At the Northwest Women's and Children's Hospital (Xi'an, China), an affected family was identified in December 2024 when the mother came for advice on having a second child. A 7-year-old girl and her mother, aged 36 years, exhibited congenital hearing loss within this family. Subsequently, clinical information was comprehensively collected from the affected individuals and peripheral blood samples (6 ml each) were obtained from the proband, and her parents and grandparents, in EDTA-anticoagulant tubes. All patients or their legal guardians provided written informed consent for both genetic counseling and molecular genetic testing. The present study was approved by the Research Ethics Committees of Northwest Women's and Children's Hospital and was conducted in accordance with the ethical principles of the Helsinki Declaration for medical research.

Genomic DNA was extracted from peripheral blood samples. Library preparation, sequencing and data analysis were performed as previously described (16). Briefly, the standard procedure included DNA fragmentation, end repair, amplification purification and library quality control. Target regions were captured and enriched to ensure sufficient coverage. Multiplexed sequencing was performed on the AmCareSeq 2000 sequencer (Guangzhou Jiajian Medical Testing Co., Ltd.) using a 2x150 bp paired-end sequencing kit (Guangzhou Jiajian Medical Testing Co., Ltd.). The sequencing covered all coding exons and flanking 20-base pair introns, generating raw FASTQ data with an average depth of ~200x. Raw FASTQ data were processed using a standard bioinformatics pipeline whereby: i) Trimmomatic software (version 0.39) (17)was used to remove low-quality reads (Phred quality score <20; read length <50 bp) and adapter-contaminated reads; ii) clean reads were aligned to the human reference genome (GRCh37/hg19) using the Burrows-Wheeler Aligner software package (version 0.7.15) (18) under default parameters; iii) aligned Binary Alignment/Map files were processed with the Genome Analysis Toolkit (version 4.4.0; Broad Institute) for duplicate marking (‘MarkDuplicates’), base quality score recalibration, variant calling (‘HaplotypeCaller’) and outputting variants in Variant Call Format files; and iv) variants were filtered by excluding those with a minor allele frequency (MAF) >1% in public databases [Exome Aggregation Consortium (https://avillach-lab.hms.harvard.ed-u/access-data/open-data); Genome Aggregation Database (https://gnomad.broadinstitute.org)] low-quality copy number variation (CNV) and synonymous or deep intronic variants not near splicing sites. Furthermore, pathogenicity and evolutionary conservation analyses were analyzed using the PolyPhen tool (version 2) and VarCards database (version 2.0) (19). The final pathogenicity assessment of variants and CNVs was conducted according to the guidelines of the American College of Medical Genetics and Genomics (ACMG) (20). Primer pairs were then designed to validate these candidate pathogenic variants using Sanger sequencing (Table I).

To predict the effect of the novel variant on mRNA transcripts of PLS1, the online software AUGUSTUS (https://bioinf.uni-greifswald.de/webaugustus/) was used with scripts and parameters set according to the official manual (21). Prediction results were visualized using the Integrative Genomics Viewer (version 2.19.7) (22). Homology modeling was employed to construct structural models of the PLS1 and β-actin (ACTB) complex using SWISS-MODEL (https://swissmodel.expasy.org) and MODELLER (version 10.8) (23). The model was constructed using a multi-template alignment strategy based on the crystal structure of human PLS1 [Protein Data Bank (PDB): 1AOA], a plant fimbrin structure resolved by cryo-electron microscopy and ACTB (PDB: 3BYH). The final structural model was visualized using PyMOL (version 3.1; Schrödinger).

Total RNA was extracted from the peripheral blood of all patients and controls using the TRIzol^® (Thermo Fisher Scientific, Inc.) method. Furthermore, 1,000 ng of mRNA was reverse-transcribed into cDNA using the PrimeScript^™ RT reagent kit according to the manufacturer's instructions (Takara Bio Inc.). Primers were designed (forward, 5'-CTTGGTCTTGGGACTTCTCTG-3'; and reverse, 5'-TGTTCTTCCCAAGTTCCACT-3') to amplify the PLS1 exons of interest in affected family members. Thermocycling conditions were an initial denaturation at 98˚C, followed by annealing at 58˚C and extension at 72˚C for 35 cycles. The amplification products were verified through 1% agarose gel electrophoresis, and visualized by gel imaging system (ChemiDoc^™XRS; Bio-Rad Inc.). and then subjected to Sanger sequencing (conducted by Beijing Tsingke Biotech Co., Ltd.).

