activation at 24 h after t-BHP-treated RBC injection, which persist at 7 days. Furthermore, significant CMH develop in the absence of blood-brain barrier leakage after t-BHP-RBC injection.

CONCLUSIONS

Our in vivo and ex vivo findings show the stalling and clearance of oxidatively stressed RBC in cerebral capillaries, highlighting the significance of microglial responses and altered RBC-brain endothelial interactions in CMH development. Our study provides novel mechanistic insight into CMH associated with pathological conditions with increased RBC-brain endothelial interactions.

De novo appearance of cerebral microbleeds in community-dwelling older adults. Neuroimaging and clinical correlates

BACKGROUND AND PURPOSE

Prospective studies on cerebral microbleeds (CMB) have departed from individuals who already have CMB at baseline. Therefore, main outcomes have usually been the composite of new lesions appearing on the follow-up among patients who already had CMB together with those who de novo developed CMB. Using the Atahualpa Project Cohort, we aimed to assess correlates of incident CMB in community-dwelling older adults free of CMB at baseline.

METHODS

Atahualpa residents aged ≥ 60 years received baseline clinical interviews and a brain MRI. Those who were free of CMB at baseline and received follow-up brain MRI at the end of the study were included. Multivariate logistic and Poisson regression models were fitted to assess the association and the incidence rate ratio (IRR) of incident CMB according to clinical and neuroimaging variables.

RESULTS

The mean age of 241 study participants was 65.6 ± 6.1 years (57% women). After 6.5 years of follow-up, 25 subjects (10.4%) developed incident CMB. A total of 37 CMB were noticed in these 25 patients. A parsimonious logistic regression model demonstrated an association between the Edmonton Frail Scale (EFS) and incident CMB (p = .043). Multivariate logistic regression models showed an association between WMH progression and incident CMB (p = .011). Using Poisson regression models, the IRR for WMH progression at follow-up was increased in subjects with incident CMB (p = .029).

CONCLUSIONS

Study results show a significant relationship between the EFS, WMH progression, and incident CMB. This knowledge will allow the implementation of preventive policies to reduce incident CMB and its consequences.

Arteriolar neuropathology in cerebral microvascular disease

Cerebral microvascular disease (MVD) is an important cause of vascular cognitive impairment. MVD is heterogeneous in aetiology, ranging from universal ageing to the sporadic (hypertension, sporadic cerebral amyloid angiopathy [CAA] and chronic kidney disease) and the genetic (e.g., familial CAA, cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy [CADASIL] and cerebral autosomal recessive arteriopathy with subcortical infarcts and leukoencephalopathy [CARASIL]). The brain parenchymal consequences of MVD predominantly consist of lacunar infarcts (lacunes), microinfarcts, white matter disease of ageing and microhaemorrhages. MVD is characterised by substantial arteriolar neuropathology involving ubiquitous vascular smooth muscle cell (SMC) abnormalities. Cerebral MVD is characterised by a wide variety of arteriolar injuries but only a limited number of parenchymal manifestations. We reason that the cerebral arteriole plays a dominant role in the pathogenesis of each type of MVD. Perturbations in signalling and function (i.e., changes in proliferation, apoptosis, phenotypic switch and migration of SMC) are prominent in the pathogenesis of cerebral MVD, making 'cerebral angiomyopathy' an appropriate term to describe the spectrum of pathologic abnormalities. The evidence suggests that the cerebral arteriole acts as both source and mediator of parenchymal injury in MVD.

Correlation of lipoprotein-associated phospholipase A2 and cerebral microbleeds in patients with acute ischaemic stroke

BACKGROUND AND AIMS

Cerebral microbleeds (CMBs) increase the risk of stroke occurrence and recurrence,and affect the prognosis of stroke patients. Therefore, identifying biological markers that predict CMBs after stroke is urgently needed. This study explored whether high levels of lipoprotein-associated phospholipase A2(Lp-PLA2) are associated with an increased risk of CMBs in patients with acute ischaemic stroke (AIS).

METHODS

From April 2020 to October 2021, we enrolled 242 patients with AIS. At admission, the plasma levels of Lp-PLA2 were measured in all patients as well as the number of CMBs and white matter lesions. According to the results of the Susceptibility Weighted Imaging (SWI), the patients were divided into a CMB group and a no-CMB group. The groups were compared with univariate and multivariate analyses to clarify the correlation between Lp-PLA2 levels and CMBs, and the optimal cut-off value of Lp-PLA2 that predicted CMBs was determined from the receiver-operating characteristic curve.

