Enrollment of subjects and exome sequencing of probands
The study was approved by the Institutional Review Board (Protocol no. 20081138) at the University of Miami (USA) and Ethics Committee (Protocol no. 012413) at Ankara University Medical School (Türkiye). Written informed consents were obtained from all participants and in the case of minors, it was obtained from parents. Audiometry was performed to measure average hearing thresholds for all participants under standard conditions and guidelines. All affected individuals were examined by a clinical geneticist and otolaryngologist.
For exome sequencing (ES), we followed a recently published protocol (Ramzan et al. 2024). Briefly, single nucleotide, indel, and copy number variants (CNVs) in all known deafness genes were analyzed. Variants were retained for further evaluation if they had an allele frequency of less than 0.01. The variants in all known genes for HL were analyzed using a larger list retrieved from hereditary hearing loss homepage (https://hereditaryhearingloss.org/) and OMIM. If there was no candidate variant identified in known deafness genes, ES data were re-examined for all genes containing variants less than 0.01 allele frequency. CNV analysis with ES data used CoNIFER v.02.2 with default parameters. It uses a singular value decomposition method to correct systematic biases and identifies a CNV call if the corrected signal reaches a predefined threshold at no less than three consecutive exons (Krumm et al. 2012). We performed Sanger sequencing to confirm candidate variants and evaluate segregation within respective families.
Candidate variants were classified according to ACMG and ClinGen Hearing Loss Expert Panel (HL-EP) specifications, aligned with the ACMG/AMP Variant Interpretation Guidelines (Oza et al. 2018; Richards et al. 2015; Tavtigian et al. 2020). AlphaMissense was included in the prediction of pathogenicity as described (Zhuo et al. 2024).
iPSC maintenance and validation
Cells from a validated male European lymphocyte-derived iPSC line (ASE9203) were cultured on Vitronectin (Thermo Scientific, USA, Cat# A2858501) coated culture ware and maintained in E8-Flex (Thermo Scientific, USA, Cat# A2858501) media supplemented with 100 µg/mL Normocin (Invivogen, USA, Cat# ant-nr-05), following the manufacturer's guidelines. Routine passages utilized 1x Revitacell (Thermo Scientific, USA, Cat# A2644501). We regularly assessed pluripotency markers OCT4 (Cell Signaling Technology, USA, Cat# 2840) and TRA-1-60 (Cell Signaling Technology, USA, Cat# 4746) through Immunocytochemistry (ICC).
Generation of CRISPR/Cas9-edited variants
We generated monoclonal cell lines for FGF3KO, FGF3c.493A>G, GREB1LKO, GREB1L c.556T>C, and PBXIP1c.1722G>A as previously described (Ramzan et al. 2024). Custom sgRNA and HDR donor blocks were designed and procured using the Alt-R-CRISPR/Cas9 platform (IDT); sequences are available in supplementary Table 1. Cas9-gRNA ribonucleoprotein complexes were assembled under sterile conditions using 140 pmol sgRNA and 40 pmol EnGen® Spy Cas9 HF1 (New England Biolabs, USA, Cat# M0667M) and incubated for 10 mins at room temperature. Following the incubation, 8x105 cells were suspended in the RNP complex and premixed 3 µM HDR donor block in 80 µL of nucleofector solution. The final volume of the nucleofection mix was 100 µL. Nucleofections are performed using the P3-primary Cell 4D-Nucleofector Kit (Lonza, USA, Cat# V4XP-3024) and preprogrammed pulse protocol CA-137 in the 4D-Nucleofector system (Lonza). Post-nucleofections, iPSCs were cultured in E8-flex media supplemented with CloneR™2 (Stem Cell Technologies, Canada, Cat# 100–0691) and 1 µM of the HDR enhancer-V2 (IDT).
DNA isolation and confirmation of CRISPR/Cas9-edited cell pools and off-targets
DNA isolation was done using QuickExtract DNA Extraction Solution (Biosearch Technologies, USA, Cat# QE09050); the resulting DNA was then PCR amplified using Phusion® High-Fidelity PCR Master Mix (New England Biolabs, USA, Cat# M0531S) and sequenced with primers flanking (Supplementary Table S1) the guide region and Sanger sequencing at Genewiz (Azenta Life Sciences). Outcomes of CRISPR/Cas9 knock-ins were assessed using ICE Analysis (Conant et al. 2022). For every guide sequence, off-target analysis was done using bioinformatic tools such as Cas-OFFinder (Bae et al. 2014) and the sgRNA guide analysis tool from IDT in Coralville, Iowa, USA. The top five off-target sites were Sanger sequenced for flanking primers and analyzed for all the clonal lines used in this study (Supplementary Table S2).
