A genetic disorder may affect one person differently from another, even if they both have the same mutation. This anomaly has long been attributed to the varying effects of environmental or other competing genetic factors. But now a new study from researchers at the Flatiron Institute and Princeton University has revealed how a third factor — randomness — can also impact disease development. This study in flies may help explain why there is so much variability among people with the same genetic mutation, especially in neuropsychiatric conditions such as autism and schizophrenia. The research team reports their findings in the March 13 issue of Current Biology.
“The question of what is sufficient for disease occurrence, in addition to a mutation, is fundamental,” says Stanislav Shvartsman, head of the Developmental Dynamics group in the Flatiron Institute’s Center for Computational Biology, and professor of molecular biology at Princeton. “We designed an experiment to rule out the effects of genetic and environmental factors on disease, and to our surprise, we saw the significant role played by randomness,” he says.
Shvartsman’s lab led the research, with postdoctoral researcher Robert Marmion as lead author and contributions from graduate students Alison Simpkins, Lena Barrett and David Denberg. Susan Zusman, Jodi Schottenfeld-Roames and Trudi Schüpbach also contributed to this work.
To investigate the issue of randomness in disease, the researchers set their sights on a class of developmental disorders associated with mutations in the RAS family of genes. These genes encode proteins that are important in signaling during development, and the same RAS mutation, or genotype, can result in many different physical effects, or phenotypes — ranging from short stature to facial abnormalities to cancer. Thanks to increasingly sophisticated DNA sequencing, a more complex RAS picture has emerged. For example, in the UK Biobank, a repository of genetic and health information assembled from half a million U.K. participants, RAS mutations have been found in people with no abnormal phenotypes.
In a series of experiments designed to disentangle the factors responsible for these varying phenotypes, scientists engineered fruit flies to have a RAS mutation. In flies, the mutation would manifest as several visible defects, such as malformed wings, that researchers could see and count. The flies were bred with each other and grown in a controlled environment, with the intention of reducing the ‘noise’ from the genetic and environmental factors that can influence disease development. To the researchers’ surprise, the vast majority (approximately 90 percent) of the flies’ wings were normal. “This was striking,” says Marmion, “and led us to consider the role of randomness more closely.”
The scientists then developed a conceptual model showing that the biochemical signal that causes a defect — in this case, a malformed wing — is not 100 percent effective. Rather, it has a probability with a random distribution. “Even in a normal or ‘wild type’ fly, every now and then you see a physical defect,” says Simpkins. In the model, a mutation simply shifts the distribution to lead to a higher probability of defects. The shape of the distribution remains unchanged, and still results in many cases where the mutation does not result in a physical defect. This could explain why most of the flies with the mutation did not have defective wings.
To further reduce any genetic and environmental variability in their experiments, researchers homed in on the tracheal system of the fruit fly. The cells in that system form on both sides of the fly’s body, giving the researchers a chance to study the system twice within the same organism. “Since within an individual the genetic makeup is the same for all organs,” Marmion explains, “you can completely rule out the effects of genetic factors.” This experiment also rules out environmental factors, as it would be unlikely that two sides of the fruit fly’s body would be subject to different conditions.
When the researchers counted the tracheal defects on the right and left sides in flies with a RAS mutation, the results were independent of each other, supporting the idea of a random distribution. Because the two tracheal systems in each fly had identical genotypes and environments, the researchers concluded that the variation seen between the animals’ right and left sides was driven by randomness.
“What’s exciting here is that they saw different things on two sides of the same organism. They showed randomness at play, and that is so hard to study,” says Bruce Gelb, professor of pediatrics and genetics and genomic sciences at the Icahn School of Medicine at Mount Sinai, where he is the dean for Child Health Research. (Gelb was not involved in the study.) “If we want to make progress in human genetics, studies like these are essential. We have to figure out what contribution each of these ‘buckets’ — genetics, environment and randomness — makes to disease,” says Gelb.
This work also lays the foundation for additional studies on the nature of phenotypic variability in disease, especially in neuropsychiatric conditions where big questions loom about the connection between genotype and phenotype. “These results will push scientists to explore the biological mechanisms that underlie this randomness and that contribute to the deep phenotypic variability we see in disease,” says Kelsey Martin, executive vice president of the Simons Foundation Autism Research Initiative and the Simons Collaborations in Neuroscience, who was not involved in the study.
Going forward, Simpkins plans to study the effects of the mutations at an earlier point in fruit fly development. “We want to know when and how the cells with defects arise,” she says. The lab is also planning additional follow-up studies looking at how the mutations affect the flies’ behavior. “It was in flies that scientists first saw that genes are arranged in chromosomes,” says Shvartsman. “Flies will continue to be extremely useful in telling us about humans.”