NEW YORK (GenomeWeb) – Researchers have linked more than a dozen genomic loci to loneliness, and nearly 40 loci to participation in social activities, in a new genome-wide association study.
About a quarter of individuals in the UK over the age of 65 suffer from loneliness, which is also associated with increased mortality.
Using data collected by the UK Biobank, a trio of University of Cambridge researchers conducted a GWAS of nearly half a million people for loneliness that homed in on 15 loci. As they reported in Nature Communications today, the researchers also identified several loci associated with regular participation in social activities.
"Further study of these traits may identify novel modifiable risk factors associated with social withdrawal and isolation," senior author John Perry from Cambridge and his colleagues wrote in their paper.
UK Biobank participants answered three questions that touched on loneliness, frequency of social interactions, and ability to confide in someone, which the researchers bundled together in a multi-trait GWAS with an effective sample size of 487,627 people.
Through this, they linked 15 genomic loci to loneliness at genome-wide significance. They also estimated the heritability of loneliness to be about 4.2 percent, and found that a genetic risk score based on the 15 loci they uncovered was able to predict loneliness in a separate set of 7,556 individuals.
Using gene expression and epigenetic data, the researchers traced these loci and found that they were enriched within regions near genes expressed in brain tissues such as the cerebellum, basal ganglia, and cortex. Further, in brain tissue samples from the Genotype-Tissue Expression (GTex) project, they uncovered eQTL effects with eight gene transcripts whose expression levels were linked to loneliness susceptibility, including BPTF, GPX1, and MTCH2. BPTF, the researchers noted, encodes a transcriptional regulator that is active in the fetal brain and has been linked to neurodegenerative diseases.
The researchers also identified 36 complex traits that were genetically correlated with loneliness, including a strong positive overlap with neuroticism and depressive symptoms and a strong negative overlap with subjective well-being and years of education. They also uncovered a positive causal effect of BMI on loneliness, though not for loneliness on BMI.
The researchers also conducted three other GWAS on links to sociability in the UK Biobank populations: regular attendance at a sports club or gym, pub or social club, or religious group. They identified 38 loci with genome-wide significance for these three traits.
The allele most strongly associated with pub or social club attendance is a missense allele in the gene that encodes alcohol dehydrogenase, which was not linked to either gym or religious group attendance. Sport club attendance, meanwhile, was more strongly associated with a signal recently linked to risk taking, while the others were not.
A number of loci, however, were shared across all three traits. For instance, one signal near BARHL2 was linked to all three, as well as to educational attainment, chronotype, and age at first sexual intercourse. Another signal near CAMV was also linked to all three social activities as well as to educational attainment, resting heart rate, and HDL cholesterol, among other traits.
All four of the traits the researchers investigated — loneliness, pub-going, gym attendance, and religious meeting participation — had genetic associations enriched for genes expressed in the central nervous system. Using the GTex data, they noted that pub-going genes were enriched for expression in the amygdala, and religious group attendance with the frontal cortex.
"We find evidence for shared genetic effects across social traits, in addition to more specific pathways that drive engagement in particular activities," the researchers wrote. "Our findings also suggest a causal relationship between cardio-metabolic health and social isolation/mental health, an observation which warrants further investigation using other experimental approaches."