Paul Shamrat Kumar, Panday Shailesh Kumar, Boccuto Luigi, Alexov Emil
Department of Physics and Astronomy, College of Science, Clemson University, Clemson, SC 29634, USA.
Medical Biophysics Graduate Program, College of Science, Clemson University, Clemson, SC 29634, USA.
Int J Mol Sci. 2025 Aug 29;26(17):8399. doi: 10.3390/ijms26178399.
Cyclin-dependent kinase-like 5 (CDKL5) deficiency disorder, which is a developmental and epileptic encephalopathy occurring in 1 in every 40,000 to 60,000 live births, was the subject of this computational investigation. This study provided a comprehensive list of missense variants (156) seen in the human population within the CDKL5 protein. Furthermore, the list of CDKL5 binding partners was updated to include four new entries. Computational modeling resulted in 3D structure models of twenty-four CDKL5-target protein complexes. The CDKL5 stability changes upon the above-mentioned missense mutations that were modeled, and it was shown that the corresponding folding free energy changes (ΔΔG) caused by pathogenic variants are much larger than the ΔΔG caused by benign variants. The same observation was made for the binding free energy change (ΔΔG). This resulted in a protocol that allowed for the reclassification of missense variants with unknown or conflicting significance into pathogenic or benign. It was demonstrated that such reclassification is more reliable than using leading tools for pathogenicity predictions, since the latter failed to correctly predict known pathogenic/benign variants. Furthermore, the study demonstrated that pathogenicity is linked with the disturbance of thermodynamics quantities such as ΔΔG and ΔΔG, paving the way for development of therapeutic solutions.
细胞周期蛋白依赖性激酶样5（CDKL5）缺陷障碍是一种发生于每40000至60000例活产婴儿中的1例的发育性和癫痫性脑病，是本次计算研究的主题。本研究提供了在人类群体中CDKL5蛋白内所见错义变体（156个）的综合列表。此外，CDKL5结合伴侣列表得到更新，增加了四个新条目。计算建模产生了24种CDKL5 - 靶蛋白复合物的三维结构模型。对上述建模的错义突变进行建模后，CDKL5稳定性发生变化，结果表明，致病性变体引起的相应折叠自由能变化（ΔΔG）远大于良性变体引起的ΔΔG。结合自由能变化（ΔΔG）也有同样的观察结果。这产生了一种方案，可将意义未知或存在冲突的错义变体重新分类为致病性或良性。结果表明，这种重新分类比使用领先的致病性预测工具更可靠，因为后者未能正确预测已知的致病性/良性变体。此外，该研究表明，致病性与诸如ΔΔG和ΔΔG等热力学量的扰动有关，为治疗方案的开发铺平了道路。
