In Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Takashi Kitani, Sushma C. Maddipatla, Ramya Madupuri, James N. Baraniuk, Christopher Greco, Jonathan Hartman, Sona Vasudevan
IIn Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Takashi Kitani , Sushma C. Maddipatla , Ramya Madupuri , James N. Baraniuk , Christopher Greco , Jonathan Hartman & Sona Vasudevan ✉ Department of Neurology, Georgetown University Hospital, 3800 Reservoir Rd NW, Washington, DC 20007, USA Department of Biochemistry, Molecular and Cellular Biology, Georgetown University Medical Center, 3900 Reservoir Road, NW, Washington DC, 20057, USA Division of Rheumatology, Immunology and Allergy, Department of Medicine, Georgetown University Medical Center, 3900 Reservoir Rd NW, Washington, DC, 20007, USA Dahlgren Memorial Library, Graduate Health & Life Sciences Research Library, Georgetown University Medical Center, 3900 Reservoir Road, NW, Washington DC 20057, USA ✉ Corresponding Author . Email: [email protected]; Phone: 202-687-2242.
Synopsis
This work presents a Systems and a Network Medicine view of Crohn’s and Ulcerative Colitis (UC). This work has identified distinctly different genes and molecular processes involved in each disease, which were used to find potential drug targets, as well as novel candidate drugs.
Abstract
Crohn’s Disease (CD) and Ulcerative Colitis (UC), both under the umbrella of Inflammatory Bowel Diseases (IBD), involve many distinct molecular processes. The difference in molecular processes is studied using the different genes involved in each disease, and explored further for drug targeting and drug repurposing.
Methods
We have created a manually curated one-stop-shop informatics database (IBD-Omics) of genes, SNPs, drugs, biomarkers and microarray studies involved in CD and UC. This curated database was created by mining published literature, published studies such as the Genome Wide Association Studies (GWAS). The identified genes were then subject to informatics analytics software such as IPA, to reveal their molecular processes. A network pharmacology approach has identified newer targets and drugs that can be repurposed.
Results
We found that CD involved the Th1 pathway, while UC involved the Th2 pathway. Further analysis on gene networks demonstrated that dyslipidemia, epigenetic modulations and dysbiosis play a key role in CD, while inflammation, angiogenesis and potential carcinogenesis processes are involved in UC. Further, this dysbiosis and dyslipidemia in CD is associated with microbiome population change, suggestive of altered levels of micronutrients. Finally, using extended networks based on metabolomics and protein-protein interactions, we were able to suggest a role for synergistic approach in guiding treatment for CD and UC.
Conclusion
We demonstrate using the Systems and Network Medicine approach that CD and UC are distinct diseases with unique sets of genes that define them. We find novel molecular connections that can be explored for therapeutic and clinical purposes.
Keywords : Systems Medicine, Network Medicine, Informatics, Crohn’s, Colitis, IBD n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of BACKGROUND n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of (Uniprot Diseasome) and GDsome (GWAS Diseasome), which leverage the UniProt database and GWAS databases [26, 27], respectively, to mine data the user wants using simple text searches. We hope this will evolve to become a one-stop shop for researchers focused on IBD, CD and UC. We utilized the data in the database to first explore the granular understanding of UC and CD, relating to molecular events and disease progression. Further, we used analytics software Ingenuity Pathway Analysis (IPA) tool to provide a more holistic view of CD and UC. Thirdly, we have applied a novel systems network pharmacology approach to provide a mechanistic view of the diseases and to point to newer drugs and targets. RESULTS Canonical Pathway Analysis of genes from Master List for CD, UC and IBD-Unclassified CD Involves Th1 while UC involves Th2 Activation and Differentiation Pathways
Following the lesson that there exists a significant relationship between activated genes and molecular events that may lead to inflammatory diseases, we analyzed the genes that were involved in CD, UC and IBD-unclassified as noted in our master list (
Supplementary_Table1.xlsx n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of Table 1: Top 5 statistically enriched canonical pathway analysis for CD, UC, IBD-Unclassified identified using IPA. The p-value signifies the statistical significance of the pathways mapped. A p-value of <0.5 is taken as statistically significant.
CD CD p-value UC UC p-value IBD-Unclassified IBD-Unclassified p-value
TREM1 Signaling 1.02E-08 Systemic Lupus Erythematosus in B Cell Signaling Pathway 7.61E-23 Th17 Activation Pathway 8.66E-19 Role of Macrophages, Fibroblasts and Endothelial Cells in Rheumatoid Arthritis 1.34E-08 Role of Pattern Recognition Receptors in Recognition of Bacteria and Viruses 1.08E-19 IL-23 Signaling Pathway 2.41E-18 Role of Cytokines in Mediating Communication between Immune Cells 8.94E-08 Hepatic Cholestasis 2.16E-18 Systemic Lupus Erythematosus In B Cell Signaling Pathway 1.86E-17 Neuroinflammation Signaling Pathway 1.70E-07 Th17 Activation Pathway 5.02E-18 Role of Cytokines in Mediating Communication between Immune Cells 2.28E-17 T Helper Cell Differentiation 4.12E-07 HMGB1 Signaling 1.28E-17 T Helper Cell Differentiation 5.75E-16 n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of Table 2:
Top 5 statistically significant upstream regulators of the top 5 pathways provided in Table 1 for CD, UC, IBD-Unclassified based on IPA. The p-value signifies the statistical significance of the pathways mapped. A p-value of <0.5 is taken as statistically significant. CD CD p-value UC UC p-value
IBD-Unclassified
IBD-Unclassified p-value
NFkB (complex) 2.49E-12 IFNG 7.43E-36 Peptidoglycan 1.86E-19 Lipopolysaccharide 1.25E-11 Lipopolysaccharide 6.44E-31 IFNG 2.03E-19 Glucocorticoid 2.56E-11 TNF 1.60E-29 SOCS3 1.76E-18 Resiquimod 3.75E-10 IL6 9.41E-29 Misoprostol 4.76E-18 TLR4 4.22E-10 IL10 4.85E-28 IL23 5.06E-18 n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of Table 3: Genes from our Master list in the Top Canonical pathways as per IPA for CD and UC
Based on the IPA results, it is clear that CD and UC have distinctive pathways that define them. CD is driven by the Th-1 pathway, and UC by Th-2 pathway and an altered epithelial barrier function [28]. The interplay between innate and adaptive immunity plays a key role in both CD and UC and both are under the cross regulation of these responses. The genes and pathways provided in Table 1 correlate well with what we know from the literature. For example, TREM-1 is an important signaling receptor on Neutrophil or Macrophage leading to chemokine and cytokine induction by interacting with DAP12 [29]. Further, TREM-1 is part of the signaling pathways induced by Pattern Recognition Receptors TLR and NLR. Through the literature we know TLR5 and TLR10 are innate immunity genes identified to be involved in CD, which are part of the TREM1 signaling pathway [29]. Specifically, the binding of flagellin to TLR5 is well known to stimulate an immune response in patients with CD. In addition, ICAM1, IL27, TYK2 all lead to the differentiation of the T cells to Th1. Other genes, for example, Tyk2 and Jak2 are involved in both CD and UC. Tyk2 and Jak2 is required for the differentiation of Th1 cells from naïve T cells [30]. CD28 promotes the production of Th2 differentiation, which then activates the humoral response, making both processes involved in UC and CD. This suggests that there seems to be several different components of inflammation involved with CD and UC and these are regulated by different sets of genes (Table 2).
