Assessing the influence of plasma metabolites on chronic skin ulcer risk: a two-sample Mendelian randomization study | Scientifi

Assessing the influence of plasma metabolites on chronic skin ulcer risk: a two-sample Mendelian randomization study | Scientific Reports
Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
Advertisement
Assessing the influence of plasma metabolites on chronic skin ulcer risk: a two-sample Mendelian randomization study
Download PDF
Download PDF
Subjects
Diseases
Risk factors
Skin diseases
Abstract
Chronic skin ulcers, although rare, pose severe and debilitating challenges. The identification of causal metabolite biomarkers presents an opportunity to refine effective risk assessment strategies for this condition. In this study, we conducted a comprehensive Two-Sample Mendelian Randomization (TSMR) investigation to delineate the potential causal effects of plasma metabolites on chronic skin ulcer risk. Exposure data comprised 14,296 participants with 913 metabolites from INTERVAL/EPIC-Norfolk, and 8,299 participants with 1,091 metabolites and 309 ratios from the Canadian Longitudinal Study on Aging (CLSA). Outcome data came from the finngen_R9_L12_CHRONICULCEROFSKIN (1,840 cases, 353,088 controls) and UK Biobank Chronic ulcer of skin (495 cases, 455,853 controls) cohorts. Leveraging the inverse-variance weighted (IVW) method, alongside MR-Egger and MR-PRESSO sensitivity analyses, we evaluated metabolite associations with chronic skin ulcer risk. Further assessment involved a phenome-wide MR (Phe-MR) analysis to explore potential repercussions of targeting identified metabolites for intervention. Our study identified 12 distinct metabolites significantly associated with chronic skin ulcers, demonstrating consistent and replicable results. Notably, X-19,141 exhibited the highest reproducibility. These findings highlight novel plasma metabolites relevant to chronic skin ulcers, offering theoretical underpinnings for mechanistic research and clinical strategies in prevention and treatment.
Similar content being viewed by others
Disease modification in inflammatory skin disorders: opportunities and challenges
Article
13 July 2023
Clinical vs. chronological skin age: exploring determinants and
stratum corneum
protein markers of differential skin ageing in 351 healthy women
Article
Open access
09 October 2024
The healing process of diabetic ulcers correlates with changes in the cutaneous microbiota
Article
Open access
12 November 2024
Introduction
Chronic skin ulcers represent a significant medical challenge, characterized by prolonged non-healing wounds that pose substantial burdens on affected individuals
. The intricate pathophysiology underlying these ulcers involves a convergence of factors including compromised local circulation, aberrant immune responses, and various comorbidities
.The genesis of chronic skin ulcers is intricately linked to compromised tissue microenvironments and a dysregulated wound healing cascade
. Impaired blood flow, often associated with conditions such as peripheral vascular disease or diabetes, contributes to tissue ischemia, hindering the delivery of essential nutrients and oxygen, thus impeding the reparative processes
Recent research endeavors have illuminated multifaceted aspects of ulcer formation and healing mechanisms. Advances in biomedical engineering have led to innovative wound dressings and therapeutic devices aimed at fostering tissue regeneration
10
11
12
. Additionally, immunotherapeutic interventions and bioinformatics analyses have shed light on potential molecular targets and prognostic indicators for ulcer severity and response to treatment
13
14
. However, amidst these strides, a critical lacuna exists: the dearth of focus on early detection and intervention strategies for chronic skin ulcers. While late-stage treatments have shown promise, the crucial phase of early identification and preemptive measures remains underexplored, leading to substantial gaps in patient care and outcomes.
This study seeks to address this pivotal research gap by adopting a Two-Sample Mendelian randomization (TSMR) approach, investigating the intricate relationship between plasma metabolites and the onset of chronic skin ulcers. Mendelian Randomization (MR) is a method that, under specific assumptions, aims to estimate causal effects by using genetic variants as instrumental variables
15
16
. This method leverages the principle of random assortment of genes during meiosis to establish causal relationships between modifiable risk factors and disease outcomes, circumventing many of the biases inherent in traditional observational studies
17
18
19
. The TSMR analysis rests on three fundamental assumptions: (1) the genetic variants are strongly associated with the plasma metabolites in question, (2) the genetic variants are independent of any confounding factors that could influence ulcer development, and (3) the genetic variants exert their effect on ulcer risk solely through the plasma metabolites, thus adhering to the exclusion restriction criterion
20
By leveraging Two-Sample Mendelian Randomization (TSMR), this research aims to elucidate potential causal associations between specific plasma metabolites and the early development of chronic skin ulcers. The specific objectives of this study are as follows:
(1)
Identification of Causal Metabolites: To identify plasma metabolites that have a causal effect on the early development of chronic skin ulcers using TSMR. The pre-specified hypothesis is that certain metabolites are directly involved in the pathogenesis of chronic skin ulcers.
(2)
Discovery of Early Predictive Markers: To identify early predictive markers that can be utilized in proactive diagnostic protocols, facilitating early detection and intervention for chronic skin ulcers.
(3)
Development of Targeted Interventions: To identify potential targets for personalized therapeutic strategies aimed at preventing or mitigating the development of chronic skin ulcers.
Through the application of this methodology, the study seeks to provide novel insights into proactive diagnostics and targeted interventions in the domain of chronic ulcerative diseases. This approach has the potential to drive paradigm shifts towards proactive, personalized strategies for early detection and intervention.
Methods
Description of studies to select metabolitegenetic instruments
We utilized GWAS
21
data from two distinct cohorts of European ancestry to select genetic instruments for plasma metabolites. The first cohort comprised 14,296 participants from INTERVAL/EPIC-Norfolk
22
23
, where 913 metabolites were identified post quality control procedures. The Metabolon HD4 platform measured these metabolites, which were transformed and adjusted for age and sex through regression analysis.
In parallel, the second cohort from the Canadian Longitudinal Study on Aging (CLSA) comprised 8,299 participants
24
. Within this cohort, 1,091 plasma metabolites underwent assessment post batch normalization, ensuring the preservation of metabolites despite missing measurements in fewer than 50% of samples among the total 1,458 quantified metabolites. Additionally, 309 metabolite ratios were integrated into our pool of exposure factors
25
. Comprehensive elucidation of the INTERVAL/EPIC-Norfolk and CLSA GWAS methodologies can be found in previously published sources. This investigation culminated in the scrutiny of 1,427 distinct metabolites (Table
S1
).
