Identification of key genes for secondary stroke in patients with periodontitis

Identification of key genes for secondary stroke in patients with periodontitis | 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 Article Identification of key genes for secondary stroke in patients with periodontitis zunke gong, hanqing zhao, zhenghao dong, suhua qi, shiyan wang, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5657155/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: The immune system plays a particularly important role in the pathogenesis of both stroke and periodontitis. The aim of this study is to identify key diagnostic candidate genes for stroke in patients with periodontitis. Methods: We searched for a periodontitis dataset and a stroke dataset in the Gene Expression Omnibus (GEO) database. Limma, Weighted Gene Co-expression Network Analysis (WGCNA), immune analysis, single-cell sequencing, and machine learning algorithms were used to identify and analyze immune-related genes associated with periodontitis and stroke. Finally, a nomogram and receiver operating characteristic (ROC) curves were used for evaluation. Results: Limma analysis produced 4252 differentially expressed genes (DEGs) from the integrated stroke dataset and 4113 DEGs from the integrated periodontitis dataset. The periodontitis dataset generated the most relevant module after WGCNA, containing 2028 genes, with a total of 169 intersecting genes among the three datasets. Enrichment analysis revealed that most immune-related genes were enriched, and immune infiltration analysis showed an imbalance of various immune cells. Single-cell sequencing was used to screen immune cells. Further machine learning screening produced three core genes, which were evaluated for their diagnostic value using a nomogram and ROC curves, showing high diagnostic value. Conclusion: Three immune-related core genes (SIGIRR, PLCL2, GLRX) were identified, and a nomogram for diagnosing periodontitis and stroke was established. This study may help identify potential peripheral blood diagnostic candidate genes for secondary stroke in patients with periodontitis. Biological sciences/Computational biology and bioinformatics Biological sciences/Computational biology and bioinformatics/Data mining Biological sciences/Computational biology and bioinformatics/Data processing Biological sciences/Computational biology and bioinformatics/Databases Differentially Expressed Genes Periodontitis Stroke Single-Cell Sequencing Machine Learning Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 1. Introduction According to the World Health Organization, stroke is the second leading cause of death globally and one of the main causes of adult acquired disability(1). Approximately 15 million people suffer from a stroke each year, with one-third of these patients dying and another one-third experiencing permanent disability(2). Strokes are generally classified into two major types: ischemic stroke and hemorrhagic stroke(3). Ischemic stroke, caused by vascular obstruction, accounts for approximately 85% of all stroke cases(4);hemorrhagic stroke, which includes intracerebral hemorrhage and subarachnoid hemorrhage, is caused by vascular rupture(5). The pathogenic factors and mechanisms of stroke are complex, involving multiple biological processes such as vascular occlusion(6), neuronal damage and cell death(6), inflammatory response(7) and hemodynamic changes(8). Currently, there are no specific therapeutic drugs for ischemic stroke, so prevention is crucial(9). In recent years, it has been recognized that chronic infections, inflammation, and immune system abnormalities are related to the risk of ischemic stroke. Periodontitis, as a chronic inflammatory disease, has a significant impact on the occurrence and development of ischemic stroke(10). Periodontitis(11) is a common chronic inflammatory disease in the oral cavity, typically involving the destruction of the supporting structures of the teeth, with a relatively high prevalence among adults worldwide. According to global health studies, about half of adults suffer from mild to moderate periodontitis, while the prevalence of severe periodontitis is approximately 10–15%(12). Periodontitis not only leads to oral health problems but is also associated with various systemic diseases, such as cardiovascular disease, diabetes, respiratory diseases, and preterm birth(13). Stroke and periodontitis have traditionally been regarded as two independent diseases in conventional medical literature, with the former primarily affecting the nervous system and the latter impacting oral health. However, increasing evidence in recent years suggests a possible biological and pathological link between these seemingly unrelated conditions(14). According to the World Health Organization(15), According to the World Health Organization, stroke has an extremely high incidence rate globally and is one of the leading causes of disability and death among adults. At the same time, periodontitis is also a widespread chronic inflammatory disease, affecting the oral health of a large number of adults worldwide(16).Research indicates that the association between periodontitis and stroke is mainly established through inflammatory and immune response mechanisms(14). Periodontitis can cause persistent local and systemic inflammatory responses, which can affect cardiovascular health through various pathways. For example, many researchers believe that chronic low-grade inflammation is a key biological mechanism linking periodontitis and stroke. Studies have found that individuals with periodontitis have significantly elevated levels of inflammatory markers in their blood, such as C-reactive protein and interleukin-6, which are closely related to the occurrence of stroke(17). Additionally, oral pathogens like Porphyromonas gingivalis and Actinomyces have been detected in carotid plaques of stroke patients, suggesting that periodontal pathogens may directly participate in the formation of arterial plaques(18). Currently, there is no definitive research indicating a direct and clear relationship between periodontitis and stroke. This study employs bioinformatics methods to screen for biomarkers related to immune infiltration in both conditions. This can help identify potential diagnostic markers associated with immunity in patients with concurrent stroke and periodontitis, providing a basis for clinical diagnosis and research. 2. Materials and Methods 2.1 Microarray Data The datasets used in this study were obtained from the NCBI Gene Expression Omnibus (GEO) database(23), The periodontitis dataset is GSE106090(24);and the stroke dataset is GSE198710(25). Figure 1 depicts the study flowchart. GSE106090: Offers extensive gene expression data on periodontitis, covering various samples of periodontitis and healthy controls, making it ideal for exploring the molecular mechanisms of periodontitis. GSE198710: Contains gene expression data from stroke patients, providing a comparison between stroke and non-stroke samples, which helps in uncovering the molecular characteristics of stroke. 2.2 Data Processing and Differential Gene Screening Differential analysis of protein expression data from different treatments was performed using the R package "limma," with a significance threshold of P < 0.05 and an absolute log fold change of ≥ 0.5. In LIMMA analysis, P-values and adjusted P-values (FDR) were calculated, with DEGs identified using thresholds like P < 0.05. This controls the false discovery rate and ensures reliability. Volcano plots for differential analysis were created using the R package "ggplot2," and heatmaps were generated using the R package "pheatmap." Intersections of different groups and Venn diagrams were created using the online tool Evenn ( http://www.ehbio.com/test/venn ). These tools were used to create volcano plots and heatmaps for visualizing DEG results, offering an intuitive view of data distribution and patterns. 2.3 Weighted Gene Co-expression Network Analysis WGCNA(26) was used to reveal expression patterns among proteins and construct co-expression networks to identify modules of protein sets related to biological phenotypes. Initial clustering analysis showed small intra-group differences and large inter-group differences among samples. The optimal soft threshold of 20 was determined by comparing scale independence and mean connectivity, ensuring the constructed network conformed more closely to a scale-free topology. Proteins with different expression patterns were clustered into different color modules, with correlation analysis indicating significantly weaker correlations between different expression modules compared to those within the same module. 2.4 Functional Enrichment Analysis Enrichment analysis of intersecting genes from the stroke, periodontitis, and periodontitis WGCNA datasets was conducted using the Enrichr database ( https://maayanlab.cloud/Enrichr/ ), focusing on pathways, biological processes (BP), cellular components (CC), and molecular functions (MF). Pathway annotation was performed using the Enrichr tool, which provides a comprehensive and user-friendly platform for gene set enrichment analysis. 2.5 Immune Infiltration Analysis The CIBERSORT algorithm(27) was used to determine the proportion of immune cells in tissues or cells(28). Bar plots displayed the proportions of various immune cell types in different samples. Correlation heatmaps of immune cells were generated using the "corrplot"(29) R package. Heatmaps created with pheatmap and ComplexHeatmap visualized patterns and relationships in the data. Stacked proportion plots illustrated the proportions of different cell types in each sample. Spearman correlations between datasets (e.g., gene expression data and CIBERSORT results) were calculated using cor.test. 2.6 Single-Cell Sequencing Single-cell sequencing (scRNA-seq) was used to study expression levels within individual cells, revealing heterogeneity among cell types. This study analyzed single-cell sequencing data from the public database GSE225948(30), Standard procedures, including quality control, genome alignment, and transcript quantification, were applied to the raw sequencing data. Transcript count data were normalized to eliminate systematic biases between samples. The t-SNE (t-Distributed Stochastic Neighbor Embedding) algorithm was used for dimensionality reduction, visualizing complex high-dimensional data in two dimensions. Cells or gene expression features were processed with t-SNE, generating two-dimensional coordinates. Points were clustered based on t-SNE results to identify different cell populations or expression patterns. Differences in t-SNE distributions under different experimental conditions (e.g., normal control group vs. experimental group) were compared to explore the impact of experimental treatments on cell or gene expression. Statistical tests such as ANOVA or t-test compared expression level differences in different cell types or genes under various experimental conditions. Scatter plots were generated using graphical software (R's ggplot2), where each point represented a cell or gene, with colors and shapes indicating different cell types or experimental conditions. GSE225948: Provides single-cell RNA sequencing data, allowing for an in-depth investigation of cell heterogeneity and cell type-specific gene expression patterns. 2.7 Machine Learning Machine learning algorithms identified core genes for diagnosing periodontitis. "glmnet"(31) was used for LASSO(32) regression, “random Forest”(33) was used for RF(34) analysis. The intersecting genes from these methods were identified as core genes for diagnosing stroke combined with periodontitis. 