Vascular dementia (VaD) is not a specific nosological entity but a syndrome encompassing a number of diseases caused by insufficient blood supply to the brain. It is widely regarded as the second most common type of dementia after Alzheimer’s disease (AD), and there are no licensed treatments for it (Iadecola et al., 2019). Carotid atherosclerotic plaques is a chronic inflammatory disease and might be implicated in VaD risk. A recent study found that subclinical carotid plaque burden is related to reduced brain metabolism (Cortes-Canteli et al., 2021). A meta-analysis of three studies assessing carotid plaque showed an association between carotid plaque and impaired cognitive function (Anbar et al., 2022). However, little is known about the underlying mechanisms responsible for this association.

Previous studies showed that VaD and carotid plaque shared some common risk factors, including diabetes, cholesterol levels, smoking, and a sedentary lifestyle (Baumgart et al., 2015). Bioinformatic analysis of transcriptomes has been instrumental in investigating shared biological mechanisms of comorbid diseases. Chung and Lee (2021) analyzed blood transcriptomic signatures between AD and diabetes mellitus (DM) and revealed common dysregulated gene modules between AD and DM. Several studies conducted transcriptome analysis on VaD and found that immunity and inflammation, oxidative stress, and apoptosis were involved in VaD (Chen et al., 2022; Shu et al., 2022; Wang et al., 2022). Recent studies also found that immune and inflammation process played vital roles in the progression (Dong et al., 2017), instability (Zheng et al., 2022), and rupture (Li et al., 2022) of carotid atherosclerotic plaques based on bioinformatic analysis. However, to our knowledge, few studies have compared transcriptome signatures between VaD and carotid atherosclerotic plaques. Therefore, our study aimed to analyze transcriptome signatures between VaD and carotid atherosclerotic plaques to shed light on the underlying pathology of cognitive decline associated with carotid atherosclerotic plaques and provide new candidate targets for therapeutic intervention.

We analyzed the two datasets GSE122063 and GSE163154 to identify commonly differentially expressed genes (DEGs). Then we performed functional enrichment analysis, immunocyte infiltration and evaluation, receiver operating characteristic (ROC) analysis, and drug-gene analysis to identify underlying common pathophysiological mechanisms and potential targets for the treatment.

In this study, we investigated gene expression patterns from the brain cortex of VaD and atheromatous carotid plaques of symptomatic patients undergoing carotid endarterectomy surgery using a systems biology approach to identify biomolecular signatures that underlie common pathophysiological mechanisms between the two diseases.

We identified 41 overlapped DEGs between the VaD and carotid atherosclerosis plaque datasets. GO enrichment analyses revealed that these overlapped DEGs were mainly enriched in inflammatory and immune-related processes. Previous studies showed that inflammatory and immune-related processes played vital roles in the development of VaD and carotid atherosclerosis plaques (Ammirati et al., 2015; Poh et al., 2021). Our results aligned with previous studies and indicated that inflammatory and immune-related processes might be a crucial common pathophysiological mechanism that VaD and atheromatous carotid plaques shared.

Considering the above functional enrichment analysis results, we further evaluated the infiltration of 22 immune cell types in two datasets. The results revealed that the top five highest infiltrating fractions in carotid plaque and VaD samples differed, indicating that although inflammatory and immune processes were a common pathophysiological mechanism between two diseases, they might be mediated by distinct immune cell types. In carotid atherosclerosis plaque samples, M0 macrophages, M2 macrophages, and T cells gamma delta have a dominant abundance, consistent with a previous study that showed that T cells and macrophages dominated the immune landscape of atherosclerotic plaques (Fernandez et al., 2019). Moreover, M0 macrophages showed a significantly different infiltration percentage between the early and advanced stage plaques group. Plaque macrophages are a functionally heterogeneous group of immune cells, and M0 macrophages were considered non-differentiated macrophages, which could polarize toward pro-inflammatory M1 macrophages or anti-inflammatory M2 macrophages. M1 macrophages could contribute to an increased and sustained inflammatory response via the secretion of pro-inflammatory cytokines and promote the progression of atherosclerotic plaques. In contrast, the M2 macrophages are considered anti-atherogenic by producing anti-inflammatory cytokines and promoting tissue repair (Colin et al., 2014). Therefore, it is necessary to elucidate the molecular mechanisms of macrophage phenotype switching in the future, which might be a potential target for delaying plaque progression.