CD Genes (CD) UC Genes (UC)
TREM1 Signaling ICAM1, IL10, IL18, NOD2, TLR10, TLR5, TNF Systemic Lupus Erythematosus in B Cell Signaling Pathway FCGR2A,IL13,IL2,IL21,IL25,IL33,IL6,IRF5,JAK2,KRAS,NFATC1,NFKB1,PIK3CG,PLCG2,PTPN11,RAC1,RELA,STAT3,TGFB1, TNFSF15 Role of Macrophages, Fibroblasts and Endothelial Cells in Rheumatoid Arthritis FCGR3A/FCGR3B, ICAM1, IL10, IL16, TNF IL18, SOCS3, TLR10, TLR5, TNFRSF1A Role of Pattern Recognition Receptors in Recognition of Bacteria and Viruses CLEC7A, IL13, IL2, IL21, IL25, IL33, IL6, NFKB1, NOD1, PIK3CG, PLCG2, RELA, TGFB1, TLR4, TNFSF15 Role of Cytokines in Mediating Communication between Immune Cells IFNA10, IFNA4, IL10, IL18, IL27 TNF Hepatic Cholestasis ABCB1, ADCY7, CD14, HNF4A, IL13, IL2, IL21, IL25, IL33, IL6, NFKB1, RELA, TGFB1, TLR4, TNF, TNFSF15 Neuroinflammation Signaling Pathway ICAM1, IFNA10, IFNA4, IL10, IL18, TLR10, TLR5, TNFRSF1A, TNF Th17 Activation Pathway CCL20,CCR6,DEFB4A/DEFB4B,IL21,IL22,IL23R,IL6,JAK2,NFATC1,NFKB1,RELA,STAT3,STAT4 T Helper Cell Differentiation IL10, IL18, IL23R, IL2RA, TNFRSF1A, TNF HMGB1 Signaling IL13, IL2, IL21, IL25, IL33, IL6, KRAS, NFKB1, PIK3CG, RAC1, RELA, TGFB1, TLR4, TNF, TNFSF15 n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of Systems and Network Pharmacology approach for identification of newer targets to enable drug-repurposing
Single target symptom based treatment of diseases with one drug for the disease has not proven effective. Here we have applied a more mechanistic network based approach that was recently published in the literature [25]. This approach is an integrative multilayered molecular network. This approach starts with one causal gene target and extends to protein metabolite interacting partners (co-targets). A causal gene (i.e. specific to CD or UC) directly interacts with multiple metabolites. Such metabolites also interact with other proteins. There are situations where such pairs of proteins that involve common metabolites may behave in synergy (guilt by association) and result in pairing synergistic mechanistic disease targets enabling systems medicine based drug repurposing. For CD, Tumor Necrosis Factor (TNF) was chosen as a causal clinical target because of its involvement in the activation of Th1 cells and its central role as a drug target in the treatment of CD. The results for CD show involvement of multiple Histone deacetylases (HDACs), specifically HDAC 2,3,4,5 (Table 4). The role of epigenetics in IBD has been known for decades. CD and UC are more common among certain sections of the population. For example, a study carried out on the geographic origins of Jewish patients with IBD found that IBD genes are more prevalent in those who originated from mid-Europe than those of Polish and Russian origin [31]. Our analysis points to an important role of epigenetics in CD. HDAC3 regulates many genes that are transcriptional targets of PPAR-nuclear receptors, which is also significantly involved as a protein interacting partner in CD (Table 4). HDACs are involved in regulation of immune systems and some have positive influence on antigen-presenting cell (APC), production of IL12 leading to differentiation of Th1. This is in concordance with the involvement of Th1 in CD (Table 1) [32]. While further studying the network associated with CD surrounding TNF, it was revealed that the immediate protein clusters are all involved with Th1 activity consistent with Table 1. The results of the network analysis led to interacting partners that all have a common metabolite and also are currently drug-targets with already approved drugs for other disease conditions, paving the way for drug-repurposing [30, 33]. Additionally, IL16 leads to Th1 mediated inflammation [34]; IL18 in the presence of IL12 promotes Th1 production of IFN γ further driving the Th balance towards Th1 [35]. ATG16L1 is a protein known to have mutant form T300A, which result in abnormal Paneth cells. These altered Paneth cells experienced increased Endoplasmic Reticulum (ER) stress markers, and patients with such abnormal Paneth cells harbored adherent-invasive E Coli [36]. Thus, we can infer that the immediate protein cluster surrounding TNF promotes epigenetic changes, as well as Th1 development, which drives the CD state. We propose that the CD state involves epigenetic changes that lead to dysbiosis and potentially altered levels of micronutrients such as SCFA that affect the colonic environment, which lead to more pro-inflammatory states. For UC, Vascular Endothelial Growth Factor A (VEGFA) was chosen as a causal target, because it is known to be a key mediator of angiogenesis and a promising target for UC. The results for UC showed proteins involved in acute phase reactants (e.g. NPPB, F10) (Table 5) [37, 38]. Heparanase was shown to be involved in UC as well as in a mouse-model of Dextran-induced n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach Page of colitis, where the protein contributed to sensitization of macrophage towards colonic flora leading to a pro-inflammatory state [39]. It is notable that one of a key set of proteins involved in external environment sensing, Dectin-1, is part of the unique UC gene-set that we found ( Supplementary_Table1.xlsx ). The results also show a cluster of proteins connected: TGFB1, FGF2, APOE, or FN1-FGF1; leading to angiogenesis. While FGF2 has a role in repair of intestinal epithelium it is notable that FGF2 can induce angiogenesis in the setting of inflammation leading to disbiosis [40, 41]. The other end of the network involves players such as LCN2, IL1B, VCAM1, or NFkB1which are known genes involved in inflammation. The protein NGAL encoded by LCN2 is shown to be closely correlated with an inflamed IBD state, suggesting that the UC network involves active inflammation derived from VEGFA [42]. Additionally, the UC network associated with VEGFA demonstrates involvement of PTPN11, SOCS3, genes involved in inflammation. Aberrant expressions of SHP2 (encoded by PTPN11) and SOCS3 expression in multiple disease models led to carcinogenesis by induction of STAT3/IL6 pathways including ulcerative colitis [43-45]. The UC disease network also contains a cluster that involves RAC1, TLR4, APRC2, and PLCG2. Many of these proteins are surface related proteins involved in inflammatory response [46]. Specifically, RAC1 has the potential to be a therapeutic biomarker for thiopurine [47] and ARPC2 is seen as a contributory locus to UC susceptibility [48]. Taken together, this suggests a disease model of UC: antigens and colonic flora cause the initial inflammatory step, sensed by ECM receptors as Dectin-1 which then release proinflammatory cytokines as IL1B. This leads to activation of macrophage, which promote more inflammatory states through heparinase activity. IL1B is a downstream effector of the Dectin-1 pathway [49]. Additionally, this inflammation results in angiogenesis and potential carcinogenesis. The results of the network pharmacology approach have identified a list of targets and drugs that work synergistically (Table 4, Table 5). This opens doors to a combination therapy approach, especially because of the complete ineffectiveness of immunotherapy and other drugs currently used in the clinic [50]. Our data suggests that anti-TNF therapy is effective for CD but not for UC. In fact, some adverse effects have been seen in UC patients treated with anti-TNF antibodies [51]. For example, the combination therapy for CD based on our analysis, might be using anti-TNF (adalimumab) and Pioglitazone in synergy and anti-VEGFA antibodies and Gallium Nitrate (targeting IL1B) for UC. Pioglitazone targets PPAR, have shown to be effective in in-vivo mouse studies on colitis models [52, 53]. Gallium Nitrate (targeting IL1B), may be a potential therapeutic agent in UC. n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of Table 4: CD-associated network proteins ranked according to their connectedness to TNF via their metabolites. Results of Semantic-similarity ranking based on Molecular Functions (MF SemSem. An R value cutoff of 0.6 was used. Many of these genes are epigenetic modulators.
