OI is a prevalent functional cardiovascular disease that affects children and adolescents [1,2,3]. A series of hemodynamic changes, such as hypotension, tachycardia, or bradycardia, and corresponding symptoms (e.g., dizziness and headache, palpitation, and even syncope) occur in OI children after prolonged standing, during emotional stress, or in crowded or unventilated environments22. Several mechanisms have been documented for pediatric OI, including autonomic nerve dysfunction, hypovolemia, and injured vascular endothelium, but the exact pathogenesis of OI remains unclear [12]. Children with OI have relatively higher plasma Hcy levels than normal children, suggesting high plasma Hcy might participate in the pathogenesis of OI [4, 5]. Hcy is an important metabolite of one-carbon metabolism, and elevated Hcy may be accompanied by a variety of metabolic changes. Therefore, to clarify the plasma metabolomic changes in OI children with high Hcy levels and investigate the potential underlying mechanism, we used UHPLC–Q-TOF MS analysis.

We found 105 significantly differential metabolites between the OI group and the control group. Several metabolites related to choline and glutamate changed significantly in the OI group. The upregulated metabolites included choline, PC(18:0/18:2(9Z,12Z)), PC(P-18:0/22:6(4Z,7Z,10Z,13Z,16Z,19Z)), histidine, isocitric acid, and DL-glutamic acid and its downstream metabolites. The downregulated metabolites included LPC(16:0), LPC(18:0), sphingomyelin (d18:1/18:0), betaine aldehyde, hydroxyproline, and GABA. Heatmaps revealed a common metabolic pattern with higher choline, PC(18:0/18:2(9Z,12Z)), and DL-glutamic acid, and lower sphingomyelin (d18:1/18:0), LPC(18:0), and LPC(16:0) in the subset of OI patients we called the special group compared with the common group and normal controls. Clinical indices analysis revealed that the max-uHR, ΔHR, and ΔHR/sHR × 100% of VVS patients were significantly higher in the special group than in the common group (P < 0.05). Choline, PC(18:0/18:2(9Z,12Z)), and DL-glutamic acid were positively correlated with ΔHR/sHR × 100% in VVS patients (P < 0.05).

Choline is an essential micronutrient for the human body, which is mainly derived from dietary intake or glycerophospholipid metabolism [13]. LPC(16:0) and LPC(18:0) are the upstream metabolites of choline in glycerophospholipid metabolism. In this study, choline was upregulated, whereas LPC(16:0) and LPC(18:0) were downregulated in the OI group, indicating that the elevated choline in OI children with high Hcy might be caused by increased dietary intake of it. Like folic acid, choline and betaine, can also be used as methyl group donors for one-carbon metabolism [14]. In mitochondria, choline dehydrogenase catalyzes the conversion of choline to betaine aldehyde, which is then transformed to betaine, thereby providing methyl groups for the remethylation of Hcy in one-carbon metabolism [14, 15]. More choline may be absorbed from the diet in the case of folate acid deficiency or utilization disorders [15,16,17]. Therefore, we suppose that OI children with high levels of Hcy might have folate acid deficiency or utilization disorders, resulting in increased choline intake and elevated plasma choline levels. We also found that betaine aldehyde was downregulated, whereas other downstream products of choline, such as PC(18:0/18:2(9Z,12Z)) and PC(P-18:0/22:6(4Z,7Z,10Z,13Z,16Z,19Z)), were upregulated in the OI group. These results suggested that OI children with high levels of Hcy might have difficulties in converting choline to betaine aldehyde, which affected one-carbon metabolism and increased Hcy levels. Further studies are necessary to confirm these speculations and investigate potential mechanisms.

