β ⁶⁵⁴ -thalassemia is caused by a point mutation in the second intron (IVS-II) of the β-globin gene that activates a cryptic 3′ splice site, leading to incorrect RNA splicing. Our previous study demonstrated that when direct deletion of the β ⁶⁵⁴ mutation sequence or the cryptic 3′ splice site in the IVS-II occurs, correct splicing of β-globin mRNA can be restored. Herein, we conducted an in-depth analysis to explore a more precise gene-editing method for treating β ⁶⁵⁴ -thalassemia. A single-base substitution of the cryptic 3′ acceptor splice site was introduced in the genome of a β ⁶⁵⁴ -thalassemia mouse model using clustered regularly interspaced short palindromic repeats (CRISPR)–CRISPR-associated protein 9(Cas9)-mediated homology-directed repair (HDR). All of the HDR-edited mice allow the detection of correctly spliced β-globin mRNA. Pathological changes were improved compared with the nonedited β ⁶⁵⁴ mice. This resulted in a more than twofold increase in the survival rate beyond the weaning age of the mice carrying the β ⁶⁵⁴ allele. The therapeutic effects of this gene-editing strategy showed that the typical β-thalassemia phenotype can be improved in a dose-dependent manner when the frequency of HDR is over 20%. Our research provides a unique and effective method for correcting the splicing defect by gene editing the reactive splicing acceptor site in a β ⁶⁵⁴ mouse model.

Background/Objectives: X-linked Alport syndrome (XLAS) arises from pathogenic variants in COL4A5. Truncating variants are generally classified as severe, but whether clinically meaningful heterogeneity exists within this group remains unclear. This study aimed to establish a novel Col4a5 knock-in mouse model based on a clinical variant and to determine whether truncating mutation position influences disease severity. Methods: A de novo COL4A5 frameshift variant, c.2440delG, was identified in a patient with severe early-onset XLAS. A Col4a5-G814fs knock-in mouse was generated by CRISPR/Cas9 on the C57BL/6J inbred mouse strain background and compared with the established Col4a5-G5X nonsense model using survival analysis, serial functional measurements, kidney histopathology, transmission electron microscopy, and RNA sequencing. Results: The Col4a5-G814fs knock-in mouse was successfully generated and showed loss of glomerular α5(IV) collagen chain expression. Compared with G5X mice, G814fs mice exhibited shorter survival (median 141 vs. 161.5 days, p = 0.0004), earlier onset of proteinuria, and more severe kidney functional decline. By 16 weeks, G814fs mice also showed more severe glomerular basement membrane abnormalities and more extensive glomerulosclerosis. RNA sequencing revealed a shared inflammatory gene signature in both models, together with selective upregulation of genes related to the PPAR signaling pathway and fatty acid metabolism in G814fs kidneys. Conclusions: This study reports a novel de novo COL4A5 frameshift variant and establishes the first Col4a5-G814fs knock-in mouse model. Direct comparison with the G5X model shows that distinct truncating COL4A5 mutations can be associated with substantially different disease severity, providing a useful platform for future mechanistic and therapeutic studies in XLAS.