RESULTS

CMBs were detected in 71 (29.3%) of the 242 AIS patients. The median Lp-PLA2 level was 182.79 ng/ml. Using the 1st quartile of Lp-PLA2 levels (the lowest levels) as the reference group, univariate logistic regression analysis showed that individuals in the 4th quartile (the highest levels) had a higher risk of CMBs (odds ratio [OR] = 1.460, 95% confidence interval [CI]: 1.188-1.795, P = 0.000). This correlation persisted after adjusting for relevant risk factors (OR = 1.370, 95% CI: 1.096-1.713, P = 0.006). The optimal cut-off value of Lp-PLA2 that predicted the occurrence of CMBs was 184.36 ng/ml; at this threshold, the sensitivity was 69.0%, and the specificity was 60.2%.

CONCLUSIONS

Our data suggest that a high level of Lp-PLA2 in patients with AIS is a potential risk factor for CMBs.

Insights Into the Mechanisms of Brain Endothelial Erythrophagocytosis

The endothelial cells which form the inner cellular lining of the vasculature can act as non-professional phagocytes to ingest and remove emboli and aged/injured red blood cells (RBCs) from circulation. We previously demonstrated an erythrophagocytic phenotype of the brain endothelium for oxidatively stressed RBCs with subsequent migration of iron-rich RBCs and RBC degradation products across the brain endothelium in vivo and in vitro, in the absence of brain endothelium disruption. However, the mechanisms contributing to brain endothelial erythrophagocytosis are not well defined, and herein we elucidate the cellular mechanisms underlying brain endothelial erythrophagocytosis. Murine brain microvascular endothelial cells (bEnd.3 cells) were incubated with tert-butyl hydroperoxide (tBHP, oxidative stressor to induce RBC aging in vitro)- or PBS (control)-treated mouse RBCs. tBHP increased the reactive oxygen species (ROS) formation and phosphatidylserine exposure in RBCs, which were associated with robust brain endothelial erythrophagocytosis. TNFα treatment potentiated the brain endothelial erythrophagocytosis of tBHP-RBCs in vitro. Brain endothelial erythrophagocytosis was significantly reduced by RBC phosphatidylserine cloaking with annexin-V and with RBC-ROS and phosphatidylserine reduction with vitamin C. Brain endothelial erythrophagocytosis did not alter the bEnd.3 viability, and tBHP-RBCs were localized with early and late endosomes. Brain endothelial erythrophagocytosis increased the bEnd.3 total iron pool, abluminal iron levels without causing brain endothelial monolayer disruption, and ferroportin levels. In vivo, intravenous tBHP-RBC injection in aged (17–18 months old) male C57BL/6 mice significantly increased the Prussian blue-positive iron-rich lesion load compared with PBS-RBC-injected mice. In conclusion, RBC phosphatidylserine exposure and ROS are key mediators of brain endothelial erythrophagocytosis, a process which is associated with increased abluminal iron in vitro. tBHP-RBCs result in Prussian blue-positive iron-rich lesions in vivo. Brain endothelial erythrophagocytosis may provide a new route for RBC/RBC degradation product entry into the brain to produce iron-rich cerebral microhemorrhage-like lesions.

Spectroscopic and deep learning-based approaches to identify and quantify cerebral microhemorrhages

Cerebral microhemorrhages (CMHs) are associated with cerebrovascular disease, cognitive impairment, and normal aging. One method to study CMHs is to analyze histological sections (5-40 μm) stained with Prussian blue. Currently, users manually and subjectively identify and quantify Prussian blue-stained regions of interest, which is prone to inter-individual variability and can lead to significant delays in data analysis. To improve this labor-intensive process, we developed and compared three digital pathology approaches to identify and quantify CMHs from Prussian blue-stained brain sections: (1) ratiometric analysis of RGB pixel values, (2) phasor analysis of RGB images, and (3) deep learning using a mask region-based convolutional neural network. We applied these approaches to a preclinical mouse model of inflammation-induced CMHs. One-hundred CMHs were imaged using a 20 × objective and RGB color camera. To determine the ground truth, four users independently annotated Prussian blue-labeled CMHs. The deep learning and ratiometric approaches performed better than the phasor analysis approach compared to the ground truth. The deep learning approach had the most precision of the three methods. The ratiometric approach has the most versatility and maintained accuracy, albeit with less precision. Our data suggest that implementing these methods to analyze CMH images can drastically increase the processing speed while maintaining precision and accuracy.