An RNA world. Last week, a simple bar graph caught my attention and prompted me to write my first blog post in over a year. I came across a figure comparing the frequency of de novo variants in RNU4-2 to other genes causing neurodevelopmental disorders. The data suggested that the recently identified ReNU syndrome may be one of the most common genetic neurodevelopmental disorders with a high frequency of seizures. This led me to take a closer look at a group of conditions that have emerged over the last few years: disorders of spliceosome function, providing some interesting insight into the dynamics of gene discovery in the post-genomic era.
Figure 1. Recurrent variants in RNU4-2 cause neurodevelopmental disorders. (A) De novo variants in RNU4-2 tend to cluster in a critical region in the RNU4-2 gene, including the recurrent n.64_65insT variant. (B) U4 snRNA binds U6 snRNA via base-pairing across two stems, forming a so-called stable duplex. The binding with U6 snRNA is important to prevent it from assuming its catalytic role within the active site of the spliceosome. (C) Traditionally, we think of RNA as the precursor for proteins. However, there is a much wider range of activities that RNA molecules assume in cellular function, including the role of small nuclear RNAs (snRNA) in splicing. (D) De novo variants in RNU4-2 are more common than other neurodevelopmental genes, even exceeding the frequency of ANKRD11, one of the most common genes reported in Kaplanis et al., 2020. [Figures modified from Chen et al., 2024 (A,B), and Greene et al., 2024 (D) under a Creative Commons Attribution 4.0 International License. Figure C was created using BioRender.]
The spliceosome. Over the last few years, an increasing number of neurodevelopmental disorders were found to be caused by variants in genes involved in the spliceosome. I was first introduced to this concept through our 2024 publication by Li and collaborators where we identified disease-causing variants in U2AF2, PRPF19, and RBFOX1. But what is this mysterious spliceosome? In brief, it is the cellular machinery responsible for removing the introns from pre-mRNA and splicing the exons together. Over 90% of human genes have introns, with an average of 8–10 introns per gene. Accordingly, the spliceosome is one of the most essential structures of the human cell. The essential components of the spliceosome are the so-called “snurps”—small ribonucleoproteins containing a small nuclear RNA (snRNA) and a set of associated proteins. The snRNAs used in the spliceosome are U1, U2, U4, U5, and U6, and the genes encoding them begin with the prefix RNU. Accordingly, RNU4-2 is one of several genes comprising the U4 snRNA.
A PRRT2 moment. In 2024, two publications identified de novo variants in RNU4-2 in individuals with neurodevelopmental disorders. Taken together, these publications represented the single largest number of individuals identified with a new neurodevelopmental gene, including 47 individuals identified by Greene and collaborators and 115 individuals identified by Chen and collaborators. In addition, Nava and collaboratorsidentified an additional 145 individuals with de novo variants in RNU4-2. With more than 20 publications since its discovery in 2024, the momentum behind RNU4-2 is reminiscent of the 2011 discovery of PRRT2 as the long-sought gene for self-limiting infantile seizures, which triggered more than 100 publications in the two years after its initial discovery. In summary, RNU4-2 is one of the more common monogenic causes of neurodevelopmental disorders. Clearly, this gene has been entirely overlooked. But why?
Exome failure. The recency of RNU4-2’s identification is attributable to two main factors. First, because RNU4-2 does not code for a protein, it is not targeted in exome sequencing, though some individuals receiving this test have off-target reads mapping to RNU4-2. Accordingly, RNU4-2 is the first true neurodevelopmental “genome-only gene”. However, a lack of coverage on exomes is not the only reason RNU4-2 had been underrecognized; limitations to identifying this disorder remain even with genome sequencing. Current methods of variant classification rely heavily on the predicted impact of a variant on protein function. Because RNU4-2 does not encode a protein, it is difficult to assess whether variants alter gene function using existing bioinformatic tools. Our current pipelines are protein-focused and do not include predicted impact on RNA structure or binding. Put differently, it is difficult to assess the impact of a variant if there is no protein-truncating or missense variant that we can put into context.
ReNU Syndrome. Our understanding of RNU4-2-related neurodevelopmental disorders has progressed rapidly. The condition, initially referred to by the somewhat bulky term neurodevelopmental disorder with hypotonia, brain anomalies, distinctive facies, and absent language (NEDHAFA) is now called ReNU syndrome and is supported by ReNU Syndrome United advocacy organization. Even though both initial reports used structured, HPO-based phenotyping to delineate the features of ReNU syndrome, the focus thus far has largely been on gene discovery rather than phenotype delineation. Encouragingly, a growing number of publications have extended the clinical understanding of ReNU Syndrome, focusing on distinct syndromic and facial features, such short stature, hooded upper eyelids and a tented philtrum.
Neurodevelopment. The neurodevelopmental and neurological features of ReNU Syndrome are not yet well delineated and are often reported alongside non-neurological features, making it difficult to distinguish them. This is particularly due to some internal “bugs” of the Human Phenotype Ontology (HPO), where features such as “drooling” and “excessive salivation” are listed as phenotypic features of the face. In reality, these features are a consequence of orofacial hypotonia, which is a frequent feature of the global hypotonia seen across neurodevelopmental disorders. The emerging picture suggests that most individuals with ReNU Syndrome have intellectual disability ranging from moderate to severe as well as a relatively high frequency of neuro-ophthalmological issues features such as nystagmus. As initially suggested in the NEDHAFA acronym, most individuals with ReNU Syndrome do not use spoken language, though expressive language only covers a subset of communicative abilities. Fully delineating the range of neurodevelopmental and neurological features in natural history studies will be important in moving ReNU towards clinical trial readiness.
Epilepsy. I was surprised that the frequency of epilepsy in ReNU Syndrome seems to be much higher than in other neurodevelopmental genes. Greene and collaborators find a frequency of 50%, while Chen and collaborators report a frequency of almost 80%. Seizure types and trajectories are highly variable: infantile spasms, focal seizures, and bilateral tonic-clonic seizures have been reported. Seizure onset varies from infancy to adolescence. The full burden of seizure phenotypes in ReNU Syndrome is currently unknown, including the frequency of refractory epilepsy and developmental and epileptic encephalopathy. However, with increased recognition of RNU4-2-related neurodevelopmental disorders, it is reasonable to expect that its seizure landscape will soon be better delineated.
What you need to know. RNU4-2-related neurodevelopmental disorder (ReNU Syndrome) is an emerging neurodevelopmental disorder typically missed on traditional exome sequencing. ReNU Syndrome is relatively common amongst monogenic neurodevelopmental disorders and appears to have a relatively high frequency of seizures, comparable to SCN2A– or STXBP1-related disorders. The genetic basis of ReNU Syndrome is unusual, as the RNU4-2 gene codes for a small nuclear RNA rather than a protein. Given its frequency, it is reasonable to expect significant progress in better delineating ReNU syndrome in the near future.
