Hearing Loss Autosomal Recessive 116

Mendelian MONDO:0033670 Pathograph 12 Show in embeddings browser Autosomal Recessive Nonsyndromic Hearing Loss

DFNB116 is autosomal recessive nonsyndromic sensorineural hearing loss caused by biallelic variants in CLDN9, which encodes claudin-9, a constituent of the bicellular tight junctions of the epithelia that line the scala media. Three families are reported with functional support - a Turkish family with the truncating c.86delT (p.Leu29ArgfsTer4), and Pakistani and Moroccan families with the hypomorphic p.(Glu159Lys) and p.(Ile124dup) alleles - plus one further individual whose p.(Cys25Trp) allele is proposed rather than established. The mechanism this entry curates is a paracellular ion barrier failure, and it is worth being precise about which compartment fails, because "tight junction gene, therefore endolymph barrier" is the intuitive story and it is not quite what was measured. In the Cldn9 nmf329 mouse the endocochlear potential and the endolymphatic potassium concentration are both normal - the stria vascularis works - while the *perilymphatic* potassium concentration is raised. Claudin-9 sits in a subapical subdomain of the hair cell junctional complex, underneath the more apical strands built by other claudins, and what its loss does is let potassium leak from the endolymph past those outer strands to the basolateral face of the hair cell. Hair cells then die, from the cochlear base upward. That the potassium is what kills them is not inferred: outer hair cell loss is prevented by culturing the explanted organ of Corti in low potassium, and in vivo by deleting pou3f4 to collapse the endocochlear driving force. None of that has been measured in a human cochlea. The human evidence is segregation in consanguineous families plus heterologous-cell and structural-modelling work on the variant proteins; the barrier mechanism reaches humans by inference from the mouse, and the pathophysiology nodes downstream of the protein defect are marked PROVISIONAL for that reason. The mouse also diverges from the human phenotype in a way that matters: it is uniformly deaf across frequencies, whereas the human audiogram characteristically spares the low frequencies and slopes steeply above 1 kHz. Clinically, DFNB116 is a mild-to-profound, often asymmetric loss whose reported onset ranges from prelingual to adolescence, and which is progressive in some families and stable over a decade in others. That milder end is the ascertainment point the second report makes explicitly - recessive deafness genetics has concentrated on profound congenital cases, and CLDN9 is a gene to consider when the loss is moderate.

Ask OpenScientist

Ask a research question about Hearing Loss Autosomal Recessive 116. OpenScientist will conduct autonomous deep research using the Disorder Mechanisms Knowledge Base and PubMed literature (typically 10-30 minutes).

Submitting...

Do not include personal health information in your question. Questions and results are cached in your browser's local storage.

1
Mappings
1
Inheritance
6
Pathophys.
6
Phenotypes
4
Gaps
12
Pathograph
1
Genes
3
Medical Actions
2
Models
5
References
1
Deep Research
🏷

Classifications

Harrison's Part
DISORDER OF EAR GENETICS ENVIRONMENT DISEASE
🔗

Mappings

MONDO
MONDO:0033670 hearing loss, autosomal recessive 116
skos:exactMatch MONDO
👪

Inheritance

1
Autosomal recessive HP:0000007
Biallelic CLDN9 variants, homozygous in all three functionally supported families, each of which is consanguineous or from a population where consanguinity is common. Heterozygous carriers are unaffected: the father in the Turkish family had a normal audiogram, and carriers in the Pakistani and Moroccan families were asymptomatic. The same recessive pattern holds in the mouse, where heterozygotes are audiometrically indistinguishable from wild type.
Autosomal recessive inheritance
Show evidence (2 references)
PMID:31175426 SUPPORT Human Clinical
"Sanger sequencing of all four family members showed co-segregation of the variant with the phenotype as an autosomal recessive trait in the family"
Recessive segregation of the truncating allele in the founding Turkish family.
PMID:34265170 SUPPORT Human Clinical
"Physical examinations and medical histories in both families did not reveal other clinically relevant phenotypes and carriers of the variants were asymptomatic"
Carriers unaffected in both subsequently reported families, and no extra-auditory features - the two claims that make this recessive and nonsyndromic.
?

Discussions and Knowledge Gaps

4
Does the cochlear paracellular potassium leak demonstrated in the Cldn9 mouse actually occur in humans with biallelic CLDN9 variants, and is it what deafens them?
KNOWLEDGE GAP cldn9_barrier_mechanism_unmeasured_in_humans
Attached to
pathophysiology#Failure of the Claudin-9 Paracellular Ion Barrier pathophysiology#Basolateral Potassium Exposure of Cochlear Hair Cells
The barrier mechanism is well demonstrated - in mice and in transfected epithelial monolayers. In humans it is entirely inferential. What has been shown in affected people is that a CLDN9 allele segregates with hearing loss and that the encoded protein behaves abnormally when overexpressed in HEK293, HeLa or MDCK cells. Nobody has measured a human perilymphatic potassium concentration, a human endocochlear potential, or the state of a human cochlear tight junction, and the tissue required is not obtainable during life. That gap is easy to overlook because the intermediate steps are individually plausible: claudin-9 is a tight junction protein, it is expressed in the cochlear epithelia, and the mouse phenotype matches. But the entry marks each downstream node PROVISIONAL rather than ESTABLISHED, because two of the three human alleles fail in a way the mouse allele does not - they never reach the membrane, whereas F35L sits in the membrane and simply stops sealing. A mistrafficked claudin could in principle disturb junction assembly more broadly than a barrier-dead one that is correctly incorporated. The tractable route is not a human measurement. It is a knock-in mouse carrying a human allele - which is exactly what the second report proposes - with the same electrode measurements repeated. If a p.(Glu159Lys) or p.(Ile124dup) knock-in shows the same isolated perilymphatic potassium rise with a preserved endocochlear potential, the inference to humans becomes much stronger; if it does not, the human alleles are doing something else.
Show evidence (2 references)
PMID:34265170 SUPPORT Human Clinical
"Identifying additional mutant alleles of CLDN9 and engineering animal models with the corresponding human variants could further our understanding of pathogenic mechanisms that expose therapeutic opportunities."
The authors propose allele-matched animal models, which is the experiment that would close this gap.
PMID:31175426 SUPPORT INDIRECT Model Organism
"Since CLDN9 is involved in paracellular permeability, this deformity is likely to impair functionality."
Included as an example of the inferential step itself. The founding paper reasons from claudin-9's known role to the significance of a murine morphological finding; marked INDIRECT because that is a plausibility argument rather than a measurement.
Why is the Cldn9 mouse uniformly and severely deaf across all frequencies from the onset of hearing, when humans with biallelic CLDN9 variants have normal low-frequency thresholds and a loss that can be as mild as moderate and may not progress at all?
HUMAN MODEL MISMATCH cldn9_mouse_human_audiometric_mismatch
Attached to
pathophysiology#Basal-Predominant Outer Hair Cell Degeneration phenotypes#High-Frequency Sensorineural Hearing Impairment animal_models#nmf329 Cldn9 F35L mouse
The mismatch is not subtle. ABR thresholds in the nmf329 mouse are uniformly elevated at 8, 16 and 32 kHz and are already severe at P16, days after hearing normally begins. The human phenotype is the opposite shape - normal at 500 Hz, steeply sloping above 1 kHz - and spans mild to profound with recognition ranging from prelingual to age 17. Two explanations are separable and neither has been tested. The first is allele strength: the mouse F35L completely eliminates barrier function in the Ussing chamber assay, while two of three human alleles are explicitly described as hypomorphic, so the human cochlea may retain enough barrier to protect the apex. The second is that the mouse histology itself is not uniform - degeneration begins basally and reaches the apex only with age - so a uniformly flat ABR deficit in an animal with a basal-predominant lesion suggests the mouse cochlea is more vulnerable overall, or that the ABR at P28 is already reporting a later stage of the same gradient the human audiogram catches early. This matters for more than tidiness. If the human low-frequency sparing reflects residual barrier function rather than a slower version of the same total loss, then the apical cochlea in a DFNB116 patient contains living hair cells for decades, and the therapeutic question is preservation rather than restoration. A knock-in of a human hypomorphic allele with frequency-resolved ABR and turn-by-turn hair cell counts would separate the two accounts.
Proposed experiments
Frequency-resolved audiometry and cochlear histology in a human-allele knock-in mouse
cldn9_human_allele_knockin_frequency_abr
Generate mice carrying the human p.(Glu159Lys) or p.(Ile124dup) allele, measure ABR thresholds by frequency across the lifespan, and count hair cells turn by turn. A hypomorphic knock-in that reproduces apical sparing would attribute the human audiogram shape to residual barrier function; one that is uniformly deaf like nmf329 would locate the difference in species biology instead.
Would support
pathophysiology#CLDN9 Hypomorphic Missense and In-Frame Alleles
Would refute
pathophysiology#CLDN9 Hypomorphic Missense and In-Frame Alleles
Supporting outcome
  • Knock-in mice carrying a human hypomorphic allele show elevated high-frequency ABR thresholds with preserved low-frequency thresholds, and hair cell counts that are reduced at the base and preserved at the apex.
Refuting outcome
  • Knock-in mice are uniformly deaf across frequencies with pan-cochlear hair cell loss, indistinguishable from nmf329, indicating that allele strength does not explain the human audiogram shape.
Show evidence (3 references)
PMID:19696885 SUPPORT Model Organism
"The hearing loss of nmf329/nmf329 mice was equally severe at high- and low-frequency, as indicated by the uniformly high ABR thresholds at 8, 16, and 32 kHz"
The mouse side of the mismatch, measured across three frequencies.
PMID:31175426 SUPPORT Human Clinical
"Hearing thresholds show a normal hearing level at 500 Hz and a steep decline after 1000 Hz in both sisters."
The human side of the mismatch, in the family with the strongest (truncating) allele.
PMID:34265170 SUPPORT Human Clinical
"here we report families with two likely pathogenic CLDN9 variants both of which appear to be hypomorphic."
The residual-function hypothesis for the mismatch - the human alleles studied so far are not nulls.
Is DFNB116 caused by having too little claudin-9, or by having claudin-9 that cannot seal - and do the hypomorphic human alleles act on the same axis as the mouse barrier-dead allele?
KNOWLEDGE GAP cldn9_dosage_versus_barrier_function
Attached to
pathophysiology#CLDN9 Hypomorphic Missense and In-Frame Alleles pathophysiology#Failure of the Claudin-9 Paracellular Ion Barrier animal_models#nmf329 Cldn9 F35L mouse animal_models#Cldn9 tet-OFF conditional knockdown mouse
Two mouse manipulations of the same gene give opposite results. The nmf329 allele leaves normal amounts of claudin-9 in the membrane but abolishes its ion barrier function, and those mice lose their hair cells and go deaf within days of hearing onset. Reducing claudin-9 protein roughly eightfold with a tet-OFF transgene leaves outer hair cell counts normal at fifteen months and shifts ABR thresholds by only 5-15 dB; what it does instead is generate an extra row of functional inner hair cells, implicating claudin-9 levels in lateral inhibition of hair cell fate during a window that closes by P14. So quantity and barrier competence are separable axes, and DFNB116 has been explained entirely on the barrier axis. That is very likely right for the truncating allele, which delivers no functional protein to the junction at all. It is less obviously right for p.(Glu159Lys), which reaches the membrane normally and is predicted to perturb a cis-interaction - a lesion that could plausibly reduce effective claudin-9 in strands rather than open a leak, which the knockdown data suggest would be relatively well tolerated. Nobody has measured paracellular permeability for any human CLDN9 allele. Two things would resolve it: Ussing-chamber permeability measurements on monolayers expressing each human variant, the assay that established the mouse F35L allele as barrier-dead; and attention to whether hypomorphic-allele carriers have any of the developmental signature - supernumerary inner hair cells are not visible on an audiogram, but the reduced afferent synapse counts that accompanied them in the mouse would show up as a wave I amplitude deficit disproportionate to threshold.
Show evidence (4 references)
PMID:37873357 SUPPORT Model Organism
"The findings suggest that Cldn9 levels coordinate embryonic and postnatal HC differentiation, making it a viable target for altering IHC development pre- and post-terminal differentiation."
The dosage axis stated by its own authors - claudin-9 level as a developmental variable, separate from the barrier role that explains the disease.
PMID:37873357 SUPPORT Model Organism
"By contrast, the P14 inner ear injected with Cldn9-shRNA produced no detectable increase in ectopic IHCs as counted by three independent blinded observers"
Bounds the developmental window to before P14, which is what makes it a separate phenomenon from the degeneration that begins after the second postnatal week.
PMID:19696885 SUPPORT In Vitro
"In an epithelial cell line, heterologous expression of wild-type claudin-9 reduced the paracellular permeability to Na+ and K+, and the nmf329 mutation eliminated this ion barrier function without affecting the plasma membrane localization of claudin-9."
The barrier axis, and the assay that would have to be run on the human alleles to place them on it.
+ 1 more reference
Is the CLDN9 c.75C>G p.(Cys25Trp) allele reported in a child with fluctuating hearing loss actually pathogenic, and does fluctuating hearing loss belong in the DFNB116 phenotype?
KNOWLEDGE GAP cldn9_cys25trp_allele_uncertain
Attached to
genetic#CLDN9 phenotypes#Bilateral Sensorineural Hearing Impairment
A fourth individual has been reported with a CLDN9 missense allele at a highly conserved cysteine and a steeply sloping, fluctuating audiogram. It is deliberately excluded from this entry's phenotype and prevalence counts. The reasons are specific rather than generic caution: the same individual carried variants in FLNA and ANKRD11 and had unrelated upper airway findings, no functional work was done on the allele, and the subsequent authors said in print that more evidence would be needed. Fluctuation is the part worth resolving, because it is not merely an extra severity tier - it would be a different mechanism. Nothing in the potassium-leak model predicts a hearing threshold that recovers, since hair cell death does not reverse. A genuinely fluctuating CLDN9 phenotype would imply a reversible functional disturbance of the barrier rather than progressive cell loss, and that would change how the whole entry reads. Establishing it needs functional characterisation of p.(Cys25Trp) and at least one more family.
Show evidence (3 references)
PMID:34265170 SUPPORT Human Clinical
"Recently, a c.75C>G;p.(Cys25Trp) CLDN9 variant was suggested to cause fluctuating hearing loss with a steeply sloping audiogram in one individual"
The report of the fourth allele and its distinctive fluctuating course.
PMID:34265170 SUPPORT Human Clinical
"The affected individual also had variants in FLNA and ANKRD11 and exhibited turbinate hypertrophy, allergic rhinitis and nasopharyngeal nodule."
The confounding genotype and phenotype that make the attribution to CLDN9 uncertain in that individual.
PMID:34265170 SUPPORT Human Clinical
"Additional experimental evidence would be useful in evaluating the predicted pathogenicity of the CLDN9 c.75C>G;p.(Cys25Trp) variant."
The subsequent authors' own judgement that the allele is not yet established.
⚙

