Tooth agenesis (OMIM #106600) is a common congenital craniofacial developmental anomaly, with the prevalence (excluding third molars) ranges from 2 % to 10 % across ethnic groups (Nadolinski et al., 2023). Individuals affected by tooth agenesis experience significant functional, psychological, and aesthetic challenges (Dujic et al., 2025). Tooth agenesis can be classified as syndromic, when associated with systemic disorders such as hypotrichosis, hypohidrosis, nail dysplasia, thin skin, cleft lip and palate, or non-syndromic tooth agenesis (NSTA) (Wu et al., 2024; Yu et al., 2022). It may follow autosomal dominant, autosomal recessive, or X-linked inheritance patterns, and in some cases, may lack a clear segregation pattern (Uchiyama et al., 2025; Yang et al., 2020). Therefore, elucidating the genetic etiology of tooth agenesis is essential for understanding the pathogenic mechanisms underlying this inherited disorder.

Over 200 genes have been implicated in tooth agenesis, with mutations affecting both syndromic and non-syndromic forms (Wu et al., 2024; Yu & Klein, 2020). Syndromic tooth agenesis is commonly caused by gene mutations in evolutionarily conserved signaling pathways such as WNT, EDA, SHH, FGF, and TGF-β/BMP, as well as mutations in key regulatory molecules including PAX9, PIXT2, IRF6, members of the p53 family, and subunits of RNA polymerase III (Lan et al., 2023). In the context of non-syndromic tooth agenesis (NSTA), critical roles have been attributed to components of the Wnt/β-catenin, EDA/EDAR/NF-κB, SHH, BMP, and TGF signaling pathways. The most commonly associated genes include MSX1, PAX9, WNT10A, and EDA (Khan et al., 2022, Letra, 2022). Of these, WNT10A is the most frequently mutated gene in NSTA, with over 60 pathogenic variants identified, accounting for up to 56 % of cases (Albu et al., 2021; Biedziak et al., 2022). For EDA, though it is primarily linked to syndromic ectodermal dysplasia, an increasing number of studies have reported mutations causing NSTA. To date, approximately 40 mutations in the EDA gene have been identified in patients with NSTA (Table 1). Among these, the c.1013 C>T (p.Thr338Met) mutation appears to be the most prevalent (Wu et al., 2024). These mutations are believed to disrupt the function of EDA protein, thereby contributing to dental anomalies. In addition to these rare pathogenic variants, recent studies have highlighted the role of common variants of the EDA gene in the etiology of NSTA. For instance, the T allele of rs12853659 and the G allele of rs2428151 have both been associated with an increased risk of hypodontia, suggesting that both rare and common EDA variants may contribute to the phenotypic spectrum of non-syndromic hypodontia (Al-Ani et al., 2021).

The ectodysplasin-A (EDA) gene, located on the long arm of the X-chromosome (Xq12–13.1), is inherited in either an X-linked recessive or X-linked dominant manner (Lee et al., 2024; Tarpey et al., 2007). Due to alternative splicing, the EDA genes encode eight distinct isoforms, among which only EDA-A1 and EDA-A2 possess receptor-binding domains (Hymowitz et al., 2003). As a member of the tumor necrosis factor (TNF)-related ligand family, EDA-A1 functions as a soluble type II transmembrane protein (Xing et al., 2024). It comprises four distinct functional domains: a short intracellular N-terminal domain, a stalk region with a furin cleavage site, a collagen-like domain with 19 G-X-Y repeats, and a C-terminal TNF homology domain (Lee et al., 2024; Ranjan & Das, 2022). EDA-A1 binds to the ectodysplasin A receptor (EDAR) through its TNF domain, initiating EDA/EDAR/NF-κB signal transduction, which plays a crucial role in tooth morphogenesis and development (Cai et al., 2021). EDA-A1 (a 391-amino-acid protein) and EDA-A2 (a 389-amino-acid protein) differ by a two-amino-acid motif within TNF domain, specifically at positions p.Glu308 and p.Val309, which are present in EDA-A1 but absent in EDA-A2. This difference alters the electrostatic potential and surface conformation of the protein, culminating in a change in receptor specificity (Yu et al., 2023). Unlike EDA-A1, EDA-A2 binds exclusively to another receptor, X-linked ectodysplasin receptor (XEDAR), which does not contribute to the etiology of tooth agenesis. Although XEDAR can activate the NF-κB pathway by recruiting TRAF3 and TRAF6, this signaling pathway cannot compensate for the disruption of the EDA/EDAR/NF-κB signaling cascade (Gao et al., 2023). The EDA gene has been definitively identified as an etiological factor in both syndromic tooth agenesis and NSTA (Liu et al., 2023; Ouyang et al., 2024). However, EDA mutations are primarily linked to syndromic tooth agenesis, manifesting as X-linked hypohidrotic ectodermal dysplasia (XLHED), whereas their association with NSTA is exceptionally rare. Therefore, studying the impact of EDA mutations on NSTA is crucial for advancing our understanding of the mechanisms underlying this disease (Andreoni et al., 2021).

In this study, through whole-exome sequencing, we revealed a novel missense mutation, c.827 G>T (p.Arg276Leu), and a previously reported mutation, c.1069 C>T (p.Arg357Trp), in EDA in two unrelated Chinese families with NSTA. Notably, the latter mutation exhibited a distinct phenotypic spectrum compared to prior reports. Three-dimensional (3D) modeling and bioinformatics analyses predicted that these two mutations may be pathogenic, likely due to alterations in hydrogen bonds and electrostatic potential.