Introducing ICAM1 as a possible drug target in treatment of eosinophilic esophagitis Authors: Alireza Omidi Farzin, Nikoo Hossein-Khannazer, Babak Arjmand, Nastaran Asri, Zahra Razzaghi, Mostafa Rezaei-Tavirani, Mohammad Javad Ehsani Ardakani, Hamid Mohaghegh-Shalmani Published: 2025 ABSTRACT AbstractAim:Determining of more details of Eosinophilic esophagitis (EoE) molecular mechanism is the main aim of this study. Background:EoE is a chronic inflammatory disease associated with eosinophilic infiltration of the squamous epithelium and malfunction of immune system. Roles of number interleukins such as IL4, IL5, and IL13 are highlighted as the dysregulated genes innEoE. Methods:The significant differentially expressed genes (DEGs) of patient sample versus health controls are explored from Gene Expression Omnibus database and evaluated via GEO2R and directed protein-protein interaction (PPI) analyses to find the key genes that promote EoE. Gene ontology enrichment was applied to screen the critical genes. Results:IL1B, ICAM1, MAPK3, RAC2, IFNG, HGF, FGF2, IRS1, CCL2, GRAP, GNAO1, TBX21, IGF1, SPI1, BMP4, FLT4, MMP2, and PDGFRB were pointed out as the critical genes associated with EoE disease. Significant roles of IL1B and FGF2 were considered as important molecular event in EoE promotion. ICAM1 was highlighted as a possible drug target in EoE treatment.Conclusion:in conclusion, ICAM1 is an excellent actor in EoE disease which have controlled by the top actors as IL1B and FGF2. However, it has controlled by IFNG. Regulation of ICAM1 is suggested as a possible therapeutic protocol to control EoE progression.Key Words: Eosinophilic esophagitis, ICAM1, Network analysis, Treatment, Drug Introduction Eosinophilic esophagitis (EoE) as an esophagus chronic immune-mediated inflammatory disease is seen in adults and children. Esophageal strictures, dysphagia, and food impaction are attributed to EoE (1). Based on literature esophageal dysfunction and histologically by predominantly eosinophilic infiltration of the squamous epithelium are the well-known characters of EoE (2). Genetic predisposition, a dysregulated immune response, and environmental triggers such as dietary antigens and allergens are highlighted as the important players in EoE. Dysregulation of cytokines such as IL4, IL5, and IL13 which are responsible for eosinophil recruitment and activation is considered as immunological dysregulation in EoE patients. Proton-pump inhibitors, dietary changes, and topical corticosteroids are the current treatment procedure against EoE (3). bulk and single-cell RNA sequencing, genome-wide association study, and whole exome sequencing are provided beneficial description of EoE disease (4). Investigations are explored several points about EoE pathogenesis including a complicated network of interactions and signaling between immune, epithelial, and mesenchymal cells, and on intracellular and molecular levels. Overactivation of thymic stromal lymphopoietin and IL13 are pointed out as the crucial molecular events in EoE (5). Using topical steroids, proton pump inhibitors, biological drugs, esophageal dilatation, and food elimination diets are introduced as therapeutic protocols against EoE (6). PPI network analysis is a useful method to explore molecular mechanism of diseases. The method that includes a set of genes or proteins in a network to discriminate them based on their binding properties to each other. In a scale free PPI network there are a set of crucial nodes such as the central node which are critical elements of the network. The directed PPI networks are the networks that, the nodes are connected via directed links such as expression, activation, inhibition, and other-directed actions. In directed PPI network the actor nodes control other elements of network or their neighbors via various types of actions (7-10). In the present study, a dysregulated gene set from EoE-like esophagitis patients versus healthy controls are downloaded from GEO database and assessed via directed PPI network to analyze molecular mechanism of disease and explore possible critical genes which are responsible for its initiation and progression. Methods Data collection Data were retrieved from GEO database by using “eosinophilic esophagitis, gastroesophageal reflux disease” keywords. RNA sequencing profiles of 13 patients with EoE-like inflammatory diseases (EoE-like esophagitis) and 7 unaffected healthy controls were considered to be analyzed. The studied samples were extracted from proximal esophageal biopsies RNA from patients and controls. None of patients and controls were under taken anti-eosinophil treatment. Data are recorded in gene set of GSE148381(https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE148381). The related Data of this gene set are published by Greuter T et al in 2022 and 2024 (11, 12). Distribution of directed PPI network actors based on outdegree value. Genes are clustered in 8 groups, from 10 < x ≤ 20, …, 80 < x ≤ 90. Pre-evaluation analysis Data were assessed via GEO2R program to compare the gene expression profiles of patients versus healthy controls. Uniform Manifold Approximation and Projection (UMAP) and box plot were administrated to show comparability of the studied samples. The significant DEGs were determined based on a P-adjusted < 0.05 and -1 > log (fold change) > 1. Directed PPI network analysis The significant DEGs were included in a directed PPI network via CluePdia application of Cytoscape software. Activation, inhibition, and expression actions were applied to form the related network. The network was analyzed via “Network analyzer” plugin of Cytoscape software. The main connected component of the constructed PPI network was layout based on outdegree value to find the crucial actor nodes. The identified DEGs were grouped based on an interval distance of outdegree value = 10, to identify the crucial top DEGs. To determined relationships between the crucial top actors (DEGs), they were included in a directed sub-network. Gene ontology enrichment The crucial DEGs were enriched via gene ontology to find the important dysregulated biochemical pathways, molecular functions, and biological processes via ClueGO application of Cytoscape software. The biological terms were grouped based on Kappa score and the related genes were screened based on the rate of contribution in the related biological terms. Results Pre-evaluation analysis: UMAP plot analysis revealed that, the studied gene expression profiles cannot discriminate the patient samples from controls (Figure 1). After more investigation, the proper samples were selected for more analysis. As it depicted in Figure 2, the patient samples were separated from the control individuals. Box plot analysis of samples is shown in Figure 3. As depicted in Figure 3, the gene expression profiles are median centric and can be compared statistically. After cleaning data, 3155 significant DEGs were identified as separators of patient samples from healthy controls. UMAP plot of gene expression profiles of patients with EoE-like esophagitis versus healthy controls. Nbrc = 9, indicates that each gene expression profiles is compared with 9 neighbors. UMAP plot of gene expression profiles of patients with EoE-like esophagitis versus healthy controls. Nbrc = 7, indicates that each gene expression profiles is compared with 7 neighbors. A set of patient samples including GSM4485305, GSM4485307, GSM4485308, GSM4485328, GSM4485330, GSM4485332, GSM4485352, and GSM4485353 were selected as the proper patient samples. Box plot presentation of gene expression corrected profiles of 8 patients with EoE-like esophagitis versus 7 healthy controls. Directed PPI network analysis The significant DEGs were assessed to find the crucial DEGs via action map. As depicted in Figure 4, there are many actor genes that control the elements of the PPI network via activation, inhibition, and expression actions. The related outdegree values of the top nodes of the central part of the main connected component of the directed PPI network are presented in Figure 5. As administrated in Figures 4 and 5, IL1B the top actor is characterized by an outdegree value of 82 and the second top gene is MAPK3 with outdegree value of 78. As depicted in Figure 5, the crucial elements of the constructed PPI network are characterized by outdegree values of 17 t0 82. Distribution of the top DEGs based on outdegree value parameter is presented in Figure 6. As administrated in Figure 6, there are two classes of actor genes including class 1 (groups 1 and 2) and the second class as well as groups 3-8. The groups of 3-8 were selected as the crucial top actors. The genes of these groups and their characters are tabulated in Table 1. The crucial top actors are mapped in a directed sub-network by using expression, activation, and inhibition actions (Figure 7). The central part of the main component of the directed PPI network of the dysregulated genes that discriminate patients with EoE-like esophagitis from healthy controls. The nodes are connected via expression, activation, and inhibition actions and are layout based on outdegree value. The central part of the main component of the directed PPI network of the dysregulated genes that discriminate the patients with EoE-like esophagitis from healthy controls. The nodes are connected via expression, activation, and inhibition actions and are visualized based on outdegree value. A sub-network of 18 crucial top actors that are connected with activation (green) and expression (yellow) links. There were not any inhibition relationships. Table 1The crucial actor genes which discriminate the patients with EoE-like esophagitis from healthy controls. FC refers to log (fold change).No.GeneLogFCOutdegreeNo.GeneLogFCOutdegree1IL1B2.78210ICAM1-1.1452MAPK3-1.07811RAC21.7423IFNG4.26012HGF2.9414FGF21.95613IRS12.3355CCL22.55214GRAP2.0356GNAO11.34815TBX212.2357IGF12.24616SPI11.3348BMP41.94617FLT42.1339MMP22.94618PDGFRB1.932 The crucial actor genes which discriminate the patients with EoE-like esophagitis from healthy controls. FC refers to log (fold change). Gene ontology findings for the 18 crucial top actors. The small circles (labled in black) refer to the associated genes. FLT4 is not appeared in this analysis. Gene ontology enrichment Result of gene ontology assessment for the 18 crucial top DEGs is shown in Figure 8. were enriched via gene ontology to find the important dysregulated biochemical pathways, molecular functions, and biological processes via ClueGO application of Cytoscape software. The biological terms were grouped based on Kappa score and the related genes were screened based on the