Treponema pallidum inhibits CD4+ T-cell proliferation through METAP2: insights from Mendelian randomization analysis

Abstract

Neurosyphilis (NS) is a chronic central nervous system infection caused by Treponema pallidum. Owing to its diverse clinical manifestations and the limited sensitivity of current diagnostic methods, NS is difficult to diagnose. Understanding the molecular mechanisms of NS and identifying reliable biomarkers are essential for improving diagnostic and therapeutic strategies. This study employed Mendelian randomization (MR) analysis to explore the causal relationships among protein ratio quantitative trait loci (rQTLs), cerebrospinal fluid (CSF) metabolites, and syphilis risk at various stages. The results revealed that several rQTLs, including CD46/TNFRSF14 and TBC1D23/TBC1D5, were closely associated with syphilis risk, whereas others, such as BANK1/HEXIM1 and GOPC/HEXIM1, exhibited protective effects. Mediation analysis further identified key CSF metabolites, such as N-acetyltaurine and bilirubin, as important mediators linking rQTLs and syphilis progression. Through integrated analysis of cis-proteins from rQTLs and transcriptomic data from CD4 + T-cells of NS patients, METAP2 was identified as a key biomarker in NS, with the potential mechanisms elucidated. Importantly, T. pallidum may inhibit CD4 + T-cell proliferation by modulating METAP2, thereby accelerating disease progression. These findings offer new insights into the pathogenesis of NS and highlight METAP2 as a potential biomarker, laying a foundation for improving diagnostic and therapeutic strategies.

Introduction

Neurosyphilis (NS), caused by Treponema pallidum, is experiencing a global resurgence and results in major health challenges. NS can occur at any stage of syphilis and has a wide array of clinical symptoms, making early and accurate diagnosis critical (Li et al. 2024). Despite the widespread use of antibiotics, NS cases have increased in recent years (Wormser and Pavia 2019; Xiong et al. 2023; Liu et al. 2023). Known as the “great imitator”, NS is often misdiagnosed due to its diverse clinical manifestations, leading to challenges in early diagnosis (2019, Shi et al. 2025; Wormser and Pavia 2019). Current diagnostic methods, such as the cerebrospinal fluid Venereal Disease Research Laboratory (CSF-VDRL) and Rapid Plasma Reagin (RPR) tests, have limited sensitivity, increasing the risk of missed diagnoses (Hamill et al. 2024). Therefore, investigating the role of proteins and cerebrospinal fluid (CSF) metabolites in NS progression is crucial for increasing diagnostic accuracy and elucidating the pathophysiology of NS.

Mendelian randomization (MR) is a method that uses genetic variations as instrumental variables (IVs) to assess causal relationships between exposures and disease outcomes, offering major advantages in epidemiological research, particularly in reducing confounding factors (Birney 2022). Previous studies have utilized MR to reveal the diversity of syphilis immune phenotypes and their roles across different disease stages (Xie et al. 2024). Additionally, in another broad MR study, researchers constructed a susceptibility map of syphilis, identifying multiple risk and protective factors, such as body mass index (BMI) and psychological factors, associated with susceptibility to syphilis (Xie, Guo Xie et al. 2024a, b). Moreover, MR studies have also identified phosphatidylcholine and sterol ester lipid species as potential causal factors in syphilis susceptibility, implicating host lipid metabolism in disease development (Shen et al. 2025). Furthermore, genetically reduced IL-6 signaling has been linked to decreased tuberculosis risk, underscoring the immunomodulatory influence of cytokine pathways in infectious disease pathogenesis(Hamilton et al. 2025). However, studies exploring the specific roles of proteins and CSF metabolites in NS through MR analysis remain limited. Genetic association studies focusing on protein quantitative trait loci (pQTLs) have revealed novel protein ratio quantitative trait loci (rQTLs) related to protein‒protein interactions (PPIs), providing new perspectives on the biological importance of these rQTLs in disease (Suhre 2024). Furthermore, a comprehensive metabolome-wide genome-wide association study (GWAS) of the CSF metabolome identified 19 brain-related phenotypic associations, revealing the genetic architecture of the CSF metabolome and its role in neurological diseases (Panyard et al. 2021). While some studies have reported abnormal expression of specific proteins and CSF metabolites in NS patients, these causal relationships have yet to be systematically validated via MR methods (Li et al. 2024; Liu et al. 2017, 2019; Qi et al. 2019). rQTLs capture genetically regulated shifts in protein interaction networks that are often central to immune signaling. CSF metabolites, on the other hand, reflect biochemical alterations within the central nervous system, including processes such as inflammation and neurodegeneration. Investigating both allows for a systems-level understanding of how host immune regulation and CNS metabolism jointly contribute to the pathogenesis of NS. Future research should focus on elucidating the causal relationships among rQTLs, CSF metabolites, and NS, potentially offering new avenues for early diagnosis and targeted therapy.

CD4 + T-cells play crucial roles in the proliferation, activation, and differentiation of various immune cells, especially during chronic and persistent infections (Zhu and Zhu 2020). CD4 + T-cells initiate the primary local immune response to syphilis, prompting macrophages and other cells to produce IFN-γ, thereby partially clearing invading T. pallidum (Leader et al. 2007). Although CD4 + T-cells are activated early and play important roles in the antibody response and macrophage activation, these cells appear unable to fully eliminate T. pallidum or prevent persistent infection. Notably, the number and function of CD4 + T-cells are significantly impaired in NS patients (Leader et al. 2007; Liu et al. 2017), which may be related to CD4 + T-cells serving as key targets in the immune evasion strategy of T. pallidum. Moreover, lncRNA-ENST00000421645 can modulate CD4 + T-cell apoptosis and cytokine production in neurosyphilis(Wu et al. 2021). However, the specific effects and mechanisms by which T. pallidum impacts CD4 + T-cells remain largely unclear.

Methionine aminopeptidase 2 (METAP2) is an enzyme that plays a critical role in protein synthesis and cell proliferation (Goya Grocin et al. 2021). METAP2 is essential for the post-translational modification and stabilization of proteins by removing the initiator methionine (iMet) from nascent polypeptide chains (Giglione et al. 2015). Additionally, METAP2 supports cell proliferation by promoting protein synthesis through the protection of eukaryotic initiation factor 2α (eIF2α) from inhibitory phosphorylation (Datta et al. 2001). Notably, METAP2 is closely related to T-cell function, and inhibiting METAP2 can significantly suppress CD4 + T-cell proliferation, thereby dampening the immune response (Yoshimura et al. 2013). However, the specific regulatory mechanisms of METAP2 in NS and its role in disease onset and progression remain unclear.

This study systematically reveals the critical roles of rQTLs and CSF metabolites in the pathogenesis of syphilis through mediation MR analyses. Importantly, METAP2 was identified as a key molecule that mediates the development of NS, and our findings demonstrated that T. pallidum inhibits CD4 + T-cell proliferation by modulating METAP2. These results provide new insights into the pathological mechanisms of NS and lay the theoretical foundation for the development of novel diagnostic and therapeutic strategies.

Materials and methods

Study design

The primary steps of this study are illustrated in Fig. 1: (1) The causal relationships between 2,821 rQTLs and syphilis were analyzed (Fig. 1A). (2) The causal relationships between 338 CSF metabolites and syphilis were assessed (Fig. 1B). (3) The effects of CSF metabolites in the pathway from rQTLs to syphilis were evaluated (Fig. 1C). We defined single nucleotide polymorphisms (SNPs) as IVs, with the validity of MR based on three core assumptions: (1) IVs are strongly associated with exposure; (2) IVs are independent of confounders; and (3) IVs influence the outcome only through exposure (Skrivankova et al. 2021).

[See PDF for image]

Fig. 1

Assumptions and design of the mediation MR analyses. A The causal effects of rQTLs (exposure) on syphilis (outcome), with β_total indicating the total effect of rQTLs on syphilis. B The causal effect of CSF metabolites (mediators) on syphilis (outcome), with β2 representing the effect of CSF metabolites on syphilis. C Mediation analysis where CSF metabolites mediate the pathway from rQTLs (exposure) to syphilis (outcome). Here, β1 represents the effect of rQTLs (exposure) on CSF metabolites (mediators). The indirect effect (β_dir = β_total - β1*β2) reflects the impact of exposure on syphilis mediated by the corresponding metabolite

Data sources

GWAS data sources for the rQTLs

Data on 2,821 rQTLs were obtained from published studies derived from the GWAS data of rQTLs in the UK Biobank (Suhre 2024). This dataset includes samples from 52,705 participants, all from the UK Biobank baseline data. Protein levels were measured via the Olink Explore 1536 proteomic platform, which covers 1,463 proteins. Samples from the participants underwent rigorous quality control, with low-quality samples excluded. The remaining samples were divided into two subsets: one for discovery analysis (43,509 participants of European ancestry) and the other for validation analysis (9,196 participants of diverse ancestries). Notably, all discovery-stage participants were of European ancestry, which ensures population homogeneity and supports the validity of MR analysis. Data were accessed via the UK Biobank Research Analysis Platform (RAP), following guidelines and procedures approved by the UK Biobank Ethics Committee.

