Integrated analysis of hub genes and miRNA- transcription factor-hub gene interaction network in necrotizing enterocolitis

doi:10.21203/rs.3.rs-2756663/v1

Download PDF

Research Article

Integrated analysis of hub genes and miRNA- transcription factor-hub gene interaction network in necrotizing enterocolitis

https://doi.org/10.21203/rs.3.rs-2756663/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

The aim of this study was to identify hub genes, related transcription factors (TFs) and miRNAs from the miRNA–TF–gene interaction network in necrotizing enterocolitis (NEC).

Methods

Three expression data sets from GEO database that compared NEC with surgical negative controls were used to calculate differentially expressed miRNAs (DEMis) and genes (DEGs). A protein-protein interaction (PPI) network was constructed using DEGs and was used to determine hub genes. miRNAs related to hub genes were identified from the intersection between DEMis and predictions of hub gene-miRNA pairs using Starbase, TFs were predicted by hub genes, TF-miRNA pairs were predicted using miRNet. Finally, the miRNA–TF–hub gene interaction network was formed using these predicted pairs.

Results

A total of 14 DEMis and 123 DEGs were identified from the GEO datasets. One hundred and twenty DEGs were found in the PPI network. A pathogenic-associated interaction network was created by intersecting miRNAs, predicted TFs and hub genes. Article-published RNAs such as hsa-miR-7 or TLR4 were shown in this network, and novel RNAs and TFs (Hsa-miR-200a, GATA3, CXCL5) were shown in the network as important regulator.

Conclusions

This analysis displayed several important hub genes, TFs and miRNAs, some of which were not fully understood in previous studies of NEC. These results may play an important role in future studies on the etiology or treatment of NEC.

Necrotizing enterocolitis

Bioinformatics

Hub gene

Transcription factor

TLR4

CXCL5

Necrotizing enterocolitis (NEC) is a gastrointestinal disease that is prevalent in preterm infants [1], especially in infants born at or prior to 28 weeks of gestation or those with a very or extremely low birth weight(VLBW < 1500 g, ELBW < 1000 g) [2]. It is the leading cause of pediatric surgery due to necrosis of the intestinal epithelium or bowel perforation, even when noninvasive treatments such as antibiotics, ionotropic/fluid support or bowel rest [3], are used to decrease the effects of NEC. NEC may be accompanied by delayed neurodevelopment, cholestasis, short bowel syndrome, infection, multiple organ failure or even death, and some of these symptoms persist into adulthood [4–6]. Well-known risk factors such as preterm birth, infant formula feeding, hypoperfusion, or abnormal microbiota contribute to the etiology of NEC [3, 4], and these factors require detailed explanations for their contributions to NEC despite decades of experimental studies[7, 8].

TFs, miRNAs and downstream targeted mRNAs have been reported in several NEC studies. For example, brain-derived neurotrophic factor (BDNF) reduced TLR4 signaling and NEC severity in mice [9]. the role of the miR-223/nuclear factor I-A (NFIA) axis was revealed while miR-223 was overexpressed in NEC [10]. Plasma miR-1290 was identified as a diagnostic biomarker in the NEC Prospective Cohort Study [11]. Fecal miR-223/miR-451a were proved as specific biomarkers in NEC [12]. These studies have provided sporadic evidence in different aspects of NEC and need more accurate or systematic methodologies to explain the occurrence and development of NEC. Current research in NEC progressively tends to utilize multi-omics and multi data processing [13–16], which may provide new predictive or therapeutic value, whereas few reports of integrated bioinformatic analysis including TF have been reported in NEC.

In this study, we developed a systematic NEC-related interaction network among hub genes, TFs and miRNAs, the ranking of each node in the network was determined by multiple topological algorithms, and the nodes connecting to published disease-related TFs or RNAs as well as potential ones were considered key nodes which represented the whole network and regulated the pathogenesis of NEC. Thus, developing an interaction network might help us uncover the pathological mechanisms or potential targets underlying NEC at the molecular and bioinformatic levels.

The study flow is shown in Fig. 1.

Data download and processing

mRNA (GSE64801 and GSE46619) [7, 8] and miRNA (GSE68054) [17] datasets were downloaded from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/). The former had a total of 14 NEC intestinal samples and nine negative controls verified as non-colitis intestinal samples on platform GPL6244 and GPL11154, the latter had four individual NEC samples and four allogeneic non-NEC negative controls on platform GPL16384. GSE46619 was obtained via profiling array, and GSE64801 data was obtained via high throughput sequencing with its raw data expressed as fragments Per kilobase million, GSE68054 was a non-coding RNA profile obtained by array.

The FIT function in the Limma package [18] was used to differentiate expression of miRNAs and mRNAs, except in the case of the GSE64801 which DEG fold changes were included directly in the additional file two of its original article[7]. Thus, GSE64801 data directly contained DEGs that

were used to validate the accuracy and authenticity of GSE46619 DEGs. Average expression values were noted and calculated if overlaps occurred between one predicted target and two or more DEMis. DEMis and DEGs were identified with parameters of adj. P. Value < 0.05 and |log2FC| ≥ 1 (adj. P. value: adjusted p value by Benjamini - Hochberg method [19]; FC: fold change). Then GSE68054 DEMis and GSE46619 DEGs were plotted by dplyr, ggrepel and ggplot2 packages in R studio and shown as volcano map[20]. Common DEGs (co-DEGs) were identified by the intersection of DEGs from GSE46619 and GSE64801 and were shown by Venn diagram using ggVennDiagram and ggplot2 packages.