Pathophysiology

6
CLDN9 Truncating Loss of Function
The founding allele, homozygous c.86delT (p.Leu29ArgfsTer4), truncates claudin-9 at the start of the first extracellular loop. Because CLDN9 has a single coding exon, a premature stop is unlikely to trigger nonsense-mediated decay, so the truncated protein may be made and simply fail to work. GFP-tagged mutant protein expressed in HEK293 cells stays in the cytosol instead of reaching the plasma membrane, so no functional claudin-9 is delivered to the junction.
CLDN9 hgnc:2051 HUGO Gene Nomenclature Committee (hgnc) Relation: this pathophysiological event involves this gene This pathophysiological event involves CLDN9 (hgnc:2051). hgnc:2051 is a gene from the HUGO Gene Nomenclature Committee.
Genetic context allele_type: single-base deletion causing a frameshift and premature termination variant_origin: GERMLINE zygosity: HOMOZYGOUS functional_impact_category: LOSS_OF_FUNCTION
CLDN9 NM_020982.3:c.86delT, p.Leu29ArgfsTer4, homozygous in three affected members of a consanguineous Turkish family.
Downstream
Failure of the Claudin-9 Paracellular Ion Barrier
Show evidence (3 references)
PMID:31175426 SUPPORT Human Clinical
"The variant is predicted to cause a frameshift and produce a truncated protein (p.Leu29ArgfsTer4) in this single-exon gene."
The predicted molecular consequence, and the single-exon architecture that bears on whether the transcript escapes decay.
PMID:31175426 SUPPORT In Vitro
"This suggests that the variant disrupts the migration of CLDN9 to the plasma membrane."
The measured cellular consequence in transfected HEK293 cells - the mutant protein does not reach the membrane where a claudin has to sit to do anything.
PMID:34265170 SUPPORT Human Clinical
"It is not known if in the inner ear truncated CLDN9 protein remained stable or was degraded."
Records the limit of what is known. The transfected-cell result does not establish what the truncated protein does in a human cochlea, and the later authors say so.
CLDN9 Hypomorphic Missense and In-Frame Alleles
Mechanism confidence: Provisional
The two subsequently reported alleles retain partial function and fail in different ways. p.(Glu159Lys) reaches the membrane normally in MDCK-II cells; structural modelling predicts it perturbs one of the two cis-interactions by which adjacent claudin protomers oligomerise within the same membrane, so the protein is present but the strand it should build is compromised. p.(Ile124dup), an in-frame single-residue duplication, is predicted to misfold locally and does not reach the HeLa cell membrane. The authors classify both as hypomorphic. Whether that residual function explains the milder end of the clinical spectrum is not settled by the data: the family carrying p.(Glu159Lys) had moderate-to-profound loss that was stable over ten years, while the family carrying the truncating allele included a profoundly deaf 46-year-old. Severity and progression do not sort cleanly by allele class across three families.
CLDN9 hgnc:2051 HUGO Gene Nomenclature Committee (hgnc) Relation: this pathophysiological event involves this gene This pathophysiological event involves CLDN9 (hgnc:2051). hgnc:2051 is a gene from the HUGO Gene Nomenclature Committee.
Genetic context allele_type: missense and in-frame single-codon duplication variant_origin: GERMLINE zygosity: HOMOZYGOUS functional_impact_category: PARTIAL_LOSS_OF_FUNCTION
CLDN9 NM_020982.4:c.475G>A p.(Glu159Lys), homozygous in a Pakistani family, and c.370_372dupATC p.(Ile124dup), homozygous in a Moroccan family.
Downstream
Failure of the Claudin-9 Paracellular Ion Barrier
Show evidence (3 references)
PMID:34265170 SUPPORT Computational
"Computational structural modeling predicts that substitution of a lysine for glutamic acid p.(Glu159Lys) alters one of two cis-interactions between CLDN9 protomers."
The proposed molecular lesion for the missense allele. Evidence source is COMPUTATIONAL because the claim rests on a structural model, not on a measurement.
PMID:34265170 SUPPORT In Vitro
"The p.(Ile124dup) variant is predicted to locally misfold CLDN9 and mCherry tagged p.(Ile124dup) CLDN9 is not targeted to the HeLa cell membrane."
Mistrafficking of the in-frame duplication measured in transfected cells, the same failure mode as the truncating allele.
PMID:34265170 SUPPORT In Vitro
"The localization of CLDN9 with the substitution of lysine at position 159 was similar to that of the wild-type, when expressed in MDCK-II cells."
Distinguishes the two hypomorphs: p.(Glu159Lys) traffics normally, so its defect has to be in what the protein does once it is in the membrane rather than in getting there.
Failure of the Claudin-9 Paracellular Ion Barrier
Mechanism confidence: Provisional
Claudin-9 is present in the bicellular tight junctions of every sensory and non-sensory epithelial cell facing the endolymph of the scala media, and within the hair cell junctional complex it occupies a specific subapical subdomain, sitting beneath the more apical strands formed by other claudins. Heterologously expressed wild-type claudin-9 lowers paracellular permeability to sodium and potassium; the deafness-causing mouse F35L substitution abolishes that barrier function while leaving the protein correctly located in the plasma membrane, which separates the barrier defect from a trafficking defect. What this node is not: a loss of the epithelial seal that maintains the endocochlear potential. The stria vascularis is unaffected in the mutant mouse. The claim is specifically that a deeper strand of the junction leaks.
cochlear outer hair cell CL:0000601 Cell Ontology (CL) Relation: this pathophysiological event involves this cell type This pathophysiological event involves cochlear outer hair cell (CL:0000601). CL:0000601 is a cell type from the Cell Ontology. cochlear inner hair cell CL:0000589 Cell Ontology (CL) Relation: this pathophysiological event involves this cell type This pathophysiological event involves cochlear inner hair cell (CL:0000589). CL:0000589 is a cell type from the Cell Ontology.
paracellular transport of potassium and sodium GO:0160184 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves increased paracellular transport of potassium and sodium, annotated with paracellular transport (GO:0160184). GO:0160184 is a biological process from the Gene Ontology. ↑ INCREASED
bicellular tight junction GO:0005923 Gene Ontology (GO) Relation: this pathophysiological event involves this cellular component This pathophysiological event involves bicellular tight junction (GO:0005923). GO:0005923 is a cellular component from the Gene Ontology.
spiral organ of cochlea UBERON:0002227 Uberon multi-species anatomy ontology (UBERON) Relation: this pathophysiological event occurs in this anatomical location This pathophysiological event occurs in spiral organ of cochlea (UBERON:0002227). UBERON:0002227 is an anatomical location from the Uberon multi-species anatomy ontology.
Downstream
Basolateral Potassium Exposure of Cochlear Hair Cells
Show evidence (4 references)
PMID:19696885 SUPPORT In Vitro
"In an epithelial cell line, heterologous expression of wild-type claudin-9 reduced the paracellular permeability to Na+ and K+, and the nmf329 mutation eliminated this ion barrier function without affecting the plasma membrane localization of claudin-9."
The direct measurement that claudin-9 is an ion barrier and that a deafness-causing substitution abolishes the barrier specifically, not the protein's delivery.
PMID:19696885 SUPPORT Model Organism
"In the tight-junction complexes of hair cells, claudin-9 is localized specifically to a subdomain that is underneath more apical tight-junction strands formed by other claudins."
Locates the lesion to a subapical strand, which is why the leak reaches the basolateral compartment rather than simply opening the apical seal.
PMID:34265170 SUPPORT Model Organism
"Staining with the antibody revealed that CLDN9 was present in all sensory and non-sensory epithelial cells facing the endolymph within the scala media, where the sensory epithelium of the inner ear is located"
Establishes where claudin-9 is expressed, in mouse. This is the anatomical premise of the barrier argument, and it has been shown in mouse rather than in human tissue.
+ 1 more reference
Basolateral Potassium Exposure of Cochlear Hair Cells
Mechanism confidence: Provisional
Potassium leaks from the endolymph through the defective subapical junction into the perilymphatic space around the basolateral membranes of the hair cells. In the Cldn9 mutant mouse the perilymphatic potassium concentration is measurably raised while the endolymphatic concentration and the endocochlear potential are both normal, so the disturbance is a redistribution of potassium across a leaky junction rather than a failure of strial potassium secretion or of the endocochlear battery. That this potassium is what damages the hair cells is established by two rescues rather than asserted. Explanted organ of Corti from mutant mice keeps all three rows of outer hair cells when cultured in low potassium, and in vivo deletion of pou3f4 - which removes the endocochlear potential and so the driving force pushing potassium through the leak - also prevents the loss.
potassium ion homeostasis GO:0055075 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves dysregulated potassium ion homeostasis (GO:0055075). GO:0055075 is a biological process from the Gene Ontology. ↕ DYSREGULATED
cochlea UBERON:0001844 Uberon multi-species anatomy ontology (UBERON) Relation: this pathophysiological event occurs in this anatomical location This pathophysiological event occurs in cochlea (UBERON:0001844). UBERON:0001844 is an anatomical location from the Uberon multi-species anatomy ontology.
Animal models
nmf329 Cldn9 F35L mouse Mouse recapitulates The only system in which the ionic disturbance has been measured at all. Endolymphatic and perilymphatic potassium and the endocochlear potential were measured directly by double-barrelled microelectrode in vivo, and the causal role of potassium was then tested by two independent rescues.
Downstream
Basal-Predominant Outer Hair Cell Degeneration
Show evidence (5 references)
PMID:19696885 SUPPORT Model Organism
"In the nmf329 mouse line, the perilymphatic K+ concentration was found to be elevated, suggesting that the cochlear tight junctions were dysfunctional."
The in vivo measurement of the leak, in the compartment the leak drains into.
PMID:19696885 SUPPORT Model Organism
"The perilymphatic K + concentration was higher in the nmf329/nmf329 cochleas, whereas the endolymphatic K + concentration was similar in cochleas of both genotypes"
The paired measurement that makes the claim specific - only the perilymphatic side moves, so this is not a global collapse of cochlear ionic compartmentalisation.
PMID:19696885 SUPPORT Model Organism
"These results show that the claudin-9 defect in the nmf329 line does not abolish the endocochlear K + and voltage gradients."
The explicit negative result. Recorded because the intuitive reading of a tight-junction deafness gene is that the endocochlear potential collapses, and here it does not.
+ 2 more references
Basal-Predominant Outer Hair Cell Degeneration
Mechanism confidence: Provisional
Hair cells are lost from the second week of life in the mutant mouse, beginning at the cochlear base and extending apically with age. At P28 the basal-turn organ of Corti is collapsed and lacks paracellular spaces and one of the three outer hair cell rows, while the apical turn is still intact; by P80 only a few basal outer hair cells remain, so the rapid early degeneration slows rather than continuing at the same rate. Human cochlear histology is not available, so the basal-to-apical gradient is an inference from the mouse - though it is the natural explanation for the human audiogram, in which the high-frequency, basally encoded region is affected and the low-frequency apex is spared.
cochlear outer hair cell CL:0000601 Cell Ontology (CL) Relation: this pathophysiological event involves this cell type This pathophysiological event involves cochlear outer hair cell (CL:0000601). CL:0000601 is a cell type from the Cell Ontology.
cell death of cochlear outer hair cells GO:0008219 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves increased cell death of cochlear outer hair cells, annotated with cell death (GO:0008219). GO:0008219 is a biological process from the Gene Ontology. ↑ INCREASED
spiral organ of cochlea UBERON:0002227 Uberon multi-species anatomy ontology (UBERON) Relation: this pathophysiological event occurs in this anatomical location This pathophysiological event occurs in spiral organ of cochlea (UBERON:0002227). UBERON:0002227 is an anatomical location from the Uberon multi-species anatomy ontology.
Animal models
nmf329 Cldn9 F35L mouse Mouse recapitulates Reproduces the cellular lesion and its base-to-apex gradient with histology that human material cannot supply.
Cldn9 tet-OFF conditional knockdown mouse Mouse fails to recapitulate Substantial reduction of claudin-9 protein does not produce the outer hair cell loss or the deafness seen with the barrier-abolishing nmf329 allele.
Downstream
Loss of Cochlear Transduction in the High-Frequency Base
Show evidence (4 references)
PMID:19696885 SUPPORT Model Organism
"We found a widespread loss of sensory hair cells in the hearing organs of nmf329 mice after the second week of life."
The primary structural lesion and its postnatal timing.
PMID:19696885 SUPPORT Model Organism
"our data indicate that claudin-9 is required for the preservation of sensory cells in the hearing organ because claudin-9-defective tight junctions fail to shield the basolateral side of hair cells from the K+-rich endolymph"
The authors' statement of the causal chain this node sits at the end of: barrier failure, basolateral potassium exposure, hair cell loss.
PMID:31175426 SUPPORT Model Organism
"In the nmf329/nmf329 cochlea at P28, the organ of Corti was collapsed at the basal turn, lacking paracellular spaces and one of the three rows of outer hair cells (OHCs)."
The base-versus-apex gradient described by the human curators of the founding family, who invoked it to explain their patients' sloping audiograms.
+ 1 more reference
Loss of Cochlear Transduction in the High-Frequency Base
Loss of outer hair cells removes cochlear amplification where those cells have died. In affected humans this presents as a sensorineural loss that spares the low frequencies and declines steeply above 1 kHz - normal thresholds at 500 Hz with a steep drop after 1000 Hz in the two Turkish sisters, and thresholds worse at high than at low frequencies in both of the later families. This node is deliberately not conformed to the module's "Progressive Sensorineural Hearing Loss" consequence: progression is inconsistent across the reported families, so a node asserting it would overstate the human data.
sensory perception of sound GO:0007605 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves decreased sensory perception of sound (GO:0007605). GO:0007605 is a biological process from the Gene Ontology. ↓ DECREASED
Downstream
Bilateral Sensorineural Hearing Impairment
High-Frequency Sensorineural Hearing Impairment
Show evidence (2 references)
PMID:31175426 SUPPORT Human Clinical
"Hearing thresholds show a normal hearing level at 500 Hz and a steep decline after 1000 Hz in both sisters."
The characteristic human audiogram shape, measured in two affected sisters.
PMID:34265170 SUPPORT Human Clinical
"The degree of hearing loss in both individuals was worse at the higher frequencies as compared to the low frequencies"
The same high-frequency-predominant configuration in an independent family.
⬡

Pathograph

Use the checkboxes to hide or show graph categories. Hover nodes for evidence and cross-linked metadata.
Pathograph: causal mechanism network for Hearing Loss Autosomal Recessive 116 Interactive directed graph showing how pathophysiology mechanisms, phenotypes, genetic factors and variants, experimental models, environmental triggers, and treatments relate through causal and linked edges.
●

