Introduction
Juvenile idiopathic arthritis (JIA) encompasses a variety of intricate and diverse disorders characterized by persistent inflammation, mainly observed in the synovial membranes and this chronic inflammatory process significantly heightens the risk of degenerative changes occurring in the osteocartilaginous tissues [
1]. The prevalence of JIA is 3.8–400 per 100,000 proportion [
1]. JIA, a condition primarily affecting children under the age of 16, is characterized by symptoms such as swelling, pain, and restricted joint movement that persists for a minimum of 6 weeks, according to research conducted by the International League of Associations for Rheumatology [
2,
3]. In the early stage, there may be severe symptom, including macrophage activation syndrome and synovitis, potentially leading to multi-organ damage [
4]. Cartilage damage and bone erosion may occur as the disease progresses, leading to joint deformities and functional impairment, affecting the quality of life and increasing morbidity.JIA is a condition whose development is thought to be influenced by various factors, such as genetic and environmental factors, as well as infections and it is believed that these factors can trigger inflammatory responses and lead to the onset of autoimmune disorders [
5‐
7]. Nonetheless, the influence of serum metabolites on disease pathogenesis is unclear [
8]. Therefore, the early identification of changes in serum metabolites can help prevent JIA.
Metabolomics has emerged as a burgeoning field that focuses on the detection, analysis, and measurement of naturally occurring small-molecule metabolites in biological samples and this field holds the potential to enhance diagnoses by identifying biomarkers and pathway components, as well as analyzing changes in serum metabolite levels [
9‐
11]. JIA is associated with changes in serum metabolites. Currently, only a limited number of molecular, immune, and clinical markers with JIA have been suggested in these studies [
12,
13]. For instance, calprotectin (also known as MRP8/14 and S100A8/A9) is useful for diagnosing JIA [
14]. In turn, circulating levels of 25-(OH)D appear to have no detectable effect on the incidence of JIA [
15]. In particular, the occurrence of JIA is linked to the activation of endothelial cells, the activation of macrophages, heightened levels of pro-inflammatory cytokines (such as IL-6, IL-10, and IFN-γ), and increased levels of adipokines [
16]. These findings highlight the close connection between lipid profiles, inflammatory responses, and the pathogenesis and progression of autoimmune diseases [
17]. High-mobility group box 1 and matrix metalloproteinase 3 are markers of JIA [
18]. Population-based observational studies have identified various metabolites associated with JIA but are influenced by potential confounding factors or limited by sample size. Therefore, large studies are needed to identify and characterize serum biomarkers that are clinically useful for the early diagnosis of JIA.
The Mendelian randomization method utilizes genetic variants, particularly single nucleotide polymorphisms (SNPs), that are strongly linked to exposure factors, serving as instrumental variables (IVs) and these instrumental variables are then used to estimate the causal effects of exposure factors on health outcomes [
19]. Moreover, MR eliminates potential confounders and reverses causation effects, making it somewhat similar to randomized controlled trials and capable of evaluating genetic correlations across complex diseases [
20]. MR studies based on genome-wide association study (GWAS) datasets utilize genetic variation data as IVs to estimate causal effects. This MR study inferred the causal relationships of 486 serum metabolites (exposure factors) with JIA (outcome), thus providing a basis for identifying JIA biomarkers and metabolic pathways.
Discussion
To investigate the causality between JIA and 486 serum metabolites in humans, this research utilized FinnGen datasets and the GWAS catalog through a two-sample MR approach.The findings suggest that elevated levels of Kyn and linolenate increase the risk of JIA, whereas increased levels of 3-dehydrocarnitine, levulinate, and X-14,208 (phenylalanylserine) protect against this condition. Furthermore, the LDSC analysis indicated no genetic correlation of these metabolites with JIA, demonstrating that the MR analysis was reliable and unaffected by pleiotropy. Moreover, seven metabolic pathways were significantly associated with JIA, including pantothenate and CoA biosynthesis and α-linolenic acid metabolism. JIA is a condition resulting from a combination of genetic and environmental factors that subsequently induces systemic immune reactions, which means timely and precise identification and intervention play a crucial role in enhancing patient outcome [
4]. In the United States, JIA affects about one in a thousand children, being the most prevalent pediatric rheumatic disease and a leading cause of disability acquired in childhood [
36,
37]. This MR research has deepened our understanding of the mechanisms behind JIA, playing a significant role in the disease’s prevention and treatment.
