Metabolomic signatures associated with pathological angiogenesis in moyamoya disease

Abstract

To the Editor: Our study proposes specific metabolomic changes and biomarkers that can identify different subtypes of moyamoya disease (MMD) for clinical diagnosis. Moreover, LPC supplementation could inhibit pathological angiogenesis, which might be meaningful for new therapeutic target in MMD. MMD is an uncommon cerebrovascular condition where the intracranial internal carotid arteries gradually become narrower, frequently leading to the occurrence of stroke.1 Currently, there are no diagnostic biomarkers for distinguishing among the different subtypes of MMD. Antiplatelet medications are commonly prescribed to prevent the formation of blood clots in narrowing arteries (a potential cause of ischaemic symptoms) in ischaemic MMD. However, these medications may induce a higher risk of bleeding in haemorrhagic MMD.2 Geng et al. have used non-targeted gas chromatography–mass spectrometry (GC–MS) to investigate serum metabolic biomarkers for MMD, including L-isoleucine and urea.3 Lipidomic analysis has suggested that patients with MMD have decreased serum levels of complex membrane glycosphingolipids.4 However, separate subgroup analyses could not be performed due to the small number of identified metabolites, with the small cohort size and lack of additional validation in previous studies. Therefore, we used ultrahigh-performance liquid chromatography high-resolution mass spectrometry (UHPLC-HRMS) untargeted metabolomics analysis to find and identify specific metabolomic changes and biomarkers of different subtypes of MMD. After detailed consultation and physical examination, we collected basic data regarding all patients (Tables S1–S6). The utility of UHPLC-HRMS untargeted metabolomics analysis5 identified 887 and 510 differential metabolite features in positive and negative ion modes with analysis of variance (ANOVA) among ischaemic, haemorrhagic MMD, atherosclerotic stenosis (AS) and health control (HC) groups, respectively. Untargeted metabolomics analysis revealed differential LPC expression between patients with MMD and HC. Specifically, LPC 16:1-2 expression was significantly lower in patients with ischaemic MMD than in HC. The area under the receiver operating characteristic curve (AUROC) (.9675) showed that LPC16:1-2 was a strong candidate biomarker for differentiating patients with ischaemic MMD from HC (Figure 1G). Notably, LPC 16:1-2 level in the positive ion mode was significantly decreased in ischaemic MMD compared to haemorrhagic MMD (Figure S3, Table S14). Differences in metabolites between other subgroups are shown in Tables S7–S18 and Figures S1 and S2. Moreover, we have whole-exome sequencing (WES) with blood of the same 53 MMD participants and the same 20 healthy controls in untargeted metabolomics group. We found that RNF213:c.14429G>A, p.R4810K variant existed in 15 MMD patients, and there was no significant correlation between RNF213, p.R4810K variant and LPC levels shown in Table S19. RNA sequencing (RNA-seq) analysis of superficial temporal artery (STA) samples revealed that PLA2G2A and PLA1A were down-expressed significantly in MMD patients compared with controls shown in Figure S4. And PLA2G expression was significantly lower in the middle cerebral artery (MCA) than in the STA in MMD patients as shown in Table S20. Moreover, LPCAT4 and MFSD2A were all up-expressed in MMD compared with non-MMD patients as shown in Figure S4. Moreover, we assessed the expression of serum LPC-related enzymes with ELISA to identify the results of RNA-seq. And the results of ELISA were consistent with RNA-seq (Figure S5). We constructed a cell model of moyamoya disease using serum stimulation to verify the function of the above metabolites.6 The apoptotic ratios of HBVSMCs in the ischaemic and haemorrhagic subgroups were significantly lower than HC (Figure 2A,B). Cell viability was significantly increased in three MMD groups. In all the MMD subgroups, monocyte chemotactic protein-1 (MCP-1) and NO levels in human brain vascular smooth muscle cells (HBVSMCs) were significantly increased, with the ischaemic subgroup showing the highest level. Levels of reactive oxygen species (ROS) were significantly increased in MMD subgroups. Moreover, the percentage of cells in the S phase was significantly higher in patients with MMD than in HC (Figure 2F,G), indicating increased cell proliferation. Additionally, the percentage of covered area, total tube length and total loops were significantly higher in the MMD groups than in HC (Figure 3 and Figure S8). To identify the effects of LPC level, we constructed cell models with different concentrations of LPC (1, 10, 25, 50 μM). Low LPC concentrations had the similar effects on HBVSMCs and human brain microvascular endothelial cells (HBMECs) to MMD groups shown in Supporting Materials Figures S6 and S7. Compared with before supplementation, supplementation with LPC 16 (Figure 2C) and LPC 22 (Figures S8–S10) significantly decreased cell viability in all three MMD subgroups. Moreover, supplementation with LPC 16 (in Figure 2D,E) and LPC 22 (Figure S8) significantly increased the MCP-1, NO levels and ROS levels in the MMD subgroups (Figure 3A,C). Supplementation with LPC 16 significantly decreased the percentage of cells in S phase in all the MMD subgroups (Figure 2F,G), and also significantly decreased covered area, total tube length and total loops in tube formation assay of MMD subgroups (Figure 3 and Figure S9). LPC, a bioactive lipid generated through pathological processes, is a major constituent of oxidized low-density lipoprotein, which is known to cause inflammation.7 It is suggested to alter the function of cell, including the expression of endothelial adhesion molecules, migration of circulating monocytes, and proliferation and migration of smooth muscle cells, which is involved in cerebral ischaemia and inflammatory diseases.8 The sources of LPC in the peripheral circulation include direct absorption from the diet, phosphatidylcholine cleavage by PLA2 in membranes, and hepatic PLA1 activity on PC.9 Additionally, LPC can disrupt the integrity of cell membranes and hinder the proper functioning of macromolecules within the membrane, leading to cellular damage in vascular smooth muscle cells.10 We found that PLA1 and PLA2 were significantly decreased in MMD, which could lead to low LPC level and decrease its effects. We present here the most detailed metabolomic data on moyamoya disease to date. LPC 16:1 is downregulated in patients with ischaemic moyamoya disease, suggesting that it may be a candidate biomarker to identify different subtypes of moyamoya disease. LPC supplementation could inhibit abnormal cell viability and cell proliferation of HBVSMCs and angiogenesis function of HBMECs, which may offer a novel therapeutic approach for managing moyamoya disease. Not Applicable. The authors thank the patients and controls who agreed to participate in the study. The authors declare they have no conflicts of interest. This research was funded by the National Natural Science Foundation of China (Grant Numbers: 81571110 and 81771234 to Yuanli Zhao; grant 82371296 to Rong Wang). These funds covered the expenses related to testing and processing, data collection, analysis and interpretation of the experiment. The study obtained the consent of the participants and parents of children participants in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of the Beijing Tiantan Hospital (KY 2020-045-02). The authors declare that all supporting data are available within the article, the Supporting Materials. WES data have been submitted to Sequence Read Archive (SRA) with the deposited number as PRJNA987311, and the deposited number of RNA-seq data is PRJNA986272. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts