From mechanism to classification: Understanding a novel model of cerebral small vessel disease

Abstract

The studies explored cerebral small vessel disease (cSVD), emphasizing the need for precise classification to improve prevention and intervention strategies. Kang et al. introduced an intra-cisterna-magna bevacizumab injection (ICM-BI) model in mice, which induced tight junction loss, microbleeds, and amyloid deposits. However, bevacizumab’s low affinity for murine vascular endothelial growth factor raises questions about its mechanism of action, suggesting potential off-target effects. While most cSVD models mimic arteriolosclerosis (type 1) or genetic variants (types 2 and 3), the ICM-BI model represents a novel approach to studying immune-mediated cSVD (type 4). The complexity and variability of cSVD remain significant research challenges.

Introduction

In recent years, the field of cerebral small vessel disease (cSVD) has gained attention through clinical trials and advances in the understanding of its pathophysiology.^1,2 Cerebrovascular disease classification should consider the mechanism of action and focus on primary and secondary prevention. For instance, diagnosing lacunar infarction is less about specific clinical signs, and more about identifying differences in pathophysiology and interventions compared with other types of stroke. In this context, the current cSVD definition presents several challenges.

Kang et al. ³ raised important, yet debatable, points regarding microvascular alterations. They demonstrated that intra-cisterna-magna bevacizumab injection (ICM-BI) caused loss of tight junctions, cerebral microbleeds, amyloid peptide deposits, and functional impairments resembling cSVD in C57/Bl6 mice. While bevacizumab is designed to target human vascular endothelial growth factor A (VEGF-A) with high specificity, it has low affinity for mouse VEGF-A. Previous studies have used alternative anti-VEGF antibodies to inhibit rodent VEGF after acute ischemic stroke, confirming protein-level inhibition. ⁴ Kang et al. did not report any alterations in VEGF levels after injection. ³ Therefore, the effect of bevacizumab on cSVD, the observed effects could potentially be due to mechanisms unrelated to VEGF-A binding, such as interactions with other molecules, or effects on the blood-brain barrier (BBB) that are not dependent on VEGF-A neutralization (off-target effects). Nonetheless, the cSVD model has the potential to understand cSVD if it can be reliably reproduced.

The definition and classification of cSVD

Kang et al. did not explicitly define cSVD in their study. Their findings showed that although bevacizumab injection did not affect large vessel blood flow (diameter > 30 μm), it reduced the diameter of small blood vessels (<15 μm), a loss of tight junction protein ZO-1, and led to amyloid deposits. Generally, cSVD refers to pathological processes affecting small blood vessels in the brain, including arterioles, venules, and capillaries, with diameters ranging from 20 to 800 μm. ⁵ These vessels, with outer diameters of 20–300 μm, are commonly affected by arteriolosclerosis, a hallmark of cSVD. Vessels under 20 to 30 μm, are mainly capillaries, represent a specific category. Hence, the ICM-BI model aligns more closely with the capillary alteration.

cSVD encompasses various conditions and is primarily defined by its hallmark features on brain MRI, including: 1. white matter hyperintensities, 2. small subcortical infarcts or lacunae, 3. visible perivascular spaces, and 4. microbleeds, 5. intracerebral hemorrhage, and 6. brain atrophy. ¹ The ICM-BI model shows tight junction disruption and microbleeding, suggesting alterations in the BBB.

According to etiological classification, ⁶ cSVD is categorized into six types: type 1, arteriolosclerosis (or age-related and vascular risk-factor-related); type 2, sporadic and hereditary cerebral amyloid angiopathy; type 3, inherited or genetic SVDs such as cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), high-temperature requirement A serine peptidase 1 (HTRA1)-related cSVD, and Fabry’s disease; type 4, inflammatory and immunologically mediated SVDs as systemic lupus erythematosus (SLE) and eosinophilic granulomatosis with polyangiitis; type 5, venous collagenosis; and type 6, other SVDs such as post-radiation angiopathy. Type 1 cSVD is commonly associated with lacunar infarction. cSVD is commonly caused by multiple factors. Animal models of cSVD have been established to evaluate potential therapeutic interventions.^1,5 The spontaneously hypertensive stroke-prone (SHRSP) rat is a well-established model for cSVD. Similarly, rodent models utilizing bilateral common carotid artery stenosis (BCAS) replicate the hypoperfusion aspect of cSVD. These models have been employed to assess treatments aimed at mitigating white matter damage and cognitive deficits resulting from reduced cerebral blood flow, resembling type 1 cSVD. Furthermore, several knockout models correspond to type 2 and 3 cSVD. Most cSVD models are categorized as types 1 through 3.

Inflammatory and immunologically mediated SVD

The ICM-BI model exhibits alterations in the cortex and striatum. Cerebral arterial small vessels have two origins: superficially, they stem from the subarachnoid circulation as the terminal vessels of medium-sized arteries, deeper at the base of the brain, and stem directly from the large vessels as arterial perforators. ⁶ After ischemia, vascular responses are completely different between the cortex and striatum. ⁷ The locations of amyloid deposition also differed. Therefore, the ICM-BI model does not reflect anatomical differences. Although cell therapies that upregulate VEGF-A promote angiogenesis, axonal outgrowth, and functional recovery in ischemic models,^8,9 the effect of bevacizumab on vessels in the ICM-BI model is not dependent on VEGF-A neutralization. In the ICM-BI model, bulk RNA sequencing of brain tissues revealed upregulation of CXCL9 and CXCL10 and interferon-alpha (IFNα), and leukocyte migration. ICM-BI induced an increasing number of IFNγ-positive microglia/monocytes. Excessive IFN activity, as seen in type 4 cSVD (e.g., Aicardi-Goutières syndrome), can drive microangiopathy with BBB leakage (Figure 1). ¹⁰ Most evaluations were performed at 7 days post-injection. That means the responses were acute, but not degenerative and arteriosclerotic. Thus, the ICM-BI model resembles type 4 cSVD, which is a novel model. Generally, cSVD is a chronic disease. Therefore, one limitation of using this acute model is its inability to fully capture the characteristics of a chronic disease.

Figure 1.

Bevacizumab injection induced cerebral small vessel disease in mice. Intra-cisterna-magna bevacizumab injection caused loss of tight junction ZO-1. It reduced the diameter of small blood vessels, cerebral microbleeds, amyloid peptide deposits, and functional impairments resembling cerebral small vessel disease in C57/Bl6 mice. A possible mechanism for this alternation is mediated by interferon and neuroinflammation by injection.

Concluding remarks

The ICM-BI model offers insights into microvascular changes resembling type 4 cSVD. However, the specificity of bevacizumab's action has not been clarified. Despite this progress, the complexity and variability of cSVD remain significant research challenges.

Availability of data and materials

All data and materials are available within the article file on reasonable request.

Consent for publication

We have obtained consent to publish from the participant to report individual data.