The causal relationship between Kyn and JIA remains unclear.However, previous research has shown that kyn is a significant byproduct of Trp catabolism through tryptophan 2,3-dioxygenase (TDO) or indoleamine 2,3-dioxygenase (IDO) [
38]. The ultimate metabolic product of Kyn is NAD+, which plays a crucial role in immune regulation [
39]. Additionally, under certain physiological conditions, Kyn can be converted into kynurenic acid and xanthurenic acid, both of which are involved in inflammation and immunity in mammals [
40,
41]. The three rate-limiting enzymes of the Trp-Kyn pathway are IDO1, IDO2, and TDO2, with IDO1 promoting inflammation in rheumatoid arthritis (RA) [
42]. Because of its high homology with IDO1, IDO2 may also be implicated in the onset and progression of autoimmune arthritis [
43]. Moreover, a significant increase in serum Kyn levels is associated with chronic low-grade inflammation [
44].JIA is a multifactorial disease with heterogeneous manifestations, including many forms of chronic arthritis [
2]. The levels of Trp-Kyn pathway metabolites are elevated in the serum, urine, and synovial fluid of RA patients [
45]. Moreover, RA is correlated with increased Trp catabolism, increased Kyn concentrations, and immune cell activation in patients and animal models [
46]. These data suggest that the Trp-Kyn metabolic pathway is involved in the pathogenesis of RA. Moreover, there is increased evidence of the role of Kyn metabolites in physiological and disease states. Therefore, increased Kyn levels may be implicated in the pathophysiology of JIA, as demonstrated in this study.
Although the causal relationship between linolenate and JIA is unclear, the findings suggest that linolenate increases the risk of JIA via alpha-linolenic acid metabolism. Research demonstrates that Wuwei Shexiang pill treatment has been linked to a reduction in γ-linolenic acid and other components of the linoleic acid metabolic pathway, suggesting its anti-inflammatory properties by inhibiting linoleic acid metabolism and affecting arachidonic acid metabolism [
47]. Alpha-linolenic acid has multiple biological functions and is involved in endoplasmic reticulum (ER) stress and lipid metabolism. Linolenate induces ER stress by inhibiting the expression of fatty acid synthase, thereby affecting fatty acid synthesis and inflammatory immune responses [
48]. Some drugs can influence the production of inflammatory mediators, including PGE2 and leukotrienes, by regulating alpha-linolenic acid and arachidonic acid metabolism [
49,
50]. These data and our findings suggest that linolenate is implicated in the development and progression of JIA.
We found that three serum metabolites—3-dehydrocarnitine, levulinate, and X-14,208 (phenylalanylserine)—protected against JIA. Nonetheless, little is known about the causal relationship of these metabolites with JIA. Specifically, 3-dehydrocarnitine affects fatty acid metabolism in gout arthritis [
51]. Levulinate has a genetic causal relationship with RA [
52]. Additionally, 5-aminolevulinic acid (5-ALA) has anti-inflammatory and immunomodulatory properties and 5-ALA combined with sodium ferrous citrate (SFC) increases the expression and release of heme oxygenase 1 (HO-1) and its metabolites in macrophages and has been utilized in the treatment of inflammatory diseases [
53‐
55]. These data suggest that the interaction between levulinate and JIA may be mediated by the upregulation of HO-1 by 5-ALA/SFC in macrophages. However, this hypothesis needs to be validated by experimental research. We found that the increased expression of X-14,208 (phenylalanylserine) was associated with a reduced risk of JIA. A study indicates that serum levels of threonine, phenylalanine, and leucine exhibit a positive correlation with the expression of synovial IL-1β and IL-8 in RA patients [
56]. Wine-processed Curculigo orchioides (pCO)’s potential anti-inflammatory actions might be due to its modulation of the phenylalanine metabolic pathway [
57]. Additionally, another study aimed at correlating serum metabolic profiles of RA patients undergoing methotrexate treatment with synovial gene expression discovered associations between serine/glycine/phenylalanine metabolism and aminoacyl-tRNA biosynthesis with TNF-α/CD3E and B cell/plasma related signatures, suggesting a role in lymphocyte regulation within the RA synovium [
58].