2.8 Nomogram Construction and Receiver Operating Characteristic Evaluation The "rms"(35) R package was used to create a nomogram for the screened genes, with clinical value evaluated by the area under the curve (AUC) and 95% confidence intervals (CI) using ROC analysis. AUC values over 0.7 indicated diagnostic value. 2.9 Statistical Analysis High-throughput sequencing data analysis in this study utilized the R package "limma" to evaluate differential expression using linear models, with empirical Bayesian methods enhancing statistical inference robustness. For single-cell data, the R package "Seurat" used z-score transformations to compare expression levels of different genes, employing Wilcoxon rank-sum tests to identify marker genes distinguishing different cell populations. 3. Results 3.1 Differentially Expressed Genes Log-transformed data from the periodontitis and stroke datasets showed a normal distribution, followed by quantile normalization for comparability between standardized data groups (Figs. 2A, 2B, 4A, 4B). Principal component analysis indicated significant differences between groups under different treatments (Figs. 2C, 4C). LIMMA analysis identified 4113 upregulated and 4340 downregulated genes in the integrated periodontitis dataset, and 4369 upregulated and 5016 downregulated genes in the stroke dataset (Figs. 4A,4B). Figures 2D, 2E, and 4D, 4E display volcano plots and heatmaps generated from these data. 3.2 Key Module Selection WGCNA identified key modules in periodontitis. Initial clustering analysis indicated small intra-group differences and large inter-group differences (Fig. 3A). The optimal soft threshold of 20 ensured the constructed network conformed more closely to a scale-free topology (Fig. 3B). Proteins with different expression patterns were clustered into various color modules (Fig. 3C). The brown module, containing 691 genes, showed the highest correlation with periodontitis (cor = 0.99, P = 4e-10), making it a key module for subsequent analysis (Fig. 3D). Further exploration of the relationship between key modules and specific experimental conditions validated the correlation of module member proteins with traits, all of which were statistically significant (Fig. 3E). Correlation analysis indicated significantly weaker correlations between different expression modules compared to those within the same module (Fig. 3F), further supporting the decisive role of proteins closely linked to experimental conditions within their modules. 3.3 Enrichment Analysis for Stroke Combined with Periodontitis The intersection of 4113 upregulated periodontitis genes, 4369 upregulated stroke genes, and 691 brown module genes resulted in 169 genes (Fig. 4F). KEGG analysis showed enrichment mainly in "Viral protein interaction with cytokine and cytokine receptor" and "Cytokine-cytokine receptor interaction" (Figs. 5A, 5B). GO analysis revealed that BP terms were mainly enriched in "Regulation of Transcription by RNA Polymerase II" and "Regulation of Cell Population Proliferation" (Figs. 5C, 5D). CC terms were primarily found in "Endoplasmic Reticulum Membrane," "Golgi Membrane," and "Late Endosome Lumen" (Figs. 5E, 5F). MF analysis showed enrichment in "Sequence-Specific DNA Binding," "RNA Polymerase II Cis-Regulatory Region Sequence-Specific DNA Binding," and "RNA Binding" (Figs. 5G, 5H). 3.4 Immune Infiltration Analysis We found that genes associated with stroke also played roles in periodontitis, primarily in immune regulation. Therefore, we conducted immune analysis on the 169 intersecting genes. Figure 6A shows the distribution of various immune cell types in different samples. Proportions of cell types, such as memory B cells, plasma cells, activated and resting T cells, NK cells, various macrophages, and dendritic cells, were detailed to compare differences between healthy controls and pathological samples. Figure 6B's heatmap used color to represent relative expression levels or activity of immune cell types in different samples. Figure 6C's boxplot compared the composition of different immune cell types in two sample groups (H group and P group), highlighting significant changes in T cells CD4 memory resting and Plasma cells as potential therapeutic targets. Figure 6D's correlation matrix displayed correlations between different immune cell types using color and numerical values. Higher positive correlations included NK cells and CD4 memory T cells (correlation coefficient ~ 0.72), monocytes and plasma cells (~ 0.83), and memory B cells and dendritic cells (~ 0.79). Higher negative correlations included monocytes and NK cells (~ 0.80) and monocytes and T cells (activated/resting) (~ 0.74). Figure 6E displayed correlations between various immune cell types and different biomarkers (mainly genes). Genes were represented in columns, while immune cell types were in rows. The heatmap used color and symbols to indicate correlation strength and statistical significance, showing T cells (activated/resting) and Plasma cells as significantly positively correlated, while resting dendritic cells were negatively correlated, suggesting T cells (activated/resting) and Plasma cells as potential therapeutic nodes. 3.5 Single-Cell Sequencing This study analyzed transcriptomic single-cell sequencing data from brain tissue in a mouse ischemia-reperfusion model. After outlier removal and normalization of single-cell sequencing data, cell types in t-SNE clusters were annotated (Fig. 7A). The study observed changes in upregulated gene levels in Bc4 and Tc2 in stroke and periodontitis. Figure 7B displayed cell distribution under different experimental conditions (e.g., mcao and sham), with regions highlighting two main cell populations, Bc4 (Plasma cells) and Tc2 (T cells CD4 memory activate). Figure 7C showed expression levels of different genes (e.g., Cd4, Igtbp4, Cd19, Cd38) in cells, with Cd19 and Cd38 as Plasma cell markers and Cd4 and Igtbp4 as T cells (activated/resting) markers. Color intensity represented gene expression levels, with dark colors indicating high expression and light colors indicating low expression. Figure 7D's Dot Plot visualized gene expression across different cell types, identifying genes highly expressed in Plasma cells and T cells (activated/resting). 3.6 Machine Learning Core genes were identified using machine learning. Lasso regression and RF algorithms identified core genes and created related nomograms for ROC analysis. Lasso regression selected 9 candidate genes (Figs. 8A, 8B), and RF identified 10 important genes (Figs. 8C, 8D). The intersection of the two methods (Fig. 8E) ultimately identified three genes: SIGIRR, PLCL2, and GLRX. 3.7 Determining Diagnostic Value ROC curves were plotted for the three candidate genes (Figs. 8F, 8G, 8H), and a nomogram was created (Fig. 8I) to evaluate the diagnostic value of each gene. Calculated AUCs and 95% confidence intervals were as follows: SIGIRR (AUC: 0.830, CI: 0.728–0.938), PLCL2 (AUC: 0.733, CI: 0.602–0.851), GLRX (AUC: 0.801, CI: 0.683–0.908). The results indicated high diagnostic value for the identified genes in diagnosing stroke combined with periodontitis. 4. Discussion We must recognize that stroke is not only one of the leading causes of death globally but also a major cause of long-term disability in adults. According to the World Health Organization, approximately 15 million people suffer from strokes each year, with one-third of these patients eventually dying, and another one-third experiencing severe long-term disabilities(36). These disabilities include language disorders, motor function impairments, and cognitive deficits(37), significantly impacting patients' quality of life and imposing substantial burdens on families and society. Recent studies have identified several biomarkers for stroke, including glutamate(38), glutamate oxaloacetate transaminase (GOT)(37),S100B(39) and the receptor for advanced glycation end products (RAGE)(39). However, research combining these two diseases is relatively rare. There are currently no biomarkers for diagnosing stroke using nomograms and machine learning methods. Therefore, we integrated bioinformatics analysis, single-cell sequencing, and machine learning, using nomograms and ROC curves to evaluate their diagnostic value. Notably, we identified three key immune-related candidate genes (SIGIRR, PLCL2, and GLRX) and developed a nomogram for diagnosing stroke in periodontitis patients. By simply detecting peripheral blood in periodontitis patients(40), we can estimate the probability of stroke secondary to periodontitis. Given that the stroke and periodontitis samples used in this study were derived from peripheral blood, patients exhibited known core gene expressions. Peripheral blood testing is widely used for diagnosing many diseases due to its simplicity and efficacy. Our next step is to develop a more refined model that accurately reflects gene expression and assigns values, enhancing diagnostic accuracy. Early monitoring and intervention for significant changes in target indicators in periodontitis patients would be valuable for diagnosing stroke combined with periodontitis. SIGIRR (Single Ig IL-1R-related receptor) is a negative regulatory immune receptor mainly expressed in epithelial cells and some immune cells(41). It reduces inflammatory responses by inhibiting Toll-like receptor (TLR) and IL-1 receptor family signaling. SIGIRR helps maintain immune balance and prevent excessive inflammation by forming heterodimers with other TIR domain proteins, thus reducing cell surface receptor signaling. As a negative regulator of IL-1 and TLR signaling pathways(42),SIGIRR can inhibit their activation, reducing inflammation caused by these pathways. In stroke contexts, this anti-inflammatory effect may help mitigate secondary brain damage, as post-stroke inflammation often exacerbates brain injury. SIGIRR also plays an important role in periodontitis. As a chronic inflammatory disease involving the destruction of periodontal tissues, periodontitis can lead to significant oral health issues. By inhibiting inflammatory responses, SIGIRR may help reduce inflammation and damage to periodontal tissues. Studies have shown that patients with periodontitis often have lower levels of SIGIRR expression, which may contribute to heightened inflammatory responses. Thus, SIGIRR's anti-inflammatory effects could be valuable in the treatment of periodontitis. PLCL2 (Phospholipase C-like 2) is a phosphatidylinositol-specific phosphatase crucial for regulating lipid metabolism. Dysregulated lipid metabolism is a primary cause of atherosclerosis. High levels of low-density lipoprotein (LDL) and low levels of high-density lipoprotein (HDL) are key factors in atherosclerosis(43). Oxidized LDL (ox-LDL) is ingested by macrophages, forming foam cells and contributing to arterial plaque formation and progression(44). Atherosclerosis results from lipid accumulation in arterial walls, triggering chronic inflammation. PLCL2 influences foam cell formation and arterial plaque development by regulating genes related to lipid metabolism. Studies show PLCL2 is significantly associated with high expression of lipid metabolism-related genes, including CD36, LPL, and PPARG, which play critical roles in atherosclerosis pathology. Atherosclerosis is a major cause of ischemic stroke(45, 46). Arterial plaque formation and rupture can interrupt blood supply to the brain, causing strokes. PLCL2 indirectly affects stroke risk by regulating lipid metabolism and atherosclerosis. Inhibiting PLCL2 expression or function may reduce ox-LDL formation and foam cell accumulation, lowering atherosclerosis and stroke risk(45, 47). In periodontitis, PLCL2 may also play a significant role. Patients with periodontitis often have an increased risk of atherosclerosis, suggesting shared pathological mechanisms between the two conditions. By modulating lipid metabolism, PLCL2 may serve as a bridge between periodontitis