Instrumental variables for plasma metabolites
We established stringent criteria to select genetic variants as instrumental variables (IVs) for investigating chronic skin ulcers. These criteria encompassed: (1) Significance of SNPs in association with each metabolite at the genome-wide level (INTERVAL/EPIC-Norfolk: P-value < 5 × 10
− 8
; CLSA: P-value < 5 × 10
− 6
26
27
28
. (2) Application of LD-based clumping to filter SNPs, with an r
threshold > 0.1 within a 1000 kb window. (3) Consistency in alleles of selected SNPs between exposure and outcome datasets. (4) Assurance that IVs for exposure were not directly linked to the outcome, maintaining a p-value > 1 × 10
− 5
. (5) Exclusion of IVs with an F-statistic < 10. Additionally, downstream analyses included only metabolites with a minimum of three SNPs as IVs and an r
> 0.001
29
(representing the proportion of metabolite-level variation explained by IVs).
To identify eligible genetic variants as IVs, adherence to three critical assumptions was essential: (1) Association of genetic variants with the exposure (metabolic traits). (2) Absence of associations between genetic variants and any known or unknown confounders. (3) Exclusive influence of genetic variants on the outcome (skin ulcers) through the exposure variable, precluding alternative pathways. To meet assumption (1), SNPs associated with the exposure were sieved at the genome-wide significant threshold. Additionally, weak instruments were excluded using an F-statistic < 10. Assumption (2) was addressed by scrutinizing phenotypes linked to the selected IVs through PhenoScanner2.0. To tackle assumption (3), MR-Egger
30
and MR-PRESSO
31
methods were utilized for evaluating horizontal pleiotropy. Moreover, SNPs directly associated with the outcome at a P-value < 1 × 10
− 5
were excluded. We utilized PhenoScanner2.0
32
to assess if the exposure SNPs we employed were associated with other potential confounders (at P value < 5 × 10
− 8
). Previous studies have suggested a correlation between chronic skin ulcers and conditions like diabetes and autoimmune diseases. We found that rs653178 was associated with autoimmune diseases, whereas rs9420589, rs7570971, rs4665972, rs58542926, rs1260326, and rs780093 were linked to type 2 diabetes. Hence, we removed these SNPs from the exposure instrumental variables.
GWAS dataset for chronic ulcer skin
In assessing associations related to the risk of chronic skin ulcers, we retrieved GWAS data from two sources: the finngen_R9_L12_CHRONICULCEROFSKIN
33
database comprising 1,840 cases and 353,088 controls of European ancestry, and the UK-Biobank Chronic ulcer of skin dataset (phenocode 707,
) including 495 European ancestry cases and 455,853 controls
34
. The study utilized data from a genome-wide genotyping array focusing on the trait of chronic ulcer of skin (PheCode 707). The study was published in the journal Nat Genet with the PubMed ID 34,737,426, authored by Jiang L et al., and the full summary statistics are available for download via FTP(
) ( Table
S2
).
Two-sample MR analysis for potential causalassociations of metabolite levels with chronic ulcer skin
To explore potential causal links between exposures (metabolites) and chronic skin ulcers, a two-sample Mendelian randomization (MR) analysis was conducted employing the Inverse-Variance Weighting
35
(IVW), MR Egger
30
, Weighted median
36
, Weighted mode
37
methods via the “TwoSampleMR“
38
R package. Significance was set at a P-value < 0.05 for MR results. Sensitivity analyses were undertaken to scrutinize horizontal pleiotropy effects and directional pleiotropy using the MR-Egger intercept test and MR Pleiotropy RESidual Sum and Outlier (MR-PRESSO) global test. Ensuring robustness, multiple methods including maximum likelihood, MR-Egger, simple median, and weighted median approaches were employed to validate causal effect directionality. Additionally, the Steiger test
39
40
was utilized to assess the directionality of each SNP’s effect. These comprehensive analyses aimed to provide consistent and reliable insights into the relationships between exposures and chronic skin ulcers (Table
S3
).
Phenome-wide MR(Phe-MR) analysis of phenotypes for identified chronic ulcer skin associated metabolites
We employed two approaches to evaluate the causal relationships between metabolite levels associated with chronic skin ulcers and other outcomes. The first method involved selecting representative instrumental variables associated with metabolite levels to proxy the metabolite levels and assess the causal relationships between these SNPs and disease-related outcomes. Associations between SNPs and outcomes were obtained from the Scalable and Accurate Implementation of the GEneralized mixed model
41
(SAIGE v.0.29)(
). Diseases with fewer than 500 positive samples (cases) were excluded, resulting in 784 diseases for the Phe-MR analysis (Supplement S1.ukb-saige phewas). Significant outcomes were identified through FDR correction (0.05/(17 × 784)).
The second method involved utilizing the selected metabolite-associated instrumental variables for two-sample Mendelian randomization analyses across 50,037 phenotypes from the Open-GWAS
42
database. We retained characteristics with P-values < 0.05 calculated via the IVW method, excluding traits with less than three associated SNPs during the computation process (Supplement S2.Open GWAS Category phewas).
Results
Identifying plasma metabolite risk factors associated with chronic skin ulcers
The overall workflow is depicted in Fig.
. We obtained a total of 1,427 unique quantitative metabolite data and 309 metabolite ratio data from GWAS studies across two large cohorts. To ensure robust conclusions, we employed datasets from UK-Biobank and finn-gen_R9, each encompassing chronic skin ulcer features, as discovery cohorts, respectively. In the INTERVAL/EPIC-Norfolk cohort, a threshold of 5 × 10
− 8
was utilized for tool variable selection, conducting Two Sample Mendelian Randomization (TSMR) analyses for chronic skin ulcer features from the UK-Biobank and Finn-gen cohorts. Within the finngen_R9_L12_CHRONICULCEROFSKIN cohort, 23 statistically significant metabolites associated with chronic skin ulcers were identified (Table
). Among these, 11 were positively correlated and 12 were inversely associated with chronic skin ulcers. In the UK-Biobank Chronic skin ulcer cohort, 21 significant metabolites were identified (Table
), with 12 positively correlated and 9 negatively associated with chronic skin ulcers. Within the CLSA cohort, a threshold of 5 × 10
− 6
was applied for tool variable selection. Through TSMR, 68 significant metabolites related to chronic skin ulcers were identified in the finngen_R9_L12_CHRONICULCEROFSKIN cohort (Table
), with 34 positively and 34 negatively associated. In the UK-Biobank Chronic skin ulcer cohort, 73 significant metabolites were discovered (Table
), comprising 35 positive and 38 negative associations with chronic skin ulcers. For the majority of identified metabolites, horizontal pleiotropy was not a major concern based on MR-Egger intercept (if P
Egger−Intercept
> 0.05) or MR-PRESSO global pleiotropy test (if P
GlobalTest
> 0.05). Sensitivity analyses indicated consistent directions for all associated metabolites.
Fig. 1
The alternative text for this image may have been generated using AI.
Full size image
Workflow
Table 1 Metabolites from the INTERVAL/EPIC-Norfolk cohort as exposure factors, with Finn_R9_L12_Chronic skin ulcer as the outcome factor, screened using the IVW method, resulting in 23 statistically significant metabolite exposures.。.