GWAS data sources of CSF metabolites

Data on 338 CSF metabolites were obtained from published studies sourced from the Wisconsin Alzheimer’s Disease Research Center (WADRC) and the Wisconsin Registry for Alzheimer’s Prevention (WRAP) GWAS data (Panyard et al. 2021). This dataset includes CSF samples from 689 participants, with 532 from WADRC and 168 from WRAP. The samples were collected via lumbar puncture, centrifuged, aliquoted, and stored at -80 °C. Metabolite analysis was conducted via ultrahigh-performance liquid chromatography‒tandem mass spectrometry (UPLC‒MS/MS). After quality control and genetic filtering, the final GWAS dataset included 291 unrelated individuals of European ancestry, ensuring consistency across MR datasets. The participant selection criteria for the WADRC included age ≥ 45 years, decision-making capacity, and fasting for 12 h before collection. The study was approved by the Health Sciences Ethics Committee of the University of Wisconsin.

GWAS data sources for syphilis

Three datasets from the FinnGen R10 GWAS were included: the overall-stage syphilis GWAS (Ncase = 697, Ncontrol = 400,197), the early-stage syphilis GWAS (Ncase = 308, Ncontrol = 400,197), and the late-stage syphilis GWAS (Ncase = 359, Ncontrol = 400,197). All GWAS data originated from European individuals, with detailed information accessible online (FinnGen 2024). This consistent European ancestry across all GWAS sources minimizes population stratification bias and enhances the reliability of MR inference.

MR analysis

Primary analysis

We conducted two-sample MR analysis to evaluate the causal effects of rQTLs, CSF metabolites, and syphilis (Fig. 1). Five prevalent MR approaches were employed: MR Egger, weighted median, inverse variance weighted (IVW), weighted mode, and simple mode. The complete results from each method are presented in Supplementary Tables S3-S5. Among these methods, the IVW method, which combines the Wald ratios of causal effects for each SNP via meta-analysis, was adopted as the principal analysis method because of its effectiveness in assessing causal effects. We harmonized the summary statistics via the harmonise_data function within the TwoSampleMR R software package. The causal impacts of rQTLs on syphilis and CSF metabolites on syphilis were documented as total effects (β_total) and β2, respectively (Fig. 1) (Papadimitriou et al. 2020; Skrivankova et al. 2021).

Mediation analysis

After completing the initial two-sample MR analysis, we conducted a mediation analysis via a two-step Mendelian randomization (TSMR) design. Significant rQTLs and CSF metabolites from the initial analysis were included in the mediation analysis (Fig. 1) to evaluate the mediation effect. β_total was decomposed into an indirect effect (β1*β2; through mediation) and a direct effect (β_dir; without mediation). The percentage mediated by the mediating effect was calculated by dividing the indirect effect by the total effect (Papadimitriou et al. 2020; Skrivankova et al. 2021).

IV selection

SNPs associated with rQTLs, CSF metabolites, and syphilis were selected with P < 5 × 10^-5 to maximize SNP inclusion. An F > 10 was used to ensure sufficient strength, avoiding weak instrument bias. SNPs with F < 10 were excluded, and linkage disequilibrium was eliminated through clumping (LD r2 < 0.001, kb = 10,000). SNPs with palindromic properties and minor allele frequency (MAF) < 0.01 were pruned (Papadimitriou et al. 2020).

Sensitivity analysis

Cochrane’s Q statistic was employed to assess the heterogeneity of SNP effects. Horizontal pleiotropy was identified via the MR‒Egger intercept method and MR-PRESSO method. Scatter plots of SNP-exposure and SNP-outcome associations were generated to visualize the MR results (Additional file: Sensitivity analysis). A leave-one-out analysis was conducted to verify the influence of each SNP on the overall causal estimate (Papadimitriou et al. 2020; Skrivankova et al. 2021; Verbanck et al. 2018).

Construction of the lncRNA‒miRNA-mRNA regulatory network

Using the GEO database, we analyzed lncRNA and mRNA expression profiles (GSE103599) from CD4 + T-cells of NS patients and miRNA expression profiles (GSE156421) from syphilis-associated peripheral blood mononuclear cells (PBMCs) (Liu et al. 2017; Yang et al. 2021). The relationships between differentially expressed lncRNAs (DElncRNAs) and differentially expressed miRNAs (DEmiRNAs) were explored via the starBase v3.0 and miRcode databases. TargetScan, miRcode, and MiRanda were used to decode the interactions between DEmiRNAs and differentially expressed mRNAs (DEmRNAs) (Liu et al. 2021). The competing endogenous RNA (ceRNA) regulatory network was constructed and visualized via Cytoscape 3.10.2.

Cell lines and preparation of the virulent T. pallidum strain

Human Jurkat T-cells were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA) and cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum. The cells were incubated at 37 °C in a 5% CO2 atmosphere. Cell authentication was conducted via short tandem repeat (STR) profiling. All the cell lines were confirmed to be free of mycoplasma contamination.

The T. pallidum Nichols strain used in this study was generously provided by Professor Tianci Yang (Zhongshan Hospital, Xiamen University, China). This strain was subsequently propagated within the testicles of adult male New Zealand white rabbits (Hu et al. 2023; Zhang et al. 2023). Following infection, T. pallidum spirochetes were harvested from inflamed testicular tissue, resuspended in PBS, and examined via dark field microscopy at 40x magnification. Uninfected samples were processed similarly to exclude potential interference. All animal experiments were approved by the Animal Care and Use Committee of the University of South China (USC2024JT031).

Plasmids and transfection

Small interfering RNAs (siRNAs) targeting METAP2, along with a negative control, were synthesized by GenePharma, and the sequences are provided in Supplementary Table S1. The cDNA expression plasmids (GV492-METAP2 and GV492-control vectors) were purchased from Genechem. The cells were transfected with Lipofectamine 3000 (Invitrogen) following the manufacturer’s instructions. Transfection efficacy was assessed via quantitative real-time PCR (qRT‒PCR) and Western blotting.

RNA extraction and qRT‒PCR

Total RNA was isolated via TRIzol reagent, and cDNA was synthesized via a RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific). qRT‒PCR was conducted with ChamQ™ Universal SYBR^® qPCR Master Mix (Vazyme Biotech). The target mRNA expression levels were normalized to those of β-actin (ACTB). The primer sequences are listed in Supplementary Table S2.

Western blot analysis

Jurkat T-cells were lysed in RIPA buffer supplemented with a protease inhibitor cocktail. Protein extracts (50 µg) were boiled, separated by SDS‒PAGE on 10% gels, and transferred to nitrocellulose membranes. The membranes were blocked with 5% nonfat dry milk and incubated overnight at 4 °C with antibodies against METAP2 (#sc-365637, Santa Cruz Biotechnology), PDCD5 (#YN1785, Immunoway), and ACTB (#sc-47778, Santa Cruz Biotechnology). The membranes were then incubated with HRP-conjugated secondary antibodies and detected via an enhanced chemiluminescence (ECL) HRP substrate.

Fluorescence-activated cell sorting (FACS)

For flow cytometric analysis, Jurkat T-cells were stained with antibodies specific for Ki67 (clone Ki-67, BioLegend). The cells were fixed and permeabilized via a FOXP3/Transcription Factor Buffer Set (EBioscience). Data collection was performed via a BD Fortessa X20 system (BD Biosciences), and data analysis was conducted with FlowJo (v10.8.1) software.

Statistical analysis

We used R (version 4.4.0) with the “TwoSampleMR,” “MRPRESSO,” “foreach,” and “ggplot2” packages for MR and sensitivity analyses. The “data.table” package was used for SNP selection, and the “grid” and “forestploter” packages were used for plotting. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were conducted via DAVID (https://david.ncifcrf.gov/) and KOBAS (http://bioinfo.org/kobas) software, with visualization via “ggplot2” (Ai and Kong 2018; Sherman et al. 2022).

Student’s t test was used to determine the significance of differences between the control and experimental groups via SPSS 26.0 and GraphPad Prism 9.0. The data are presented as the means ± standard deviations (SDs) from three independent experiments. P < 0.05 was considered statistically significant.

Results

MR analysis reveals rQTLs linked to syphilis risk across different stages

We conducted MR analysis via the IVW method to explore the causal relationships between rQTLs and syphilis risk. We analyzed syphilis cases from the FinnGen database, which included overall, early, and late stages of syphilis. Additionally, this analysis incorporated data from 2,821 rQTLs derived from Olink proteomics in the UK Biobank to identify key rQTLs involved in syphilis pathogenesis.