In VaD samples, resting CD4 memory T cells, M2 macrophages, and naive B cells were the top three highest infiltrating fractions. Furthermore, B cells and NK cells showed a different infiltration percentage between VaD and matched controls. Previous studies have rarely focused on the infiltration of immune cell types in VaD. Busse et al. (2017, 2021) investigated the frequency of innate and adaptive immune cell populations in whole blood and CSF from different forms of dementia, including VaD. They found alterations of innate and adaptive immune cells in the CSF and peripheral immune system. Compared to healthy persons, B cells, T cells, monocytes, and NK cells in whole blood were diminished in VaD, and central memory CD4 + T cells in CSF were reduced, and NK and NKT cells were enhanced in VaD; the number of B lymphocytes was unchanged between the VaD and controls. Together, these results indicated that these immune cell types might be involved in the pathogenesis of VaD, and their role needs to clarify in the future. We identified 12 DEIRGs and investigated the correlation between these DEIRGs and differentially expressed immune cells of the corresponding groups. The results revealed that these DEIRGs were closely related to differentially expressed immune cells, indicating that these DEIRGs might have an essential role in altering the infiltration percentage of these immune cells. Figure 11 briefly summarizes the association between inflammation pathways and the immune system as well as the development of vascular dementia and carotid plaque.

We identified five key DEIRGs based on ROC analysis. The gene HMOX1 also named HO-1, encodes the enzyme that could exert its anti-oxidant and anti-inflammatory effects by several pathways, including degrading the pro-inflammatory heme and producing potent anti-oxidants (Nath et al., 2000). Moreover, as one of the Nrf2 targets, HO-1 had neuroprotective effects. HO-1 knockout worsens infarcts (Shah et al., 2011), and HO-1 overexpression reduces infarcts in mice (Zhang et al., 2012). Mao et al. (2019) induced experimental vascular cognitive impairment in rats by permanent occlusion of both common carotid arteries for 6 weeks. They found sulforaphane could alleviate neuronal damage, white matter injury, blood-brain barrier disruption cognitive impairment by activating the Nrf2/HO-1 pathway. HO-1 also played a protective role in the pathophysiology of atherosclerosis. In APOE-/- mice, a lack of HO-1 accelerated atherosclerosis (Yet et al., 2003), whereas HO-1 induction inhibited atherosclerotic lesion formation in low-density lipoprotein receptor knockout mice (Ishikawa et al., 2001). A study found that plasma HO-1 levels were significantly higher in subjects with carotid plaque than without plaque and were a significant factor for carotid plaque independent of atherosclerotic risk factors (Kishimoto et al., 2018). The protein CD88 encoded by the gene C5AR1 is a classical C5a receptor that has a role in CNS disease processes by binding to C5a. A previous study showed increased expression of CD88 in VaD brain tissue compared to age-matched controls or AD samples by Western blot and immunohistochemistry, which was consistent with our results (Fonseca et al., 2013). Another study showed that treatment with a CD88-specific antagonist enhanced behavioral performance in murine AD models, suggesting that C5a/CD88 interaction had an essential role in the progression of AD (Fonseca et al., 2009). However, the role of CD88 in VaD was not clear. Pavlovski et al. (2012) found that the C5a and CD88 were up-regulated in ischemic neurons and promoted neuronal apoptosis. Pretreatment of the cells with the specific CD88 receptor antagonist PMX53 significantly blocked ischemia-induced apoptosis. They also found that CD88–/– mice subjected to middle cerebral artery occlusion (MCAO) had significantly reduced infarct volumes and improved neurological scores. Stroke is one of the leading causes of VaD; thus, reducing ischemic injury might help prevent or delay the progression of VaD. Accumulating evidence in humans and experimental animal models revealed that the C5a–C5aR axis is involved in the development of atherosclerosis lesions (An et al., 2014). A recent study found that C5aR1 was higher in carotid plaques than in control arteries (Niyonzima et al., 2020), which was in line with our results. The gene CCR1 encodes a member of beta chemokine receptor family that belongs to G protein-coupled receptors (White et al., 2013). The activation of CCR1 was pro-inflammatory in several animal models of neurological diseases. A previous study reported that the expression of CCR1 was increased in the brain after MCAO (Offner et al., 2006) and intracerebral hemorrhage (ICH) in mice (Yan et al., 2020). In the mouse model of ICH, CCR1 activation could promote acute neuroinflammation, and the inhibition of CCR1 improved neurobehavioral deficits and attenuated brain edema and neuroinflammation (Yan et al., 2020). These results indicated that CCR1 might be involved in the pathophysiological mechanism of stroke. There are few studies on the role of CCR1 in atherosclerosis. A study found that blood-borne CCR1 altered the immuno-inflammatory response in atherosclerosis and prevented excessive plaque growth and inflammation (Potteaux et al., 2005). A recent study also found that CCR1 was up-regulated in carotid atherosclerosis, but the exact mechanism was unclear (Meng et al., 2021). The gene SPP1 encodes the protein secreted phosphoprotein 1, also known as osteopontin (OPN). OPN is an extracellular phosphoprotein expressed in various tissues and cells in response to stress and injury and is also a soluble cytokine involved in inflammation. Previous studies revealed that it was upregulated in symptomatic carotid atherosclerosis (Golledge et al., 2004), stroke (Carbone et al., 2015), and vascular cognitive impairment (Chai et al., 2021), suggesting the involvement of OPN in these inflammation-associated neurological diseases, which supported our results. The gene CYBB encodes the protein cytochrome b-245 heavy chain, also named NADPH oxidase 2 (Nox 2), which is one of the NADPH oxidase (Nox) isoforms. It was reported that Nox 2 had been implicated in the VaD and atherosclerotic plaques. Previous studies indicated that Nox2 hereditary deficiency in humans was associated with reduced atherosclerotic burden (Violi et al., 2013). Specific Nox2 inactivation would arrest atheroma plaque progression and instability in animal models (Quesada et al., 2015). Another study revealed that Ling-Yang-Gou-Teng -Decoction, a well-known traditional Chinese formula, exhibited beneficial effects on the VaD by decreasing NOX2 expression (Zhao et al., 2018). Collectively, these results indicated that the five key genes played vital roles in the pathogenesis of VaD and carotid atherosclerotic plaques and might be potential targets for both diseases’ treatment.