and atherosclerosis. Research suggests that PLCL2 expression levels in patients with periodontitis may influence lipid metabolism, thereby exacerbating the progression of atherosclerosis. Therefore, PLCL2 could be a potential target for the prevention of both periodontitis and stroke. GLRX (Glutaredoxin) is a key antioxidant enzyme in the thioredoxin family, primarily maintaining redox balance in cells. GLRX controls intracellular free radical levels by catalyzing reactions between thiol groups in proteins and glutathione (GSH), protecting cells from oxidative damage(48). In stroke, GLRX's antioxidant function is crucial for protecting brain cells from free radical damage due to increased oxidative stress(48). Oxidative stress exacerbates brain injury in stroke, so GLRX may help mitigate this damage through its antioxidant mechanism. In periodontitis, GLRX's antioxidant function is equally important. Patients with periodontitis often experience oxidative stress, which can lead to damage of periodontal tissues. Through its antioxidant effects, GLRX can effectively clear free radicals and reduce oxidative stress-induced damage to periodontal tissues. Studies have shown that GLRX expression levels are often lower in patients with periodontitis, which may exacerbate oxidative stress. Therefore, GLRX's antioxidant functions could be valuable in the treatment of periodontitis. Enhancing GLRX expression may help reduce oxidative damage to periodontal tissues and promote their repair and regeneration. Immune response, inflammatory response, and oxidative stress response are present in all stages of stroke and significantly impact disease progression. Stroke leads to neuron death and inflammation, triggering immune cell migration from peripheral tissues to the brain, causing secondary damage. These immune and inflammatory pathways are potential stroke treatment targets. For instance, platelet activation and aggregation in the acute phase of stroke promote neurodamage progression by enhancing inflammation and thrombosis. Moreover, reperfusion therapy may help restore blood flow and reduce brain tissue damage, but the immune system may maintain inflammation post-reperfusion. Research shows post-stroke immune responses involve brain tissue infiltration by neutrophils, macrophages, B cells, T cells, and dendritic cells. These cells activate downstream signaling pathways through pattern recognition receptors like TLRs and NLRs, producing pro-inflammatory cytokines, reactive oxygen species, and chemokines, exacerbating post-ischemia inflammation. Immune response regulation plays a crucial role in improving stroke outcomes. Appropriate immunomodulatory treatments, such as immunomodulators, can effectively reduce brain damage, lesion size, hemorrhage, and improve neurological recovery. Understanding these mechanisms is essential for developing new treatment strategies and enhancing stroke management effectiveness. 5. Limitations The current study has several limitations that should be acknowledged. Firstly, the sample size was relatively small, which may limit the statistical power and generalizability of the findings. The datasets used in this study were sourced from public databases, which, while valuable resources, may have inherent limitations such as variability in data collection protocols, potential batch effects, and missing clinical information for some samples. These factors could introduce noise or bias into the analysis, affecting the reliability of the results. Additionally, given the complexity of both stroke and periodontitis, involving multiple genetic, environmental, and lifestyle factors, it is possible that not all relevant genes and biological processes were captured in this study. The analysis focused on specific datasets and methodologies, which might have excluded other important genes or interactions that could play a role in the pathogenesis of these diseases. Furthermore, the cross-sectional nature of the data limits the ability to establish causal relationships and assess the temporal dynamics of gene expression changes in relation to disease progression. 6. Future Directions Future research in this area should aim to address the limitations identified in this study. One critical step is to expand the sample size by including more diverse patient cohorts from different geographic regions, ethnic backgrounds, and clinical settings. This would enhance the representativeness of the data and improve the robustness of the findings. Additionally, incorporating multi-omics data integration could provide a more comprehensive understanding of the molecular mechanisms underlying stroke and periodontitis. For instance, combining transcriptomics with proteomics and metabolomics data would allow for the identification of not only differentially expressed genes but also the corresponding proteins and metabolites, offering insights into the functional implications of these changes at various biological levels. Furthermore, functional studies in vitro and in vivo are necessary to validate the roles of the identified genes (e.g., SIGIRR, PLCL2, GLRX) in disease pathogenesis. This could involve experiments such as gene knockdown or overexpression studies to assess their impact on relevant cellular processes and disease models. Moreover, exploring the therapeutic potential of these genes through drug target validation and preclinical studies could pave the way for the development of novel treatment strategies. Finally, leveraging advanced computational models and machine learning algorithms with improved interpretability could help in uncovering more nuanced patterns in the data and identifying additional biomarkers or therapeutic targets that may have been overlooked in this study. 7. Conclusion Through bioinformatics analysis and machine learning, we systematically identified three relevant candidate genes (SIGIRR, PLCL2, GLRX) and provided a template for diagnosing periodontitis combined with stroke. We also noted immune system imbalances in stroke patients with periodontitis, where immune cell percentages are influenced by the immune microenvironment. The screened genes can be used for clinical diagnosis and treatment. References Gorelick PB, Whelton PK, Sorond F, Carey RM. Blood Pressure Management in Stroke. Hypertension. 2020;76(6):1688-95. Kainz A, Meisinger C, Linseisen J, Kirchberger I, Zickler P, Naumann M, et al. Changes of Health-Related Quality of Life Within the 1st Year After Stroke–Results From a Prospective Stroke Cohort Study. Frontiers in Neurology. 2021;12. Hasan TF, Hasan H, Kelley RE. Overview of Acute Ischemic Stroke Evaluation and Management. Biomedicines. 2021;9(10):1486-. Potter TBH, Tannous J, Vahidy FS. A Contemporary Review of Epidemiology, Risk Factors, Etiology, and Outcomes of Premature Stroke. Current Atherosclerosis Reports. 2022;24(12):939-48. Chen S, Zeng L, Hu Z. Progressing haemorrhagic stroke: categories, causes, mechanisms and managements. Journal of Neurology. 2014;261(11):2061-78. Salaudeen MA, Bello N, Danraka RN, Ammani ML. Understanding the Pathophysiology of Ischemic Stroke: The Basis of Current Therapies and Opportunity for New Ones. Biomolecules. 2024;14(3). Choi YH, Laaker C, Hsu M, Cismaru P, Sandor M, Fabry Z. Molecular Mechanisms of Neuroimmune Crosstalk in the Pathogenesis of Stroke. International Journal of Molecular Sciences. 2021;22(17). Asil T, Utku U, Balci K, Uzunca I. Changing cerebral blood flow velocity by transcranial Doppler during head up tilt in patients with diabetes mellitus. Clinical neurology and neurosurgery. 2007;109(1):1-6. Zheng X, Li X, Zhen J, Xue D, Hu J, Cao Q, et al. Periodontitis is associated with stroke. Journal of Translational Medicine. 2023;21(1):697. Sharma A, Sharma A, Chauhan R, Chauhan AS, Parmar M, Thakur S. The Relationship Between Periodontal Disease (PD) and Recurrent Vascular Events in Ischemic Stroke/Transient Ischemic Attack (TIA) Patients: A Hospital-Based Cohort Study. Cureus. 2023;15(3):e36530. Könönen E, Gursoy M, Gursoy U. Periodontitis: A Multifaceted Disease of Tooth-Supporting Tissues. Journal of Clinical Medicine. 2019;8(8). Pink C, Holtfreter B, Völzke H, Nauck M, Dörr M, Kocher T. Periodontitis and systemic inflammation as independent and interacting risk factors for mortality: evidence from a prospective cohort study. BMC Medicine. 2023;21(1). Bhuyan R, Bhuyan SK, Mohanty JN, Das S, Juliana N, Juliana IF. Periodontitis and Its Inflammatory Changes Linked to Various Systemic Diseases: A Review of Its Underlying Mechanisms. Biomedicines. 2022;10(10). Baniulyte G, Piela K, Culshaw S. How strong is the link between periodontitis and stroke? Evidence-Based Dentistry. 2021;22(1):10-1. Du M, Mi D, Liu M, Liu J. Global trends and regional differences in disease burden of stroke among children: a trend analysis based on the global burden of disease study 2019. BMC Public Health. 2023;23(1):2120. Pink C, Holtfreter B, Völzke H, Nauck M, Dörr M, Kocher T. Periodontitis and systemic inflammation as independent and interacting risk factors for mortality: evidence from a prospective cohort study. BMC Medicine. 2023;21(1):430. Fagundes NCF, Almeida APCPSC, Vilhena KFB, Magno MB, Maia LC, Lima RR. Periodontitis As A Risk Factor For Stroke: A Systematic Review And Meta-Analysis. Vascular Health and Risk Management. 2019;Volume 15:519-32. Guo J, Xu R, Liu R, Lai W, Hu C, He H, et al. Association between the systemic immune inflammation index and periodontitis: a cross-sectional study. Journal of Translational Medicine. 2024;22(1):96. Panda AK, Jin Q, Teng F, Cheng Z. Association between common polymorphisms in IL-1 and TNFα and risk of peri-implant disease: A meta-analysis. Plos One. 2021;16(10). Martínez-García M, Hernández-Lemus E. Periodontal Inflammation and Systemic Diseases: An Overview. Frontiers in Physiology. 2021;12. Aleksandrowicz P, Brzezińska-Błaszczyk E, Kozłowska E, Żelechowska P, Borgonovo AE, Agier J. Analysis of IL-1β, CXCL8, and TNF-α levels in the crevicular fluid of patients with periodontitis or healthy implants. BMC Oral Health. 2021;21(1):120. Wang RP-H, Huang J, Chan KWY, Leung WK, Goto T, Ho Y-S, et al. IL-1β and TNF-α play an important role in modulating the risk of periodontitis and Alzheimer’s disease. Journal of Neuroinflammation. 2023;20(1):71. Barrett T, Wilhite SE, Ledoux P, Evangelista C, Kim IF, Tomashevsky M, et al. NCBI GEO: archive for functional genomics data sets--update. Nucleic acids research. 2013;41(Database issue):D991-5. Liu Y, Liu Q, Li Z, Acharya A, Chen D, Chen Z, et al. Long non-coding RNA and mRNA expression profiles in peri-implantitis vs periodontitis. Journal of periodontal research. 2020;55(3):342-53. Jiang W, Li J, Cai Y, Liu W, Chen M, Xu X, et al. The Novel lncRNA ENST00000530525 Affects ANO1, Contributing to Blood-Brain Barrier Injury in Cultured hCMEC/D3 Cells Under OGD/R Conditions. Frontiers in genetics. 2022;13:873230. Langfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC bioinformatics. 2008;9:559. Chen B, Khodadoust MS, Liu CL, Newman AM, Alizadeh AA. Profiling Tumor Infiltrating Immune Cells with CIBERSORT. Methods in molecular biology (Clifton, NJ). 2018;1711:243-59. Newman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, et al. Robust enumeration of cell subsets from tissue expression profiles. Nature methods. 2015;12(5):453-7. Hu K. Become Competent within One Day in Generating Boxplots and Violin Plots for a Novice without Prior R Experience. Methods and protocols. 2020;3(4). Garcia-Bonilla L, Shahanoor Z, Sciortino R, Nazarzoda O, Racchumi G, Iadecola C, et al. Analysis of brain and blood single-cell transcriptomics in acute and subacute phases after experimental stroke. Nature immunology. 2024;25(2):357-70. Yang C, Delcher C, Shenkman E, Ranka S. Machine learning approaches for predicting high cost high need patient expenditures in health care. Biomedical engineering online. 