MACC1 seen as biomarker for PAH in bioinformatics study

Machine learning, a form of AI, also used to make comparisons of cellular data

Margarida Maia, PhD avatar

by Margarida Maia, PhD |

Share this article:

Share article via email
This illustration of a DNA strand highlights its double helix structure.

MACC1, a gene that’s been linked to overgrowth of cancer cells, is more active in the lungs of people with pulmonary arterial hypertension (PAH) than in healthy individuals, a study out of China found, suggesting it could be used as a diagnostic biomarker for the disease.

Researchers also observed that knocking down MACC1 to reduce its activity in lab-grown smooth muscle cells — cells that wrap around blood vessels to control blood flow — made them grow less, move slower, and become more prone to death.

It’s thought that smooth muscle cells act up early in PAH, so “these findings suggest that targeting MACC1 may offer a promising therapeutic strategy for treating PAH,” the researchers wrote. “Further clinical studies are warranted to evaluate its efficacy.”

The study, “Identification of MACC1 as a potential biomarker for pulmonary arterial hypertension based on bioinformatics and machine learning,” was published in Computers in Biology and Medicine.

Recommended Reading

Two genes ID’d as therapeutic targets for PAH

PAH occurs because of a narrowing of the pulmonary arteries, the vessels that carry blood from the heart to the lungs. The resulting high blood pressure leads to trouble breathing and forces the right side of the heart to work harder to pump blood to the lungs.

What causes the pulmonary arteries to narrow isn’t entirely clear, but it’s known to involve an overgrowth of pulmonary artery smooth muscle cells (PASMCs), which control blood flow into the lungs by relaxing or contracting to adjust the size of the blood vessels.

The study and its findings

To find out what drives PASMCs into overgrowth, the researchers used machine learning (a branch of artificial intelligence that enables a computer to learn from data) and bioinformatics to compare sets of genetic data from people with PAH versus healthy individuals.

The goal was to identify differentially expressed genes, which are genes that are more active (upregulated) or less active (downregulated) in a certain condition. This identifications helps scientists understand which genes are involved in the molecular mechanisms leading up to disease.

The sets (GSE113439 and GSE117261) were from the Gene Expression Omnibus public repository and included genetic data from lung samples of 73 people with PAH and 36 healthy individuals. Hundreds of genes were differentially expressed in PAH.

To narrow down the list of genes, the researchers made use of machine learning and weighted gene co-expression network analysis, a type of bioinformatics analysis that groups genes that work together in biological processes.

The list was reduced to four hub genes — ANKRD36CCOL14A1, MACC1, and POSTN — which play central roles in those networks. These genes are involved in the structure of the extracellular matrix (a web of proteins and other molecules that gives support to cells in the body), cell migration and inflammation, among other processes.

To validate their findings, the researchers drew on another set, GSE53408, which included genetic data from lung samples of 12 people with severe PAH and 11 healthy individuals. They found that all four hub genes were indeed upregulated in people with PAH.

One of the genes, MACC1, was upregulated in circulating blood cells from people with PAH associated with scleroderma, a disease of the body’s connective tissue. These data came from yet another dataset, GSE22356. The levels of MACC1 protein were higher in the blood of people with PAH than in healthy controls.

The higher the MACC1 levels, the higher the World Health Organization (WHO) functional class, which indicates more severe disease. Higher levels of MACC1 were also linked to higher pulmonary artery pressure and shorter distance walked in six minutes.

“MACC1 concentration was a risk factor for clinical manifestations,” wrote the researchers, leading them to perform studies in animal models of PAH to better understand the role of MACC1 in disease progression.

They found that levels of MACC1 increased in response to platelet-derived growth factor-BB, also known as PDGF-BB, a signaling protein that helps PASMCs grow in number, thereby contributing to the progression of PAH.

Knocking down MACC1, that is, reducing the gene’s activity so that less protein is produced, made PASMCs grow less, migrate (move toward a stimulus) slower, and become more prone to a type of cell death called apoptosis. These findings suggest that MACC1 may prime PASMCs to adopt a cancerlike behavior (for example, overgrowth) in PAH.

Combining machine learning and bioinformatics made it possible to identify key genes in PAH, with “MACC1 emerging as a promising biomarker for early diagnosis and treatment of the disease,” the researchers wrote.

“However,” they concluded, “further investigations are warranted to validate our findings and ascertain the therapeutic potential of MACC1 as a therapeutic target for PAH intervention.”

A Conversation With Rare Disease Advocates