Phenotypes

6
Ear 5
Bilateral Sensorineural Hearing Impairment OBLIGATE HP:0008619 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Bilateral sensorineural hearing impairment (HP:0008619). HP:0008619 is a phenotype from the Human Phenotype Ontology.
Sequelae: Tinnitus
Show evidence (3 references)
PMID:31175426 SUPPORT Human Clinical
"Three affected members of the family had sensorineural hearing loss (SNHL) ranging from moderate to profound in severity."
Severity range in the founding family.
PMID:34265170 SUPPORT Human Clinical
"Audiological examinations revealed an asymmetric mild to profound hearing loss with childhood or adolescent onset."
Extends the severity range down to mild and records the asymmetry, which is the feature that separates the later families from the founding one.
PMID:34265170 SUPPORT Human Clinical
"Audiometric testing in family HLRBS10 revealed asymmetric moderate to profound and moderate to severe bilateral hearing loss in the affected individuals IV:2 and IV:3, respectively"
Per-individual audiometric findings in the Pakistani family.
High-Frequency Sensorineural Hearing Impairment VERY_FREQUENT HP:0001757 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is High-frequency sensorineural hearing impairment (HP:0001757). HP:0001757 is a phenotype from the Human Phenotype Ontology.
Show evidence (2 references)
PMID:31175426 SUPPORT Human Clinical
"We, therefore, conclude that CLDN9 is essential for the hair cells in the base of cochlea from early on and that c.86delT variant results in steeply sloping high frequency moderate to profound SNHL."
The authors' characterisation of the audiometric phenotype in the founding family.
PMID:34265170 SUPPORT Human Clinical
"Similar to the report of young affected individuals with a frameshift variant of CLDN9 (Sineni et al., 2019), mild to moderate degrees of hearing loss at lower frequencies were observed in affected individuals of family HLRBS10 and F7285."
Replication of the low-frequency-sparing configuration across all three families, stated as such by the later authors.
Progressive Sensorineural Hearing Impairment FREQUENT HP:0000408 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Progressive sensorineural hearing impairment (HP:0000408). HP:0000408 is a phenotype from the Human Phenotype Ontology.
Show evidence (3 references)
PMID:34265170 SUPPORT Human Clinical
"The hearing loss was mild to begin with for the proband (IV:4; age 35 years), however, it gradually progressed over a period of 16 years to moderate."
Documented longitudinal progression in one individual.
PMID:34265170 REFUTE Human Clinical
"However, progression of hearing loss was not observed over a period of 10 years for the affected members in family HLRBS10."
Ten years of stable thresholds in two affected individuals, which contradicts progression as a general feature of DFNB116 and is why this phenotype is not graded VERY_FREQUENT.
PMID:31175426 SUPPORT INDIRECT Human Clinical
"In the 46-year-old mother, both 500 Hz and 1000 Hz hearing levels show significant HL, suggesting that HL has progressed in the mother."
Marked INDIRECT because the inference is cross-sectional - the mother's audiogram is compared with her daughters' rather than with her own earlier one. The same paper notes the sisters' thresholds were unchanged three years apart.
Adolescent to Adult Recognition of Hearing Loss FREQUENT Postlingual sensorineural hearing impairment HP:0008596 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Postlingual sensorineural hearing impairment (HP:0008596). HP:0008596 is a phenotype from the Human Phenotype Ontology.
Show evidence (3 references)
PMID:31175426 SUPPORT Human Clinical
"All three affected members of the family were diagnosed after age 10 years, while age of onset was not clearly delineated."
Late diagnosis, and the explicit statement that onset was not established - the reason this phenotype is about recognition rather than about onset.
PMID:34265170 SUPPORT Human Clinical
"Her hearing loss was first noticed at the age of 17 years and it worsened progressively."
Adolescent recognition in the Moroccan proband.
PMID:34265170 REFUTE Human Clinical
"The parents indicated that their children had a prelingual hearing loss."
Prelingual onset in the Pakistani family, which contradicts postlingual recognition as a constant feature and shows the onset range spans the whole of childhood.
Tinnitus OCCASIONAL HP:0000360 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Tinnitus (HP:0000360). HP:0000360 is a phenotype from the Human Phenotype Ontology.
Show evidence (2 references)
PMID:34265170 SUPPORT Human Clinical
"She also reported the presence of tinnitus."
The one affected individual reporting tinnitus.
PMID:34265170 REFUTE Human Clinical
"Tinnitus was not reported by these individuals."
Absent in the two affected Pakistani siblings despite being asked about, which is what keeps this graded OCCASIONAL.
Other 1
Normal Vestibular Function VERY_FREQUENT
No HP term is bound. HP codes abnormal phenotypes, and there is no appropriate term for a normal vestibular examination; binding an abnormality term with a negating description would be worse than leaving it unbound.
Show evidence (3 references)
PMID:31175426 SUPPORT Human Clinical
"Gross motor development was normal with no history of balance problems, vertigo, dizziness, or nystagmus."
Clinical vestibular assessment in the founding family.
PMID:34265170 SUPPORT Human Clinical
"Romberg and tandem gait tests in both individuals revealed a grossly normal balance function."
Formal bedside vestibular testing in an independent family.
PMID:19696885 SUPPORT INDIRECT Model Organism
"Equally good balance was demonstrated by wild-type, nmf329/+, and nmf329/nmf329 mice"
Marked INDIRECT because it is a rotarod result in mice, cited in support of a human clinical negative. The gene is Cldn9 and the finding is concordant with the human examinations above.
🧬

Genetic Associations

1
CLDN9
Gene: CLDN9 hgnc:2051 HUGO Gene Nomenclature Committee (hgnc) Relation: this disease-associated gene is this gene This disease-associated gene is CLDN9 (hgnc:2051). hgnc:2051 is a gene from the HUGO Gene Nomenclature Committee. relationship_type: CAUSATIVE variant_origin: GERMLINE
Show evidence (6 references)
PMID:31175426 SUPPORT Human Clinical
"Through whole-genome sequencing, we identified a one base pair deletion (c.86delT) in CLDN9 in a consanguineous family from Turkey with autosomal recessive nonsyndromic hearing loss."
The founding gene-disease observation for DFNB116.
PMID:31175426 SUPPORT Human Clinical
"It is absent in public databases as well as in over 1000 Turkish individuals, and co-segregates with SNHL in the family."
Population-frequency evidence for pathogenicity, checked against a large population-matched control set rather than gnomAD alone.
PMID:34265170 SUPPORT Human Clinical
"Exome sequencing of probands identified a homozygous c.475G>A;p.(Glu159Lys) variant of CLDN9 (NM_020982.4) in one family and a homozygous c.370_372dupATC;p.(Ile124dup) CLDN9 variant in an affected individual of a second family."
The two independently ascertained families that replicated the gene-disease relationship.
+ 3 more references
💊

Medical Actions

3
Hearing Amplification
Action: hearing aid fitting and auditory rehabilitationNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is hearing aid fitting and auditory rehabilitation, annotated with Rehabilitation (NCIT:C15315), qualified as medical device hearing aid. NCIT:C15315 is a clinical intervention from the NCI Thesaurus. Ontology label: Rehabilitation NCIT:C15315
Platform: Device
Hearing aids are the first-line management for a mild-to-severe sloping sensorineural loss, fitted to the audiogram. No reported DFNB116 patient's hearing aid use is described in the primary literature, and no disease-modifying therapy exists.
Cochlear Implantation
Action: cochlear device implantationNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is cochlear device implantation, annotated with Surgical Procedure (NCIT:C15329), qualified as medical device cochlear implant. NCIT:C15329 is a clinical intervention from the NCI Thesaurus. Ontology label: Surgical Procedure NCIT:C15329
Platform: Device
Standard of care for the profound end of the severity range in autosomal recessive nonsyndromic hearing loss generally. The 46-year-old mother in the Turkish family was profoundly deaf and so would fall within usual candidacy criteria, but no implantation is reported in any DFNB116 patient, and there is nothing in the mechanism to predict an outcome different from other cochlear causes of deafness.
Genetic Counselling
Action: genetic counselingNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is genetic counseling (NCIT:C15240). NCIT:C15240 is a clinical intervention from the NCI Thesaurus. Ontology label: Genetic Counseling NCIT:C15240
Platform: Other
Recurrence risk is 25% per pregnancy for the parents of an affected child. Counselling is material in the populations where DFNB116 has been found, all three families being consanguineous or from populations where consanguineous union is common, and carriers are entirely asymptomatic so there is no clinical way to identify them.
Show evidence (1 reference)
PMID:34265170 SUPPORT Human Clinical
"Physical examinations and medical histories in both families did not reveal other clinically relevant phenotypes and carriers of the variants were asymptomatic"
Carriers are clinically silent, which is what makes molecular carrier testing rather than examination the only route to identifying them.
🔬

Diagnosis

2
Audiometry and Vestibular Examination
Pure-tone audiometry establishes the sensorineural loss and its configuration; the steeply sloping shape with preserved low frequencies is what makes CLDN9 worth considering. Balance is assessed clinically - Romberg and tandem gait - and has been normal in every reported patient. Temporal bone CT in the founding proband showed no inner ear malformation, so imaging serves to exclude structural causes rather than to identify this one.
pure-tone audiometry NCIT:C38036 NCI Thesaurus (NCIT)
Markers: Bilateral sensorineural loss with normal thresholds at 500 Hz and a steep decline above 1 kHz; normal Romberg and tandem gait; no inner ear anomaly on temporal bone CT.
Show evidence (2 references)
PMID:31175426 SUPPORT Human Clinical
"The diagnosis of sensorineural HL (SNHL) was established following standard audiometric testing in a soundproofed room in accordance with current clinical standards."
How the auditory diagnosis was made in the founding family.
PMID:31175426 SUPPORT Human Clinical
"High resolution computed tomography scan of the temporal bone did not show inner ear anomalies in the proband."
Imaging is normal, so it excludes structural causes rather than contributing a positive finding.
Genome or Exome Sequencing after Exclusion of Common Deafness Genes
DFNB116 is not clinically distinguishable from other causes of sloping sensorineural loss, so it is found by sequencing. All three families were solved that way - whole-genome sequencing in the Turkish family after every known deafness gene had been excluded, and exome sequencing of the probands in the other two. Homozygosity mapping is what makes it tractable: the Turkish variant sat in the longest of six autozygous runs. Both later families were connected through GeneMatcher, which is worth noting for a gene this rare - a single family is not enough to publish a novel gene-disease relationship.
whole genome sequencing NCIT:C101294 NCI Thesaurus (NCIT)
Markers: Biallelic CLDN9 variants in a run of homozygosity, after exclusion of GJB2 and the other established nonsyndromic deafness genes.
Show evidence (2 references)
PMID:31175426 SUPPORT Human Clinical
"After filtering and excluding variants in all known deafness genes, only one variant remained mapping to an autozygous region: CLDN9 NM_020982.3:c.86delT (p.Leu29ArgfsTer4)."
The diagnostic route in the founding family, and the reason autozygosity mapping is the tool that made it work.
PMID:34265170 SUPPORT Human Clinical
"A research collaboration between SN and CP was established using GeneMatcher (Sobreira et al., 2015)."
Records how the two replicating families were brought together, which is the practical mechanism by which an ultra-rare gene accumulates enough families to be believed.
📊

Prevalence

1
Worldwide, published families
Cases In Literature Ultra Rare
Three families with functional support - one Turkish, one Pakistani, one Moroccan - comprising seven affected individuals, plus one further individual with a proposed p.(Cys25Trp) allele that the second study judged insufficiently evidenced. No incidence or carrier-frequency estimate exists for any population. Every statement in this entry about severity, onset and progression rests on those three families, and the frequency gradings on the phenotypes should be read as descriptions of that handful of people rather than as population estimates.
Show evidence (2 references)
PMID:34265170 SUPPORT Human Clinical
"Together with the Cldn9 deaf mouse and a homozygous frameshift of CLDN9 previously associated with deafness, the two bi-allelic variants of CLDN9 described here point to CLDN9 as a bona fide human deafness gene."
Establishes the size of the reported evidence base at the time of the most recent report: two families here plus the one previously published.
PMID:34265170 SUPPORT Human Clinical
"Our work suggests that CLDN9 represents one of a few genes, variants of which should be considered for recessively inherited mild to moderate hearing loss in humans."
The ascertainment argument for why the count is likely an underestimate - milder recessive hearing loss has been systematically under-sequenced.
🐁