CD-associated Genes Protein Name R values DRUGS DISEASE
TNF Tumor necrosis factor (Causal Gene) 1 Adalimumab Rheumatoid arthritis; Intermittent claudication; HDAC5 Histone deacetylase 5 0.712 PMID29671355- Compound-62 NA HDAC4 Histone deacetylase 4 0.699 Quisinostat Huntington disease; ALPL Alkaline phosphatase, tissue-nonspecific isozyme 0.645 Asfotase alfa Genetic disease; Parasitic infection; AHCY Adenosylhomocysteinase 0.645 Neplanocin A NA HDAC3 Histone deacetylase 3 0.624 CHR-3996 Solid tumour/cancer; NGF Beta-nerve growth factor 0.623 Tanezumab Alzheimer disease; Chronic pain; HDAC2 Histone deacetylase 2 0.608 CHR-3996 Dermatitis; Solid tumour/cancer; PPARG Peroxisome proliferator-activated receptor gamma 0.6 Pioglitazone Type-2 diabetes; Hepatosteatosis n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of Table 5: UC-associated network proteins ranked according to their connectedness to VEGFA via their metabolites. Results of Semantic-similarity ranking based on Molecular Functions (MF SemSem. An R value cutoff of 0.61 was used. Many of these genes are acute-phase reactants, as well as pro-inflammatory molecules.
UC- associated Gene Name R Values DRUGS DISEASE
VEGFA Vascular endothelial growth factor A 1 Ranibizumab Solid tumour/cancer; FN1 Fibronectin 0.736 Ocriplasmin Colorectal cancer; Symptomatic vitreomacular adhesion IL1B Interleukin-1 beta 0.716 Gallium nitrate Immune System disease; Hypercalcaemia; NPPB Natriuretic peptides B 0.693 BD-NP Hypertension APP Amyloid beta A4 protein 0.691 Florbetapir F-18 Alzheimer disease; Cerebral amyloid angiopathy ANG Angiogenin 0.674 Citric Acid APOB Apolipoprotein B-100 0.659 SPC4955 High blood cholesterol level FGF1 Heparin-binding growth factor 1 0.657 CVBT-141H Coronary artery disease; Peripheral arterial disease SERPINC1 Antithrombin-III 0.648 Enoxaparin Venous thrombosis; Thrombosis APOE Apolipoprotein E 0.641 AEM-28 Hyperlipidaemia; Familial hypercholesterolemia; FGF2 Heparin-binding growth factor 2 0.633 Sucralfate Neurodegenerative disorder; Diabetic retinopathy; HPSE Heparanase 0.625 PG-545 Angiogenesis disorder; Ocular disease; ADRA1A Alpha-1A adrenergic receptor 0.622 Xatral Major depressive disorder; Glaucoma/ocular hypertension; VCAM1 Vascular cell adhesion protein 1 0.615 Carvedilol Blood pressure; heart failure F10 Coagulation factor X 0.611 Apixaban Thrombosis; Hemophilia n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of Protein-Protein Interacting networks: Putting it all together
Combining data from our master list and the results of our network pharmacology approach (Tables 1, 2, 3, 4, and 5) we created a combined network to further understand the mechanism in CD and UC (Figures 1a and 1b).
Figure 1a: Protein-protein interaction network of all identified genes for CD using String-DB [54]. Genes identified by our network pharmacology approach is starred (*Druggable targets). Colored nodes indicate first shell interactions. Edges represent protein-protein associations. Blue indicates that the information is from curated databases; pink experimentally determined; Green is gene neighborhood; dark blue represents gene co-occurrence, yellow is from text mining; black represents co-expression. Proteins with no interactions are not included in the figure for clarity. n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of Figure 1b: Protein-protein interaction network of all identified genes for UC using String-DB [54]. Genes identified by our network pharmacology approach is starred (*Druggable targets). Colored nodes indicate first shell interactions. Edges represent protein-protein associations. Blue indicates that the information is from curated databases; pink experimentally determined; Green is gene neighborhood; dark blue represents gene co-occurrence, yellow is from text mining; black represents co-expression. Proteins with no interactions are not included in the figure for clarity.
Altered microbiome of intestine may be associated with altered micronutrient in the gut
The Enteric environment becomes significantly altered in the aforementioned manner in the context of IBD, leading to altered microbiome when compared to the healthy counterparts, especially with CD. We were hoping to characterize this dysbiosis, in hopes of identifying more changes that could explain the pro-inflammatory states seen in CD and UC. Disbiome is a database that houses microbial composition changes in 354 different diseases [55]. We extracted fecal microbiome as a surrogate for intestinal microbiome; and from this we studied the intestinal microbiome of CD and UC. The significantly changed microbiome species and their relative elevation or reduction from the control has been summarized (Table 6). n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of Table 6:
Fecal microbiota for CD and UC from Disbiome database. “Elevated” refers to species elevated compared to control groups. “Reduced” refers to species reduced compared to control groups.
