Investigating the role of sympathetic neurons in type 1 diabetes

Graphical Abstract

INTRODUCTION

Type 1 diabetes (T1D) is an autoimmune disease characterised by loss of beta cells in the islets of Langerhans within the pancreas. Sympathetic neurons innervate pancreatic islets and release neurotransmitters, modulating glucagon and insulin secretion. Glucose metabolism is affected by β-cell function through changes in the microenvironment and epigenome.^[1] The correlation between sympathetic nervous system activation and T1D progression is increasingly recognised. The impact of sympathetic signals on the islet cell environment in T1D highlights the need for a better understanding of molecular pathways that have remained largely unexplored.^[2]

The autonomic nervous system, particularly sympathetic efferent nerves, regulates insulin secretion and beta-cell proliferation.^[3] Studies have shown that loss of sympathetic innervation in islets could contribute to impaired responses to hypoglycaemia and the destruction of pancreatic beta cells in T1D.^[4] While others report that, despite beta cell loss in T1D, pancreatic capillaries and nerve fibres remain intact.^[5] The precise mechanisms by which it interacts with the genome to influence insulin production and immune response remain elusive.^[6] A significant gap exists in understanding the genes altered due to sympathetic mechanisms driving T1D.^[7,8] It is hypothesised that sympathetic neurotransmitters or neuropeptides interact with specific receptors on pancreatic beta cells or the cellular microenvironment. Thereby triggering intracellular signalling cascades that ultimately influence T1D gene expression. Literature reports an interesting study that compared sympathetic nerve fibres (marked by tyrosine hydroxylase, TH) isolated from the pancreas of T1D patients post-mortem.^[9] However, the identity of these receptors, the downstream signalling molecules, and the exact genes affected are yet to be fully elucidated. So, data from accession number GSE181674, deposited in the Gene Expression Omnibus (GEO) was selected for this study.

This previously reported dataset was analysed to fill the research gap, focusing on the nervous system's role within the pancreatic islets in T1D. The dataset GSE181674 comprised samples from control subjects and donors with T1D. Our study used bioinformatic tools to identify differentially expressed genes (DEGs) within these sympathetic nerve fibres. We also sought to validate our results on DEGs from post-mortem pancreas using blood soluble biomarker levels in T1D patients from another cohort.^[10] Data on blood biomarker levels can significantly bolster trends observed in DEGs. There is a growing recognition of the need for integrative approaches in diabetes research. This study combines biomarker analyses and gene expression datasets in T1D research. Integrating these data types allows the study to identify novel intervention points overlooked in traditional single-approach studies.

MATERIAL AND METHODS

We obtained gene expression data from the GEO database at the National Centre for Biotechnology Information (NCBI). The GSE181674 dataset includes samples from control islets and T1D patients from human pancreas samples obtained from deceased organ donors.^[9] To elucidate the biological process predictions, Shiny GO v0.61, a user-friendly tool, was employed for Gene Ontology (GO) enrichment analysis.^[11]A dataset on the analysis of blood biomarker levels was obtained from a publicly available repository containing information on patients with T1D.^[10]

RESULTS

Differentially expressed genes

The initial quality checks are shown in Figure 1. The PCA plot illustrates that the ND are control samples and T1D cases clustering patterns of the samples based on their gene expression profiles. The Gene Set Enrichment Analysis (GSEA) graph shows a moderate to strong enrichment of the gene set in the phenotype being analysed. A descriptive analysis was conducted to compare low glycated haemoglobin (HbA1c) to high HbA1c, which was not affected by the duration since the patient was diagnosed with T1D. Figure 2 shows a heat map of DEGs in T1D patients' islet sympathetic innervation. Many upregulated genes like MYCN, KLHL1, and BPI, shown in the heatmap, are involved in immune system function. Their role in cellular stress responses is central to the autoimmune destruction of pancreatic beta cells in T1D. Downregulated genes like FABP6, CTNNA3, and DNMT3B are related to metabolism, cell signalling, and epigenetic regulation. Together, Kyoto Encyclopedia of Genes and Genomes (KEGG) and Reactome pathways show that sympathetic neurons overexpress cytokines and chemokines that can influence T1D severity by altering lipid metabolism and insulin resistance. Identifying novel DEGs not previously associated with T1D, such as MAS-related GPR family member E and family with sequence similarity 182 member B, opens up new avenues.

Close

Figure 1:

A comprehensive visualisation of gene expression data, incorporating heatmaps of differentially expressed genes (DEGs), and gene set enrichment analysis (GSEA) of upregulated genes.

Export to PPT

Close

Figure 2:

Pathway analysis visualisation of gene expression data, with differentially expressed genes (DEGs). KEGG: Kyoto Encyclopedia of Genes and Genomes.

