Exploring the potential common genetic characteristics and molecular mechanisms between inflammatory bowel disease and atherosclerosis using bioinformatics analysis and machine learning

doi:10.21203/rs.3.rs-3116876/v1

Download PDF

Research Article

Exploring the potential common genetic characteristics and molecular mechanisms between inflammatory bowel disease and atherosclerosis using bioinformatics analysis and machine learning

https://doi.org/10.21203/rs.3.rs-3116876/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

The prevalence of inflammatory bowel disease (IBD) is increasing worldwide. According to recent research, IBD is a separate risk element for atherosclerosis (AS),however the cause of IBD combined with AS is still not clear. Through a thorough bioinformatics analysis, this study investigated the potential mechanisms of AS in conjunction with IBD and aimed to pinpoint biomarkers for patients with IBD and AS complications.

Methods

From two microarray datasets, we obtained differentially expressed genes(DEGs) for AS and IBD using the "Limma" package. The CDEGs underwent enrichment analysis, leading to the construction of a network for protein-protein interaction (PPI). Afterward, two algorithms based on machine learning were utilized to search for hub genes. A diagnostic nomogram was created using the Hub genes. To assess the dependability of the nomogram, the ROC curve was employed.qPCR was used to analyze the expression of hub genes in animal models. The AS dataset underwent immune infiltration analysis and consensus clustering analysis in the end.

Results

A total of 51 CDEGs were obtained. Further screening yielded three hub genes (LCP2, MMP9, and NCF2). The nomogram demonstrated good diagnostic performance. The disease group exhibited markedly elevated expression levels of hub genes compared to the control group, as revealed by the qPCR findings.In AS, the analysis of immune infiltration showed irregularity in the infiltration of immune cell. Two molecular subtypes were identified through consensus clustering analysis, with subtype B exhibiting higher expression levels of hub genes and immune checkpoint genes compared to subtype A.

Conclusion

Our study revealed the common inflammatory immune pathways in IBD and AS and constructed a nomogram with good diagnostic performance based on hub genes.

Inflammatory bowel disease

Atherosclerosis

Immune infiltration

Machine learning

Potential crucial biomarker

Bioinformatics analysis

Two primary subtypes of inflammatory bowel disease (IBD) are Crohn's disease and ulcerative colitis, both of which are chronic and recurrent gastrointestinal inflammatory diseases.Crohn's disease typically has its highest occurrence rate between the ages of 20 and 30, while ulcerative colitis tends to affect individuals in their 30s and 40s(1). IBD patients usually experience a persistent systemic inflammatory state because IBD typically affects young people and is characterized by recurrent episodes(2).AS, characterized by the gradual accumulation of lipids, immune cells, and connective tissue within the arterial wall, is the primary factor behind cardiovascular disease, leading to the narrowing and hardening of blood vessels. From the beginning of plaque development to the thrombus rupture in coronary artery disease, inflammation has been demonstrated to play a crucial role in the start and progression of AS(3).It has been reported that IBD patients are more likely to suffer from cardiovascular disease, suggesting they may be related (4–6). Numerous possible mechanisms have been suggested to clarify this correlation. Chronic systemic inflammation, which is a defining feature of IBD, could potentially encourage the advancement and onset of AS by causing endothelial dysfunction and the formation of thrombosis (7, 8). A patient with cardiovascular disease has elevated amounts of pro-inflammatory markers, such as C-reactive protein (CRP) and homocysteine, which are also markers of chronic systemic inflammation(9). In IBD patients, Elevated levels of these markers may reflect an exacerbated inflammatory state that contributes to the development and progression of AS(10, 11)

However, there is currently a lack of research indicating the common inflammatory immune pathways and biomarkers between the two diseases.With the advancement of technology, the use of bioinformatics and machine learning to screen for biomarkers of diseases has been proven to be a reliable method.To our knowledge, little genetic research has been done on the mechanisms underlying IBD-induced AS.We utilized Limma to identify CDEGs and employed PPI networks and machine learning algorithms to further screen for hub genes.In addition, we performed functional enrichment analysis, immune infiltration analysis, and consensus clustering analysis to elucidate the common inflammatory immune pathways and pathophysiological mechanisms between the two diseases.

2.1 Data source

Figure 1 depicts the research flowchart.From the GEO database, we obtained three high-throughput datasets and four microarray datasets.GSE100927 and GSE75214 were used for data analysis.To confirm the dependability of the analysis findings, GSE28829, GSE57691, GSE87473, GSE208303, and GSE120521 were employed.Each dataset's specific information is played in Table 1.