Enrichment analysis of co-DEGs

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were used to predict potential biological or pathological functions in co-DEGs. Firstly, co-DEGs were transformed to ENTREZID using bitr and org.Hs.eg.db package. Secondly, the enrichment analyses were conducted using openxlsx, ggplot2, stringr, enrichplot, clusterProfiler [21], GOplot, DOSE [22], ggnewscale, topGO, circlize [23] and ComplexHeatmap [20] packages in R studio. The adj. P. value cutoff was set as 0.05, and the Q value cutoff was set as 0.05.

Protein-protein interaction (PPI) network, clusters and hub genes

The STRING database V11.5 [24] (string-db.org/) was used to evaluate the interactions of DEGs, and the interaction score was set at a medium confidence of 0.400 to screen out unwanted nodes. The final result was exported as short tabular text output and imported in Cytoscape V3.9.1 [25] to construct the PPI network. Disconnected nodes were filtered in Cytoscape. In the PPI network, key clusters were found using the MCODE [26] plugin in Cytoscape, and analysis parameters (Node Density Cutoff: 0.1, Node Score Cutoff: 0.2, Degree Cutoff: 2, K-Core: 2, Haircut: Yes, Fluff: No, Max Depth: 100) were set. The cluster with the highest score was set as the key cluster, and the genes within the key cluster were considered hub genes. GO and KEGG analyses of genes in the key cluster were performed the same as that described above in the enrichment analysis of co-DEGs section.

Prediction of the miRNA–TF–hub gene interaction network

Starbase [27] has retrieve data of the miRNA-target interactions by intersecting the predicting target sites of miRNAs with binding sites of Ago protein, so it was used to predict miRNA-hub gene pairs. miRNet V2.0 [28] (https://www.mirnet.ca/) was used to predict pairs, including miRNA–TF pairs and hub gene–TF pairs using the JASPAR database [29]. The intersection of DEMis from GSE68054 and Starbase predictions represented miRNAs that pair with hub genes and TFs in the interaction network. TFs that commonly appeared both in miRNA–TF pairs and TF–hub gene pairs were chosen for inclusion in the interaction network. Feed-forward loops (FFL, where a TF regulates both miRNA and target gene expression while the same miRNA also regulates the same gene) and feed-backward loops (FBL, a miRNA regulates both the TF and gene while the same TF also regulates the same gene) were predicted by JASPAR database in miRNet. Collectively, these interaction relationships were constructed into a miRNA–TF–hub gene interaction network using Cytoscape. The significance of these nodes was calculated by CytoHubba [30], five rank methods, including Maximal Clique Centrality (MCC), Degree, Edge Percolated Component (EPC), Closeness and Betweenness were used for ranking nodes.

Display of DEMis and DEGs.

At the threshold previously described, there were 1168 down-regulated DEGs and 371 up-regulated DEGs in GSE46619, 421 down-regulated DEGs and 382 up-regulated DEGs in GSE64801, 12 down-regulated DEMis and two up-regulated DEMis in GSE68504. There was an explanation for GSE64801, 383 up-regulated genes were described in original article but 382 up-regulated genes were found in additional file two, the latter was chosen for further processing. The volcano map of the two datasets expect GSE64801 are displayed in Fig. 2A, B. In co-DEGs, there were 44 up-regulated genes and 79 down-regulated genes, Venn diagram were shown in Fig. 2C, D.

Enrichment analysis of co-DEGs

120 genes that originated from the co-DEGs were identified by GO database of org.Hs.eg.db packages, ARSE, MFI2 and FAM65B were failed to mapped. Then 120 genes were used in GO and KEGG analyses to provide overall guidance of co-DEM functional roles in NEC. The top five biological processes (GO-BP category) were plasma lipoprotein particle organization, protein-lipid complex subunit organization, plasma lipoprotein particle assembly, protein-lipid complex assembly and regulation of plasma lipoprotein particle levels. The top five cellular components (GO-CC category) were brush border membrane, brush border, cluster of actin-based cell projections, cell projection membrane and apical plasma membrane. The top five molecular function of GO analysis (GO-MF category) were secondary active transmembrane transporter activity, active ion transmembrane transporter activity, active transmembrane transporter activity, pattern recognition receptor activity and symporter activity. KEGG pathway analysis involved 93 pathways, but none of their Q value is lower than 0.05 so no KEGG bar plot. these results are shown in Fig. 3.

PPI network, clusters and hub genes

The PPI network constructed in Cytoscape contained 80 nodes and 172 edges (Figure. 4A). The key cluster of the PPI network had 10 genes counted by MCODE (Figure. 4B). Its enrichment analysis were shown in Fig. 5. KEGG pathway analysis involved 29 pathways, but none of their Q value is lower than 0.05 so no KEGG bar plot. The result of GO analysis showed great similarity between the PPI network and key cluster.

TF prediction and miRNA–TF–gene Network construction

Using the aforementioned methods, 69 TF–hub gene pairs, 54 miRNA–TF pairs and 14 miRNA–hub gene pairs predictions were identified. Next, the result of the intersection between the DEMis and miRNA predictions from Starbase was the miRNAs that related to hub genes. Then, 9 TFs that commonly appeared both in miRNA–TF pairs and TF–hub gene pairs were chosen for inclusion in the interaction network. Altogether, 13 TF–hub gene pairs, 21 miRNA–TF pairs and 14 miRNA–hub gene pairs were merged to construct the miRNA–TF–hub gene network (Figure. 6). In the constructed interaction network, CXCL5 had the highest level for all set parameters and was thought as the top hub gene in this network. The JUN(TF) rank was slightly lower than CXCL5 and was thought as the top TF in this network, Hsa-miR-200a was a lower rank than JUN and considered the top miRNA. FBL and FFL were not validated in the miRNAs from the interaction network. Ranks are shown in Table 1.