2018;17(Suppl 1):131. Ellis K, Kerr J, Godbole S, Lanckriet G, Wing D, Marshall S. A random forest classifier for the prediction of energy expenditure and type of physical activity from wrist and hip accelerometers. Physiological measurement. 2014;35(11):2191-203. Zhang M, Zhu K, Pu H, Wang Z, Zhao H, Zhang J, et al. An Immune-Related Signature Predicts Survival in Patients With Lung Adenocarcinoma. Frontiers in oncology. 2019;9:1314. Alderden J, Pepper GA, Wilson A, Whitney JD, Richardson S, Butcher R, et al. Predicting Pressure Injury in Critical Care Patients: A Machine-Learning Model. American journal of critical care : an official publication, American Association of Critical-Care Nurses. 2018;27(6):461-8. Pan X, Jin X, Wang J, Hu Q, Dai B. Placenta inflammation is closely associated with gestational diabetes mellitus. American journal of translational research. 2021;13(5):4068-79. Shahi S, Farhoudi M, Dizaj SM, Sharifi S, Sadigh-Eteghad S, Goh KW, et al. The Link between Stroke Risk and Orodental Status—A Comprehensive Review. Journal of Clinical Medicine. 2022;11(19). Hansen PR, Holmstrup P. Cardiovascular Diseases and Periodontitis. In: Santi-Rocca J, editor. Periodontitis: Advances in Experimental Research. Cham: Springer International Publishing; 2022. p. 261-80. Naderi S, Merchant AT. The Association Between Periodontitis and Cardiovascular Disease: an Update. Current Atherosclerosis Reports. 2020;22(10):52. Shahi S, Farhoudi M, Dizaj SM, Sharifi S, Sadigh-Eteghad S, Goh KW, et al. The Link between Stroke Risk and Orodental Status—A Comprehensive Review. 2022;11(19):5854. Wang H, Zhang S, Xie L, Zhong Z, Yan F. Neuroinflammation and peripheral immunity: Focus on ischemic stroke. International immunopharmacology. 2023;120:110332. Riva F, Bonavita E, Barbati E, Muzio M, Mantovani A, Garlanda C. TIR8/SIGIRR is an Interleukin-1 Receptor/Toll Like Receptor Family Member with Regulatory Functions in Inflammation and Immunity. 2012;3. Liu Z, Zhu L, Lu Z, Chen H, Fan L, Xue Q, et al. IL-37 Represses the Autoimmunity in Myasthenia Gravis via Directly Targeting Follicular Th and B Cells. Journal of immunology (Baltimore, Md : 1950). 2020;204(7):1736-45. Pan X, Liu J, Zhong L, Zhang Y, Liu C, Gao J, et al. Identification of lipid metabolism-related biomarkers for diagnosis and molecular classification of atherosclerosis. Lipids in Health and Disease. 2023;22(1):96. Gaggini M, Gorini F, Vassalle C. Lipids in Atherosclerosis: Pathophysiology and the Role of Calculated Lipid Indices in Assessing Cardiovascular Risk in Patients with Hyperlipidemia. 2023;24(1):75. Chow Y-L, Teh Kuan L, Chyi Huey L, Lim Fang L, Yee Chu C, Wei Keat L. Lipid Metabolism Genes in Stroke Pathogenesis: The Atherosclerosis. Current Pharmaceutical Design. 2020;26(34):4261-71. Duan H, Song P, Li R, Su H, He L. Attenuating lipid metabolism in atherosclerosis: The potential role of Anti-oxidative effects on low-density lipoprotein of herbal medicines. 2023;14. Kloska A, Malinowska M, Gabig-Cimińska M, Jakóbkiewicz-Banecka J. Lipids and Lipid Mediators Associated with the Risk and Pathology of Ischemic Stroke. 2020;21(10):3618. Burns M, Rizvi SHM, Tsukahara Y, Pimentel DR, Luptak I, Hamburg NM, et al. Role of Glutaredoxin-1 and Glutathionylation in Cardiovascular Diseases. 2020;21(18):6803. Additional Declarations No competing interests reported. Supplementary Files GSE198710.zip GSE106090.zip GSE225948.zip KEGGPERMISSION251070.pdf Cite Share Download PDF Status: Posted Version 1 posted 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-5657155","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":449726025,"identity":"c1d13e24-20ba-41b7-b604-5ed1e9bd6a4c","order_by":0,"name":"zunke gong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAx0lEQVRIiWNgGAWjYBADAwb2xsaHH0jTwnO42ViCNC0S6W0CPMQolXfvPfzyZ9sdY37Jh20MEgx2croNBLQYnjmXZiHZ9sxMcnZi24MChmRjswOEtMzIMTMwbDtsY3A7sd1AguFA4jaitCQCtdjfPNgmwUOMFnmJHOMHB9sOmxlIMBKpxYDnjBljw7nDxhJnEoGBbECEX+Tbe4w//ig7bNjffvzhww8VdnIEtRgcYGBDikADAsrBtjQwMJOUTEbBKBgFo2AEAgAaVkQSXYvSQgAAAABJRU5ErkJggg==","orcid":"","institution":"Xuzhou Rehabilitation Hospital Affiliated to Xuzhou Medical University, Xuzhou Central Hospital","correspondingAuthor":true,"prefix":"","firstName":"zunke","middleName":"","lastName":"gong","suffix":""},{"id":449726026,"identity":"00485c70-68bf-4fee-a79a-c32c11d78867","order_by":1,"name":"hanqing zhao","email":"","orcid":"","institution":"Xuzhou Medical College","correspondingAuthor":false,"prefix":"","firstName":"hanqing","middleName":"","lastName":"zhao","suffix":""},{"id":449726027,"identity":"cdac37e4-16b9-42c8-a173-cb1e415f910c","order_by":2,"name":"zhenghao dong","email":"","orcid":"","institution":"Xuzhou Medical College","correspondingAuthor":false,"prefix":"","firstName":"zhenghao","middleName":"","lastName":"dong","suffix":""},{"id":449726028,"identity":"4c4837ae-4cdb-44d1-a6d3-c5b6abe1842b","order_by":3,"name":"suhua qi","email":"","orcid":"","institution":"School of Medical Technology, Xuzhou Medical University, Xuzhou Key Laboratory of Laboratory Diagnostics","correspondingAuthor":false,"prefix":"","firstName":"suhua","middleName":"","lastName":"qi","suffix":""},{"id":449726029,"identity":"b92abd49-1356-4b88-adf6-8b4b0ab03af8","order_by":4,"name":"shiyan wang","email":"","orcid":"","institution":"Xuzhou Rehabilitation Hospital Affiliated to Xuzhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"shiyan","middleName":"","lastName":"wang","suffix":""},{"id":449726030,"identity":"9c51209e-4c3f-44ce-ae6f-5cf58bebe5f3","order_by":5,"name":"xiang wang","email":"","orcid":"","institution":"Xuzhou Rehabilitation Hospital Affiliated to Xuzhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"xiang","middleName":"","lastName":"wang","suffix":""}],"badges":[],"createdAt":"2024-12-17 00:53:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5657155/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5657155/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":82044112,"identity":"28b5fd1b-6b02-409b-8e32-b3480c07d7c3","added_by":"auto","created_at":"2025-05-06 09:28:49","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":428274,"visible":true,"origin":"","legend":"\u003cp\u003eStudy flowchart, GSE, gene expression omnibus series; WGCNA, weighted gene co-expression network analysis; Limma, linear models for microarray data; DEGs, differentially expressed genes;\u003cstrong\u003e \u003c/strong\u003eSingle-Cell Sequencing.\u003c/p\u003e","description":"","filename":"Fig1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5657155/v1/fe919e89be170eb92b655e54.jpg"},{"id":82044114,"identity":"3b8934c3-7acd-4618-b976-deceff118a8a","added_by":"auto","created_at":"2025-05-06 09:28:49","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":564396,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eHeatmap and valcano plot for the DEGs identified from the integrated Periodontitis dataset.\u003c/strong\u003e (A) Bar plot showing the gene expression levels (log2 scale) across 12 samples before normalization. (B) Bar plot showing the gene expression levels (log2 scale) across 12 samples after normalization.(C) Principal Component Analysis (PCA) of gene expression data distinguishing between control (H) and periodontitis (P) groups. (D) Red and Blue plot triangles represent DEGs with upregulated and downregulated gene expression. (E)Each row shows the DEGs, and each column refers to one of the samples of Periodontitis cases or controls. The red and blue represent DEGs with upregulated and downregulated gene expression, respectively.\u003c/p\u003e","description":"","filename":"Fig2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5657155/v1/86bb1f2492eecaf14ddc7e38.jpg"},{"id":82044113,"identity":"bad4378c-a8ad-490e-8d96-d852e806be8f","added_by":"auto","created_at":"2025-05-06 09:28:49","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":912709,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eWeighted Gene Co-expression Network Analysis (WGCNA) and module analysis.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A) \u003c/strong\u003eA hierarchical clustering dendrogram of samples based on gene expression profiles is shown.\u003cstrong\u003e (B) \u003c/strong\u003eThe left plot displays scale-free topology fit index (R²) as a function of the soft threshold (power).\u003cstrong\u003e (C) \u003c/strong\u003eA dendrogram showing the hierarchical clustering of genes based on their co-expression patterns is depicted.\u003cstrong\u003e (D)\u003c/strong\u003e A heatmap illustrates the correlation between gene modules and various traits.\u003cstrong\u003e (E) \u003c/strong\u003eA scatter plot shows the relationship between module membership (MM) and gene significance (GS) for the brown module.\u003cstrong\u003e (F) \u003c/strong\u003eA heatmap-based network plot visualizes the interactions among selected genes.\u003c/p\u003e","description":"","filename":"Fig3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5657155/v1/3f3aebe4a863632765164241.jpg"},{"id":82044119,"identity":"c49b0bd1-0a9a-4e86-8dcd-fd1def967f60","added_by":"auto","created_at":"2025-05-06 09:28:50","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":725006,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eHeatmap and valcano plot for the DEGs identified from the integrated Stroke dataset.\u003c/strong\u003e (A) Bar plot showing the gene expression levels (log2 scale) across 10 samples before normalization. Bar plot showing the gene expression levels (log2 scale) across 10 samples after normalization.(C) Principal Component Analysis (PCA) of gene expression data distinguishing between control (C) and Stroke (S) groups. (D) Red and Blue plot triangles represent DEGs with upregulated and downregulated gene expression. (E)Each row shows the DEGs, and each column refers to one of the samples of Stroke cases or controls. The red and blue represent DEGs with upregulated and downregulated gene expression, respectively. (F) Identifying the common elements within three groups.\u003c/p\u003e","description":"","filename":"Fig4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5657155/v1/cc1b2559e362fbca72d285ee.jpg"},{"id":82047948,"identity":"540c6cc3-1dcd-486f-b250-b1bf884a361b","added_by":"auto","created_at":"2025-05-06 09:44:50","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1208496,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEnrichment Analysis\u003c/strong\u003e (A) Pathway Enrichment Analysis Dot Plot(B) Pathway Interaction Chord Diagram(C) Cellular Component Enrichment Analysis Dot Plot(D) Cellular Component Interaction Chord Diagram(E) Molecular Function Enrichment Analysis Dot Plot (F) Molecular Function Interaction Chord Diagram (G) Biological Process Enrichment Analysis Dot Plot(H) Biological Process Interaction Chord Diagram\u003c/p\u003e","description":"","filename":"Fig5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5657155/v1/7543e5eddd7752a49da72cc3.jpg"},{"id":82045578,"identity":"0f6032cc-de7a-4d67-aba0-8ce088d787c2","added_by":"auto","created_at":"2025-05-06 09:36:50","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1224774,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eComprehensive analysis of immune cell infiltration and gene expression in periodontitis and control samples. \u003c/strong\u003e(A) Heatmap of immune cell infiltration. (B) Comparison of immune cell infiltration between periodontitis and control groups. (C) Boxplot of cell composition. (D) Correlation matrix of immune cell infiltration. (E) Heatmap of gene expression in immune cell types. (F) Heatmap of gene expression in periodontitis and control samples.