Export to PPT

Analysis of biomarkers in type 1 diabetes patients’ blood

The dataset was obtained from^[10], which contained 1222 T1D patient samples, with 620 females and 602 males, of which 198 had a relative with T1D. The patients' ages ranged from 2 to 87 years, and their glycated haemoglobin levels were between 5.45 and 14.63. Table 1 includes creatinine levels, blood urea nitrogen, and immune markers in this dataset, which reveals a positive correlation with HbA1c levels. Blood levels of biomarkers are often reported as percentiles (25th, 50th, and 75th) rather than just the mean and standard deviation. It adds robustness to outliers and gives a clear picture of the dataset. Various genetic and environmental lifestyle factors can influence blood biomarker levels, and percentiles are less sensitive to these outliers than the mean and standard deviation. By focusing on percentiles, we better understand the distribution of values for most of the population, excluding extreme cases. The interquartile numbers are reported in Figure 3. This twofold picture helps interpret the biomarker levels in our research context.

Close

Figure 3:

Distribution of immune marker levels between low and high HbA1c cases

Export to PPT

Most biomarkers show differences between the low- and high-HbA1c groups, but the magnitude of the difference varies [Figure 3]. Each box plot compares the distribution of the immune marker levels between the low and high HbA1c categories. Some biomarkers, like CRP (C-Reactive Protein), sIL2Ra (Soluble Interleukin-2 Receptor α), and IGFBP1, appear to have notably higher levels in the high HbA1c group. Other biomarkers, like IL8 (Interleukin-8) and MIP1B (Macrophage Inflammatory Protein-1β), show less pronounced differences between the two groups.

The pattern of DEGs in T1D sympathetic neurons is indeed reflected in changes in the levels of soluble biomarkers. The heatmap provides a visual representation of the correlations between the immune markers in the blood of T1D patients. Strong positive correlations were observed in sTNFRI and sTNFRII (0.87), sIL2Ra and sTNFRII (0.71), and sIL2Ra and sTNFRI (0.70), highlighting the significance of sympathetic innervation in the pathogenesis of T1D. Figure 3 suggests that an increase in glycated haemoglobin is associated with an increase in these soluble immune factors, chemokines, matrix metallopeptidase, interleukins, and adhesion molecules related to DEGs. Overall, the observed concordance between pathways predicted by upregulated gene expression and associated increased serum levels of linked biomarkers strengthens the validity of our findings. Our results lay a foundation for understanding the sympathetic molecular pathways of diabetic effects on nerve cells and propose avenues for future studies.

DISCUSSION

Sympathetic neurons affect both endocrine function and immune responses within the pancreas. The GSEA results suggest that in T1D, the complex interplay of immune system dysregulation, altered cell signalling and adhesion, and metabolic changes contributes to the development and progression of the disease. TH (tyrosine hydroxylase) is an enzyme that synthesises catecholamines, which can repress insulin secretion. Upregulation of TH in pancreatic β-cells is restricted by DNA methylation during development. Loss of TH promoter methylation in response to chronic overnutrition increases the number of TH+ β-cells and correlates with impaired β-cell function. The upregulation of MYCN could be a compensatory mechanism in response to the autoimmune destruction of pancreatic beta cells. MYCN (MYCN protooncogene, bHLH transcription factor) is implicated in cell growth and differentiation. Its upregulation may indicate an increased inflammatory response, which could be associated with more aggressive disease progression. Genes involved in metabolic processes can also influence disease severity. For example, upregulation of Fatty Acid Binding Protein 6 (FABP6) may indicate altered lipid handling, which could exacerbate insulin resistance. The 6-Phosphofructo-2-kinase/fructose-2,6-bisphosphatase 2 (PFKFB2) regulates glycolysis and may indicate a compensatory mechanism in response to impaired insulin secretion. Results show sympathetic neurons’ roles in cell signalling and survival pathways. Catenin Alpha 3 (CTNNA3) is involved in cell adhesion and associated with changes in cellular interactions affecting beta cell survival and function. Sodium Voltage-Gated Channel Beta Subunit 1 (SCN1B) is involved in neuronal excitability and may influence altered electrical activity in beta cells, impacting insulin secretion.

Differential expression analysis of gene expression profiles in intrinsic sympathetic nerve fibres between diabetic and non-diabetic patients identified several genes associated with viral protein interactions and cytokine/cytokine receptor pathways. The virus-related pathways likely play a role in the immune response, as confirmed by a recent report on zinc finger NFX1, which is identified as a critical player in the antiviral response in beta cells.^[18] Molecular function analysis revealed peptidoglycan binding, oxidoreductase activity acting on paired donors with incorporation or monooxygenase activity, and heme binding (tetrapyrrole and iron ion binding). Chronic exposure to tetrachlorodibenzop-dioxin can impair glucose regulation and insulin secretion, and the aryl hydrocarbon receptor (AhR) in β-cells plays a crucial role in this process. This suggests that aryl hydrocarbons' effects on β-cell function drive metabolic changes in the body.^[19] Further damage by these molecular functions can lead to the formation of adducts, resulting in permanent mutations.