Differentially expressed RNAs, such as genes, miRNAs, lncRNAs and ceRNAs, have been discovered in several published studies on NEC, but limited integrated analysis among different NEC datasets and studies involving TFs in NEC have been established. NEC is a multiple etiologically guided intestinal inflammatory disease. Thus, changes in more than one gene or non-coding RNA may lead to inflammatory response, and it is necessary to establish an integrated network that allows discovery of critical disease-related genes, hub genes or pathogenicity of key clusters involved in the pathological processes of NEC. To our knowledge, this is the first study to construct a miRNA–TF–gene network to predict hub genes and clusters which show higher relevance than other differentially expressed RNAs or TFs in NEC. One hundred and twenty-three co-DEGs and 14 DEMis were identified. After hub gene identification, TF prediction, pairs foundation and FFL/FBL filtration, the miRNA–TF–hub gene interaction network was formed. Based on this network, previously published NEC-related candidate genes (TLR4 [9, 31], NFKB1 [32])and RNAs (Hsa-miR-223 [10]) and novel miRNA–TF–gene triplets that showed great connection towards several fundamental pathogenic pathways were identified. We would like to discuss some of the novel NEC miRNA–gene pairs, TF–gene pairs and gene families in the interaction network, hoping that they will be validated in the future and lead to greater comprehension towards the pathogenesis of NEC.

Table 1 ranks of 15 nodes in the interaction network by five parameters

MCC Maximal Clique Centrality, EPC Edge Percolated Component

In the present study, hsa-miR-200a had the highest MCC/degree/EPC/Closeness/Betweenness among all miRNAs in the interaction network. Hsa-miR-200b, hsa-miR-200c, hsa-miR-429 and hsa-miR-141 also showed adequate closeness and betweenness. These results suggest us that the miR-200 family might have great value or function in NEC. The miR-200 family, composed of 5 members [33] (141, 200a, 200b,200c and 429), were associated by gene expression in various tissues at corresponding times and spaces. Expression of miR-200 family members is mainly associated with neurodegenerative diseases such as Parkinson`s disease or Alzheimer`s disease [34–37] and renal/pulmonary fibrosis [38, 39]. Moreover, miR-200 family have shown tumor-promoting as well as tumor-suppressive effects in multiple cancer types, including colorectal cancer [40, 41], bladder cancer [42, 43], prostate cancer [44, 45], ovarian cancer [46] and so on. In NEC, miR-141 expression is altered during the inflammatory response along with TLR4 [47, 48]. Additionally, miR-200a-3p/141-3p regulate RIPK1 [49, 50], and has been well-studied for its participation in inflammation and cell death [51, 52]. Interestingly, even though all these miR-200 families show great relevance in inflammation, miR-200a has specifically been shown to be related to inflammation-related genes, e.g., TLR4 or CXCL5 [53, 54]. To our knowledge, only one published pathological experiment regarding miR-200a in NEC via RIPK1 has been published [50]. Thus, we would like to discuss further about miR-200a and its downstream TFs or genes.

In the predicted miRNA–TF–hub gene interaction network, 4 hub genes (AGT, CXCL5, PLEK and TLR4) and 3 TFs (GATA3, JUN, NFKB1) were related to miR-200a. Firstly, we would like to discuss 4 hub genes. C-X-C motif chemokine ligand 5 (CXCL5), also named as neutrophil activating peptide 78 (ENA-78), is from the CXC subfamily of chemokines. CXCL5 has been known as a crucial biomarker in most of cancers and had different bioprocess in each type of cancer. CXCL5 was up-regulated in inflammatory bowel disease (IBD) models and patients [55, 56], but the mechanism of CXCL5 in IBD or NEC were hardly validated and few articles regarding CXCL5 and Hsa-miR-200a have been published. TLR4 is a well-known gene in the pathogenesis of NEC, its activation by lipopolysaccharide develops the dysfunction of intestinal epithelium and the activation of inflammatory storm in NEC. In recent years, TLR4 has been further demonstrated to be involved in the regeneration of intestinal epithelial cells (IEC) mediated by IGF-1 produced by macrophages [57], and the TLR4-NFKB signaling axis where TLR4 is located was confirmed to be regulated by MD2 and TRPM7 [58, 59]. The relationship between TLR4 and miRNA has been well studied including Hsa-miR-200a, so few articles about this relationship were found in recent years. Pleckstrin (PLEK) is a mediator of in several pathways and located in cytoplasm and ruffle membrane. In the published articles related to intestinal inflammation, PLEK only appeared in the articles calculating the DEGs, the mechanism of PLEK in intestinal inflammation is still unclear. In GO terms of co-DEGs, PLEK was found in the metabolic processes of various chemicals. Angiotensinogen (AGT) is the precursor of the angiotensin converting enzyme (ACE) which has been known to be in the renin-angiotensin system and mediate sodium homeostasis, blood pressure and inflammation. In GO analysis of co-DEGs, GO terms containing AGT mainly included protein-lipid complex subunit organization, plasma lipoprotein particle organization, lipid localization and lipid transport. Most of these GO terms were accompanied with APOC1, APOC3 and SOAT2. Published articles confirmed that AGT is responsible for acute renal injury [60] and acute pancreatitis [61], and overexpressed AGT correlates with the worst survival of rectum and stomach adenocarcinoma [62]. But none of articles regarding AGT and intestinal disease have been published. Similarly, three TFs (GATA3 [63], JUN [64], NFKB1 [65]) that was related to miR-200a were all validated in the inflammatory response. Because these TFs were only predicted without validated data, they would be discussed based on set parameters from the interaction network in the following paragraph.

