A meta-analysis of genome-wide gene expression differences identifies promising targets for type 2 diabetes mellitus

Aims/introduction: Due to the heterogeneous nature of type 2 diabetes mellitus and its complex effects on hemodynamics, there is a need to identify new candidate markers which are involved in the development of type 2 diabetes mellitus (DM) and can serve as potential targets. As the global diabetes prevalence in 2019 was estimated as 9.3% (463 million people), rising to 10.2% (578 million) by 2030 and 10.9% (700 million) by 2045, the need to limit this rapid prevalence is of concern. The study aims to identify the possible biomarkers of type 2 diabetes mellitus with the help of the system biology approach using R programming.

Materials and methods: Several target proteins that were found to be associated with the source genes were further curated for their role in type 2 diabetes mellitus. The differential expression analysis provided 50 differentially expressed genes by pairwise comparison between the biologically comparable groups out of which eight differentially expressed genes were short-listed. These DEGs were as follows: MCL1, PTX3, CYP3A4, PTGS1, SSTR2, SERPINA3, TDO2, and GALNT7.

Results: The cluster analysis showed clear differences between the control and treated groups. The functional relationship of the signature genes showed a protein–protein interaction network with the target protein. Moreover, several transcriptional factors such as DBX2, HOXB7, POU3F4, MSX2, EBF1, and E4F1 showed association with these identified differentially expressed genes.

Conclusions: The study highlighted the important markers for diabetes mellitus that have shown interaction with other proteins having a role in the progression of diabetes mellitus that can serve as new targets in the management of DM.

Background

Type 2 diabetes mellitus is known to be a life-long disease that is linked to disturbances in glucose levels in the blood. This disease acts as a silent killer that is affecting a wider population of the world. The need to develop potential target hypoglycemic agents as well as new biological candidates and possible treatment strategies has been emphasized in order to combat this disease (1). Type 2 diabetes mellitus has become prevalent day by day, and at present, doctors have an increasing number of patients who suffer from this disease (2). Both insulin resistance and defects in the secretion of insulin have shown a role in the pathogenesis of type 2 diabetes mellitus (3). Individuals suffering from type 2 diabetes mellitus encounter microvascular complications such as retinopathy, nephropathy, and neuropathy and also macrovascular complications including cardiovascular comorbidities, contributing to the development of hyperglycemia and insulin resistance (metabolic) syndrome (4).

The system biology approach aids in understanding integrative physiological responses through the incorporation of experimental and computational approaches. It provides a powerful foundation from which to address complex scientific questions (5). The R software is a recent, functional programming language that creates high-quality graphical productivity, and all the stages of a study, from analysis to publication, can be undertaken within R (6). R is a very powerful statistical tool and is especially used for data manipulation, calculations, and plots (7). In-silico or docking studies are those research tools that are “run on or performed on a computer or done via computer simulation.” Drug designing and discovery involves the development of new potential targets for drug molecules through computer simulation, and it is a rapidly emerging trend nowadays (8). Several mechanisms are under the control of selective gene expressions in molecular biology. The difference in the expression between normal tissue and diseased tissue aids in identifying the disease mechanism. The dysregulation of genes and pathways in normal tissue occasionally leads to a disease state that after recognition can guide in making therapy decisions. Functional enrichment analysis and gene ontology studies provide the affiliation data of such proteins and the collaborating proteins that help in isolating these topographies (9).

The main aim of this study is the screening of potential biomarkers for type 2 diabetes mellitus for the better management of this disease. The target-based treatment strategies have proven fruitful due to their site-specific efficacy and reduced side effects (10). For the development of target-based drugs, the system biology approach has been found useful. This approach aids in the identification of important gene signatures that are known to play an essential role in disease etiology. Targeting these genes can slow down the pace of the progression of DM. Therefore, type 2 diabetes mellitus differentially expressed genes have been identified and short-listed. The role of these genes has been curated and the gene ontology and annotation of these genes will also be studied. Moreover, a gene network analysis of signature genes with other potential interacting proteins will also be mapped to further clarify the role of these genes in the progression of type 2 diabetes mellitus. The miRNA target hits for these identified DEGs will also be predicted. This will aid particularly in targeting these specific signature genes or associated genes involved in the pathogenesis of this disease, thus opening new treatment strategies.