\u003c/p\u003e","description":"","filename":"Fig6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5657155/v1/a6d1ee83350b602da58d2e81.jpg"},{"id":82044118,"identity":"cc4c78f1-8e3d-446e-b1ad-02b59b812948","added_by":"auto","created_at":"2025-05-06 09:28:49","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":1121508,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAnalysis of cellular subtypes and gene expression patterns in different conditions.\u003c/strong\u003e(A) t-SNE plot of cellular subtypes. (B) t-SNE plot of group distribution. (C) Violin plots of gene expression. (D) Dot plot of differentially expressed genes (DEGs).\u003c/p\u003e","description":"","filename":"Fig7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5657155/v1/67396465332271c8c780ec1a.jpg"},{"id":82045576,"identity":"88ece589-52cf-43c9-9f82-3b706bb7cc17","added_by":"auto","created_at":"2025-05-06 09:36:50","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":591938,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eModel fitting, variable selection, and diagnostic performance metrics.\u003c/strong\u003e(A) Cross-validation curve for LASSO.(B) Cross-validation curve for ridge regression. (C) Mean decrease in Gini for variable selection. (D) Error rate for random forest model. (E) Venn diagram of selected genes. (F) ROC curve for SIGIRR. (G) ROC curve for PLCL2. (H) ROC curve for GLRX. (I) Nomogram for risk prediction.\u003c/p\u003e","description":"","filename":"Fig8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5657155/v1/1cc4a879bc1ef6221c1a4274.jpg"},{"id":92944800,"identity":"6e1cd6c3-82f9-4e90-8128-31a10f543c19","added_by":"auto","created_at":"2025-10-07 12:29:36","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":7698971,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5657155/v1/8d08c32a-7aba-45c2-96bc-a57e2b13b092.pdf"},{"id":82045583,"identity":"e0e21f3b-126d-410c-9321-4bb02d601b63","added_by":"auto","created_at":"2025-05-06 09:36:50","extension":"zip","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":12993572,"visible":true,"origin":"","legend":"","description":"","filename":"GSE198710.zip","url":"https://assets-eu.researchsquare.com/files/rs-5657155/v1/12d29b36e6efdb4364328150.zip"},{"id":82044139,"identity":"fb3ef09b-b9f6-412d-8db4-29d4d8a00423","added_by":"auto","created_at":"2025-05-06 09:28:53","extension":"zip","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":50486234,"visible":true,"origin":"","legend":"","description":"","filename":"GSE106090.zip","url":"https://assets-eu.researchsquare.com/files/rs-5657155/v1/d27e22a965c34f35d0578af4.zip"},{"id":82044241,"identity":"ce5a99e5-9612-4830-a647-a6beafb03672","added_by":"auto","created_at":"2025-05-06 09:29:28","extension":"zip","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":538356414,"visible":true,"origin":"","legend":"","description":"","filename":"GSE225948.zip","url":"https://assets-eu.researchsquare.com/files/rs-5657155/v1/21dae400316b49130b22068f.zip"},{"id":82044116,"identity":"6bd0e481-d78c-4ded-9802-9fa4346623eb","added_by":"auto","created_at":"2025-05-06 09:28:49","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":173066,"visible":true,"origin":"","legend":"","description":"","filename":"KEGGPERMISSION251070.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5657155/v1/d17f72005bd6358105c8ff7f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Identification of key genes for secondary stroke in patients with periodontitis","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eAccording to the World Health Organization, stroke is the second leading cause of death globally and one of the main causes of adult acquired disability(1). Approximately 15\u0026nbsp;million people suffer from a stroke each year, with one-third of these patients dying and another one-third experiencing permanent disability(2).\u003c/p\u003e \u003cp\u003eStrokes are generally classified into two major types: ischemic stroke and hemorrhagic stroke(3). Ischemic stroke, caused by vascular obstruction, accounts for approximately 85% of all stroke cases(4);hemorrhagic stroke, which includes intracerebral hemorrhage and subarachnoid hemorrhage, is caused by vascular rupture(5). The pathogenic factors and mechanisms of stroke are complex, involving multiple biological processes such as vascular occlusion(6), neuronal damage and cell death(6), inflammatory response(7) and hemodynamic changes(8). Currently, there are no specific therapeutic drugs for ischemic stroke, so prevention is crucial(9). In recent years, it has been recognized that chronic infections, inflammation, and immune system abnormalities are related to the risk of ischemic stroke. Periodontitis, as a chronic inflammatory disease, has a significant impact on the occurrence and development of ischemic stroke(10).\u003c/p\u003e \u003cp\u003ePeriodontitis(11) is a common chronic inflammatory disease in the oral cavity, typically involving the destruction of the supporting structures of the teeth, with a relatively high prevalence among adults worldwide. According to global health studies, about half of adults suffer from mild to moderate periodontitis, while the prevalence of severe periodontitis is approximately 10\u0026ndash;15%(12). Periodontitis not only leads to oral health problems but is also associated with various systemic diseases, such as cardiovascular disease, diabetes, respiratory diseases, and preterm birth(13).\u003c/p\u003e \u003cp\u003eStroke and periodontitis have traditionally been regarded as two independent diseases in conventional medical literature, with the former primarily affecting the nervous system and the latter impacting oral health. However, increasing evidence in recent years suggests a possible biological and pathological link between these seemingly unrelated conditions(14). According to the World Health Organization(15), According to the World Health Organization, stroke has an extremely high incidence rate globally and is one of the leading causes of disability and death among adults. At the same time, periodontitis is also a widespread chronic inflammatory disease, affecting the oral health of a large number of adults worldwide(16).Research indicates that the association between periodontitis and stroke is mainly established through inflammatory and immune response mechanisms(14). Periodontitis can cause persistent local and systemic inflammatory responses, which can affect cardiovascular health through various pathways. For example, many researchers believe that chronic low-grade inflammation is a key biological mechanism linking periodontitis and stroke. Studies have found that individuals with periodontitis have significantly elevated levels of inflammatory markers in their blood, such as C-reactive protein and interleukin-6, which are closely related to the occurrence of stroke(17). Additionally, oral pathogens like \u003cem\u003ePorphyromonas gingivalis\u003c/em\u003e and \u003cem\u003eActinomyces\u003c/em\u003e have been detected in carotid plaques of stroke patients, suggesting that periodontal pathogens may directly participate in the formation of arterial plaques(18).\u003c/p\u003e \u003cp\u003eCurrently, there is no definitive research indicating a direct and clear relationship between periodontitis and stroke. This study employs bioinformatics methods to screen for biomarkers related to immune infiltration in both conditions. This can help identify potential diagnostic markers associated with immunity in patients with concurrent stroke and periodontitis, providing a basis for clinical diagnosis and research.\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Microarray Data\u003c/h2\u003e \u003cp\u003eThe datasets used in this study were obtained from the NCBI Gene Expression Omnibus (GEO) database(23), The periodontitis dataset is GSE106090(24);and the stroke dataset is GSE198710(25). Figure\u0026nbsp;1 depicts the study flowchart. GSE106090: Offers extensive gene expression data on periodontitis, covering various samples of periodontitis and healthy controls, making it ideal for exploring the molecular mechanisms of periodontitis. GSE198710: Contains gene expression data from stroke patients, providing a comparison between stroke and non-stroke samples, which helps in uncovering the molecular characteristics of stroke.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Data Processing and Differential Gene Screening\u003c/h2\u003e \u003cp\u003eDifferential analysis of protein expression data from different treatments was performed using the R package \"limma,\" with a significance threshold of P\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and an absolute log fold change of \u0026ge;\u0026thinsp;0.5. In LIMMA analysis, P-values and adjusted P-values (FDR) were calculated, with DEGs identified using thresholds like P\u0026thinsp;\u0026lt;\u0026thinsp;0.05. This controls the false discovery rate and ensures reliability. Volcano plots for differential analysis were created using the R package \"ggplot2,\" and heatmaps were generated using the R package \"pheatmap.\" Intersections of different groups and Venn diagrams were created using the online tool Evenn (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.ehbio.com/test/venn\u003c/span\u003e\u003cspan address=\"http://www.ehbio.com/test/venn\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). These tools were used to create volcano plots and heatmaps for visualizing DEG results, offering an intuitive view of data distribution and patterns.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Weighted Gene Co-expression Network Analysis\u003c/h2\u003e \u003cp\u003eWGCNA(26) was used to reveal expression patterns among proteins and construct co-expression networks to identify modules of protein sets related to biological phenotypes. Initial clustering analysis showed small intra-group differences and large inter-group differences among samples. The optimal soft threshold of 20 was determined by comparing scale independence and mean connectivity, ensuring the constructed network conformed more closely to a scale-free topology. Proteins with different expression patterns were clustered into different color modules, with correlation analysis indicating significantly weaker correlations between different expression modules compared to those within the same module.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Functional Enrichment Analysis\u003c/h2\u003e \u003cp\u003eEnrichment analysis of intersecting genes from the stroke, periodontitis, and periodontitis WGCNA datasets was conducted using the Enrichr database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://maayanlab.cloud/Enrichr/\u003c/span\u003e\u003cspan address=\"https://maayanlab.cloud/Enrichr/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), focusing on pathways, biological processes (BP), cellular components (CC), and molecular functions (MF). Pathway annotation was performed using the Enrichr tool, which provides a comprehensive and user-friendly platform for gene set enrichment analysis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Immune Infiltration Analysis\u003c/h2\u003e \u003cp\u003eThe CIBERSORT algorithm(27) was used to determine the proportion of immune cells in tissues or cells(28). Bar plots displayed the proportions of various immune cell types in different samples. Correlation heatmaps of immune cells were generated using the \"corrplot\"(29) R package. Heatmaps created with pheatmap and ComplexHeatmap visualized patterns and relationships in the data. Stacked proportion plots illustrated the proportions of different cell types in each sample. Spearman correlations between datasets (e.g., gene expression data and CIBERSORT results) were calculated using cor.test.