Within the present study, the two affected individuals exhibiting hearing loss, were members of a three-generation Chinese family from Shanxi (Fig. 1A). The proband (III-1), a 7-year-old girl, was diagnosed with congenital hearing loss, failing newborn hearing screening, which included otoacoustic emissions and automatic auditory brainstem response (Fig. 1B). The patient also exhibited an abnormal V wave response in the right ear. At age 6, pure-tone audiometry revealed bilateral symmetric severe hearing loss at medium-to-high frequencies (250-8000 Hz), with abnormal air and bone conduction thresholds (Fig. 1C). Tympanometry showed a ‘type As’ tympanogram in the right ear and a ‘type B’ in the left. Typical inflammatory symptoms were absent, ruling out conditions such as otitis media. The probands mother, aged 36 years, also exhibited congenital hearing loss, with pure-tone audiometry demonstrating levels markedly above the normal threshold (≤25 dB HL) (Fig. 1D). At present, both mother and daughter require hearing aids for daily life. Furthermore, the father and maternal grandparents of the proband currently exhibit no hearing abnormalities.

To determine the genetic etiology of the family's congenital hearing loss, whole-exome sequencing was performed on the proband. After filtering out variants with an MAF >0.01 in public databases, the hearing loss-related genes matching the patient phenotype were focused on. Subsequent bioinformatics analysis identified a novel mutation, PLS1 c.981+5G>A (reference sequence: NM_001145319.2), as the most likely candidate variant. Sanger sequencing was applied to further demonstrate segregation of this mutation within the disease presented by the family (Fig. 2A). The mutation was found to be present in the affected mother but absent in the healthy grandparents (data not shown), indicating the de novo nature of the mutation within the mother.

According to the ACMG guidelines and specialized criteria for hereditary hearing loss (24), the PLS1 c.981+5G>A variant is classified as a Variant of Uncertain Significance (VUS) based on the following evidence codes: i) PM2 (absent or very low frequency in normal population databases); ii) PM6-P (assumed de novo without confirmed maternity and paternity. A phenotype consistent with the genotype but not highly specific and with a high degree of genetic heterogeneity); and iii) PP1 (co-segregation with disease). Supporting its potential pathogenicity, the mutation was predicted to alter PLS1 splicing by SpliceAI (Illumina, Inc.).

The PLS1 c.981+5G>A variant is located near a splicing site. Therefore, using the AUGUSTUS program, this mutation was predicted to cause the skipping of exon 9 (Fig. 2B). To validate this, total RNA was extracted from the lymphocytes of all patients, including the control (the father) and fragments were analyzed through reverse transcription PCR and agarose gel electrophoresis. In addition to the normal band, the proband and her affected mother exhibited an unexpected, lower band that was absent in the control (Fig. 2C). Sanger sequencing demonstrated that this extra band corresponded to a transcript lacking the entire exon 9 (Fig. 2D). These findings demonstrate that the PLS1 c.981+5G>A mutation causes exon 9 skipping, generating an aberrant splicing product.

To understand the physiological impact of the exon 9 skipping, a structural model of the actin-binding domain 1 (ABD1) of PLS1 complexed with ACTB was constructed based on homology modeling (Fig. 3A) using the SWISS-MODEL program. Structural analysis revealed that the 31 amino acid residues (p.S297_N327) encoded by exon 9 were part of the calponin-homology (CH)-2 domain within ABD1. The deletion of this region results in the loss of an α-helix (residues 293-296), which directly faces the ACTB (Fig. 3A). Therefore, the present study proposes that this structural alteration disrupts the PLS1-ACTB interaction. Furthermore, residues 293-327 are evolutionarily conserved across species, underscoring their functional importance (Fig. 3B). Collectively, these data suggest that the PLS1 c.981+5G>A mutation impairs the binding between PLS1 and ACTB.