Methods

Derivation of the GEO datasets

In this study, the potential type 2 diabetes mellitus targets were identified through a differential screening method. Ten datasets of type 2 diabetes mellitus were retrieved using the NCBI database Gene Expression Omnibus. The dataset used was “organism: Homo sapiens” and “experiment type: expression profiling by array”. The Affymetrix GeneChip Human Genome U133 Plus 2.0 Array was used (Affymetrix, Inc., Santa Clara, CA, 95051, USA). All the cDNA datasets included various information including platform, GEO accession number, number of samples, sample type, and genetic expression data. The differentially expressed genes were identified using the array platform and the annotation platform hgu133plus2 probes. For the computational analysis, the R programming and the Bioconductor packages, affy QC Report, affy, annotate, AnnotationDbi, limma, Biobase, AffyRNAdegradation, hgu133plus2cdf, and hgu133a2cdf, were used (11).

Processing of cDNA datasets

The phenodata files for each dataset having the “.cel format” were prepared for initial preprocessing. For normalization, the Bioconductor Array Quality Metrics package was employed in R programming to a median expression level for each gene (version 3.1.3). After the normalization, the second step was the background correction for perfect match (pm) and mismatch (mm) using the robust multiarray analysis (RMA). This method reduces the local noise present in the background. Afterward, the RMA algorithm summarization was performed to measure the averages among the probes in a probe set and to attain the summary of intensities.

The degradation of RNA was analyzed in order to determine the quality of RNA in these microarray datasets measured by using the AffyRNAdegradation package of Bioconductor. Lastly, the differentially expressed genes in each dataset were recognized using the pairwise comparison. The Benjamini–Hochberg approach was applied for multiple testing correction in order to shortlist the DEGs. The shortlisting of DEGs was done on the basis of the resulting scores and p-values that were then ranked. The cutoff values for shortlisting the genes were p-value ≤0.05, false discovery rate (FDR)  <0.05, and absolute log fold change logFC  >1 (11).

Data mapping and screening

The differentially expressed genes obtained after shortlisting were further mapped to confirm their role in type 2 diabetes mellitus. This was done with the help of varied data sources such as PubMed (http://www.ncbi.nlm.nih.gov/pubmed), Medical Subject Headings (MeSH) (http://www.ncbi.nlm.nih.gov/mesh), Online Mendelian Inheritance in Man (OMIM) (http://www.ncbi.nlm.nih.gov/omim), and PubMed Central (PMC) databases (http://www.ncbi.nlm.nih.gov/pmc). Biomedical text mining was done to filter the potential genes specific to the disease.

Cluster analysis

The expression values present in each dataset were evaluated for cluster analysis using the CIMminer tool applying the absolute Pearson correlation analysis. A significant variation in the gene expression levels among the control and treated samples was observed in the cluster analysis (12).

Protein–protein analysis and network construction

A network analysis was performed to study the interaction pattern of each protein with other genes having variable physiological functions in the pathological conditions when compared to normal signifying a protein–protein interaction of the signature genes. The Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) and the Human Annotated and Predicted Protein Interaction (HAPPI) databases helped in evaluating the interaction of proteins with each other having a confidence score of 0.999. The Cytoscape software (version 3.2.1, Temple Place, Suite 330, Boston, MA 02111-1307, USA) was utilized to visualize and analyze the molecular interactions of signature genes with the target genes. Various databases including OMIM, MeSH, and PMC were used to recognize the role of target genes in type 2 diabetes mellitus, and the associated gene signatures cause a pathological phenotype by dysregulation. The online tool Database for Annotation Visualization and Integrated Discovery (DAVID) and the functional enrichment tool FunRich were used to study the physiological functions of signature genes including functional annotation and gene ontology (11, 12).

miRNA target prediction

miRNAs have shown a potential influence on gene targets, and hence, the prediction of miRNAs has proven beneficial in better understanding disease etiology. Several diseases are linked to the dysregulation of miRNAs causing disruption of signaling pathways. Therefore, RNA prediction was done using the miRDB target predictor (www.mirdb.org). It is an online database for the target prediction and functional annotation of miRNAs. The target score (≤99) was used to select miRNAs.