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Single-Cell Sequencing\u003c/h2\u003e \u003cp\u003eSingle-cell sequencing (scRNA-seq) was used to study expression levels within individual cells, revealing heterogeneity among cell types. This study analyzed single-cell sequencing data from the public database GSE225948(30), Standard procedures, including quality control, genome alignment, and transcript quantification, were applied to the raw sequencing data. Transcript count data were normalized to eliminate systematic biases between samples. The t-SNE (t-Distributed Stochastic Neighbor Embedding) algorithm was used for dimensionality reduction, visualizing complex high-dimensional data in two dimensions. Cells or gene expression features were processed with t-SNE, generating two-dimensional coordinates. Points were clustered based on t-SNE results to identify different cell populations or expression patterns. Differences in t-SNE distributions under different experimental conditions (e.g., normal control group vs. experimental group) were compared to explore the impact of experimental treatments on cell or gene expression. Statistical tests such as ANOVA or t-test compared expression level differences in different cell types or genes under various experimental conditions. Scatter plots were generated using graphical software (R's ggplot2), where each point represented a cell or gene, with colors and shapes indicating different cell types or experimental conditions. GSE225948: Provides single-cell RNA sequencing data, allowing for an in-depth investigation of cell heterogeneity and cell type-specific gene expression patterns.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e2.7 Machine Learning\u003c/h2\u003e \u003cp\u003eMachine learning algorithms identified core genes for diagnosing periodontitis. \"glmnet\"(31) was used for LASSO(32) regression, \u0026ldquo;random Forest\u0026rdquo;(33) was used for RF(34) analysis. The intersecting genes from these methods were identified as core genes for diagnosing stroke combined with periodontitis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e2.8 Nomogram Construction and Receiver Operating Characteristic Evaluation\u003c/h2\u003e \u003cp\u003eThe \"rms\"(35) R package was used to create a nomogram for the screened genes, with clinical value evaluated by the area under the curve (AUC) and 95% confidence intervals (CI) using ROC analysis. AUC values over 0.7 indicated diagnostic value.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e2.9 Statistical Analysis\u003c/h2\u003e \u003cp\u003eHigh-throughput sequencing data analysis in this study utilized the R package \"limma\" to evaluate differential expression using linear models, with empirical Bayesian methods enhancing statistical inference robustness. For single-cell data, the R package \"Seurat\" used z-score transformations to compare expression levels of different genes, employing Wilcoxon rank-sum tests to identify marker genes distinguishing different cell populations.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Differentially Expressed Genes\u003c/h2\u003e \u003cp\u003eLog-transformed data from the periodontitis and stroke datasets showed a normal distribution, followed by quantile normalization for comparability between standardized data groups (Figs.\u0026nbsp;2A, 2B, 4A, 4B). Principal component analysis indicated significant differences between groups under different treatments (Figs.\u0026nbsp;2C, 4C). LIMMA analysis identified 4113 upregulated and 4340 downregulated genes in the integrated periodontitis dataset, and 4369 upregulated and 5016 downregulated genes in the stroke dataset (Figs.\u0026nbsp;4A,4B). Figures\u0026nbsp;2D, 2E, and 4D, 4E display volcano plots and heatmaps generated from these data.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Key Module Selection\u003c/h2\u003e \u003cp\u003eWGCNA identified key modules in periodontitis. Initial clustering analysis indicated small intra-group differences and large inter-group differences (Fig.\u0026nbsp;3A). The optimal soft threshold of 20 ensured the constructed network conformed more closely to a scale-free topology (Fig.\u0026nbsp;3B). Proteins with different expression patterns were clustered into various color modules (Fig.\u0026nbsp;3C). The brown module, containing 691 genes, showed the highest correlation with periodontitis (cor\u0026thinsp;=\u0026thinsp;0.99, P\u0026thinsp;=\u0026thinsp;4e-10), making it a key module for subsequent analysis (Fig.\u0026nbsp;3D). Further exploration of the relationship between key modules and specific experimental conditions validated the correlation of module member proteins with traits, all of which were statistically significant (Fig.\u0026nbsp;3E). Correlation analysis indicated significantly weaker correlations between different expression modules compared to those within the same module (Fig.\u0026nbsp;3F), further supporting the decisive role of proteins closely linked to experimental conditions within their modules.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Enrichment Analysis for Stroke Combined with Periodontitis\u003c/h2\u003e \u003cp\u003eThe intersection of 4113 upregulated periodontitis genes, 4369 upregulated stroke genes, and 691 brown module genes resulted in 169 genes (Fig.\u0026nbsp;4F). KEGG analysis showed enrichment mainly in \"Viral protein interaction with cytokine and cytokine receptor\" and \"Cytokine-cytokine receptor interaction\" (Figs.\u0026nbsp;5A, 5B). GO analysis revealed that BP terms were mainly enriched in \"Regulation of Transcription by RNA Polymerase II\" and \"Regulation of Cell Population Proliferation\" (Figs.\u0026nbsp;5C, 5D). CC terms were primarily found in \"Endoplasmic Reticulum Membrane,\" \"Golgi Membrane,\" and \"Late Endosome Lumen\" (Figs.\u0026nbsp;5E, 5F). MF analysis showed enrichment in \"Sequence-Specific DNA Binding,\" \"RNA Polymerase II Cis-Regulatory Region Sequence-Specific DNA Binding,\" and \"RNA Binding\" (Figs.\u0026nbsp;5G, 5H).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Immune Infiltration Analysis\u003c/h2\u003e \u003cp\u003eWe found that genes associated with stroke also played roles in periodontitis, primarily in immune regulation. Therefore, we conducted immune analysis on the 169 intersecting genes. Figure\u0026nbsp;6A shows the distribution of various immune cell types in different samples. Proportions of cell types, such as memory B cells, plasma cells, activated and resting T cells, NK cells, various macrophages, and dendritic cells, were detailed to compare differences between healthy controls and pathological samples. Figure\u0026nbsp;6B's heatmap used color to represent relative expression levels or activity of immune cell types in different samples. Figure\u0026nbsp;6C's boxplot compared the composition of different immune cell types in two sample groups (H group and P group), highlighting significant changes in T cells CD4 memory resting and Plasma cells as potential therapeutic targets. Figure\u0026nbsp;6D's correlation matrix displayed correlations between different immune cell types using color and numerical values. Higher positive correlations included NK cells and CD4 memory T cells (correlation coefficient\u0026thinsp;~\u0026thinsp;0.72), monocytes and plasma cells (~\u0026thinsp;0.83), and memory B cells and dendritic cells (~\u0026thinsp;0.79). Higher negative correlations included monocytes and NK cells (~\u0026thinsp;0.80) and monocytes and T cells (activated/resting) (~\u0026thinsp;0.74). Figure\u0026nbsp;6E displayed correlations between various immune cell types and different biomarkers (mainly genes). Genes were represented in columns, while immune cell types were in rows. The heatmap used color and symbols to indicate correlation strength and statistical significance, showing T cells (activated/resting) and Plasma cells as significantly positively correlated, while resting dendritic cells were negatively correlated, suggesting T cells (activated/resting) and Plasma cells as potential therapeutic nodes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Single-Cell Sequencing\u003c/h2\u003e \u003cp\u003eThis study analyzed transcriptomic single-cell sequencing data from brain tissue in a mouse ischemia-reperfusion model. After outlier removal and normalization of single-cell sequencing data, cell types in t-SNE clusters were annotated (Fig.\u0026nbsp;7A). The study observed changes in upregulated gene levels in Bc4 and Tc2 in stroke and periodontitis. Figure\u0026nbsp;7B displayed cell distribution under different experimental conditions (e.g., mcao and sham), with regions highlighting two main cell populations, Bc4 (Plasma cells) and Tc2 (T cells CD4 memory activate). Figure\u0026nbsp;7C showed expression levels of different genes (e.g., Cd4, Igtbp4, Cd19, Cd38) in cells, with Cd19 and Cd38 as Plasma cell markers and Cd4 and Igtbp4 as T cells (activated/resting) markers. Color intensity represented gene expression levels, with dark colors indicating high expression and light colors indicating low expression. Figure\u0026nbsp;7D's Dot Plot visualized gene expression across different cell types, identifying genes highly expressed in Plasma cells and T cells (activated/resting).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e3.6 Machine Learning\u003c/h2\u003e \u003cp\u003eCore genes were identified using machine learning. Lasso regression and RF algorithms identified core genes and created related nomograms for ROC analysis. Lasso regression selected 9 candidate genes (Figs.\u0026nbsp;8A, 8B), and RF identified 10 important genes (Figs.\u0026nbsp;8C, 8D). The intersection of the two methods (Fig.\u0026nbsp;8E) ultimately identified three genes: SIGIRR, PLCL2, and GLRX.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e3.7 Determining Diagnostic Value\u003c/h2\u003e \u003cp\u003eROC curves were plotted for the three candidate genes (Figs.\u0026nbsp;8F, 8G, 8H), and a nomogram was created (Fig.\u0026nbsp;8I) to evaluate the diagnostic value of each gene. Calculated AUCs and 95% confidence intervals were as follows: SIGIRR (AUC: 0.830, CI: 0.728\u0026ndash;0.938), PLCL2 (AUC: 0.733, CI: 0.602\u0026ndash;0.851), GLRX (AUC: 0.801, CI: 0.683\u0026ndash;0.908). The results indicated high diagnostic value for the identified genes in diagnosing stroke combined with periodontitis.\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eWe must recognize that stroke is not only one of the leading causes of death globally but also a major cause of long-term disability in adults. According to the World Health Organization, approximately 15\u0026nbsp;million people suffer from strokes each year, with one-third of these patients eventually dying, and another one-third experiencing severe long-term disabilities(36). These disabilities include language disorders, motor function impairments, and cognitive deficits(37), significantly impacting patients' quality of life and imposing substantial burdens on families and society.\u003c/p\u003e \u003cp\u003eRecent studies have identified several biomarkers for stroke, including glutamate(38), glutamate oxaloacetate transaminase (GOT)(37),S100B(39) and the receptor for advanced glycation end products (RAGE)(39). However, research combining these two diseases is relatively rare. There are currently no biomarkers for diagnosing stroke using nomograms and machine learning methods. Therefore, we integrated bioinformatics analysis, single-cell sequencing, and machine learning, using nomograms and ROC curves to evaluate their diagnostic value. Notably, we identified three key immune-related candidate genes (SIGIRR, PLCL2, and GLRX) and developed a nomogram for diagnosing stroke in periodontitis patients. By simply detecting peripheral blood in periodontitis patients(40), we can estimate the probability of stroke secondary to periodontitis.