For overall-stage syphilis, regardless of disease stage, 52 rQTLs were positively associated with syphilis risk, whereas 59 were negatively associated. In early-stage syphilis, 57 rQTLs were positively associated, and 72 rQTLs were negatively associated. In contrast, 37 rQTLs were positively associated with late-stage syphilis, and 50 rQTLs were negatively associated with late-stage syphilis (Supplementary Table S3). We highlighted the top five significant positive and negative associations for each stage on the basis of the lowest p values (Fig. 2). In overall-stage syphilis, rQTLs such as CD46/TNFRSF14 and AXL/IL18BP were strongly linked to an increased risk of syphilis, whereas rQTLs such as BANK1/HEXIM1 and GOPC/HEXIM1 were associated with a reduced risk, suggesting their potential protective roles (Fig. 2A, B). Early-stage syphilis was significantly positively associated with CD46/TNFRSF14 and negatively associated with BANK1/HEXIM1 (Fig. 2A, B). In late-stage syphilis, notable rQTLs, such as IL15/IL18BP and CRELD2/ERP44, were positively correlated with disease risk, whereas CA2/TGM2 and DPP7/LGMN were negatively associated with disease risk (Fig. 2A, B). These results indicate that different rQTLs play distinct roles in the pathogenesis of syphilis across various stages of the disease.

[See PDF for image]

Fig. 2

MR analysis reveals rQTLs linked to syphilis risk across different stages. A rQTLs positively associated with syphilis risk [odds ratio (OR) > 1]. B rQTLs negatively associated with syphilis risk (OR < 1). Red and green dots represent positive and negative syphilis risk, respectively, from the IVW analysis. Each panel displays the top five significant rQTLs with the lowest p values across overall-stage syphilis, early-stage syphilis, and late-stage syphilis. The OR and 95% confidence interval (CI) were used to quantify the association between each rQTLs and syphilis risk

Functional pathways and biological roles of rQTL-associated cis-proteins in syphilis

To further elucidate the roles of rQTLs in various biological processes and molecular functions across different stages of syphilis, we performed KEGG and GO enrichment analyses on the cis-proteins associated with these rQTLs.

For overall-stage syphilis, KEGG enrichment analysis revealed that the cis-proteins were involved primarily in the MAPK signaling pathway, the PI3K-Akt signaling pathway, the HIF-1 signaling pathway, metabolic pathways, and purine metabolism (Fig. 3A). GO enrichment analysis highlighted their critical roles in inflammatory responses, lipid metabolic processes, lymphocyte chemotaxis, positive regulation of the ERK1 and ERK2 cascades, protein phosphorylation, and macrophage differentiation (Fig. 3D). In early-stage syphilis, KEGG analysis indicated that these cis-proteins were associated mainly with the T-cell receptor signaling pathway, the chemokine signaling pathway, metabolic pathways, the PI3K-Akt signaling pathway, and purine metabolism (Fig. 3B). GO enrichment analysis revealed their roles in adaptive immune responses, inflammatory responses, fatty acid transport, negative regulation of T-cell proliferation, and regulation of the MAPK cascade (Fig. 3E). For late-stage syphilis, KEGG enrichment analysis revealed that the cis-proteins were predominantly involved in the MAPK signaling pathway, the PI3K‒Akt signaling pathway, cellular senescence, metabolic pathways, and purine metabolism (Fig. 3C). GO analysis revealed that these proteins were associated mainly with axon development, neuronal cell body organization, CD4 receptor binding, cytokine binding, and protein tyrosine kinase activity (Fig. 3F). These findings highlight the multifaceted roles of rQTLs in diverse biological processes and molecular functions across the various stages of syphilis.

[See PDF for image]

Fig. 3

Functional pathways and biological roles of rQTL-associated cis-proteins in syphilis. A KEGG pathway analysis and D GO enrichment analysis of cis-proteins from rQTLs associated with overall-stage syphilis risk. B KEGG pathway analysis and E GO enrichment analysis of cis-proteins from rQTLs associated with early-stage syphilis risk. C KEGG pathway analysis and F GO enrichment analysis of cis-proteins from rQTLs associated with late-stage syphilis risk. The horizontal axis represents the proportion of candidate gene sets relative to background genes. The size of the dots indicates the number of cis-proteins involved in each pathway, and the color of the dots corresponds to different p value ranges

MR analysis identifies key CSF metabolites associated with syphilis risk

CSF metabolites play crucial roles in the pathogenesis and progression of syphilis, particularly NS (Liu et al. 2019; Qi et al. 2019). We performed MR analysis via the IVW method to investigate the causal relationships between CSF metabolites and the risk of developing syphilis. The analysis included syphilis cases from the FinnGen database, covering overall, early, and late stages, to identify key CSF metabolites involved in syphilis pathogenesis. The CSF metabolite data were sourced from the WADRC and WRAP studies, encompassing a total of 338 CSF metabolites. In overall-stage syphilis, three metabolites were positively associated with syphilis risk, whereas five demonstrated negative associations. For early-stage syphilis, four metabolites were positively associated, and the other four were negatively associated. In late-stage syphilis, five metabolites were positively associated with increased disease risk, whereas ten were negatively associated (Fig. 4, Supplementary Table S4).

Specifically, in the overall-stage syphilis analysis, certain CSF metabolites were significantly linked to syphilis risk. Orotidine, 3-methoxytyramine sulfate, and N-acetylaspartate (NAA) were found to significantly increase the overall risk of syphilis. Conversely, CSF metabolites such as N-acetyltaurine, 4-chlorobenzoic acid, and S-1-pyrroline-5-carboxylate were associated with a lower risk of syphilis, suggesting a potential protective effect (Fig. 4A, B). In early-stage syphilis, bilirubin (Z, Z), maleate, and alpha-hydroxyisocaproate were significantly associated with increased risk, indicating their possible roles in the early pathogenesis of syphilis. Moreover, urate, X-12,411, and homocarnosine were linked to a reduced risk, suggesting that these metabolites might play a protective role in early syphilis (Fig. 4A, B). For late-stage syphilis, several metabolites exhibited significant associations with disease risk. Orotidine, dimethylmalonic acid, and phenyllactate (PLA) were linked to a greater risk of late-stage syphilis, whereas sphingomyelin (d18:2/16:0, d18:1/16:1), serine, and N-acetylglutamine demonstrated protective effects (Fig. 4A, B). These findings underscore the critical and stage-dependent roles of CSF metabolites in syphilis pathogenesis, with different metabolites either promoting or inhibiting disease progression.

[See PDF for image]

Fig. 4

MR analysis identifies key CSF metabolites associated with syphilis risk. A CSF metabolites positively associated with syphilis risk (OR > 1). B CSF metabolites are negatively associated with syphilis risk (OR < 1). Red and green dots represent positive and negative syphilis risk, respectively, from the IVW analysis. Each panel displays the top five significant rQTLs with the lowest p values across overall-stage syphilis, early-stage syphilis, and late-stage syphilis. The OR and 95% CI were used to quantify the association between each rQTLs and syphilis risk

Mediation analysis of the effect of rQTLs on syphilis risk through CSF metabolites

We performed TSMR analyses to evaluate the mediation effects of CSF metabolites on rQTLs that exhibited statistically significant associations across different stages of syphilis (Supplementary Table S5). The complete mediation analysis results are provided in Supplementary Table S6. We highlighted the top five associations with the highest mediation effect proportions for each stage (Table 1). In the overall-stage syphilis risk analysis, rQTLs such as PDGFA/SNAP23, BACH1/IKBKG, METAP2/PDCD5, LTBR/TNFRSF14, and EIF4EBP1/PDCD5 were found to be negatively associated with syphilis risk via N-acetyltaurine levels. Among these factors, PDGFA/SNAP23 had the greatest mediating effect at 21.8%, followed by BACH1/IKBKG, METAP2/PDCD5, LTBR/TNFRSF14, and EIF4EBP1/PDCD5, with mediating effect proportions of 18.2%, 14.7%, 14.2%, and 13.7%, respectively. In the early-stage syphilis analysis, rQTLs such as CA2/NSFL1C and ARG1/TGM2 demonstrated significant negative mediation effects through urate levels, with proportions of 14.9% and 13.2%, respectively, with CA2/NSFL1C showing the highest mediation effect proportion. Additionally, PARK7/TXNRD1, HSPG2/SMOC2, and IGFBP6/SMOC2 exhibited positive mediation effects through alpha-hydroxyisocaproate levels, with mediation effect proportions of 13.3%, 12.4%, and 12.4%, respectively. In the late-stage syphilis analysis, CA5A/GSTA1 demonstrated a significant negative mediation effect through 2-aminophenol sulfate levels, with a proportion of 17.3%, the highest among the late-stage associations. MVK/SERPINE1 had a negative mediation effect on N-acetyltaurine levels, with a proportion of 11.4%. Additionally, DCTN1/RWDD1 and PARK7/TXNDC5 had positive mediating effects on 2-aminophenol sulfate levels, with percentages of 13.9% and 12.7%, respectively. Finally, ACY1/SCLY demonstrated a significant positive mediation effect through orotidine levels, with a proportion of 10.8%.