Isolation of monoclonal lines
Single-cell monoclonal lines were isolated using Poisson distribution (Sanjurjo-Soriano et al. 2022). CRISPR/Cas9 pools were briefly dissociated using StemPro accutase (Thermo Scientific, USA, Cat# A1110501). The resulting cell suspension was collected in E8-Flex media (Thermo Scientific, USA, Cat# A2858501) containing CloneR™2 (Stem Cell Technologies, Canada, Cat# 100–0691) and centrifuged at 300 rpm. After centrifugation, the single-cell suspension was passed through 70 µm strainers (SP Bel-Art, USA, Cat# H13680-0070), and the cells were then counted using a Countess II cell counter (Thermo Scientific, USA). Depending on the cell concentrations, serial dilutions were performed to achieve a typical concentration of 1 cell per 100 µL of media. The cells were then plated in 96-well plates precoated with rhLaminin (Thermo Scientific, USA, Cat# A29249), and media changes were done with 1x CloneR™2-containing E8 media every other day for the first week and then with 0.5x CloneR™2 containing E8-Flex media till the single cell colonies reached a passaging confluence. Clonal lines were identified using Sanger sequencing, and from the same experiments, clonal lines bearing large frameshift deletions disrupting the open reading frame were used as knockouts for FGF3 and GREB1L (Supplementary Figure S1).
Generation of inner ear organoids
IEOs were generated using the previously described protocol by (Moore et al. 2023). Briefly, on the first day of differentiation (day − 2), cells were dissociated with StemPro accutase (Thermo Scientific, USA, A1110501). The cells were then passed through 70 µm strainers (SP Bel-Art, USA, Cat# H13680-0070), and 3500 cells/well were plated in 96-well Nunc-Sphera U-bottom plates (Thermo Scientific, USA, Cat# 174925) in 100 µL of E8-Flex media containing 20 µM Y-27632 (Stem Cell Technologies, Canada, Cat# 72304). After 4 hours of initial plating, 100 µL of fresh E8-Flex media was added to each well, bringing the final concentration of Y-27632 to 10 µM. All the aggregates were collected on day 0, washed thrice with E6-Medium (Thermo Scientific, USA, Cat# A1516401) and IEO differentiation media (E6-Medium, 2% GFR-Matrigel (Corning, Sigma Aldrich, Cat# 354230) 10 µM SB431542 (Stem Cell Technologies, Canada, Cat# 72232), 4 ng/mL FGF2 (Peprotech, Thermo, USA, Cat# 100-18B) and 2.5 ng/mL BMP4 (Stemgent, USA, Cat# 03–0007)). Cell aggregates were transferred to a new 96-well Nunc-Sphera U-bottom plate in 100 µL of differentiation media. On day 3, 25 µL E6-Medium containing 100 µg/mL Normocin, 250 ng/mL FGF-2, and 200 nM LDN193189 (Stem Cell Technologies, Canada, Cat# 72147) were added, bringing the volume to 125 µL. On days 6 and 9 of differentiation, cell aggregates were washed thrice with E6 media and then thrice with E6 media containing 100 µg/mL Normocin, 3 mM CHIR99021 (Stemgent, USA, Cat# 04-0004-02), 200 nM LDN193189, and 50 ng/mL FGF2. The cell aggregates were transferred to fresh 96-well U-bottom plates in 250 µL of new media.
Cell aggregates were washed thrice on the 11th day of differentiation with Advanced DMEM/F12 (Thermo Scientific, USA, Cat# 12634010). Then they were transferred to 90 mm Nunc low-attachment plates (Thermo Scientific, USA, Cat# 174945) in 10 mL of organoid maturation media (OMM) comprising Advanced DMEM/F12, Neurobasal medium (Thermo Scientific, USA, Cat# 21103049), 1X Glutamax, 0.5X B-27 supplement without Vitamin A, 0.5X N-2 supplement, 0.1 mM 2-Mercaptoethanol, 100 µg/mL Normocin, and supplemented with 1% GFR-Matrigel and 3 µM CHIR99021. On days 13 and 15, culture media was changed to OMM + 3 mM CHIR99021 + 1 mM Purmorphamine (Stem Cell Technologies, Canada, Cat# 72202). On day 18, aggregates were washed to eliminate CHIR99021, and the media was changed to OMM + 3 mM IWP-2 (Stem Cell Technologies, Canada, Cat# 72122) + 1 mM Purmorphamine. The media was changed on day 20, with fresh media used on day 18. On day 22, cultures were washed and transferred to an anti-adhere solution coated with a low attachment 100 mm culture dish in OMM-only. The cultures received half media change every 3rd day and complete media change every 7th day until day 60. Samples were collected on day 25 and day 35 for RNA sequencing and on day 25, day 35, and day 60 for immunohistochemistry.
