October 1, 2024 | In a first for genetic diagnostics, a pair of Nobel-Prize-winning technologies are being repurposed for obtaining RNA from disease-causing genes that are expressed in neither blood nor skin. The feat comes from the Neurobiology Research Group at the University of Adelaide’s School of Biomedicine, led by Lachlan Jolly, Ph.D., which aims to resolve the pathogenicity of 300 variants of uncertain significance (VUSs) under a government-funded, Australia-wide study.
The overarching mission is to provide a diagnosis for the roughly 30% of individuals whose DNA test returns a variant which is likely to be the cause of their disorder but requires confirmation at the level of RNA, says Jolly. But for one-third of all disease genes, RNA is not expressed in clinically accessible tissues and therefore the second-tier RNA sequencing test won’t solve the case. These are termed “silent genes.”
For the roughly 1,500 silent disease-causing genes, the new RNA diagnostic method now under investigation could be a gamechanger, he explains. One of two approaches can be applied, either engineering a skin fibroblast to turn the silent gene on or directing a skin fibroblast to morph into another cell type where the silent gene is naturally switched on. But in both instances the process begins with a small sample of skin taken from a patient.
Most recently, the “transactivation” and “transdifferentiation” methods for RNA variant assessment were used in a study where rare neurodevelopmental disorders were being identified (American Journal of Human Genetics, DOI: 10.1016/j.ajhg.2024.06.018). Although the approach can be applied to any genetic disease, researchers started with brain-related disorders associated with the largest portion of silent Mendelian genes that are insufficiently expressed in clinically accessible tissue, says Jolly. The same access issue also occurs for other tissue-specific conditions such as those affecting the kidney, liver, or heart, he adds.
The transdifferentiation technique—based on the discovery that mature cells can be directly reprogrammed into a different cell identity—turns skin cells into cell types (in this case neurons) found in disease-manifesting tissue so the silent disease becomes switched on. In contrast, the transactivation approach—based on CRISPR technology used in the field of functional genetics—engineers skin cells to “turn on any gene we like,” says Jolly. “It’s relevant to the entire 1,500 disease genes.”
Currently, the only other way to obtain the RNA transcript of a disease-causing brain gene is through a sample of patient brain tissue, which is rarely available or advisable, Jolly says. Such individuals therefore often never receive a genetic diagnosis, and they number in the “hundreds of thousands” worldwide.
The value of a genetic diagnosis for patients is that it provides them access to clinical trials, around 13,000 of which are underway testing treatments targeting the genetic cause of a rare disease, as well as any relevant precision therapies already on the market, says Jolly. Studying the mechanisms by which the RNA molecule is affected is also opening avenues for development of a new generation of therapeutics targeting the process of RNA splicing.
Most of the variants being studied by Jolly and his team are caused by abnormal alternative RNA splicing, which results in either the complete loss or dysfunction of proteins contributing to disease development. The access to RNA from these disease genes also enables the design and development of antisense oligonucleotide (ASOs) therapies, he reports. Several ASOs have already received marketing approval in Australia and the U.S., including nusinersen (Spinraza) for the treatment of spinal muscular atrophy and eteplirsen (Exondys 51) for the treatment of Duchenne muscular dystrophy.
Jolly and his colleagues have spent the past five years optimizing and modifying their two-pronged diagnostic technique on human fibroblasts to make it possible to get RNA from genes which otherwise is only available from inaccessible tissues. For the latest study, they succeeded in inducing the expression of 40 silent genes using the transactivation approach, and 193 silent genes using the transdifferentiation approach, in human dermal fibroblasts.
“For transactivation, we did it one gene at a time,” he says, with each gene test taking just a matter of days. For patients with epilepsy suspected to be caused by variants in the SCN1A gene, which is expressed primarily in the brain, the researchers “delivered the synthetic gene transactivation tool to the patient’s fibroblasts and... activated the expression of the SCN1A gene. Once the gene was activated, we could get the RNA and do the RNA-based diagnostic tests.”