To ensure the robustness of our findings, we conducted additional validation and sensitivity analyses. The results of the Q statistic and MR‒Egger intercept analyses indicated no significant heterogeneity (Q statistic p values ranged from 0.061 to 0.977; Table 1). Similarly, the MR‒Egger intercept analysis did not reveal any substantial horizontal pleiotropy (the intercept p values ranged from 0.090 to 0.933; Table 1). These findings suggest that our results are both robust and reliable. In summary, the results of these mediation MR analyses illustrate how rQTLs influence the pathogenesis of syphilis through specific CSF metabolites.

Table 1. TSMR analysis of causal effects among rQTLs, CSF metabolites, and syphilis

Exposure	Mediator	β1 ± SE	β2 ± SE	Beta_dir (β_total - β1*β2)	Mediated proportion	P	Q-statistics	Ph	Egger intercept	Pintercept
OVERALL_SYPHILIS
PDGFA/SNAP23	N-acetyltaurine levels	0.100 ± 0.034	-0.642 ± 0.247	-0.229 ± 0.159	0.218	0.003	26.983	0.927	-0.001	0.892
BACH1/IKBKG	N-acetyltaurine levels	0.155 ± 0.055	-0.642 ± 0.247	-0.448 ± 0.159	0.182	0.005	16.361	0.567	0.005	0.463
METAP2/PDCD5	N-acetyltaurine levels	0.081 ± 0.041	-0.642 ± 0.247	-0.299 ± 0.159	0.147	0.048	24.257	0.061	0.003	0.656
LTBR/TNFRSF14	N-acetyltaurine levels	-0.060 ± 0.023	-0.642 ± 0.247	0.232 ± 0.159	0.142	0.010	20.416	0.977	-0.004	0.225
EIF4EBP1/PDCD5	N-acetyltaurine levels	0.061 ± 0.028	-0.642 ± 0.247	-0.248 ± 0.159	0.137	0.029	6.807	0.744	0.008	0.242
EARLY_SYPHILIS
CA2/NSFL1C	Urate levels	-0.165 ± 0.063	-0.747 ± 0.316	0.701 ± 0.236	0.149	0.009	12.878	0.378	0.008	0.393
PARK7/TXNRD1	Alpha-hydroxyisocaproate levels	0.060 ± 0.027	0.857 ± 0.363	0.333 ± 0.311	0.133	0.025	9.044	0.875	-0.001	0.788
ARG1/TGM2	Urate levels	0.117 ± 0.043	-0.747 ± 0.316	-0.577 ± 0.236	0.132	0.007	35.636	0.220	-0.004	0.449
HSPG2/SMOC2	Alpha-hydroxyisocaproate levels	-0.046 ± 0.020	0.857 ± 0.363	-0.277 ± 0.311	0.124	0.023	37.040	0.762	< -0.001	0.862
IGFBP6/SMOC2	Alpha-hydroxyisocaproate levels	-0.052 ± 0.023	0.857 ± 0.363	-0.314 ± 0.311	0.124	0.027	55.208	0.192	-0.002	0.490
LATE_SYPHILIS
CA5A/GSTA1	2-aminophenol sulfate levels	-0.141 ± 0.056	-0.272 ± 0.125	0.184 ± 0.035	0.173	0.012	20.325	0.160	0.024	0.090
DCTN1/RWDD1	2-aminophenol sulfate levels	0.331 ± 0.134	-0.272 ± 0.125	-0.556 ± 0.056	0.139	0.013	15.317	0.640	-0.004	0.825
PARK7/TXNDC5	2-aminophenol sulfate levels	0.279 ± 0.134	-0.272 ± 0.125	-0.522 ± 0.051	0.127	0.037	11.123	0.518	-0.033	0.121
MVK/SERPINE1	N-acetyltaurine levels	-0.085 ± 0.040	-0.764 ± 0.348	0.503 ± 0.266	0.114	0.034	38.252	0.368	0.007	0.203
ACY1/SCLY	Orotidine levels	-0.110 ± 0.045	0.834 ± 0.325	-0.752 ± 0.271	0.108	0.015	19.355	0.562	0.001	0.907

This table presents the beta (β), standard error (SE), and p values obtained through TSMR analysis. β1 represents the effect of rQTL (exposure) on CSF metabolites (mediators); β2 represents the effect of CSF metabolites (mediators) on syphilis (outcome); β_total represents the total effect of rQTLs (exposure) on syphilis (outcome). The p value was used to assess the significance of the impact of rQTLs (exposure) on CSF metabolites (mediators) via the IVW method. The indirect effect (β_dir = β_total - β1*β2) reflects the impact of exposure on syphilis mediated by the corresponding metabolite. The proportion mediated is calculated as the “indirect effect/total effect”. Ph refers to the p value for heterogeneity, and Pintercept refers to the p value for the intercept of the MR‒Egger regression.

Enrichment of biological pathways and interactions of rQTLs in syphilis mediation

To further elucidate the roles of the rQTLs associated with CSF metabolites in the biological processes and molecular functions related to syphilis, we conducted KEGG and GO enrichment analyses on the cis-proteins associated with these rQTLs. KEGG enrichment analysis revealed that these cis-proteins are involved primarily in key pathways, such as the PI3K-Akt signaling pathway, cholesterol metabolism, purine metabolism, the HIF-1 signaling pathway, and the Jak-STAT signaling pathway (Fig. 5A). The GO enrichment analysis indicated that these cis-proteins are involved mainly in processes such as axon structure, calcium ion binding, membrane raft formation, signaling receptor binding, and virus receptor activity (Fig. 5B). Additionally, PPI analysis of cis-proteins associated with rQTLs in syphilis mediation revealed complex interactions within biological networks, further supporting their critical roles in syphilis pathogenesis (Fig. 5C).

[See PDF for image]

Fig. 5

KEGG and GO enrichment analyses of cis-proteins from rQTLs in mediation analysis. A KEGG pathway analysis and B GO enrichment analysis of cis-proteins from rQTLs in mediation analysis. (C) PPI analysis of cis-proteins from rQTLs via mediation analysis. The horizontal axis represents the proportion of candidate gene sets to background genes, the size of the dots indicates the number of cis-proteins involved in each pathway, and the color of the dots corresponds to different p value ranges.

METAP2 as a potential key molecule in regulating CD4 + T-cells in NS

To identify critical protein molecules involved in the CSF metabolite-mediated processes of NS, we screened cis-proteins from rQTLs identified in the mediation analysis, focusing on those with consistent directions in both indirect and direct effects (Supplementary Table S5). Through an integrated analysis of these cis-proteins and transcriptomic data from CD4 + T-cells of NS patients (GSE103599, p < 0.05), METAP2 emerged as a uniquely differentially expressed protein (Fig. 6A).

To further explore the specific mechanisms through which METAP2 contributes to NS, we conducted an integrated analysis of lncRNA expression profiles (GSE103599) from CD4 + T-cells of NS patients and miRNA expression profiles (GSE156421) from syphilis-associated PBMCs. The upregulated miR-98-3p and miR-449b-3p in syphilis-associated PBMCs may interact with METAP2. Moreover, we found that miR-98-3p and miR-449b-3p have binding sites on LINC00501 and LINC00334, respectively, both of which are downregulated in the CD4 + T-cells of NS patients (Fig. 6B). Notably, Jurkat T-cells were treated for 24 h with PBS (resuspended from noninfected rabbit testes), active T. pallidum [LTP, multiplicity of infection (MOI) of 2:1], or inactivated T. pallidum (DTP, MOI of 2:1, 56 °C). qRT‒PCR and Western blot analyses revealed that T. pallidum treatment substantially decreased the mRNA levels of LINC00501, LINC00334, and METAP2, as well as the protein expression levels of METAP2 (Fig. 6C-E). These findings suggest that METAP2 is a critical molecule in NS that potentially modulates CD4 + T-cell function in this disease. Further mechanistic analysis revealed that the LINC00501/miR-98-3p/METAP2 and LINC00334/miR-449b-3p/METAP2 axes may constitute a potential ceRNA regulatory network in NS.