Table 1

Dataset information
GSE Series	Disease	Samples	Type	Group
GSE100927	AS	69 samples of atherosclerotic vascular tissue and 35 control samples	Microarray dataset	Discovery cohort
GSE75214	IBD	106 colon tissue samples from IBD patients and 11 control samples	Microarray dataset	Discovery cohort
GSE28829	AS	16 advanced atherosclerosis samples and 13 early atherosclerosis samples	Microarray dataset	Validation cohort
GSE57691	AS	9 atherosclerotic aorta samples and 10 control samples	Microarray dataset	Validation cohort
GSE87473	IBD	106 colon tissue samples from UC patients and 21 control samples	Microarray dataset	Validation cohort
GSE120521	AS	4 stable plaque sampless and 4 unstable plaque samples	High-throughput dataset	Validation cohort
GSE208303	IBD	26 colon tissue samples from CD patients and 9 control samples	High-throughput dataset	Validation cohort

2.2 Identification of DEGs

DEGs were identified using the Limma package with the criteria of |log2 fold change (FC)| > 1 and P-value < 0.05.CDEGs were obtained by taking the intersection of Venn diagrams.

2.3 Functional enrichment analysis and construction of PPI network

To analyze Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, we used the “clusterProfiler” package. Results were displayed using the “ggplot2” package. A PPI network was constructed using the STRING database, and the network was visualized using Cytoscape. We utilized the CytoHubba extension in Cytoscape, which encompasses 12 algorithms, to screen hub genes(12). The intersection of the top 15 nodes of genes from 5 algorithms has been preliminarily identified as hub genes.

2.4 Machine learning was used to screen for hub genes.

LASSO regression and SVM are two commonly used feature selection methods. By incorporating a regularization term into the loss function, LASSO regression is able to achieve feature selection and parameter sparsity through the reduction of certain regression coefficients to zero, making it a linear regression technique. The SVM-RFE algorithm is based on SVM and eliminates features recursively by training an SVM model repeatedly, assessing the significance of each feature using support vectors or weight vectors, and gradually removing the least significant features.LASSO regression was performed using the “glmnet” package, while SVM-RFE analysis was conducted using the “e1071 package”.The genes at the intersection of the two methods were considered hub genes.

2.5 Validation of hub genes and construction of a nomogram

Hub genes were subjected to expression validation (with a significance level of P < 0.05) and ROC analysis. We created a nomogram utilizing the “rms” package to elucidate the diagnostic significance of hub genes. To assess the dependability of the nomogram, the ROC curve was employed.

2.6 Immune Infiltration Analysis

Analysis of immune infiltration was conducted using the "CIBERSORT" package. The proportions of immune infiltration in the AS and control groups were analyzed, as well as the correlation between immune cells and hub genes. "ggplot2" and "corrplot" packages were used for graphing.

2.7 Consensus clustering analysis

The “ConsensusClusterPlus” package was used to perform consensus clustering analysis. To confirm the dependability of the subtyping, Principal Component Analysis (PCA) was employed, and the visualization of the PCA outcomes was done with the “ggplot2” package.

2.8 Inflammatory bowel disease model

Beijing HFK Biotech Co., Ltd. provided male C57BL/6J mice that were between 6 and 8 weeks old. The mice were randomly divided into two groups of five each: normal and dextran sulfate sodium (DSS). After being adaptively fed for a week, the control mice were allowed unrestricted access to food and water, while the mice in the DSS group were given a 2.5% DSS solution in their drinking water for seven days straight to establish an IBD model. After fasting for 12 hours on the eighth day, Carbon dioxide euthanasia for mice. Afterwards, the PBS solution was used to cleanse the mice's colons and rectums, and the colon tissues were subsequently preserved at a temperature of -80°C.

2.9 qPCR

Validation of hub gene mRNA expression levels was performed using qPCR.Using Trizol ND, we extract RNA from tissues, and TIANSc(TIANScript RT Kit, China) is used for reverse transcription.GAPDH was utilized as an internal reference for mRNA. At least three measurements were taken and analyzed using the 2-ΔΔCT method to collect data. Table 2 displays the primer pairs utilized for quantitative qPCR in both forward and reverse directions. Supplementary Table S1 shows the expression level of hub genes.

