Is there a relationship between psoriasis and hepatitis C? A meta-analysis and bioinformatics investigation

We conducted a systematic review and meta-analysis of observational studies on the association between psoriasis and hepatitis C, which was studied in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [12] and Meta-analysis of Observational Studies in Epidemiology (MOOSE) guidelines [13].

For this meta-analysis, we searched PubMed, EMBASE and Cochrane library on December 31, 2020 without language or publication date restrictions, all relevant clinical studies were included in our study. Using the following keyword combinations: ("psoriasis" OR "psoriases" OR "psoriasi") AND ("hepatitis C" OR "hepacivirus"). We also searched the reference lists, relevant reviews and meta-analysis of included studies to find other potential studies. After searching the electronic database, studies were selected based on pre-established inclusion criteria. (1) Observational studies examining the correlation between psoriasis and hepatitis C, including cross-sectional studies, case–control studies or cohort studies; (2) The study participants are humans; (3) The case group is composed of patients with psoriasis, and the control group is composed of individual composition without psoriasis; (4) Studies must have provided odds ratios (ORs) or sufficient raw data to calculate ORs. Research screening was conducted by two independent investigators based on the above criteria, and a third investigator arbitrated whether the study was suitable for inclusion. First, the investigator independently screened all titles and abstracts to exclude duplicate and obviously irrelevant articles. Second, we checked the full texts of all remaining papers and included studies that met the inclusion criteria.

The following data were extracted from the included studies: first author, year of publication, country, study design, and quantitative estimates including odds ratio (OR) with 95%CIs on the association of hepatitis C with psoriasis. We used the Newcastle–Ottawa Scale to assess the risk of bias of included studies [14]. A study could be rated as low risk, uncertain risk, or high risk based on its effects, a study with one or more red lights indicated a lower-quality study, while a study with no red lights suggested a higher-quality study (Additional file 1). All included studies were independently evaluated by two investigators, and differences were resolved through discussions with the third investigator.

All analyses were conducted by using the Review Manager 5.3 [15]. Odds ratios (OR) were calculated for case control/cross-sectional studies. The I² statistic was used to assess the statistical heterogeneity of the included studies. I² value greater than 50% was considered to performed a random-effects model meta-analysis and using fixed effects model meta-analysis with I² less than 50%. All statistical tests were two-sided, and a p value < 0.05 was considered statistically significant.

Using keywords “hepatitis C” and “psoriasis” to search for protein targets of hepatitis C and psoriasis-related diseases in the DisGeNET and GeneCards databases. GeneCards (https://​www.​genecards.​org/​) knowledge base automatically integrates gene-centric data from 150 network resources including genome, transcriptome, proteome, genetic, clinical, and functional information. It is an integrable, searchable database and provides comprehensive, user-friendly information about all annotated and predicted human genes. DisGeNET (http://​www.​disgenet.​org/​) is a discovery platform containing one of the largest publicly available collections of genes that can be used for different research purposes, including the investigation of the molecular underpinnings of human diseases and their comorbidities, the analysis of the properties of disease genes etc. [16]. Furthermore, the intersection of hepatitis C target and psoriasis target was determined with the online tool Venny 2.1 (http: // bioinfogp.cnb.csic.es/tools/venny/index.html).

The STRING database (https://​string-db.​org/​) covers almost all functional interactions between expressed proteins [17], and the species can be set to homo sapiens to construct a protein–protein interaction (PPI) network. Importing the target interaction information results obtained from the analysis into Cytoscape (version 3.6.1; (https://​www.​cytoscape.​org/​) where the interactive network is drawn and analyzed [18]. The database defines PPI with confidence ranges for data scores (low confidence: scores > 0.15; medium > 0.4; and high: > 0.7). In the present study, PPIs with combined scores higher than 0.4 were selected for further research. At the same time, we can use the Cytoscape plugin cytohubba plug-in according to our previous research to select a degree greater than twice the median degree of all nodes as the hub to identify the central element of the biological network [19].