Results

Identification of differentially expressed genes

Ten datasets of type 2 diabetes mellitus having the “.cel format” were retrieved from the NCBI GEO database. Each dataset had a size of ArrayBatch object 1,164 × 1,164 features with associated affyIDs (Table 1). The normalization and background correction was done using quantile normalization in order to avoid systematic variation. The MA plot was obtained after normalization showing the quality of arrays in each dataset defining the probe level data (Figure 1). The degradation of RNA and the severity were represented by the AffyRNAdeg plot of the affy package of R (Figure 2). The AffyRNAdeg package provided the single summary statistic for each array in the batch (Additional File 1; Table S1). Additional File 2; Table S2 enlists the databases, tools, and software used in this study.

TABLE 1

Table 1 cDNA datasets obtained from the GEO databases.

FIGURE 1

Figure 1 The normalization and the quality array metrics are represented by the MA plots showing the log intensity ratio (M) vs. log intensity averages (A). Usually, the mass of distribution in the MA plot is likely to be intense along the M = 0 axis.

FIGURE 2

Figure 2 RNA degradation plots obtained by the plot AffyRNAdeg package of R demonstrating RNA quality and severity of degradation.

Screening of differentially expressed genes

About 50 DEGs were obtained in each dataset through differential expression analysis by pairwise comparison among biologically similar groups. From the 10 datasets, 20 DEGs were selected. This was done on the basis of p-value and logFC factors. These 20 DEGs were curated to eight common signature genes that were considered as candidate markers for type 2 diabetes mellitus (Additional File 3; Table S3).

Mapping of DEGs

The curation and mapping of DEGs was done using different databases such as PubMed, OMIM, MeSH, and PMC giving the following potential type 2 diabetes mellitus-associated genes: MCL1, PTX3, CYP3A4, PTGS1, SSTR2, SERPINA3, TDO2, and GALNT7. The function of these signature genes was curated and counted. Table 2 shows the differentially expressed type 2 diabetes mellitus-associated genes which were curated.

TABLE 2

Table 2 List of differentially expressed type 2 diabetes-related signature genes after curation.

Cluster analysis

The genetic expression of type 2 diabetes samples shows obvious differences among the control and treated samples. Figure 3 shows the cluster analysis of type 2 diabetes mellitus-related differentially expressed signature genes. The blue color displays a large distance, while the red color indicates a small distance. Lines represent the cluster boundaries at the level of the tree.

FIGURE 3

Figure 3 Cluster analysis of type 2 diabetes-associated differentially expressed signature genes using the CIMminer tool. Blue indicates a large distance, while red represents a small distance. Lines show the cluster boundaries in the level of the tree.

Functional enrichment analysis

The FunRich tool provided information regarding the percentage of our selected genes that were involved in various biological processes. Figure 4A shows the biological processes involved in type 2 diabetes mellitus. The FunRich tool also provided the transcriptional factors that are linked to these differentially expressed genes. The transcription factors identified were DBX2, HOXB7, POU3F4, MSX2, EBF1, and E4F1. The transcription factors (TFs) showed 20% abundance with the known type 2 diabetes mellitus genes. These transcriptional factors also showed a role as targets in type 2 diabetes mellitus. Figure 4B shows the transcription factors for type 2 diabetes mellitus-related genes that can modify the genetic expression in a host cell.

FIGURE 4

Figure 4 (A) Biological processes involved in type 2 diabetes mellitus. (B) Transcription factors for type 2 diabetes mellitus-related signature genes that can modify the gene expression in a host cell.

DAVID is an online tool that was used to determine the gene ontology and functional annotation of the identified differentially expressed genes. In functional enrichment analysis, the signature genes were short-listed based on fold change and p-value cutoff. Table 3 shows the gene ontology of type 2 diabetes-related DEGs.

TABLE 3

Table 3 Gene ontology of type 2 diabetes mellitus-associated DEGs.

