Xiaozhi Gao
Postdoctoral Scholar, Cardiovascular Institute
Contact
https://orcid.org/0000-0001-5843-2832All Publications
-
Clinical Utility of Protein Language Models in Resolution of Variants of Uncertain Significance inKCNQ1, KCNH2, andSCN5ACompared With Patch-Clamp Functional Characterization.
Circulation. Genomic and precision medicine
2024: e004584
Abstract
BACKGROUND: Genetic testing for cardiac channelopathies is the standard of care. However, many rare genetic variants remain classified as variants of uncertain significance (VUS) due to lack of epidemiological and functional data. Whether deep protein language models may aid in VUS resolution remains unknown. Here, we set out to compare how 2 deep protein language models perform at VUS resolution in the 3 most common long-QT syndrome-causative genes compared with the gold-standard patch clamp.METHODS: A total of 72 rare nonsynonymous VUS (9 KCNQ1, 19 KCNH2, and 50 SCN5A) were engineered by site-directed mutagenesis and expressed in either HEK293 cells or TSA201 cells. Whole-cell patch-clamp technique was used to functionally characterize these variants. The protein language models, ESM1b and AlphaMissense, were used to predict the variant effect of missense variants and compared with patch clamp.RESULTS: Considering variants in all 3 genes, the ESM1b model had a receiver operator curve-area under the curve of 0.75 (P=0.0003). It had a sensitivity of 88% and a specificity of 50%. AlphaMissense performed well compared with patch-clamp with an receiver operator curve-area under the curve of 0.85 (P<0.0001), sensitivity of 80%, and specificity of 76%.CONCLUSIONS: Deep protein language models aid in VUS resolution with high sensitivity but lower specificity. Thus, these tools cannot fully replace functional characterization but can aid in reducing the number of variants that may require functional analysis.
View details for DOI 10.1161/CIRCGEN.124.004584
View details for PubMedID 39119706
-
SGK1 inhibition attenuated the action potential duration in patient- and genotype-specific re-engineered heart cells with congenital long QT syndrome.
Heart rhythm O2
2023; 4 (4): 268-274
Abstract
Background: Long QT syndrome (LQTS) stems from pathogenic variants in KCNQ1 (LQT1), KCNH2 (LQT2), or SCN5A (LQT3) and is characterized by action potential duration (APD) prolongation. Inhibition of serum and glucocorticoid regulated kinase-1 (SGK1) is proposed as a novel therapeutic for LQTS.Objective: The study sought to test the efficacy of novel, selective SGK1 inhibitors in induced pluripotent stem cell-derived cardiomyocyte (iPSC-CM) models of LQTS.Methods: The mexiletine (MEX)-sensitive SCN5A-P1332L iPSC-CMs were tested initially compared with a CRISPR (clustered regularly interspaced short palindromic repeats)/Cas9 SCN5A-P1332L variant-corrected isogenic control (IC). The SGK1-I1 therapeutic efficacy, compared with MEX, was tested for APD at 90% repolarization (APD90) shortening in SCN5A-P1332L, SCN5A-R1623Q, KCNH2-G604S, and KCNQ1-V254M iPSC-CMs using FluoVolt.Results: The APD90 was prolonged in SCN5A-P1332L iPSC-CMs compared with its IC (646 ± 7 ms vs 482 ± 23 ms; P < .0001). MEX shortened the APD90 to 560 ± 7 ms (52% attenuation, P < .0001). SGK1-I1 shortened the APD90 to 518 ± 5 ms (78% attenuation, P < .0001) but did not shorten the APD90 in the IC. SGK1-I1 shortened the APD90 of the SCN5A-R1623Q iPSC-CMs (753 ± 8 ms to 475 ± 19 ms compared with 558 ± 19 ms with MEX), the KCNH2-G604S iPSC-CMs (666 ± 10 ms to 574 ± 18 ms vs 538 ± 15 ms after MEX), and the KCNQ1-V254M iPSC-CMs (544 ± 10 ms to 475 ± 11ms; P = .0004).Conclusions: Therapeutically inhibiting SGK1 effectively shortens the APD in human iPSC-CM models of the 3 major LQTS genotypes. These preclinical data support development of SGK1 inhibitors as novel, first-in-class therapy for patients with congenital LQTS.
View details for DOI 10.1016/j.hroo.2023.02.003
View details for PubMedID 37124559