Computational Model Identifies Candidate Therapies Based on MG Risk Genes

Ana Pena, PhD avatar

by Ana Pena, PhD |

Share this article:

Share article via email
myasthenia gravis severity

Ice Test For MG

Researchers have identified five candidate therapies for treating myasthenia gravis (MG) by using a new computational model that also predicts key biological processes involved in the disease.

The study, “Building the drug-GO function network to screen significant candidate drugs for myasthenia gravis,” was published in the journal PLOS ONE.

Drug repurposing — using old drugs for new indications — “has become an attractive form of drug discovery that can save time and money compared to developing new drugs,” researchers said.

In this study, they created a computational model for uncovering important biological functions linked with MG and screening for potential therapies from a database of existing compounds based on the biological processes they target.

In genetics, any given gene can be assigned into one or more categories called gene ontology (GO) classes, which represent that gene’s function, location within the cell, and the broader biological processes in which that gene is involved.

To create the computational model, researchers started with 258 genes associated with MG risk in prior studies to identify risk-related GO biological processes. Most of these risk genes give instructions for the production of cytokines — signaling proteins that serve as important messengers of the immune system.

From there, researchers build a computational network that displayed the connections among all biological functions predicted to play a role in MG and quantified the magnitude of these interactions.

This network uncovered 238 biological processes that appeared to be key for the disease, as they spanned many MG risk genes.

Specifically, they discovered that “the positive regulation of NF-kappaB transcription factor activity may be one of the most important GO functions in the mechanism of MG.”

NF-kappaB is a family of proteins termed transcription factors, which work as positive switchers of gene activity. They are critical regulators of many biological functions, including immune and inflammatory responses. They are persistently active in diseases such as arthritis, chronic inflammation and neurodegenerative diseases.

Next, researchers built a drug-GO function network to weigh the relationships between a list of existing drugs, whose targets are known, and the biological functions associated with MG.

They used this network to screen for drugs with strong associations with MG risk that could, therefore, represent potential candidates for treating the condition.

Five candidate drugs were identified — glucosamine, apremilast (brand name Otezla), adalimumab (brand name Humira), and etanercept (brand names Enbrel, Erelzi, Brenzys) and polaprezinc.

Two of them, adalimumab and etanercept, have already been investigated to treat MG; these and apremilast are FDA-approved to treat many inflammatory conditions including arthritis. Glucosamine is a naturally occurring sugar protein involved in the formation of cartilage. It has been used to relieve joint pain, swelling, and stiffness caused by arthritis. Polaprezinc is a compound that increases the levels of various antioxidant enzymes. It is approved in Japan for treating stomach ulcers.

Based on the common links between candidate drugs and the GO functions affected by them, researchers predict that the identified compounds may target MG via two main pathways: graft-versus-host disease (GVHD), a complication where bone marrow or stem cell transplants begin attacking the host body, and type I diabetes.

These two pathways can involve self-reacting immune responses. In fact, myasthenic symptoms are frequently associated with other symptoms of chronic GVHD, and MG has been reported as a rare complication of GVHD after stem cell transplants. Moreover, type 1 diabetes increases the risk for other autoimmune diseases and is related to a genetic risk for these diseases.

“In conclusion, we compiled a catalog of MG risk genes and identified risk GO functions, drugs and risk pathways,” researchers said. “We also investigated the complex connection among MG risk genes, drugs, and GO functions by constructing a [drug-GO function network] and we identified five candidate drugs.” 

The study, they say, will “reveal a new perspective on the mechanisms and novel treatment strategies of MG” as well as “provide strong support for research on GO functions.”