Network integrome analysis

The network analysis exposed the interaction of type 2 diabetes mellitus-related signature genes with other genes. The network showed 156 nodes and 158 edges. The network was characterized in three neighborhoods: the blue nodes show the type 2 diabetes mellitus-associated potential biomarkers, the pink nodes show the proteins involved in diabetes mellitus, and the remaining yellow nodes display the non-type 2 diabetes mellitus target proteins. Figure 5 shows the gene network of type 2 diabetes-related differentially expressed genes.

FIGURE 5

Figure 5 Genetic network of type 2 diabetes-associated differentially expressed signature genes. The blue nodes indicate the type 2 diabetes mellitus-associated potential biomarkers, the yellow nodes represent the non-type 2 diabetes mellitus target proteins, and the pink nodes represent type 2 diabetes mellitus-related DEGs.

miRNA target prediction

Multiple diabetes mellitus-associated miRNA targets for each gene were identified by the computational algorithm miRDB. The miRNAs identified were hsa-miR-3686, hsa-miR-1299, hsa-miR-3163, hsa-let-7a-2-3p, hsa-miR-4306, hsa-miR-4277, hsa-miR-5680, and hsa-miR-296-5p (Table 4). The progression and development of diabetes mellitus is linked with the dysregulation of these miRNA targets. The genes MCL1, PTGS1, CYP3A4, SERPINA3, TDO2, and GALNT7 predicted 99, 107, 98, 89, 95, and 98 miRNA hits, respectively.

TABLE 4

Table 4 miRNA targets for diabetes mellitus-associated genes.

Discussion

Owing to the heterogeneous nature of type 2 diabetes mellitus and its complex effects on hemodynamics, there is a need to identify new targets to overcome the effects which are involved in the progression of type 2 diabetes mellitus (23). To reduce the worldwide burden of type 2 diabetes mellitus, targeted policies are required to be put into practice with better beneficial efforts. The main objective of this study was to find potential biomarkers for type 2 diabetes mellitus which can later be utilized in antidiabetic therapy.

In this study, the affy package was implemented in the R statistical program and scripting language for several reasons (24). The software is free of cost and is very effective in the differential expression analysis of several genes that can be used as targets for the better management of disease (11, 12). Several gene signatures were identified through the system biology approach, and their functional annotation and protein–protein interactions for a better understanding of type 2 diabetes mellitus progression were performed. The differential expression analysis resulted in 50 differentially expressed genes by pairwise comparison among the physiologically similar groups. From the 50 differentially expressed genes, the topmost 2 genes were graded and nominated from each dataset. The DAVID tool was used to retrieve the gene symbol, UniProt ID, and protein name. Data curation was done to shortlist the genes on the basis of high count by using various databases, i.e., PubMed, PMC, OMIM, and MeSH. The differential study exposed eight potential gene signatures out of the 20 DEGs on the basis of physicochemical and functional evidence, which play a role in type 2 diabetes. MCL1, PTX3, CYP3A4, PTGS1, SSTR2, SERPINA3, TDO2, and GALNT7 were the identified DEGs, out of which seven were upregulated (MCL1, PTX3, PTGS1, SSTR2, SERPINA3, TDO2, and GALNT7) and one was downregulated (CYP3A4). The cluster analysis was done by using the CIMminer tool that showed variations between the control and treated groups using yellow and blue colors.