A number of cardiovascular risk factors, including elevated SBP, increased BMI, insulin resistance, and decreased high-density lipoprotein cholesterol, have been linked to high choline levels and low betaine levels [18]. In Chinese adults with hypertension, Song et al. suggested that the serum choline concentration was positively correlated with all-cause mortality risk. The risk of all-cause death in hypertensive patients with serum choline level ≥ 4.00 µg/mL was 1.79 times higher than that in other hypertensive patients (P < 0.05) [16]. Prospective cohort studies of black people, white Americans, and Chinese adults have shown that high choline intake increased the mortality of individuals with cardiovascular diseases, especially diabetes and ischemic heart disease [19]. The underlying reasons for this risk remain unclear but may partly involve the increased generation of trimethylamine-N-oxide [18, 19]. Studies on choline and its upstream and downstream metabolites in OI are rare. Autonomic nerve dysfunction is a vital mechanism of OI, and the autonomic nuclei that control the autonomic nervous system are mostly located in the dorsal medulla [12, 20]. Using proton magnetic resonance spectroscopy, Wagoner et al. [21] found that OI children had higher levels of total choline in the dorsal medulla while in a supine position and that total choline was negatively correlated with the high-frequency α-index and sequence all. In our study, the plasma choline levels were also higher in the OI group than in the control group. Choline and its downstream product, PC(18:0/18:2(9Z,12Z)), were positively correlated with ΔHR/sHR × 100% in VVS patients. The high-frequency α-index, sequence all, and ΔHR/sHR × 100% are indices that reflect autonomic nerve function and the severity of OI [4, 21]. Therefore, the above studies indicate that high choline levels are associated with autonomic nerve dysfunction and might be involved in the pathogenesis of OI in children.

In the central nervous system (CNS), glutamate acts as the main excitatory neurotransmitter, while GABA acts as the main inhibitory neurotransmitter. The balance between glutamate and GABA is essential for maintaining the normal function of the nervous system. In experimental models, Hcy induced glutamatergic alterations in astrocytes, which was related to the decrease in the activity of Na+/K+-ATPase, an enzyme responsible for glutamate uptake [6]. Hcy also altered glutamate and GABA levels by activating the N-methyl-d-aspartate receptor in the CNS, whereas changes in glutamate and GABA levels differ between different regions of the CNS [22, 23]. In methylenetetrahydrofolate reductase–deficient mice, Jadavji et al. [23] reported that glutamate was elevated in the amygdala, GABA dropped in the thalamus, and both were decreased in the hippocampus. The essential enzyme responsible for converting glutamic acid into GABA is glutamic acid decarboxylase (GAD). After 15 days of l-methionine feeding, there was an increase in S-adenosyl-homocysteine and a decrease in GAD67 mRNA in mouse brain tissues, which might be caused by CpG island hypermethylation in the GAD67 promoter region [24]. Afterward, the downregulation of GAD67 mRNA expression affected the glutamate and GABA levels. The aforementioned studies suggest that high Hcy can cause an imbalance between glutamate and GABA in the CNS.

The glutamate–GABA imbalance has been linked to various neurological disorders in children, such as migraine, depression, and attention deficit and hyperactivity disorder (ADHD) [25,26,27]. Using magnetic resonance spectroscopy (MRS), Bell et al. found that the glutamate–GABA imbalance could occur in the early stage of pediatric migraine. The glutamate concentration in the thalamus and the GABA/Glx (a combination of glutamate and glutamine) ratio in the sensorimotor cortex increased with the duration of migraine [25]. Proton magnetic resonance spectroscopy showed that young patients with depression had notably lower GABA levels in the anterior cingulate cortex compared to healthy controls [26]. The levels of amino acid neurotransmitters in serum are closely related to those in the brain and cerebrospinal fluid. Miniksar et al. [27] found that individuals with ADHD had elevated serum levels of Hcy, glutamate, and GABA levels, which could be used as predictor biomarkers for ADHD.

OI children are often concomitant with neurological disorders, which can be explained by potential shared mechanisms [1, 28, 29]. Wang et al. [30] reported that among 85 VVS children with comorbidities, 14.1% (12/85) had psychological disorders and 3.6% (3/85) had migraine. In individuals with POTS, the proportion of migraine was higher: 50.1% (95% CI 9.9–90.3) in adolescents and 31.2% (95% CI 5.4–57.0) in adult patients [31]. As observed in neurological disorders, plasma DL-glutamic acid was higher and plasma GABA was lower in our OI group than our control group, indicating that the glutamate–GABA imbalance also occurred in OI children with high Hcy. DL-glutamic acid was positively correlated with ΔHR/sHR × 100% in VVS patients (P < 0.05), suggesting that glutamate–GABA imbalance would result in negative effects on OI children. Glutamatergic excitotoxicity or other mechanisms might be involved in this process, requiring further studies [32].

The current research has shortcomings. For instance, the relatively small sample size might have resulted in bias, and OI patients with a normal level of plasma Hcy were not included, which made it unfeasible to investigate the plasma metabolomic changes in OI children with different levels of Hcy. The causes of high plasma Hcy in children with OI, such as deficiencies in folic acid and B-group vitamins, declines in Hcy-metabolizing enzyme function, or mutations in gene-encoding enzymes, were not investigated.