For the interaction network, parameters (MCC, Degree, EPC, Closeness and Betweenness) calculated by topology algorithm in interaction network determined which node was critical in the pathogenesis of NEC. CXCL5 had the highest MCC, degree and EPC amongst the hub genes and was regulated by three TFs (GATA3, JUN and NKFB1) that were all related to miR-200a, showing its significance in the interaction network. Overview of CXCL5 has been briefly discussed in the previous paragraph, furthermore, CXCL5 is mainly produced by immune cells, mesenchymal cells and fibroblasts [66–68]. After binding with CXCR2 receptor, CXCL5 recruits T lymphocytes, B lymphocytes and eosinophils to participate in immune response [69], e.g. neutrophil recruitment and tissue injury in acute inflammatory states [70]. In intestinal inflammatory diseases, CXCL5 affects immune cell aggregation during intestinal infection mainly by affecting neutrophil extravasation in the submucosa and neutrophil interstitial migration [71]. Altogether, it is strongly suggested that an implicit correlation between CXCL5 and NEC should be explored, especially in the immune tolerance foundation of preterm infants.

Because TFs were predicted in the interaction network and miRNAs and genes were measured by array with expression profiling, we used the set parameters mentioned above to discriminate the significance among TFs in the interaction network. As a result, RELA, JUN and GATA3 were ranked as top three TFs which were screen out for further discussion. RELA is a gene encoding p65 that constructs a p65/p50 heterodimer complexed with IκB proteins [72]. This p65/p50 heterodimer is the inactive state of NF-κB, which is a well-known factor involved in almost all inflammatory responses. Interestingly, recent finding suggest that RELA transcription activation domains (TADs) increase DNA-binding affinity, especially the binding to nonspecific DNA sequences that leads to overwhelming gene translation and severe inflammatory responses [73]. JUN has a JUN family and participates in the c-Jun/fos and activator protein-1 (AP-1) signaling pathways that control the degree of inflammatory response [74]. Unfortunately, JUN has multiple isoforms, each with different downstream target genes or upstream target miRNAs. Thus, a limited amount of discussion about NEC in regard to JUN can be accomplished. GATA3 is mainly expressed in group 2 innate lymphoid cells, adaptive CD4⁺T helper type 2 (Th2) cells and regulatory T (Treg) cell as a crucial transcription factor [75, 76], GATA3 mediates the development and function of these two cells, and BRD4 has been shown to inhibit the expression of GATA3 and thus affect the differentiation and normal function of Th2 cells [77]. GATA3 has also been shown in regulating Treg cell accompanied with Foxp3 and Setd2 [63, 76]. In lamina propria of small intestine, Lymphoid tissue inducer cell could be dysfunctional when GATA3 was deficient and led to fate determination failure [78]. Moreover, CXCL5 is expressed in immune cells where GATA3 co-located, Hsa-miR-200a was also related to both CXCL5 and GATA3. Hsa-miR-200a-GATA3-CXCL5 interactions are likely to play a key role in the inflammatory response and epithelial dysregulation in NEC.

Some limitations exist in this study. Limited human NEC colon tissue expression data are available. Thus, it is hard to limit type II error in this study. None of the classifications regarding severity or survival in NEC patients exists, results of DEGs and DEMis might be influenced by these clinical differences. Although the relationships in the interaction network are mainly based on scientific experiments, some of them are just predicted interactions. Relationships among miRNA–TF, miRNA–mRNA and TF–mRNA pairs need to be validated by cross-linking immunoprecipitation or qRT-PCR in NEC samples and identified by proper functional experiment. There are multiple miRNAs, TFs and mRNAs to analyze, such as those involved in pyroptosis, preterm immune tolerance status and TF family isoforms. RNA binding protein (RBP) were failed to map in the interaction network due to none of circRNA expression array data availability and RBP is associated with circRNA and miRNA, we believed that it would be nice to add RNA binding proteins when circRNA was present in the network. The database containing miRNA degradation is rare and incomplete so miRNA degradation was not predicted in the interaction network. Discovery and discussion of these miRNAs, TFs and mRNAs may lead to worthwhile information regarding the entire network. Finally, several methods were used for network analysis, and inevitable loss by other methods might lead to some new undiscovered findings.

We constructed a miRNA–TF–hub gene interaction network based on the expression levels of miRNAs and mRNAs in NEC. This network identified hub genes, TFs, miRNAs or clusters which were highly related to the pathogenesis of NEC. The network identifications included well-studied TFs and RNAs, as well as novel ones, implicating their therapeutic potential for further functional studies.

TF: transcription factor; NEC: necrotizing enterocolitis; DEMis: differentially expressed miRNAs; DEGs: differentially expressed genes; PPI: protein-protein interaction; VLBW: very low birth weight; ELBW: extremely low birth weight; BDNF: brain-derived neurotrophic factor; NFIA: nuclear factor I-A; TLR4: toll-like receptor 4; GEO: Gene Expression Omnibus; adj. P. value: adjusted p value; FC: fold change; co-DEGs: Common DEGs; GO: Gene Ontology; KEGG: Kyoto Encyclopedia of Genes and Genomes; FFL: Feed-forward loops; FBL: feed-backward loops; MCC: Maximal Clique Centrality; EPC: Edge Percolated Component; BP: biological processes; CC: cellular components; MF: molecular function; CXCL5: C-X-C motif chemokine ligand 5; ENA-78: neutrophil activating peptide 78; IBD: inflammatory bowel disease; IEC: intestinal epithelial cells; PLEK: Pleckstrin; AGT: Angiotensinogen; ACE: angiotensin converting enzyme; AP1: activator protein-1; Th2: Thelper type 2; Treg: regulatory T.