We performed a drug-gene interaction analysis to detect potential target drugs/compounds for VaD and carotid atherosclerotic plaques treatment and identified five drugs that were predicted to have specific types of interactions with these key genes. We checked the five candidate drugs in the ClinicalTrials.gov registry. Four drugs (avacopan, CCX354, BMS-817399, and ASK-8007) could be potential drugs for VaD and carotid atherosclerotic plaques treatment. Avacopan was a C5AR1 antagonist, CCX354 and BMS-817399 were CCR1 antagonists, and ASK-8007 was an SPP1 inhibitor. A randomized, placebo-controlled trial showed that C5a receptor inhibition with avacopan was effective in replacing high-dose glucocorticoids in treating antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis. CCX354-C, BMS-817399, and ASK-8007 were all studied to treat Rheumatoid arthritis (RA). ANCA-associated vasculitis and RA were both chronic inflammatory diseases and associated with atherosclerosis and VaD (Atzeni et al., 2017). A systematic review with meta-analysis showed a statistically significant increase in the risk of dementia among patients with RA compared with non-RA controls (Ungprasert et al., 2016). Bilateral carotid plaque was more than twice in RA than in controls (Wah-Suarez et al., 2019). In future studies, these four drugs need to be evaluated as potential drugs in VaD and carotid atherosclerotic plaques treatment.

There are some limitations to the present study. Firstly, the study is only based on bioinformatics analysis; no clinical or experimental confirmation of the roles of key genes was attempted. Secondly, the sample size in this study is not very large, and it will be necessary to replicate our results in more datasets with larger sample size when available. Notwithstanding these limitations, this study offers some insights into the underlying common pathophysiological mechanism shared by VaD and carotid plaques and might provide potential targets for the treatment.

We identified overlapping DEGs between VaD and carotid plaques. These overlapping DEGs were enriched mainly in inflammatory and immune-related processes, indicating it might be a crucial common pathophysiological mechanism shares by VaD and carotid plaques. We identified DEIRGs, evaluated the infiltration of immune cells, and investigated the correlation between DEIRGs and differentially expressed immune cells to gain greater insight into the role of inflammatory and immune-related processes in vascular dementia and carotid plaque progression. We identified five key DEIRGs and performed gene-drug analysis, which might provide potential targets and drugs for therapeutic intervention.

The original contributions presented in this study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.

JS, YR, WT, WW, LZ, and JC conceived and designed the study, analyzed the data, wrote the manuscript, contributed to the designs of the methods used in this study, read, and agreed to the publication of this version of the manuscript.