RELA as a Diagnostic Biomarker for Parkinson's Disease by Integrating Ferroptosis, Lipid Metabolism, and Neuroinflammation

RELA as a Diagnostic Biomarker for Parkinson's Disease by Integrating Ferroptosis, Lipid Metabolism, and Neuroinflammation | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article RELA as a Diagnostic Biomarker for Parkinson's Disease by Integrating Ferroptosis, Lipid Metabolism, and Neuroinflammation Qingfa Chen, Yajun Jing, Weiting Bu, Junli Zhang, Wenqing Liu, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8361812/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 11 Mar, 2026 Read the published version in Inflammation → Version 1 posted 9 You are reading this latest preprint version Abstract Parkinson’s disease (PD) is a progressive neurodegenerative disorder, in which ferroptosis and dysregulation of lipid metabolism are believed to play critical roles. However, the key genes and molecular mechanisms underlying these processes remain to be fully elucidated. We combined bioinformatics analyses with experimental validation to identify and characterize central genes involved in ferroptosis and lipid metabolism in PD. Integrated bioinformatics analysis identified 44 candidate genes associated with ferroptosis and lipid metabolism in PD. Machine learning refined these to three core genes: CBS, PRKAR2B, and RELA. Expression analysis revealed significant upregulation of CBS and RELA and downregulation of PRKAR2B in PD samples. ROC analysis indicated strong diagnostic potential for all three genes. Functional enrichment suggested their involvement in neuroinflammation, energy metabolism, and neuroprotective pathways. Immune infiltration analysis revealed significant correlations between these genes and specific immune cell types. In cellular models, knockdown of RELA attenuated MPP⁺-induced oxidative stress, ferroptosis, and inflammatory activation, while also restoring dopaminergic neuronal function. ELISA results from PD patient cerebrospinal fluid samples further confirmed the dysregulation of these genes, supporting their clinical relevance. This study identifies CBS, PRKAR2B, and RELA as key genes linking ferroptosis and lipid metabolism in Parkinson’s disease. These genes demonstrate strong diagnostic value and are closely associated with neuroinflammation and immune responses. Experimental validation underscores the protective effect of RELA knockdown against ferroptosis and inflammatory damage. Our findings suggest that RELA may serve as a promising diagnostic biomarker and therapeutic target for PD. Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Introduction Parkinson’s disease(PD) constitutes a major neurodegenerative disorder, clinically delineated by a triad of motor impairments collectively termed parkinsonism. These hallmark features include resting tremor, progressive muscular rigidity, bradykinesia(characterized by hypokinesia and delayed movement initiation), and postural instability with gait disturbances [ 1 , 2 ]. Epidemiological investigations demonstrate that PD exhibits a prevalence of approximately 1% in populations aged ≥ 60 years, with incidence rates ranging from 1–2 cases per 1,000 person-years in general populations [ 3 ]. At the molecular level, PD is pathognomically defined by the intraneuronal accumulation of α-synuclein(α-Syn) aggregates, forming Lewy bodies, concomitant with selective degeneration of dopaminergic neurons in the substantia nigra pars compacta[ 4 ]. Emerging mechanistic studies implicate multifactorial pathogenic pathways, including oxidative stress-mediated neuronal injury, mitochondrial dysfunction, lysosomal degradation impairment, and chronic neuroinflammatory processes in disease progression[ 5 ]. Research has found that many pathogenic risk factors for PD are associated with ferroptosis. For example, DJ-1 can negatively regulate ferroptosis by maintaining the biosynthesis of cysteine and GSH, while the SLC7A11 gene, which is significantly downregulated in PD, is also considered a key regulator of ferroptosis and can protect dopaminergic neurons by activating the transsulfuration pathway[ 6 ]. A recent study using human induced pluripotent stem cells(HiPSCs) showed that triploid mutations in SNCA(the gene encoding α-syn) and the accumulation of α-syn oligomers can specifically induce ferroptosis in differentiated neurons[ 7 ]. Beyond the high correlation with genetic factors, PD also exhibits typical pathological features related to ferroptosis, including excessive iron accumulation, lipid peroxidation, and redox homeostasis imbalance(changes in levels of glutamate-cystine antiporter(xCT), GSH, DJ-1, and coenzyme Q10(CoQ10), among others). In PD models induced by neurotoxins, such as 1-methyl-4-phenyl-pyridinium ion (MPP+) and 6-hydroxydopamine(6-OHDA)-treated human neuroblastoma (SH-SY5Y) cells and 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-damaged mice, obvious ferroptosis processes and characteristics are observed. Inhibitory therapeutic measures targeting ferroptosis can also significantly improve motor dysfunction and the loss of dopaminergic neurons in PD model mice[ 2 ]. Thus, ferroptosis plays an undeniable role in the genetics, pathological progression, pathogenesis, and potential treatment of PD. In this study, we systematically characterized the diagnostic biomarkers and elucidated the underlying mechanisms of ferroptosis-lipid metabolism-related genes in PD through comprehensive bioinformatics analysis. The findings were further validated using both cellular models and cerebrospinal fluid(CSF) samples from PD patients, demonstrating the translational potential of these molecular targets for both early diagnosis and therapeutic intervention in PD. Materials and Methods Data acquisition and preprocessing Gene expression datasets(GSE49036, GSE20292, GSE7621, and GSE20164) were obtained from the Gene Expression Omnibus(GEO). A total of 74 samples(42 PD and 32 controls) from GSE49036, GSE20292, and GSE7621 were merged and normalized. The ComBat algorithm was applied to remove batch effects. The dataset GSE20164 was used as an independent validation set. Identification of differentially expressed genes(DEGs) DEGs between PD and control samples were identified using the limma R package, with an adjusted p-value 0 set as significance thresholds. Weighted gene co-expression network analysis(WGCNA) WGCNA was performed to identify gene modules significantly associated with PD. A soft-thresholding power of 4 was selected to ensure a scale-free network. Modules with |r|>0.4 and p < 0.05 were considered clinically significant. Screening of ferroptosis and lipid metabolism-related genes Ferroptosis-related genes(FRGs) and lipid metabolism-related genes(LMRGs) were retrieved from public databases. Venn analysis was used to identify overlapping genes among DEGs, WGCNA module genes, FRGs, and LMRGs. Protein-protein interaction(PPI) network and functional enrichment The PPI network was constructed using the STRING database(confidence score > 0.4) and visualized in Cytoscape. Functional enrichment analysis of GO terms and KEGG pathways was performed using the clusterProfiler R package. Machine learning-based feature selection LASSO regression and random forest algorithms were applied to screen key diagnostic genes[ 8 ]. The optimal lambda(λ) value in LASSO was determined via 10-fold cross-validation. The top 10 genes ranked by mean decrease accuracy were selected from the random forest model. The intersection of genes identified by both methods was considered the final key gene set. Gene expression and diagnostic validation Expression levels of key genes were validated in the independent dataset GSE20164. Receiver operating characteristic(ROC) curves were generated to evaluate diagnostic performance. Immune infiltration analysis To investigate the immune microenvironment in PD, immune cell infiltration abundances were estimated from the transcriptomic data using the CIBERSORT algorithm. The relative proportions of 22 immune cell types were calculated for each sample. Differences in immune cell infiltration between PD and control groups were assessed using the Wilcoxon rank-sum test, with a significance threshold of P 0.5 and P < 0.01 were considered statistically significant. Cell culture, transfection and cell death analysis Human SH-SY5Y cells were purchased from the Cell Bank/Stem Cell Bank, Chinese Academy of Sciences (Shanghai, China). Briefly, the cells were cultured in high-glucose DMEM (Thermo Fisher Scientific, CA, USA) supplemented with 10% fetal bovine serum (Gibco, Life Technologies) in an incubator at 37°C and 5% CO2. SH-SY5Y cells were treated with MPP + to establish PD models. RELA knockdown was performed using siRNA transfection. Cell viability was assessed using the Cell Counting Kit-8 (CCK-8) assay. Briefly, cells were seeded in 96-well plates and after treatment, the CCK-8 reagent was added to each well. Following incubation at 37°C for 2 hours, the absorbance was measured at a wavelength of 450 nm using a microplate reader. Apoptosis was quantified by flow cytometry with an Annexin V-PE/7-AAD apoptosis detection kit. Cells were stained with Annexin V-PE and 7-AAD in the dark for 15 minutes and then analyzed to distinguish early and late apoptotic populations. Intracellular ROS levels were determined using the fluorescent probe DCFH-DA. Cells were loaded with DCFH-DA for 30 minutes at 37°C. After washing, the fluorescence intensity was immediately measured using flow cytometry. To assess oxidative stress, the content of malondialdehyde (MDA) and the activity of superoxide dismutase (SOD) were measured using commercial enzyme-linked immunosorbent assay (ELISA) kits, strictly following the manufacturer's protocols. Furthermore, to directly investigate ferroptosis, intracellular Fe²⁺ levels were determined using a specific iron assay kit according to the manufacturer's instructions. Western blot Western blot analysis was conducted as described in our previous study [ 9 ]. Whole-cell lysates were extracted using radioimmunoprecipitation assay (RIPA) lysis buffer supplemented with a protease inhibitor cocktail (Sigma, USA). The proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS‒PAGE) and then transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore, Bedford, USA). The membranes were incubated with primary antibodies against Achaete-scute family bHLH transcription factor 4(ASCL4) (1:1000), CD40 (1:1000), C-X-C motif chemokine ligand 6 (CXCL6)(1:1000), C-X-C motif chemokine ligand 9 (CXCL9)(1:1000), Glutathione peroxidase 4 (GPX4)(1:1000), Transferrin receptor (TFRC)(1:1000), and Tyrosine hydroxylase (TH)(1:1000) at 4°C overnight. The membranes were then probed with an HRP-conjugated secondary antibody (Beyotime, Shanghai, China) at 25°C for 1 h, followed by washing with Tris Buffered Saline and 0.05% Tween 20 (TBST) three times for 10 min each. Finally, the membranes were visualized and imaged with an enhanced chemiluminescence system. With GAPDH as the loading control, the expression of proteins was quantified using ImageJ Pro Plus 6.0 Software. ELISA validation using human CSF samples CSF samples were collected from 10 PD patients and 10 controls(patients with facial spasm). Protein levels of CBS, PRKAR2B, and RELA were quantified using commercial ELISA kits according to manufacturers’ instructions. Statistical differences were assessed using Student’s t-test, and ROC analysis was performed to evaluate diagnostic utility. The study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of the Institutional Review Board of Jinan Central Hospital. Statistical analysis All statistical analyses were conducted in R(v4.2.1). Differential expression was analyzed using the limma package. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.001 were considered statistically significant. Results 1. Identification of ferroptosis and lipid metabolism-related core genes in PD The schematic representation of the experimental workflow is presented in Fig. 1 . Before going further, batch effect correction was performed on 74 samples(42 PD and 32 controls) from datasets GSE49036, GSE20292, and GSE7621 using the ComBat algorithm. After normalization, the median expression values across datasets showed high consistency(Fig. 2 A), indicating successful removal of batch effects. Differential expression analysis with the limma package identified 1,493 significantly dysregulated genes in PD compared to controls(P 0)(Fig. 2 B). WGCNA was applied to identify co-expression modules associated with PD, using a soft-thresholding power of 4(Fig. 2 C). Seventeen modules were identified, among which the blue, brown, and royalblue modules demonstrated significant correlation with disease status(|r|>0.4, P < 0.05)(Fig. 2 D), yielding 4376 core module genes. By integrating 382 ferroptosis-related genes and 867 lipid metabolism-related genes from public databases, Venn analysis revealed 20 ferroptosis-related and 60 lipid metabolism-related DEGs. Correlation analysis further identified 44 candidate genes significantly associated with both processes(|r|>0.6, P < 0.05)(Fig. 2 E-F), which were selected for subsequent analyses. 2. Functional enrichment and protein interaction analysis of candidate genes A protein-protein interaction network was constructed using the STRING database(confidence threshold: 0.4), resulting in 34 interaction pairs. The network was visualized in Cytoscape(Fig. 3 A). Functional enrichment analysis via clusterProfiler revealed significant involvement of the candidate genes in 331 biological processes(BP), 26 cellular components(CC), 76 molecular functions(MF), and 24 KEGG pathways. Key enriched BPs included biosynthesis and metabolic processes related to alcohols, terpenoids, aerobic respiration, TCA cycle, and sphingolipid/phospholipid metabolism. CC terms were associated with autolysosomes, autophagosomes, organelle lumens, and mitochondrial membrane complexes. MF analysis highlighted enzymatic activities such as acetylgalactosamine transferase, lipid kinase, and oxidoreductase, as well as ferrous iron and NADP/NADPH binding. KEGG pathway analysis indicated enrichment in pyruvate metabolism, TCA cycle, inositol phosphate metabolism, cysteine and methionine metabolism, and terpenoid backbone biosynthesis, among others(Fig. 3 B–E). 3. Machine learning-based identification of diagnostic biomarkers LASSO regression and random forest were employed to refine candidate genes. The LASSO model identified five feature genes with non-zero coefficients at the optimal λ value(λ.min = 0.1428043)(Fig. 4 A-B). Random forest analysis ranked genes by importance using mean decrease accuracy, and the top 10 genes were selected(Fig. 4 C-D). The intersection of genes identified by both methods yielded three key genes: CBS, PRKAR2B, and RELA (Fig. 4 E). 4. Differential expression and diagnostic performance of key genes Expression analysis of the three key genes was conducted in both the training and validation sets(GSE20164). CBS and RELA were significantly upregulated in PD samples, while PRKAR2B was downregulated(Fig. 5 A–B). ROC curve analysis demonstrated strong diagnostic ability for all three genes, with AUC values exceeding 0.75 in both cohorts(Fig. 5 C–D). 5. Pathway enrichment analysis of key genes Gene Set Enrichment Analysis(GSEA) revealed that CBS was primarily associated with the Notch signaling pathway, ribosome, spliceosome, and pathways in cancer(Fig. 6 A). PRKAR2B was enriched in glycolysis/gluconeogenesis, JAK-STAT signaling, oxidative phosphorylation, pyruvate metabolism, and TCA cycle(Fig. 6 B). RELA was significantly linked to apoptosis, chemokine signaling, JAK-STAT pathway, oxidative phosphorylation, Toll-like receptor signaling, and cytokine-cytokine receptor interactions(Fig. 6 C). 6. Immune infiltration combined with proteomics data reveals the mechanism of inflammatory response in PD To investigate the immune microenvironment in PD samples and the role of key genes, immune cell infiltration analysis was performed. The results revealed significant differences in the infiltration levels of 12 immune cell types between PD and control samples(Fig. 7 A). Among these, dendritic cells, effector memory CD8⁺T cells, MDSCs, natural killer T cells, helper T cells, macrophages, B cells, mast cells, and neutrophils were significantly upregulated in PD samples, indicating a disrupted immune microenvironment and suggesting the involvement of specific immune cells in disease progression. Spearman correlation analysis was conducted to examine the relationship between the key genes(CBS, PRKAR2B, RELA) and the differentially infiltrated immune cells. The results demonstrated that CBS was strongly correlated with follicular helper T cells(cor = 0.62, P < 0.01). PRKAR2B showed a strong correlation with immature dendritic cells(cor = 0.73, P < 0.01)(Fig. 7 B). RELA was significantly correlated with multiple immune cells, including follicular helper T cells(cor = 0.55, P < 0.01), plasmacytoid dendritic cells(cor = 0.56, P < 0.01), natural killer T cells(cor = 0.54, P < 0.01), myeloid-derived suppressor cells(cor = 0.53, P < 0.01), mast cells(cor = 0.58, P < 0.01), and effector memory CD8⁺T cells(cor = 0.54, P < 0.01)(Fig. 7 B). These findings suggest that CBS, PRKAR2B, and RELA may play important roles in regulating immune responses in PD. ELISA was performed on CSF samples from 10 PD patients and 10 controls with facial spasm. Consistent with bioinformatic predictions, CBS and RELA protein levels were significantly elevated in PD, while PRKAR2B was reduced(Fig. 7 C-E). ROC analysis based on CSF protein expression confirmed the diagnostic utility of all three markers, with AUC values above 0.7(Fig. 7 F). 7. RELA participates in the pathological process of PD by regulating oxidative stress and ferroptosis Given that RELA is known to be a core transcription factor of NF-κB, and the NF-κB pathway is a central axis of neuroinflammation in PD. Through clinical cerebrospinal fluid proteomic analysis, it was discovered that inflammation-related differentially expressed proteins in PD patients form a synergistic network and exhibit significant overlap with RELA enrichment pathways. To further verify the regulatory role of RELA on PD-related pathological phenotypes at the cellular level, RELA was first knocked down in cells. The relative mRNA expression level of RELA was detected by qRT-PCR to evaluate the knockdown efficiency of si-RELA (Fig. 8 A). SH-SY5Y cells were induced with MPP + to construct a PD cell model. Compared with the Control group, cell viability was significantly reduced after MPP + treatment (Fig. 8 B, P < 0.001), indicating successful simulation of neurotoxic damage in PD and the establishment of a PD cell model. Compared with the MPP + + si-NC group, cell viability in the MPP + + si-RELA group was significantly restored (Fig. 8 B, P < 0.001). This suggests that knocking down RELA can alleviate MPP+-induced cytotoxicity. Flow cytometry showed that compared with the Control group (3.67%), the proportion of apoptotic cells significantly increased after MPP + treatment (11.29%). The proportion of apoptotic cells in the MPP + + si-RELA group (7.8%) was significantly lower than that in the MPP + + si-NC group (11.49%) (Fig. 8 C). This indicates that knocking down RELA can inhibit MPP+-induced cell apoptosis. 8. Effects of RELA knockdown on oxidative stress, ferroptosis, and dopaminergic neuronal function in PD cell models Oxidative stress is a core pathologica characteristic in PD neuronal injury. This study investigated the regulatory role of RELA in oxidative stress by detecting ROS, the lipid peroxidation product malondialdehyde (MDA), and the antioxidant enzyme superoxide dismutase (SOD). Flow cytometry showed significant shifts in the fluorescence intensity distribution peaks among the groups. Compared to the Control group, the Mean X was significantly increased in the MPP⁺ group, while the Mean X in the MPP⁺+si-RELA group was significantly lower than that in the MPP⁺+si-NC group (Fig. 9 A). MDA is a characteristic product of lipid peroxidation, and its content directly reflects the degree of lipid damage. ELISA detection of MDA content revealed that compared to the Control group, MPP⁺ treatment significantly increased MDA content (Fig. 9 B, P < 0.01), whereas the MDA content in the MPP⁺+si-RELA group was significantly decreased compared to the MPP⁺+si-NC group (Fig. 9 B, P < 0.05), suggesting that RELA knockdown can alleviate MPP⁺-induced lipid peroxidation damage. SOD is a key intracellular antioxidant enzyme responsible for scavenging superoxide anion radicals. ELISA detection of SOD activity showed that the Control group had the highest SOD activity. After MPP⁺ treatment, SOD activity was significantly reduced (Fig. 9 C, P < 0.001), but it was significantly restored in the MPP⁺+si-RELA group compared to the MPP⁺+si-NC group (Fig. 9 C, P < 0.01), indicating that knocking down RELA enhances cellular antioxidant defense capacity. Ferroptosis is a form of programmed cell death dependent on lipid peroxidation, and the intracellular Fe 2+ level is a key factor initiating ferroptosis. The regulatory role of RELA was explored by detecting core indicators of ferroptosis. Results from the iron ion assay kit showed that intracellular Fe 2+ content significantly increased after MPP + treatment, while the Fe 2+ content in the MPP⁺+si-RELA group was lower than that in the MPP + +si-NC group (Fig. 9 D, P < 0.05). This suggests that RELA knockdown can inhibit MPP + -induced Fe 2+ accumulation, thereby regulating the ferroptosis process. The expression of inflammatory chemokines (CD40, CXCL6, CXCL9), ferroptosis-related proteins (GPX4, ACSL4, TFRC), and the dopaminergic neuronal marker (TH) was detected by Western blot. After MPP + treatment, the expression of CD40, CXCL6, and CXCL9 was significantly upregulated (Fig. 9 E, J-I, P < 0.01). In the MPP + +si-RELA group, the expression of these genes was significantly reduced compared to the MPP + +si-NC group (Fig. 9 E, J-I, P < 0.05). Compared to the Control group, the expression of the ferroptosis-related proteins ACSL4 and TFRC was significantly increased after MPP + treatment (Fig. 9 E-F, K, P < 0.001), and their expression in the MPP + +si-RELA group was lower than that in the MPP + +si-NC group (Fig. 9 E-F, K, P < 0.05). Compared to the Control group, GPX4 expression was significantly decreased after MPP + treatment (Fig. 9 J, P < 0.001), but its expression in the MPP + +si-RELA group was higher than that in the MPP + +si-NC group (Fig. 9 J, P < 0.01). TH is the key rate-limiting enzyme for dopamine synthesis, and its expression reflects dopaminergic neuronal function. As shown in Fig. 9 E, L, TH expression was highest in the Control group, significantly decreased after MPP + treatment (P < 0.001), and significantly increased in the MPP + +si-RELA group compared to the MPP + +si-NC group (P < 0.05). In summary, these results indicate that knocking down RELA can inhibit MPP + -induced inflammatory chemokine expression, regulate ferroptosis, and restore dopaminergic neuronal function. Discussion Parkinson's disease is a complex neurodegenerative disorder whose pathogenesis involves multiple genetic and metabolic pathways[ 10 ]. In this study, we identified three key genes-CBS, PRKAR2B, and RELA-that are closely associated with ferroptosis and lipid metabolism, demonstrating significant diagnostic value in both bioinformatic predictions and experimental validation. The integration of multiple transcriptomic datasets revealed 44 candidate genes significantly associated with both ferroptosis and lipid metabolism. Functional enrichment analysis indicated their strong involvement in critical metabolic pathways including the TCA cycle, pyruvate metabolism, and sphingolipid/phospholipid metabolism, suggesting a tight coupling between energy metabolism dysregulation and ferroptosis susceptibility in PD. The convergence of these pathways highlights the complex metabolic rewiring that characterizes PD pathology. Importantly, the diagnostic potential of these biomarkers was validated across multiple levels. Both transcriptomic and proteomic analyses consistently showed CBS and RELA upregulation and PRKAR2B downregulation in PD. The strong AUC values (> 0.75 in transcriptomic data, > 0.7 in CSF protein levels) across independent cohorts support their robustness as diagnostic markers. Among these, CBS attracts particular attention due to its dual role in neuroprotection and neuroinflammation regulation[ 11 – 13 ]. As the rate-limiting enzyme in the transsulfuration pathway, CBS catalyzes the conversion of homocysteine to cysteine. Its deficiency is closely