Animal Models

2
nmf329 Cldn9 F35L mouse
An ENU-mutagenesis mouse recovered on a deafness screen and mapped by positional cloning to a missense substitution in Cldn9. It is the system in which the entire mechanism of this entry was established, and it preceded the human gene discovery by a decade. The allele is instructive in its own right: F35L abolishes claudin-9's paracellular ion barrier function in transfected epithelial monolayers while leaving the protein correctly inserted in the plasma membrane, so the mouse isolates loss of barrier function from loss of the protein - which is not what two of the three human alleles do, since those mistraffic.
Species
Mouse
Genotype
Cldn9 nmf329 (p.Phe35Leu) homozygous, ENU-induced
Genes
CLDN9 hgnc:2051 HUGO Gene Nomenclature Committee (hgnc) Relation: this experimental model concerns this gene This experimental model concerns CLDN9 (hgnc:2051). hgnc:2051 is a gene from the HUGO Gene Nomenclature Committee.
Publication
PMID:19696885
Pathograph links
Basolateral Potassium Exposure of Cochlear Hair Cells recapitulates fidelity: high observes: tissue The only system in which the ionic disturbance has been measured at all. Endolymphatic and perilymphatic potassium and the endocochlear potential were measured directly by double-barrelled microelectrode in vivo, and the causal role of potassium was then tested by two independent rescues.
Limitations: The measurements are murine and have no human counterpart - cochlear fluids are not accessible in living patients - so the claim that this ionic disturbance occurs in affected humans is inference from a shared gene, not observation. The mouse allele is also a missense that preserves membrane targeting, whereas the human truncating and in-frame-duplication alleles fail to reach the membrane; the endpoint is likely the same but the molecular route differs.
Readouts
Perilymphatic potassium concentration
Direction: INCREASED
Interpretation: Raised perilymphatic potassium with normal endolymphatic potassium and normal endocochlear potential, localising the leak to the junction rather than the stria.
Show evidence (1 reference)
PMID:19696885 SUPPORT Model Organism
"The perilymphatic K + concentration was higher in the nmf329/nmf329 cochleas, whereas the endolymphatic K + concentration was similar in cochleas of both genotypes"
The measurement itself, with the internal control that makes it interpretable.
Endocochlear potential
Direction: UNCHANGED
Interpretation: A real negative result. The endocochlear battery is intact, so this is not a strial or endolymph-generation defect.
Show evidence (1 reference)
PMID:19696885 SUPPORT Model Organism
"The EPs were not significantly different in the two groups"
Direct in vivo EP measurement at P70-P80, showing no difference from littermate controls.
Show evidence (1 reference)
PMID:19696885 SUPPORT Model Organism
"Furthermore, the hair-cell loss in the claudin-9-defective cochlea was rescued in vitro when the explanted hearing organs were cultured in a low-K+ milieu and in vivo when the endocochlear K+-driving force was diminished by deletion of the pou3f4 gene."
Two independent rescues, one ex vivo and one genetic, which is what makes this model informative about causation rather than only about correlation.
Basal-Predominant Outer Hair Cell Degeneration recapitulates fidelity: moderate observes: tissue Reproduces the cellular lesion and its base-to-apex gradient with histology that human material cannot supply.
Limitations: Fidelity is graded MODERATE rather than HIGH because the audiometric consequence diverges: the mouse is uniformly deaf across 8, 16 and 32 kHz and severely so within days of hearing onset, whereas affected humans have normal low-frequency thresholds and a range from mild to profound with onset from prelingual to adolescence. The cellular lesion is shared; its severity and frequency distribution are not.
Readouts
Outer hair cell survival by cochlear turn
Direction: DECREASED
Interpretation: Loss of outer hair cells beginning at the base after the second postnatal week and extending apically with age.
Show evidence (1 reference)
PMID:19696885 SUPPORT Model Organism
"We found a widespread loss of sensory hair cells in the hearing organs of nmf329 mice after the second week of life."
The histological readout and its timing.
Auditory brainstem response threshold
Direction: DECREASED
Interpretation: Severe threshold elevation already present at P16, three to four days after the normal onset of hearing, and equally severe at every frequency tested.
Show evidence (1 reference)
PMID:19696885 SUPPORT Model Organism
"The hearing loss of nmf329/nmf329 mice was equally severe at high- and low-frequency, as indicated by the uniformly high ABR thresholds at 8, 16, and 32 kHz"
The functional readout, and the point at which the model departs from the human audiogram, which spares low frequencies.
Show evidence (1 reference)
PMID:19696885 SUPPORT Model Organism
"We found that the hearing loss of homozygous mutant mice was already severe at P16"
Establishes early, near-congenital severity in the mouse - the timing that the human phenotype does not match.
Show evidence (3 references)
PMID:19696885 SUPPORT Model Organism
"Positional cloning revealed that the nmf329 strain carries a missense mutation in the claudin-9 gene, which encodes a tight junction protein with unknown biological function."
Establishes that the deafness phenotype maps to Cldn9, which is what makes this a model of DFNB116.
PMID:19696885 SUPPORT Model Organism
"Thus, the ion barrier function of claudin-9 is essential in the cochlea, but appears to be dispensable in other organs."
Organ specificity in the mouse, which is the model's counterpart of the nonsyndromic presentation in humans.
PMID:19696885 SUPPORT Model Organism
"claudin-9 mutant mice exhibited no signs of vestibular, hepatic, or renal defects"
The specific negative findings behind that organ specificity, including the vestibular system where claudin-9 is expressed.
Cldn9 tet-OFF conditional knockdown mouse
A dosage model rather than a disease model, and it is included here because of what it fails to show. Doxycycline-controlled tet-OFF transgenic mice with roughly eightfold reduced cochlear claudin-9 protein are not deaf: their ABR thresholds are elevated by only about 5-15 dB, stay constant over months of monitoring, and their outer hair cell counts are indistinguishable from wild type at P14 and at fifteen months. Instead they develop an extra row of functional, mechanotransducing, long-surviving ectopic inner hair cells, apparently because claudin-9 levels participate in lateral inhibition of hair cell fate during a postnatal window that has closed by P14. The bearing on DFNB116 is that reducing how much claudin-9 there is does not reproduce what abolishing its barrier function does. That is a substantive constraint on reading the human hypomorphic alleles, and it makes claudin-9 dosage a second, developmental axis distinct from the barrier axis this entry's pathophysiology chain describes.
Species
Mouse
Genotype
Cldn9+/T tet-OFF transgenic on doxycycline, approximately 8-fold reduced claudin-9 protein
Genes
CLDN9 hgnc:2051 HUGO Gene Nomenclature Committee (hgnc) Relation: this experimental model concerns this gene This experimental model concerns CLDN9 (hgnc:2051). hgnc:2051 is a gene from the HUGO Gene Nomenclature Committee.
Publication
PMID:37873357
Pathograph links
Basal-Predominant Outer Hair Cell Degeneration fails to recapitulate fidelity: low observes: tissue Substantial reduction of claudin-9 protein does not produce the outer hair cell loss or the deafness seen with the barrier-abolishing nmf329 allele.
Limitations: This is knockdown, not a disease allele: the residual protein is presumably normal and still forms functional junctions, whereas F35L and the human variants deliver a protein that is dysfunctional or absent from the membrane. So the negative result constrains how claudin-9 quantity relates to the phenotype and says nothing directly about a mistrafficked or barrier-dead protein. The transgene is also on a CBA-CaJ background chosen to avoid age-related hearing loss, and the mild threshold elevation that is present is not mechanistically attributed.
Readouts
Outer hair cell counts by cochlear turn
Direction: UNCHANGED
Interpretation: No outer hair cell loss at any cochlear location, in contrast to the widespread basal loss in the barrier-defective nmf329 mouse.
Show evidence (1 reference)
PMID:37873357 REFUTE Model Organism
"For OHCs, the numbers along the cochlear contour, apex, middle, and base were not significantly different among the two genotypes at P14"
The negative structural readout. Graded REFUTE against the degeneration node because reduced claudin-9 does not produce it.
Auditory brainstem response threshold
Direction: UNCHANGED
Interpretation: A small, non-progressive threshold elevation instead of the roughly 60 dB deficit of the barrier-defective mouse.
Show evidence (1 reference)
PMID:37873357 REFUTE Model Organism
"with ~5-15 dB threshold elevation in the Cldn9+/T mice"
The functional readout. Graded REFUTE because a mild stable shift is not the hearing loss this disease entry describes.
Show evidence (1 reference)
PMID:37873357 REFUTE Model Organism
"The pattern of hearing threshold remained virtually constant from 2-8 months of monitoring"
Stability over months, against the progressive degeneration that defines the node this link points at.
Show evidence (2 references)
PMID:37873357 SUPPORT Model Organism
"the pharmacogenetic downregulation of Cldn9, a tight junction protein, generates robust supernumerary inner HCs (IHCs) in mice"
What the model does show - a developmental effect on hair cell number, not degeneration.
PMID:37873357 SUPPORT Model Organism
"Immunoelectron microscopic analysis showed that Cldn9 levels reduced by ~8-fold in the Cldn9+/T cochlea"
Quantifies the knockdown, which is what makes the absence of degeneration interpretable rather than a failure of the manipulation.
{ }

Source YAML

click to show 
name: Hearing Loss Autosomal Recessive 116
category: Mendelian
creation_date: "2026-09-07T00:00:00Z"
synonyms:
- DFNB116
- deafness, autosomal recessive 116
- CLDN9-related hearing loss
- autosomal recessive nonsyndromic deafness 116
description: >-
  DFNB116 is autosomal recessive nonsyndromic sensorineural hearing loss caused by biallelic
  variants in CLDN9, which encodes claudin-9, a constituent of the bicellular tight junctions
  of the epithelia that line the scala media. Three families are reported with functional
  support - a Turkish family with the truncating c.86delT (p.Leu29ArgfsTer4), and Pakistani
  and Moroccan families with the hypomorphic p.(Glu159Lys) and p.(Ile124dup) alleles - plus
  one further individual whose p.(Cys25Trp) allele is proposed rather than established.