Full size table
Table 2 Metabolites from the INTERVAL/EPIC-Norfolk cohort as exposure factors, with UKB-Chronic skin ulcer as the outcome factor, screened using the IVW method, resulting in 21 statistically significant metabolite exposures.
Full size table
Table 3 Metabolites from the CLSA cohort as exposure factors, with Finn_R9_L12_Chronic skin ulcer as the outcome factor, screened using the IVW method, resulting in 68 statistically significant metabolite exposures.
Full size table
Table 4 Metabolites from the CLSA cohort as exposure factors, with UKB-Chronic skin ulcer as the outcome factor, screened using the IVW method, resulting in 73 statistically significant metabolite exposures.
Full size table
Compare the results from two independent plasma metabolomics cohorts with two independent chronic skin ulcer cohorts.
Through Two Sample Mendelian Randomization (TSMR) analyses using the Canadian Longitudinal Study on Aging (CLSA) as the exposure factor in two independent chronic skin ulcer cohorts, we identified 137 significantly unique metabolite-associated indicators related to chronic skin ulcers. Among these, four metabolites (GCST90199889: 2R,3R-dihydroxybutyrate levels, GCST90200660: X-25957 levels, GCST90200711: X-19141 levels, GCST90200919: Caffeine to theophylline ratio) exhibited significant correlations (IVW P-value < 0.05) in both independent outcomes (UK-Biobank Chronic skin ulcer, finngen_R9_L12_CHRONICULCEROFSKIN). However, only X-19141 levels showed consistent directionality (finn_gen_R9: OR(95%CI) = 0.90(0.82 ~ 0.98), UK-Biobank: OR(95%CI) = 0.82(0.69 ~ 0.97)), alongside the Caffeine to theophylline ratio (finn_gen_R9: OR(95%CI) = 1.42(1.08 ~ 1.85), UK-Biobank: OR(95%CI) = 1.73(1.04 ~ 2.87). Analyzing metabolites from the INTERVAL/EPIC-Norfolk cohort as exposure factors across two independent datasets, we identified a total of 43 significantly unique metabolite-associated indicators related to chronic skin ulcers. Among these, only one metabolite (erythronate, finn_gen_R9: OR(95%CI) = 1.41(1.05 ~ 1.89), UK-Biobank: OR(95%CI) = 1.85(1.05 ~ 3.26)) displayed a significant impact in both independent outcome cohorts (Fig.
).
Fig. 2
The alternative text for this image may have been generated using AI.
Full size image
Scatter plot illustrating the relationships between SNP effects on X-19,141, Caffeine to theophylline ratio and their impact on chronic ulcer incidence. Each circle represents the marginal genetic association between a specific variant and the risk of chronic ulcer. Error bars depict the 95% confidence intervals. MR refers to mendelian randomization, while IVW stands for inverse-variance weighted. (
) Utilizing FinnGen-R9-L12-Chronic Ulcer as the outcome. (
) Employing GCST90044533 chronic ulcer as the outcome. (
) FinnGene-R9-L12-Chronic Ulcer as the outcome, and the X-19,141(INTERVAL/EPIC-Norfolk) as the exposure. (
) Utilizing Caffeine to theophylline ratio as the exposure, and FinnGen-R9-L12-Chronic Ulcer as the outcome. (
) Employing GCST90044533 chronic ulcer as the outcome and the Caffeine to theophylline ratio as the exposure.
Subsequently, a comprehensive analysis was conducted on all samples exhibiting positive outcomes, filtering those with at least two positive results (IVW) for the exposure outcome. During this process, consistent findings were observed across the CLSA and INTERVAL/EPIC-Norfolk studies. Firstly, the association between Imidazole lactate as an exposure factor and finngen_R9_L12_CHRONICULCEROFSKIN as an outcome remained consistent across both CLSA and INTERVAL/EPIC-Norfolk studies (CLSA: OR(95%CI) = 0.889(0.801 ~ 0.986), INTERVAL/EPIC-Norfolk: OR(95%CI) = 0.876(0.786–0.978)). Similarly, Homocitrulline as another exposure factor demonstrated consistent associations with UK-Biobank Chronic skin ulcer as an outcome in both CLSA and INTERVAL/EPIC-Norfolk studies (CLSA: OR(95%CI) = 1.597(1.13 ~ 2.257), INTERVAL/EPIC-Norfolk: OR(95%CI) = 1.995(1.076 ~ 3.7)). The metabolite 5alpha-androstan-3alpha,17alpha-diol monosulfate, as an outcome, displayed consistent results between CLSA and INTERVAL/EPIC-Norfolk analyses (CLSA: OR(95%CI) = 1.181(1.065 ~ 1.31), INTERVAL/EPIC-Norfolk: OR(95%CI) = 1.16(1.031 ~ 1.306)). Moreover, Methionine sulfone as another outcome showed consistency in the analysis across finngen_R9_L12_CHRONICULCEROFSKIN (CLSA: OR(95%CI) = 0.855(0.757 ~ 0.966), INTERVAL/EPIC-Norfolk: OR(95%CI) = 0.876(0.776 ~ 0.991)). Additionally, Sphingomyelin (d18:1/17:0, d17:1/18:0, d19:1/16:0) and 3-methoxytyrosine as outcomes in finngen_R9_L12_CHRONICULCEROFSKIN also demonstrated consistent results across both CLSA and INTERVAL/EPIC-Norfolk studies (Sphingomyelin: CLSA: OR(95%CI) = 0.794(0.644 ~ 0.979), INTERVAL/EPIC-Norfolk: OR(95%CI) = 0.687(0.502–0.941); 3-methoxytyrosine: CLSA: OR(95%CI) = 0.825(0.689 ~ 0.989), INTERVAL/EPIC-Norfolk: OR(95%CI) = 0.783(0.628 ~ 0.977)) (Table
).
Table 5 Common metabolites across two exposure cohorts (UKB-Chronic ulcer, Finn_R9_L12_Chronic_Ucler) and two outcome cohorts (INTERVAL/EPIC-Norfolk, CLSA).
Full size table
Phe-MR analysis of chronic skin ulcer associated metabolites
To further explore the relationships between the screened metabolites and other diseases or clinical information, we employed two distinct approaches for the Phenome-wide MR (Phe-MR) analysis. In the first method, we utilized the SAIGE database, which encompasses 784 diseases. We performed a Phewas analysis on 17 metabolite markers associated with chronic skin ulcers, including the Caffeine to theophylline ratio, X − 19,141 levels, and others. To ensure the reliability of our findings, we applied a strict filtering criterion, retaining only features with a P value < 0.5/(17 × 784). Through this analysis, we discovered that, in the INTERVAL/EPIC-Norfolk cohort, only the X − 19,141 levels showed a significant promotive effect (effect size > 0, adjusted
< 0.05) in five diseases: Disorders of iron metabolism, Gout, Gout and other crystal arthropathies, Disorders of mineral metabolism, and Celiac disease. This observation implies potential adverse effects of X − 19,141 in these conditions. However, when we examined the effect size of X − 19,141 in the same five diseases within the CLSA cohort, no obvious adverse effects were manifested (see Supplement S1.ukb - saige phewas.xlsx and Figure
S1
for detailed data).
