Changes in Oil on Skin’s Surface May Help to Diagnose Parkinson’s
Changes in the genetic material of the sebum — the “oil” on the skin’s surface — are evident in people with Parkinson’s, and analyzing this skin surface lipid may aid in diagnosing the disease, scientists report.
Their study, “Non-invasive diagnostic tool for Parkinson’s disease by sebum RNA profile with machine learning,” was published in Scientific Reports.
Sebum is an oily substance secreted onto the skin to help it stay moist. Parkinson’s patients commonly have abnormally high sebum secretion, for reasons that are not well understood.
Recent studies have shown that sebum contains measurable amounts of genetic material, specifically RNA. When a gene in the cell’s DNA is “read,” the cell makes a copy of the gene out of RNA, which is then usually sent to the cell’s protein-making machinery to provide instructions for protein production. Some RNA molecules have additional functions, helping to regulate various aspects of cellular activity.
By analyzing the RNA contained in sebum — termed the skin surface lipid RNA, or SSL-RNA — investigators may be able to make inferences about the health of an individual. For example, prior research has indicated that skin inflammation may lead to abnormal SSL-RNA.
A team of researchers in Japan examined the SSL-RNA in people with and without Parkinson’s disease.
The scientists analyzed SSL-RNA collected using an oil-blotting film from two groups: one containing 15 people with Parkinson’s and 15 healthy people serving as controls, and another with 50 Parkinson’s patients and 50 healthy controls. The first cohort contained both male and female participants, while the second had only males; all participants were matched for age.
By analyzing the SSL-RNA profiles of these groups using advanced computer algorithms, the researchers demonstrated that “the SSL-RNA expression profiles of PD [Parkinson’s disease] had different characteristics from the profiles of healthy controls.”
Specifically, examining particular genes expressed at different levels indicated abnormal function of mitochondria in Parkinson’s. Mitochondria are the so-called powerhouse of a cell, responsible for generating energy; mitochondrial dysfunction is thought to play a role in Parkinson’s development and progression.
Abnormalities were also found in genes related to immune function in the Parkinson’s SSL-RNA. Increased immune activity (i.e., inflammation) has also been implicated in Parkinson’s development.
The researchers then tested whether analyzing the SSL-RNA profiles with machine learning could be used to discriminate between individuals with or without Parkinson’s — a test that would be useful in a diagnostic setting.
To do this, the team calculated a statistical measure called the area under the receiver operating characteristic curve, or AUC. Simply put, AUC measures how well a given test can distinguish between two groups — in this case, Parkinson’s or not. AUC values can range from 0.5 to 1; higher values indicate better ability to differentiate.
The AUC of the researchers’ algorithm, which considered SSL-RNA alongside factors like age and sex, was just over 0.8, indicating relatively robust discriminatory ability.
This supports the “potential use of SSL-RNAs in a novel, non-invasive method for the detection of PD,” the researchers wrote.
Researchers stressed that this was a preliminary study done at a single institution; more research is needed to understand the role of SSL-RNA in Parkinson’s, and its potential uses.