Background

Voltage-gated calcium channels (VGCCs) are representative transmembrane proteins that convert electrical signals into chemical signals in excitable cells, including neurons and cardiomyocytes. They sense changes in cell membrane voltage and allow calcium ions to flow in, thereby precisely controlling cell survival and signal transduction. In particular, it is well known that mutations in the CACNA1D gene, which encodes the pore-forming α1-subunit of L-type calcium channels, cause various clinical diseases. These include neurodevelopmental disorders such as autism, endocrine disorders, and even life-threatening conditions such as deafness or sinoatrial node dysfunction.

Existing genomic diagnostic techniques have relied on assessing the risk of variants based on previously reported information in patient databases. Attention has been focused on well-defined regions with verified functions, such as domains that directly control channel activity. However, the majority of amino acid residues distributed throughout the channel protein have remained largely unexplored, with their structural values or functional significance unknown. As a result, when previously unreported rare variants appear in clinical settings, it has been nearly impossible for clinicians to determine whether a patient will develop the disease.

Key Findings

A study published in the international journal PNAS provided a breakthrough by focusing on the phenomenon of sequence covariation, which appears in protein evolutionary history. Sequence covariation modeling is a computational biology technique that retrospectively analyzes the relationship between pairs of amino acids that have co-evolved over the long evolutionary history of proteins to maintain their original three-dimensional structure and function. The research team, led by Professor Klaus R. Liedl at the University of Innsbruck, Austria, applied this technique to the Cav1.3 calcium channel α1-subunit, identifying key functional sites with extremely high resolution at the single amino acid level.

The evolutionary model they designed successfully identified a number of novel amino acid sites with significant biophysical properties but previously unknown functions, in addition to already verified regulatory regions. To assess the actual clinical validity of the model's predictions, the researchers verified the model by comparing it to a control group of data on gain-of-function (GoF) variants associated with neurodevelopmental disorders found in actual patient populations. Subsequently, they selected five novel amino acid variant candidates that the model indicated had a high risk of causing disease and conducted electrophysiological experiments to reproduce them at the cellular level.

The results of recording the ion current on the channel surface using a patch-clamp device were surprising. The amino acid variants predicted by the model exhibited a wide range of gating disturbances, from actual channel mutations that completely blocked the channel from opening to mutations that delayed the closing speed of the channel or caused it to open irregularly, resulting in excessive calcium influx. This confirmed that the evolutionary genetics model constructed by the researchers was successful in pre-identifying both loss-of-function (LoF) variants that cause deafness or sinoatrial node paralysis and GoF variants that cause neurodevelopmental diseases such as autism with high confidence.

Significance and Prospects

This computer model provides a unique and precise diagnostic framework that can be immediately applied in actual hospitals or genetic testing companies to determine the potential harm of newly identified rare variants in patients. It has been praised for filling the gap in clinical genetics by deciphering structural clues engraved in evolutionary history through physical calculations. Moreover, even if the types of ions that interact are different, the overall skeleton and the fundamental mechanism of opening and closing are evolutionarily closely conserved in ion channel protein families. This modeling framework is expected to be expanded as a useful prototype to assist in the interpretation of mutations in other transmembrane proteins directly related to human diseases, such as sodium or potassium channels.

However, this evolutionary model has the limitation of being developed by focusing on sequence covariation information of a single target channel protein. In reality, when calcium channels function in the body, they communicate organically with various auxiliary subunits, and the physiological environment inside the cell membrane also changes in a complex manner. In order to ensure the reliability of clinical applications and to further improve targeted drug screening, it is necessary to supplement this with follow-up studies that comprehensively integrate these in vivo interaction factors into the evolutionary prediction algorithm.