CD_Elevated
Species
Ruminococcus gnavus Clostridium cluster XIVa Dipodascus Lactobacillales Bifidobacterium Entyloma Bacteroides-Prevotella group Actinomyces Candida albicans Enterococcus Akkermansia Municiphila
CD_Reduced
Clostridium leptum Bacteroides Subdoligranulum Parabacteroides Citrobacter Bacteriodes uniformis Enterobacter Lactobacillus Prevotella Eubacterium Faecalibacterium
UC_Elevated
Dipodascus Lactobacillales Entyloma Wallemia
UC_Reduced
Roseburia hominis Lactobacillus Hanseniaspora Blautia coccoides Bifidobacterium Faecalibacterium Prausnitzii Clostridium leptum Faecalibacterium Bacteroides n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of There were some taxa that were elevated in CD while reduced in UC (e.g. bifidobacterium ), suggesting there may be disease-state specific changes conducive to such population change. This may be associated with the difference in the activated disease proteins and pathways, e.g. activation of macrophage and altered lipidome in CD vs heparinase in UC. Also,
Faecalibacterium, associated with healthy state microbiome, were decreased in both CD and UC states (Table 6). On the contrary,
Ruminococcus gnavus is often found in IBD, associated with inflammatory states and shown in-vitro to promote inflammation [56]. This organism was found to be elevated in CD, but not in UC (Table 6). Other literature reported that CD seemed to experience more dysbiosis and unstable microbiome composition than UC – our data shows CD seems to have more species with altered composition than UC. Further in-depth analysis is needed before fully understanding the specific roles of species in disease states.
Discussion
Our integrated informatics approach sheds light on the significant differences between CD and UC, as well as involvement of intestinal dysbiosis that may contribute to disease states. This analysis derived a potential mechanistic view of the CD and UC states, as well as of multi-drug therapeutic regimens. As far as we know, this is one of the first works to suggest a polypharmacological therapy approach to both CD and UC. The drug therapy suggestions were in concordance with prior published works, as our therapy suggestions are based on the drugs that showed efficacy in animal trials of colitis. The dysbiosis are also in concordance with other prior works, and offer an insight into how dysbiosis is playing a role in the disease state as well as how the disease state affects the microbiome. We found that CD involved the Th1 pathway, while UC involved the Th2 pathway. Further analysis on the metabolite-shared protein-protein network led us to demonstrate that dyslipidemia and dysbiosis may play a key role in CD, while inflammation, angiogenesis, and potential carcinogenesis processes are involved in UC.
Notably our data on CD sheds light on the potential involvement of altered micronutrient level in the gut environment. HDAC (as shown to be involved in CD from out Network Pharmacology approach) is also involved with changes in lipid metabolism in the gut environment. HDAC3 deletion in the mouse model led to increased mitochondrial and peroxisomal b-oxdation, fatty acid oxidation, and reduction in long-chain fatty acids [57]. This led to regulation of lipid metabolism in intestinal epithelium, resulting in remodeling of lipidome of intestinal enterocytes resulting in protection from diet-induced obesity. The same histone deacetylatses involved in lipid dysbiosis are also affected by the Short Chain Fatty Acids (SCFA) generated by
Akkermansia Municiphila, or sulforaphane cysteine by
Bacteroides thetaiotaomicron [58, 59]. This suggests that HDAC may be involved in a feedback loop to itself by regulating the host enteric environment.
For UC, it is interesting to emphasize the involvement of IL1B in its disease state. Dectin-1 is activated through fungi/mycobacterial pathogens leading to transcription of IL1B, which is important in Th17 differentiation [49]. This is consistent with our protein disease network analysis that showed involvement of IL1B. Furthermore, the finding regarding UC is in-line with our conventional knowledge of UC, as we see significant association of UC with cancerous events, e.g. colonic cancer [60]. This is consistent with our finding that UC is associated molecularly with angiogenic events. n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of IBD-Omics, the database introduced here, contains literature validated data for genes, proteins, SNPs and miRNA from various sources as GWAS, relevant GEO datasets, drugs and current clinical trials, and other pharmacological databases. Marrying such data with analytic algorithms allowed molecular understanding of CD and UC. We hope that this approach further advances the efforts to utilize the Systems and Network approach for understanding diseases. This also clearly has demonstrated, via examples, how layered approaches are needed in delineating CD and UC and has led to the identifications of newer targets. Furthermore, as a community, we must focus the effort on collecting disparate types of information, e.g. drug information associated with their molecular targets, more verified and focused molecular processes, and cast a wider net of gene lists to increase our knowledge for understanding these diseases. Such an approach made possible the probable polypharmaceutical regimen presented here, as well as the understanding into the dysbiosis that IBD may cause. Such an approach would facilitate the identification of new therapeutic targets for IBD and the understanding of the molecular basis of these targets.
Materials and Methods
Literature Mining : Genes, miRNA, SNPs and biomarkers implicated in UC, CD and IBD were mined from literature using the text-mining tool Linguimatics ( ) and direct searches using Pubmed ( . Curated databases like DisGenet, Uniprot and NCBI-Refseq were used to filter the huge amount of articles retrieved. Each of the genes and its sources thus obtained was reviewed manually to ensure the validity of the gene’s involvement in the disease. Table 4 gives a total count of articles retrieved and the results of curation of the articles over the years 1966-2019.
Table 7: Total count of Articles curated for Genes, biomarkers, SNPs and miRNA in IBD, CD, UC
Search terms used IBD CD UC
Genes 3060 972 526 Biomarkers 191 59 48 SNP 1470 1094 531 miRNA 189 82 75
Curated Information
Genes 67 31 61 Biomarkers 24 15 18 SNP 24 91 87 miRNA 22 39 37 The process of curation involved reading each article to ascertain its relevance and role in IBD, CD and UC, respectively, as proven by experimentation. It must be pointed out that during the pre-genomic era, CD and UC were simply referred to as IBD. The distinction between them was adapted in the literature only in the last decade after the completion of the human project and the availability of data from other projects. n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of GWAS Catalog
Master List : A master list of Genes was created by combining the list of genes obtained from the manually curated set derived from the literature, Curated Genes from DisGenet database [63] and those obtained from GWAS. The duplicate genes between the different data sorces were removed from the master list. From this master list, a unique set of genes for CD and UC was obtained for disease specific analysis. Only genes that mapped to a Uniprot Accession were retained. The master list of genes is in
Supplementary_Table1.xlsx . A total of 50 genes were obtained for CD, 83 for UC and 57 genes for IBD. The IBD genes are the common set of genes between CD and UC. Some of the genes could give us clues about IBD-indeterminate genes. However, identification of IBD-indeterminate gene analysis is beyond the scope of this paper. IPA was used to map the genes in our master list to pathways. IPA provides detailed mapping to top canonical pathways and provides upstream and downstream regulators.
