Introduction
Chronic cerebral hypoperfusion (CCH) and blood–brain barrier (BBB) dysfunction are two significant pathological features of the aging brain [
1‐
3]. Older age is the single most important risk factor for cerebral small vessel disease (CSVD) [
4]. CSVD is one of the most common causes of vascular dementia (VD) [
5]. VD is a neurodegenerative disease that is second only to Alzheimer’s disease (AD) in prevalence [
6]. CSVD imposes a serious burden on the development of society. The pathogenesis of CSVD has not been clearly established. Although several pathological changes, including CCH, BBB impairment, oxidative stress, inflammation and white matter hyperintensities (WMHs), have been shown to be related to CSVD [
7], the cascade of pathological changes that occur in CSVD is still not fully understood. Therefore, we approached this topic by exploring the cellular and molecular mechanisms that regulate the relationship between CCH and BBB function.
The adult brain relies mostly on the continuous influx of glucose from the blood for energy. The rates of glucose and oxygen metabolic decrease in normal aging and are further exacerbated in AD. CCH alters cerebral blood flow (CBF) and brain energy metabolism. Metabolic alterations strongly influence the progression of AD and VD [
8,
9]. CCH has also been suggested to cause of BBB dysfunction and WMHs [
10]. By restricting the free diffusion of circulating toxins or pathogens, the BBB provides a homeostatic brain microenvironment for healthy neural function [
11,
12]. Cross-sectional studies have revealed that CCH is correlated with BBB impairment. CCH is also related to the severity of WMHs [
10]. BBB impairment is more severe in areas near WMHs than in areas of apparently normal white matter (WM) in CSVD [
10,
13]. This indicates that BBB impairment is a key factor linking CCH and WMHs in CSVD.
BBB integrity is maintained by endothelial cell (EC), pericyte, astrocyte, microglia, tight junction (TJ) and extracellular basement membrane (BM) [
14]. BBB components form a complex, dynamic structure, and BBB impairment therefore involves these various components [
15,
16]. The precise responses of all BBB components to CCH have not been thoroughly characterized. Meanwhile, it is unclear whether BBB breakdown is the primary cause of brain parenchyma damage following CCH or secondary to this damage. Furthermore, BBB breakdown leads to inflammation, oxidative stress, neural injury, loss of neuronal connectivity and neurodegeneration [
17]. However, whether ameliorating BBB disruption executes a protective role on neural function following CCH is still poorly understood.
Bilateral common carotid artery occlusion (2VO) is used in rats to mimic the chronic hypoperfusive state of CSVD, and rat subjected to 2VO are used as animal model to assess the mechanisms of CSVD [
18]. Based on the evidence presented above, we first used the 2VO rat model in this study to determine the effects of CCH on changes in BBB permeability and BBB components and brain parenchyma damage. It has been demonstrated that imatinib, a tyrosine kinase inhibitor, counteracts brain oedema, stroke lesion volume and haemorrhagic transformation in rodent stroke model [
19]. Secondly, we observed the BBB protective role of imatinib following CCH. Lastly, we explored the molecular mechanisms that regulate neural injury after BBB breakdown and determined whether BBB impairment is the key pathophysiological mechanism following CCH.
Our results indicate that BBB impairment occurs early in the disease process, preceding neuroinflammatory responses and white matter lesions (WMLs). The mechanism of BBB disruption appears to be pericyte loss. Toxins are able to enter the brain parenchyma due to the increased endothelial transcytosis after BBB impairment. Imatinib can ameliorate BBB disruption via suppression of endothelial transcytosis. Maintenance of BBB integrity via imatinib alleviates oligodendrocyte progenitor cell (OPC) activation, microglial activation, and aberrant TGF-β/Smad2 signaling activation. Imatinib shows neuroprotective function. This study helps explain the effects of BBB injury following CCH and identifies a new potential therapeutic target for BBB integrity maintenance, providing a theoretical basis for the development of targeted treatment strategies.
Material and methods
Animals
Adult male Sprague Dawley rats (weighing 280–300 g, aged 8–12 weeks) selected for this study were housed at a temperature of 24–26 °C on a 12-h light/dark cycle with free access to food and water. Forty-eight rats were used for immunohistochemistry, which was performed to assess serum protein leakage and changes in BBB components in the corpus callosum (CC) at various time intervals following CCH. Thirty rats were injected with Evans blue (EB) dye via the tail vein and used for the assessment of BBB permeability, protein extraction and transmission electron microscopy(TEM). Twenty rats were used to measure CBF. Eighty rats were used to assess BBB protection role of imatinib. Twenty rats were used for TEM after imatinib treatment. All experimental procedures were approved by and performed in accordance with the standards of the Experimental Animal Committee of Henan University and Henan Provincial People’s Hospital.