For the second method, we utilized the Open GWAS database, which contains 50,037 phenotype records. Using the selected metabolite - associated instrumental variables, we conducted two - sample Mendelian randomization analyses across these phenotypes. The results revealed that 3 - methoxytyrosine had significant positive effects on several clinical characteristics, such as Height, Other osteochondrodysplasias, and Inflammatory disorders of male genital organs. Conversely, it had significant negative effects on traits like Platelet count, Standard tea intake, and Time spent doing light physical activity. Additionally, 5alpha - androstan − 3alpha,17alpha - diol monosulfate exhibited significant positive effects on clinical traits including CD45RA + CD28 - CD8 + T cell %CD8 + T cell, Central corneal thickness, and Asthma (adult onset), while showing significant negative effects on Breast cancer, Systemic sclerosis (strict definition), and Bipolar disorder. Metabolites erythronate, GCST90199639 (Imidazole lactate levels), and GCST90199682 (Homocitrulline levels) also demonstrated diverse associations with various clinical phenotypes, with both positive and negative effects across different diseases and health - related traits. The detailed results are presented in Supplement S2.
Discussion
Chronic skin ulcer refers to non-healing ulcers persisting on the skin surface for an extended duration, often remaining unhealed for weeks or months
43
. This type of ulcer may be associated with various conditions such as diabetes, venous ulcers, pressure ulcers, and autoimmune disorders
44
45
. In this TSMR study, we comprehensively analyzed the association between plasma metabolite markers and chronic skin ulcers using GWAS summary data from two independent sources for metabolic traits and chronic skin ulcer cohorts. From all cohorts, we identified 17 metabolites exhibiting significant effects in at least two exposure-outcome pairs, indicating significant implications for chronic skin ulcers.
Imidazole lactate (from histidine metabolism) is a compound containing an imidazole ring and serves as a metabolic product of the amino acid histidine, a constituent of proteins
46
. Histidine serves various crucial functions in the human body, including participating in hemoglobin synthesis, regulating acid-base balance, scavenging reactive oxygen and nitrogen compounds, and acting as a precursor to the neurotransmitter histamine
47
48
. Histidine metabolism primarily occurs in the liver and kidneys
49
, where imidazole lactate is an intermediate produced through the catalysis of histidine by imidazole glycerol phosphate synthase (IGPS)
50
. Levels of imidazole lactate can reflect the metabolic status of histidine and may be associated with certain diseases. Elevated levels of imidazole lactate in the plasma of Amyotrophic Lateral Sclerosis (ALS) patients have been observed, potentially linked to neurodegenerative damage and muscle atrophy
51
Homocitrulline, structurally similar to citrulline but with an additional methylene group, originates from lysine reacting with cyanate. This metabolite has garnered attention as a potential antigen confounding rheumatoid arthritis (RA) antibodies targeting citrullinated proteins/peptides. Notably, antibodies binding to sequences containing homocitrulline have been identified in sera from RA patients
52
53
. Recent findings indicate the presence of homocitrulline-containing proteins within RA joints
54
. Patients with rheumatoid arthritis (RA) have been shown to be associated with chronic skin ulcers
55
56
5alpha-Androstane-3alpha,17a-diol, a steroid present in fecal matter of pregnant women
57
, occurs in monosulfate and disulfate fractions in normal human feces
58
, and forms glucuronide, mono-, and disulphate conjugates of neutral steroids in human bile
59
. 5alpha-androstane-3alpha,17alpha-diol monosulfate is a metabolite linked to endurance sports, notably elevated in athletes
60
. Its biological significance extends to hormonal regulation, elucidating its role in assessing athletic performance and hormonal equilibrium in individuals active in endurance-based activities
61
. 3-Methoxytyrosine, or 3-O-Methyldopa, stands as a crucial metabolite stemming from L-DOPA via COMT-mediated O-methylation, an enzymatic process by catechol-O-methyltransferase
62
. This compound holds significance in dopamine metabolism, serving as a precursor to 3-methoxytyramine, a notable dopamine metabolite
63
. As a primary derivative of L-DOPA, its identification proves pivotal in comprehending dopaminergic mechanisms
64
. To summarize, 3-Methoxytyrosine assumes a pivotal role in dopamine metabolism, functioning as a vital intermediate in the synthesis of critical dopamine metabolites.
Kynurenate, arising from L-Tryptophan breakdown, presents a correlation with Vitamin B6 deficiency when detected in urine
65
. This compound exerts control over glutamate and dopamine release, influencing neurological functionalities
66
. Its precursor, kynurenic acid, plays pivotal roles in brain function and is linked to conditions such as schizophrenia and neurodegenerative disorders
67
. Elevated kynurenic acid levels have been identified in patients with tick-borne encephalitis, schizophrenia, and HIV-related illnesses
68
69
70
. Erythronate was shown to be significantly enriched in cancer cells
71
. In our investigations, outcomes from independent TSMR analyses across distinct endpoints consistently delineated a direct causative link between erythronate metabolism and chronic skin ulcers. Notably, the topSNP for Erythronate, rs114030816, is proximal to the gene TKT, which encodes Transketolase (TKT). This enzyme holds a pivotal position in the Pentose Phosphate Pathway (PPP), regulating the generation of fundamental cellular components necessary for proliferation
72
Our research highlights that elevated levels of X-19,141 correlate with a reduced incidence of chronic skin ulcers. In the INTERVAL/EPIC-Norfolk cohort, the SNP rs34728672 proximal to the gene TMPRSS11E corresponds to the elevation of X-19,141. Similarly, in the CLSA cohort, the SNP rs2603187 linked to X-19,141 aligns with TMPRSS11E. TMPRSS11E represents a serine protease. Existing studies substantiate the association between TMPRSS11E gene expression and LPS-induced inflammation, demonstrating significant perturbations in animal models of chronic skin ulcers
73
. Among these 17 metabolites, five—3-methoxytyrosine, 5alpha-androstan-3alpha,17alpha-diol monosulfate, Homocitrulline, Imidazole lactate, Kynurenate, Methionine sulfone, and Sphingomyelin (d18:1/17:0, d17:1/18:0, d19:1/16:0), X-19,141 show significant effects on chronic skin ulcers across two independent metabolite cohorts. However, only X-19,141 exhibits consistent significant effects in three exposure-outcome analyses across all cohorts.