Protein-Protein Interactions Network : String database was used to create protein-protein networks [54]. n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of Systems Medicine and Network Pharmacology Analysis workflow:
Figure 2: Workflow used for target identification using Systems and Network pharmacology approach. The pipeline is adapted from Ana I, Casas et al. [25]. Briefly, the pipeline consists of 3 interdependent layers of data extraction, the Protein-Metabolite Network layer, Protein-Protein Network layer and a Semantic Similarity layer. The Protein-Metabolite layer extracts metabolites interacting with the causal targets (TNF for CD and VEGFA for UC) from the Human Metabolome Database (HMDB). For each of the metabolites, proteins interacting with them is extracted and “The Integrated Interaction Database” (IID) is used to extract protein-protein interactions. The Therapeutic Target Database (TTD) is then used to filter the extracted proteins based on the availability of drugs. The proteins thus extracted that have drugs serves as the input for the Semantic Similarity layer, which calculates ontology-based semantic similarity scores using molecular function (MF) annotations. It provides a ranking based on the similarity scores [64, 65]. n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of IBD-OMICS DATABASE CONTENTS
The HOME tab links to CCFA (The Crohn’s and Colitis Foundation), the widely used resource for Crohn’s and Colitis. Within each of the tabs, IBD, Crohn’s and Ulcerative Colitis, we have the following information: All of the information collected for each of the following tabs is based on data mined and validated from the literature. • The Genes tab contains gene centric information. The major sources of this data are from NCBI gene database and GWAS. Under the GWAS tab there are two sub-tabs that provide a network view of other diseases sharing a common SNP or a common gene with the disease (IBD, CD or UC). • The protein tab provides protein centric information from Uniprot database ( . • The SNPs tab provides all the curated SNPs from dbsnp database from NCBI ( ). • miRNA tab provides microRNAs which were filtered after literature analysis. These miRNAs were further investigated using mirBASE ( ) and TarBASE databases. http://carolina.imis.athenainnovation.gr/diana_tools/web/index.php?r=tarbasev8%2Fi dex • The Microarrays tab provides expression analysis from GEO database. • Biomarker Tab provides literature validated biomarkers. • Literature Tab provides a direct link to literature articles relating to IBD, UC or CD from Pubmed and Medline • The GDSome (GWAS-DiseaSome) tab is a tool that leverages the GWAS database by allowing users to search for SNP (rs identifier), Gene, or disease of interest—returning all diseases related to the search terms, either by common gene or common SNP. For instance, searching the GWAS Diseasome for the disease, Crohn’s Disease, by common gene, we return 186 results, 62 unique diseases/traits associated by gene. This can then be used to create disease networks to better visualize how certain diseases correspond to one another. • The UDSome (Uniprot-DiseaSome) tab is a tool that leverages the Uniprot database by allowing users to search for SNP (rs identifier), Gene, or disease of interest—returning all diseases related to the search terms, either by common gene or common SNP. For instance, searching the UDSome for Crohn’s Disease, by common gene, we return 186 results.
Both UDSome and GDSome enable us to see disease associations by mining the databases in their entirety that is otherwise not possible. n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of Acknowledgements
We thank Ritika Kundra who, as a graduate student at Georgetown University, contributed to data mining for setting up IBD-OMICS. We thank Sarma Dittakavi, Professor Emeritus, Laboratory Medicine and Pathobiology, University of Toronto, and Rajaram Gana, Georgetown University, for the many stimulating discussions that we have had together. We thank Elliott Crooke, Professor and Chair, Department of Biochemistry and Molecular & Cellular Biology, and Senior Associate Dean, Faculty and Academic Affairs, Georgetown University, for his support and encouragement. We thank Limelight Services LLC, Toronto, Canada, for building the IBDOmics database, which has not yet been made public. This work is dedicated to my sister, the late Shanthi Sitaraman, who was a physician-scientist at Emory School of Medicine.
Funding : No funding was received for this work.
Authors’ contributions
Sona Vasudevan conceived the idea, validated the data, drove the analyses and discussions, and wrote the manuscript. Takashi Kitani contributed to pathway analysis, and several important discussions of the manuscript. Sushma Maddipatla carried out the Network Pharmacology approach analysis. Ramya Madupuri developed the GDSome and UDsome tools and helped with data uploads to the IBDOmics Website. Christopher Greco carried out curation of the literature data. Jonathan Hartman provided text mining data using the Linguamatics tool. James Baraniuk read the manuscript and participated in some discussions.
Ethics approval and consent to participate
Not applicable. This study does not utilize any data that requires ethical approval or consent to participate.
Consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests. n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of REFERENCES [1] Flynn S, Eisenstein S. Inflammatory Bowel Disease Presentation and Diagnosis. Surg Clin North Am 2019;99(6):1051-62. [2] Mahadevan U, Silverberg MS. Inflammatory Bowel Disease-Gastroenterology Diamond Jubilee Review. Gastroenterology 2018;154(6):1555-8. [3] Malik TA. Inflammatory Bowel Disease: Historical Perspective, Epidemiology, and Risk Factors. Surg Clin North Am 2015;95(6):1105-22, v. [4] Nanki K, Mizuno S, Naganuma M, Kanai T. [Inflammatory bowel disease and the intestinal microbiota]. Nihon Shokakibyo Gakkai Zasshi 2015;112(11):1947-55. [5] Coskun M. Intestinal epithelium in inflammatory bowel disease. Front Med (Lausanne) 2014;1:24. [6] Arnold GL, Beaves MR, Pryjdun VO, Mook WJ. Preliminary study of ciprofloxacin in active Crohn's disease. Inflamm Bowel Dis 2002;8(1):10-5. [7] Call ME, Wucherpfennig KW. Molecular mechanisms for the assembly of the T cell receptor-CD3 complex. Mol Immunol 2004;40(18):1295-305. [8] Damiao A, de Azevedo MFC, Carlos AS, Wada MY, Silva TVM, Feitosa FC. Conventional therapy for moderate to severe inflammatory bowel disease: A systematic literature review. World J Gastroenterol 2019;25(9):1142-57. [9] Jigaranu AO, Nedelciuc O, Blaj A, Badea M, Mihai C, Diculescu M, Cijevschi-Prelipcean C. Is rifaximin effective in maintaining remission in Crohn's disease? Dig Dis 2014;32(4):378-83. [10] Kolls JK, Zhang Z. Anti-interleukin-12 antibody for active Crohn's disease. N Engl J Med 2005;352(6):627-8; author reply -8. [11] Liao W, Lin JX, Leonard WJ. IL-2 family cytokines: new insights into the complex roles of IL-2 as a broad regulator of T helper cell differentiation. Curr Opin Immunol 2011;23(5):598-604. [12] MacDermott RP. Progress in understanding the mechanisms of action of 5-aminosalicylic acid. Am J Gastroenterol 2000;95(12):3343-5. [13] Shafran I, Burgunder P. Adjunctive antibiotic therapy with rifaximin may help reduce Crohn's disease activity. Dig Dis Sci 2010;55(4):1079-84. [14] Arruda JM, Bogetz AL, Vellanki S, Wren A, Yeh AM. Yoga as adjunct therapy for adolescents with inflammatory bowel disease: A pilot clinical trial. Complement Ther Med 2018;41:99-104. [15] Eckert KG, Abbasi-Neureither I, Koppel M, Huber G. Structured physical activity interventions as a complementary therapy for patients with inflammatory bowel disease - a scoping review and practical implications. BMC Gastroenterol 2019;19(1):115. [16] Ewais T, Begun J, Kenny M, Rickett K, Hay K, Ajilchi B, Kisely S. A systematic review and meta-analysis of mindfulness based interventions and yoga in inflammatory bowel disease. J Psychosom Res 2019;116:44-53. [17] Tysk C, Lindberg E, Jarnerot G, Floderus-Myrhed B. Ulcerative colitis and Crohn's disease in an unselected population of monozygotic and dizygotic twins. A study of heritability and the influence of smoking. Gut 1988;29(7):990-6. [18] Thompson NP, Driscoll R, Pounder RE, Wakefield AJ. Genetics versus environment in inflammatory bowel disease: results of a British twin study. BMJ 1996;312(7023):95-6. [19] Orholm M, Binder V, Sorensen TI, Rasmussen LP, Kyvik KO. Concordance of inflammatory bowel disease among Danish twins. Results of a nationwide study. Scand J Gastroenterol 2000;35(10):1075-81. [20] Halme L, Paavola-Sakki P, Turunen U, Lappalainen M, Farkkila M, Kontula K. Family and twin studies in inflammatory bowel disease. World J Gastroenterol 2006;12(23):3668-72. n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach
Page of [21] Halfvarson J, Brislawn CJ, Lamendella R, Vazquez-Baeza Y, Walters WA, Bramer LM, D'Amato M, Bonfiglio F, McDonald D, Gonzalez A, McClure EE, Dunklebarger MF, Knight R, Jansson JK. Dynamics of the human gut microbiome in inflammatory bowel disease. Nat Microbiol 2017;2:17004. [22] Neurath MF. Host-microbiota interactions in inflammatory bowel disease. Nat Rev Gastroenterol Hepatol 2020;17(2):76-7. [23] Sheehan D, Shanahan F. The Gut Microbiota in Inflammatory Bowel Disease. Gastroenterol Clin North Am 2017;46(1):143-54. [24] Jostins L, Ripke S, Weersma RK, Duerr RH, McGovern DP, Hui KY, Lee JC, Schumm LP, Sharma Y, Anderson CA, Essers J, Mitrovic M, Ning K, Cleynen I, Theatre E, Spain SL, Raychaudhuri S, Goyette P, Wei Z, Abraham C, Achkar JP, Ahmad T, Amininejad L, Ananthakrishnan AN, Andersen V, Andrews JM, Baidoo L, Balschun T, Bampton PA, Bitton A, Boucher G, Brand S, Buning C, Cohain A, Cichon S, D'Amato M, De Jong D, Devaney KL, Dubinsky M, Edwards C, Ellinghaus D, Ferguson LR, Franchimont D, Fransen K, Gearry R, Georges M, Gieger C, Glas J, Haritunians T, Hart A, Hawkey C, Hedl M, Hu X, Karlsen TH, Kupcinskas L, Kugathasan S, Latiano A, Laukens D, Lawrance IC, Lees CW, Louis E, Mahy G, Mansfield J, Morgan AR, Mowat C, Newman W, Palmieri O, Ponsioen CY, Potocnik U, Prescott NJ, Regueiro M, Rotter JI, Russell RK, Sanderson JD, Sans M, Satsangi J, Schreiber S, Simms LA, Sventoraityte J, Targan SR, Taylor KD, Tremelling M, Verspaget HW, De Vos M, Wijmenga C, Wilson DC, Winkelmann J, Xavier RJ, Zeissig S, Zhang B, Zhang CK, Zhao H, International IBDGC, Silverberg MS, Annese V, Hakonarson H, Brant SR, Radford-Smith G, Mathew CG, Rioux JD, Schadt EE, Daly MJ, Franke A, Parkes M, Vermeire S, Barrett JC, Cho JH. Host-microbe interactions have shaped the genetic architecture of inflammatory bowel disease. Nature 2012;491(7422):119-24. [25] Casas AI, Hassan AA, Larsen SJ, Gomez-Rangel V, Elbatreek M, Kleikers PWM, Guney E, Egea J, Lopez MG, Baumbach J, Schmidt H. From single drug targets to synergistic network pharmacology in ischemic stroke. Proc Natl Acad Sci U S A 2019;116(14):7129-36. [26] Buniello A, MacArthur JAL, Cerezo M, Harris LW, Hayhurst J, Malangone C, McMahon A, Morales J, Mountjoy E, Sollis E, Suveges D, Vrousgou O, Whetzel PL, Amode R, Guillen JA, Riat HS, Trevanion SJ, Hall P, Junkins H, Flicek P, Burdett T, Hindorff LA, Cunningham F, Parkinson H. The NHGRI-EBI GWAS Catalog of published genome-wide association studies, targeted arrays and summary statistics 2019. Nucleic Acids Res 2019;47(D1):D1005-D12. [27] The UniProt C. UniProt: the universal protein knowledgebase. Nucleic Acids Res 2017;45(D1):D158-D69. [28] Oh DY, Koh SJ. Cross-regulation of innate and adaptive immunity: a new perspective for the pathogenesis of inflammatory bowel disease. Gut Liver 2015;9(3):263-4. [29] Arts RJ, Joosten LA, van der Meer JW, Netea MG. TREM-1: intracellular signaling pathways and interaction