Candidate Biomarkers for Targeting in Type 1 Diabetes; A Bioinformatic Analysis of Pancreatic Cell Surface Antigens

Abstract

Objective: Type 1 diabetes (T1Ds) is an autoimmune disease in which the immune system invades and destroys insulin-producing cells. Nevertheless, at the time of diagnosis, about 30-40% of pancreatic beta cells are healthy and capable of producing insulin. Bi-specific antibodies, chimeric antigen receptor regulatory T cells (CAR-Treg cells), and labeled antibodies could be a new emerging option for the treatment or diagnosis of type I diabetic patients. The aim of the study is to choose appropriate cell surface antigens in the pancreas tissue for generating an antibody for type I diabetic patients.

Materials and methods: In this bioinformatics study, we extracted pancreas-specific proteins from two large databases; the Human Protein Atlas (HPA) and Genotype-Tissue Expression (GTEx) Portal. Pancreatic-enriched genes were chosen and narrowed down by Protter software for the investigation of accessible extracellular domains. The immunohistochemistry (IHC) data of the protein atlas database were used to evaluate the protein expression of selected antigens. We explored the function of candidate antigens by using the GeneCards database to evaluate the potential dysfunction or activation/hyperactivation of antigens after antibody binding.

Results: The results showed 429 genes are highly expressed in the pancreas tissue. Also, eighteen genes encoded plasma membrane proteins that have high expression in the microarray (GEO) dataset. Our results introduced four structural proteins, including NPHS1, KIRREL2, GP2, and CUZD1, among all seventeen candidate proteins.

Conclusion: The presented antigens can potentially be used to produce specific pancreatic antibodies that guide CARTreg, bi-specific, or labeling molecules to the pancreas for treatment, detection, or other molecular targeted therapy scopes for type I diabetes.

Keywords: Bioinformatics; Cell Surface Antigens; Molecular Targeted Therapies; Pancreatic Islets; Type 1 Diabetes.

Fig.1

The tissue-specific gene enrichment plot of pancreatic chosen genes. The graph shows the enrichment expression of 429 genes that were evaluated in different 35 tissues.

Fig.2

The Heatmap plot of pancreas enriched expression genes. The plot showed all 68 genes have a high preferential expression level in the pancreas. Some genes, such as insulin (INS), were only detected in the pancreas, and some others, such as Glycine Amidinotransferase (GATM), had a basic expression level in other body tissues.

Fig.3

RNA expression level of plasma membrane proteins of the pancreatic islet cells and other tissues. Islets of the pancreas, acinar cells of the pancreas, adipose tissue, heart, small intestine, diaphragm tissues, bone marrow, and spleen tissues were evaluated by microarray data analysis (accession No. GSE24207). All 8 candidate genes have acceptable expression levels in the Pancreas beta cells. Five genes also have a good expression level in the Pancreas acini cells. The results showed that AQP8, CUZD1, GP2, SLC30A8, SLC39A5, and PTPRN genes were significantly expressed in the Pancreatic islet and acini cells. Also, the data showed that GRPR and NPHS1 were not significantly expressed in the Pancreas islets and acini cells in compression with other tissues. According to the results of analysis, P Value of AQP8, CUZD1, GP2, SLC30A8, SLC39A5, and PTPRN genes were 3×10^-3, 5×10-5, 7×10^-5, 3×10^-5 and 2×10^-5, respectively. ****; P<0.0001, **; P<0.01, and ns; Not significant.

Fig.4

Two examples of the Protter outputs of candidate plasma membrane antigens. A. The results showed that glycoprotein 2 (GP2) had an extracellular domain of about 510 amino acids which binds to the plasma membrane by a glycosylphosphatidylinositol (GIP) anchor. B. In comparison, the SLC30A8 protein did have not an adequate extracellular domain (just about 15 amino acids) for being targeted by antibodies or immune Treg cells.

Fig.5

Two examples of immunohistochemistry (IHC) images and localized area for candidate antigens. A. IHC data showed that the CUZD1 protein had a high expression in the pancreas exocrine cells and displayed a cytoplasmic/membranous subcellular localization. B. IHC data showed that the SLC30A8 protein had a high expression in the pancreatic islet endocrine cells and showed a cytoplasmic/membranous subcellular localization. Red arrows indicate the islets of Langerhans, endocrine cells including insulin-producing beta cells, and green arrows reveal the exocrine cells, including acini.