Table 2

Forward and reverse primer pairs
Gene name	forward primer	reverse primer
LCP2	TCCACCCTCCAACAACGAG	GAGACAGTGGCGAGTGATTCC
MMP9	GCAGAGGCATACTTGTACCG	TGATGTTATGATGGTCCCACTTG
NCF2	GAAGATACCTCTCCAGAATCCG	TTCTTAGACACCATGTTCCGAA

3.1 Identification of CDEGs in IBD and AS

In the IBD dataset GES75214, we obtained 785DEGs for UC and 636 DEGs for CD(Figs. 2B and C). The intersection of these two sets of genes, totaling 532, was considered to be the DEGs of IBD (Fig. 2D). The AS dataset revealed 481 DEGs, comprising 355 genes that were up-regulated and 126 genes that were down-regulated(Fig. 2A).a total of 51 CDEGs exhibiting a shared expression pattern were identified through cross-referencing the up-regulated and down-regulated 47 genes from both datasets. (Figs. 2E and F).

3.2 Functional enrichment analysis of CDEGs and construction of PPI network

The biological functions of CDEGs were clarified using enrichment analysis. The GO analysis revealed that CDEGs(Fig. 3A) were mainly enriched in biological processes (BP) such as “leukocyte-mediated immunity,” “T cell activation,” “regulation of T cell activation,” “cytokine-mediated signaling pathway,” and “leukocyte cell-cell adhesion.” Molecular functions (MF) included “endocytic vesicle membrane,” “MHC class II protein complex,” “endocytic vesicle,” “MHC protein complex,” and “tertiary granule membrane.” Cellular components (CC) included “serine-type endopeptidase activity,” “serine-type peptidase activity,” “serine hydrolase activity,” “immune receptor activity,” and “MHC protein complex binding.” The KEGG pathway analysis revealed (Figure. 3B) that CDEGs were mainly enriched in the FcεRI signaling pathway and B cell receptor signaling pathway. Furthermore, we found that CDEGs were related to the development of inflammatory diseases (“lipid and atherosclerosis”, “asthma”, and “inflammatory bowel disease”), infectious diseases (“tuberculosis”, “leishmaniasis”, and “Yersinia infection”), and Inflammatory response(“osteoclast differentiation” and “platelet activation”).

CDEGs were imported into the STRING database to construct a PPI network containing 51 nodes and 191 edges(Fig. 3C). The STRING database was used to import CDEGs to construct a PPI network with 51 nodes and 191 edges. Initially, the cytoHubba plugin was utilized to analyze the PPI network with the aid of five algorithms (Closeness, EPC, MCC, Stress, and Radia), resulting in the acquisition of the top 15 node genes for each algorithm. Using 5 algorithms, 11 hub genes were preliminarily determined by intersecting their results(Fig. 3D). Finally, The Metascape platform was utilized to conduct enrichment analysis on the 11 hub genes(Fig. 3E). The findings indicated that the hub genes played a role in the osteoclast differentiation, microglial phagocytosis pathway of pathogens, fluid shear stress, atherosclerosis, and neutrophil extracellular traps. This result emphasized the important role of inflammation and immune response in IBD-induced AS.

(C) PPI network diagram of CDEGs. (D) Venn diagram of the results screened by five algorithms of CytoHubba. (E)Functional enrichment analysis of hub genes.

3.3 Machine learning was used to screen for hub genes

In the GSE100927 and GSE75214 datasets, LASSO regression and SVM-RFE were employed for more refinement of the selection of hub genes. The results showed that 6 and 8 hub genes were obtained from the GSE100927 dataset using LASSO regression and SVM-RFE, respectively(Figures. 4A and C). In the GSE75214 dataset, we obtained 5 and 11 hub genes (Figures. 4B and D). The intersection of these results determined 3 hub genes (NFC2, MMP9, and LCP2) (Figure. 4E). To better understand the biological purposes of the hub genes, we constructed a gene co-expression network using the GeneMANIA database, and its associated functions were displayed. Physical interactions accounted for 77.64%, co-expression for 8.01%, predict for 5.37%, co-localization for 3.63%, genetic interactions for 2.83%, pathways for 1.88%, and shared protein domains for 0.6% (Fig. 4F). As expected, the hub genes were mainly related to oxidative stress and immune response. In addition, we performed ssGSEA correlation analysis on the hub genes in the IBD dataset. The results demonstrated that “APOPTOSIS,” “IL2_STAT5_SIGNALING,” “IL6_JAK_STAT3_SIGNALING,” “HYPOXIA,” and “INFLAMMATORY_RESPONSE” were significantly positively correlated, and these five hallmark gene sets were closely related to the development of AS (Fig. 4G). This finding suggests that the three hub genes are closely related to the development of AS in IBD patients.