\u003c/p\u003e \u003cp\u003eGiven that the stroke and periodontitis samples used in this study were derived from peripheral blood, patients exhibited known core gene expressions. Peripheral blood testing is widely used for diagnosing many diseases due to its simplicity and efficacy. Our next step is to develop a more refined model that accurately reflects gene expression and assigns values, enhancing diagnostic accuracy. Early monitoring and intervention for significant changes in target indicators in periodontitis patients would be valuable for diagnosing stroke combined with periodontitis.\u003c/p\u003e \u003cp\u003e \u003cb\u003eSIGIRR (Single Ig IL-1R-related receptor)\u003c/b\u003e is a negative regulatory immune receptor mainly expressed in epithelial cells and some immune cells(41). It reduces inflammatory responses by inhibiting Toll-like receptor (TLR) and IL-1 receptor family signaling. SIGIRR helps maintain immune balance and prevent excessive inflammation by forming heterodimers with other TIR domain proteins, thus reducing cell surface receptor signaling. As a negative regulator of IL-1 and TLR signaling pathways(42),SIGIRR can inhibit their activation, reducing inflammation caused by these pathways. In stroke contexts, this anti-inflammatory effect may help mitigate secondary brain damage, as post-stroke inflammation often exacerbates brain injury. SIGIRR also plays an important role in periodontitis. As a chronic inflammatory disease involving the destruction of periodontal tissues, periodontitis can lead to significant oral health issues. By inhibiting inflammatory responses, SIGIRR may help reduce inflammation and damage to periodontal tissues. Studies have shown that patients with periodontitis often have lower levels of SIGIRR expression, which may contribute to heightened inflammatory responses. Thus, SIGIRR's anti-inflammatory effects could be valuable in the treatment of periodontitis.\u003c/p\u003e \u003cp\u003e \u003cb\u003ePLCL2 (Phospholipase C-like 2)\u003c/b\u003e is a phosphatidylinositol-specific phosphatase crucial for regulating lipid metabolism. Dysregulated lipid metabolism is a primary cause of atherosclerosis. High levels of low-density lipoprotein (LDL) and low levels of high-density lipoprotein (HDL) are key factors in atherosclerosis(43). Oxidized LDL (ox-LDL) is ingested by macrophages, forming foam cells and contributing to arterial plaque formation and progression(44). Atherosclerosis results from lipid accumulation in arterial walls, triggering chronic inflammation. PLCL2 influences foam cell formation and arterial plaque development by regulating genes related to lipid metabolism. Studies show PLCL2 is significantly associated with high expression of lipid metabolism-related genes, including CD36, LPL, and PPARG, which play critical roles in atherosclerosis pathology. Atherosclerosis is a major cause of ischemic stroke(45, 46). Arterial plaque formation and rupture can interrupt blood supply to the brain, causing strokes. PLCL2 indirectly affects stroke risk by regulating lipid metabolism and atherosclerosis. Inhibiting PLCL2 expression or function may reduce ox-LDL formation and foam cell accumulation, lowering atherosclerosis and stroke risk(45, 47). In periodontitis, PLCL2 may also play a significant role. Patients with periodontitis often have an increased risk of atherosclerosis, suggesting shared pathological mechanisms between the two conditions. By modulating lipid metabolism, PLCL2 may serve as a bridge between periodontitis and atherosclerosis. Research suggests that PLCL2 expression levels in patients with periodontitis may influence lipid metabolism, thereby exacerbating the progression of atherosclerosis. Therefore, PLCL2 could be a potential target for the prevention of both periodontitis and stroke.\u003c/p\u003e \u003cp\u003e \u003cb\u003eGLRX (Glutaredoxin)\u003c/b\u003e is a key antioxidant enzyme in the thioredoxin family, primarily maintaining redox balance in cells. GLRX controls intracellular free radical levels by catalyzing reactions between thiol groups in proteins and glutathione (GSH), protecting cells from oxidative damage(48). In stroke, GLRX's antioxidant function is crucial for protecting brain cells from free radical damage due to increased oxidative stress(48). Oxidative stress exacerbates brain injury in stroke, so GLRX may help mitigate this damage through its antioxidant mechanism. In periodontitis, GLRX's antioxidant function is equally important. Patients with periodontitis often experience oxidative stress, which can lead to damage of periodontal tissues. Through its antioxidant effects, GLRX can effectively clear free radicals and reduce oxidative stress-induced damage to periodontal tissues. Studies have shown that GLRX expression levels are often lower in patients with periodontitis, which may exacerbate oxidative stress. Therefore, GLRX's antioxidant functions could be valuable in the treatment of periodontitis. Enhancing GLRX expression may help reduce oxidative damage to periodontal tissues and promote their repair and regeneration.\u003c/p\u003e \u003cp\u003eImmune response, inflammatory response, and oxidative stress response are present in all stages of stroke and significantly impact disease progression. Stroke leads to neuron death and inflammation, triggering immune cell migration from peripheral tissues to the brain, causing secondary damage. These immune and inflammatory pathways are potential stroke treatment targets. For instance, platelet activation and aggregation in the acute phase of stroke promote neurodamage progression by enhancing inflammation and thrombosis. Moreover, reperfusion therapy may help restore blood flow and reduce brain tissue damage, but the immune system may maintain inflammation post-reperfusion.\u003c/p\u003e \u003cp\u003eResearch shows post-stroke immune responses involve brain tissue infiltration by neutrophils, macrophages, B cells, T cells, and dendritic cells. These cells activate downstream signaling pathways through pattern recognition receptors like TLRs and NLRs, producing pro-inflammatory cytokines, reactive oxygen species, and chemokines, exacerbating post-ischemia inflammation. Immune response regulation plays a crucial role in improving stroke outcomes. Appropriate immunomodulatory treatments, such as immunomodulators, can effectively reduce brain damage, lesion size, hemorrhage, and improve neurological recovery. Understanding these mechanisms is essential for developing new treatment strategies and enhancing stroke management effectiveness.\u003c/p\u003e"},{"header":"5. Limitations","content":"\u003cp\u003eThe current study has several limitations that should be acknowledged. Firstly, the sample size was relatively small, which may limit the statistical power and generalizability of the findings. The datasets used in this study were sourced from public databases, which, while valuable resources, may have inherent limitations such as variability in data collection protocols, potential batch effects, and missing clinical information for some samples. These factors could introduce noise or bias into the analysis, affecting the reliability of the results. Additionally, given the complexity of both stroke and periodontitis, involving multiple genetic, environmental, and lifestyle factors, it is possible that not all relevant genes and biological processes were captured in this study. The analysis focused on specific datasets and methodologies, which might have excluded other important genes or interactions that could play a role in the pathogenesis of these diseases. Furthermore, the cross-sectional nature of the data limits the ability to establish causal relationships and assess the temporal dynamics of gene expression changes in relation to disease progression.\u003c/p\u003e"},{"header":"6. Future Directions","content":"\u003cp\u003eFuture research in this area should aim to address the limitations identified in this study. One critical step is to expand the sample size by including more diverse patient cohorts from different geographic regions, ethnic backgrounds, and clinical settings. This would enhance the representativeness of the data and improve the robustness of the findings. Additionally, incorporating multi-omics data integration could provide a more comprehensive understanding of the molecular mechanisms underlying stroke and periodontitis. For instance, combining transcriptomics with proteomics and metabolomics data would allow for the identification of not only differentially expressed genes but also the corresponding proteins and metabolites, offering insights into the functional implications of these changes at various biological levels. Furthermore, functional studies in vitro and in vivo are necessary to validate the roles of the identified genes (e.g., SIGIRR, PLCL2, GLRX) in disease pathogenesis. This could involve experiments such as gene knockdown or overexpression studies to assess their impact on relevant cellular processes and disease models. Moreover, exploring the therapeutic potential of these genes through drug target validation and preclinical studies could pave the way for the development of novel treatment strategies. Finally, leveraging advanced computational models and machine learning algorithms with improved interpretability could help in uncovering more nuanced patterns in the data and identifying additional biomarkers or therapeutic targets that may have been overlooked in this study.\u003c/p\u003e"},{"header":"7. Conclusion","content":"\u003cp\u003eThrough bioinformatics analysis and machine learning, we systematically identified three relevant candidate genes (SIGIRR, PLCL2, GLRX) and provided a template for diagnosing periodontitis combined with stroke. We also noted immune system imbalances in stroke patients with periodontitis, where immune cell percentages are influenced by the immune microenvironment. The screened genes can be used for clinical diagnosis and treatment.\u003c/p\u003e "},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eGorelick PB, Whelton PK, Sorond F, Carey RM. Blood Pressure Management in Stroke. Hypertension. 2020;76(6):1688-95.\u003c/li\u003e\n\u003cli\u003eKainz A, Meisinger C, Linseisen J, Kirchberger I, Zickler P, Naumann M, et al. Changes of Health-Related Quality of Life Within the 1st Year After Stroke\u0026ndash;Results From a Prospective Stroke Cohort Study. Frontiers in Neurology. 2021;12.\u003c/li\u003e\n\u003cli\u003eHasan TF, Hasan H, Kelley RE. Overview of Acute Ischemic Stroke Evaluation and Management. Biomedicines. 2021;9(10):1486-.\u003c/li\u003e\n\u003cli\u003ePotter TBH, Tannous J, Vahidy FS. A Contemporary Review of Epidemiology, Risk Factors, Etiology, and Outcomes of Premature Stroke. Current Atherosclerosis Reports. 2022;24(12):939-48.\u003c/li\u003e\n\u003cli\u003eChen S, Zeng L, Hu Z. Progressing haemorrhagic stroke: categories, causes, mechanisms and managements. Journal of Neurology. 2014;261(11):2061-78.\u003c/li\u003e\n\u003cli\u003eSalaudeen MA, Bello N, Danraka RN, Ammani ML. Understanding the Pathophysiology of Ischemic Stroke: The Basis of Current Therapies and Opportunity for New Ones. Biomolecules. 2024;14(3).\u003c/li\u003e\n\u003cli\u003eChoi YH, Laaker C, Hsu M, Cismaru P, Sandor M, Fabry Z. Molecular Mechanisms of Neuroimmune Crosstalk in the Pathogenesis of Stroke. International Journal of Molecular Sciences. 2021;22(17).\u003c/li\u003e\n\u003cli\u003eAsil T, Utku U, Balci K, Uzunca I. Changing cerebral blood flow velocity by transcranial Doppler during head up tilt in patients with diabetes mellitus. Clinical neurology and neurosurgery. 2007;109(1):1-6.\u003c/li\u003e\n\u003cli\u003eZheng X, Li X, Zhen J, Xue D, Hu J, Cao Q, et al. Periodontitis is associated with stroke. Journal of Translational Medicine. 2023;21(1):697.\u003c/li\u003e\n\u003cli\u003eSharma A, Sharma A, Chauhan R, Chauhan AS, Parmar M, Thakur S. The Relationship Between Periodontal Disease (PD) and Recurrent Vascular Events in Ischemic Stroke/Transient Ischemic Attack (TIA) Patients: A Hospital-Based Cohort Study. Cureus. 