linked to hyperhomocysteinemia, which has been implicated in neurodegeneration[ 12 ]. More importantly, CBS is a major endogenous producer of hydrogen sulfide(H 2 S), a gaseous neuromodulator that exhibits anti-inflammatory and neuroprotective properties. Previous studies have shown that H 2 S donors can inhibit neuroinflammation through modulation of the NF-κB and AMPK pathways, and exert protective effects in a mouse model of PD[ 14 , 15 ]. Our findings align with these reports, showing significant upregulation of CBS in PD samples. This elevation may reflect a compensatory mechanism against ongoing neuroinflammation and oxidative stress. Notably, recent evidence suggests that microglial-specific deletion of CBS exacerbates NLRP3 inflammasome activation and dopaminergic neurodegeneration, whereas its overexpression or H 2 S supplementation attenuates these effects[ 11 ]. Therefore, CBS may serve not only as a biomarker but also as a promising therapeutic target for modulating neuroinflammation in PD. In contrast, PRKAR2B was significantly downregulated in PD patients. As a regulatory subunit of protein kinase A(PKA), PRKAR2B is involved in cAMP signaling, which plays crucial roles in neuronal survival, synaptic plasticity, and neurotransmitter release [ 16 ]. Although its specific function in PD remains underexplored, the dysregulation of PKA signaling has been implicated in Rett syndrome[ 17 ] and several neurodegenerative diseases[ 18 ]. The observed downregulation may indicate impaired cAMP-mediated neuroprotection, contributing to dopaminergic neuron vulnerability. Further studies are warranted to elucidate the mechanistic role of PRKAR2B in PD pathogenesis. RELA (p65), a core component of the NF-κB pathway, was significantly upregulated in PD, consistent with its well-established role in regulating inflammatory responses[ 19 ]. NF-κB activation is known to drive the expression of pro-inflammatory cytokines and chemokines, contributing to chronic neuroinflammation and neuronal death in PD[ 20 ]. Our cellular experiments demonstrated that RELA knockdown significantly alleviated MPP⁺-induced cytotoxicity, apoptosis, oxidative stress, and ferroptosis. The reduction in ROS, MDA, and Fe²⁺ accumulation, coupled with restored SOD activity and GPX4 expression, provides compelling evidence that RELA modulates redox homeostasis and ferroptosis sensitivity. Furthermore, RELA knockdown suppressed inflammatory chemokine expression (CD40, CXCL6, CXCL9) and restored tyrosine hydroxylase (TH) levels, indicating its dual role in regulating neuroinflammation and dopaminergic function. Our GSEA results further support its involvement in immune-related pathways such as cytokine-cytokine receptor interaction, Toll-like receptor signaling, and the JAK-STAT pathway. The immune infiltration analysis further revealed extensive correlations between these key genes and specific immune cell populations. The broad association of RELA with multiple immune cells—including follicular helper T cells, plasmacytoid dendritic cells, and myeloid-derived suppressor cells—underscores its central role in shaping the neuroimmune landscape in PD. The consistent upregulation of various immune cell types in PD samples confirms immune dysregulation as a hallmark of PD pathology. Several limitations should be acknowledged. First, although we integrated multiple datasets and validated results in an independent cohort and human CSF samples, the sample size remains moderate. Future studies with larger clinical cohorts are necessary to strengthen the generalizability of our findings. Second, The precise mechanisms through which CBS and PRKAR2B regulate the cross-talk between ferroptosis and lipid metabolism require further investigation, particularly using cell and animal models. Finally, longitudinal studies would help evaluate whether these biomarkers can reflect disease progression or treatment response. In conclusion, our study systematically identified and validated CBS, PRKAR2B, and RELA as key genes linking ferroptosis and lipid metabolism dysregulation in PD. Our study also confirm that the pathways regulated by RELA align with the enrichment pathways of differentially expressed inflammatory proteins in the cerebrospinal fluid of PD patients. This provides direct evidence supporting RELA as a clinically relevant target. Declarations Author Contributions Q.C., C.S. and D.S. contributed to the study design and conceptual framework. Q.C., Y.J., W.B., and W.L. conducted data acquisition, performed statistical analysis, and interpreted the results. Technical support was provided by Y.J., W.B., and W.L. Manuscript composition and critical revision were completed by Q.C., C.S., C.L. and D.S. All authors read and approved the final manuscript. Clinical trial number Not applicable. Conflicts of Interest The authors declare no conflicts of interest. Data Availability Statement The data that support the findings of this study are available in the methods of this article. Funding Declaration This study was supported by Shandong Provincial Medical and Health Science and Technology Development Plan Project (No. 202204040490), Jinan Municipal Health Commission Science and Technology Plan Project (NO.2024202003). References Khan, M. S., S. Nasiripour, and J. C. Bopassa. 2025. Parkinson Disease Signaling Pathways, Molecular Mechanisms, and Potential Therapeutic Strategies: A Comprehensive Review. International journal of molecular sciences ; 26. Gao, X. D., J. E. Ding, J. X. Xie, and H. M. Xu. 2025. Epigenetic regulation of iron metabolism and ferroptosis in Parkinson's disease: Identifying novel epigenetic targets. Acta pharmacologica Sinica 46:2075–2092. Tysnes, O. B., and A. Storstein. 1996. Epidemiology of Parkinson's disease. Journal of neural transmission (Vienna, Austria: 2017; 124:901–905. Gan, Y. H., L. Z. Ma, Y. Zhang, J. You, Y. Guo, and Y. He et al. 2025. Large-scale proteomic analyses of incident Parkinson's disease reveal new pathophysiological insights and potential biomarkers. Nature aging 5:642–657. Kia, D. A., D. Zhang, S. Guelfi, C. Manzoni, L. Hubbard, and R. H. Reynolds et al. 2021. Identification of Candidate Parkinson Disease Genes by Integrating Genome-Wide Association Study, Expression, and Epigenetic Data Sets. JAMA neurology 78:464–472. Fan, H., Y. Li, J. Huang, J. Jiang, F. Feng, and N. Huang et al. 2025. DJ-1 in Parkinson's disease: Its important role at endoplasmic reticulum-mitochondria contact sites. Behavioural brain research 495:115775. Angelova, P. R., M. L. Choi, A. V. Berezhnov, M. H. Horrocks, C. D. Hughes, and S. De et al. 2020. Alpha synuclein aggregation drives ferroptosis: an interplay of iron, calcium and lipid peroxidation. Cell death and differentiation 27:2781–2796. Xi, L. J., Z. Y. Guo, X. K. Yang, and Z. G. Ping. 2023. [Application of LASSO and its extended method in variable selection of regression analysis]. Zhonghua yu fang yi xue za zhi [Chinese journal of preventive medicine] ; 57:107–111. Wang, H., Z. Zhou, H. Zhong, S. Wang, K. Shen, and R. Huang et al. 2025. A pyrimidine metabolism-related gene signature for prognosis prediction and immune microenvironment description of breast cancer. Journal of translational medicine 23:698. Diaz-Ortiz, M. E., Y. Seo, M. Posavi, M. Carceles Cordon, E. Clark, and N. Jain et al. 2022. GPNMB confers risk for Parkinson's disease through interaction with α-synuclein . vol. 377 eabk0637. New York, NY: Science. Mou, Y. J., Y. T. Ma, X. Yuan, M. Wang, Y. Liu, and C. S. Pei et al. 2023. Cystathionine β-Synthase Suppresses NLRP3 Inflammasome Activation via Redox Regulation in Microglia . Antioxidants & redox signaling. Liu, Y., Z. Wang, Y. T. Ma, Y. X. Chen, X. Y. Qian, and X. O. Hou et al. 2025. STAT1 and STAT6 orchestrate Cbs transcription and transsulfur metabolism in microglia and contribute to parkinson's disease-related neuroinflammation. Cellular and molecular life sciences: CMLS. ; 82:294. Lee, M., C. Schwab, S. Yu, E. McGeer, and P. L. McGeer. 2009. Astrocytes produce the antiinflammatory and neuroprotective agent hydrogen sulfide. Neurobiology of aging 30:1523–1534. Kida, K., M. Yamada, K. Tokuda, E. Marutani, M. Kakinohana, and M. Kaneki et al. 2011. Inhaled hydrogen sulfide prevents neurodegeneration and movement disorder in a mouse model of Parkinson's disease. Antioxidants & redox signaling 15:343–352. Hu, L. F., M. Lu, P. T. Hon Wong, and J. S. Bian. 2011. Hydrogen sulfide: neurophysiology and neuropathology. Antioxidants & redox signaling 15:405–419. Wang, Y., S. Cao, D. Tone, H. Fujishima, R. G. Yamada, and R. I. Ohno et al. 2024. Postsynaptic competition between calcineurin and PKA regulates mammalian sleep-wake cycles. Nature 636:412–421. Weise, S. C., G. Arumugam, A. Villarreal, P. Videm, S. Heidrich, and N. Nebel et al. 2019. FOXG1 Regulates PRKAR2B Transcriptionally and Posttranscriptionally via miR200 in the Adult Hippocampus. Molecular neurobiology 56:5188–5201. Benjamin-Zukerman, T., V. Pane, R. Safadi-Safa, M. Solomon, V. Lev-Ram, and M. Aboraya et al. Allosteric modulation of protein kinase A in individuals affected by NLPD-PKA, a neurodegenerative disease in which the PRKAR1B L50R variant is expressed. The FEBS journal 2025. Figat, M., A. Wisniewska, J. Plichta, J. Milkowska-Dymanowska, S. Majewski, and M. S. Karbownik et al. 2025. Asthma and COPD Beyond the Airways: Exploring Neurocognitive Links Through NF-κB Subunits c-Rel and p65. International journal of molecular sciences ; 26. Si, Y., S. Yan, X. Li, W. Ding, X. Zhang, and K. Jia et al. 2025. P2X7R activation promotes ferroptosis in dopaminergic neurons via NF-κB signaling pathway in vitro and in vivo models of MPP(+)/MPTP-induced Parkinson's disease. Brain research 1865:149824. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 11 Mar, 2026 Read the published version in Inflammation → Version 1 posted Editorial decision: Revision requested 07 Jan, 2026 Reviews received at journal 07 Jan, 2026 Reviews received at journal 05 Jan, 2026 Reviewers agreed at journal 19 Dec, 2025 Reviewers agreed at journal 18 Dec, 2025 Reviewers invited by journal 18 Dec, 2025 Editor assigned by journal 15 Dec, 2025 Submission checks completed at journal 15 Dec, 2025 First submitted to journal 14 Dec, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8361812","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":563516578,"identity":"0c9152c1-9d63-444c-8199-4e30a8dac5af","order_by":0,"name":"Qingfa Chen","email":"","orcid":"","institution":"Jinan Central Hospital Affiliated to Shandong First Medical University","correspondingAuthor":false,"prefix":"","firstName":"Qingfa","middleName":"","lastName":"Chen","suffix":""},{"id":563516579,"identity":"4a343359-1f14-4da6-b0da-0d9001cd1381","order_by":1,"name":"Yajun Jing","email":"","orcid":"","institution":"The First Affiliated Hospital, Fujian Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yajun","middleName":"","lastName":"Jing","suffix":""},{"id":563516580,"identity":"6377f714-734b-4a90-87a1-b4ffe9f95499","order_by":2,"name":"Weiting Bu","email":"","orcid":"","institution":"Jinan Central Hospital Affiliated to Shandong First Medical University","correspondingAuthor":false,"prefix":"","firstName":"Weiting","middleName":"","lastName":"Bu","suffix":""},{"id":563516581,"identity":"ec30bcce-7430-4682-9411-0020c6867e6d","order_by":3,"name":"Junli Zhang","email":"","orcid":"","institution":"Shandong University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Junli","middleName":"","lastName":"Zhang","suffix":""},{"id":563516582,"identity":"894a8de5-cad6-4792-936d-e3a2d130c900","order_by":4,"name":"Wenqing Liu","email":"","orcid":"","institution":"Jinan Central Hospital Affiliated to Shandong First Medical University","correspondingAuthor":false,"prefix":"","firstName":"Wenqing","middleName":"","lastName":"Liu","suffix":""},{"id":563516583,"identity":"83600ffb-8da1-4bba-aeea-041e8238de6b","order_by":5,"name":"Chuanying Shi","email":"","orcid":"","institution":"Jinan Central Hospital Affiliated to Shandong First Medical University","correspondingAuthor":false,"prefix":"","firstName":"Chuanying","middleName":"","lastName":"Shi","suffix":""},{"id":563516584,"identity":"b1be2685-81eb-4632-90eb-e7c304f92e22","order_by":6,"name":"Cheng Liu","email":"","orcid":"","institution":"Liaocheng