  The mechanism this entry curates is a paracellular ion barrier failure, and it is worth
  being precise about which compartment fails, because "tight junction gene, therefore
  endolymph barrier" is the intuitive story and it is not quite what was measured. In the
  Cldn9 nmf329 mouse the endocochlear potential and the endolymphatic potassium
  concentration are both normal - the stria vascularis works - while the *perilymphatic*
  potassium concentration is raised. Claudin-9 sits in a subapical subdomain of the hair
  cell junctional complex, underneath the more apical strands built by other claudins, and
  what its loss does is let potassium leak from the endolymph past those outer strands to
  the basolateral face of the hair cell. Hair cells then die, from the cochlear base
  upward. That the potassium is what kills them is not inferred: outer hair cell loss is
  prevented by culturing the explanted organ of Corti in low potassium, and in vivo by
  deleting pou3f4 to collapse the endocochlear driving force.

  None of that has been measured in a human cochlea. The human evidence is segregation in
  consanguineous families plus heterologous-cell and structural-modelling work on the
  variant proteins; the barrier mechanism reaches humans by inference from the mouse, and
  the pathophysiology nodes downstream of the protein defect are marked PROVISIONAL for
  that reason. The mouse also diverges from the human phenotype in a way that matters: it
  is uniformly deaf across frequencies, whereas the human audiogram characteristically
  spares the low frequencies and slopes steeply above 1 kHz.

  Clinically, DFNB116 is a mild-to-profound, often asymmetric loss whose reported onset
  ranges from prelingual to adolescence, and which is progressive in some families and
  stable over a decade in others. That milder end is the ascertainment point the second
  report makes explicitly - recessive deafness genetics has concentrated on profound
  congenital cases, and CLDN9 is a gene to consider when the loss is moderate.
disease_term:
  preferred_term: hearing loss, autosomal recessive 116
  term:
    id: MONDO:0033670
    label: hearing loss, autosomal recessive 116
mappings:
  mondo_mappings:
  - term:
      id: MONDO:0033670
      label: hearing loss, autosomal recessive 116
    mapping_predicate: skos:exactMatch
    mapping_source: MONDO
classifications:
  harrisons_chapter:
  - classification_value: DISORDER_OF_EAR
  - classification_value: GENETICS_ENVIRONMENT_DISEASE
parents:
- Autosomal Recessive Nonsyndromic Hearing Loss
inheritance:
- name: Autosomal recessive
  description: >-
    Biallelic CLDN9 variants, homozygous in all three functionally supported families, each
    of which is consanguineous or from a population where consanguinity is common.
    Heterozygous carriers are unaffected: the father in the Turkish family had a normal
    audiogram, and carriers in the Pakistani and Moroccan families were asymptomatic. The
    same recessive pattern holds in the mouse, where heterozygotes are audiometrically
    indistinguishable from wild type.
  inheritance_term:
    preferred_term: Autosomal recessive inheritance
    term:
      id: HP:0000007
      label: Autosomal recessive inheritance
  evidence:
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Sanger sequencing of all four family members showed co-segregation of the variant with the phenotype as an autosomal recessive trait in the family"
    explanation: Recessive segregation of the truncating allele in the founding Turkish family.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Physical examinations and medical histories in both families did not reveal other clinically relevant phenotypes and carriers of the variants were asymptomatic"
    explanation: >-
      Carriers unaffected in both subsequently reported families, and no extra-auditory
      features - the two claims that make this recessive and nonsyndromic.
genetic:
- name: CLDN9
  notes: >-
    CLDN9 (16p13.3) is a single-coding-exon gene encoding claudin-9, a tetraspan integral
    membrane protein of bicellular tight junctions with two extracellular loops. EL1 carries
    the charged residues that set paracellular ionic selectivity and mediates trans-interaction
    with the claudin on the opposing membrane; EL2 mediates cis-interaction between claudins
    within the same membrane. Three disease alleles have functional support: the truncating
    c.86delT (p.Leu29ArgfsTer4), which falls at the start of EL1 and is a probable functional
    null - notable because a premature stop in a single-exon gene is unlikely to trigger
    nonsense-mediated decay, so the truncated protein may persist without functioning; and two
    hypomorphs, p.(Glu159Lys), which is predicted to disturb one of the two cis-interactions
    while leaving membrane targeting intact, and p.(Ile124dup), which locally misfolds the
    protein and fails to reach the membrane.

    Do not read across from the other claudins. CLDN14 (hgnc:2035) causes DFNB29 and is a
    separate nonsyndromic deafness entity; CLDN1 causes NISCH syndrome and CLDN19
    hypomagnesemia 5. CLDN9 is also similar in sequence to CLDN6, which matters practically:
    the 2021 study had to show its Cldn9 in situ probe was not detecting Cldn6.
  relationship_type: CAUSATIVE
  variant_origin: GERMLINE
  gene_term:
    preferred_term: CLDN9
    term:
      id: hgnc:2051
      label: CLDN9
  evidence:
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Through whole-genome sequencing, we identified a one base pair deletion (c.86delT) in CLDN9 in a consanguineous family from Turkey with autosomal recessive nonsyndromic hearing loss."
    explanation: The founding gene-disease observation for DFNB116.
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "It is absent in public databases as well as in over 1000 Turkish individuals, and co-segregates with SNHL in the family."
    explanation: >-
      Population-frequency evidence for pathogenicity, checked against a large
      population-matched control set rather than gnomAD alone.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Exome sequencing of probands identified a homozygous c.475G>A;p.(Glu159Lys) variant of CLDN9 (NM_020982.4) in one family and a homozygous c.370_372dupATC;p.(Ile124dup) CLDN9 variant in an affected individual of a second family."
    explanation: The two independently ascertained families that replicated the gene-disease relationship.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Together with the Cldn9 deaf mouse and a homozygous frameshift of CLDN9 previously associated with deafness, the two bi-allelic variants of CLDN9 described here point to CLDN9 as a bona fide human deafness gene."
    explanation: >-
      The authors' own summary of the strength of the gene-disease relationship after
      replication - three families plus the mouse.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "here we report families with two likely pathogenic CLDN9 variants both of which appear to be hypomorphic."
    explanation: >-
      Establishes that the second and third alleles retain partial function, which is the
      basis for treating them as a distinct molecular node from the truncating allele.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: "nonsyndromic deafness DFNB29 (CLDN14, MIM# 605608)"
    explanation: >-
      Recorded to keep the paralogue boundary explicit: the claudin deafness gene most
      likely to be confused with CLDN9 is CLDN14, and it is a different disease entity.
      Evidence source is OTHER because the statement is a background review sentence.
pathophysiology:
- name: CLDN9 Truncating Loss of Function
  description: >-
    The founding allele, homozygous c.86delT (p.Leu29ArgfsTer4), truncates claudin-9 at the
    start of the first extracellular loop. Because CLDN9 has a single coding exon, a
    premature stop is unlikely to trigger nonsense-mediated decay, so the truncated protein
    may be made and simply fail to work. GFP-tagged mutant protein expressed in HEK293 cells
    stays in the cytosol instead of reaching the plasma membrane, so no functional claudin-9
    is delivered to the junction.
  biological_scale: MOLECULAR
  genes:
  - preferred_term: CLDN9
    term:
      id: hgnc:2051
      label: CLDN9
  genetic_context:
    zygosity: HOMOZYGOUS
    variant_origin: GERMLINE
    allele_type: single-base deletion causing a frameshift and premature termination
    functional_impact_category: LOSS_OF_FUNCTION
    description: >-
      CLDN9 NM_020982.3:c.86delT, p.Leu29ArgfsTer4, homozygous in three affected members of
      a consanguineous Turkish family.
  downstream:
  - target: Failure of the Claudin-9 Paracellular Ion Barrier
    causal_link_type: DIRECT
  evidence:
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The variant is predicted to cause a frameshift and produce a truncated protein (p.Leu29ArgfsTer4) in this single-exon gene."
    explanation: The predicted molecular consequence, and the single-exon architecture that bears on whether the transcript escapes decay.
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "This suggests that the variant disrupts the migration of CLDN9 to the plasma membrane."
    explanation: >-
      The measured cellular consequence in transfected HEK293 cells - the mutant protein
      does not reach the membrane where a claudin has to sit to do anything.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "It is not known if in the inner ear truncated CLDN9 protein remained stable or was degraded."
    explanation: >-
      Records the limit of what is known. The transfected-cell result does not establish
      what the truncated protein does in a human cochlea, and the later authors say so.
- name: CLDN9 Hypomorphic Missense and In-Frame Alleles
  description: >-
    The two subsequently reported alleles retain partial function and fail in different ways.
    p.(Glu159Lys) reaches the membrane normally in MDCK-II cells; structural modelling
    predicts it perturbs one of the two cis-interactions by which adjacent claudin protomers
    oligomerise within the same membrane, so the protein is present but the strand it should
    build is compromised. p.(Ile124dup), an in-frame single-residue duplication, is predicted
    to misfold locally and does not reach the HeLa cell membrane. The authors classify both
    as hypomorphic.

