DISCUSSION
Previous studies have indicated that CIC interaction with ATXN1 contributes to the pathogenesis of spinocerebellar ataxia type 1 (Rousseaux et al., 2018). Disruption of CIC-ATXN1 complex results in hyperactivity, impaired learning and memory, and abnormal maturation and maintenance of upper-layer cortical neurons in Emx1-cre mice (Cao et al., 2021). Forebrain–specific deletion of CIC in Foxg1-cre mice caused abnormal increases in oligodendrocyte progenitor and immature oligodendrocytes populations (Yang et al., 2017). Brain-specific deletion of CIC compromised neuroblasts transition to immature neurons in mouse hippocampus and compromises normal neuronal differentiation (Hwang et al., 2020). While CIC deletion clearly impacts numerous aspects of neurodevelopment, ours is the first evidence indicating a potential relationship between CIC and neural tube development.
In the present study, we performed and support the first genetic association analysis between CIC variants and an increased risk for NTDs. We also explored the possible underlying mechanisms by which CIC could contribute to NTDs in humans. We initially focused on the folate receptors, as they are membrane proteins that mediate cellular uptake of folates. FOLR1 is important for neural tube closure during early embryogenesis, as inactivation of this gene leads to embryoic death and NTDs in mice which can be rescued by folate supplementation (Piedrahita et al., 1999). In humans, homozygous biallelic LoF FOLR1 mutations lead to extremely low 5-methyltetrohydrofolate level in the cerebrospinal fluid (CFD) but have not been observed to result in NTDs (Steinfeld et al., 2009). However, it has been reported that increased level of maternal serum FOLR1 autoantibody during pregnancy is associated with the occurrence of fetal NTDs (Rothenberg et al., 2004). Although evidence is lacking to show that isolated FOLR1 gene variants are associated with human NTDs, Saitsu (2017) found that altered spatial and temporal Folr1 expression patterns in mice are associated with anterior neural tube closure, and this expression pattern is conserved between human and mice engineered to express a lacZ reporter transgene. Additionally, there have been studies of human folate transport genes that described 12 novel variants in FOLR1, FOLR2, and FOLR3 (Findley et al., 2017) found in NTD cases. This included four large insertion deletion variants in FOLR3 as well as a single stop gain variant. While far from being conclusive evidence, it is suggestive that FOLR1 abnormalities might be involved in a subset of human NTDs. According to our previous study of CIC and CFD, CIC could regulate FOLR1 expression through binding to its promoter region. In this study, we confirmed that CIC variants can decrease FOLR1 expression levels in human cell lines. While we previously found that CIC binding motifs lie within the human FOLR1 promoter region, they do not exist in the mouse Folr1promoter region.
Convergent extension (CE) is a crucial process during neural tube closure by which the neural plate undergoes narrowing along its mediolateral axis and extends along anteroposterior axis (Tada & Heisenberg, 2012). The progression of convergent extension is driven by planar polarized cell intercalation, which in turn is reported to be driven by subcellular processes including extension of mediolaterally directed cellular protrusions and shrinkage of mediolaterally oriented cell-cell junctions (Butler & Wallingford, 2018; Blankenship, Backovic, Sanny, Weitz & Zallen, 2006). The PCP signaling pathway is the most well-characterized regulator of cell intercalation (Butler & Wallingford, 2017). Several genes in the PCP pathway have been associated with NTDs (Humphries, Narang & Mlodzik, 2020). Over 300 genes are known to be causative of NTDs in mice, and many of them participate in the PCP pathway (Wang etal., 2018). Core PCP genes are highly conserved from invertebrates to mammals, among which Van Gogh (Vang), Vangl1/2 in mammals, is the critical regulator for normal extension conversion. Other PCP genes include FZD3/FZD6, CELSR1/CELSR2/CELSR3, DVL1/DVL2/DVL3, PRICKLE1/PRICKLE2/PRICKLE3/PRICKLE4 and ANKRD6. Vangl2 was the first genetically mapped PCP gene through studying Loop-tail mutant mice. Homozygous Vangl2 mutants cause craniorachischisis, the most severe type of NTD (Kibar et al., 2001), while Vangl1 and Vangl2 compound heterozygous mice also exhibit a craniorachischisis phenotype (Torban et al., 2008). Besides Vangl2, inactivation of Dvl1 and Dvl2, Celsr1, and Fzd3 or both Fzd3 and Fzd6 also lead to severe NTD phenotypes, primarily craniorachischisis and exencephaly in mice (Kibar et al., 2001; Wang et al., 2006; Ybot-Gonzalez et al., 2007; Curtin et al., 2003). In addition to these core PCP genes, some noncore PCP genes also exhibit severe NTD in mice when gene targeted, such as protein tyrosine kinase 7 (PTK7), scribbled PCP protein, the gene responsible for the circle tail mouse phenotype, Scrib, and dishevelled binding antagonist of beta-catenin 1 (Dact-1) (Mohd-Zin, Marwan, Abou Chaar, Ahmad-Annuar & Abdul-Aziz, 2017). In humans, PCP genes have been examined in case–control association studies or directly sequenced in mutation screens (Juriloff & Harris, 2012). Vangl2 variants in human NTD samples were reported by Lei et al.(2010) and Kibar et al.(2011). In the present study, we found that all eight CIC variants lead to decreased Vangl2 protein levels in human cell lines, and the CIC LoF variant also limited Vangl2 protein levels in a murine cell line. We suspect this contributes to the observed NTD phenotype.
Acting downstream of Vangl2, RhoA plays a crucial role in regulating cell shape and movement (Oishi, Makita, Sato & Iiri, 2012). In this study, protein levels of Vangl2 and RhoA showed similar trends for different CIC variants among groups in both human and mouse cell lines. Our in vitro assays showed that CIC variants diminished both FOLR1 expression level and PCP signals, both of which could increase the likelihood of human NTDs. However, the relationship between inhibiting Folr1 and associated PCP pathway gene expression and the underlying mechanisms by which CIC affects Vangl2, remain unclear. Balashova, Visina and Borodinsky (2017) found that FOLR1 is enriched in the apical surface of the neural plate in Xenopus laevis , colocalized with C-cadherin and β-catenin, and that FOLR1 is necessary for neural plate cell apical constriction during Xenopus neural tube formation. The seemingly contradictory evidence emphasizes the importance of continued experimentation on this system.
In this study, we identified a novel gene contributing to human NTDs. We demonstrated the possible underlying mechanisms. Future studies are needed to further elucidate these cellular and molecular mechanisms, which will help us to better understand the genetic etiology of NTDs and develop effective strategies to prevent these severe birth defects.