The differentially expressed gene MCL1 belongs to the BCL2 family of antiapoptotic proteins, that is, myeloid cell leukemia sequence 1 (Mcl-1). It is an apoptosis regulator and type 2 diabetes mellitus is triggered by impaired β-cell function. The pro-inflammatory cytokines downregulated the MCL1 gene causing β-cell apoptosis (25). The role of MCL1 in diabetes mellitus needs to be characterized. However, due to its restricted role in the control of apoptosis, it can serve as a promising target in the management of diabetes by preventing β-cell apoptosis. The downregulation of MCL1 is a crucial event in β-cell apoptosis and its role has been studied by Cardozo et al. (26). PTX3, another differentially expressed gene, is a pentraxin 3 gene that is involved in the progression of diabetic complications including diabetic nephropathy and retinopathy by the activation of immunological and inflammatory mechanisms (27). The complications of diabetes mellitus are the major obstacles in its treatment including both microvascular and macrovascular complications. Recently, in one study, PTX3 has been shown to be an accurate marker in revealing diabetic neuropathic progression (28, 29). The involvement of PTX3 in diabetic complications serves to be an attractive target. Another identified differentially expressed gene, CYP3A4, belongs to the cytochrome P450 family, and the presence of these enzymes contributes to low chronic inflammation in type 2 diabetes mellitus (13). The changes in the expression levels of some P450 isoenzymes have shown an association with increased cytokine levels in type 2 diabetes mellitus (14). Induction of CYP3A4 expression levels has also been observed in hepatic cells due to high serum fatty acid levels clearly showing upregulation of CYP3A4 in diabetic conditions (15). PTGS1 is prostaglandin endoperoxide synthase 1 that activates the prostaglandin pathway, and through TNF alpha signaling, the immune system will be upregulated or downregulated and will cause inflammation in type 2 diabetes mellitus (16). The upregulation of the expression of the inflammatory PTGS gene in pancreatic islets might be contributing to the dysfunction of islets in diabetes (17). SSTR2 is somatostatin receptor 2 and inhibits insulin and glucagon release from pancreatic islets by interacting with membrane somatostatin receptors, and it is expressed at high levels on alpha cells and suggested a selective role in the regulation of glucagon release (18). SERPINA3 belongs to the serpin family A member 3. It increases the trans-endothelial permeability of retinal microvascular endothelial cells and is involved in the pathogenesis of diabetic retinopathy (19). TDO2 is tryptophan 2,3-dioxygenase and makes hepatic changes in metabolism (20). GALNT7 is a polypeptide N-acetylgalactosaminyltransferase. It downregulates the expression of type 2 diabetes mellitus (21). The role of the identified differentially expressed genes in the development of diabetes mellitus and their association with the development of diabetic complications have opened new insights for the better management of this disease.

The FunRich software was used to determine the transcriptional factors and biological processes involved in diabetes mellitus. The transcriptional factors identified were DBX2, HOXB7, POU3F4, MSX2, EBF1, and E4F1. The TFs showed 20% abundance with the known type 2 diabetic genes. The identified transcriptional factors also showed a significant role in type 2 diabetes mellitus, serving as targets in the development of new treatment strategies. The biological processes identified were immune response, metabolism, energy pathways, apoptosis, lipid metabolism, and anti-apoptosis. The DAVID tool was used to determine gene ontology. The gene ontology of these genes showed inflammatory response, oxidation–reduction process, xenobiotic metabolic process, lipid metabolic process, organelle membrane, heme binding, and oxygen binding. The role of the identified differentially expressed genes was extensively studied opening new paths in the better management of diabetes mellitus.

The gene network analysis was performed in order to analyze the interaction of the identified seeder genes with other proteins that may be involved in the progression of diabetes mellitus. The HAPPI and STRING databases were used to determine the proteins associated with each other and other proteins. The Cytoscape software was used to visualize the network. The role of target proteins in type 2 diabetes was further confirmed by using various databases, i.e., PubMed, PMC, OMIM, and MeSH. The databases identified the proteins having a role in diabetes mellitus (colored in red and pink) and also the non-diabetes-associated proteins (colored in yellow). Several interacting proteins that showed association with MCL1 were BCL2, BAD, CASP2, CCND1, TP53, etc. PTX3 showed association with TNFAIP, MMP9, CXCL, CAMP, and TF; CYP3A4 with FMO1, UGT, and CYB5; and PTGS1 with ALOX1, PTGE, LOX5, HEM1, MSH3, and TNFA. SSTR2 showed interactions with NPY, POMC, SST, and SSR1. The identified signature genes and the interacting proteins have shown roles in the development of diabetes mellitus. The interaction analysis has shown how these potential biomarkers interact with each other as well as with other proteins in causing diabetes mellitus. The literature has also confirmed their role in the progression of type 2 diabetes mellitus; hence, these genes can further be used as potential targets to treat diabetes mellitus by formulating drugs that would specifically target these genes. The secondary genes TNFAIP, MMP9, and CXCL showing interaction with PTX3 have a potential role in diabetes mellitus (22, 30, 31). Several lines of evidence have also supported the role of the secondary genes FMO1 and ALOX1 having interaction with CYP3A4 and PTGS1 in diabetes mellitus (31, 32). Most of the signature genes that do not show a role in diabetes mellitus provided evidence of the contribution of these genes interacting with DEGs in causing diabetes mellitus. In conclusion, the study has identified the potential candidate biomarkers for diabetes mellitus that are recently emerging as new targets in disease management, and the interaction of these genes with other proteins through network analysis has further clarified the role of these genes in the progression of diabetes mellitus.