    Whether that residual function explains the milder end of the clinical spectrum is not
    settled by the data: the family carrying p.(Glu159Lys) had moderate-to-profound loss that
    was stable over ten years, while the family carrying the truncating allele included a
    profoundly deaf 46-year-old. Severity and progression do not sort cleanly by allele class
    across three families.
  biological_scale: MOLECULAR
  mechanism_confidence: PROVISIONAL
  genes:
  - preferred_term: CLDN9
    term:
      id: hgnc:2051
      label: CLDN9
  genetic_context:
    zygosity: HOMOZYGOUS
    variant_origin: GERMLINE
    allele_type: missense and in-frame single-codon duplication
    functional_impact_category: PARTIAL_LOSS_OF_FUNCTION
    description: >-
      CLDN9 NM_020982.4:c.475G>A p.(Glu159Lys), homozygous in a Pakistani family, and
      c.370_372dupATC p.(Ile124dup), homozygous in a Moroccan family.
  downstream:
  - target: Failure of the Claudin-9 Paracellular Ion Barrier
    causal_link_type: DIRECT
  evidence:
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: COMPUTATIONAL
    snippet: "Computational structural modeling predicts that substitution of a lysine for glutamic acid p.(Glu159Lys) alters one of two cis-interactions between CLDN9 protomers."
    explanation: >-
      The proposed molecular lesion for the missense allele. Evidence source is COMPUTATIONAL
      because the claim rests on a structural model, not on a measurement.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "The p.(Ile124dup) variant is predicted to locally misfold CLDN9 and mCherry tagged p.(Ile124dup) CLDN9 is not targeted to the HeLa cell membrane."
    explanation: >-
      Mistrafficking of the in-frame duplication measured in transfected cells, the same
      failure mode as the truncating allele.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "The localization of CLDN9 with the substitution of lysine at position 159 was similar to that of the wild-type, when expressed in MDCK-II cells."
    explanation: >-
      Distinguishes the two hypomorphs: p.(Glu159Lys) traffics normally, so its defect has to
      be in what the protein does once it is in the membrane rather than in getting there.
- name: Failure of the Claudin-9 Paracellular Ion Barrier
  description: >-
    Claudin-9 is present in the bicellular tight junctions of every sensory and non-sensory
    epithelial cell facing the endolymph of the scala media, and within the hair cell
    junctional complex it occupies a specific subapical subdomain, sitting beneath the more
    apical strands formed by other claudins. Heterologously expressed wild-type claudin-9
    lowers paracellular permeability to sodium and potassium; the deafness-causing mouse
    F35L substitution abolishes that barrier function while leaving the protein correctly
    located in the plasma membrane, which separates the barrier defect from a trafficking
    defect.

    What this node is not: a loss of the epithelial seal that maintains the endocochlear
    potential. The stria vascularis is unaffected in the mutant mouse. The claim is
    specifically that a deeper strand of the junction leaks.
  biological_scale: CELLULAR
  mechanism_confidence: PROVISIONAL
  conforms_to: "sensorineural_hair_cell_loss#Cochlear Sensory Epithelium Insult"
  cell_types:
  - preferred_term: cochlear outer hair cell
    term:
      id: CL:0000601
      label: cochlear outer hair cell
  - preferred_term: cochlear inner hair cell
    term:
      id: CL:0000589
      label: cochlear inner hair cell
  cellular_components:
  - preferred_term: bicellular tight junction
    term:
      id: GO:0005923
      label: bicellular tight junction
  biological_processes:
  - preferred_term: paracellular transport of potassium and sodium
    modifier: INCREASED
    term:
      id: GO:0160184
      label: paracellular transport
  locations:
  - preferred_term: spiral organ of cochlea
    term:
      id: UBERON:0002227
      label: spiral organ of cochlea
  downstream:
  - target: Basolateral Potassium Exposure of Cochlear Hair Cells
    causal_link_type: DIRECT
  evidence:
  - reference: PMID:19696885
    reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "In an epithelial cell line, heterologous expression of wild-type claudin-9 reduced the paracellular permeability to Na+ and K+, and the nmf329 mutation eliminated this ion barrier function without affecting the plasma membrane localization of claudin-9."
    explanation: >-
      The direct measurement that claudin-9 is an ion barrier and that a deafness-causing
      substitution abolishes the barrier specifically, not the protein's delivery.
  - reference: PMID:19696885
    reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "In the tight-junction complexes of hair cells, claudin-9 is localized specifically to a subdomain that is underneath more apical tight-junction strands formed by other claudins."
    explanation: >-
      Locates the lesion to a subapical strand, which is why the leak reaches the
      basolateral compartment rather than simply opening the apical seal.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Staining with the antibody revealed that CLDN9 was present in all sensory and non-sensory epithelial cells facing the endolymph within the scala media, where the sensory epithelium of the inner ear is located"
    explanation: >-
      Establishes where claudin-9 is expressed, in mouse. This is the anatomical premise of
      the barrier argument, and it has been shown in mouse rather than in human tissue.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: OTHER
    snippet: "Claudin 9 (CLDN9) is an integral membrane protein and constituent of epithelial bicellular tight junctions (TJs) that form semipermeable, paracellular barriers between inner ear perilymphatic and endolymphatic compartments."
    explanation: >-
      The general statement of what claudin-9 does. Marked INDIRECT because it describes the
      protein class and compartment boundary rather than demonstrating that this barrier
      fails in an affected human. Evidence source is OTHER because the sentence is a
      background assertion, not this study's own result.
- name: Basolateral Potassium Exposure of Cochlear Hair Cells
  description: >-
    Potassium leaks from the endolymph through the defective subapical junction into the
    perilymphatic space around the basolateral membranes of the hair cells. In the Cldn9
    mutant mouse the perilymphatic potassium concentration is measurably raised while the
    endolymphatic concentration and the endocochlear potential are both normal, so the
    disturbance is a redistribution of potassium across a leaky junction rather than a
    failure of strial potassium secretion or of the endocochlear battery.