Immunohistochemistry and sectioning
IEOs on day 25 and day 60 were collected, washed with PBS twice, fixed with 4% paraformaldehyde, and processed at the Cancer Modeling Shared Resource core, Sylvester Cancer Center, University of Miami. Serial sections of 5 µm thickness were antigen-retrieved using 10 mM citrate buffer pH6. For immunohistochemistry, sections were permeabilized with 0.4% triton-X for 10 mins and blocked with 5% BSA + 0.01% Tween20. Primary antibody incubations for MYO7A 1:50 ((MYO7A 138-1, deposited to the DSHB by Orten, D.J. (DSHB Hybridoma Product MYO7A 138-1)), and SOX2 1:100 (Cell Signaling Technology, USA, Cat# 3579) was done in the blocking buffer for overnight at 4oC. The following sections were washed thrice with PBS the following day, and secondary antibody incubations at 1:500 dilutions for Anti-Mouse AlexaFluor-647 (Thermo Scientific, USA, Cat# A32728TR) and anti-Rabbit AlexaFluor-488 (Thermo Scientific, USA, Cat# A32787TR) were done for 1hr at room temperature, respectively. Slides were mounted with ProLong™ Glass Antifade (Thermo, USA, Cat# P36980) and images were acquired using Zeiss LSM 980 with AiryScan 2 (Zeiss, Germany) at Flow Cytometry Shared Resource (FCSR), University of Miami. All the images were analyzed using the Fiji Image Analysis Tool or ImageJ (Schindelin et al. 2012).
Western blot analysis
Cells were harvested in RIPA buffer supplemented with 1x HALT protease and phosphatase inhibitor (Thermo Scientific, USA, Cat# 78441). Protein quantification was performed using a Thermo Scientific™ Pierce™ BCA kit (Thermo Scientific, USA, Cat# 23227). Equal amounts of protein were then reduced and loaded onto a 4–20% Tris-Glycine gradient gel for separation, following the method described by Laemmli (Laemmli 1970). Subsequently, proteins were transferred onto a 0.22 µm PVDF membrane using the Turbo-trans Blot system (Biorad, USA). The membranes were then blocked in 5% BSA for 1.5 hours and incubated overnight at 4°C with primary PBXIP1 antibody (Proteintech, Thermo Scientific, USA, Cat# 12102-1-AP) diluted at 1:1000 in 5% BSA + TBST (TBS with 0.5% Tween). After washing with TBST, the blots were incubated with HRP-conjugated anti-rabbit goat secondary antibody (1:3000) diluted in 5% BSA + TBST for 1.5 hours at room temperature. Following the termination of antibody reactions, the blots were washed three times with TBST and developed using the West Pico Super-Signal ECL substrate (Thermo Scientific, USA, 37069). Finally, visualization was performed using FluorChemE (ProteinSimple, USA).
mRNA expression analysis
The expression of GREB1L, FGF3, and PBXIP1 mRNA in monoclonal lines was analyzed using qRT-PCR. Total RNA was isolated with TRIzol Reagent (Thermo, USA, Cat# 15596026) according to the manufacturer’s instructions. cDNA was synthesized using qScript XLT cDNA SuperMix (Quanta Biosciences, USA, Cat# 9516-025). The primers amplifying the transcript were GREB1L sense 5’-CAGTTTCCTGGCATCACATTTC-3’ antisense 5’-GTAACCACACTGTCTCCTCTTC-3’; FGF3 sense 5’-ATTGCTCCTGGGTGGAAATTA-3’ antisense 5’-AGAGAGAAAGAGAGGGAGAGTG-3’; PBXIP1 sense 5’-GGCCTCTCTGCTAAGAACATAC-3’ antisense 5’-GATGCCATCCTCACCAAAGA-3’.
Quantification and statistical analysis
All the statistical analyses are performed using GraphPad Prism 10. Paired analyses were done using the student’s t-test. Multiple comparisons were performed using one-way ANOVA with Tukey’s multiple comparisons test. The results are expressed as Mean ± SEM; a statistical difference of p ≤ 0.05 was considered significant. The significant differences are marked with (*) whenever comparisons were made between GREB1Lc.556T>C, FGF3c.493A>G, FGF3KO, GREB1LKO, PBXIP1c.1722G>A and their respective controls GREB1LWT, FGF3WT, PBXIP1WT. Details about the number of replicates and significance notation are provided in the figure legends.