3.4 Validation of hub genes and construction of nomogram

In order to verify the reliability of the above results, we validated the expression levels of hub genes in external AS datasets and IBD datasets. The results showed that the expression levels of hub genes were higher than those of the control group(Figs. 5A-D). ROC analysis showed that the area under the ROC curve (AUC) of 3 hub genes was > 0.7, indicating good diagnostic performance(Figs. 6B-L). In addition, in the IBD mouse model, we also detected the expression levels of hub genes, which were consistent with previous analysis results and were higher than those of the control group(Figs. 5E-G). Finally, based on the three hub genes, we constructed a diagnostic nomogram (Fig. 6A) and validated its predictive performance in the GSE57691 dataset (Fig. 6M). This result demonstrated the reliability of the nomogram for diagnosis.

Figuer5: Validation of hub genes. (A) Expression differences of 3 hub genes in the GSE28829 dataset. (B) Expression differences of 3 hub genes in the GSE9dataset. (C) Expression differences of 3 hub genes in the GSE208303. (D) Expression differences of 3 hub genes in the GSE120521. (D-F) The relative expressions of hub genes were validated by q-PCR

3.5 Immune infiltration analysis

An immune infiltration analysis was conducted to gain a better understanding of the immune regulation of AS, as the enrichment analysis showed that CDEGs were strongly associated with immune response. The strip chart shown in Fig. 7A illustrates the composition of immune cells in both the atherosclerosis patients and healthy controls. In the AS group, the proportion of memory B cells, M0 macrophages, activated mast cells, CD4 naive T cells, and T cells were found to be higher compared to the control group. The proportion of resting mast cells, M1 macrophages, and monocytes was lower (Fig. 7B). The heatmap of the correlation between immune cells showed that resting mast cells were significantly positively correlated with resting dendritic cells (r = 0.8) while resting memory T cells were significantly negatively correlated with M0 macrophages (r = 0.65)(Fig. 7C). Furthermore, 3 hub genes were found to be significantly positively associated with M0 macrophages, γ-δT cells, regulatory T cells, and activated mast cells. They were significantly negatively correlated with resting memory T cells and naive B cells (Figs. 7D-F), and had the strongest positive correlation with M0 macrophages(Figs. 7G-I). Therefore, we believe that hub genes may play an important role in the concurrent AS occurrence in IBD by affecting immune cell infiltration.

3.6 Identification of molecular subtypes

Utilizing the expression profile of CDEGs in AS samples, consensus clustering analysis was performed and two subtypes, subtype A and subtype B, were identified(Figs. 8A and B). The PCA analysis was used to further validate the reliability of the classification (Fig. 8C). The expression levels of the 3 hub genes in subtype B were significantly lower than those in subtype A (Fig. 8D). The proportions of M0 macrophages, activated mast cells, and γ-δ T cells in subtype B were significantly lower than those in subtype A (Fig. 8E). In addition, most immune checkpoint genes were highly expressed in subtype A than in subtype B (Fig. 8F).

IBD is a group of chronic non-specific diseases with an unknown cause. IBD affects at least 6.8 million people worldwide, and its inflammatory response involves the intestines but causes systemic inflammatory reactions(13).AS development is closely associated with chronic inflammation according to recent observations (14). The risk of developing atherosclerotic cardiovascular disease in IBD patients has been reported to be significantly higher than in healthy adults(15). In addition, clinical evidence indicates that the risk of acute coronary syndrome significantly increases in IBD patients(16). The immune response, inflammatory reaction, endothelial dysfunction, and changes in intestinal flora are possible mechanisms for the development of AS in IBD patients. However, there seems to be little research on the pathogenesis of AS in IBD patients at the genetic level.

Our study results suggest that the CDEGs were enriched in KEGG pathways such as “Tuberculosis”, “Lipid and Atherosclerosis”, “Osteoclast Differentiation”, “Platelet Activation”, “FcεRI Signaling Pathway”, “B Cell Receptor Signaling Pathway”, and “Asthma”. Tuberculosis is related to immune response and is known to be associated with elevated serum IgG levels. Asthma is associated with elevated IgE levels. The development of B cells relies on the potency of the signaling of the B cell receptor. B cell differentiation is driven by B cell receptor signaling throughout all stages of B cell development and is associated with B cell autoreactivity upon recognition of self-antigens(17). Previous studies have reported that increased genetic risk for autoimmunity is associated with enhanced B cell receptor signaling(18, 19). Takahiro et al. demonstrated that CD40L/B Tg mice spontaneously develop colitis at 8–15 weeks of age, suggesting that the pathogenesis of IBD may be primarily established by a B cell-triggered mechanism(20). Strong BCR signaling in transitional B cells leads to the development of follicular (FO) B cells, which support Th1 responses and the production of pro-inflammatory cytokines(21). Moreover, upon the stimulation of Tfh cells, FO B cells transform into the germinal center(GC) B cells which are involved in the creation of GC(22). GC B cell proliferation and affinity maturation lead to increased production of IgG and IgE(23). It has been demonstrated in several studies that FO B cells contribute to the promotion of AS primarily through the production of IgG and activation of Th1 cells(24). The FcεRI signaling pathway is related to the development of AS and regulates allergic reactions(25). Activated mast cells through IgE antigen complex cross-linking FcεRI can release and secrete histamine, chemokines, platelet-activating factor, lipid mediators, and proteoglycans, causing inflammation(26). The analysis of immune infiltration also indicated that AS has a greater proportion of memory B cells and mast cells that are activated..