University","correspondingAuthor":false,"prefix":"","firstName":"Cheng","middleName":"","lastName":"Liu","suffix":""},{"id":563516585,"identity":"1bb49636-023d-482e-9e78-366f43cd1b28","order_by":7,"name":"Daoqing Su","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyElEQVRIiWNgGAWjYDCCA0CcAGKwNzY+/ECaFp7DzcYSRGsBA4n0NgEeYnTwHT988MbDHbX2BjcftjFIMNjJ6TYQ0CJ5Ji3ZIvHM8cQNtxPbHhQwJBubHSCgxeBAjplEYtuxBIPbie0GEgwHErcR1HL+DVgL0GEH2yR4iNJyA2xLDeOGG4xEapG88Qzol7YDiTPPJAID2YAIv/CdTz5482dbnT3f8eMPH36osJMjqAUEgBF4GOZOIpRDtdQRqXQUjIJRMApGJAAAmZ9KeycPsEoAAAAASUVORK5CYII=","orcid":"","institution":"Jinan Central Hospital Affiliated to Shandong First Medical University","correspondingAuthor":true,"prefix":"","firstName":"Daoqing","middleName":"","lastName":"Su","suffix":""}],"badges":[],"createdAt":"2025-12-15 04:53:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8361812/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8361812/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10753-026-02478-7","type":"published","date":"2026-03-11T15:59:35+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":98821129,"identity":"7a69ea4d-76f6-47a6-9880-3b963217a1eb","added_by":"auto","created_at":"2025-12-22 17:20:29","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5346176,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/d6028edf496d2c96d8e89a51.docx"},{"id":98821122,"identity":"96003e26-e6b2-421e-a44a-3b16465d0237","added_by":"auto","created_at":"2025-12-22 17:20:29","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9100,"visible":true,"origin":"","legend":"","description":"","filename":"e367af6bb63d45d79ab917512214522f.json","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/22e8f29ffb22dcf780a6ce39.json"},{"id":98821124,"identity":"7aa08427-1ac8-4bc9-b973-b10a3795b023","added_by":"auto","created_at":"2025-12-22 17:20:29","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":92053,"visible":true,"origin":"","legend":"","description":"","filename":"e367af6bb63d45d79ab917512214522f1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/09a63d6397d3595754bebe7d.xml"},{"id":99307994,"identity":"3fb01c3d-87b2-49cf-84ed-9738b772c126","added_by":"auto","created_at":"2025-12-31 16:07:22","extension":"jpeg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":119349,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/c233e729035cdf769d2bf078.jpeg"},{"id":99308044,"identity":"36a87f9c-7341-4296-89c9-a6ffc1730fc5","added_by":"auto","created_at":"2025-12-31 16:07:37","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":320169,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/4cd5c69e15cb3bdc38108703.jpeg"},{"id":98821132,"identity":"f1686096-a151-45b2-873c-5b3d78f54e24","added_by":"auto","created_at":"2025-12-22 17:20:29","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1648935,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/e02bad6ad35306a4c94556e1.jpeg"},{"id":99307888,"identity":"835f7057-becc-4430-baa7-fbda524022c1","added_by":"auto","created_at":"2025-12-31 16:06:58","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":122324,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/baa818e7535cb91cce8498bf.jpeg"},{"id":98821137,"identity":"9c9e06bf-145d-4de2-881b-51ae46ee43cd","added_by":"auto","created_at":"2025-12-22 17:20:29","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":463705,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/a7bf3e7a6e475f949e8cc459.jpeg"},{"id":99307911,"identity":"5252f19e-ed2d-4546-94eb-7c86387eee21","added_by":"auto","created_at":"2025-12-31 16:07:03","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5760212,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/9a9296e1c45248f2721c9d32.jpeg"},{"id":98821143,"identity":"28a5bea8-59d2-402a-a4e8-47355bca98b4","added_by":"auto","created_at":"2025-12-22 17:20:29","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5581414,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/7bfa8b30cfa07ffb7e7b7292.jpeg"},{"id":98821148,"identity":"f506589a-0f4f-472a-8d4a-2394016d27d2","added_by":"auto","created_at":"2025-12-22 17:20:29","extension":"jpeg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2596688,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/6873250ff74dd307c824f768.jpeg"},{"id":99307759,"identity":"5e8401e3-322d-432b-b971-5fb9bc8a9fa1","added_by":"auto","created_at":"2025-12-31 16:06:42","extension":"jpeg","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1547852,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage9.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/cc08a56d92cd136e388a064c.jpeg"},{"id":99308049,"identity":"9ff5e30d-c3dd-44d9-ba1a-48dd694576c8","added_by":"auto","created_at":"2025-12-31 16:07:37","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":59202,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/bbcfe513dc60c77042332576.png"},{"id":98821138,"identity":"ffca723e-990c-4581-a964-e7a5f4baf9a0","added_by":"auto","created_at":"2025-12-22 17:20:29","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":197269,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/c474686ba641321d886270b7.png"},{"id":99307369,"identity":"1850520c-b4eb-4cbb-a277-8a3130936d1e","added_by":"auto","created_at":"2025-12-31 16:06:05","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":268204,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/f2efbc96e360ff817c6875cc.png"},{"id":98821146,"identity":"59d830a2-ee59-40d6-b0db-30efeec1787b","added_by":"auto","created_at":"2025-12-22 17:20:29","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":68389,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/05e8b21f81e4e0e2ac0d1f9b.png"},{"id":99307988,"identity":"5750ba25-57f4-45e9-b834-d9a131f4f502","added_by":"auto","created_at":"2025-12-31 16:07:21","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":60696,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/d3ff43a1e714bcb90baf3658.png"},{"id":98821149,"identity":"49edfba5-6f59-420a-be89-a6f12c2dc66a","added_by":"auto","created_at":"2025-12-22 17:20:29","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":109844,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/7e42f4db8faa01d49c1d67a6.png"},{"id":99307337,"identity":"fad9b5f5-22b7-4ce2-9075-7394b96ae167","added_by":"auto","created_at":"2025-12-31 16:06:03","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":77122,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/2c1d8f197a7144721ee7dab2.png"},{"id":99307447,"identity":"df6bef75-89c0-4046-988f-4455a736cac9","added_by":"auto","created_at":"2025-12-31 16:06:16","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":180560,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/5372161738a0a003e91cdaa8.png"},{"id":98821153,"identity":"1e079976-b1b5-4d30-bca8-e2317d6c1ed7","added_by":"auto","created_at":"2025-12-22 17:20:29","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":296246,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/50c667b49abd2d9281d03beb.png"},{"id":99307644,"identity":"e2347bb9-c016-4581-9e4c-b430b934f969","added_by":"auto","created_at":"2025-12-31 16:06:28","extension":"xml","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":89241,"visible":true,"origin":"","legend":"","description":"","filename":"e367af6bb63d45d79ab917512214522f1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/baeb1d5b8d4cb76fef9f36ed.xml"},{"id":98821152,"identity":"9de92d60-77f9-4fb2-ad8f-eed238deb20b","added_by":"auto","created_at":"2025-12-22 17:20:29","extension":"html","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":102539,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/f6887aad14aa455bc5480c4d.html"},{"id":99307487,"identity":"9f71e2e8-81d2-4740-943a-6e302aa083c9","added_by":"auto","created_at":"2025-12-31 16:06:20","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":260117,"visible":true,"origin":"","legend":"\u003cp\u003eThe schematic representation of the experimental workflow\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/035923301709a99f539e8709.png"},{"id":99307966,"identity":"72650752-9221-4991-b681-7fe957f5f9ff","added_by":"auto","created_at":"2025-12-31 16:07:16","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1115165,"visible":true,"origin":"","legend":"\u003cp\u003eScreening of ferroptosis-related genes associated with lipid metabolism. (A) Box plot after batch effect adjustment of the dataset. (B) Volcano plot of differentially expressed genes between sample groups, where each point represents a gene. (C) Module clustering dendrogram, with different colors representing different modules. (D) Heatmap of module-clinical trait correlations. (E) Venn diagram of ferroptosis-lipid metabolism-related intersection genes. (F) Heatmap of ferroptosis-lipid metabolism-related correlations.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/bfb3298859646d1ce788277d.png"},{"id":98821154,"identity":"190d4418-0eae-4500-bd6e-a5b385b3527a","added_by":"auto","created_at":"2025-12-22 17:20:30","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1617369,"visible":true,"origin":"","legend":"\u003cp\u003eInteraction and enrichment analysis of candidate genes. (A) Protein interaction network diagram of candidate genes. (B) Network diagram of KEGG TOP10 enrichment results for candidate genes. Gene node colors represent gene fold changes, with darker colors indicating larger fold changes; node sizes represent gene counts, with larger nodes indicating more genes. (C) Bubble chart of GO-CC TOP20 enrichment results for candidate genes. (D) Bubble chart of the top 20 enriched results of candidate genes in GO-MF. (E) Bubble chart of the top 20 enriched results of candidate genes in GO-BP.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/7f18cae8ded8359cc5fcdd02.png"},{"id":99307967,"identity":"6ca316de-36d6-42a4-975d-26727f1646f7","added_by":"auto","created_at":"2025-12-31 16:07:16","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":443168,"visible":true,"origin":"","legend":"\u003cp\u003eMachine learning screening of key genes. (A) Lasso model λ selection plot. (B) LASSO non-zero coefficient feature map. Each curve corresponds to a dynamic change trajectory of the coefficient of an independent variable. (C) Random forest model tree diagram. (D) Bar chart of the top 10 ranked feature genes. (E) Venn diagram of the intersection analysis of machine learning screening results.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/654a4aeecaedbea23cc0b7af.png"},{"id":98821127,"identity":"17f7b28b-19e3-43a6-9b4f-4ee468f3b9ae","added_by":"auto","created_at":"2025-12-22 17:20:29","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":511634,"visible":true,"origin":"","legend":"\u003cp\u003eThe expression of key genes and validation of diagnostic efficacy. (A) Expression of key genes in the training set, **** indicates P\u0026lt;0.0001. (B) Expression of key genes in the validation set, * indicates P\u0026lt;0.05, ** indicates P\u0026lt;0.01. (C) ROC curve of key genes in the training set. (D) ROC curve of key genes in the validation set.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/06d31f1d245be3cecdd028e0.png"},{"id":99307609,"identity":"b7687dd3-dcca-4724-8645-f76cda82291f","added_by":"auto","created_at":"2025-12-31 16:06:27","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":763749,"visible":true,"origin":"","legend":"\u003cp\u003eGSEA enrichment analysis of key genes. (A) Pathway bubble chart and GSEA curve chart of the GSEA analysis results for CBS. (B) Pathway bubble plot and GSEA curve plot of the GSEA analysis results for PRKAR2B. (C) Pathway bubble plot and GSEA curve plot of the GSEA analysis results for RELA. The colors in the pathway bubble plot represent P-values, with redder colors indicating smaller P-values; bubble size represents gene count, with larger bubbles indicating more genes. The vertical axis of the GSEA curve plot represents the enrichment score; a positive enrichment score indicates that the functional pathway is enriched at the front of the gene ranking sequence, while a negative enrichment score indicates that the functional pathway is enriched at the back of the gene ranking sequence. The horizontal axis represents the ranking of gene fold changes, with each vertical line representing a gene.