with pattern recognition receptors. J Leukoc Biol 2013;93(2):209-15. [30] Ghoreschi K, Laurence A, O'Shea JJ. Janus kinases in immune cell signaling. Immunol Rev 2009;228(1):273-87. [31] Roth MP, Petersen GM, McElree C, Feldman E, Rotter JI. Geographic origins of Jewish patients with inflammatory bowel disease. Gastroenterology 1989;97(4):900-4. [32] Shakespear MR, Halili MA, Irvine KM, Fairlie DP, Sweet MJ. Histone deacetylases as regulators of inflammation and immunity. Trends Immunol 2011;32(7):335-43. [33] Cao Y, Doodes PD, Glant TT, Finnegan A. IL-27 induces a Th1 immune response and susceptibility to experimental arthritis. J Immunol 2008;180(2):922-30. [34] Skundric DS, Cai J, Cruikshank WW, Gveric D. Production of IL-16 correlates with CD4+ Th1 inflammation and phosphorylation of axonal cytoskeleton in multiple sclerosis lesions. J Neuroinflammation 2006;3:13. n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach Page of [35] Nakanishi K. Unique Action of Interleukin-18 on T Cells and Other Immune Cells. Front Immunol 2018;9:763. [36] Deuring JJ, Fuhler GM, Konstantinov SR, Peppelenbosch MP, Kuipers EJ, de Haar C, van der Woude CJ. Genomic ATG16L1 risk allele-restricted Paneth cell ER stress in quiescent Crohn's disease. Gut 2014;63(7):1081-91. [37] Gabay C, Kushner I. Acute-phase proteins and other systemic responses to inflammation. N Engl J Med 1999;340(6):448-54. [38] Yagci R, Totan Y, Ozluk E, Ozyurt H, Akbay G, Eksioglu M. Serum levels of natriuretic peptides in patients with Behcet's disease. Clin Rheumatol 2008;27(9):1153-8. [39] Lerner I, Hermano E, Zcharia E, Rodkin D, Bulvik R, Doviner V, Rubinstein AM, Ishai-Michaeli R, Atzmon R, Sherman Y, Meirovitz A, Peretz T, Vlodavsky I, Elkin M. Heparanase powers a chronic inflammatory circuit that promotes colitis-associated tumorigenesis in mice. J Clin Invest 2011;121(5):1709-21. [40] Presta M, Andres G, Leali D, Dell'Era P, Ronca R. Inflammatory cells and chemokines sustain FGF2-induced angiogenesis. Eur Cytokine Netw 2009;20(2):39-50. [41] Song X, Dai D, He X, Zhu S, Yao Y, Gao H, Wang J, Qu F, Qiu J, Wang H, Li X, Shen N, Qian Y. Growth Factor FGF2 Cooperates with Interleukin-17 to Repair Intestinal Epithelial Damage. Immunity 2015;43(3):488-501. [42] Oikonomou KA, Kapsoritakis AN, Theodoridou C, Karangelis D, Germenis A, Stefanidis I, Potamianos SP. Neutrophil gelatinase-associated lipocalin (NGAL) in inflammatory bowel disease: association with pathophysiology of inflammation, established markers, and disease activity. J Gastroenterol 2012;47(5):519-30. [43] Chen YY, Ma ZB, Xu HY, Shi LJ, Li DY, Sun LY, Yin XH, Sang GY, Xu D, Tang YH, Wang X, Li P, Wu F, Zhou J. IL-6/STAT3/SOCS3 signaling pathway playing a regulatory role in ulcerative colitis carcinogenesis. Int J Clin Exp Med 2015;8(8):12009-17. [44] Li C, Li H, Zhang P, Yu LJ, Huang TM, Song X, Kong QY, Dong JL, Li PN, Liu J. SHP2, SOCS3 and PIAS3 Expression Patterns in Medulloblastomas: Relevance to STAT3 Activation and Resveratrol-Suppressed STAT3 Signaling. Nutrients 2016;9(1). [45] Li Y, de Haar C, Chen M, Deuring J, Gerrits MM, Smits R, Xia B, Kuipers EJ, van der Woude CJ. Disease-related expression of the IL6/STAT3/SOCS3 signalling pathway in ulcerative colitis and ulcerative colitis-related carcinogenesis. Gut 2010;59(2):227-35. [46] Uhlig HH, Muise AM. Clinical Genomics in Inflammatory Bowel Disease. Trends Genet 2017;33(9):629-41. [47] Seinen ML, van Nieuw Amerongen GP, de Boer NK, Mulder CJ, van Bezu J, van Bodegraven AA. Rac1 as a Potential Pharmacodynamic Biomarker for Thiopurine Therapy in Inflammatory Bowel Disease. Ther Drug Monit 2016;38(5):621-7. [48] Franke A, Balschun T, Karlsen TH, Sventoraityte J, Nikolaus S, Mayr G, Domingues FS, Albrecht M, Nothnagel M, Ellinghaus D, Sina C, Onnie CM, Weersma RK, Stokkers PC, Wijmenga C, Gazouli M, Strachan D, McArdle WL, Vermeire S, Rutgeerts P, Rosenstiel P, Krawczak M, Vatn MH, group Is, Mathew CG, Schreiber S. Sequence variants in IL10, ARPC2 and multiple other loci contribute to ulcerative colitis susceptibility. Nat Genet 2008;40(11):1319-23. [49] Gringhuis SI, Kaptein TM, Wevers BA, Theelen B, van der Vlist M, Boekhout T, Geijtenbeek TB. Dectin-1 is an extracellular pathogen sensor for the induction and processing of IL-1beta via a noncanonical caspase-8 inflammasome. Nat Immunol 2012;13(3):246-54. [50] Targan SR, Hanauer SB, van Deventer SJ, Mayer L, Present DH, Braakman T, DeWoody KL, Schaible TF, Rutgeerts PJ. A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor alpha for Crohn's disease. Crohn's Disease cA2 Study Group. N Engl J Med 1997;337(15):1029-35. n Search of Newer Targets for IBD: A Systems and a Network Medicine Approach Page of [51] Ierardi E, Valle ND, Nacchiero MC, De Francesco V, Stoppino G, Panella C. Onset of liver damage after a single administration of infliximab in a patient with refractory ulcerative colitis. Clin Drug Investig 2006;26(11):673-6. [52] Byrav DSP, Medhi B, Prakash A, Chakrabarti A, Vaiphei K, Khanduja KL. Comparative evaluation of different doses of PPAR-gamma agonist alone and in combination with sulfasalazine in experimentally induced inflammatory bowel disease in rats. Pharmacol Rep 2013;65(4):951-9. [53] Da Silva S, Keita AV, Mohlin S, Pahlman S, Theodorou V, Pahlman I, Mattson JP, Soderholm JD. A Novel Topical PPARgamma Agonist Induces PPARgamma Activity in Ulcerative Colitis Mucosa and Prevents and Reverses Inflammation in Induced Colitis Models. Inflamm Bowel Dis 2018;24(4):792-805. [54] Szklarczyk D, Gable AL, Lyon D, Junge A, Wyder S, Huerta-Cepas J, Simonovic M, Doncheva NT, Morris JH, Bork P, Jensen LJ, Mering CV. STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res 2019;47(D1):D607-D13. [55] Janssens Y, Nielandt J, Bronselaer A, Debunne N, Verbeke F, Wynendaele E, Van Immerseel F, Vandewynckel YP, De Tre G, De Spiegeleer B. Disbiome database: linking the microbiome to disease. BMC