While previous studies have established sympathetic neuronal changes in T1D, the linked mechanisms and pathways remain poorly understood.^[8] Research has shown that specific gene expression signatures can predict disease progression in T1D.^[12] For example, studies have identified soluble biomarker levels in whole blood that correlate with the decline in beta-cell function and the progression of autoimmunity.^[13] We observed upregulation of chemokines, matrix metalloproteinases, interleukins, and adhesion molecules, which aligns with findings in T1D patient blood samples. This study identified the most robust relationships observed between sTNFRI, sTNFRII, and sIL2Ra, reflecting a common pathway or response in the immune system. The findings demonstrated a remarkable concordance for the involvement of inflammatory processes, such as chemokines, cytokines, and interleukins.^[14] Changes in sympathetic signalling affect how immune cells interact with islet cells, which leads to beta-cell destruction.^[20] While the specific link to T1D is not explicitly mentioned, tyrosine kinase inhibitors have been shown to reverse T1D in nonobese diabetic mice, suggesting a potential role in the disease.

B cells' role in the development of T1D is to present islet autoantigens to T cells. Disrupting B-cell function can prevent T1D, and targeting autoantigen-specific B cells may be a potential therapeutic approach. B cell receptor somatic hypermutation increases affinity for autoantigen. However, immune tolerance mechanisms may limit highly autoreactive clones, as reviewed in the role of B cells in T1D pathogenesis.^[21] Pathways like viral protein interaction with cytokine and cytokine receptors suggest a possible role of viral interactions in the disease. Immune checkpoint modulators are being studied for their potential to treat T1D by targeting the immune response. However, conflicting results have been reported, and further research is needed to fully understand the disease's complex pathophysiology.^[22] The downregulation of genes in T1D indicates a complex interplay among immune dysregulation, metabolic dysfunction, and epigenetic modifications. Downregulation of DNMT3B in T1D may alter methylation patterns in other genes, potentially contributing to disease pathogenesis. The downregulated genes, such as PDX1-associated lncRNA and IAPP suggests that PDX-1 could be a potential target for early interventions in autoimmune diabetes.^[23] Additionally, pathways related to energy production (citrate cycle) and protein processing appear to be downregulated, potentially contributing to beta cell dysfunction. In the future, single-cell sequencing can map cellular heterogeneity and analyse IL6, IFN-α, and IFN-γ expression during cell dedifferentiation.^[14]

Genetic and environmental factors, including viral infections, cause T1D. The viruses can directly infect insulin-producing cells, modulate the immune system, and increase stress on the cells. The recent COVID-19 pandemic may also impact the development of T1D. Understanding the relationship between viruses and autoimmunity is essential for developing treatments and prevention strategies.^[24] The effectiveness of Interleukin-2 therapies in treating T1D was assessed through a study using an ultra-low dose of IL-2.^[12] However, this approach did not protect against hyperglycemia or improve Treg and Breg cell response. Alternative strategies or immunotherapies may be more effective in treating T1D.^[25] The current analysis highlights statistical associations but lacks the temporal resolution to infer causality. Our research informs prospective clinical designs and experimental models to substantiate sympathetic neuronal function in the pancreas and to understand T1D aetiology.

Further research is needed to understand the sympathetic mechanisms of insulin secretion. Therefore, our results on the pathogenesis of T1D and the potential role of sympathetic neuron activity are essential. The present study has provided valuable insights into the pathogenesis of T1D and implicated the potential role of sympathetic neuron activity. Future studies should include larger, more diverse cohorts and conduct functional analyses to better understand how sympathetic neuronal activity influences the immunological processes implicated in T1D.

CONCLUSION

Our study bridges the gap in studying nervous system involvement in Type-1 diabetes. We comprehensively analysed sympathetic innervation gene expression. We provide evidence linking sympathetic neurons to the pathogenesis of T1D. To elucidate this role, this study employed bioinformatics and statistical analysis of blood test results. We examined the transcriptional profiles of deceased T1D patients and discovered significant upregulation of genes implicated in inflammatory chemokines, matrix metallopeptidases, interleukins, and adhesion molecules. Evidence from biomarker levels supports post-transcriptional modifications and metabolic changes relevant to disease pathology. These congruent data strands suggest that neuronal processes mediated by the sympathetic nervous system may contribute to the changes in biomarkers and T1D. The study enables future investigations to pinpoint the precise roles of the identified genes. Overall, the findings advocate integrating our data insights toward developing targeted therapies for T1D management.