Establishment of the CCH model
As previously described [
20], CCH was induced in the rats by 2VO. In brief, the surgical procedure was performed under sterile conditions. The rats were anesthetized via intraperitoneal (i.p.) injection of a combination of ketamine (50 mg/kg) and xylazine (10 mg/kg) and placed ventral side up. A 2-cm-long midline incision was made on the ventral side of the cervical neck of each rat. Following careful separation of muscle tissue, nerves and other adjacent tissue, the bilateral common carotid arteries were identified and permanently ligated using a silk ligature. For the sham operation, the same procedure was performed to expose the common carotid arteries, but no ligation was performed. Afterward, the muscle tissue and skin were sutured in layers. Finally, postoperative rats were placed on a warm blanket for recovery.
Measurement of regional CBF
Rats were anesthetized and fixed on a stereotaxic frame (RWD Lifescience, China). Under sterile conditions, a bone window was made at 1-1.5 mm anterior to the bregma and 0-3 mm left to the midline. A laser Doppler flowmetry (Moor Instruments, UK) probe was fixed to the center of the bone window and CBF was recorded 1, 3, 7, and 28 days postoperation.
Measurement of brain water content and BBB permeability
Brain water content and BBB permeability were examined 1, 3, 7, and 28 days postoperation. EB extravasation was used to assess BBB permeability [
21]. In brief, 2% EB (3 mL/kg, Sigma) was injected via the tail vein at various timepoints as indicated. After the EB was allowed to circulate for 2 h, the rats were anesthetized and then perfused transcardially with normal saline solution. The whole brains were collected and divided into the left and right hemispheres. The left hemisphere was used for the measurement of brain water content. The right hemisphere was further cut into 1-mm-thick sections using a stainless steel rat brain matrices (RWD Life Science, China). One section was used for Western blotting, and one section was used for TEM. The other sections were used to assess EB extravasation. To measure brain water content, the left hemisphere was weighed before and after being dehydrated in an oven for 24 h at 100 ℃. The wet/dry brain weight ratio was used to quantify brain water content for statistical analysis. For EB extravasation assessment, sections of the right hemisphere were weighed, homogenized in 1 ml of 50% trichloroacetic acid, and then centrifuged at 10,000 rpm for 30 min. The supernatant was collected and mixed with an equal volume of ethanol. The concentration of EB was determined by spectrophotometry at an absorbance of 620 nm. EB content (μg/g) was calculated according to the standard curve to evaluate BBB permeability.
Histology and immunohistochemistry
At different timepoints after operation, rats were anesthetized and perfused transcardially with 100 ml normal saline solution, followed by 500 ml phosphate-buffered fixative solution composed of 4% paraformaldehyde (PFA, pH 7.4). Next, the brains were removed, postfixed overnight, and finally cryoprotected in phosphate-buffered sucrose (30%) for 3–5 days. Frozen sections (20 µm) were prepared using a cryostat (Leica) and processed for histological examination. Immunohistochemistry staining was performed as previously described [
22,
23]. The following primary antibodies were used: rabbit anti-Olig2 (1:200, Millipore); mouse anti-PCNA (1:200, Invitrogen); rabbit anti-collagen IV (COIV, 1:200, Abcam); chicken anti-albumin (ALB, 1:500, Abcam); mouse anti-immunoglobulin G (IgG, 1:200, Jackson ImmunoResearch); mouse anti-platelet-derived growth factor receptor beta (PDGFR-β, 1:200, Abcam); rabbit anti-desmin (1:100, Cell Signaling Technology); mouse anti-Glut1 (1:200, Abcam); mouse anti-GFAP (1:200, Sigma); rabbit anti-Iba1 (1:300, Wako); rabbit anti-phosphorylated Smad2 (pSmad2, 1:500, Millipore); and mouse anti-myelin basic protein (MBP, 1:200, Biolegend). Alexa Fluor 488- and 594- conjugated goat secondary antibodies (1:500, Thermo Fisher) were also used. Nuclear staining was performed using 4’,6’-diamidino-2-phenylindole dihydrochloride (DAPI, 1:2000, Thermo Fisher). The sections were examined using a confocal laser scanning microscope (TCS SP8, Leica, German).