Metabolomics, an advanced technology, provides a comprehensive means to explore variations in metabolite levels within biological frameworks, offering invaluable insights into the disruption of metabolic pathways across various diseases [
59]. For instances, a study highlights the pivotal role of metabolomics in revealing that exclusive enteral nutrition can effectively trigger remission in JIA by significantly altering the microbiome and metabolome [
60]. This approach has pinpointed seven critical metabolic pathways associated with JIA, encompassing the biosynthesis of pantothenate and CoA, aminoacyl-tRNA, phenylalanine, tyrosine, and tryptophan, as well as the metabolism of thiamine, taurine and hypotaurine, phenylalanine, and α-linolenic acid. These pathways play essential roles in cellular energy metabolism and the activation of inflammatory cells, highlighting their importance in the pathogenesis of diseases. Employing Ultra Performance Liquid Chromatography-Tandem Mass Spectrometry technology, research has shown that Munziq Balgam, a herbal medicine, can modulate collagen-induced arthritis (CIA) in rat models by affecting these specific pathways, including linoleic acid, alpha-linolenic acid, and the biosynthesis of pantothenate and CoA [
61]. Additionally, Wuwei Shexiang pills have been observed to indirectly affect mitochondrial function and the tricarboxylic acid cycle by altering the synthesis of pantothenic acid and CoA, influencing phenylalanine metabolism [
47]. Further investigations into blood metabolomics of RA rats revealed the impact of Phellodendri Amurensis Cortex, berberine, and palmatine on aminoacyl-tRNA biosynthesis, phenylalanine metabolism, tryptophan metabolism, and the biosynthesis of pantothenic acid and coenzyme A, showcasing anti-RA effect [
62]. Thiamine enhances neurotransmission, muscle function, and immune response in CIA by adjusting metabolism to meet increased energy needs and reduce cellular stress, highlighting the importance of thiamine and arachidonic acid levels in CIA treatment [
63]. Furthermore, research indicates that pCO treatment targeting taurine metabolism can mitigate RA inflammation and bone degradation by modulating anti-inflammatory responses and protecting against oxidative stress [
57]. These studies indicate that targeted regulation of specific metabolic pathways provides a meaningful pathway for the treatment and comprehension of JIA.
This study has strengths. First, IVs and exposure factors were strongly correlated (F > 10). Second, we obtained JIA genetic variation data from multiple sources, conducted several MR analyses, and combined the results of two MR studies to enhance the confidence of MR estimates. Third, MR can eliminate confounding factors and is unaffected by reverse causality. Fourth, the results of MR studies are more robust than those of traditional observational studies. Fifth, we addressed the problem of endogeneity due to reverse causation. To address multiple comparisons, we implemented the Bonferroni correction. The significance level was adjusted to p < 0.00024 (0.05/486) for the analysis.
This study also has limitations. First, factors such as the Beavis effect, compensatory mechanisms (e.g., canalization), low statistical power, and genetic complexity can limit the application of MR studies [
35]. Second, causal inferences were drawn from the results of MR studies and should be further validated through molecular experiment and real-world clinical studies. Third, the SNP data were obtained from European populations, limiting the generalizability of the findings. Fourth, data on age, gender, and other demographic characteristics were unavailable. Fifth, there was heterogeneity in the MR analysis of serum metabolites and JIA. Although heterogeneity was reduced to acceptable levels after removing outliers, the results should be interpreted with caution.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.