In order to better understand the co-expression process related to hepatitis C and psoriasis, we performed Gene Ontology (GO) item and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of co-expressed genes. GO enrichment analysis provides a structured network of three aspects: biological process (BP), molecular function (MF) and cell composition (CC) to describe gene attributes. KEGG (http://​www.​genome.​jp/​kegg/​) is a database for large-scale systematic analysis of gene or protein molecular interaction networks [20]. DAVID bioinformatics resources include a comprehensive biological knowledge base and analysis tools designed to systematically extract biological meaning from large gene or protein lists [21]. We use the web-based search engine DAVID to determine over-represented GO terms and KEGG pathways with enrichment scores > 2, counts > 5, and P < 0.05 threshold. Using the free online data analysis platform imageGP (http: //www. ehbio. com/ ImageGP/ index. php/ Home/Index/index.html) tool to draw the bubble chart of the GO and KEGG analysis results. The important path items of KEGG have been mapped to the bubble chart. Larger and redder bubbles represent those highly abundant path items.

After screening 1205 records from preliminary search, a total of 11 studies were finally included in the study [22‐32]. The study selection process is summarized using the PRISMA flow chart in Fig. 1. 11 studies (8 cross-sectional and 3 case–control) examined the association of hepatitis C in psoriasis patients, as shown in Table 1. Risk of bias assessment revealed nine studies of higher quality [23‐25, 27‐32] and two studies of lower quality, as shown in Fig. 2A [22, 26].

A total of 11 studies, including 3 case control and 8 cross-sectional studies with 25,047 psoriasis patients and 4,091,631 controls provided data for this outcome [22‐32]. We identified substantial statistical heterogeneity across these 11 studies (I² = 89%). Psoriasis was associated with a significant increase of prevalent hepatitis C (OR 1.72; 95% confidence interval [CI] (1.17–2.52)) (Fig. 2B).

Target genes in hepatitis C and psoriasis were obtained from the GeneCard and DisGeNET databases (Additional file 2: Table 1–4). A total of 389 significant genes were common to both hepatitis C and psoriasis, which were associated with two diseases each other (Additional file 2: Table 5).

A total of 389 common potential genes associated with hepatitis C and psoriasis were uploaded to the STRING database for analysis. The network of PPI was established through the STRING database and the results were used for further analysis through Cytoscape software (Fig. 3B). From the analysis results, a total of 389 nodes and 1,375 edges were acquired, and the average node degree is 66.94. The edges represent the association between a pair of action targets, the nodes represent the action target, and the degree value represents its action intensity. The top ten targets IL-6, Tumor necrosis factor (TNF), IL10, Signal transducer and activator of transcription 3 (STAT3), Albumin(ALB), C-X-C Motif Chemokine Ligand (CXCL)8, Glyceraldehyde-3-Phosphate Dehydrogenase (GAPDH), IL4, AKT Serine/Threonine Kinase 1 (AKT1) and Vascular Endothelial Growth Factor A (VEGFA) have higher degree in this process, which explained their significance in the network (Fig. 3C).

To verify whether the 389 genes are related to hepatitis C and psoriasis, we conducted GO enrichment analysis to clarify the relevant biological processes (Additional file 3: Table 1). The redder the color, the bigger the value of -log10 p-value which also means greater credibility and more importance. The results of the GO analysis indicated that inflammatory response, immune response, and cellular response to lipopolysaccharide were the most significant terms related to hepatitis C and psoriasis in the BP category. Furthermore, the significant terms related to hepatitis C and psoriasis were cytokine activity, protein binding, and chemokine activity. The results of GO enrichment showed in Fig. 4A–C.

The KEGG pathway enrichment analysis was shown in Fig. 4D. Larger and redder bubbles represent the highly, significantly enriched pathway terms. In this study, the 389 overlapping gene symbols were mapped to 59 pathways following KEGG pathway enrichment. We sorted the first 25 pathways from small to large according to the p-value for a brief demonstration. The details of the KEGG pathway enrichment analysis are described in Additional file 3: Table 2. According to the results, an integrated overlapping signal pathways between psoriasis and hepatitis C has been constructed by the KEGG enrichment analysis, Of these pathways include: Immune system (e.g., Cytokine–cytokine receptor interaction, Toll-like receptor signaling pathway) (Additional file 3: Fig. 1), Signal transduction (e.g., Jak-STAT signaling pathway), Inflammatory disease (e.g., Inflammatory bowel disease (IBD)), Human disease-bacterial infectious disease pathways (e.g., Tuberculosis), human disease-viral infection (e.g., Influenza A, Herpes simplex infection, Measles), human disease-parasitic infectious disease pathway (e.g., Leishmaniasis, Chagas disease (American trypanosomiasis)). These signal pathways can affect the immune response, inflammatory response.