Ethics approval and consent to participate

Not applicable

Consent for publication

Not applicable

Availability of data and materials

Publicly available datasets were analyzed in this study. These data can be found here: https://www.ncbi.nlm.nih.gov/geo/. The DEM fold change data of GSE64801 can be found in https://bmcmedgenomics.biomedcentral.com/articles/10.1186/s12920-016-0166-9.

Competing Interests

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Funding

This study was supported by the Guangdong Basic and Applied Basic Research Foundation of China (2019A1515 011086).

Authors` contributions

WK and YL contributed to conception and design of the study. SY, WC and YY organized the database and performed the statistical analysis. SY wrote the manuscript and prepared figures. All authors have read and agreed to the published version of the manuscript.

Acknowledgements

We would like to thank all the members performing this study to accomplish this article.

Cai X, Golubkova A, Hunter CJ: Advances in our understanding of the molecular pathogenesis of necrotizing enterocolitis. BMC PEDIATR 2022, 22(1):225.
Yee WH, Soraisham AS, Shah VS, Aziz K, Yoon W, Lee SK, Canadian NN: Incidence and timing of presentation of necrotizing enterocolitis in preterm infants. Pediatrics (Evanston) 2012, 129(2):e298.
Meister AL, Doheny KK, Travagli RA: Necrotizing enterocolitis: It's not all in the gut. Exp Biol Med (Maywood) 2020, 245(2):85-95.
Neu J: Necrotizing Enterocolitis: The Future. NEONATOLOGY 2020, 117(2):240-244.
Gephart SM, Hanson C, Wetzel CM, Fleiner M, Umberger E, Martin L, Rao S, Agrawal A, Marin T, Kirmani K et al: NEC-zero recommendations from scoping review of evidence to prevent and foster timely recognition of necrotizing enterocolitis. Maternal Health, Neonatology and Perinatology 2017, 3(1):23.
Parker R: Probiotic Guideline for Necrotizing Enterocolitis Prevention in Very Low-Birth-Weight Neonates. ADV NEONAT CARE 2014, 14(2):88-95.
Tremblay E, Thibault MP, Ferretti E, Babakissa C, Bertelle V, Bettolli M, Burghardt KM, Colombani JF, Grynspan D, Levy E et al: Gene expression profiling in necrotizing enterocolitis reveals pathways common to those reported in Crohn's disease. BMC MED GENOMICS 2016, 9:6.
Chan KYY, Leung KT, Tam YH, Lam HS, Cheung HM, Ma TPY, Lee KH, To KF, Li K, Ng PC: Genome-wide Expression Profiles of Necrotizing Enterocolitis Versus Spontaneous Intestinal Perforation in Human Intestinal Tissues. ANN SURG 2014, 260(6):1128-1137.
Kovler ML, Gonzalez Salazar AJ, Fulton WB, Lu P, Yamaguchi Y, Zhou Q, Sampah M, Ishiyama A, Prindle T, Wang S et al: Toll-like receptor 4–mediated enteric glia loss is critical for the development of necrotizing enterocolitis. SCI TRANSL MED 2021, 13(612):g3459.
Wu YZ, Chan K, Leung KT, Lam HS, Tam YH, Lee KH, Li K, Ng PC: The miR-223/nuclear factor I-A axis regulates inflammation and cellular functions in intestinal tissues with necrotizing enterocolitis. FEBS OPEN BIO 2021, 11(7):1907-1920.
Ng PC, Chan K, Yuen TP, Sit T, Lam HS, Leung KT, Wong R, Chan L, Pang Y, Cheung HM et al: Plasma miR-1290 Is a Novel and Specific Biomarker for Early Diagnosis of Necrotizing Enterocolitis-Biomarker Discovery with Prospective Cohort Evaluation. J Pediatr 2019, 205(2):83-90.
Ng PC, Chan K, Lam HS, Wong R, Ma T, Sit T, Leung KT, Chan L, Pang Y, Cheung HM et al: A Prospective Cohort Study of Fecal miR-223 and miR-451a as Noninvasive and Specific Biomarkers for Diagnosis of Necrotizing Enterocolitis in Preterm Infants. NEONATOLOGY 2020, 117(5):555-561.
Guo J, Ning Y, Su Z, Guo L, Gu Y: Identification of hub genes and regulatory networks in histologically unstable carotid atherosclerotic plaque by bioinformatics analysis. BMC MED GENOMICS 2022, 15(1):145.
Mahapatra S, Bhuyan R, Das J, Swarnkar T: Integrated multiplex network based approach for hub gene identification in oral cancer. Heliyon 2021, 7(7):e7418.
Jiang M, Cheng Y, Wang D, Lu Y, Gu S, Wang C, Huang Y, Li Y: Transcriptional network modulated by the prognostic signature transcription factors and their long noncoding RNA partners in primary prostate cancer. EBIOMEDICINE 2021, 63(1):103150.
Ma J, Wang P, Huang L, Qiao J, Li J: Bioinformatic analysis reveals an exosomal miRNA-mRNA network in colorectal cancer. BMC MED GENOMICS 2021, 14(1):60.
Ng PC, Chan KYY, Leung KT, Tam YH, Ma TPY, Lam HS, Cheung HM, Lee KH, To KF, Li K: Comparative MiRNA Expressional Profiles and Molecular Networks in Human Small Bowel Tissues of Necrotizing Enterocolitis and Spontaneous Intestinal Perforation. PLOS ONE 2015, 10(8):e135737.