Conclusion

The study has helped us in identifying candidate biomarker genes for diabetes mellitus using differential expression analysis. MCL1, PTX3, CYP3A4, PTGS1, SSTR2, SERPINA3, TDO2, and GALNT7 were the identified DEGs. Several lines of evidence have shown the role of these genes in the development of diabetes mellitus. Moreover, the study has also explored the interaction of these genes with other proteins involved in crosstalks with these genes. The crosstalks between these genes are responsible for the progression and the complications associated with diabetes mellitus. The gene ontology and functional annotation analysis have also helped us in understanding the biological processes, pathways, and transcriptional factors associated with these genes. The study has opened new insights related to the role of these DEGs in diabetes mellitus, and targeting these genes could potentially help us in the better management of this disease.

Data availability statement

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.

Author contributions

Conceptualization: TH, UI, RA, and MA. Writing—original draft preparation: BN, UI, RA, and MA. Writing—review and editing: BZ, HZ, AP-S, NN, and MI. Supervision: UI, MA, and YD. All authors have read and agreed to the published version of the manuscript.

Funding

This study was financed by Riphah International University (Riphah-ORIC-21-22/FPS-51 and Riphah-ORIC-21-22/FPS-56).

Conflict of interest

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

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fendo.2022.985857/full#supplementary-material

Additional file 1 | Table S1: A single summary statistic for each array provided the function summary AffyRNAdeg of Bioconductor package in each dataset.

Additional file 2 | Table S2: Databases, software, and tools used in the study.

Additional file 3 | Table S3: Initial examination of common and related differential genes of each microarray dataset.

References

1. Mane PB, Antre RV, Oswal RJ. Antidiabetic drugs: An overview. Int J Pharm Chem Sci (2012) 1(1):301–6.

Google Scholar

2. Stumvoll M, Goldstein BJ, Van Haeften TW. Type 2 diabetes: principles of pathogenesis and therapy. Lancet (2005) 365(9467):1333–46. doi: 10.1016/S0140-6736(05)61032-X

PubMed Abstract | CrossRef Full Text | Google Scholar

3. Tripathi A, Misra K. Molecular docking: A structure-based drug designing approach. JSM Chem (2017) 5(2):1042–7.

Google Scholar

4. DeFronzo RA, Ferrannini E, Groop L, Henry RR, Herman WH, Holst JJ, et al. Type 2 diabetes mellitus. Nat Rev Dis primers. (2015) 1(1):1–22. doi: 10.1038/nrdp.2015.19

CrossRef Full Text | Google Scholar

5. Kitano H. Computational systems biology. Nature. (2002) 420(6912):206–10. doi: 10.1038/nature01254

PubMed Abstract | CrossRef Full Text | Google Scholar

6. Eglen SJ. A quick guide to teaching r programming to computational biology students. PloS Comput Biol (2009) 5(8):1000482. doi: 10.1371/journal.pcbi.1000482

CrossRef Full Text | Google Scholar

7. de Micheaux PL, Drouilhet R, Liquet B. Programming in r. In: The r software statistics and computing. New York: Springer (2013). p. 193–282.

Google Scholar

8. Abdollahpour N, Asoodeh A, Saberi MR, Chamani J. Separate and simultaneous binding effects of aspirin and amlodipine to human serum albumin based on fluorescence spectroscopic and molecular modeling characterizations: A mechanistic insight for determining usage drugs doses. J luminescence. (2011) 131(9):1885–99. doi: 10.1016/j.jlumin.2011.04.043

CrossRef Full Text | Google Scholar

9. Rachlin J, Cohen DD, Cantor C, Kasif S. Biological context networks: a mosaic view of the interactome. Mol Syst Biol (2006) 2(1):66. doi: 10.1038/msb4100103