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/a907bd4d6e29430ae4d08142.png"},{"id":99307494,"identity":"da8afede-ac20-4525-83ab-bf7f6d8dae9d","added_by":"auto","created_at":"2025-12-31 16:06:21","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":496608,"visible":true,"origin":"","legend":"\u003cp\u003eIntegrated analysis of immune infiltration and proteomics data. (A) Box plots of the abundance of 28 immune cell types between sample groups. * indicates P\u0026lt;0.05, ** indicates P\u0026lt;0.01, and *** indicates P\u0026lt;0.001. (B) Heatmap showing the correlation between key genes and differentially expressed immune cells. (C) Expression of CBS in cerebrospinal fluid of controls and PD patients, with * indicating P\u0026lt;0.05. (D) Expression of PRKAR2B in cerebrospinal fluid from controls and PD patients, *** indicates P\u0026lt;0.001. (E) Expression of RELA in cerebrospinal fluid from controls and PD patients, * indicates P\u0026lt;0.05. (F) ROC curve for key genes in cerebrospinal fluid from controls and PD patients.\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/5758ee61bfabeaaf1d166366.png"},{"id":98821136,"identity":"a8a1dcae-c267-46dd-b78d-9aa410ab3f68","added_by":"auto","created_at":"2025-12-22 17:20:29","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":2095208,"visible":true,"origin":"","legend":"\u003cp\u003eEffects of RELA knockdown on cell viability and apoptosis in PD cell models. (A) qRT-PCR assay to detect the efficiency of RELA knockdown. (B) CCK8 assay to detect cell viability in each group. (C) Detection of apoptotic cells by Annexin V/7-AAD double-staining flow cytometry. ***\u003cem\u003eP\u003c/em\u003e\u0026lt;0.001, ns indicates no statistical difference.\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/7787959421eb30e2df732074.png"},{"id":99307327,"identity":"63939e0c-925a-402f-8079-5b50e189e6d4","added_by":"auto","created_at":"2025-12-31 16:06:01","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":3365036,"visible":true,"origin":"","legend":"\u003cp\u003eRegulatory effects of si-RELA on oxidative stress, ferroptosis, and dopaminergic neuronal function in PD cell models. (A) Flow cytometry analysis of intracellular ROS levels. (B) ELISA assay for intracellular MDA content. (C) ELISA assay for intracellular SOD activity. (D) Iron ion assay kit to measure intracellular Fe\u003csup\u003e2+ \u003c/sup\u003elevels. (E-L) Western blot assay to detect the expression of inflammation-related proteins, ferroptosis-related proteins, and TH. *P \u0026lt; 0.05, **P \u0026lt; 0.01, ***P \u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/773e710459865b26d141e697.png"},{"id":104740300,"identity":"a15cf21c-fae4-4e48-871a-bb2c370ff296","added_by":"auto","created_at":"2026-03-16 16:16:34","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":11623764,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8361812/v1/19703689-c2f8-4851-b97f-782ccec93b5e.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"RELA as a Diagnostic Biomarker for Parkinson's Disease by Integrating Ferroptosis, Lipid Metabolism, and Neuroinflammation","fulltext":[{"header":"Introduction","content":"\u003cp\u003eParkinson\u0026rsquo;s disease(PD) constitutes a major neurodegenerative disorder, clinically delineated by a triad of motor impairments collectively termed parkinsonism. These hallmark features include resting tremor, progressive muscular rigidity, bradykinesia(characterized by hypokinesia and delayed movement initiation), and postural instability with gait disturbances [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Epidemiological investigations demonstrate that PD exhibits a prevalence of approximately 1% in populations aged\u0026thinsp;\u0026ge;\u0026thinsp;60 years, with incidence rates ranging from 1\u0026ndash;2 cases per 1,000 person-years in general populations [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. At the molecular level, PD is pathognomically defined by the intraneuronal accumulation of α-synuclein(α-Syn) aggregates, forming Lewy bodies, concomitant with selective degeneration of dopaminergic neurons in the substantia nigra pars compacta[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Emerging mechanistic studies implicate multifactorial pathogenic pathways, including oxidative stress-mediated neuronal injury, mitochondrial dysfunction, lysosomal degradation impairment, and chronic neuroinflammatory processes in disease progression[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eResearch has found that many pathogenic risk factors for PD are associated with ferroptosis. For example, DJ-1 can negatively regulate ferroptosis by maintaining the biosynthesis of cysteine and GSH, while the SLC7A11 gene, which is significantly downregulated in PD, is also considered a key regulator of ferroptosis and can protect dopaminergic neurons by activating the transsulfuration pathway[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. A recent study using human induced pluripotent stem cells(HiPSCs) showed that triploid mutations in SNCA(the gene encoding α-syn) and the accumulation of α-syn oligomers can specifically induce ferroptosis in differentiated neurons[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Beyond the high correlation with genetic factors, PD also exhibits typical pathological features related to ferroptosis, including excessive iron accumulation, lipid peroxidation, and redox homeostasis imbalance(changes in levels of glutamate-cystine antiporter(xCT), GSH, DJ-1, and coenzyme Q10(CoQ10), among others). In PD models induced by neurotoxins, such as 1-methyl-4-phenyl-pyridinium ion (MPP+) and 6-hydroxydopamine(6-OHDA)-treated human neuroblastoma (SH-SY5Y) cells and 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-damaged mice, obvious ferroptosis processes and characteristics are observed. Inhibitory therapeutic measures targeting ferroptosis can also significantly improve motor dysfunction and the loss of dopaminergic neurons in PD model mice[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Thus, ferroptosis plays an undeniable role in the genetics, pathological progression, pathogenesis, and potential treatment of PD.\u003c/p\u003e \u003cp\u003eIn this study, we systematically characterized the diagnostic biomarkers and elucidated the underlying mechanisms of ferroptosis-lipid metabolism-related genes in PD through comprehensive bioinformatics analysis. The findings were further validated using both cellular models and cerebrospinal fluid(CSF) samples from PD patients, demonstrating the translational potential of these molecular targets for both early diagnosis and therapeutic intervention in PD.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData acquisition and preprocessing\u003c/h2\u003e \u003cp\u003eGene expression datasets(GSE49036, GSE20292, GSE7621, and GSE20164) were obtained from the Gene Expression Omnibus(GEO). A total of 74 samples(42 PD and 32 controls) from GSE49036, GSE20292, and GSE7621 were merged and normalized. The ComBat algorithm was applied to remove batch effects. The dataset GSE20164 was used as an independent validation set.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eIdentification of differentially expressed genes(DEGs)\u003c/h3\u003e\n\u003cp\u003eDEGs between PD and control samples were identified using the limma R package, with an adjusted p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and |log₂FC|\u0026gt;0 set as significance thresholds.\u003c/p\u003e\n\u003ch3\u003eWeighted gene co-expression network analysis(WGCNA)\u003c/h3\u003e\n\u003cp\u003eWGCNA was performed to identify gene modules significantly associated with PD. A soft-thresholding power of 4 was selected to ensure a scale-free network. Modules with |r|\u0026gt;0.4 and p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were considered clinically significant.\u003c/p\u003e\n\u003ch3\u003eScreening of ferroptosis and lipid metabolism-related genes\u003c/h3\u003e\n\u003cp\u003eFerroptosis-related genes(FRGs) and lipid metabolism-related genes(LMRGs) were retrieved from public databases. Venn analysis was used to identify overlapping genes among DEGs, WGCNA module genes, FRGs, and LMRGs.\u003c/p\u003e\n\u003ch3\u003eProtein-protein interaction(PPI) network and functional enrichment\u003c/h3\u003e\n\u003cp\u003eThe PPI network was constructed using the STRING database(confidence score\u0026thinsp;\u0026gt;\u0026thinsp;0.4) and visualized in Cytoscape. Functional enrichment analysis of GO terms and KEGG pathways was performed using the clusterProfiler R package.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eMachine learning-based feature selection\u003c/h2\u003e \u003cp\u003eLASSO regression and random forest algorithms were applied to screen key diagnostic genes[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. The optimal lambda(λ) value in LASSO was determined via 10-fold cross-validation. The top 10 genes ranked by mean decrease accuracy were selected from the random forest model. The intersection of genes identified by both methods was considered the final key gene set.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eGene expression and diagnostic validation\u003c/h3\u003e\n\u003cp\u003eExpression levels of key genes were validated in the independent dataset GSE20164. Receiver operating characteristic(ROC) curves were generated to evaluate diagnostic performance.\u003c/p\u003e\n\u003ch3\u003eImmune infiltration analysis\u003c/h3\u003e\n\u003cp\u003eTo investigate the immune microenvironment in PD, immune cell infiltration abundances were estimated from the transcriptomic data using the CIBERSORT algorithm. The relative proportions of 22 immune cell types were calculated for each sample. Differences in immune cell infiltration between PD and control groups were assessed using the Wilcoxon rank-sum test, with a significance threshold of P\u0026thinsp;\u0026lt;\u0026thinsp;0.05. Spearman correlation analysis was performed to evaluate associations between key genes(CBS, PRKAR2B, RELA) and significantly altered immune cell types. Correlations with |r|\u0026gt;0.5 and P\u0026thinsp;\u0026lt;\u0026thinsp;0.01 were considered statistically significant.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eCell culture, transfection and cell death analysis\u003c/h2\u003e \u003cp\u003eHuman SH-SY5Y cells were purchased from the Cell Bank/Stem Cell Bank, Chinese Academy of Sciences (Shanghai, China). Briefly, the cells were cultured in high-glucose DMEM (Thermo Fisher Scientific, CA, USA) supplemented with 10% fetal bovine serum (Gibco, Life Technologies) in an incubator at 37\u0026deg;C and 5% CO2. SH-SY5Y cells were treated with MPP\u0026thinsp;+\u0026thinsp;to establish PD models. RELA knockdown was performed using siRNA transfection.\u003c/p\u003e \u003cp\u003eCell viability was assessed using the Cell Counting Kit-8 (CCK-8) assay. Briefly, cells were seeded in 96-well plates and after treatment, the CCK-8 reagent was added to each well. Following incubation at 37\u0026deg;C for 2 hours, the absorbance was measured at a wavelength of 450 nm using a microplate reader. Apoptosis was quantified by flow cytometry with an Annexin V-PE/7-AAD apoptosis detection kit. Cells were stained with Annexin V-PE and 7-AAD in the dark for 15 minutes and then analyzed to distinguish early and late apoptotic populations. Intracellular ROS levels were determined using the fluorescent probe DCFH-DA. Cells were loaded with DCFH-DA for 30 minutes at 37\u0026deg;C. After washing, the fluorescence intensity was immediately measured using flow cytometry. To assess oxidative stress, the content of malondialdehyde (MDA) and the activity of superoxide dismutase (SOD) were measured using commercial enzyme-linked immunosorbent assay (ELISA) kits, strictly following the manufacturer's protocols. Furthermore, to directly investigate ferroptosis, intracellular Fe\u0026sup2;⁺ levels were determined using a specific iron assay kit according to the manufacturer's instructions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eWestern blot\u003c/h2\u003e \u003cp\u003eWestern blot analysis was conducted as described in our previous study [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Whole-cell lysates were extracted using radioimmunoprecipitation assay (RIPA) lysis buffer supplemented with a protease inhibitor cocktail (Sigma, USA). The proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS‒PAGE) and then transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore, Bedford, USA). The membranes were incubated with primary antibodies against Achaete-scute family bHLH transcription factor 4(ASCL4) (1:1000), CD40 (1:1000), C-X-C motif chemokine ligand 6 (CXCL6)(1:1000), C-X-C motif chemokine ligand 9 (CXCL9)(1:1000), Glutathione peroxidase 4 (GPX4)(1:1000), Transferrin receptor (TFRC)(1:1000), and Tyrosine hydroxylase (TH)(1:1000) at 4\u0026deg;C overnight. The membranes were then probed with an HRP-conjugated secondary antibody (Beyotime, Shanghai, China) at 25\u0026deg;C for 1 h, followed by washing with Tris Buffered Saline and 0.05% Tween 20 (TBST) three times for 10 min each. Finally, the membranes were visualized and imaged with an enhanced chemiluminescence system. With GAPDH as the loading control, the expression of proteins was quantified using ImageJ Pro Plus 6.0 Software.