Western blotting
To determine the change in protein levels in the CC, CC tissue was precisely isolated from the right hemisphere closed to 1.0 mm anterior to the bregma on ice. Once weighed, the tissue was digested in RIPA lysis buffer and homogenized. The protein concentration was quantified, and then the protein were separated on 10% SDS–PAGE gels and transferred to nitrocellulose membranes (Invitrogen, USA). After three times washes in TBS with 0.05% Tween-20 (TBST), the membranes were blocked in TBST with 5% skim milk for 2 h at room temperature. The membranes were incubated with primary antibodies at 4 ℃ overnight and then further incubated with HRP-conjugated secondary antibodies (1:2000) for 1 h at room temperature. The following primary antibodies were used: mouse anti-PDGFR-β (1:1000, Abcam); rabbit anti-TGF-β1 (1:2000, Abcam); rabbit anti-pSmad2 (1:1000, Millipore); rabbit anti-occludin (1:2000, Thermo Fisher); rabbit anti-claudin 5 (1:2000, Thermo Fisher); rabbit anti-ZO-1 (1:1000, Thermo Fisher) and rabbit anti-MBP (1:2000, Abcam). Western Bright ECL solution was used to develop the blots, which were analyzed using GelPro Analyzer 6.0 software (Media Cybernetics, Rockville, MD, USA).
TEM
After the right hemisphere was divided into 1-mm sections, tissue closed to the bregma was selected and further cut into 1 × 1 × 1-mm tissue blocks. The tissue blocks were incubated with 2.5% glutaraldehyde for 6 h, dehydrated and embedded in epoxy resin. Ultrathin sections were cut at 60 nm thickness and observed under an electron microscope (Hitachi TEM system, Japan).
Using stainless steel rat brain matrices (RWD Life Science, China), the whole brains were cut into 1-mm-thick section. The sections closed to 1-1.5 mm anterior to the bregma was selected. The CC was precisely dissected and collected in TRIzol for RNA sequencing. The CC of three rats was collected together as one biological replicates. Three biological replicates were used in the Sham, 1 day, 3 day and 7 day groups respectively.
RNA sequencing and data analysis
Total RNA was extracted in the CC to perform RNA sequencing analysis. The products were sequenced using Illumina HiSeq™ 4000 by Gene Denovo Biotechnology Co. (Guangzhou, China). RNA sequencing raw reads were mapped to the reference genome using HISAT2 [
24] (version 2.1.0). Differentially expressed genes were identified using DESeq2 [
25] based on the criteria that P value < 0.05 and |log
2Fold Change|> 1.
BBB protection
Imatinib inhibits signaling of platelet-derived growth factor receptor (PDGFR) by inducing receptor dimerization via binding to RTK phosphorylation sites [
26]. Imatinib has been found to maintain BBB integrity [
27]. After CCH, rats were administered imatinib (150 mg/kg) by i.p. injections every 12 h for 3 days. The lesion control rats were given normal saline after CCH. After the final injection, rats were euthanized and perfused.
Cell counts
To quantify the number of various cell types in the CC, five fields were randomly chosen from each section and imaged under a confocal microscope (TCS SP8, Leica, German) with a 40 × or 63 × oil immersion objective in 10–12 µm thick z-stacks. A total of five sections were analyzed. Every cell expressing the selected marker was manually counted using Image-Pro Plus 7 (Media Cybernetics, USA). The data are presented as the average cell number in a single field per section.
Quantification of vessel diameter and pericyte coverage
Confocal images were acquired under a 40 × objective. Using Image-Pro Plus 7, the Glut1-positive vessel diameter in each image was measured manually. The COIV-positive brain capillary length and PDGFR-β-positive pericyte length were measured manually using Image-Pro Plus 7. The ratio of PDGFR-β-positive pericyte length to COIV-positive brain capillary length was calculated and coverted to percentage as an indicator of pericyte coverage.
Statistical analysis
Comparisons between multiple group comparisons were made by one-way ANOVA followed by Dunnett’s post hoc test or two-way ANOVA. Data normality was assessed using the Shapiro–Wilk test. The data are presented as the mean ± SD, and the boxplots show the maximum and minimum values. Statistical analysis was performed and graphs were made using GraphPad Prism 8.0 software. Values were considered significant at p < 0.05.