2023;15(3):e36530.\u003c/li\u003e\n\u003cli\u003eK\u0026ouml;n\u0026ouml;nen E, Gursoy M, Gursoy U. Periodontitis: A Multifaceted Disease of Tooth-Supporting Tissues. Journal of Clinical Medicine. 2019;8(8).\u003c/li\u003e\n\u003cli\u003ePink C, Holtfreter B, V\u0026ouml;lzke H, Nauck M, D\u0026ouml;rr M, Kocher T. Periodontitis and systemic inflammation as independent and interacting risk factors for mortality: evidence from a prospective cohort study. BMC Medicine. 2023;21(1).\u003c/li\u003e\n\u003cli\u003eBhuyan R, Bhuyan SK, Mohanty JN, Das S, Juliana N, Juliana IF. Periodontitis and Its Inflammatory Changes Linked to Various Systemic Diseases: A Review of Its Underlying Mechanisms. Biomedicines. 2022;10(10).\u003c/li\u003e\n\u003cli\u003eBaniulyte G, Piela K, Culshaw S. How strong is the link between periodontitis and stroke? Evidence-Based Dentistry. 2021;22(1):10-1.\u003c/li\u003e\n\u003cli\u003eDu M, Mi D, Liu M, Liu J. Global trends and regional differences in disease burden of stroke among children: a trend analysis based on the global burden of disease study 2019. BMC Public Health. 2023;23(1):2120.\u003c/li\u003e\n\u003cli\u003ePink C, Holtfreter B, V\u0026ouml;lzke H, Nauck M, D\u0026ouml;rr M, Kocher T. Periodontitis and systemic inflammation as independent and interacting risk factors for mortality: evidence from a prospective cohort study. BMC Medicine. 2023;21(1):430.\u003c/li\u003e\n\u003cli\u003eFagundes NCF, Almeida APCPSC, Vilhena KFB, Magno MB, Maia LC, Lima RR. \u0026lt;p\u0026gt;Periodontitis As A Risk Factor For Stroke: A Systematic Review And Meta-Analysis\u0026lt;/p\u0026gt;. Vascular Health and Risk Management. 2019;Volume 15:519-32.\u003c/li\u003e\n\u003cli\u003eGuo J, Xu R, Liu R, Lai W, Hu C, He H, et al. Association between the systemic immune inflammation index and periodontitis: a cross-sectional study. Journal of Translational Medicine. 2024;22(1):96.\u003c/li\u003e\n\u003cli\u003ePanda AK, Jin Q, Teng F, Cheng Z. Association between common polymorphisms in IL-1 and TNF\u0026alpha; and risk of peri-implant disease: A meta-analysis. Plos One. 2021;16(10).\u003c/li\u003e\n\u003cli\u003eMart\u0026iacute;nez-Garc\u0026iacute;a M, Hern\u0026aacute;ndez-Lemus E. Periodontal Inflammation and Systemic Diseases: An Overview. Frontiers in Physiology. 2021;12.\u003c/li\u003e\n\u003cli\u003eAleksandrowicz P, Brzezińska-Błaszczyk E, Kozłowska E, Żelechowska P, Borgonovo AE, Agier J. Analysis of IL-1\u0026beta;, CXCL8, and TNF-\u0026alpha; levels in the crevicular fluid of patients with periodontitis or healthy implants. BMC Oral Health. 2021;21(1):120.\u003c/li\u003e\n\u003cli\u003eWang RP-H, Huang J, Chan KWY, Leung WK, Goto T, Ho Y-S, et al. IL-1\u0026beta; and TNF-\u0026alpha; play an important role in modulating the risk of periodontitis and Alzheimer\u0026rsquo;s disease. Journal of Neuroinflammation. 2023;20(1):71.\u003c/li\u003e\n\u003cli\u003eBarrett T, Wilhite SE, Ledoux P, Evangelista C, Kim IF, Tomashevsky M, et al. NCBI GEO: archive for functional genomics data sets--update. Nucleic acids research. 2013;41(Database issue):D991-5.\u003c/li\u003e\n\u003cli\u003eLiu Y, Liu Q, Li Z, Acharya A, Chen D, Chen Z, et al. Long non-coding RNA and mRNA expression profiles in peri-implantitis vs periodontitis. Journal of periodontal research. 2020;55(3):342-53.\u003c/li\u003e\n\u003cli\u003eJiang W, Li J, Cai Y, Liu W, Chen M, Xu X, et al. The Novel lncRNA ENST00000530525 Affects ANO1, Contributing to Blood-Brain Barrier Injury in Cultured hCMEC/D3 Cells Under OGD/R Conditions. Frontiers in genetics. 2022;13:873230.\u003c/li\u003e\n\u003cli\u003eLangfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC bioinformatics. 2008;9:559.\u003c/li\u003e\n\u003cli\u003eChen B, Khodadoust MS, Liu CL, Newman AM, Alizadeh AA. Profiling Tumor Infiltrating Immune Cells with CIBERSORT. Methods in molecular biology (Clifton, NJ). 2018;1711:243-59.\u003c/li\u003e\n\u003cli\u003eNewman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, et al. Robust enumeration of cell subsets from tissue expression profiles. Nature methods. 2015;12(5):453-7.\u003c/li\u003e\n\u003cli\u003eHu K. Become Competent within One Day in Generating Boxplots and Violin Plots for a Novice without Prior R Experience. Methods and protocols. 2020;3(4).\u003c/li\u003e\n\u003cli\u003eGarcia-Bonilla L, Shahanoor Z, Sciortino R, Nazarzoda O, Racchumi G, Iadecola C, et al. Analysis of brain and blood single-cell transcriptomics in acute and subacute phases after experimental stroke. Nature immunology. 2024;25(2):357-70.\u003c/li\u003e\n\u003cli\u003eYang C, Delcher C, Shenkman E, Ranka S. Machine learning approaches for predicting high cost high need patient expenditures in health care. Biomedical engineering online. 2018;17(Suppl 1):131.\u003c/li\u003e\n\u003cli\u003eEllis K, Kerr J, Godbole S, Lanckriet G, Wing D, Marshall S. A random forest classifier for the prediction of energy expenditure and type of physical activity from wrist and hip accelerometers. Physiological measurement. 2014;35(11):2191-203.\u003c/li\u003e\n\u003cli\u003eZhang M, Zhu K, Pu H, Wang Z, Zhao H, Zhang J, et al. An Immune-Related Signature Predicts Survival in Patients With Lung Adenocarcinoma. Frontiers in oncology. 2019;9:1314.\u003c/li\u003e\n\u003cli\u003eAlderden J, Pepper GA, Wilson A, Whitney JD, Richardson S, Butcher R, et al. Predicting Pressure Injury in Critical Care Patients: A Machine-Learning Model. American journal of critical care : an official publication, American Association of Critical-Care Nurses. 2018;27(6):461-8.\u003c/li\u003e\n\u003cli\u003ePan X, Jin X, Wang J, Hu Q, Dai B. Placenta inflammation is closely associated with gestational diabetes mellitus. American journal of translational research. 2021;13(5):4068-79.\u003c/li\u003e\n\u003cli\u003eShahi S, Farhoudi M, Dizaj SM, Sharifi S, Sadigh-Eteghad S, Goh KW, et al. The Link between Stroke Risk and Orodental Status\u0026mdash;A Comprehensive Review. Journal of Clinical Medicine. 2022;11(19).\u003c/li\u003e\n\u003cli\u003eHansen PR, Holmstrup P. Cardiovascular Diseases and Periodontitis. In: Santi-Rocca J, editor. Periodontitis: Advances in Experimental Research. Cham: Springer International Publishing; 2022. p. 261-80.\u003c/li\u003e\n\u003cli\u003eNaderi S, Merchant AT. The Association Between Periodontitis and Cardiovascular Disease: an Update. Current Atherosclerosis Reports. 2020;22(10):52.\u003c/li\u003e\n\u003cli\u003eShahi S, Farhoudi M, Dizaj SM, Sharifi S, Sadigh-Eteghad S, Goh KW, et al. The Link between Stroke Risk and Orodental Status\u0026mdash;A Comprehensive Review. 2022;11(19):5854.\u003c/li\u003e\n\u003cli\u003eWang H, Zhang S, Xie L, Zhong Z, Yan F. Neuroinflammation and peripheral immunity: Focus on ischemic stroke. International immunopharmacology. 2023;120:110332.\u003c/li\u003e\n\u003cli\u003eRiva F, Bonavita E, Barbati E, Muzio M, Mantovani A, Garlanda C. TIR8/SIGIRR is an Interleukin-1 Receptor/Toll Like Receptor Family Member with Regulatory Functions in Inflammation and Immunity. 2012;3.\u003c/li\u003e\n\u003cli\u003eLiu Z, Zhu L, Lu Z, Chen H, Fan L, Xue Q, et al. IL-37 Represses the Autoimmunity in Myasthenia Gravis via Directly Targeting Follicular Th and B Cells. Journal of immunology (Baltimore, Md : 1950). 2020;204(7):1736-45.\u003c/li\u003e\n\u003cli\u003ePan X, Liu J, Zhong L, Zhang Y, Liu C, Gao J, et al. Identification of lipid metabolism-related biomarkers for diagnosis and molecular classification of atherosclerosis. Lipids in Health and Disease. 2023;22(1):96.\u003c/li\u003e\n\u003cli\u003eGaggini M, Gorini F, Vassalle C. Lipids in Atherosclerosis: Pathophysiology and the Role of Calculated Lipid Indices in Assessing Cardiovascular Risk in Patients with Hyperlipidemia. 2023;24(1):75.\u003c/li\u003e\n\u003cli\u003eChow Y-L, Teh Kuan L, Chyi Huey L, Lim Fang L, Yee Chu C, Wei Keat L. Lipid Metabolism Genes in Stroke Pathogenesis: The Atherosclerosis. Current Pharmaceutical Design. 2020;26(34):4261-71.\u003c/li\u003e\n\u003cli\u003eDuan H, Song P, Li R, Su H, He L. Attenuating lipid metabolism in atherosclerosis: The potential role of Anti-oxidative effects on low-density lipoprotein of herbal medicines. 2023;14.\u003c/li\u003e\n\u003cli\u003eKloska A, Malinowska M, Gabig-Cimińska M, Jak\u0026oacute;bkiewicz-Banecka J. Lipids and Lipid Mediators Associated with the Risk and Pathology of Ischemic Stroke. 2020;21(10):3618.\u003c/li\u003e\n\u003cli\u003eBurns M, Rizvi SHM, Tsukahara Y, Pimentel DR, Luptak I, Hamburg NM, et al. Role of Glutaredoxin-1 and Glutathionylation in Cardiovascular Diseases. 2020;21(18):6803.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Differentially Expressed Genes, Periodontitis, Stroke, Single-Cell Sequencing, Machine Learning","lastPublishedDoi":"10.21203/rs.3.rs-5657155/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5657155/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground:\u003c/strong\u003e The immune system plays a particularly important role in the pathogenesis of both stroke and periodontitis. The aim of this study is to identify key diagnostic candidate genes for stroke in patients with periodontitis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e We searched for a periodontitis dataset and a stroke dataset in the Gene Expression Omnibus (GEO) database. Limma, Weighted Gene Co-expression Network Analysis (WGCNA), immune analysis, single-cell sequencing, and machine learning algorithms were used to identify and analyze immune-related genes associated with periodontitis and stroke. Finally, a nomogram and receiver operating characteristic (ROC) curves were used for evaluation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e Limma analysis produced 4252 differentially expressed genes (DEGs) from the integrated stroke dataset and 4113 DEGs from the integrated periodontitis dataset. The periodontitis dataset generated the most relevant module after WGCNA, containing 2028 genes, with a total of 169 intersecting genes among the three datasets. Enrichment analysis revealed that most immune-related genes were enriched, and immune infiltration analysis showed an imbalance of various immune cells. Single-cell sequencing was used to screen immune cells. Further machine learning screening produced three core genes, which were evaluated for their diagnostic value using a nomogram and ROC curves, showing high diagnostic value.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion:\u003c/strong\u003e Three immune-related core genes (SIGIRR, PLCL2, GLRX) were identified, and a nomogram for diagnosing periodontitis and stroke was established. This study may help identify potential peripheral blood diagnostic candidate genes for secondary stroke in patients with periodontitis.\u003c/p\u003e","manuscriptTitle":"Identification of key genes for secondary stroke in patients with periodontitis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-06 09:28:45","doi":"10.21203/rs.3.rs-5657155/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"2ac9ba8a-c3c6-4aed-bb65-24626b34cf40","owner":[],"postedDate":"May 6th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":47855754,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":47855755,"name":"Biological sciences/Computational biology and bioinformatics/Data mining"},{"id":47855756,"name":"Biological sciences/Computational biology and bioinformatics/Data processing"},{"id":47855757,"name":"Biological sciences/Computational biology and bioinformatics/Databases"}],"tags":[],"updatedAt":"2025-10-03T05:08:47+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-06 09:28:45","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5657155","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5657155","identity":"rs-5657155","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}