Through the PPI network and machine learning, we screened 3 hub genes (LCP2, MMP9, NCF2) with high diagnostic values and obtained reliable validation in external datasets. Lymphocyte cytosolic protein (LCP2) is capable of regulating T cell receptor signaling in stimulated T lymphocytes and is intimately associated with the progression of IBD(27). Although bioinformatics studies have reported that LCP2 expression is closely related to AS, there is still a lack of experimental research to prove this association(28–30). One possible explanation is that LCP2 is a target gene of STAT1. Upon activation by IFN-γ, STAT1 translocates to the nucleus and interacts with the enhancers of LCP2 and TNFAIP2.EP300 collaborates with p-STAT1 to enhance H3K27ac levels on the enhancer, leading to the transcription of LCP2 and TNFAIP2 and ultimately facilitating the progression of chronic inflammation (31). The intestinal barrier is remodeled by a significant group of enzymes called matrix metalloproteinases (MMPs) that can break down extracellular matrix (ECM) and basement membrane components(32). In the intestines of IBD patients, bacteria or their products trigger the epithelial layer to release large amounts of chemokine (CXCCL8) to mediate neutrophil migration and produce large amounts of MMP9(33). S Arihiro et al. reported that MMP9 was highly expressed in activated fibroblasts and vascular smooth muscle cells in IBD patients and participated in biological processes such as angiogenesis, remodeling, and inflammation(34). In the MMP-9 transgenic (Tg) rabbits model, increased sensitivity to cholesterol diet-induced AS was observed, and vascular wall calcification was more likely to occur compared to the wild type(35). Neutrophil cytoplasmic factor 2 (NCF2) is a subunit of NADPH oxidase that produces superoxide as a major source of reactive oxygen species (ROS)(36, 37). Missense mutations in NCF2 have been reported to be closely related to the development of IBD, and potential destructive mistranslation mutations in NCF2 are associated with reduced neutrophil ROS production and reduced bacterial killing(38). In CD patients with low ROS production, co-enhancement of hypoxia-inducible factor 1, TNF, NF-κB, and TNF signaling was observed, and glucose energy utilization was more dependent on fatty acid metabolism because of glucose energy utilization disorder, leading to AS development(39). Consistent with our nomogram results, NCF2 acts as a protective factor. Notably, univariate regression analysis indicated that an increase in the expression of NCF2 was a risk factor for AS. We speculate that NCF2 may be secondarily elevated after the illness. Furthermore, ssGSEA analysis of hub genes in the IBD dataset revealed that hub genes were significantly positively correlated with “APOPTOSIS”, “IL2_STAT5_SIGNALING”, “IL6_JAK_STAT3_SIGNALING”, “HYPOXIA”, “INFLAMMATORY_RESPONSE”. The immune infiltration analysis revealed that hub genes were significantly positively correlated with MO macrophages, γ-δT cells, regulatory T cells, and activated mast cells, supporting the reliability of the aforementioned views.

In summary, we believe that IBD patients have high sensitivity and that the B cell receptor signaling pathway is up-regulated. Reducing the production of excessive B cells promotes the differentiation of B cells into FO B cells, mediating Th1 responses and the production of pro-inflammatory cytokines, which is related to the development of AS. Furthermore, the B cell receptor signaling pathway is also closely related to the secretion of IgG and IgE. IgG and IgE activate the FcεRI signaling pathway; subsequently, activated mast cells release and secrete histamine, chemokines, platelet-activating factor, lipid mediators, and proteoglycans, causing inflammation and leading to the development of AS. LCP2, MMP9, and NCF2 may play a central role in the aforementioned mechanisms.