Discussion
CSVD refers to a group of pathological changes with various aetiologies, including age, hypertension, heredity and others [
4]. All these factors can affect CBF [
5]. While it has been found that CCH and BBB dysfunction are the common pathophysiology in different types of CSVD [
10,
39], which BBB component becomes impaired, how neurotoxic molecules are able to enter into the parenchyma, and whether maintenance of BBB integrity can be used as a treatment strategy for CSVD are unclear. WMLs, in contrast, are well understood to be the hallmark of CSVD pathology [
10,
35]. The molecular mechanisms that lead to brain parenchymal damage after BBB breakdown following CCH have not yet been properly examined. In this study, we observed the timeline of both BBB dysfunction and the formation of WMLs following CCH. We revealed that BBB leakage occurs earlier than other pathological events, including OPC activation, mature oligodendrocyte loss, astrocytic activation, and microglial activation, following CCH. Furthermore, we thoroughly examined the components of the BBB. The key change in BBB components following CCH is pericyte loss, which is apparently the leading cause of BBB impairment. Blood-derived pathogens enter into the brain parenchymal through increased endothelial transcytosis rather than endothelial paracellular passage. TGF-β signaling regulates the consequences of BBB breakdown following CCH. Our findings suggest that disease development resulting from CCH unfolds as follows (Fig.
8h). CCH leads to a reduction in pericyte coverage, which induces increased BBB permeability. Following BBB impairment, blood-derived neurotoxic substances enter the brain parenchyma via endothelial endocytosis. Blood-derived neurotoxic substances initiate the inflammatory response, OPC activation and other pathological events by regulating the TGF-β/pSmad2 signaling pathway. Ultimately, homeostasis of oligodendroglial lineage cells is disturbed, resulting in the formation of irreversible WMLs. Furthermore, imatinib treatment ameliorated endothelial endocytosis and brain parenchyma injury. Together, these findings demonstrate that BBB impairment plays a predominant regulatory role in the occurrence of brain damage following CCH, suggesting that BBB compromise is the primary driving factor leading to progressive neural dysfunction. Reversal of BBB dysfunction may be a promising strategy for treating CSVD.
The BBB limits the free diffusion of molecules from the blood into the parenchymal to maintain brain microenvironment homeostasis [
31,
40]. BBB dysfunction contributes to the pathology of many neurological diseases, including TBI [
37], stroke [
41], AD [
17], aging [
42] and CSVD [
10]. Recent studies have shown that BBB dysfunction in the hippocampus occurs earlier than cognitive impairment in AD [
43]. BBB dysfunction is an early biomarker of AD [
43]. On the other hand, during normal aging, WM integrity is maintained after BBB impairment [
44]. These results suggest that BBB dysfunction occurs in the early stage of neurological diseases and that BBB impairment is the key factor leading to brain parenchymal injury. In this study, we found that BBB impairment precedes a series of pathological events, including astrocyte activation, microglial activation, OPC activation and the formation of WMLs. BBB dysfunction is the link between blood-derived pathogens and neural dysfunction.
The cellular components of the BBB include EC and pericyte [
31]. It has been reported that EC dysfunction is the primary cause of BBB dysfunction in stroke-prone spontaneously hypertensive rats [
45]. This rat model is known to be a model of human sporadic CSVD [
46]. On the other hand, Ding et al. analyzed the frontal WM in the postmortem brains from 124 subjects with poststroke dementia (PSD), VD, AD, or AD-VD (mixed), poststroke nondemented (PSND) stroke survivors and normal aging controls [
47]. The researchers found that capillary pericyte loss was a common characteristic among these patients [
47]. Ding et al. indicated that capillary pericyte loss is the structural basis of BBB dysfunction in aging-related dementias [
47]. Furthermore, Bell et al. also found that pericytes control key neurovascular functions in the adult brain and during normal aging [
48]. Our study revealed that CCH does not alter the microvascular number or endothelial TJ expression of claudin-5 and ZO-1. While microvascular length, occludin protein expression, and pericyte coverage are reduced following CCH, microvascular diameter and BM thickness are increased. Pericyte loss may be the leading cause of BBB impairment, and other structural changes may be secondary to pericyte loss following CCH. First, there is a strong positive correlation between pericyte loss and BBB permeability. Second, significant loss of pericytes occurs from 1 day postoperation, i.e., in the early stage of CCH. Most of the other structural changes occur from 3 days postoperation. Third, it has been reported that pericytes play a critical role in maintaining BBB integrity [
27,
29]. Therefore, on the basis of previous studies and our study, we can infer that the structural bases of BBB impairment in a variety of neurodegenerative diseases are somewhat similar. This finding has important implications for the development of new therapeutic strategies.