Microbiol 2018;18(1):50. [56] Parada Venegas D, De la Fuente MK, Landskron G, Gonzalez MJ, Quera R, Dijkstra G, Harmsen HJM, Faber KN, Hermoso MA. Short Chain Fatty Acids (SCFAs)-Mediated Gut Epithelial and Immune Regulation and Its Relevance for Inflammatory Bowel Diseases. Front Immunol 2019;10:277. [57] Davalos-Salas M, Montgomery MK, Reehorst CM, Nightingale R, Ng I, Anderton H, Al-Obaidi S, Lesmana A, Scott CM, Ioannidis P, Kalra H, Keerthikumar S, Togel L, Rigopoulos A, Gong SJ, Williams DS, Yoganantharaja P, Bell-Anderson K, Mathivanan S, Gibert Y, Hiebert S, Scott AM, Watt MJ, Mariadason JM. Deletion of intestinal Hdac3 remodels the lipidome of enterocytes and protects mice from diet-induced obesity. Nat Commun 2019;10(1):5291. [58] Bhat MI, Kapila R. Dietary metabolites derived from gut microbiota: critical modulators of epigenetic changes in mammals. Nutr Rev 2017;75(5):374-89. [59] Lukovac S, Belzer C, Pellis L, Keijser BJ, de Vos WM, Montijn RC, Roeselers G. Differential modulation by Akkermansia muciniphila and Faecalibacterium prausnitzii of host peripheral lipid metabolism and histone acetylation in mouse gut organoids. mBio 2014;5(4). [60] Levin B. Inflammatory bowel disease and colon cancer. Cancer 1992;70(5 Suppl):1313-6. [61] Li MJ, Liu Z, Wang P, Wong MP, Nelson MR, Kocher JP, Yeager M, Sham PC, Chanock SJ, Xia Z, Wang J. GWASdb v2: an update database for human genetic variants identified by genome-wide association studies. Nucleic Acids Res 2016;44(D1):D869-76. [62] MacArthur J, Bowler E, Cerezo M, Gil L, Hall P, Hastings E, Junkins H, McMahon A, Milano A, Morales J, Pendlington ZM, Welter D, Burdett T, Hindorff L, Flicek P, Cunningham F, Parkinson H. The new NHGRI-EBI Catalog of published genome-wide association studies (GWAS Catalog). Nucleic Acids Res 2017;45(D1):D896-D901. [63] Pinero J, Ramirez-Anguita JM, Sauch-Pitarch J, Ronzano F, Centeno E, Sanz F, Furlong LI. The DisGeNET knowledge platform for disease genomics: 2019 update. Nucleic Acids Res 2020;48(D1):D845-D55. [64] Pesquita C, Faria D, Bastos H, Ferreira AE, Falcao AO, Couto FM. Metrics for GO based protein semantic similarity: a systematic evaluation. BMC Bioinformatics 2008;9 Suppl 5:S4. [65] Wang JZ, Du Z, Payattakool R, Yu PS, Chen CF. A new method to measure the semantic similarity of GO terms. Bioinformatics 2007;23(10):1274-81.[51] Ierardi E, Valle ND, Nacchiero MC, De Francesco V, Stoppino G, Panella C. Onset of liver damage after a single administration of infliximab in a patient with refractory ulcerative colitis. Clin Drug Investig 2006;26(11):673-6. [52] Byrav DSP, Medhi B, Prakash A, Chakrabarti A, Vaiphei K, Khanduja KL. Comparative evaluation of different doses of PPAR-gamma agonist alone and in combination with sulfasalazine in experimentally induced inflammatory bowel disease in rats. Pharmacol Rep 2013;65(4):951-9. [53] Da Silva S, Keita AV, Mohlin S, Pahlman S, Theodorou V, Pahlman I, Mattson JP, Soderholm JD. A Novel Topical PPARgamma Agonist Induces PPARgamma Activity in Ulcerative Colitis Mucosa and Prevents and Reverses Inflammation in Induced Colitis Models. Inflamm Bowel Dis 2018;24(4):792-805. [54] Szklarczyk D, Gable AL, Lyon D, Junge A, Wyder S, Huerta-Cepas J, Simonovic M, Doncheva NT, Morris JH, Bork P, Jensen LJ, Mering CV. STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res 2019;47(D1):D607-D13. [55] Janssens Y, Nielandt J, Bronselaer A, Debunne N, Verbeke F, Wynendaele E, Van Immerseel F, Vandewynckel YP, De Tre G, De Spiegeleer B. Disbiome database: linking the microbiome to disease. BMC Microbiol 2018;18(1):50. [56] Parada Venegas D, De la Fuente MK, Landskron G, Gonzalez MJ, Quera R, Dijkstra G, Harmsen HJM, Faber KN, Hermoso MA. Short Chain Fatty Acids (SCFAs)-Mediated Gut Epithelial and Immune Regulation and Its Relevance for Inflammatory Bowel Diseases. Front Immunol 2019;10:277. [57] Davalos-Salas M, Montgomery MK, Reehorst CM, Nightingale R, Ng I, Anderton H, Al-Obaidi S, Lesmana A, Scott CM, Ioannidis P, Kalra H, Keerthikumar S, Togel L, Rigopoulos A, Gong SJ, Williams DS, Yoganantharaja P, Bell-Anderson K, Mathivanan S, Gibert Y, Hiebert S, Scott AM, Watt MJ, Mariadason JM. Deletion of intestinal Hdac3 remodels the lipidome of enterocytes and protects mice from diet-induced obesity. Nat Commun 2019;10(1):5291. [58] Bhat MI, Kapila R. Dietary metabolites derived from gut microbiota: critical modulators of epigenetic changes in mammals. Nutr Rev 2017;75(5):374-89. [59] Lukovac S, Belzer C, Pellis L, Keijser BJ, de Vos WM, Montijn RC, Roeselers G. Differential modulation by Akkermansia muciniphila and Faecalibacterium prausnitzii of host peripheral lipid metabolism and histone acetylation in mouse gut organoids. mBio 2014;5(4). [60] Levin B. Inflammatory bowel disease and colon cancer. Cancer 1992;70(5 Suppl):1313-6. [61] Li MJ, Liu Z, Wang P, Wong MP, Nelson MR, Kocher JP, Yeager M, Sham PC, Chanock SJ, Xia Z, Wang J. GWASdb v2: an update database for human genetic variants identified by genome-wide association studies. Nucleic Acids Res 2016;44(D1):D869-76. [62] MacArthur J, Bowler E, Cerezo M, Gil L, Hall P, Hastings E, Junkins H, McMahon A, Milano A, Morales J, Pendlington ZM, Welter D, Burdett T, Hindorff L, Flicek P, Cunningham F, Parkinson H. The new NHGRI-EBI Catalog of published genome-wide association studies (GWAS Catalog). Nucleic Acids Res 2017;45(D1):D896-D901. [63] Pinero J, Ramirez-Anguita JM, Sauch-Pitarch J, Ronzano F, Centeno E, Sanz F, Furlong LI. The DisGeNET knowledge platform for disease genomics: 2019 update. Nucleic Acids Res 2020;48(D1):D845-D55. [64] Pesquita C, Faria D, Bastos H, Ferreira AE, Falcao AO, Couto FM. Metrics for GO based protein semantic similarity: a systematic evaluation. BMC Bioinformatics 2008;9 Suppl 5:S4. [65] Wang JZ, Du Z, Payattakool R, Yu PS, Chen CF. A new method to measure the semantic similarity of GO terms. Bioinformatics 2007;23(10):1274-81.