Furthermore, we performed a co-clustering analysis to identify two subtypes of AS. Subtype B exhibited a significantly greater expression of hub genes compared to subtype A.Subtype B exhibited a significantly greater percentage of activated mast cells, γ-δT cells, and M0 macrophages compared to subtype A.. Most immune checkpoint genes were also highly expressed in subtype B, suggesting that subtype B patients have a higher correlation with immune infiltration and are more likely to benefit from immunotherapy. This finding provides direction for future precision medicine treatment.

Limitation

The co-expressed differential genes obtained in our study are derived from the intersection of two diseases. Although the reliability of the analysis results has been validated in external datasets, it should be verified by differential genes in patients with IBD complicated by AS. The diversity of sample types can cause bias, but we standardized the samples to minimize errors. Additionally, while this study lacks validation of AS hub gene expression levels through clinical specimens or animal models, the high expression of hub genes has been validated in multiple datasets, ensuring the reliability of the analysis results to the greatest extent.

Our study revealed the common inflammatory immune pathways in IBD and AS and constructed a nomogram with good diagnostic performance based on hub genes.

Ethical Approval and Consent to Participate

All animal experiments were approved by Tianjin Medical University General Hospital Experimental Animal Welfare Ethics Committee (Approval no. IBR2022-DWFL-004) for the use of animals and all methods were carried out in accordance with relevant guidelines and regulations.The study was carried out in compliance with the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines.

Funding

This work was funded by the grants 82000338 ,from the National Natural Science Foundation of China,the project of Tianjin Municipal Science and Technology Bureau (19JCYBJC26700).

Consent for publication

Not applicable

Data Availability Statement

The microarray dataset and high-throughput sequencing dataset of this study are available in the GEO database(Home - GEO - NCBI (nih.gov)). PCR data can be obtained from Supplementary Table S1.

Conflict of interest

The authors declare that they have no competing interests.

Author contributions

XW, ZZ, and ZR contributed equally to this work.XW analyzed the data and wrote the article. ZZ and XW acquired the data. ZR completed animal experiments.ZZ and TW searched the literature. BB and XY conceived and revised the paper. All authors contributed to the article and approved the submitted version.

Acknowledgments

Not applicable.

Authors' information

Xuezhu Wang：Master,Department of Cardiology, Tianjin Medical University General Hospital,Tianjin,China. Address:Tianjin Medical University General Hospital, 154 Anshan St, Heping District, Tianjin, China. Email: [email protected].

Zihan Zhai:Master,Department of Gastroenterology, Tianjin Medical University General Hospital,Tianjin,China. Address:Tianjin Medical University General Hospital, 154 Anshan St, Heping District, Tianjin, China. Email: [email protected].

Zhen Rong:Master,Department of Gastroenterology, Tianjin Medical University General Hospital,Tianjin,China. Address:Tianjin Medical University General Hospital, 154 Anshan St, Heping District, Tianjin, China. Email:[email protected].

Tingting Wei:Master,Department of Cardiology, Tianjin Medical University General Hospital,Tianjin,China. Address:Tianjin Medical University General Hospital, 154 Anshan St, Heping District, Tianjin, China. Email: [email protected].

Xuefang Yu:PhD,Department of Cardiology, Tianjin Medical University General Hospital,Tianjin,China. Address:Tianjin Medical University General Hospital, 154 Anshan St, Heping District, Tianjin, China. Email: [email protected].

Bo Bian:PhD,Department of General Practice, Tianjin Medical University General Hospital,Tianjin,China. Address:Tianjin Medical University General Hospital, 154 Anshan St, Heping District, Tianjin, China. Email:[email protected].