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eELISA validation using human CSF samples\u003c/h2\u003e \u003cp\u003eCSF samples were collected from 10 PD patients and 10 controls(patients with facial spasm). Protein levels of CBS, PRKAR2B, and RELA were quantified using commercial ELISA kits according to manufacturers\u0026rsquo; instructions. Statistical differences were assessed using Student\u0026rsquo;s t-test, and ROC analysis was performed to evaluate diagnostic utility. The study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of the Institutional Review Board of Jinan Central Hospital.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eAll statistical analyses were conducted in R(v4.2.1). Differential expression was analyzed using the limma package. *P\u0026thinsp;\u0026lt;\u0026thinsp;0.05, **P\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ***P\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ****P\u0026thinsp;\u0026lt;\u0026thinsp;0.001 were considered statistically significant.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003e \u003cb\u003e1. Identification of ferroptosis and lipid metabolism-related core genes in PD\u003c/b\u003e \u003c/p\u003e \u003cp\u003eThe schematic representation of the experimental workflow is presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Before going further, batch effect correction was performed on 74 samples(42 PD and 32 controls) from datasets GSE49036, GSE20292, and GSE7621 using the ComBat algorithm. After normalization, the median expression values across datasets showed high consistency(Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e2\u003c/span\u003eA), indicating successful removal of batch effects. Differential expression analysis with the limma package identified 1,493 significantly dysregulated genes in PD compared to controls(P\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and |log₂FC|\u0026gt;0)(Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). WGCNA was applied to identify co-expression modules associated with PD, using a soft-thresholding power of 4(Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). Seventeen modules were identified, among which the blue, brown, and royalblue modules demonstrated significant correlation with disease status(|r|\u0026gt;0.4, P\u0026thinsp;\u0026lt;\u0026thinsp;0.05)(Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e2\u003c/span\u003eD), yielding 4376 core module genes. By integrating 382 ferroptosis-related genes and 867 lipid metabolism-related genes from public databases, Venn analysis revealed 20 ferroptosis-related and 60 lipid metabolism-related DEGs. Correlation analysis further identified 44 candidate genes significantly associated with both processes(|r|\u0026gt;0.6, P\u0026thinsp;\u0026lt;\u0026thinsp;0.05)(Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e2\u003c/span\u003eE-F), which were selected for subsequent analyses.\u003c/p\u003e \u003cp\u003e \u003cb\u003e2. Functional enrichment and protein interaction analysis of candidate genes\u003c/b\u003e \u003c/p\u003e \u003cp\u003eA protein-protein interaction network was constructed using the STRING database(confidence threshold: 0.4), resulting in 34 interaction pairs. The network was visualized in Cytoscape(Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). Functional enrichment analysis via clusterProfiler revealed significant involvement of the candidate genes in 331 biological processes(BP), 26 cellular components(CC), 76 molecular functions(MF), and 24 KEGG pathways. Key enriched BPs included biosynthesis and metabolic processes related to alcohols, terpenoids, aerobic respiration, TCA cycle, and sphingolipid/phospholipid metabolism. CC terms were associated with autolysosomes, autophagosomes, organelle lumens, and mitochondrial membrane complexes. MF analysis highlighted enzymatic activities such as acetylgalactosamine transferase, lipid kinase, and oxidoreductase, as well as ferrous iron and NADP/NADPH binding. KEGG pathway analysis indicated enrichment in pyruvate metabolism, TCA cycle, inositol phosphate metabolism, cysteine and methionine metabolism, and terpenoid backbone biosynthesis, among others(Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e3\u003c/span\u003eB\u0026ndash;E).\u003c/p\u003e \u003cp\u003e \u003cb\u003e3. Machine learning-based identification of diagnostic biomarkers\u003c/b\u003e \u003c/p\u003e \u003cp\u003eLASSO regression and random forest were employed to refine candidate genes. The LASSO model identified five feature genes with non-zero coefficients at the optimal λ value(λ.min\u0026thinsp;=\u0026thinsp;0.1428043)(Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e4\u003c/span\u003eA-B). Random forest analysis ranked genes by importance using mean decrease accuracy, and the top 10 genes were selected(Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e4\u003c/span\u003eC-D). The intersection of genes identified by both methods yielded three key genes: CBS, PRKAR2B, and RELA (Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e4\u003c/span\u003eE).\u003c/p\u003e \u003cp\u003e \u003cb\u003e4. Differential expression and diagnostic performance of key genes\u003c/b\u003e \u003c/p\u003e \u003cp\u003eExpression analysis of the three key genes was conducted in both the training and validation sets(GSE20164). CBS and RELA were significantly upregulated in PD samples, while PRKAR2B was downregulated(Fig.\u0026nbsp;\u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e5\u003c/span\u003eA\u0026ndash;B). ROC curve analysis demonstrated strong diagnostic ability for all three genes, with AUC values exceeding 0.75 in both cohorts(Fig.\u0026nbsp;\u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e5\u003c/span\u003eC\u0026ndash;D).\u003c/p\u003e \u003cp\u003e \u003cb\u003e5. Pathway enrichment analysis of key genes\u003c/b\u003e \u003c/p\u003e \u003cp\u003eGene Set Enrichment Analysis(GSEA) revealed that CBS was primarily associated with the Notch signaling pathway, ribosome, spliceosome, and pathways in cancer(Fig.\u0026nbsp;\u003cspan refid=\"Fig14\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). PRKAR2B was enriched in glycolysis/gluconeogenesis, JAK-STAT signaling, oxidative phosphorylation, pyruvate metabolism, and TCA cycle(Fig.\u0026nbsp;\u003cspan refid=\"Fig14\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). RELA was significantly linked to apoptosis, chemokine signaling, JAK-STAT pathway, oxidative phosphorylation, Toll-like receptor signaling, and cytokine-cytokine receptor interactions(Fig.\u0026nbsp;\u003cspan refid=\"Fig14\" class=\"InternalRef\"\u003e6\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003e \u003cb\u003e6. Immune infiltration combined with proteomics data reveals the mechanism of inflammatory response in PD\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo investigate the immune microenvironment in PD samples and the role of key genes, immune cell infiltration analysis was performed. The results revealed significant differences in the infiltration levels of 12 immune cell types between PD and control samples(Fig.\u0026nbsp;\u003cspan refid=\"Fig15\" class=\"InternalRef\"\u003e7\u003c/span\u003eA). Among these, dendritic cells, effector memory CD8⁺T cells, MDSCs, natural killer T cells, helper T cells, macrophages, B cells, mast cells, and neutrophils were significantly upregulated in PD samples, indicating a disrupted immune microenvironment and suggesting the involvement of specific immune cells in disease progression.\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eSpearman correlation analysis was conducted to examine the relationship between the key genes(CBS, PRKAR2B, RELA) and the differentially infiltrated immune cells. The results demonstrated that CBS was strongly correlated with follicular helper T cells(cor\u0026thinsp;=\u0026thinsp;0.62, P\u0026thinsp;\u0026lt;\u0026thinsp;0.01). PRKAR2B showed a strong correlation with immature dendritic cells(cor\u0026thinsp;=\u0026thinsp;0.73, P\u0026thinsp;\u0026lt;\u0026thinsp;0.01)(Fig.\u0026nbsp;\u003cspan refid=\"Fig15\" class=\"InternalRef\"\u003e7\u003c/span\u003eB). RELA was significantly correlated with multiple immune cells, including follicular helper T cells(cor\u0026thinsp;=\u0026thinsp;0.55, P\u0026thinsp;\u0026lt;\u0026thinsp;0.01), plasmacytoid dendritic cells(cor\u0026thinsp;=\u0026thinsp;0.56, P\u0026thinsp;\u0026lt;\u0026thinsp;0.01), natural killer T cells(cor\u0026thinsp;=\u0026thinsp;0.54, P\u0026thinsp;\u0026lt;\u0026thinsp;0.01), myeloid-derived suppressor cells(cor\u0026thinsp;=\u0026thinsp;0.53, P\u0026thinsp;\u0026lt;\u0026thinsp;0.01), mast cells(cor\u0026thinsp;=\u0026thinsp;0.58, P\u0026thinsp;\u0026lt;\u0026thinsp;0.01), and effector memory CD8⁺T cells(cor\u0026thinsp;=\u0026thinsp;0.54, P\u0026thinsp;\u0026lt;\u0026thinsp;0.01)(Fig.\u0026nbsp;\u003cspan refid=\"Fig15\" class=\"InternalRef\"\u003e7\u003c/span\u003eB). These findings suggest that CBS, PRKAR2B, and RELA may play important roles in regulating immune responses in PD.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eELISA was performed on CSF samples from 10 PD patients and 10 controls with facial spasm. Consistent with bioinformatic predictions, CBS and RELA protein levels were significantly elevated in PD, while PRKAR2B was reduced(Fig.\u0026nbsp;\u003cspan refid=\"Fig15\" class=\"InternalRef\"\u003e7\u003c/span\u003eC-E). ROC analysis based on CSF protein expression confirmed the diagnostic utility of all three markers, with AUC values above 0.7(Fig.\u0026nbsp;\u003cspan refid=\"Fig15\" class=\"InternalRef\"\u003e7\u003c/span\u003eF).\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e7. RELA participates in the pathological process of PD by regulating oxidative stress and ferroptosis\u003c/h2\u003e \u003cp\u003eGiven that RELA is known to be a core transcription factor of NF-κB, and the NF-κB pathway is a central axis of neuroinflammation in PD. Through clinical cerebrospinal fluid proteomic analysis, it was discovered that inflammation-related differentially expressed proteins in PD patients form a synergistic network and exhibit significant overlap with RELA enrichment pathways. To further verify the regulatory role of RELA on PD-related pathological phenotypes at the cellular level, RELA was first knocked down in cells. The relative mRNA expression level of RELA was detected by qRT-PCR to evaluate the knockdown efficiency of si-RELA (Fig.\u0026nbsp;\u003cspan refid=\"Fig16\" class=\"InternalRef\"\u003e8\u003c/span\u003eA). SH-SY5Y cells were induced with MPP\u0026thinsp;+\u0026thinsp;to construct a PD cell model. Compared with the Control group, cell viability was significantly reduced after MPP\u0026thinsp;+\u0026thinsp;treatment (Fig.\u0026nbsp;\u003cspan refid=\"Fig16\" class=\"InternalRef\"\u003e8\u003c/span\u003eB, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), indicating successful simulation of neurotoxic damage in PD and the establishment of a PD cell model. Compared with the MPP\u0026thinsp;+\u0026thinsp;+\u0026thinsp;si-NC group, cell viability in the MPP\u0026thinsp;+\u0026thinsp;+\u0026thinsp;si-RELA group was significantly restored (Fig.\u0026nbsp;\u003cspan refid=\"Fig16\" class=\"InternalRef\"\u003e8\u003c/span\u003eB, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001). This suggests that knocking down RELA can alleviate MPP+-induced cytotoxicity.