Approximately 1-2% of the global population is affected by intellectual disability (ID), representing a serious medical, social and economic problem. The clinical symptoms and aetiology of ID are highly heterogeneous, making diagnosis and treatment of the disease difficult. X-linked ID caused by UPF3B mutations is well-suited for more detailed investigation due to its clearly defined genetic cause. In this project, we will study the function of UPF3B to better understand the development of ID.
Normal brain development depends on the precise operation of molecular programs that lead to proliferation, migration, and maturation of neuronal and glial cells. Alterations in these programs and processes can impair the development of the brain and lead to intellectual disability. A varying proportion of intellectual disability cases (15% to 50%) is caused by genetic factors, such as mutations in certain genes.
Several mutations in the gene encoding UPF3B have been identified as the cause of syndromic and non-syndromic forms of X-linked intellectual disability in more than 10 families. The identified mutations include missense, frameshift and nonsense mutations presumably leading to a loss of UPF3B function. Some UPF3B patients show additional disease states, including schizophrenia, autism and attention deficit hyperactivity disorder. Thus, the UPF3B protein is crucial for normal brain development and functions by a yet unknown mechanism.
Here, we propose to dissect the mechanism of X-linked intellectual disability caused by UPF3B mutations. To gain insights into the molecular function of UPF3B, we will mutate UPF3B in cultured cells by CRISPR/Cas9 and analyse the molecular effects using different high-throughput analyses.
In the current funding period we wanted to gain insights into the molecular functions of UPF3B and why mutations in the encoding gene result in neurological phenotypes. We first set out to characterize the functional impact on human cells when UPF3B is downregulated or missing. This task was challenging due to the fact that mammalian cells express a paralog of UPF3B, named UPF3A, which shows significant sequence similarity to UPF3B. In patients with mutated UPF3B, it was shown that the protein levels of UPF3A correlate negatively with the severity of the patient’s phenotype. This indicated a certain “back-up ability” of UPF3A, suggesting at least partial functional redundancy between both paralogs.
However, UPF3A was also reported to be an antagonist of UPF3B in the nonsense-mediated mRNA decay (NMD) pathway. Therefore, it remained to be determined to which extent UPF3A and UPF3B act redundantly or inhibit each other in human cells. To address these contradicting theories, we used different approaches (e.g. CRISPR-Cas9) to manipulate UPF3A and UPF3B expression levels and analyzed the interplay between both paralogs. In contrast to the reported NMD-inhibiting role, we found that UPF3A can actually functionally compensate the absence of its paralog UPF3B. Neither the single knockout of UPF3A nor UPF3B substantially altered the expression of NMD substrates, only co-depletion of the two paralogs resulted in a marked NMD inhibition.
These results support redundant functions of UPF3A and UPF3B.
Further experiments regarding the different domains of UPF3B revealed that UPF3 is not only fault-tolerant due to the redundant paralogs, but also because mutation of a single domain does not inactivate the protein (Figure 1). We published the results earlier this year in the EMBO Journal (Wallmeroth et al., 2022)
In addition to the function of UPF3 in NMD, we aimed to investigate the role of UPF3B in neuronal cells. We proposed to use SH-SY5Y cells, but they turned out to be non-optimal for the planned experimental approaches. Therefore, we concentrated our efforts directly on the work with human induced pluripotent stem cells (hiPSCs). Despite supply shortage and other obstacles, we set up a stem-cell suitable cell culture and were able to successfully differentiate the hiPSCs into cerebral organoids (Figure 2). First experiments with a new pharmacological NMD inhibitor showed encouraging results, indicating that this system has great potential for future research.
In the following weeks and months, we will analyze these wild-type cerebral organoids for general properties and start the generation of UPF3B-KO hiPSCs using the CRISPaint system. The deeper analyses like single cell RNA-seq to e.g. identify cell types that are most severely affected have to be carried out in a later project. Furthermore, we decided to broaden the investigation to global NMD in human cerebral organoid formation. We will modulate NMD activity in either direction (inhibition or hyperactivity) at different timepoints during organoid maturation (Figure 3). The results will provide important information about the relevance of NMD in brain development.