Phe-MR analysis helps explore cross-phenotype associations of the identified metabolites linked to chronic skin ulcers with other phenotypes. Employing two Phewas scanning methods, we initially scanned 784 disease phenotypes from the SAIGE database using the 17 selected metabolites’ instrumental variables. Of these, only X-19,141 appeared to significantly influence five diseases: Disorders of iron metabolism, Gout, Gout and other crystal arthropathies, Disorders of mineral metabolism, and Celiac disease. To comprehensively assess potential side effects of these 17 metabolites, instrumental variables derived from these markers were utilized for Two-Sample Mendelian Randomization (IVW, MR-Egger) against 50,037 phenotypes from the Open GWAS database. Results revealed concordant side effects for most outcomes across both independent metabolite cohorts. For instance, X-19,131 (INTERVAL/EPIC-Norfolk cohort) displayed 603 significant positive associations, with 283 overlapping phenotypes like Colorectal Cancer and LDL cholesterol levels. Most of these associations centered around cholesterol-related traits. In negative associations, X-19,131 (INTERVAL/EPIC-Norfolk cohort) showed 579 significant negative associations, with 236 overlapping phenotypes such as Chloride levels and Disorders of psychological development, indicating broad effects across diverse physiological and behavioral traits. Further exploration is warranted to understand these wide-ranging impacts.
Our study leverages two expansive cohorts boasting high-quality metabolomics data, showcasing a robust exploration into novel metabolite associations with chronic skin ulcers, thereby broadening our comprehension of the genetic underpinnings of this condition. Employing Two-Sample Mendelian Randomization across diverse cohorts for exposure-outcome analyses provided robust causal inference outcomes. Additionally, utilizing GWAS summary data bolstered statistical power in our Two-Sample MR analyses, potentially revealing underlying mechanisms between plasma metabolites and chronic skin ulcers. Furthermore, employing two Phewas methodologies expanded our understanding of the relationships between these chronic skin ulcer-associated instrumental variables and various phenotypes. Such investigations elucidate potential associations of these metabolite markers with diverse clinical traits, unveiling their possible side effects as prospective drugs.
In terms of early diagnosis, our study identified 12 metabolites significantly associated with chronic skin ulcer risk, offering promising biomarkers for early detection. For instance, X-19,141 exhibited consistent and significant associations with chronic skin ulcer risk across different cohorts, with strong reproducibility. This suggests that plasma levels of X-19,141 could be used to identify individuals at high risk for chronic skin ulcers in the early stages of the disease. Integrating X-19,141 into early diagnostic tools, along with existing clinical methods, may enhance early detection rates and provide critical time for subsequent interventions. Additionally, a multi-metabolite diagnostic approach could offer advantages over single-metabolite testing. Several metabolites, such as Imidazole lactate and Homocitrulline, showed strong associations with chronic skin ulcers. By integrating these metabolites, we could develop a multi-metabolite diagnostic model that accounts for interactions and synergistic effects, enabling more accurate risk assessment and reducing misdiagnosis or missed diagnosis. This strategy has shown promise in the diagnosis of other complex diseases and warrants further exploration in chronic skin ulcers.
From a therapeutic and preventive perspective, our findings provide insights into potential interventions targeting metabolites associated with chronic skin ulcer risk. For metabolites positively correlated with disease risk, such as some identified in this study, further research into their role in disease progression could lead to the development of targeted drugs or therapies to reduce their levels and block pathogenic pathways, thereby preventing the development of chronic skin ulcers. Conversely, for metabolites like X-19,141 that are negatively correlated with disease risk, increasing their levels may have preventive potential. Strategies such as drug interventions, nutritional supplements, or dietary adjustments could be considered to enhance the levels of such metabolites, although specific intervention measures require further clinical trials to establish their safety and effectiveness.
A personalized approach to treatment and prevention based on individual metabolite profiles is another important application of our findings. Given the metabolic differences between individuals, which may influence chronic skin ulcer risk and treatment responses, physicians can offer tailored advice based on a patient’s metabolite profile
74
75
. For patients with elevated levels of a particular metabolite, dietary adjustments to reduce its precursor intake could be recommended. For those with low levels of a metabolite, supplementation or drug interventions to promote its synthesis could be considered. This personalized approach could lead to more targeted interventions, improving treatment outcomes and preventing disease development.
However, our study encounters limitations that warrant cautious interpretation. Notably, discrepancies in instrumental variables for the same metabolites derived from two exposure cohorts could stem from varying selection thresholds or differing SNP inclusivity in the original data. Moreover, while X-19,141 consistently showed significant effects in three exposure-outcome TSMR analyses, other exposure factors lacked consistent, significant outcomes across both independent cohorts. Inconsistent IVs across studies may stem from technical variances (e.g., lab protocols, sample processing, or data quality control) and statistical power, influenced by sample sizes (
= 14,296 for INTERVAL/EPIC-Norfolk cohort,
= 8,299 for CLSA cohort). Lastly, our exploration of the impacts of instrumental variables for the 17 metabolite levels using two Phewas methodologies relied solely on topSNPs from the SAIGE database or IVW and MR-Egger analyses within the Open GWAS cohort, requiring further rigorous validation.
Conclusion
In this extensive metabolite MR study, we unveiled associations between 12 independent metabolites and the risk of chronic skin ulcers. Among these, X-19,141 demonstrated consistent associations with the risk of chronic skin ulcers. Further analysis revealed that these 12 metabolites, examined through Phewas studies comprising 17 sets of instrumental variables, exhibit correlations with various diseases or other traits. Our research yields novel insights into the prevention, treatment, and underlying mechanisms of chronic skin ulcers, offering potential clinical applications.
Data availability
All bioinformatics data and analysis codes were sourced from publicly available data queues and data packages. The original experimental data can be obtained by applying to the corresponding author.Summary-level data of SNPs associated with the human metabolome in the INTERVAL/EPIC-Norfolk study were extracted from the OMICSCIENCE website (https://omicscience.org/apps/mgwas/mgwas.table.php). GWAS summary statistics of SNP-metabolite associations from the CLSA study were obtained from the NHGRI-EBI. GWAS Catalog (https://www.ebi.ac.uk/gwas/) with accession number GCST90199621-90201020. The GWAS from UK-Biobank and finn-R9 cohorts. Further information is available from the corresponding author upon request.
References
Darwin, E. & Tomic-Canic, M. Healing chronic wounds: current challenges and potential solutions.
Curr. Dermatol. Rep.
, 296–302 (2018).
Article
Google Scholar
Agale, S. V. Chronic leg ulcers: epidemiology, aetiopathogenesis, and management. Ulcers. (2013).
Gomes, F. et al. The skin Microbiome of infected pressure ulcers: A review and implications for health professionals.
Eur. J. Clin. Invest.