[See PDF for image]

Fig. 6

METAP2 as a potential key molecule in regulating CD4 + T-cells in NS. A Venn diagram showing the overlap between cis-proteins from rQTLs identified via mediation analysis and RNA-seq data from CD4 + T-cells of NS patients (GSE103599). B The relationships between DelncRNAs and DEmiRNAs identified in GSE103599 and GSE156421 were determined via the starBase v3.0 database. The interactions between DEmiRNAs and METAP2 were predicted via TargetScan, miRcode, and MiRanda. On the basis of the predicted interactions among lncRNAs, miRNAs, and mRNAs, a ceRNA regulatory network was constructed and visualized via Cytoscape 3.10.2. C qRT‒PCR analysis was used to detect changes in the mRNA expression levels of LINC00501, LINC00334, METAP2, and PDCD5 in Jurkat T-cells treated with LTP, DTP, or PBS for 24 h. D Western blotting was performed to assess changes in the protein expression levels of METAP2 and PDCD5 in Jurkat T-cells treated with LTP, DTP, or PBS for 24 h. *p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant

T. pallidum inhibits CD4 + T-cell proliferation through METAP2

To confirm the role of METAP2 in CD4 + T-cell proliferation, we transfected two siRNAs with optimal efficacy for METAP2 knockdown and overexpression into Jurkat T-cells, and their effects were validated (Fig. 7A-D). Ki67 staining results revealed that METAP2 knockdown significantly inhibited Jurkat T-cell proliferation, whereas METAP2 overexpression significantly promoted T-cell proliferation (Fig. 7E, F). Notably, METAP2 overexpression reversed the T. pallidum-induced inhibition of proliferation in Jurkat T-cells (Fig. 7G). These findings suggest that T. pallidum inhibits CD4 + T-cell proliferation by modulating METAP2, thereby playing a critical role in the immune response during NS.

[See PDF for image]

Fig. 7

T. pallidum inhibits CD4 + T-cell proliferation through METAP2. A-D qRT-PCR and Western blotting were used to measure METAP2 mRNA and protein expression levels in Jurkat T-cells 48 h after METAP2 knockdown or overexpression to assess the transfection efficacy. E, F Flow cytometry was used to assess Ki67 levels to evaluate the impact of METAP2 gene knockdown or overexpression on Jurkat T-cell proliferation. G Flow cytometry was used to determine Ki67 levels to assess the effects of METAP2 overexpression and T. pallidum treatment on Jurkat T-cell proliferation. **p < 0.01; ***p < 0.001; ns, not significant

Discussion

This study systematically assessed the causal relationships among rQTLs, CSF metabolites, and syphilis risk via MR analysis. The results revealed that several rQTLs are significantly associated with syphilis risk at various stages of the disease. Additionally, mediation MR analysis highlighted the intermediary role of specific rQTLs via CSF metabolites in syphilis pathogenesis. Importantly, all three GWAS datasets used in this study—including those for rQTLs, CSF metabolites, and syphilis—were derived from individuals of European ancestry. This population homogeneity minimizes the potential for confounding due to population stratification and thereby increases the robustness of causal inferences made through Mendelian randomization. Notably, METAP2 was identified as a key molecule in the development of NS, with evidence showing that T. pallidum inhibits CD4 + T-cell proliferation by modulating METAP2. These findings provide a crucial foundation for further exploration of the molecular mechanisms underlying NS.

In this study, several rQTLs, particularly the ratios of CD46/TNFRSF14, TBC1D23/TBC1D5, and IL15/IL18BP, were positively associated with syphilis risk at different stages, suggesting their potential roles in increasing susceptibility to syphilis. Conversely, rQTLs such as BANK1/HEXIM1, CNPY4/TBC1D23, and COL6A3/SMOC2 were negatively associated with syphilis risk at various stages, indicating their potential protective roles. These findings underscore the importance of these rQTLs in regulating the pathogenesis of syphilis. To gain deeper insights into the biological processes and molecular functions involved in syphilis at various stages, we performed KEGG and GO enrichment analyses on the cis-proteins associated with these rQTLs. In the overall syphilis cohort, these analyses revealed key pathways, including the MAPK and PI3K-Akt signaling pathways, which play central roles in immune regulation, inflammation, and cell survival (Glaviano et al. 2023; Ullah et al. 2022). In contrast, during early-stage syphilis, KEGG and GO analyses revealed significant associations with the PI3K-Akt signaling pathway, the PPAR signaling pathway, the T-cell receptor signaling pathway, and negative regulation of T-cell proliferation. This finding strongly suggests that in the early stages of syphilis, not only the regulation of host metabolism and cell survival processes but also the potential suppression of the T-cell immune response occur (Glaviano et al. 2023; Pinna 2023; Shah et al. 2021). In late-stage syphilis, KEGG and GO analyses revealed enrichment in pathways related to cellular stress, aging, and neurointeraction, such as the mTOR and PI3K-Akt signaling pathways, cellular senescence, and longevity regulation. These findings suggest a shift toward managing energy balance, survival, and the stress response, reflecting the chronic nature of late-stage syphilis (Glaviano et al. 2023; Liu et al. 2023). A comparison of the KEGG and GO analyses across different stages clearly revealed that while metabolism, the immune response, and inflammation (e.g., PI3K-Akt signaling) are involved at all stages, the specific pathways enriched at each stage reflect unique biological demands. Early-stage syphilis is characterized by strong immune activation, whereas late-stage syphilis shifts toward managing cellular stress, neurointeractions, and chronic inflammation. These analyses offer valuable insights into the roles of rQTLs in syphilis progression, highlighting key signaling pathways that may drive the disease at different stages.

CSF metabolites play crucial roles in the pathogenesis and progression of syphilis, particularly NS (Liu et al. 2019; Qi et al. 2019). These molecules may influence syphilis development by regulating neuroinflammation, metabolic pathways, or neuroprotective mechanisms (Wakamatsu et al. 2022). Previous studies have shown that NS patients have elevated levels of N-acetyl-L-tyrosine, bilirubin, and L-histidine and reduced levels of L-gulonolactone-γ-lactone, L-methionine, and serine (Liu et al. 2019; Qi et al. 2019). In this study, several CSF metabolites were significantly associated with syphilis risk at different stages. For example, orotidine, NAA, bilirubin, and histidine increased syphilis risk, whereas N-acetyltaurine, L-methionine, urate, and serine had protective effects. Notably, the elevated bilirubin and histidine levels, which increase syphilis risk, are consistent with their higher levels in NS (Qi et al. 2019), whereas the reduced L-methionine and serine levels, which lower syphilis risk, are consistent with their lower levels in NS (Qi et al. 2019). N-acetyltaurine is a derivative of taurine that is known for its osmoregulatory and neuroprotective functions. Its decreased levels may impair neuronal membrane stability and increase susceptibility to neuroinflammatory injury (Jafri et al. 2017; Wei et al. 2024). Additionally, elevated NAA levels in CSF are associated with poor prognosis in traumatic brain injury (Osier et al. 2019), whereas increased urate levels are linked to a reduced risk of Parkinson’s disease and Alzheimer’s disease (Paganoni and Schwarzschild 2017). Collectively, CSF metabolites may serve as important biomarkers for early syphilis diagnosis or monitoring disease progression.

Notably, our mediation analysis also provided genetic evidence for the associations among rQTLs, CSF metabolites, and syphilis. Our results suggest that certain rQTLs may influence syphilis development and progression through specific CSF metabolites, such as the mechanisms by which PDGFA/SNAP23, BACH1/IKBKG, and METAP2/PDCD5 reduce syphilis risk through N-acetyltaurine.

Additionally, we conducted KEGG and GO enrichment analyses on the cis-proteins associated with these rQTLs across different stages of syphilis. The significant enrichment of the PI3K-Akt signaling pathway, HIF-1 signaling pathway, cholesterol metabolism, and nitrogen metabolism suggested that these proteins play critical roles in regulating cell survival, metabolism, and immune responses (Glaviano et al. 2023; Luo et al. 2020; Taylor and Colgan 2017; Vander Heiden et al. 2009). Furthermore, the enrichment of axon structure, calcium ion binding, and membrane rafts indicates that these proteins may be crucial in signal transmission and intercellular communication within the nervous system (Shah et al. 2017). Notably, multiple enriched pathways such as the PI3K-Akt, MAPK, and T-cell receptor (TCR) signaling pathways, are known to directly regulate CD4 + T-cell survival, proliferation, and differentiation(Bartleson et al. 2020; Na et al. 2020; Tang et al. 2023). For example, PI3K-Akt signaling is essential for IL-2-mediated T-cell expansion and effector function, whereas MAPK cascades (e.g., ERK1/2) modulate cytokine production and activation thresholds(Bartleson et al. 2020; Na et al. 2020). Previous studies have shown that the PI3K-Akt and MAPK signaling pathways are closely associated with T. pallidum and its proteins(Li et al. 2023; Liu et al. 2024). The enrichment of these immune-regulatory pathways among rQTL-associated proteins and transcriptomic signals suggested that T. pallidum may subvert CD4 + T-cell responses by interfering with upstream regulators such as METAP2 or rQTL-related molecular networks. These findings support our central hypothesis that T. pallidum disrupts host immunity through inhibition of CD4 + T-cell function and reveal candidate molecular nodes for therapeutic targeting.