    That this potassium is what damages the hair cells is established by two rescues rather
    than asserted. Explanted organ of Corti from mutant mice keeps all three rows of outer
    hair cells when cultured in low potassium, and in vivo deletion of pou3f4 - which
    removes the endocochlear potential and so the driving force pushing potassium through
    the leak - also prevents the loss.
  biological_scale: TISSUE
  mechanism_confidence: PROVISIONAL
  conforms_to: "sensorineural_hair_cell_loss#Cochlear Ionic Homeostasis Disruption and Oxidative Stress"
  biological_processes:
  - preferred_term: potassium ion homeostasis
    modifier: DYSREGULATED
    term:
      id: GO:0055075
      label: potassium ion homeostasis
  locations:
  - preferred_term: cochlea
    term:
      id: UBERON:0001844
      label: cochlea
  downstream:
  - target: Basal-Predominant Outer Hair Cell Degeneration
    causal_link_type: DIRECT
  evidence:
  - reference: PMID:19696885
    reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "In the nmf329 mouse line, the perilymphatic K+ concentration was found to be elevated, suggesting that the cochlear tight junctions were dysfunctional."
    explanation: The in vivo measurement of the leak, in the compartment the leak drains into.
  - reference: PMID:19696885
    reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "The perilymphatic K + concentration was higher in the nmf329/nmf329 cochleas, whereas the endolymphatic K + concentration was similar in cochleas of both genotypes"
    explanation: >-
      The paired measurement that makes the claim specific - only the perilymphatic side
      moves, so this is not a global collapse of cochlear ionic compartmentalisation.
  - reference: PMID:19696885
    reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "These results show that the claudin-9 defect in the nmf329 line does not abolish the endocochlear K + and voltage gradients."
    explanation: >-
      The explicit negative result. Recorded because the intuitive reading of a
      tight-junction deafness gene is that the endocochlear potential collapses, and here it
      does not.
  - reference: PMID:19696885
    reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "In contrast to the in vivo loss of OHCs, all three rows of OHCs survived when the organ of Corti samples from nmf329/nmf329 mice were cultured ex vivo"
    explanation: >-
      The low-potassium organ culture rescue - the intervention that turns the potassium
      association into a causal claim.
  - reference: PMID:19696885
    reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Thus, deletion of the pou3f4 gene in the nmf329 line prevented the loss of OHCs."
    explanation: >-
      The in vivo genetic rescue. Removing the endocochlear driving force, without touching
      claudin-9, prevents the hair cell loss.
- name: Basal-Predominant Outer Hair Cell Degeneration
  description: >-
    Hair cells are lost from the second week of life in the mutant mouse, beginning at the
    cochlear base and extending apically with age. At P28 the basal-turn organ of Corti is
    collapsed and lacks paracellular spaces and one of the three outer hair cell rows, while
    the apical turn is still intact; by P80 only a few basal outer hair cells remain, so the
    rapid early degeneration slows rather than continuing at the same rate. Human cochlear
    histology is not available, so the basal-to-apical gradient is an inference from the
    mouse - though it is the natural explanation for the human audiogram, in which the
    high-frequency, basally encoded region is affected and the low-frequency apex is spared.
  biological_scale: TISSUE
  mechanism_confidence: PROVISIONAL
  conforms_to: "sensorineural_hair_cell_loss#Hair Cell Mechanotransduction Failure and Death"
  cell_types:
  - preferred_term: cochlear outer hair cell
    term:
      id: CL:0000601
      label: cochlear outer hair cell
  biological_processes:
  - preferred_term: cell death of cochlear outer hair cells
    modifier: INCREASED
    term:
      id: GO:0008219
      label: cell death
  locations:
  - preferred_term: spiral organ of cochlea
    term:
      id: UBERON:0002227
      label: spiral organ of cochlea
  downstream:
  - target: Loss of Cochlear Transduction in the High-Frequency Base
    causal_link_type: DIRECT
  evidence:
  - reference: PMID:19696885
    reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "We found a widespread loss of sensory hair cells in the hearing organs of nmf329 mice after the second week of life."
    explanation: The primary structural lesion and its postnatal timing.
  - reference: PMID:19696885
    reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "our data indicate that claudin-9 is required for the preservation of sensory cells in the hearing organ because claudin-9-defective tight junctions fail to shield the basolateral side of hair cells from the K+-rich endolymph"
    explanation: >-
      The authors' statement of the causal chain this node sits at the end of: barrier
      failure, basolateral potassium exposure, hair cell loss.
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "In the nmf329/nmf329 cochlea at P28, the organ of Corti was collapsed at the basal turn, lacking paracellular spaces and one of the three rows of outer hair cells (OHCs)."
    explanation: >-
      The base-versus-apex gradient described by the human curators of the founding family,
      who invoked it to explain their patients' sloping audiograms.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "the degeneration of hair cells occurs initially in the basal turn and gradually progresses with age to the apical turn leading to profound deafness"
    explanation: Independent restatement of the basal-to-apical progression in the mouse.
- name: Loss of Cochlear Transduction in the High-Frequency Base
  description: >-
    Loss of outer hair cells removes cochlear amplification where those cells have died. In
    affected humans this presents as a sensorineural loss that spares the low frequencies and
    declines steeply above 1 kHz - normal thresholds at 500 Hz with a steep drop after 1000 Hz
    in the two Turkish sisters, and thresholds worse at high than at low frequencies in both
    of the later families. This node is deliberately not conformed to the module's
    "Progressive Sensorineural Hearing Loss" consequence: progression is inconsistent across
    the reported families, so a node asserting it would overstate the human data.
  biological_scale: ORGANISM
  biological_processes:
  - preferred_term: sensory perception of sound
    modifier: DECREASED
    term:
      id: GO:0007605
      label: sensory perception of sound
  downstream:
  - target: Bilateral Sensorineural Hearing Impairment
    causal_link_type: DIRECT
  - target: High-Frequency Sensorineural Hearing Impairment
    causal_link_type: DIRECT
  evidence:
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Hearing thresholds show a normal hearing level at 500 Hz and a steep decline after 1000 Hz in both sisters."
    explanation: The characteristic human audiogram shape, measured in two affected sisters.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The degree of hearing loss in both individuals was worse at the higher frequencies as compared to the low frequencies"
    explanation: The same high-frequency-predominant configuration in an independent family.
phenotypes:
- name: Bilateral Sensorineural Hearing Impairment
  category: Auditory
  description: >-
    Bilateral sensorineural loss in every reported affected individual, ranging from mild to
    profound across families and frequently asymmetric between ears. Severity within a family
    is not uniform: the Turkish family spanned moderate in the younger sister, severe in the
    elder, and profound in the 46-year-old mother.
  frequency: OBLIGATE
  phenotype_term:
    preferred_term: Bilateral sensorineural hearing impairment
    term:
      id: HP:0008619
      label: Bilateral sensorineural hearing impairment
  sequelae:
  - target: Tinnitus
    causal_link_type: UNKNOWN
  evidence:
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Three affected members of the family had sensorineural hearing loss (SNHL) ranging from moderate to profound in severity."
    explanation: Severity range in the founding family.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Audiological examinations revealed an asymmetric mild to profound hearing loss with childhood or adolescent onset."
    explanation: >-
      Extends the severity range down to mild and records the asymmetry, which is the feature
      that separates the later families from the founding one.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Audiometric testing in family HLRBS10 revealed asymmetric moderate to profound and moderate to severe bilateral hearing loss in the affected individuals IV:2 and IV:3, respectively"
    explanation: Per-individual audiometric findings in the Pakistani family.
- name: High-Frequency Sensorineural Hearing Impairment
  category: Auditory
  description: >-
    A steeply sloping audiogram with preserved low-frequency hearing is the characteristic
    configuration, and is the clinical signature that the basal cochlea is affected first.
  frequency: VERY_FREQUENT
  phenotype_term:
    preferred_term: High-frequency sensorineural hearing impairment
    term:
      id: HP:0001757
      label: High-frequency sensorineural hearing impairment
  evidence:
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "We, therefore, conclude that CLDN9 is essential for the hair cells in the base of cochlea from early on and that c.86delT variant results in steeply sloping high frequency moderate to profound SNHL."
    explanation: The authors' characterisation of the audiometric phenotype in the founding family.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Similar to the report of young affected individuals with a frameshift variant of CLDN9 (Sineni et al., 2019), mild to moderate degrees of hearing loss at lower frequencies were observed in affected individuals of family HLRBS10 and F7285."