52
e13688. (2022).
Martinengo, L. et al. Prevalence of chronic wounds in the general population: systematic review and meta-analysis of observational studies.
Ann. Epidemiol.
29
, 8–15 (2019).
Article
PubMed
MATH
Google Scholar
Lei, X., Liu, B., Huang, Z. & Wu, J. A clinical study of photodynamic therapy for chronic skin ulcers in lower limbs infected with Pseudomonas aeruginosa.
Arch. Dermatol. Res.
307
, 49–55 (2015).
Article
CAS
PubMed
MATH
Google Scholar
Coelho, G. A., Secretan, P-H., Tortolano, L., Charvet, L. & Yagoubi, N. Evolution of the chronic venous leg ulcer microenvironment and its impact on medical devices and wound care therapies.
J. Clin. Med.
12
, 5605 (2023).
Article
CAS
PubMed
PubMed Central
Google Scholar
Sim, S. L., Kumari, S., Kaur, S. & Khosrotehrani, K. Macrophages in skin wounds: functions and therapeutic potential.
Biomolecules
12
, 1659 (2022).
Article
CAS
PubMed
PubMed Central
MATH
Google Scholar
Baltzis, D., Eleftheriadou, I. & Veves, A. Pathogenesis and treatment of impaired wound healing in diabetes mellitus: new insights.
Adv. Therapy
31
, 817–836 (2014).
Article
CAS
Google Scholar
Jünger, M., Klyscz, T., Hahn, M. & Rassner, G. Disturbed blood flow regulation in venous leg ulcers.
Int. J. Microcirculation
16
, 259–265 (1996).
Article
Google Scholar
Syam, A. F., Sadikin, M., Wanandi, S. I. & Rani, A. A. Molecular mechanism on healing process of peptic ulcer.
Acta Med. Indones
41
, 95–98 (2009).
PubMed
Google Scholar
Rahmani Del Bakhshayesh, A. et al. Recent advances on biomedical applications of scaffolds in wound healing and dermal tissue engineering.
Artif. Cells Nanomed. Biotechnol.
46
, 691–705 (2018).
Article
CAS
PubMed
MATH
Google Scholar
Chouhan, D., Dey, N., Bhardwaj, N. & Mandal, B. B. Emerging and innovative approaches for wound healing and skin regeneration: current status and advances.
Biomaterials
216
, 119267 (2019).
Article
CAS
PubMed
Google Scholar
Heydari, P., Kharaziha, M., Varshosaz, J. & Javanmard, S. H. Current knowledge of Immunomodulation strategies for chronic skin wound repair.
J. Biomedical Mater. Res. Part. B: Appl. Biomaterials
110
, 265–288 (2022).
Article
CAS
Google Scholar
Shalhout, S. Z., Kaufman, H. L., Sullivan, R. J., Lawrence, D. & Miller, D. M. Immune checkpoint Inhibition in Marjolin ulcer: a case series.
J. Immunother.
44
, 234–238 (2021).
Article
CAS
PubMed
Google Scholar
Zhou, F., Li, S. & Xu, H. Insomnia, sleep duration, and risk of anxiety: A two-sample Mendelian randomization study.
J. Psychiatr. Res.
155
, 219–225 (2022).
Article
PubMed
MATH
Google Scholar
Wang, S., Chen, J., Jin, Z., Xing, Y. & Wang, R. Natural hair color and skin cancers: A two-sample Mendelian randomization study.
Gene
893
, 147940 (2024).
Article
CAS
PubMed
Google Scholar
Liu, Z., Mi, J. & Wu, H. Relationships between Circulating metabolites and facial skin aging: a Mendelian randomization study.
Hum. Genomics
17
, 23 (2023).
Article
CAS
PubMed
PubMed Central
Google Scholar
Long, J. et al. Exploring the association between gut microbiota and inflammatory skin diseases: A two-sample Mendelian randomization analysis.
Microorganisms
11
, 2586 (2023).
Article
PubMed
PubMed Central
MATH
Google Scholar
Luo, Q., Cao, Q., Guo, J., Chang, S. & Wu, Y. Genetically predicted levels of circulating cytokines and the risk of six immune skin diseases: a two-sample Mendelian randomization study.
Front. Immunol.
14
, 1240714 (2023).
Article
CAS
PubMed
PubMed Central
Google Scholar
Yao, S. et al. Bidirectional two-sample Mendelian randomization analysis identifies causal associations between relative carbohydrate intake and depression.
Nat. Hum. Behav.
, 1569–1576 (2022).
Article
PubMed
MATH
Google Scholar
Gallagher, M. D. & Chen-Plotkin, A. S. The post-GWAS era: from association to function.
Am. J. Hum. Genet.
102
, 717–730 (2018).
Article
CAS
PubMed
PubMed Central
MATH
Google Scholar
Zhong, H. et al. Elucidating the role of blood metabolites on pancreatic cancer risk using two-sample Mendelian randomization analysis.
Int. J. Cancer
(2023).
Littlejohns, T. J. et al. Visual impairment and risk of dementia in 2 population-based prospective cohorts: UK biobank and EPIC-Norfolk.
J. Gerontol. Ser. A
77
, 697–704 (2022).
CAS
Google Scholar
Raina, P. S. et al. The Canadian longitudinal study on aging (CLSA).
Can. J. Aging/La Revue Canadienne Du Vieillissement
28
, 221–229 (2009).
Article
MATH
Google Scholar
Raina, P. et al. Cohort profile: the Canadian longitudinal study on aging (CLSA).
Int. J. Epidemiol.
48
, 1752–1753j (2019).
Article
PubMed
PubMed Central
MATH
Google Scholar
Wu, L., Zhu, J. & Wu, C. Mendelian randomization analysis to characterize causal association between coronary artery disease and COVID-19.
MedRxiv
2020
, 2005–2029. 20117309 (2020).
MATH
Google Scholar
Ghoneim, D. H. et al. Mendelian randomization analysis of n-6 polyunsaturated fatty acid levels and pancreatic cancer risk.
Cancer Epidemiol. Biomarkers Prev.
29
, 2735–2739 (2020).
Article
CAS
PubMed
PubMed Central
MATH
Google Scholar
Shu, X. et al. Associations of obesity and Circulating insulin and glucose with breast cancer risk: a Mendelian randomization analysis.
Int. J. Epidemiol.
48
, 795–806 (2019).
Article
PubMed
MATH
Google Scholar
Yun, Z. et al. Genetically predicted 486 blood metabolites in relation to risk of colorectal cancer: A Mendelian randomization study.
Cancer Med.
(2023).
Burgess, S. & Thompson, S. G. Interpreting findings from Mendelian randomization using the MR-Egger method.
Eur. J. Epidemiol.
32
, 377–389 (2017).