Elucidating the regulatory mechanisms of key molecules in NS is crucial for understanding how T. pallidum manipulates host immune responses. By integrating cis-proteins from rQTLs with transcriptomic data from CD4 + T-cells of NS patients, we identified METAP2 as a uniquely differentially expressed protein, highlighting its critical role in NS immune regulation. The bioinformatic analysis further revealed that the upregulated miR-98-3p and miR-449b-3p in PBMCs from individuals with syphilis may interact with METAP2. Additionally, miR-98-3p and miR-449b-3p have binding sites on LINC00501 and LINC00334, respectively, both of which are downregulated in the CD4 + T-cells of NS patients. Previous studies have shown that LINC00501 promotes gastric cancer cell proliferation, whereas miR-449b-3p inhibits endometrial stromal cell (ESC) proliferation (Liu et al. 2018; Pei et al. 2023). Notably, T. pallidum treatment substantially decreased the mRNA levels of LINC00501, LINC00334, and METAP2, as well as the protein expression levels of METAP2. These findings suggest that the LINC00501/miR-98-3p/METAP2 and LINC00334/miR-449b-3p/METAP2 axes may form a potential ceRNA regulatory network in NS. In this study, METAP2 knockdown significantly inhibited Jurkat T-cell proliferation, whereas METAP2 overexpression promoted T-cell proliferation. Notably, METAP2 overexpression reversed the T. pallidum-induced inhibition of Jurkat T-cell proliferation. These results strongly suggest that T. pallidum inhibits CD4 + T-cell proliferation via METAP2.

Despite the important findings of this study, several limitations exist. First, although we used the largest syphilis GWAS dataset, the absence of a specific NS GWAS dataset, combined with the relatively low proportion of cases in the syphilis GWAS dataset, might introduce bias and reduce the statistical power to detect certain exposures. Second, the small sample size and limited variety of metabolites in the CSF GWAS dataset may have led to overlooking other relevant CSF metabolites, resulting in insufficient statistical power. Finally, functional validation of the predicted ceRNAs and their noncoding RNAs is lacking. Therefore, further studies are needed to elucidate the regulatory network of METAP2 and explore its potential as a diagnostic biomarker for NS, along with prospective clinical trials to validate the effectiveness of METAP2 and CSF metabolites as diagnostic and therapeutic targets.

Through mediation MR analyses, this study systematically revealed the critical roles of rQTLs and CSF metabolites in the pathogenesis of syphilis, particularly NS. Importantly, this study identified METAP2 as a key molecule in NS development and revealed that T. pallidum inhibits CD4 + T-cell proliferation by regulating METAP2. These findings not only provide new insights into the pathogenesis of NS but also offer an important theoretical foundation for the development of novel diagnostic and therapeutic strategies, particularly by establishing a genetically supported immune‒metabolic regulatory axis and identifying METAP2 as a mechanistically validated upstream target of CD4 + T-cell dysfunction.

Acknowledgements

We thank all participants and investigators involved in the UK Biobank study, the Wisconsin Alzheimer’s Disease Research Center (WADRC), the Wisconsin Registry for Alzheimer’s Prevention (WRAP), and the FinnGen study for sharing data. We specifically acknowledge Suhre K. et al. for the rQTL GWAS data from the UK Biobank, and Panyard DJ. et al. for providing the cerebrospinal fluid metabolomics GWAS data from WADRC and WRAP. Additionally, we appreciate the FinnGen study for the syphilis GWAS data.

Author contributions

ZL, XZ, and TL performed the research; FZ and YW contributed to the conception and design, financial support, administrative support, and provision of materials; ZL, TL, XD, HY, JY, and KG analyzed and interpreted the data; ZL, FZ and YW wrote and revised the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (no. 81971980), Major Scientific and Technological Projects for collaborative prevention and control of birth defects in Hunan Province (no. 2019SK1010), Hunan Province Natural Science Foundation (no. 2023JJ30530 and 2024JJ5343), Health High-Level Talents Major Scientific Research Project of Hunan Provincial Health Commission (R2023004), Health Research Key Project of Hunan Provincial Health Commission (no. 20221064723), Research funding of the First Affiliated Hospital of University of South China (no. 20230005-1015) and Scientific Research and Innovation Project of Postgraduates in Hunan Province (no. CX20220974).

Data availability

Data is provided within the manuscript or supplementary information files.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

Ai, C; Kong, L. CGPS: A machine learning-based approach integrating multiple gene set analysis tools for better prioritization of biologically relevant pathways. J Genet genomics = Yi Chuan Xue Bao; 2018; 45, pp. 489-504. [DOI: https://dx.doi.org/10.1016/j.jgg.2018.08.002] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30292791]

Bartleson, JM; Viehmann Milam, AA; Donermeyer, DL; Horvath, S; Xia, Y; Egawa, T; Allen, PM. Strength of tonic T cell receptor signaling instructs T follicular helper cell-fate decisions. Nat Immunol; 2020; 21, pp. 1384-1396. [DOI: https://dx.doi.org/10.1038/s41590-020-0781-7] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32989327][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7578106]

Birney E (2022) Mendelian randomization. Cold Spring Harbor Perspect Med 12. https://doi.org/10.1101/cshperspect.a041302

Datta, R; Choudhury, P; Bhattacharya, M; Soto Leon, F; Zhou, Y; Datta, B. Protection of translation initiation factor eIF2 phosphorylation correlates with eIF2-associated glycoprotein p67 levels and requires the lysine-rich domain I of p67. Biochimie; 2001; 83, pp. 919-931. [DOI: https://dx.doi.org/10.1016/s0300-9084(01)01344-x] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/11728629]

FinnGen (2024) Documentation of R10 release. https://finngen.gitbook.io/documentation/v/r10/

Giglione, C; Fieulaine, S; Meinnel, T. N-terminal protein modifications: bringing back into play the ribosome. Biochimie; 2015; 114, pp. 134-146. [DOI: https://dx.doi.org/10.1016/j.biochi.2014.11.008] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25450248]

Glaviano, A; Foo, ASC; Lam, HY; Yap, KCH; Jacot, W; Jones, RH; Eng, H; Nair, MG; Makvandi, P; Geoerger, B; Kulke, MH; Baird, RD; Prabhu, JS; Carbone, D; Pecoraro, C; Teh, DBL; Sethi, G; Cavalieri, V; Lin, KH; Javidi-Sharifi, NR; Toska, E; Davids, MS; Brown, JR; Diana, P; Stebbing, J; Fruman, DA; Kumar, AP. PI3K/AKT/mTOR signaling transduction pathway and targeted therapies in cancer. Mol Cancer; 2023; 22, 138. [DOI: https://dx.doi.org/10.1186/s12943-023-01827-6] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37596643][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10436543]

Goya Grocin, A; Kallemeijn, WW; Tate, EW. Targeting methionine aminopeptidase 2 in cancer, obesity, and autoimmunity. Trends Pharmacol Sci; 2021; 42, pp. 870-882. [DOI: https://dx.doi.org/10.1016/j.tips.2021.07.004] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34446297]

Hamill, MM; Ghanem, KG; Tuddenham, S. State-of-the-Art review: neurosyphilis. Clin Infect Diseases: Official Publication Infect Dis Soc Am; 2024; 78, pp. e57-e68. [DOI: https://dx.doi.org/10.1093/cid/ciad437]

Hamilton, F; Schurz, H; Yates, TA; Gilchrist, JJ; Möller, M; Naranbhai, V; Ghazal, P; Timpson, NJ; Parks, T; Pollara, G. Altered IL-6 signalling and risk of tuberculosis: a multi-ancestry Mendelian randomisation study. Lancet Microbe; 2025; 6, 100922. [DOI: https://dx.doi.org/10.1016/s2666-5247(24)00162-9] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39579785]

Hu, YT; Wu, KX; Wang, XT; Zhao, YY; Jiang, XY; Liu, D; Tong, ML; Liu, LL. Treponema pallidum promoted microglia apoptosis and prevented itself from clearing by human microglia via blocking autophagic flux. PLoS Pathog; 2023; 19, e1011594. [DOI: https://dx.doi.org/10.1371/journal.ppat.1011594] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37611054][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10446187]

