A three-family clinical genetic study identified distinct pathogenic or likely pathogenic variants in PAX6 among 11 members of three Chinese families with congenital aniridia. The probands carried either two large chromosome 11 deletions or a single-base deletion that shifts the reading frame of the eye-development gene. Array comparative genomic hybridization, whole-exome sequencing, Sanger sequencing, quantitative PCR and chromosome karyotyping were used in combination, revealing that standard chromosome analysis alone missed both large deletions. Two deletions also removed WT1, a tumor-suppressor gene associated with Wilms tumor risk, linking the eye findings to the genetic criteria for WAGR syndrome.
The three probands all had absent irises and nystagmus, but additional findings differed among families. The female proband in Family 1 had bilateral iris absence, punctate posterior synechiae and largely normal fundi. The female proband in Family 2 had bilateral iris absence, anterior subcapsular lens opacities, poor vision and photophobia, while her mother had bilateral iris hypoplasia. The male proband in Family 3 had iris absence, photophobia, nystagmus and foveal hypoplasia confirmed by optical coherence tomography. He also had mild intellectual disability and had been diagnosed with Wilms tumor before presenting for ophthalmic evaluation. Because only three families were studied, the authors cautioned against systematic comparisons of severity between variant classes.
In Family 1, array comparative genomic hybridization identified an approximately 3.19-megabase heterozygous deletion at 11p13, designated arr[GRCh37] 11p13(31037165_34227327) × 1. The interval encompassed all of PAX6 as well as WT1, ELP4, RCN1, CCDC73 and other genes. High-resolution karyotyping at 550–700-band resolution showed a normal 46,XX complement and did not detect the deletion. Quantitative PCR across PAX6 exons 2, 7 and 14 confirmed that the proband had a single copy, while her parents and younger sister had normal copy numbers. The variant was therefore considered de novo and was classified as pathogenic under the cited ACMG and ClinGen copy-number-variant standards.
Family 2 had a different molecular architecture. Whole-exome sequencing detected a heterozygous single-base deletion in exon 14 of PAX6, reported on the MANE Select transcript NM_001368894.2 as c.1234del, or p.Ser412ProfsTer127. The deletion occurred within a thymine homopolymer, and the resulting frameshift was predicted to begin at serine 412. Because exon 14 is the terminal coding exon, the altered transcript was predicted to escape nonsense-mediated messenger RNA decay and produce a C-terminally extended protein of 537 amino acids rather than the wild-type 436. The authors therefore considered disruption of the PST transactivation domain more likely than RNA-degradation-mediated haploinsufficiency as the pathogenic mechanism. The variant was absent from ClinVar and gnomAD at the time of analysis.
Sanger sequencing confirmed that the Family 2 proband and her affected mother carried c.1234del, whereas her father did not, establishing maternal inheritance consistent with autosomal dominant aniridia. The variant was classified as likely pathogenic using a point-based framework combining strong PVS1 evidence, supporting evidence for absence from population databases, computational prediction and phenotype specificity. Conservation analysis found the affected serine 412 residue to be highly conserved across humans, chimpanzees, rhesus macaques, pigs, cattle, rats and chickens, although the authors did not present direct functional experiments. Targeted sequencing of amniotic fluid during a subsequent pregnancy showed that the fetus did not carry the family variant. This result enabled prenatal exclusion of the known familial mutation, but it was a targeted test rather than a genome-wide assessment of all possible fetal abnormalities.
Family 3 carried another de novo deletion, detected through read-depth analysis of whole-exome data rather than by the initial karyotype. The approximately 2.998-megabase interval, seq[GRCh37] del(11)(p14.1p13) NC_000011.9:g.30352495_33350179del, encompassed all 14 exons of PAX6 and all 10 exons of WT1, together with DCDC1, DNAJC24, ELP4, RCN1 and other genes. Copy-number estimates were consistent with loss of one copy, and quantitative PCR independently confirmed the deletion in PAX6 exons 2, 7 and 14. Normal results in both parents supported a de novo origin. Like the Family 1 deletion, this copy-number variant was classified as pathogenic and met the molecular diagnostic criteria for WAGR syndrome because it included WT1.
The contrast between methods was central to the diagnosis. High-resolution chromosome karyotyping produced apparently normal results in all three probands, including the two children with megabase-scale deletions. Array comparative genomic hybridization identified the Family 1 deletion, while whole-exome read-depth analysis detected the Family 3 deletion and also identified the small frameshift in Family 2. Quantitative PCR supplied independent confirmation and helped assess carrier status within families, while Sanger sequencing validated the single-base variant, supported co-segregation and provided the targeted prenatal result. The findings therefore demonstrate diagnostic complementarity in this small sample, but they do not establish the sensitivity of each platform for aniridia more broadly. The authors noted that a prior Chinese study reported a 98.3% overall detection rate when next-generation sequencing and multiplex ligation-dependent probe amplification were combined, although that result came from a separate cohort.
The PAX6 variants all converge on loss of normal gene function, but the study suggests that genetic context can determine the broader phenotype. PAX6 encodes a transcription factor involved in development of the lens, cornea, retina and iris, and PAX6 haploinsufficiency is the principal mechanism underlying congenital aniridia. The two large deletions produced the eye phenotype by removing one copy of the gene and surrounding genomic material. The Family 2 frameshift was instead interpreted as a disruption of the terminal transactivation domain. Yet the limited sample did not support broad claims that deletion variants are consistently more severe than point variants. The extraocular findings in Family 3, including Wilms tumor and mild intellectual disability, were more plausibly associated with the larger deleted region and additional genes than with PAX6 alone, although the researchers called for further neurodevelopmental follow-up.
The inheritance patterns also created different reproductive implications. Both large deletions were de novo, leaving the parents with a low recurrence risk, subject to the possibility of gonadal mosaicism, and a 50% chance of transmission if either proband later reproduced. In Family 2, the affected mother was a heterozygous carrier and therefore had a 50% risk of passing the variant to a child; prenatal testing of the fetus examined in this study showed that it did not carry the variant. WAGR-spectrum findings in Families 1 and 3 prompted the authors to emphasize coordinated ophthalmic, renal, oncological and neurodevelopmental follow-up, including abdominal ultrasound surveillance for the child in Family 1. Overall, the report shows how integrating chromosome-level, sequencing, array and targeted molecular methods can identify aniridia variants that differ greatly in size and mechanism, while clarifying whether a deletion also carries WT1 and therefore an associated tumor risk.
Subject of Research: Genetic diagnosis of PAX6-related congenital aniridia and WAGR syndrome in three Chinese families
Article Title: Diverse PAX6 variants in three Chinese families with congenital aniridia: genotype–phenotype correlations and diagnostic implications
Article References: Ku, W.-J., Wei, T.-Y., Zhang, J.-L., Kang, M.-X., Yu, G., Liu, J.-E., Zhang, J., Hu, H.-Y., & Zhang, F. (2026). Diverse PAX6 variants in three Chinese families with congenital aniridia: genotype–phenotype correlations and diagnostic implications. Frontiers in Medicine, 13, 1978944. https://doi.org/10.3389/fmed.2026.1978944
Image Credits: AI Generated
DOI: 10.3389/fmed.2026.1978944
Keywords: PAX6, congenital aniridia, genetic diagnosis, whole-exome sequencing, copy-number variants, frameshift variant, WAGR syndrome, WT1, Wilms tumor, genetic counseling, Diverse, variants