EC is key plays in direct communication between the blood and the brain parenchyma. EC exhibits two distinctive features in maintaining BBB integrity [
28]. One is a specialized TJ that blocks paracellular passage between the blood and the brain parenchyma [
11]. The other is exhibited unusually low levels of transcytotic vesicles that limit transcellular transport [
49,
50]. Our results show that TJ is not damaged and endothelial endocytosis is increased following CCH. In addition, BBB TJ molecular signatures are not significantly altered. Whereas, transcytosis molecular signatures are significantly increased. This indicates that blood-derived pathogens enter the brain parenchyma through increased endothelial endocytosis after pericyte loss. This finding is consistent with the results of previous research. Using many model of adult viable pericyte loss, Armulik et al. also found that pericytes maintain BBB integrity via transcytosis [
27]. On the other hand, pericyte loss occurs as early as 1 days postoperation. OPC, astrocytic, microglial and TGF-β signaling activation are followed by pericyte loss. Until 28 days postoperation, the homeostasis of oligodendroglial lineage cells are disrupted resulting in irreversible WMLs. It was also demonstrated that CCH induced neuronal cell death was gradually progressed with time in the CA1 subfield of hippocampus [
18]. At 12 weeks following CCH, it exhibited obviously neuronal injury in the CA1 [
20]. Although pericyte loss is a transient process and can be recovered spontaneously following CCH, brain parenchymal injury caused by pericyte loss is persistent.
Although little is known about the regulatory pathways that trigger brain parenchymal injury after BBB breakdown in CSVD, the relevant molecular mechanisms that regulate brain damage in other neurological diseases involving BBB dysfunction can provide some clues. Most studies on the interaction between blood-derived proteins and neural structure have involved AD [
17], TBI [
38] and aging [
42]. The TGF-β signaling pathway regulates progressive neural dysfunction after BBB breakdown in TBI and aging [
38,
42]. However, the brain region of interest in TBI and aging is not the CC. TBI and aging also have different pathologies [
51]. However, although the triggers of mouse models of stroke, multiple sclerosis, TBI and seizure are different, they all involve profound BBB disruption [
52]. Interestingly, EC RNA sequencing revealed similar gene expression changes in EC in these four diseases [
52]. We also examined whether the TGF-β signaling pathway is the mechanism that leads to brain damage following CCH. We found that TGF-β/pSmad2 signaling is promoted after BBB breakdown following CCH. An increase in TGF-β/pSmad2 signaling is responsible for brain damage. Reversal of BBB dysfunction ameliorates TGF-β/pSmad2 signaling activation and brain damage. Therefore, the TGF-β/pSmad2 signaling pathway regulates brain parenchymal injury after BBB breakdown following CCH. These studies also suggest that although BBB dysfunction is triggered by distinct factors in different neurological disorders, the regulatory pathways leading to neurovascular dysfunction after BBB breakdown exhibit similar responses.
BBB disruption is associated with severe brain parenchyma injury following CCH. Increased BBB permeability involves various components [
11]. Thus, currently, there is no well documented therapeutic drugs for BBB integrity maintenance. However, imatinib has shown promising therapeutic candidate drug to restore BBB integrity. Imatinib inhibits several tyrosine kinases and is primarily used to treat chronic myeloid leukemia and gastrointestinal stromal tumor [
53]. Apart from its well documented antitumor applications, imatinib also has been demonstrated its therapeutic effect in a range of neurological diseases including autoimmune encephalomyelitis, ischemic stroke, brain hemorrhage, multiple sclerosis, Parkinson’s disease, AD, Huntington’s disease and spinal cord injury [
53]. The therapeutic target of imatinib is different in these conditions. Furthermore, study has demonstrated imatinib binds to phosphorylation sites on the PDGFR-α and blocks PDGFR-α signaling to maintain BBB integrity in model of ischemic stroke [
19]. In the clinical trial, Wahlgren et al. did not observe any serious adverse events in acute ischemic stroke patients treated with imatinib and imatinib treatment improved neurological and functional outcomes [
54]. Armulik et al. also found imatinib blocks endothelial endocytosis in models of pericyte loss [
27]. In our study, we find imatinib treatment reduces BBB permeability. In addition, imatinib executes a protective role on neural function following CCH. It also should be note that imatinib treatment for BBB protection is very different from antitumor long-term applications. There will be more studies to assess the dose and time requirement for its unusual usage following CCH.
It should be noted that although our findings suggest that a reduction in pericyte coverage leads to BBB dysfunction through increased endothelial transcytosis following CCH, it is not clear why EC, which represent a direct link between blood and neural function, are not the primary cause of BBB dysfunction [
55]. However, BBB dysfunction is related to EC transcytosis after pericyte loss following CCH. Therefore, pericyte and EC are closely linked in BBB function. In future studies, we plan to isolate brain microvessel fragments from the CC and then generate single-cell suspensions for single-cell RNA sequencing. Furthermore, we will use single-cell analysis to study the cause of pericyte loss, the response of EC after pericyte loss and the crosstalk between pericyte and EC. Ultimately, we will analyze the relationship between pericyte and EC in health and disease.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.