\u003c/p\u003e \u003cp\u003eFlow cytometry showed that compared with the Control group (3.67%), the proportion of apoptotic cells significantly increased after MPP\u0026thinsp;+\u0026thinsp;treatment (11.29%). The proportion of apoptotic cells in the MPP\u0026thinsp;+\u0026thinsp;+\u0026thinsp;si-RELA group (7.8%) was significantly lower than that in the MPP\u0026thinsp;+\u0026thinsp;+\u0026thinsp;si-NC group (11.49%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig16\" class=\"InternalRef\"\u003e8\u003c/span\u003eC). This indicates that knocking down RELA can inhibit MPP+-induced cell apoptosis.\u003c/p\u003e\u003cp\u003e \u003cb\u003e8. Effects of RELA knockdown on oxidative stress, ferroptosis, and dopaminergic neuronal function in PD cell models\u003c/b\u003e \u003c/p\u003e \u003cp\u003eOxidative stress is a core pathologica characteristic in PD neuronal injury. This study investigated the regulatory role of RELA in oxidative stress by detecting ROS, the lipid peroxidation product malondialdehyde (MDA), and the antioxidant enzyme superoxide dismutase (SOD). Flow cytometry showed significant shifts in the fluorescence intensity distribution peaks among the groups. Compared to the Control group, the Mean X was significantly increased in the MPP⁺ group, while the Mean X in the MPP⁺+si-RELA group was significantly lower than that in the MPP⁺+si-NC group (Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e9\u003c/span\u003eA). MDA is a characteristic product of lipid peroxidation, and its content directly reflects the degree of lipid damage. ELISA detection of MDA content revealed that compared to the Control group, MPP⁺ treatment significantly increased MDA content (Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e9\u003c/span\u003eB, P\u0026thinsp;\u0026lt;\u0026thinsp;0.01), whereas the MDA content in the MPP⁺+si-RELA group was significantly decreased compared to the MPP⁺+si-NC group (Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e9\u003c/span\u003eB, P\u0026thinsp;\u0026lt;\u0026thinsp;0.05), suggesting that RELA knockdown can alleviate MPP⁺-induced lipid peroxidation damage. SOD is a key intracellular antioxidant enzyme responsible for scavenging superoxide anion radicals. ELISA detection of SOD activity showed that the Control group had the highest SOD activity. After MPP⁺ treatment, SOD activity was significantly reduced (Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e9\u003c/span\u003eC, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), but it was significantly restored in the MPP⁺+si-RELA group compared to the MPP⁺+si-NC group (Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e9\u003c/span\u003eC, P\u0026thinsp;\u0026lt;\u0026thinsp;0.01), indicating that knocking down RELA enhances cellular antioxidant defense capacity.\u003c/p\u003e \u003cp\u003eFerroptosis is a form of programmed cell death dependent on lipid peroxidation, and the intracellular Fe\u003csup\u003e2+\u003c/sup\u003e level is a key factor initiating ferroptosis. The regulatory role of RELA was explored by detecting core indicators of ferroptosis. Results from the iron ion assay kit showed that intracellular Fe\u003csup\u003e2+\u003c/sup\u003e content significantly increased after MPP\u003csup\u003e+\u003c/sup\u003e treatment, while the Fe\u003csup\u003e2+\u003c/sup\u003e content in the MPP⁺+si-RELA group was lower than that in the MPP\u003csup\u003e+\u003c/sup\u003e+si-NC group (Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e9\u003c/span\u003eD, P\u0026thinsp;\u0026lt;\u0026thinsp;0.05). This suggests that RELA knockdown can inhibit MPP\u003csup\u003e+\u003c/sup\u003e-induced Fe\u003csup\u003e2+\u003c/sup\u003e accumulation, thereby regulating the ferroptosis process.\u003c/p\u003e \u003cp\u003eThe expression of inflammatory chemokines (CD40, CXCL6, CXCL9), ferroptosis-related proteins (GPX4, ACSL4, TFRC), and the dopaminergic neuronal marker (TH) was detected by Western blot. After MPP\u003csup\u003e+\u003c/sup\u003e treatment, the expression of CD40, CXCL6, and CXCL9 was significantly upregulated (Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e9\u003c/span\u003eE, J-I, P\u0026thinsp;\u0026lt;\u0026thinsp;0.01). In the MPP\u003csup\u003e+\u003c/sup\u003e+si-RELA group, the expression of these genes was significantly reduced compared to the MPP\u003csup\u003e+\u003c/sup\u003e+si-NC group (Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e9\u003c/span\u003eE, J-I, P\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Compared to the Control group, the expression of the ferroptosis-related proteins ACSL4 and TFRC was significantly increased after MPP\u003csup\u003e+\u003c/sup\u003e treatment (Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e9\u003c/span\u003eE-F, K, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and their expression in the MPP\u003csup\u003e+\u003c/sup\u003e+si-RELA group was lower than that in the MPP\u003csup\u003e+\u003c/sup\u003e+si-NC group (Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e9\u003c/span\u003eE-F, K, P\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Compared to the Control group, GPX4 expression was significantly decreased after MPP\u003csup\u003e+\u003c/sup\u003e treatment (Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e9\u003c/span\u003eJ, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), but its expression in the MPP\u003csup\u003e+\u003c/sup\u003e+si-RELA group was higher than that in the MPP\u003csup\u003e+\u003c/sup\u003e+si-NC group (Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e9\u003c/span\u003eJ, P\u0026thinsp;\u0026lt;\u0026thinsp;0.01). TH is the key rate-limiting enzyme for dopamine synthesis, and its expression reflects dopaminergic neuronal function. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig17\" class=\"InternalRef\"\u003e9\u003c/span\u003eE, L, TH expression was highest in the Control group, significantly decreased after MPP\u003csup\u003e+\u003c/sup\u003e treatment (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and significantly increased in the MPP\u003csup\u003e+\u003c/sup\u003e+si-RELA group compared to the MPP\u003csup\u003e+\u003c/sup\u003e+si-NC group (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05). In summary, these results indicate that knocking down RELA can inhibit MPP\u003csup\u003e+\u003c/sup\u003e-induced inflammatory chemokine expression, regulate ferroptosis, and restore dopaminergic neuronal function.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eParkinson's disease is a complex neurodegenerative disorder whose pathogenesis involves multiple genetic and metabolic pathways[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. In this study, we identified three key genes-CBS, PRKAR2B, and RELA-that are closely associated with ferroptosis and lipid metabolism, demonstrating significant diagnostic value in both bioinformatic predictions and experimental validation.\u003c/p\u003e \u003cp\u003eThe integration of multiple transcriptomic datasets revealed 44 candidate genes significantly associated with both ferroptosis and lipid metabolism. Functional enrichment analysis indicated their strong involvement in critical metabolic pathways including the TCA cycle, pyruvate metabolism, and sphingolipid/phospholipid metabolism, suggesting a tight coupling between energy metabolism dysregulation and ferroptosis susceptibility in PD. The convergence of these pathways highlights the complex metabolic rewiring that characterizes PD pathology. Importantly, the diagnostic potential of these biomarkers was validated across multiple levels. Both transcriptomic and proteomic analyses consistently showed CBS and RELA upregulation and PRKAR2B downregulation in PD. The strong AUC values (\u0026gt;\u0026thinsp;0.75 in transcriptomic data, \u0026gt;\u0026thinsp;0.7 in CSF protein levels) across independent cohorts support their robustness as diagnostic markers.\u003c/p\u003e \u003cp\u003eAmong these, CBS attracts particular attention due to its dual role in neuroprotection and neuroinflammation regulation[\u003cspan additionalcitationids=\"CR12\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. As the rate-limiting enzyme in the transsulfuration pathway, CBS catalyzes the conversion of homocysteine to cysteine. Its deficiency is closely linked to hyperhomocysteinemia, which has been implicated in neurodegeneration[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. More importantly, CBS is a major endogenous producer of hydrogen sulfide(H\u003csub\u003e2\u003c/sub\u003eS), a gaseous neuromodulator that exhibits anti-inflammatory and neuroprotective properties. Previous studies have shown that H\u003csub\u003e2\u003c/sub\u003eS donors can inhibit neuroinflammation through modulation of the NF-κB and AMPK pathways, and exert protective effects in a mouse model of PD[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Our findings align with these reports, showing significant upregulation of CBS in PD samples. This elevation may reflect a compensatory mechanism against ongoing neuroinflammation and oxidative stress. Notably, recent evidence suggests that microglial-specific deletion of CBS exacerbates NLRP3 inflammasome activation and dopaminergic neurodegeneration, whereas its overexpression or H\u003csub\u003e2\u003c/sub\u003eS supplementation attenuates these effects[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Therefore, CBS may serve not only as a biomarker but also as a promising therapeutic target for modulating neuroinflammation in PD.\u003c/p\u003e \u003cp\u003eIn contrast, PRKAR2B was significantly downregulated in PD patients. As a regulatory subunit of protein kinase A(PKA), PRKAR2B is involved in cAMP signaling, which plays crucial roles in neuronal survival, synaptic plasticity, and neurotransmitter release [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Although its specific function in PD remains underexplored, the dysregulation of PKA signaling has been implicated in Rett syndrome[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] and several neurodegenerative diseases[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. The observed downregulation may indicate impaired cAMP-mediated neuroprotection, contributing to dopaminergic neuron vulnerability. Further studies are warranted to elucidate the mechanistic role of PRKAR2B in PD pathogenesis.\u003c/p\u003e \u003cp\u003eRELA (p65), a core component of the NF-κB pathway, was significantly upregulated in PD, consistent with its well-established role in regulating inflammatory responses[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. NF-κB activation is known to drive the expression of pro-inflammatory cytokines and chemokines, contributing to chronic neuroinflammation and neuronal death in PD[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Our cellular experiments demonstrated that RELA knockdown significantly alleviated MPP⁺-induced cytotoxicity, apoptosis, oxidative stress, and ferroptosis. The reduction in ROS, MDA, and Fe\u0026sup2;⁺ accumulation, coupled with restored SOD activity and GPX4 expression, provides compelling evidence that RELA modulates redox homeostasis and ferroptosis sensitivity. Furthermore, RELA knockdown suppressed inflammatory chemokine expression (CD40, CXCL6, CXCL9) and restored tyrosine hydroxylase (TH) levels, indicating its dual role in regulating neuroinflammation and dopaminergic function. Our GSEA results further support its involvement in immune-related pathways such as cytokine-cytokine receptor interaction, Toll-like receptor signaling, and the JAK-STAT pathway. The immune infiltration analysis further revealed extensive correlations between these key genes and specific immune cell populations. The broad association of RELA with multiple immune cells\u0026mdash;including follicular helper T cells, plasmacytoid dendritic cells, and myeloid-derived suppressor cells\u0026mdash;underscores its central role in shaping the neuroimmune landscape in PD. The consistent upregulation of various immune cell types in PD samples confirms immune dysregulation as a hallmark of PD pathology.\u003c/p\u003e \u003cp\u003eSeveral limitations should be acknowledged. First, although we integrated multiple datasets and validated results in an independent cohort and human CSF samples, the sample size remains moderate. Future studies with larger clinical cohorts are necessary to strengthen the generalizability of our findings. Second, The precise mechanisms through which CBS and PRKAR2B regulate the cross-talk between ferroptosis and lipid metabolism require further investigation, particularly using cell and animal models. Finally, longitudinal studies would help evaluate whether these biomarkers can reflect disease progression or treatment response.