The biochemical hallmark of SSADH-D is an increase in the concentration of GABA and GHB in body fluids, such as blood, urine, and cerebrospinal fluid. These two substances mainly affect the nervous system. Thus, SSADH-D is often characterized by nervous system lesions [9, 10]. The SSADH-D is primarily studied using simplified model symptoms such as HEK293 cells overexpressing genes of interest, but such overexpression can result in protein aggregation or pathway saturation that may not represent actual underlying disease phenotypes[3]. A short life span and low expression levels of SSADH further limit the application of dermal fibroblasts in a patient. Thus, more in vitro approaches with cellular models closer to humans should be developed. For instance, iPSC models of various diseases have emerged [6, 11–14]. In the present study, a successful strategy was proposed to generate two SSADH-D iPSC lines via CRISPR/Cas9 genome engineering combined with flow cytometry and a clonal loop. The advantage of this model was that iPSCs could be differentiated into all cell types of the body, especially those representing the target tissue cells, such as NSCs. Through the induction of iPSCs, the NSCs of the wild type and mutants showed the characteristic changes in SSADH-D. Notably, comparing the expression of ALDH5A1 and the amount of GABA between iPSCs and NSCs is attractive. (Fig. 2F and 2G) When iPSCs were induced into NSCs, the ALDH5A1 expression increased in the wild type and mutant types, but the degree of increase differed, and the increase in mutant cell lines was more apparent (Fig. 2C). The GABA accumulation multiple of the mutants of the wild type NSC is more serious than that of the mutant than that of the wild-type in iPSC lines (Fig. 2F and 2G). However, in comparison with NSCs, iPSC likely tolerated higher GABA concentrations. This observation emphasizes that more SSADH was needed to help metabolize GABA in NSCs. GABA accumulation could not be reduced even when the gene expression increased because of ALDH5A1 mutation.
Furthermore, the generation of L243_S245del and A244_Q246del allowed us to conclude about the in vitro function of the splicing region between exons 4 and 5 in ALDH5A1. In Fig. 2B, the expression of the A244_Q246del mutant was only 30%-40% of that of the wild-type before and after iPSCs induced to NSCs. However, the decrease in the ALDH5A1 expression in the L243_S245del mutant was not distinguishing (about 75%) in iPSC lines compared with that in the wild-type. Even after induction, the ALDH5A1 expression of the L243_S245del mutant was higher than that of the wild-type (Fig. 2C). These results indicated that only the A244_Q246del mutation affected the gene expression, although only one amino acid difference existed between these two mutations in our study. This hypothesis was tested by sequencing the cDNA of the two different mutants in the target gene. As expected, the A244_Q246del mutant (exon 5) severely affected the RNA splicing, resulting in 545bp deletion encompassing exons 4–7 of ALDH5A1 (Fig. 3B). Another mutation, i.e., L243_S245del, only affected the deletion of 12 bases (CTTGCAAGCCAG) in exon 5 (Fig. 3B). Thus, this observation was the fundamental reason for the different changes in the ALDH5A1 expression before and after iPSC induction between the two mutant cell lines. Our literature search revealed that Akaboshi et al. [15] concluded that the mutations in a short stretch between aa 223 and 268 (encoded by exons 4 and 5) are not entirely random, and a vital region of the gene may be present. This finding was helpful for further studying the mechanism of the vital region of the ALDH5A1 gene. Therefore, the construction of an iPSC disease model could be used as a good disease model for research on SSADH-D mechanisms.
A correct diagnosis is the first step in treating rare diseases, and it is the basis of disease mechanism research. For rare diseases, the availability of validated samples from patients with a specific disease is usually low, limiting the possibilities of using these samples. iPSC disease models can be differentiated into various target cells because of the pluripotency of iPSCs. In our study, the iPSC disease model could be differentiated into NSCs. This model could be utilized to detect changes in the ALDH5A1 expression and the GABA accumulation in the two cell lines, which should be a good cell model for drug screening because the therapy-induced reduction of GABA in the periphery may be a vital issue for the development of future therapies for SSADH-D [16].
In conclusion, the CRISPR-based genome editing of iPSCs shows potential for future studies on the pathogenicity of diseases. Our research concluded that the iPSCs could be helpful for SSADH-D disease modelling.