PubMed Abstract | CrossRef Full Text | Google Scholar

10. Muhammad SA, Ahmed S, Ali A, Huang H, Wu X, Yang XF, et al. Prioritizing drug targets in clostridium botulinum with a computational systems biology approach. Genomics. (2014) 104(1):24–35. doi: 10.1016/j.ygeno.2014.05.002

PubMed Abstract | CrossRef Full Text | Google Scholar

11. Altaf R, Nadeem H, Babar MM, Ilyas U, Muhammad SA. Genome-scale meta-analysis of breast cancer datasets identifies promising targets for drug development. J Biol Research-Thessaloniki. (2021) 28(1):5. doi: 10.1186/s40709-021-00136-7

CrossRef Full Text | Google Scholar

12. Ilyas U, Zaman S, Altaf R, Nadeem H, Muhammad SA. Genome wide meta-analysis of cDNA datasets reveals new target gene signatures of colorectal cancer based on systems biology approach. J Biol Res (Thessalon). (2020) 27(1):1–13. doi: 10.1186/s40709-020-00118-1

PubMed Abstract | CrossRef Full Text | Google Scholar

13. Gravel S, Chiasson JL, Turgeon J, Grangeon A, Michaud V. Modulation of CYP 450 activities in patients with type 2 diabetes. Clin Pharmacol Ther (2019) 106(6):1280–9. doi: 10.1002/cpt.1496

PubMed Abstract | CrossRef Full Text | Google Scholar

14. Darakjian L, Deodhar M, Turgeon J, Michaud V. Chronic inflammatory status observed in patients with type 2 diabetes induces modulation of cytochrome P450 expression and activity. Int J Mol Sci (2021) 22(9):4967. doi: 10.3390/ijms22094967

PubMed Abstract | CrossRef Full Text | Google Scholar

15. Hu N, Hu M, Duan R, Liu C, Guo H, Zhang M, et al. Increased levels of fatty acids contributed to induction of hepatic CYP3A4 activity induced by diabetes — In vitro evidence from HepG2 cell and Fa2N-4 cell lines. J Pharmacol Sci (2014) 124(4):433–44. doi: 10.1254/jphs.13212FP

PubMed Abstract | CrossRef Full Text | Google Scholar

16. Blair PJ, Hwang SJ, Shonnard MC, Peri LE, Bayguinov Y, Sanders KM, et al. The role of prostaglandins in disrupted gastric motor activity associated with type 2 diabetes. Diabetes. (2019) 68(3):637–47. doi: 10.2337/db18-1064

PubMed Abstract | CrossRef Full Text | Google Scholar

17. Shanmugam N, Todorov IT, Nair I, Omori K, Reddy MA, Natarajan R. Increased expression of cyclooxygenase-2 in human pancreatic islets treated with high glucose or ligands of the advanced glycation endproduct-specific receptor (AGER), and in islets from diabetic mice. Diabetologia. (2006) 49(1):100–7. doi: 10.1007/s00125-005-0065-7

PubMed Abstract | CrossRef Full Text | Google Scholar

18. Strowski M, Parmar R, Blake A, Schaeffer J. Somatostatin inhibits insulin and glucagon secretion via two receptor subtypes: an in vitro study of pancreatic islets from somatostatin receptor 2 knockout mice. Endocrinology. (2000) 141(1):111–7. doi: 10.1210/endo.141.1.7263

PubMed Abstract | CrossRef Full Text | Google Scholar

19. Sánchez-Navarro A, González-Soria I, Caldiño-Bohn R, Bobadilla NA. An integrative view of serpins in health and disease: The contribution of SerpinA3. Am J Physiology-Cell Physiol (2021) 320(1):C106–C18. doi: 10.1152/ajpcell.00366.2020

CrossRef Full Text | Google Scholar

20. Rebnord EW, Strand E, Midttun Ø, Svingen GF, Christensen MH, Ueland PM, et al. The kynurenine: tryptophan ratio as a predictor of incident type 2 diabetes mellitus in individuals with coronary artery disease. Diabetologia. (2017) 60(9):1712–21. doi: 10.1007/s00125-017-4329-9