Phipson B, Lee S, Majewski IJ, Alexander WS, Smyth GK: Robust hyperparameter estimation protects against hypervariable genes and improves power to detect differential expression. The Annals of Applied Statistics 2016, 10(2):946-963.
Glickman ME, Rao SR, Schultz MR: False discovery rate control is a recommended alternative to Bonferroni-type adjustments in health studies. J CLIN EPIDEMIOL 2014, 67(8):850-857.
Gu Z, Eils R, Schlesner M: Complex heatmaps reveal patterns and correlations in multidimensional genomic data. BIOINFORMATICS 2016, 32(18):2847-2849.
Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, Feng T, Zhou L, Tang W, Zhan L et al: clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. The Innovation 2021, 2(3):100141.
Yu G, Wang LG, Yan GR, He QY: DOSE: an R/Bioconductor package for disease ontology semantic and enrichment analysis. BIOINFORMATICS 2015, 31(4):608-609.
Gu Z, Gu L, Eils R, Schlesner M, Brors B: circlize Implements and enhances circular visualization in R. BIOINFORMATICS 2014, 30(19):2811-2812.
Szklarczyk D, Gable AL, Nastou KC, Lyon D, Kirsch R, Pyysalo S, Doncheva NT, Legeay M, Fang T, Bork P et al: The STRING database in 2021: customizable protein–protein networks, and functional characterization of user-uploaded gene/measurement sets. NUCLEIC ACIDS RES 2021, 49(D1):D605-D612.
Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B, Ideker T: Cytoscape: a software environment for integrated models of biomolecular interaction networks. GENOME RES 2003, 13(11):2498-2504.
Bader GD, Hogue CW: An automated method for finding molecular complexes in large protein interaction networks. BMC BIOINFORMATICS 2003, 4:2.
Li JH, Liu S, Zhou H, Qu LH, Yang JH: starBase v2.0: decoding miRNA-ceRNA, miRNA-ncRNA and protein-RNA interaction networks from large-scale CLIP-Seq data. NUCLEIC ACIDS RES 2014, 42(Database issue):D92-D97.
Chang L, Zhou G, Soufan O, Xia J: miRNet 2.0: network-based visual analytics for miRNA functional analysis and systems biology. NUCLEIC ACIDS RES 2020, 48(W1):W244-W251.
Castro-Mondragon JA, Riudavets-Puig R, Rauluseviciute I, Lemma RB, Turchi L, Blanc-Mathieu R, Lucas J, Boddie P, Khan A, Manosalva PN et al: JASPAR 2022: the 9th release of the open-access database of transcription factor binding profiles. NUCLEIC ACIDS RES 2022, 50(D1):D165-D173.
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(Suppl 4):S11.
Shaw AG, Sim K, Rose G, Wooldridge DJ, Li M, Misra RV, Gharbia S, Kroll JS: Premature neonatal gut microbial community patterns supporting an epithelial TLR-mediated pathway for necrotizing enterocolitis. BMC MICROBIOL 2021, 21(1):225.
Wijendran V, Brenna JT, Wang DH, Zhu W, Meng D, Ganguli K, Kothapalli KSD, Requena P, Innis S, Walker WA: Long-chain polyunsaturated fatty acids attenuate the IL-1β-induced proinflammatory response in human fetal intestinal epithelial cells. PEDIATR RES 2015, 78(6):626-633.
Hoefert JE, Bjerke GA, Wang D, Yi R: The microRNA-200 family coordinately regulates cell adhesion and proliferation in hair morphogenesis. J CELL BIOL 2018, 217(6):2185-2204.
Zheng Y, Dong L, Liu N, Luo X, He Z: Mir-141-3p Regulates Apoptosis and Mitochondrial Membrane Potential via Targeting Sirtuin1 in a 1-Methyl-4-Phenylpyridinium in vitro Model of Parkinson's Disease. BIOMED RES INT 2020, 2020:7239895.
Liu CG, Wang JL, Li L, Xue LX, Zhang YQ, Wang PC: MicroRNA-135a and -200b, potential Biomarkers for Alzheimers disease, regulate beta secretase and amyloid precursor protein. BRAIN RES 2014, 1583:55-64.
Salimian N, Peymani M, Ghaedi K, Nasr EM: Modulation in miR-200a/SIRT1axis is associated with apoptosis in MPP(+)-induced SH-SY5Y cells. GENE 2018, 674:25-30.
Rostamian DM, Baghi M, Safaeinejad Z, Kiani-Esfahani A, Ghaedi K, Nasr-Esfahani MH: Differential expression of miR-34a, miR-141, and miR-9 in MPP+-treated differentiated PC12 cells as a model of Parkinson's disease. GENE 2018, 662:54-65.
Hajarnis S, Yheskel M, Williams D, Brefort T, Glaudemans B, Debaix H, Baum M, Devuyst O, Patel V: Suppression of microRNA Activity in Kidney Collecting Ducts Induces Partial Loss of Epithelial Phenotype and Renal Fibrosis. J AM SOC NEPHROL 2018, 29(2):518-531.
Chilosi M, Calio A, Rossi A, Gilioli E, Pedica F, Montagna L, Pedron S, Confalonieri M, Doglioni C, Ziesche R et al: Epithelial to mesenchymal transition-related proteins ZEB1, beta-catenin, and beta-tubulin-III in idiopathic pulmonary fibrosis. Mod Pathol 2017, 30(1):26-38.