On the other hand, the transdifferentiation of skin cells to neurons took only two to three weeks, which is significantly shorter than other reprograming methods. Given that more than a third of silent Mendelian genes are neurological, Jolly says, investigators reasoned that the transdifferentiation approach could induce the expression of a large proportion of them with a single method, as indeed it did (193 total). The technique was used to investigate the RNA impact of a set of three variants found in the intellectual disability gene PAK3 using patient skin fibroblasts.
Both tools are useful. “With transactivation we can just pick and choose whatever gene, whatever disease, whatever tissue [we want] and get access to it, so it’s a more diverse tool,” says Jolly. “We can tackle many genes, but it’s one by one. Transactivation turns on a whole host of genes but is specific to a particular tissue [e.g., the brain].”
The final step in both cases is RNA sequencing, an established technology in many diagnostic laboratories around the world, rendering an outcome as to whether a variant of uncertain significance is or is not disease-causing, he adds. The longer-term hope here is creation of an off-the-shelf test kit that labs could use when encountering a VUS found in genes not expressed in accessible tissue to “turn those genes on in skin cells,” so they can readily get the RNA they need for that second-tier test.
Variants in silent Mendelian genes account for more than 22.2% of VUSs in ClinVar, a public archive of human genetic variations and their clinical significance hosted by the National Center for Biotechnology Information in the U.S., the study team reports. This equates to 283,353 individuals without a diagnosis, and the count continues to climb.
The latest findings of Jolly and his team are part of the PERSYST (Pathogenic Evaluation of Recalcitrant Variants by Systematic Transactivation) study, a national program in Australia actively recruiting individuals living with a medical condition caused by VUSs within rare silent disease genes. PERSYST aims to provide RNA-based functional evidence of the pathogenicity of up to 300 VUSs predicted to alter splicing in genes not expressed in clinically accessible tissues.
Skin samples obtained from enrolled individuals are being sent to Jolly’s lab where the transactivation technique will be used on patient skin cell lines to obtain the RNA and then sequence it and submit the findings back to the referring clinician. Recruitment for the study began last fall and will continue through October 2027, when grant funding from Australia’s Department of Health and Aged Care ends.
Research reports have already been prepared and delivered that have had a clinical impact, achieving reclassification of variants for Australian individuals, Jolly says. Resolving the pathogenicity of VUSs opens avenues for different types of therapies and care and, potentially, access to clinical trials. The prerequisite for trial participation is having a “solid genetic diagnosis,” he notes.
Outside of PERSYST, Jolly and his team have also helped resolve VUS from individuals in other parts of the world on a research basis. Over the short term, he says, the RNA variant assessment assay will remain a set of research tools aiding in the quest to resolve the genetic architecture of genetic disease. Work toward an off-the-shelf diagnostic will require a series of benchmarking studies against RNA extracted from disease tissue and to demonstrate the solution’s robustness in terms of test results leading to reclassification of genetic variants.
The research team is currently conducting tests involving transactivation of a muscle gene, side by side with the RNA from a patient’s muscle, Jolly reports. Acceptance of assay results by clinical pathology labs has to date been uniformly positive and clinically impactful.
The research pipeline has been set up with “good purpose,” he says, so that in another 10 or 15 years the gene activation technology might be so robust that the process could be standardized and fully automated. As imagined, diagnostic labs would set up a liquid handling robot to culture patients’ skin cells and add reagents to cells. With the push of a button, clinical geneticists could extract RNA to analyze via their sequencing pipeline.
Reclassifying a VUS used to take months if not years or decades and, if there was not enough evidence for labs to draw a definitive conclusion, sometimes never happened, Jolly says. “Without knowledge of what is causing the disorder, there is very little hope they will receive appropriate or even precision treatment.”
The kind of genomic solutions now being built could vastly reduce the number of people with a rare genetic condition facing a lifetime of unknowns. “Probably millions of individuals have variants which are under-interpreted,” says Jolly, noting that about 200,000 individuals represented in ClinVar alone have a VUS in a disease-causing gene expressed in neither skin nor blood and this is “probably [just] the tip of the iceberg.”