MLe-KCNQ2: An Artificial Intelligence Model for the Prognosis of Missense KCNQ2 Gene Variants.

Despite the increasing availability of genomic data and enhanced data analysis procedures, predicting the severity of associated diseases remains elusive in the absence of clinical descriptors. To address this challenge, we have focused on the KV7.2 voltage-gated potassium channel gene (KCNQ2), known for its link to developmental delays and various epilepsies, including self-limited benign familial neonatal epilepsy and epileptic encephalopathy. Genome-wide tools often exhibit a tendency to overestimate deleterious mutations, frequently overlooking tolerated variants, and lack the capacity to discriminate variant severity. This study introduces a novel approach by evaluating multiple machine learning (ML) protocols and descriptors. The combination of genomic information with a novel Variant Frequency Index (VFI) builds a robust foundation for constructing reliable gene-specific ML models. The ensemble model, MLe-KCNQ2, formed through logistic regression, support vector machine, random forest and gradient boosting algorithms, achieves specificity and sensitivity values surpassing 0.95 (AUC-ROC > 0.98). The ensemble MLe-KCNQ2 model also categorizes pathogenic mutations as benign or severe, with an area under the receiver operating characteristic curve (AUC-ROC) above 0.67. This study not only presents a transferable methodology for accurately classifying KCNQ2 missense variants, but also provides valuable insights for clinical counseling and aids in the determination of variant severity. The research context emphasizes the necessity of precise variant classification, especially for genes like KCNQ2, contributing to the broader understanding of gene-specific challenges in the field of genomic research. The MLe-KCNQ2 model stands as a promising tool for enhancing clinical decision making and prognosis in the realm of KCNQ2-related pathologies.

Molecular dynamics simulations of the calmodulin-induced α-helix in the SK2 calcium-gated potassium ion channel.

The family of small-conductance Ca2+-activated potassium ion channels (SK channels) is composed of four members (SK1, SK2, SK3, and SK4) involved in neuron-firing regulation. The gating of these channels depends on the intracellular Ca2+ concentration, and their sensitivity to this ion is provided by calmodulin (CaM). This protein binds to a specific region in SK channels known as the calmodulin-binding domain (CaMBD), an event which is essential for their gating. While CaMBDs are typically disordered in the absence of CaM, the SK2 channel subtype displays a small prefolded α-helical region in its CaMBD even if CaM is not present. This small helix is known to turn into a full α-helix upon CaM binding, although the molecular-level details for this conversion are not fully understood yet. In this work, we offer new insights on this physiologically relevant process by means of enhanced sampling, atomistic Hamiltonian replica exchange molecular dynamics simulations, providing a more detailed understanding of CaM binding to this target. Our results show that CaM is necessary for inducing a full α-helix along the SK2 CaMBD through hydrophobic interactions with V426 and L427. However, it is also necessary that W431 does not compete for these interactions; the role of the small prefolded α-helix in the SK2 CaMBD would be to stabilize W431 so that this is the case. In conclusion, our findings provide further insight into a key interaction between CaM and SK channels that is important for channel sensitivity to Ca2+.

An epilepsy-causing mutation leads to co-translational misfolding of the Kv7.2 channel.

BACKGROUND: The amino acid sequence of proteins generally carries all the necessary information for acquisition of native conformations, but the vectorial nature of translation can additionally determine the folding outcome. Such consideration is particularly relevant in human diseases associated to inherited mutations leading to structural instability, aggregation, and degradation. Mutations in the KCNQ2 gene associated with human epilepsy have been suggested to cause misfolding of the encoded Kv7.2 channel. Although the effect on folding of mutations in some domains has been studied, little is known of the way pathogenic variants located in the calcium responsive domain (CRD) affect folding. Here, we explore how a Kv7.2 mutation (W344R) located in helix A of the CRD and associated with hereditary epilepsy interferes with channel function. RESULTS: We report that the epilepsy W344R mutation within the IQ motif of CRD decreases channel function, but contrary to other mutations at this site, it does not impair the interaction with Calmodulin (CaM) in vitro, as monitored by multiple in vitro binding assays. We find negligible impact of the mutation on the structure of the complex by molecular dynamic computations. In silico studies revealed two orientations of the side chain, which are differentially populated by WT and W344R variants. Binding to CaM is impaired when the mutated protein is produced in cellulo but not in vitro, suggesting that this mutation impedes proper folding during translation within the cell by forcing the nascent chain to follow a folding route that leads to a non-native configuration, and thereby generating non-functional ion channels that fail to traffic to proper neuronal compartments. CONCLUSIONS: Our data suggest that the key pathogenic mechanism of Kv7.2 W344R mutation involves the failure to adopt a configuration that can be recognized by CaM in vivo but not in vitro.