A Gene-ius Approach is Opening Doors to a More Colorful Set of T1D Markers

A Gene-ius Approach is Opening Doors to a More Colorful Set of T1D Markers

Neha Majety

More often than not, you associate the term “beta” with a Greek life chapter or software that has not officially been released, but in the context of T1D, it is used to describe insulin-producing pancreatic beta-cells that face autoimmune destruction as a result of T1D onset. Produced by the immune system, islet autoantibodies (IAs) are also associated with T1D and are commonly used for its clinical diagnosis, as well as for predicting future T1D onset.  

Research has opened doors into both understanding and preventing beta-cell autoimmune destruction as a means to delay or avert T1D onset, as well as predicting T1D onset by identifying markers that lead to persistent islet autoantibodies. Read on to learn about recent breakthroughs in these fields of T1D research:

Researchers from Spain mapped stimulus-responsive enhancers linked to changes associated with human beta cells. They found:

  • Beta cells have a high plasticity based on changes observed in their transcriptome, proteome and 3D chromatin structure after exposure to pro-inflammatory cytokines.
  • To study chromatin dynamics, a model of islet cells in early-stage T1D was created based on observations in murine macrophages and dendritic cells. However, it has not been demonstrated in pancreatic islets thus far:
    • Model: Chromatin accessibility at some sites is due to tissue-specific transcription factors (TFs) binding. These sites are then activated by induction of inflammatory-response TFs.
  • T1D risk regions were also found to be enriched for human islet cytokine-responsive regulatory elements
  • Overall, this research opens the door to identification of T1D molecular mechanisms at the islet level

Weary of waiting for a solution? Well then you will be excited to hear that Wallkill Biopharma Inc. published a US patent for treating diabetes directed at functioning and senescent transgenic human beta cells. These cells can be used to express insulin in a hyperglycemic environment like that found in T1D patients. Here are some highlights of the invention:

  • CXCL12 (or CXCL13), acting as a fugetactic agent, is added endogenously – or exogenously – to the beta cells, since it has been reported to repel effector T-cells and recruit immune-suppressive regulatory T-cells
  • In one of several embodiments of the invention, the beta cells include a functional fugetactic agent expression vector that is designed to express the agent at levels that creates a buffer around the cell, allowing beta cells to resist immune cell attack while still functioning properly to produce insulin. 
    • Embodiments for administration: either via autologous from a diabetic patient or allogeneic methods

Moving past discussions of the beta cell, another group of researchers developed a machine learning model to predict the development of persistent IAs by using the TEDDY study, which collected environmental and biological data from subjects at an increased genetic risk for T1D from birth. In contrast to the common approach of regression modelling, the researcher’s model was a data-driven machine learning model that evaluated time-varying metabolics data compared with time-invariant risk factors. This method allowed for a large collection of potential risk factors and molecular markers to be used, as opposed to the few markers regression modelling uses. Their model found:

  • 42 features to predict the development of autoimmunity in TEDDY participants including short nucleotide polymorphisms (SNPs), traditional risk factors, metabolites and lipids 
  • The features were mostly (86%) associated with the three biological pathways of lipid oxidation, phospholipase A2 signaling and the pentose phosphate pathway, all of which suggests that these processes are key predictors for IA development.

Looking Forward:

Future work studying other immune-mediated stresses affecting beta cells at different stages of T1D may uncover more signals acting at the islet cell level. Additionally, data-driven machine learning models will give researchers a better understanding of the underlying biological mechanisms of T1D and the relationship between environmental factors and T1D onset.