rate of contribution in the related biological terms. The associated genes were grouped based on their association with the biological term groups. As depicted in Figure 9, five actors are connected to two biological terms, while two genes are linked to six biological terms of groups. The 17 associated genes of gene ontology assessment. The genes are grouped based on nomber of associated biological term groups (each color refers to a certain biological term). Discussion Results of pre-evaluation analysis (Figures 1-3) indicates the selected patient and control samples are comparable and PPI network analysis demonstrated that there are 18 crucial DEGs which discriminate patient samples from the control individuals (see Figures 4-6 and Table 1). Adel-Patient K et al published a document about systemic signature of EoE in childeren. Based on this report, 4767 DEGs including 464 up-regulated genes with fold change > 2 and 646 down-regulated individuals with fold change < -2, separate EoE patients from controls (13). Range of fold change in this report as like investigations of Blanchard et al (14) is determined as -8 < fold change < 8. As administrated in Table 1, range of fold change in this study is approximately, same as the reported values of the two published reports (see Table 1). There are several common genes (as like MAPK3 and ICAM3) between our study and results of Patient K et al. However, some crucial genes (for example critical gene of IL1B) are not included in the results of their investigations. As depicted in Figure 7, the 18 crucial DEGs are organized in two subnetworks including a main connected component of 15 genes and other subnetwork of 3 individuals (RAC2, MAPK3, and GNAO1. Results of Figure 9 indicate that RAC2 and GNAO1 are associated with two groups of biological terms (minimum dependency to the biological term groups). Therefore RAC2, MAPK3, and GNAO1 were ignored for more assessment. It should be mentioned that deletion of these genes is a tool to find the genes with higher impact in EoE progression. For example, there is a document that MAPK3 is a drug target candidate in treatment of gastroesophageal reflux disease (15). Since FLT4 was not connected to the determined biological terms, this gene also was disregarded for more investigation. IL1B and FGF2 are appeared as the genes that are connected with the maximum number of the biological term groups (see Figure 9). Therefore, IL1B and FGF2 and their directly or indirectly connected genes (including SPI1, CCL2, ICAM1, IGF1, and IRS1 the directly linked genes and TBX21 IFNG, MMP2 as the indirectly connected individuals) were selected as the critical genes of EoE disease. IL1B as a member of IL1 family is involved in innate immune response and plays a significant role in autoinflammatory diseases (16). Up-regulation of IL1B in EoE patients is reported by Kleuskens MT et al (17). Insulin-like growth factor 1 (IGF-1) expression is suppressed or up-regulated directly by IL1B and FGF2 (see Figure 9). Regulation of cell division, proliferation, and survival is attributed to IGF1 as an anabolic growth factor. Role of IGF1 in improving cardiac function and mediation in action of many growth hormones, improvement of lipid profile, increment of insulin sensitivity together with glucose metabolism improvement is investigated and confirmed (18). Assessment showed polymorphism of IGF genes is associated with esophageal adenocarcinoma or its predecessors (19). As like IGF1, expression of ICAM1 is suppressed and up-regulated directly by IL1B and FGF2. Role of intercellular adhesion molecule 1 (ICAM-1) in inflammation and also tumor progression and prognosis is pointed out in literature (20). It seems, up-regulation of ICAM1 by IL1B and FGF2 is a significant event in EoE. As depicted in Figure 9, IFNG effects on ICAM1 as like IL1B and FGF2. Based on Table 1, IFNG is up-regulated as like IL1B and FGF2 in patients with EoE. It can be concluded that effects of IL1B, FGF2, and IFNG on expression of ICAM1 is a noteworthy molecular event in progress of EoE. C-C motif ligand 2 (CCL2) or monocytic chemotactic protein 1 (MCP-1) is an element of main connected component in Figure 9; it is up-regulated directly by IL1B. This integral chemotactic in EoE patients is demonstrated in Table 1. Investigation of Bhardwaj N et al revealed that, treatment of adults with eosinophilic esophagitis in adults is associated with down-regulation of CCL2 (21). The other high impact gene is salmonella pathogenicity island 1 (SPI1) which up-regulates directly IL1B (see Figure 9). In conclusion, ICAM1 as a powerful actor and well-controlled actor by the main actors can be a suitable drug target against EoE. Conclusion It can be concluded that, EoE disease is well-characterized by immune system mal-function especially dysregulation of IL1B and FGF2 genes. Since ICAM1 as an actor gene was controlled by IL1B and FGF2 beside IFNG (other critical gene); it seems that this gene is a suitable individual to be studied as a drug target in treatment of EoE disease. Both experimental validation and sample size limitation are two restrictions about this project. Acknowledgment This project is supported by Shahid Beheshti University of Medical Sciences. Conflict of interests There is no conflict of interest for authors of this article. References