Article
PubMed
PubMed Central
MATH
Google Scholar
Verbanck, M., Chen, C-Y., Neale, B. & Do, R. Detection of widespread horizontal Pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases.
Nat. Genet.
50
, 693–698 (2018).
Article
CAS
PubMed
PubMed Central
Google Scholar
Kamat, M. A. et al. PhenoScanner V2: an expanded tool for searching human genotype–phenotype associations.
Bioinformatics
35
, 4851–4853 (2019).
Article
MathSciNet
CAS
PubMed
PubMed Central
MATH
Google Scholar
Kurki, M. I. et al. FinnGen: unique genetic insights from combining isolated population and National health register data.
MedRxiv
, (2022). 2022.2003. 2003.22271360.
Bycroft, C. et al. The UK biobank resource with deep phenotyping and genomic data.
Nature
562
, 203–209 (2018).
Article
ADS
CAS
PubMed
PubMed Central
MATH
Google Scholar
Marín-Martínez, F. & Sánchez-Meca, J. Weighting by inverse variance or by sample size in random-effects meta-analysis.
Educ. Psychol. Meas.
70
, 56–73 (2010).
Article
MATH
Google Scholar
Bowden, J., Davey Smith, G., Haycock, P. C. & Burgess, S. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator.
Genet. Epidemiol.
40
, 304–314 (2016).
Article
PubMed
PubMed Central
Google Scholar
Wang, R. Mendelian randomization study updates the effect of 25-hydroxyvitamin D levels on the risk of multiple sclerosis.
J. Translational Med.
20
, 1–10 (2022).
Google Scholar
Hartwig, F. P., Davies, N. M., Hemani, G. & Davey Smith, G.
Two-sample Mendelian Randomization: Avoiding the Downsides of a Powerful, Widely Applicable but Potentially Fallible Technique
1717–1726 (Oxford University Press, 2016).
Wang, K. & Han, S. Effect of selection bias on two sample summary data based Mendelian randomization.
Sci. Rep.
11
, 7585 (2021).
Article
ADS
CAS
PubMed
PubMed Central
MATH
Google Scholar
Lutz, S. M., Wu, A. C., Hokanson, J. E., Vansteelandt, S. & Lange, C. Caution against examining the role of reverse causality in Mendelian randomization.
Genet. Epidemiol.
45
, 445–454 (2021).
Article
PubMed
PubMed Central
Google Scholar
Zheng, X. & Davis, J. W. SAIGEgds—an efficient statistical tool for large-scale PheWAS with mixed models.
Bioinformatics
37
, 728–730 (2021).
Article
CAS
PubMed
MATH
Google Scholar
Mountjoy, E. et al. An open approach to systematically prioritize causal variants and genes at all published human GWAS trait-associated loci.
Nat. Genet.
53
, 1527–1533 (2021).
Article
CAS
PubMed
PubMed Central
MATH
Google Scholar
Uccioli, L. et al. Non-healing foot ulcers in diabetic patients: general and local interfering conditions and management options with advanced wound dressings.
J. Wound Care
24
, 35–42 (2015).
Article
PubMed
Google Scholar
Gary Sibbald, R. & Woo, K. Y. The biology of chronic foot ulcers in persons with diabetes.
Diab./Metab. Res. Rev.
24
, S25–S30 (2008).
Article
MATH
Google Scholar
Magen, E. & Delgado, J-S. Helicobacter pylori and skin autoimmune diseases.
World J. Gastroenterology: WJG
20
, 1510 (2014).
Article
CAS
PubMed Central
MATH
Google Scholar
Leiter, L. The metabolism of imidazoles. studies from the Otho SA sprague memorial institute: Collected reprints. 13. (1925).
Barrouin-Melo, S. M., Morejón Terán, Y. A., Anturaniemi, J. & Hielm-Björkman, A. K. Interaction between nutrition and metabolism.
Metabolic Interact. Infect.
, 29–114. (2018).
Wade, A. M. & Tucker, H. N. Antioxidant characteristics of L-histidine.
J. Nutr. Biochem.
, 308–315 (1998).
Article
CAS
MATH
Google Scholar
Stifel, F. B. & Herman, R. H. Histidine metabolism.
Am. J. Clin. Nutr.
24
, 207–217 (1971).
Article
CAS
PubMed
Google Scholar
Chaudhuri, B. N. et al. Crystal structure of imidazole glycerol phosphate synthase: a tunnel through a (β/α) 8 barrel joins two active sites.
Structure
, 987–997 (2001).
Article
CAS
PubMed
Google Scholar
Goutman, S. A. et al. Untargeted metabolomics yields insight into ALS disease mechanisms.
J. Neurol. Neurosurg. Psychiatry
91
, 1329–1338 (2020).
Article
PubMed
MATH
Google Scholar
Evered, D. & Vadgama, J. Absorption of homocitrulline from the gastrointestinal tract.
Br. J. Nutr.
49
, 35–42 (1983).
Article
CAS
PubMed
Google Scholar
Hommes, F., Roesel, R., Metoki, K., Hartlage, P. & Dyken, P. Studies on a case of HHH-syndrome (hyperammonemia, hyperornithinemia, homocitrullinuria).
Neuropediatrics
17
, 48–52 (1986).
Article
CAS
PubMed
Google Scholar
Kato, T., Sano, M. & Mizutani, N. Homocitrullinuria and homoargininuria in lysinuric protein intolerance.
J. Inherit. Metab. Dis.
12
, 157–161 (1989).
Article
CAS
PubMed
MATH
Google Scholar
McRorie, E., Jobanputra, P., Ruckley, C. & Nuki, G. Leg ulceration in rheumatoid arthritis.
Rheumatology
33
, 1078–1084 (1994).
Article
CAS
Google Scholar
McRorie, E. The assessment and management of leg ulcers in rheumatoid arthritis.
J. Wound Care
, 289–292 (2000).
Article
CAS
PubMed
Google Scholar
Eriksson, H., Gustafsson, J. Å. & Sjövall, J. Excretion of steroid hormones in adults: C19 and C21 steroids in faeces from pregnant women.
Eur. J. Biochem.
12
, 520–526 (1970).
Article
CAS
PubMed
Google Scholar
Laatikainen, T. & Vihko, R. Identification of C19O2 and C21O2 steroids in the Mono-and disulphate fractions of human faeces.
Eur. J. Biochem.
13
, 534–538 (1970).
Article
CAS
PubMed
Google Scholar
Laatikainen, T. Quantitative studies on the excretion of glucuronide and mono-and disulphate conjugates of neutral steroids in human bile.
Annals Clin. Res.
, 338–349 (1970).
CAS
Google Scholar
Abraham, D., Hingorani, A., Albagha, O. & Georgakopoulos, C. Metabolic GWAS of elite athletes reveals novel genetically-influenced metabolites associated with athletic performance.