PubMed Abstract | CrossRef Full Text | Google Scholar

21. Zhou X, Xu Y, Yin D, Zhao F, Hao Z, Zhang J, et al. Type 2 diabetes mellitus facilitates endometrial hyperplasia progression by activating the proliferative function of mucin O-glycosylating enzyme GALNT2. Biomedicine Pharmacotherapy. (2020) 131:110764. doi: 10.1016/j.biopha.2020.110764

CrossRef Full Text | Google Scholar

22. Tejasvi T, Stuart PE, Chandran V, Voorhees JJ, Gladman DD, Rahman P, et al. TNFAIP3 gene polymorphisms are associated with response to TNF blockade in psoriasis. J Invest Dermatol (2012) 132(3, Part 1):593–600. doi: 10.1038/jid.2011.376

PubMed Abstract | CrossRef Full Text | Google Scholar

23. Hu FB, Manson JE, Stampfer MJ, Colditz G, Liu S, Solomon CG, et al. Diet, lifestyle, and the risk of type 2 diabetes mellitus in women. New Engl J Med (2001) 345(11):790–7. doi: 10.1056/NEJMoa010492

PubMed Abstract | CrossRef Full Text | Google Scholar

24. Gautier L, Cope L, Bolstad BM, Irizarry RA. Affy–analysis of affymetrix GeneChip data at the probe level. Bioinformatics. (2004) 20(3):307–15. doi: 10.1093/bioinformatics/btg405

PubMed Abstract | CrossRef Full Text | Google Scholar

25. Allagnat F, Cunha D, Moore F, Vanderwinden J, Eizirik DL, Cardozo AK. Mcl-1 downregulation by pro-inflammatory cytokines and palmitate is an early event contributing to β-cell apoptosis. Cell Death Differentiation. (2011) 18(2):328–37. doi: 10.1038/cdd.2010.105

CrossRef Full Text | Google Scholar

26. Cardozo AK, Allagnat F, Cunha DA, Moore F, Vanderwinden J-M, Eizirik DL. Mcl-1 downregulation by pro-inflammatory cytokines and palmitate is an early event contributing beta cell apoptosis. Cell Death Differentiation (2011) 18:328–37. doi: 10.1038/cdd.2010.105

PubMed Abstract | CrossRef Full Text | Google Scholar

27. Uzun S, Ozari M, Gursu M, Karadag S, Behlul A, Sari S, et al. Changes in the inflammatory markers with advancing stages of diabetic nephropathy and the role of pentraxin-3. Renal failure. (2016) 38(8):1193–8. doi: 10.1080/0886022X.2016.1209031

PubMed Abstract | CrossRef Full Text | Google Scholar

28. Dawood AA, Kamel MA, Omar TA, Agaba AAM. Study of serum pentraxin 3 level in patients with diabetic nephropathy. Egyptian J Internal Med (2020) 32(1):3. doi: 10.1186/s43162-020-00002-3

CrossRef Full Text | Google Scholar

29. Mutlu M, Yuksel N, Takmaz T, Dincel AS, Bilgihan A, Altınkaynak H. Aqueous humor pentraxin-3 levels in patients with diabetes mellitus. Eye. (2017) 31(10):1463–7. doi: 10.1038/eye.2017.87

CrossRef Full Text | Google Scholar

30. Liu Y, Li J, Yao H, Zhang M, Lian J, Zhang H, et al. Effects of upregulation of TNFAIP3 on diabetic neuropathic pain in mice. Dis Markers. (2021) 2021:3470950. doi: 10.1155/2021/3470950

PubMed Abstract | CrossRef Full Text | Google Scholar

31. Lieb DC, Brotman JJ, Hatcher MA, Aye MS, Cole BK, Haynes BA, et al. Adipose tissue 12/15 lipoxygenase pathway in human obesity and diabetes. J Clin Endocrinol Metab (2014) 99(9):E1713–E20. doi: 10.1210/jc.2013-4461

PubMed Abstract | CrossRef Full Text | Google Scholar

32. Miao J, Ling AV, Manthena PV, Gearing ME, Graham MJ, Crooke RM, et al. Flavin-containing monooxygenase 3 as a potential player in diabetes-associated atherosclerosis. Nat Commun (2015) 6(1):1–10. doi: 10.1038/ncomms7498

CrossRef Full Text | Google Scholar