Jafri, AJA; Arfuzir, NNN; Lambuk, L; Iezhitsa, I; Agarwal, R; Agarwal, P; Razali, N; Krasilnikova, A; Kharitonova, M; Demidov, V; Serebryansky, E; Skalny, A; Spasov, A; Yusof, APM; Ismail, NM. Protective effect of magnesium acetyltaurate against NMDA-induced retinal damage involves restoration of minerals and trace elements homeostasis. J Trace Elem Med Biology: Organ Soc Minerals Trace Elem (GMS); 2017; 39, pp. 147-154. [DOI: https://dx.doi.org/10.1016/j.jtemb.2016.09.005]

Leader, BT; Godornes, C; VanVoorhis, WC; Lukehart, SA. CD4 + lymphocytes and gamma interferon predominate in local immune responses in early experimental syphilis. Infect Immun; 2007; 75, pp. 3021-3026. [DOI: https://dx.doi.org/10.1128/iai.01973-06] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/17403876][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1932874]

Li, W; Li, S; Wang, J; Yu, M; Yang, H; He, Z; Tang, Y; Liu, J; Guo, N; Xie, D; Liu, Z; Zheng, K; Xu, M; Wu, Y. The outer membrane protein Tp92 of Treponema pallidum delays human neutrophil apoptosis via the ERK, PI3K/Akt, and NF-κB pathways. Mol Microbiol; 2023; 120, pp. 684-701. [DOI: https://dx.doi.org/10.1111/mmi.15164] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37718557]

Li, J; Ma, J; Liu, M; Li, M; Zhang, M; Yin, W; Wu, M; Li, X; Zhang, Q; Zhang, H; Zheng, H; Mao, C; Sun, J; Wang, W; Lyu, W; Yue, X; Weng, W; Li, J; Chen, F; Zhu, Y; Leng, L. Large-Scale proteome profiling identifies biomarkers associated with suspected neurosyphilis diagnosis. Advanced science (Weinheim. Baden-Wurttemberg Germany); 2024; 11, e2307744. [DOI: https://dx.doi.org/10.1002/advs.202307744]

Liu, LL; Zhu, SG; Jiang, XY; Ren, J; Lin, Y; Zhang, NN; Tong, ML; Zhang, HL; Zheng, WH; Fu, HJ; Luo, HJ; Lin, LR; Yan, JH; Yang, TC. LncRNA expression in CD4 + T cells in neurosyphilis patients. Front Cell Infect Microbiol; 2017; 7, 461. [DOI: https://dx.doi.org/10.3389/fcimb.2017.00461] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29167762][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5682391]

Liu, Y; Chen, J; Zhu, X; Tang, L; Luo, X; Shi, Y. Role of miR–449b–3p in endometriosis via effects on endometrial stromal cell proliferation and angiogenesis. Mol Med Rep; 2018; 18, pp. 3359-3365. [DOI: https://dx.doi.org/10.3892/mmr.2018.9341] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30066926][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6102748]

Liu, LL; Lin, Y; Chen, W; Tong, ML; Luo, X; Lin, LR; Zhang, HL; Yan, JH; Niu, JJ; Yang, TC. Metabolite profiles of the cerebrospinal fluid in neurosyphilis patients determined by untargeted metabolomics analysis. Front NeuroSci; 2019; 13, 150. [DOI: https://dx.doi.org/10.3389/fnins.2019.00150] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30863278][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6399405]

Liu, Z; Wang, Y; Xu, Z; Yuan, S; Ou, Y; Luo, Z; Wen, F; Liu, J; Zhang, J. Analysis of CeRNA networks and identification of potential drug targets for drug-resistant leukemia cell K562/ADR. PeerJ; 2021; 9, e11429. [DOI: https://dx.doi.org/10.7717/peerj.11429] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34113488][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8162247]

Liu, Z; Liang, Q; Ren, Y; Guo, C; Ge, X; Wang, L; Cheng, Q; Luo, P; Zhang, Y; Han, X. Immunosenescence: molecular mechanisms and diseases. Signal Transduct Target Therapy; 2023; 8, 200. [DOI: https://dx.doi.org/10.1038/s41392-023-01451-2]

Liu, Z; Zhang, X; Xiong, S; Huang, S; Ding, X; Xu, M; Yao, J; Liu, S; Zhao, F. Endothelial dysfunction of syphilis: pathogenesis. J Eur Acad Dermatology Venereology: JEADV; 2024; 38, pp. 1478-1490. [DOI: https://dx.doi.org/10.1111/jdv.19899]

Luo, J; Yang, H; Song, BL. Mechanisms and regulation of cholesterol homeostasis. Nat Rev Mol Cell Biol; 2020; 21, pp. 225-245. [DOI: https://dx.doi.org/10.1038/s41580-019-0190-7] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31848472]

Na, N; Zhao, D; Zhang, J; Wu, J; Miao, B; Li, H; Luo, Y; Tang, Z; Zhang, W; Bellanti, JA; Zheng, SG. Carbamylated erythropoietin regulates immune responses and promotes long-term kidney allograft survival through activation of PI3K/AKT signaling. Signal Transduct Target Therapy; 2020; 5, 194. [DOI: https://dx.doi.org/10.1038/s41392-020-00232-5]

Neurosyphilis. N Engl J Med 381: 1789. https://doi.org/10.1056/NEJMx190030

Osier, ND; Ziari, M; Puccio, AM; Poloyac, S; Okonkwo, DO; Minnigh, MB; Beers, SR; Conley, YP. Elevated cerebrospinal fluid concentrations of N-acetylaspartate correlate with poor outcome in a pilot study of severe brain trauma. Brain Injury; 2019; 33, pp. 1364-1371. [DOI: https://dx.doi.org/10.1080/02699052.2019.1641743] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31305157][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6675639]

Paganoni, S; Schwarzschild, MA. Urate as a marker of risk and progression of neurodegenerative disease. Neurotherapeutics: J Am Soc Experimental Neurother; 2017; 14, pp. 148-153. [DOI: https://dx.doi.org/10.1007/s13311-016-0497-4]

Panyard, DJ; Kim, KM; Darst, BF; Deming, YK; Zhong, X; Wu, Y; Kang, H; Carlsson, CM; Johnson, SC; Asthana, S; Engelman, CD; Lu, Q. Cerebrospinal fluid metabolomics identifies 19 brain-related phenotype associations. Commun Biology; 2021; 4, 63. [DOI: https://dx.doi.org/10.1038/s42003-020-01583-z]

Papadimitriou, N; Dimou, N; Tsilidis, KK; Banbury, B; Martin, RM; Lewis, SJ; Kazmi, N; Robinson, TM; Albanes, D; Aleksandrova, K; Berndt, SI; Timothy Bishop, D; Brenner, H; Buchanan, DD; Bueno-de-Mesquita, B; Campbell, PT; Castellví-Bel, S; Chan, AT; Chang-Claude, J; Ellingjord-Dale, M; Figueiredo, JC; Gallinger, SJ; Giles, GG; Giovannucci, E; Gruber, SB; Gsur, A; Hampe, J; Hampel, H; Harlid, S; Harrison, TA; Hoffmeister, M; Hopper, JL; Hsu, L; María Huerta, J; Huyghe, JR; Jenkins, MA; Keku, TO; Kühn, T; La Vecchia, C; Le Marchand, L; Li, CI; Li, L; Lindblom, A; Lindor, NM; Lynch, B; Markowitz, SD; Masala, G; May, AM; Milne, R; Monninkhof, E; Moreno, L; Moreno, V; Newcomb, PA; Offit, K; Perduca, V; Pharoah, PDP; Platz, EA; Potter, JD; Rennert, G; Riboli, E; Sánchez, MJ; Schmit, SL; Schoen, RE; Severi, G; Sieri, S; Slattery, ML; Song, M; Tangen, CM; Thibodeau, SN; Travis, RC; Trichopoulou, A; Ulrich, CM; van Duijnhoven, FJB; Van Guelpen, B; Vodicka, P; White, E; Wolk, A; Woods, MO; Wu, AH; Peters, U; Gunter, MJ; Murphy, N. Physical activity and risks of breast and colorectal cancer: a Mendelian randomisation analysis. Nat Commun; 2020; 11, 597. [DOI: https://dx.doi.org/10.1038/s41467-020-14389-8] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32001714][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6992637]

Pei, M; Xu, N; Li, J; Liu, X; Zhang, L; Xiao, W; Yu, Z; Yang, P; Peng, Y; Zhang, J; Hong, L; Wu, X; Tang, W; Lin, Z; Zhi, F; Li, G; Li, A; Liu, S; Chen, Y; Xiang, L; Lin, J; Wang, J. The LINC00501-HSP90B1-STAT3 positive feedback loop promotes malignant behavior in gastric cancer cells. Cell Signal; 2023; 108, 110711. [DOI: https://dx.doi.org/10.1016/j.cellsig.2023.110711] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37156452]