Paterson EL, Kazenwadel J, Bert AG, Khew-Goodall Y, Ruszkiewicz A, Goodall GJ: Down-regulation of the miRNA-200 family at the invasive front of colorectal cancers with degraded basement membrane indicates EMT is involved in cancer progression. NEOPLASIA 2013, 15(2):180-191.
Eslamizadeh S, Heidari M, Agah S, Faghihloo E, Ghazi H, Mirzaei A, Akbari A: The Role of MicroRNA Signature as Diagnostic Biomarkers in Different Clinical Stages of Colorectal Cancer. CELL J 2018, 20(2):220-230.
Tolle A, Blobel CC, Jung K: Circulating miRNAs in blood and urine as diagnostic and prognostic biomarkers for bladder cancer: an update in 2017. BIOMARK MED 2018, 12(6):667-676.
Miah S, Dudziec E, Drayton RM, Zlotta AR, Morgan SL, Rosario DJ, Hamdy FC, Catto JW: An evaluation of urinary microRNA reveals a high sensitivity for bladder cancer. Br J Cancer 2012, 107(1):123-128.
Brase JC, Johannes M, Schlomm T, Falth M, Haese A, Steuber T, Beissbarth T, Kuner R, Sultmann H: Circulating miRNAs are correlated with tumor progression in prostate cancer. INT J CANCER 2011, 128(3):608-616.
Bryant RJ, Pawlowski T, Catto JW, Marsden G, Vessella RL, Rhees B, Kuslich C, Visakorpi T, Hamdy FC: Changes in circulating microRNA levels associated with prostate cancer. Br J Cancer 2012, 106(4):768-774.
Marton E, Lukacs J, Penyige A, Janka E, Hegedus L, Soltesz B, Mehes G, Poka R, Nagy B, Szilagyi M: Circulating epithelial-mesenchymal transition-associated miRNAs are promising biomarkers in ovarian cancer. J BIOTECHNOL 2019, 297:58-65.
Lin X, Wang Y: miR-141 is negatively correlated with TLR4 in neonatal sepsis and regulates LPS-induced inflammatory responses in monocytes. BRAZ J MED BIOL RES 2021, 54(7):e10603.
Qian WH, Liu YY, Li X, Pan Y: MicroRNA-141 ameliorates alcoholic hepatitisinduced intestinal injury and intestinal endotoxemia partially via a TLR4-dependent mechanism. INT J MOL MED 2019, 44(2):569-581.
Li X, Wang Y, Wang Y, He X: MiR-141-3p ameliorates RIPK1-mediated necroptosis of intestinal epithelial cells in necrotizing enterocolitis. Aging (Albany NY) 2020, 12(18):18073-18083.
Liu Y, Wang Z, Huang H, Shou K: miR-200a-3p improves neonatal necrotizing enterocolitis by regulating RIPK1. AM J TRANSL RES 2021, 13(11):12662-12672.
Calvo-Rodriguez M, de la Fuente C, Garcia-Durillo M, Garcia-Rodriguez C, Villalobos C, Nunez L: Aging and amyloid beta oligomers enhance TLR4 expression, LPS-induced Ca(2+) responses, and neuron cell death in cultured rat hippocampal neurons. J Neuroinflammation 2017, 14(1):24.
Newton K: RIPK1 and RIPK3: critical regulators of inflammation and cell death. TRENDS CELL BIOL 2015, 25(6):347-353.
Wendlandt EB, Graff JW, Gioannini TL, McCaffrey AP, Wilson ME: The role of microRNAs miR-200b and miR-200c in TLR4 signaling and NF-kappaB activation. Innate Immun 2012, 18(6):846-855.
Cantini L, Isella C, Petti C, Picco G, Chiola S, Ficarra E, Caselle M, Medico E: MicroRNA-mRNA interactions underlying colorectal cancer molecular subtypes. NAT COMMUN 2015, 6:8878.
Zhang J, Cen L, Zhang X, Tang C, Chen Y, Zhang Y, Yu M, Lu C, Li M, Li S et al: MPST deficiency promotes intestinal epithelial cell apoptosis and aggravates inflammatory bowel disease via AKT. REDOX BIOL 2022, 56:102469.
Zhou Z, Plug LG, Patente TA, de Jonge-Muller E, Elmagd AA, van der Meulen-de JA, Everts B, Barnhoorn MC, Hawinkels L: Increased stromal PFKFB3-mediated glycolysis in inflammatory bowel disease contributes to intestinal inflammation. FRONT IMMUNOL 2022, 13:966067.
Yan X, Managlia E, Zhao YY, Tan XD, De Plaen IG: Macrophage-derived IGF-1 protects the neonatal intestine against necrotizing enterocolitis by promoting microvascular development. Commun Biol 2022, 5(1):320.
An L, Li J, Liu B, Hui J, Zhang Q, Zhang X, Wang Q: Knockdown of TRPM7 attenuates apoptosis and inflammation in neonatal necrotizing enterocolitis model cell IEC-6 via modulating TLR4/NF-kappaB and MEK/ERK pathways. IRAN J BASIC MED SCI 2022, 25(8):947-953.
Huang D, Wang P, Chen J, Li Y, Zhu M, Tang Y, Zhou W: Selective targeting of MD2 attenuates intestinal inflammation and prevents neonatal necrotizing enterocolitis by suppressing TLR4 signaling. FRONT IMMUNOL 2022, 13:995791.
Mu L, Hu G, Liu J, Chen Y, Cui W, Qiao L: Protective Effects of Naringenin in a Rat Model of Sepsis-Triggered Acute Kidney Injury via Activation of Antioxidant Enzymes and Reduction in Urinary Angiotensinogen. Med Sci Monit 2019, 25:5986-5991.
Sun B, Chen Z, Chi Q, Zhang Y, Gao B: Endogenous tRNA-derived small RNA (tRF3-Thr-AGT) inhibits ZBP1/NLRP3 pathway-mediated cell pyroptosis to attenuate acute pancreatitis (AP). J CELL MOL MED 2021, 25(22):10441-10453.