Malik TA. Inflammatory Bowel Disease: Historical Perspective, Epidemiology, and Risk Factors. Surg Clin North Am. 2015;95(6):1105–22.
Rogler G, Singh A, Kavanaugh A, Rubin DT. Extraintestinal Manifestations of Inflammatory Bowel Disease: Current Concepts, Treatment, and Implications for Disease Management. Gastroenterology. 2021;161(4):1118–32.
Libby P. The changing landscape of atherosclerosis. Nature. 2021;592(7855):524–33.
Feng W, Chen G, Cai D, Zhao S, Cheng J, Shen H. Inflammatory Bowel Disease and Risk of Ischemic Heart Disease: An Updated Meta-Analysis of Cohort Studies. J Am Heart Assoc. 2017;6(8).
Sun HH, Tian F. Inflammatory bowel disease and cardiovascular disease incidence and mortality: A meta-analysis. Eur J Prev Cardiol. 2018;25(15):1623–31.
Panhwar MS, Mansoor E, Al-Kindi SG, Sinh P, Katz J, Oliveira GH, et al. Risk of Myocardial Infarction in Inflammatory Bowel Disease: A Population-based National Study. Inflamm Bowel Dis. 2019;25(6):1080–7.
Cibor D, Domagala-Rodacka R, Rodacki T, Jurczyszyn A, Mach T, Owczarek D. Endothelial dysfunction in inflammatory bowel diseases: Pathogenesis, assessment and implications. World J Gastroenterol. 2016;22(3):1067–77.
Chen B, Collen LV, Mowat C, Isaacs KL, Singh S, Kane SV, et al. Inflammatory Bowel Disease and Cardiovascular Diseases. Am J Med. 2022;135(12):1453–60.
Cainzos-Achirica M, Glassner K, Zawahir HS, Dey AK, Agrawal T, Quigley EMM, et al. Inflammatory Bowel Disease and Atherosclerotic Cardiovascular Disease: JACC Review Topic of the Week. J Am Coll Cardiol. 2020;76(24):2895–905.
Rungoe C, Nyboe Andersen N, Jess T. Inflammatory bowel disease and risk of coronary heart disease. Trends Cardiovasc Med. 2015;25(8):699–704.
Libby P, Loscalzo J, Ridker PM, Farkouh ME, Hsue PY, Fuster V, et al. Inflammation, Immunity, and Infection in Atherothrombosis: JACC Review Topic of the Week. J Am Coll Cardiol. 2018;72(17):2071–81.
Chin CH, Chen SH, Wu HH, Ho CW, Ko MT, Lin CY. cytoHubba: identifying hub objects and sub-networks from complex interactome. BMC Syst Biol. 2014;8(Suppl 4):11.
Collaborators GBDIBD. The global, regional, and national burden of inflammatory bowel disease in 195 countries and territories, 1990–2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet Gastroenterol Hepatol. 2020;5(1):17–30.
Lee MT, Mahtta D, Chen L, Hussain A, Al Rifai M, Sinh P, et al. Premature Atherosclerotic Cardiovascular Disease Risk Among Patients with Inflammatory Bowel Disease. Am J Med. 2021;134(8):1047–51. e2.
Fumery M, Xiaocang C, Dauchet L, Gower-Rousseau C, Peyrin-Biroulet L, Colombel JF. Thromboembolic events and cardiovascular mortality in inflammatory bowel diseases: a meta-analysis of observational studies. J Crohn's colitis. 2014;8(6):469–79.
Kirchgesner J, Beaugerie L, Carrat F, Andersen NN, Jess T, Schwarzinger M, et al. Increased risk of acute arterial events in young patients and severely active IBD: a nationwide French cohort study. Gut. 2018;67(7):1261–8.
Khan WN. B cell receptor and BAFF receptor signaling regulation of B cell homeostasis. J Immunol. 2009;183(6):3561–7.
Burn GL, Svensson L, Sanchez-Blanco C, Saini M, Cope AP. Why is PTPN22 a good candidate susceptibility gene for autoimmune disease? FEBS Lett. 2011;585(23):3689–98.
Rawlings DJ, Dai X, Buckner JH. The role of PTPN22 risk variant in the development of autoimmunity: finding common ground between mouse and human. J Immunol. 2015;194(7):2977–84.
Kawamura T, Kanai T, Dohi T, Uraushihara K, Totsuka T, Iiyama R, et al. Ectopic CD40 ligand expression on B cells triggers intestinal inflammation. J Immunol. 2004;172(10):6388–97.
Rawlings DJ, Metzler G, Wray-Dutra M, Jackson SW. Altered B cell signalling in autoimmunity. Nat Rev Immunol. 2017;17(7):421–36.
Tsiantoulas D, Kiss M, Bartolini-Gritti B, Bergthaler A, Mallat Z, Jumaa H, et al. Secreted IgM deficiency leads to increased BCR signaling that results in abnormal splenic B cell development. Sci Rep. 2017;7(1):3540.
Mauri C, Bosma A. Immune Regulatory Function of B Cells. Annu Rev Immunol. 2012;30(1):221–41.
Ma SD, Mussbacher M, Galkina EV. Functional Role of B Cells in Atherosclerosis. Cells [Internet]. 2021; 10(2).