Pinna, G. Role of PPAR-Allopregnanolone signaling in behavioral and inflammatory Gut-Brain axis communications. Biol Psychiatry; 2023; 94, pp. 609-618. [DOI: https://dx.doi.org/10.1016/j.biopsych.2023.04.025] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37156350]

Qi, S; Xu, Y; Luo, R; Li, P; Huang, Z; Huang, S; Nie, T; Zhang, Q; Li, Q. Novel biochemical insights in the cerebrospinal fluid of patients with neurosyphilis based on a metabonomics study. J Mol Neuroscience: MN; 2019; 69, pp. 39-48. [DOI: https://dx.doi.org/10.1007/s12031-019-01320-0]

Shah, SZ; Hussain, T; Zhao, D; Yang, L. A central role for calcineurin in protein misfolding neurodegenerative diseases. Cell Mol Life Sci; 2017; 74, pp. 1061-1074. [DOI: https://dx.doi.org/10.1007/s00018-016-2379-7] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27682820]

Shah, K; Al-Haidari, A; Sun, J; Kazi, JU. T cell receptor (TCR) signaling in health and disease. Signal Transduct Target Therapy; 2021; 6, 412. [DOI: https://dx.doi.org/10.1038/s41392-021-00823-w]

Shen, W; Zhou, H; Wang, W; Li, W. Mendelian randomization study of lipid species reveals causal relationship with syphilis. AMB Express; 2025; 15, 63. [DOI: https://dx.doi.org/10.1186/s13568-025-01873-x] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/40205291][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11981995]

Sherman, BT; Hao, M; Qiu, J; Jiao, X; Baseler, MW; Lane, HC; Imamichi, T; Chang, W. DAVID: a web server for functional enrichment analysis and functional annotation of gene lists (2021 update). Nucleic Acids Res; 2022; 50, pp. W216-w221. [DOI: https://dx.doi.org/10.1093/nar/gkac194] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35325185][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9252805]

Shi, X; Shi, J; Zou, F; Cao, Q; Yan, X; Liu, S; Li, Y; Lan, X. Omics detection and treatment of syphilis. Clin Chim Acta; 2025; 565, 120008. [DOI: https://dx.doi.org/10.1016/j.cca.2024.120008] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39427935]

Skrivankova, VW; Richmond, RC; Woolf, BAR; Yarmolinsky, J; Davies, NM; Swanson, SA; VanderWeele, TJ; Higgins, JPT; Timpson, NJ; Dimou, N; Langenberg, C; Golub, RM; Loder, EW; Gallo, V; Tybjaerg-Hansen, A; Davey Smith, G; Egger, M; Richards, JB. Strengthening the reporting of observational studies in epidemiology using Mendelian randomization: the STROBE-MR statement. JAMA; 2021; 326, pp. 1614-1621. [DOI: https://dx.doi.org/10.1001/jama.2021.18236] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34698778]

Suhre, K. Genetic associations with ratios between protein levels detect new pQTLs and reveal protein-protein interactions. Cell Genomics; 2024; 4, 100506. [DOI: https://dx.doi.org/10.1016/j.xgen.2024.100506] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38412862][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10943581]

Tang, Y; Zhou, Y; He, B; Cao, T; Zhou, X; Ning, L; Chen, E; Li, Y; Xie, X; Peng, B; Hu, Y; Liu, S. Investigation of the immune escape mechanism of Treponema pallidum. Infection; 2023; 51, pp. 305-321. [DOI: https://dx.doi.org/10.1007/s15010-022-01939-z] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36260281]

Taylor, CT; Colgan, SP. Regulation of immunity and inflammation by hypoxia in immunological niches. Nat Rev Immunol; 2017; 17, pp. 774-785. [DOI: https://dx.doi.org/10.1038/nri.2017.103] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28972206][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5799081]

Ullah, R; Yin, Q; Snell, AH; Wan, L. RAF-MEK-ERK pathway in cancer evolution and treatment. Sem Cancer Biol; 2022; 85, pp. 123-154. [DOI: https://dx.doi.org/10.1016/j.semcancer.2021.05.010]

Vander Heiden, MG; Cantley, LC; Thompson, CB. Understanding the Warburg effect: the metabolic requirements of cell proliferation. Sci (New York NY); 2009; 324, pp. 1029-1033. [DOI: https://dx.doi.org/10.1126/science.1160809]

Verbanck, M; Chen, CY; Neale, B; Do, R. Detection of widespread horizontal Pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet; 2018; 50, pp. 693-698. [DOI: https://dx.doi.org/10.1038/s41588-018-0099-7] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29686387][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6083837]

Wakamatsu K, Chiba Y, Murakami R, Miyai Y, Matsumoto K, Kamada M, Nonaka W, Uemura N, Yanase K, Ueno M (2022) Metabolites and biomarker compounds of neurodegenerative diseases in cerebrospinal fluid. Metabolites 12. https://doi.org/10.3390/metabo12040343

Wei, W; Lyu, X; Markhard, AL; Fu, S; Mardjuki, RE; Cavanagh, PE; Zeng, X; Rajniak, J; Lu, N; Xiao, S; Zhao, M; Moya-Garzon, MD; Truong, SD; Chou, JC; Wat, LW; Chidambaranathan-Reghupaty, S; Coassolo, L; Xu, D; Shen, F; Huang, W; Ramirez, CB; Jang, C; Li, L; Svensson, KJ; Fischbach, MA; Long, JZ. PTER is a N-acetyltaurine hydrolase that regulates feeding and obesity. Nature; 2024; 633, pp. 182-188. [DOI: https://dx.doi.org/10.1038/s41586-024-07801-6] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39112712][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11374699]

Wormser, GP; Pavia, CS. Neurosyphilis. N Engl J Med; 2019; 381, pp. 2376-2377. [DOI: https://dx.doi.org/10.1056/NEJMc1914502] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31826357]

Wu, KX; Wang, XT; Hu, XL; Jiang, XY; Zhuang, JC; Xu, YZ; Lin, LR; Tong, ML; Yang, TC; Liu, LL. LncRNA-ENST00000421645 upregulates Kank1 to inhibit IFN-γ expression and promote T cell apoptosis in neurosyphilis. Front Microbiol; 2021; 12, 749171. [DOI: https://dx.doi.org/10.3389/fmicb.2021.749171] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34917045][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8669649]

Xie, JW; Guo, YF; Wang, M; Tong, ML; Zhu, XZ; Lin, LR. Syphilis susceptibility factors atlas: A wide-angled Mendelian randomization study. Prev Med; 2024; 185, 108033. [DOI: https://dx.doi.org/10.1016/j.ypmed.2024.108033] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38851401]

Xie, Q; Tang, Y; Shen, L; Yang, D; Zhang, J; Luo, Q. Immunophenotypic variations in syphilis: insights from Mendelian randomization analysis. Front Immunol; 2024; 15, 1380720. [DOI: https://dx.doi.org/10.3389/fimmu.2024.1380720] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38694502][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11061532]

Xiong, S; Liu, Z; Zhang, X; Huang, S; Ding, X; Zhou, J; Yao, J; Li, W; Liu, S; Zhao, F. Resurgence of syphilis: focusing on emerging clinical strategies and preclinical models. J Translational Med; 2023; 21, 917. [DOI: https://dx.doi.org/10.1186/s12967-023-04685-4]

Yang, Z; Yang, X; Gan, Z; Yuan, L; Liu, W; Si, C. Genome-Wide MicroRNA analysis of peripheral blood mononuclear cells reveals elevated miR-142-3p expression as a potential biomarker for secondary syphilis. Biomed Res Int; 2021; 2021, 5520053. [DOI: https://dx.doi.org/10.1155/2021/5520053] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34337017][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8317471]

Yoshimura, T; Benny, O; Bazinet, L; D’Amato, RJ. Suppression of autoimmune retinal inflammation by an antiangiogenic drug. PLoS ONE; 2013; 8, e66219. [DOI: https://dx.doi.org/10.1371/journal.pone.0066219] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23785488][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3681944]

Zhang, X; Duan, J; Wang, Y; Xie, B; Zhou, J; Zhao, S; Yin, W; Liu, P; Zhao, F. Insight into the invasion process and immune-protective evaluation of Tp0971, a membrane lipoprotein from Treponema pallidum. Microbiol Spectr; 2023; 11, e0004723. [DOI: https://dx.doi.org/10.1128/spectrum.00047-23] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37855609]

Zhu X, Zhu J (2020) CD4 T helper cell subsets and related human immunological disorders. Int J Mol Sci 21. https://doi.org/10.3390/ijms21218011