Hu W, Abdel-Maksoud MA, Khalid SY, Malik A, Alkholief M, Mubarak A, Aufy M: Angiotensinogen, a promising gene signature for rectum and stomach adenocarcinoma patients. AM J TRANSL RES 2022, 14(12):8879-8892.
Ding Z, Cai T, Tang J, Sun H, Qi X, Zhang Y, Ji Y, Yuan L, Chang H, Ma Y et al: Setd2 supports GATA3(+)ST2(+) thymic-derived Treg cells and suppresses intestinal inflammation. NAT COMMUN 2022, 13(1):7468.
Zhong X, Zheng L, Shen J, Zhang D, Xiong M, Zhang Y, He X, Tanyi JL, Yang F, Montone KT et al: Suppression of MicroRNA 200 Family Expression by Oncogenic KRAS Activation Promotes Cell Survival and Epithelial-Mesenchymal Transition in KRAS-Driven Cancer. MOL CELL BIOL 2016, 36(21):2742-2754.
Gonzalez-Granado LI, Ruiz-Garcia R, Blas-Espada J, Moreno-Villares JM, German-Diaz M, Lopez-Nevado M, Paz-Artal E, Toldos O, Rodriguez-Gil Y, de Inocencio J et al: Acquired and Innate Immunity Impairment and Severe Disseminated Mycobacterium genavense Infection in a Patient With a NF-kappaB1 Deficiency. FRONT IMMUNOL 2018, 9:3148.
Rousselle A, Qadri F, Leukel L, Yilmaz R, Fontaine JF, Sihn G, Bader M, Ahluwalia A, Duchene J: CXCL5 limits macrophage foam cell formation in atherosclerosis. J CLIN INVEST 2013, 123(3):1343-1347.
Peng Y, Kajiyama H, Yuan H, Nakamura K, Yoshihara M, Yokoi A, Fujikake K, Yasui H, Yoshikawa N, Suzuki S et al: PAI-1 secreted from metastatic ovarian cancer cells triggers the tumor-promoting role of the mesothelium in a feedback loop to accelerate peritoneal dissemination. CANCER LETT 2019, 442:181-192.
Xu WW, Li B, Guan XY, Chung SK, Wang Y, Yip YL, Law SY, Chan KT, Lee NP, Chan KW et al: Cancer cell-secreted IGF2 instigates fibroblasts and bone marrow-derived vascular progenitor cells to promote cancer progression. NAT COMMUN 2017, 8:14399.
Griffith JW, Sokol CL, Luster AD: Chemokines and chemokine receptors: positioning cells for host defense and immunity. ANNU REV IMMUNOL 2014, 32:659-702.
Madalli S, Beyrau M, Whiteford J, Duchene J, Singh NI, Patel NS, Motwani MP, Gilroy DW, Thiemermann C, Nourshargh S et al: Sex-specific regulation of chemokine Cxcl5/6 controls neutrophil recruitment and tissue injury in acute inflammatory states. BIOL SEX DIFFER 2015, 6:27.
Sullivan DP, Bui T, Muller WA, Butin-Israeli V, Sumagin R: In vivo imaging reveals unique neutrophil transendothelial migration patterns in inflamed intestines. MUCOSAL IMMUNOL 2018, 11(6):1571-1581.
Hayden MS, Ghosh S: Shared principles in NF-kappaB signaling. CELL 2008, 132(3):344-362.
Baughman H, Narang D, Chen W, Villagran SA, Lee J, Bachochin MJ, Gunther TR, Wolynes PG, Komives EA: An intrinsically disordered transcription activation domain increases the DNA binding affinity and reduces the specificity of NFkappaB p50/RelA. J BIOL CHEM 2022, 298(9):102349.
Wan P, Zhang S, Ruan Z, Liu X, Yang G, Jia Y, Li Y, Pan P, Wang W, Li G et al: AP-1 signaling pathway promotes pro-IL-1beta transcription to facilitate NLRP3 inflammasome activation upon influenza A virus infection. VIRULENCE 2022, 13(1):502-513.
Kasal DN, Liang Z, Hollinger MK, O'Leary CY, Lisicka W, Sperling AI, Bendelac A: A Gata3 enhancer necessary for ILC2 development and function. Proc Natl Acad Sci U S A 2021, 118(32).
Rudra D, DeRoos P, Chaudhry A, Niec RE, Arvey A, Samstein RM, Leslie C, Shaffer SA, Goodlett DR, Rudensky AY: Transcription factor Foxp3 and its protein partners form a complex regulatory network. NAT IMMUNOL 2012, 13(10):1010-1019.
Zhao L, Wang Y, Jaganathan A, Sun Y, Ma N, Li N, Han X, Sun X, Yi H, Fu S et al: BRD4-PRC2 represses transcription of T-helper 2-specific negative regulators during T-cell differentiation. EMBO J 2023, 42(6):e111473.
Zhong C, Zheng M, Cui K, Martins AJ, Hu G, Li D, Tessarollo L, Kozlov S, Keller JR, Tsang JS et al: Differential Expression of the Transcription Factor GATA3 Specifies Lineage and Functions of Innate Lymphoid Cells. IMMUNITY 2020, 52(1):83-95.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Integrated analysis of hub genes and miRNA- transcription factor-hub gene interaction network in necrotizing enterocolitis

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Background

Methods

Data download and processing

Enrichment analysis of co-DEGs

Protein-protein interaction (PPI) network, clusters and hub genes

Prediction of the miRNA–TF–hub gene interaction network

Results

Enrichment analysis of co-DEGs

PPI network, clusters and hub genes

TF prediction and miRNA–TF–gene Network construction

Discussion

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1