Alemany, M. The roles of androgens in humans: biology, metabolic regulation and health.
Int. J. Mol. Sci.
23
, 11952 (2022).
Article
CAS
PubMed
PubMed Central
MATH
Google Scholar
Dorszewska, J. et al. Polymorphism of the COMT, MAO, DAT, NET and 5-HTT genes, and biogenic amines in Parkinson’s disease.
Curr. Genom.
14
, 518–533 (2013).
Article
CAS
Google Scholar
Goldstein, D. S., Eisenhofer, G. & Kopin, I. J. Sources and significance of plasma levels of catechols and their metabolites in humans.
J. Pharmacol. Exp. Ther.
305
, 800–811 (2003).
Article
CAS
PubMed
MATH
Google Scholar
De Deurwaerdère, P., Di Giovanni, G. & Millan, M. J. Expanding the repertoire of L-DOPA’s actions: A comprehensive review of its functional neurochemistry.
Prog. Neurobiol.
151
, 57–100 (2017).
Article
PubMed
Google Scholar
McCormick, D. B. Two interconnected B vitamins: riboflavin and pyridoxine.
Physiol. Rev.
69
, 1170–1198 (1989).
Article
CAS
PubMed
MATH
Google Scholar
Schwarcz, R. Kynurenines and glutamate: multiple links and therapeutic implications.
Adv. Pharmacol.
76
, 13–37 (2016).
Article
CAS
PubMed
PubMed Central
MATH
Google Scholar
Nemeth, H., Toldi, J. & Vecsei, L. Role of kynurenines in the central and peripherial nervous systems.
Curr. Neurovasc. Res.
, 249–260 (2005).
Article
PubMed
MATH
Google Scholar
Holtze, M., Mickiené, A., Atlas, A., Lindquist, L. & Schwieler, L. Elevated cerebrospinal fluid kynurenic acid levels in patients with tick-borne encephalitis.
J. Intern. Med.
272
, 394–401 (2012).
Article
CAS
PubMed
Google Scholar
Nilsson, L. et al. Elevated levels of kynurenic acid in the cerebrospinal fluid of male patients with schizophrenia.
Schizophr. Res.
80
, 315–322 (2005).
Article
CAS
PubMed
MATH
Google Scholar
Drivsholm, N. et al. Alterations in the kynurenine pathway of Tryptophan metabolism are associated with depression in people living with HIV.
JAIDS J. Acquir. Immune Defic. Syndr.
87
, e177–e181 (2021).
Article
CAS
PubMed
MATH
Google Scholar
Zhang, J. et al. Non-targeted metabolomics identifies erythronate accumulation in cancer cells.
BioRxiv
, (2022). 2022.2012. 2004.519010.
Ricciardelli, C. et al. Transketolase is upregulated in metastatic peritoneal implants and promotes ovarian cancer cell proliferation.
Clin. Exp. Metastasis
32
, 441–455 (2015).
Article
CAS
PubMed
Google Scholar
Zhang, W. et al. Increased expression of TMPRSS11E is involved in LPS-or Poly (I: C)-mediated inflammation.
Am. J. Respir. Cell Mol. Biol.
68
, 406–416 (2023).
Article
CAS
PubMed
MATH
Google Scholar
Wang, J., Wang, B., Liang, C., Jin, C. & Shen, H. Analysis of Lower-Limb ulcers in participants with leprosy sequelae using metabolomics and 16S ribosomal DNA sequencing.
Clin. Cosmet. Invest. Dermatol.
3465–3480. (2023).
Kuikko, K., Salmi, T., Huhtala, H. & Kimpimäki, T. Characteristics of chronic ulcer patients by gender and ulcer aetiology from a multidisciplinary wound centre.
Int. Wound J.
21
e70012. (2024).
Download references
Funding
This work was supported in part by the National Natural Science Foundation of China (No. 62072128 and 62002079), the Natural Science Foundation of Guangdong Province of China (No. 2023A1515011401), the Open Project of Guangdong Provincial Key Laboratory of Artifificial Intelligence in Medical Image Analysis and Application (No. 2022B1212010011) and the Municipal School Joint Fund of Guangzhou Science and Technology Bureau (No. SL2022A03J00935).
Author information
Authors and Affiliations
School of Mathematics and Information Science, Guangzhou University, Guangzhou, 510006, China
Zheng Ye, Deqing Hong, Jiaqi Yuan, Peng Xu & Wenbin Liu
School of Computer Science of Information Technology, Qiannan Normal University for Nationalities, Duyun, 558000, Guizhou, China
Zheng Ye & Peng Xu
Institute of Computational Science and Technology, Guangzhou University, Guangzhou, China
Zheng Ye, Deqing Hong, Jiaqi Yuan, Peng Xu & Wenbin Liu
Authors
Zheng Ye
View author publications
Search author on:
PubMed
Google Scholar
Deqing Hong
View author publications
Search author on:
PubMed
Google Scholar
Jiaqi Yuan
View author publications
Search author on:
PubMed
Google Scholar
Peng Xu
View author publications
Search author on:
PubMed
Google Scholar
Wenbin Liu
View author publications
Search author on:
PubMed
Google Scholar
Contributions
Conceptualization, ZY, JQY; Methodology, ZY, DQH; Formal analysis, ZY, PX; Software, ZY, WBL; Validation, ZY, PX; Investigation, DQH, WBL. The author has read and agreed to the published version of the manuscript.
Corresponding authors
Correspondence to
Peng Xu
or
Wenbin Liu
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Supplementary Material 1 (download XLSX
Supplementary Material 2 (download CSV
Supplementary Material 3 (download XLSX
Supplementary Material 4 (download XLSX
Supplementary Material 5 (download XLSX
Supplementary Material 6 (download CSV
Supplementary Material 7 (download XLSX
Supplementary Material 8 (download PDF
Supplementary Material 9 (download DOCX
Rights and permissions
Open Access
This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit
Reprints and permissions
About this article
Cite this article
Ye, Z., Hong, D., Yuan, J.
et al.
Assessing the influence of plasma metabolites on chronic skin ulcer risk: a two-sample Mendelian randomization study.
Sci Rep
15
, 10001 (2025). https://doi.org/10.1038/s41598-025-94311-8
Download citation
Received
05 November 2024
Accepted
12 March 2025
Published
22 March 2025
Version of record
22 March 2025
DOI
Share this article
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
Keywords
Chronic skin ulcers
Plasma metabolites
Two-sample Mendelian randomization
Risk
Phewas
Download PDF
Advertisement
Advanced search
Quick links
Explore articles by subject
Find a job
Guide to authors
Editorial policies
Sign up for the
Nature Briefing
newsletter — what matters in science, free to your inbox daily.
Get the most important science stories of the day, free in your inbox.
Sign up for Nature Briefing