New research: saturation genome editing of RNU4-2
Written by Nicky Whiffin
Last week, we posted a preprint describing an experimental approach to study the effects of genetic variation in the RNU4-2 gene.
This blog post aims to break down the key findings from this work, primarily for those who are not used to reading academic papers.
What is a preprint?
A preprint is an early version of a scientific paper which has not yet gone through the process of ‘peer review’. During peer review, scientists who are unconnected to the original research team (‘reviewers’) check that all of the findings of the paper are robust and presented clearly, and make suggestions for improvement. Only after these reviewers are happy, is the paper published in a scientific journal.
We post our work as preprints as the peer review process can take months. A preprint allows other scientists to see and build upon research far sooner - enabling faster scientific progress.
Background
We first shared an earlier preprint on RNU4-2 in April 2024, which was later published in the journal Nature. In this previous paper, we showed that DNA changes (also called ‘variants’) in RNU4-2 cause a neurodevelopmental disorder. With the help of patient families, this disorder was later named ReNU syndrome.
It might be useful to read the blog post I wrote about this previous paper before reading the rest of this post.
What is this new paper about?
We used an experimental approach (that means lab coats and pipettes rather than computers) to look at over 500 different DNA changes in RNU4-2. We made cells with each of these DNA changes (using an approach called CRISPR to edit the DNA). We then looked at how well the cells with each DNA change survived in a dish - we know that RNU4-2 is a super important gene, so if the variant causes RNU4-2 to not function properly, the cells will not survive. If fewer of the cells with a particular DNA change survive, we therefore assume that that DNA change is more damaging to the cells as it has a greater effect on the function of RNU4-2. We then create a ‘score’ for every one of these DNA changes, based on how damaging they are in the experiment, allowing us to compare different variants.
Figure explaining the experimental approach.
What did we find?
Here are three key things that we discovered:
(1) DNA changes that cause ReNU syndrome are more damaging than DNA changes that we see in healthy individuals
This is not a surprising result. In fact, it is exactly what we were hoping to see. But it tells us that the experiment is working: it can distinguish the DNA changes that disrupt RNU4-2 function from those that don’t.
This allows us to predict which of the variants that we haven’t seen yet in an individual with ReNU syndrome, we might see in a patient in the future. This will help us diagnose them with ReNU syndrome more quickly. It also means that for some individuals with a DNA change in RNU4-2, we can say that they are unlikely to have ReNU syndrome, as their variant looks more like those that we see in healthy individuals.
We can also refine what we know about which parts of the RNU4-2 gene are important for ReNU syndrome. In our previous paper, we reported that DNA changes in individuals with ReNU syndrome were found in a stretch of 18 DNA base-pairs (represented as letters) in the middle of the RNU4-2 gene. In this preprint, we show that ReNU variants are found in two smaller regions of 9 base-pairs and 4 base-pairs.
It is remarkable that around 100,000 individuals globally who are predicted to have ReNU syndrome have DNA changes in just 13 of the over 3 billion base-pairs of our genetic code!
(2) Different DNA changes may cause ReNU syndrome that is more or less severe
As we have a score for every DNA change, we can look at how those scores compare for the different DNA changes that we know cause ReNU syndrome. We split the variants into two, based on those with really strong experimental scores (not many cells survived), and those with only moderate scores (more of the cells survived). We then showed that individuals with the variants with stronger scores (including the n.64_65insT variant that most ReNU individuals have), had more severe symptoms.
For example, 93% of ReNU individuals with strongly scoring variants are non-verbal or speak only a few words. In contrast, 94% of ReNU individuals with moderate variants could speak simple sentences or had normal speech. We also saw a difference in the reported severity of developmental delay more broadly, but interestingly, the proportion of individuals with seizures was not different between the two groups.
(3) ReNU syndrome is not the only neurodevelopmental disorder caused by variants in RNU4-2
ReNU syndrome is a dominant disorder. That means that only one copy of the RNU4-2 gene (we each have two copies, one from mum and one from dad) needs to be disrupted to cause the disorder. We never see ReNU syndrome variants in both copies of the gene.
We found some DNA changes that were not in the ReNU syndrome region of RNU4-2 that were damaging to cells, but we also saw these variants in healthy individuals. In these healthy individuals, however, the variants only disrupted one copy of RNU4-2. When these variants were found on both copies of the gene, individuals had developmental delay and intellectual disability (as we see in ReNU syndrome). As both gene copies need to be disrupted to cause the disorder, it is said to be recessive. Despite there being some similarities in symptoms, this new recessive disorder has some differences to ReNU syndrome, particularly in the changes we see in the brain from MRI images.
DNA changes, or variants, in different regions of RNU4-2 cause dominant ReNU syndrome (only one gene copy is disrupted) and a novel recessive disorder (both gene copies are disrupted).
We are still learning about this new recessive disorder, but at the moment, we believe:
It is less common than ReNU syndrome, potentially impacting about 10% of the number of individuals.
While the frequency of ReNU syndrome is likely the same everywhere around the world (as it is caused by de novo variants which are chance events) the frequency of the recessive disorder will be higher in some population groups, especially those where marriages between cousins are more common.
The recessive disorder and ReNU syndrome are different. Individuals have some different symptoms, and this may mean that there are differences in how the disorders progress over time.
This new knowledge may help development of therapies for ReNU syndrome, as it gives us clues into the importance of different variants and regions across RNU4-2.
The figures in this post were created using BioRender.com