\u003c/p\u003e \u003cp\u003eIn conclusion, our study systematically identified and validated CBS, PRKAR2B, and RELA as key genes linking ferroptosis and lipid metabolism dysregulation in PD. Our study also confirm that the pathways regulated by RELA align with the enrichment pathways of differentially expressed inflammatory proteins in the cerebrospinal fluid of PD patients. This provides direct evidence supporting RELA as a clinically relevant target.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eQ.C., C.S. and D.S. contributed to the study design and conceptual framework. Q.C., Y.J., W.B., and W.L. conducted data acquisition, performed statistical analysis, and interpreted the results. Technical support was provided by Y.J., W.B., and W.L. Manuscript composition and critical revision were completed by Q.C., C.S., C.L. and D.S. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical trial number\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of Interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no conflicts of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data that support the findings of this study are available in the methods of this article.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding Declaration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported by Shandong Provincial Medical and Health Science and Technology Development Plan Project (No. 202204040490), Jinan Municipal Health Commission Science and Technology Plan Project (NO.2024202003).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eKhan, M. S., S. Nasiripour, and J. C. Bopassa. 2025. Parkinson Disease Signaling Pathways, Molecular Mechanisms, and Potential Therapeutic Strategies: A Comprehensive Review. \u003cem\u003eInternational journal of molecular sciences\u003c/em\u003e ; 26.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao, X. D., J. E. Ding, J. X. Xie, and H. M. Xu. 2025. Epigenetic regulation of iron metabolism and ferroptosis in Parkinson's disease: Identifying novel epigenetic targets. \u003cem\u003eActa pharmacologica Sinica\u003c/em\u003e 46:2075\u0026ndash;2092.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTysnes, O. B., and A. Storstein. 1996. Epidemiology of Parkinson's disease. Journal of neural transmission (Vienna, Austria: 2017; 124:901\u0026ndash;905.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGan, Y. H., L. Z. Ma, Y. Zhang, J. You, Y. Guo, and Y. He et al. 2025. Large-scale proteomic analyses of incident Parkinson's disease reveal new pathophysiological insights and potential biomarkers. \u003cem\u003eNature aging\u003c/em\u003e 5:642\u0026ndash;657.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKia, D. A., D. Zhang, S. Guelfi, C. Manzoni, L. Hubbard, and R. H. Reynolds et al. 2021. Identification of Candidate Parkinson Disease Genes by Integrating Genome-Wide Association Study, Expression, and Epigenetic Data Sets. \u003cem\u003eJAMA neurology\u003c/em\u003e 78:464\u0026ndash;472.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFan, H., Y. Li, J. Huang, J. Jiang, F. Feng, and N. Huang et al. 2025. DJ-1 in Parkinson's disease: Its important role at endoplasmic reticulum-mitochondria contact sites. \u003cem\u003eBehavioural brain research\u003c/em\u003e 495:115775.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAngelova, P. R., M. L. Choi, A. V. Berezhnov, M. H. Horrocks, C. D. Hughes, and S. De et al. 2020. Alpha synuclein aggregation drives ferroptosis: an interplay of iron, calcium and lipid peroxidation. \u003cem\u003eCell death and differentiation\u003c/em\u003e 27:2781\u0026ndash;2796.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXi, L. J., Z. Y. Guo, X. K. Yang, and Z. G. Ping. 2023. [Application of LASSO and its extended method in variable selection of regression analysis]. Zhonghua yu fang yi xue za zhi [Chinese journal of preventive medicine] ; 57:107\u0026ndash;111.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, H., Z. Zhou, H. Zhong, S. Wang, K. Shen, and R. Huang et al. 2025. A pyrimidine metabolism-related gene signature for prognosis prediction and immune microenvironment description of breast cancer. \u003cem\u003eJournal of translational medicine\u003c/em\u003e 23:698.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiaz-Ortiz, M. E., Y. Seo, M. Posavi, M. Carceles Cordon, E. Clark, and N. Jain et al. 2022. \u003cem\u003eGPNMB confers risk for Parkinson's disease through interaction with α-synuclein\u003c/em\u003e. vol. 377 eabk0637. New York, NY: Science.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMou, Y. J., Y. T. Ma, X. Yuan, M. Wang, Y. Liu, and C. S. Pei et al. 2023. \u003cem\u003eCystathionine β-Synthase Suppresses NLRP3 Inflammasome Activation via Redox Regulation in Microglia\u003c/em\u003e. Antioxidants \u0026amp; redox signaling.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, Y., Z. Wang, Y. T. Ma, Y. X. Chen, X. Y. Qian, and X. O. Hou et al. 2025. STAT1 and STAT6 orchestrate Cbs transcription and transsulfur metabolism in microglia and contribute to parkinson's disease-related neuroinflammation. Cellular and molecular life sciences: CMLS. ; 82:294.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee, M., C. Schwab, S. Yu, E. McGeer, and P. L. McGeer. 2009. Astrocytes produce the antiinflammatory and neuroprotective agent hydrogen sulfide. \u003cem\u003eNeurobiology of aging\u003c/em\u003e 30:1523\u0026ndash;1534.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKida, K., M. Yamada, K. Tokuda, E. Marutani, M. Kakinohana, and M. Kaneki et al. 2011. Inhaled hydrogen sulfide prevents neurodegeneration and movement disorder in a mouse model of Parkinson's disease. \u003cem\u003eAntioxidants \u0026amp; redox signaling\u003c/em\u003e 15:343\u0026ndash;352.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu, L. F., M. Lu, P. T. Hon Wong, and J. S. Bian. 2011. Hydrogen sulfide: neurophysiology and neuropathology. \u003cem\u003eAntioxidants \u0026amp; redox signaling\u003c/em\u003e 15:405\u0026ndash;419.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, Y., S. Cao, D. Tone, H. Fujishima, R. G. Yamada, and R. I. Ohno et al. 2024. Postsynaptic competition between calcineurin and PKA regulates mammalian sleep-wake cycles. \u003cem\u003eNature\u003c/em\u003e 636:412\u0026ndash;421.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeise, S. C., G. Arumugam, A. Villarreal, P. Videm, S. Heidrich, and N. Nebel et al. 2019. FOXG1 Regulates PRKAR2B Transcriptionally and Posttranscriptionally via miR200 in the Adult Hippocampus. \u003cem\u003eMolecular neurobiology\u003c/em\u003e 56:5188\u0026ndash;5201.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBenjamin-Zukerman, T., V. Pane, R. Safadi-Safa, M. Solomon, V. Lev-Ram, and M. Aboraya et al. Allosteric modulation of protein kinase A in individuals affected by NLPD-PKA, a neurodegenerative disease in which the PRKAR1B L50R variant is expressed. The FEBS journal 2025.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFigat, M., A. Wisniewska, J. Plichta, J. Milkowska-Dymanowska, S. Majewski, and M. S. Karbownik et al. 2025. Asthma and COPD Beyond the Airways: Exploring Neurocognitive Links Through NF-κB Subunits c-Rel and p65. \u003cem\u003eInternational journal of molecular sciences\u003c/em\u003e ; 26.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSi, Y., S. Yan, X. Li, W. Ding, X. Zhang, and K. Jia et al. 2025. P2X7R activation promotes ferroptosis in dopaminergic neurons via NF-κB signaling pathway in vitro and in vivo models of MPP(+)/MPTP-induced Parkinson's disease. \u003cem\u003eBrain research\u003c/em\u003e 1865:149824.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"inflammation","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ifla","sideBox":"Learn more about [Inflammation](https://www.springer.com/journal/10753)","snPcode":"10753","submissionUrl":"https://submission.nature.com/new-submission/10753/3","title":"Inflammation","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8361812/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8361812/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eParkinson\u0026rsquo;s disease (PD) is a progressive neurodegenerative disorder, in which ferroptosis and dysregulation of lipid metabolism are believed to play critical roles. However, the key genes and molecular mechanisms underlying these processes remain to be fully elucidated. We combined bioinformatics analyses with experimental validation to identify and characterize central genes involved in ferroptosis and lipid metabolism in PD. Integrated bioinformatics analysis identified 44 candidate genes associated with ferroptosis and lipid metabolism in PD. Machine learning refined these to three core genes: CBS, PRKAR2B, and RELA. Expression analysis revealed significant upregulation of CBS and RELA and downregulation of PRKAR2B in PD samples. ROC analysis indicated strong diagnostic potential for all three genes. Functional enrichment suggested their involvement in neuroinflammation, energy metabolism, and neuroprotective pathways. Immune infiltration analysis revealed significant correlations between these genes and specific immune cell types. In cellular models, knockdown of RELA attenuated MPP⁺-induced oxidative stress, ferroptosis, and inflammatory activation, while also restoring dopaminergic neuronal function. ELISA results from PD patient cerebrospinal fluid samples further confirmed the dysregulation of these genes, supporting their clinical relevance. This study identifies CBS, PRKAR2B, and RELA as key genes linking ferroptosis and lipid metabolism in Parkinson\u0026rsquo;s disease. These genes demonstrate strong diagnostic value and are closely associated with neuroinflammation and immune responses. Experimental validation underscores the protective effect of RELA knockdown against ferroptosis and inflammatory damage. Our findings suggest that RELA may serve as a promising diagnostic biomarker and therapeutic target for PD.\u003c/p\u003e","manuscriptTitle":"RELA as a Diagnostic Biomarker for Parkinson's Disease by Integrating Ferroptosis, Lipid Metabolism, and Neuroinflammation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-22 17:20:18","doi":"10.21203/rs.3.rs-8361812/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-01-07T16:36:57+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-07T13:56:27+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-05T20:54:28+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"288688597201087156625448287382089353623","date":"2025-12-19T14:39:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"135663477758257568060834199014056397832","date":"2025-12-18T16:55:48+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-12-18T16:38:48+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-12-15T22:08:40+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-12-15T22:08:39+00:00","index":"","fulltext":""},{"type":"submitted","content":"Inflammation","date":"2025-12-15T04:36:18+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"inflammation","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ifla","sideBox":"Learn more about [Inflammation](https://www.springer.com/journal/10753)","snPcode":"10753","submissionUrl":"https://submission.nature.com/new-submission/10753/3","title":"Inflammation","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"96465ba0-ecf4-4fc3-a2cf-078fa95efc1e","owner":[],"postedDate":"December 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-03-16T16:14:38+00:00","versionOfRecord":{"articleIdentity":"rs-8361812","link":"https://doi.org/10.1007/s10753-026-02478-7","journal":{"identity":"inflammation","isVorOnly":false,"title":"Inflammation"},"publishedOn":"2026-03-11 15:59:35","publishedOnDateReadable":"March 11th, 2026"},"versionCreatedAt":"2025-12-22 17:20:18","video":"","vorDoi":"10.1007/s10753-026-02478-7","vorDoiUrl":"https://doi.org/10.1007/s10753-026-02478-7","workflowStages":[]},"version":"v1","identity":"rs-8361812","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8361812","identity":"rs-8361812","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}