Nai W, Threapleton D, Lu J, Zhang K, Wu H, Fu Y, et al. Identification of novel genes and pathways in carotid atheroma using integrated bioinformatic methods. Sci Rep. 2016;6:18764.
Li Y, Leung PSC, Gershwin ME, Song J. New Mechanistic Advances in FcεRI-Mast Cell-Mediated Allergic Signaling. Clin Rev Allergy Immunol. 2022;63(3):431–46.
Liu H, Schneider H, Recino A, Richardson C, Goldberg MW, Rudd CE. The Immune Adaptor SLP-76 Binds to SUMO-RANGAP1 at Nuclear Pore Complex Filaments to Regulate Nuclear Import of Transcription Factors in T Cells. Mol Cell. 2015;59(5):840–9.
Zhang Q, Liu H, Zhang M, Liu F, Liu T. Identification of co-expressed central genes and transcription factors in atherosclerosis-related intracranial aneurysm. Front Neurol. 2023;14:1055456.
Su W, Zhao Y, Wei Y, Zhang X, Ji J, Yang S. Exploring the Pathogenesis of Psoriasis Complicated With Atherosclerosis via Microarray Data Analysis. Front Immunol. 2021;12:667690.
Zheng Y, Qi B, Gao W, Qi Z, Liu Y, Wang Y, et al. Macrophages-Related Genes Biomarkers in the Deterioration of Atherosclerosis. Front Cardiovasc Med. 2022;9:890321.
Yu Y-L, Chen M, Zhu H, Zhuo M-X, Chen P, Mao Y-J, et al. STAT1 epigenetically regulates LCP2 and TNFAIP2 by recruiting EP300 to contribute to the pathogenesis of inflammatory bowel disease. Clin Epigenetics. 2021;13(1):127.
Coufal S, Galanova N, Bajer L, Gajdarova Z, Schierova D, Jiraskova Zakostelska Z et al. Inflammatory Bowel Disease Types Differ in Markers of Inflammation, Gut Barrier and in Specific Anti-Bacterial Response. Cells [Internet]. 2019; 8(7).
Koelink PJ, Overbeek SA, Braber S, Morgan ME, Henricks PA, Abdul Roda M, et al. Collagen degradation and neutrophilic infiltration: a vicious circle in inflammatory bowel disease. Gut. 2014;63(4):578–87.
Arihiro S, Ohtani H, Hiwatashi N, Torii A, Sorsa T, Nagura H. Vascular smooth muscle cells and pericytes express MMP-1, MMP-9, TIMP-1 and type I procollagen in inflammatory bowel disease. Histopathology. 2001;39(1):50–9.
Chen Y, Waqar AB, Nishijima K, Ning B, Kitajima S, Matsuhisa F, et al. Macrophage-derived MMP-9 enhances the progression of atherosclerotic lesions and vascular calcification in transgenic rabbits. J Cell Mol Med. 2020;24(7):4261–74.
Liu J, Kang R, Tang D. Signaling pathways and defense mechanisms of ferroptosis. FEBS J. 2022;289(22):7038–50.
Martner A, Aydin E, Hellstrand K. NOX2 in autoimmunity, tumor growth and metastasis. J Pathol. 2019;247(2):151–4.
Muise AM, Xu W, Guo CH, Walters TD, Wolters VM, Fattouh R, et al. NADPH oxidase complex and IBD candidate gene studies: identification of a rare variant in NCF2 that results in reduced binding to RAC2. Gut. 2012;61(7):1028–35.
Denson LA, Jurickova I, Karns R, Shaw KA, Cutler DJ, Okou DT, et al. Clinical and Genomic Correlates of Neutrophil Reactive Oxygen Species Production in Pediatric Patients With Crohn's Disease. Gastroenterology. 2018;154(8):2097–110.

No competing interests reported.

SupplementaryTableS1.xls

Download PDF

Version 1

posted

You are reading this latest preprint version

Exploring the potential common genetic characteristics and molecular mechanisms between inflammatory bowel disease and atherosclerosis using bioinformatics analysis and machine learning

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

1 Introduction

2 Methods

2.1 Data source

2.2 Identification of DEGs

2.3 Functional enrichment analysis and construction of PPI network

2.4 Machine learning was used to screen for hub genes.

2.5 Validation of hub genes and construction of a nomogram

2.6 Immune Infiltration Analysis

2.7 Consensus clustering analysis

2.8 Inflammatory bowel disease model

2.9 qPCR

3 Result

3.1 Identification of CDEGs in IBD and AS

3.2 Functional enrichment analysis of CDEGs and construction of PPI network

3.3 Machine learning was used to screen for hub genes

3.4 Validation of hub genes and construction of nomogram

3.5 Immune infiltration analysis

3.6 Identification of molecular subtypes

4 Discussion

5 Conclusion

Declarations

Ethical Approval and Consent to Participate

Funding

Consent for publication

Data Availability Statement

Conflict of interest

Author contributions

Acknowledgments

Authors' information

References

Additional Declarations

Supplementary Files

Status:

Version 1