    explanation: >-
      Replication of the low-frequency-sparing configuration across all three families,
      stated as such by the later authors.
- name: Progressive Sensorineural Hearing Impairment
  category: Auditory
  description: >-
    Progression is real but inconsistent, and this is the least settled part of the clinical
    picture. In the Moroccan proband the loss was mild at first and worsened over sixteen
    years; in the Turkish family cross-sectional severity rose with age across three
    individuals and the mother's low-frequency thresholds had deteriorated, which the authors
    read as progression. Against that, the two affected siblings in the Pakistani family were
    stable over ten years of follow-up. Frequency is graded FREQUENT rather than higher for
    that reason.
  frequency: FREQUENT
  phenotype_term:
    preferred_term: Progressive sensorineural hearing impairment
    term:
      id: HP:0000408
      label: Progressive sensorineural hearing impairment
  evidence:
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The hearing loss was mild to begin with for the proband (IV:4; age 35 years), however, it gradually progressed over a period of 16 years to moderate."
    explanation: Documented longitudinal progression in one individual.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: REFUTE
    evidence_source: HUMAN_CLINICAL
    snippet: "However, progression of hearing loss was not observed over a period of 10 years for the affected members in family HLRBS10."
    explanation: >-
      Ten years of stable thresholds in two affected individuals, which contradicts
      progression as a general feature of DFNB116 and is why this phenotype is not graded
      VERY_FREQUENT.
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    snippet: "In the 46-year-old mother, both 500 Hz and 1000 Hz hearing levels show significant HL, suggesting that HL has progressed in the mother."
    explanation: >-
      Marked INDIRECT because the inference is cross-sectional - the mother's audiogram is
      compared with her daughters' rather than with her own earlier one. The same paper notes
      the sisters' thresholds were unchanged three years apart.
- name: Adolescent to Adult Recognition of Hearing Loss
  category: Auditory
  description: >-
    Age at recognition is late relative to most recessive nonsyndromic deafness. All three
    affected members of the Turkish family were diagnosed after age 10 and the true onset was
    never established; the Moroccan proband first noticed her loss at 17. The exception is the
    Pakistani family, where the parents described a prelingual loss. Late recognition is
    consistent with a loss that begins in the high frequencies and spares speech-critical low
    frequencies for a time, but no reported patient has documented normal audiometry followed
    by a measured decline, so onset as distinct from recognition is genuinely unknown.
  frequency: FREQUENT
  phenotype_term:
    preferred_term: Postlingual sensorineural hearing impairment
    term:
      id: HP:0008596
      label: Postlingual sensorineural hearing impairment
  evidence:
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "All three affected members of the family were diagnosed after age 10 years, while age of onset was not clearly delineated."
    explanation: >-
      Late diagnosis, and the explicit statement that onset was not established - the reason
      this phenotype is about recognition rather than about onset.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Her hearing loss was first noticed at the age of 17 years and it worsened progressively."
    explanation: Adolescent recognition in the Moroccan proband.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: REFUTE
    evidence_source: HUMAN_CLINICAL
    snippet: "The parents indicated that their children had a prelingual hearing loss."
    explanation: >-
      Prelingual onset in the Pakistani family, which contradicts postlingual recognition as
      a constant feature and shows the onset range spans the whole of childhood.
- name: Tinnitus
  category: Auditory
  description: >-
    Reported by the Moroccan proband and specifically not reported by the two affected
    individuals in the Pakistani family or by her affected sister, so it is an occasional
    accompaniment rather than part of the core phenotype. Both studies asked about it
    explicitly, which is why the negatives are informative.
  frequency: OCCASIONAL
  phenotype_term:
    preferred_term: Tinnitus
    term:
      id: HP:0000360
      label: Tinnitus
  evidence:
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "She also reported the presence of tinnitus."
    explanation: The one affected individual reporting tinnitus.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: REFUTE
    evidence_source: HUMAN_CLINICAL
    snippet: "Tinnitus was not reported by these individuals."
    explanation: >-
      Absent in the two affected Pakistani siblings despite being asked about, which is what
      keeps this graded OCCASIONAL.
- name: Normal Vestibular Function
  category: Vestibular
  description: >-
    Balance is spared. Formal Romberg and tandem gait testing was normal in the Pakistani
    family, and the Turkish family had normal gross motor development with no vertigo,
    dizziness or nystagmus and a negative Romberg. This is a curated negative, not an
    unexamined assumption, and it matches the mouse, which has normal rotarod balance despite
    claudin-9 expression in the vestibular system.
  frequency: VERY_FREQUENT
  evidence:
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Gross motor development was normal with no history of balance problems, vertigo, dizziness, or nystagmus."
    explanation: Clinical vestibular assessment in the founding family.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Romberg and tandem gait tests in both individuals revealed a grossly normal balance function."
    explanation: Formal bedside vestibular testing in an independent family.
  - reference: PMID:19696885
    reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: MODEL_ORGANISM
    snippet: "Equally good balance was demonstrated by wild-type, nmf329/+, and nmf329/nmf329 mice"
    explanation: >-
      Marked INDIRECT because it is a rotarod result in mice, cited in support of a human
      clinical negative. The gene is Cldn9 and the finding is concordant with the human
      examinations above.
  notes: >-
    No HP term is bound. HP codes abnormal phenotypes, and there is no appropriate term for
    a normal vestibular examination; binding an abnormality term with a negating description
    would be worse than leaving it unbound.
prevalence:
- population: Worldwide, published families
  measure_type: CASES_IN_LITERATURE
  prevalence_class: ULTRA_RARE
  notes: >-
    Three families with functional support - one Turkish, one Pakistani, one Moroccan -
    comprising seven affected individuals, plus one further individual with a proposed
    p.(Cys25Trp) allele that the second study judged insufficiently evidenced. No incidence
    or carrier-frequency estimate exists for any population. Every statement in this entry
    about severity, onset and progression rests on those three families, and the frequency
    gradings on the phenotypes should be read as descriptions of that handful of people
    rather than as population estimates.
  evidence:
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Together with the Cldn9 deaf mouse and a homozygous frameshift of CLDN9 previously associated with deafness, the two bi-allelic variants of CLDN9 described here point to CLDN9 as a bona fide human deafness gene."
    explanation: >-
      Establishes the size of the reported evidence base at the time of the most recent
      report: two families here plus the one previously published.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Our work suggests that CLDN9 represents one of a few genes, variants of which should be considered for recessively inherited mild to moderate hearing loss in humans."
    explanation: >-
      The ascertainment argument for why the count is likely an underestimate - milder
      recessive hearing loss has been systematically under-sequenced.
diagnosis:
- name: Audiometry and Vestibular Examination
  description: >-
    Pure-tone audiometry establishes the sensorineural loss and its configuration; the
    steeply sloping shape with preserved low frequencies is what makes CLDN9 worth
    considering. Balance is assessed clinically - Romberg and tandem gait - and has been
    normal in every reported patient. Temporal bone CT in the founding proband showed no
    inner ear malformation, so imaging serves to exclude structural causes rather than to
    identify this one.
  diagnosis_term:
    preferred_term: pure-tone audiometry
    term:
      id: NCIT:C38036
      label: Audiometric Test
  markers: >-
    Bilateral sensorineural loss with normal thresholds at 500 Hz and a steep decline above
    1 kHz; normal Romberg and tandem gait; no inner ear anomaly on temporal bone CT.
  evidence:
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The diagnosis of sensorineural HL (SNHL) was established following standard audiometric testing in a soundproofed room in accordance with current clinical standards."
    explanation: How the auditory diagnosis was made in the founding family.
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "High resolution computed tomography scan of the temporal bone did not show inner ear anomalies in the proband."
    explanation: >-
      Imaging is normal, so it excludes structural causes rather than contributing a positive
      finding.
- name: Genome or Exome Sequencing after Exclusion of Common Deafness Genes
  description: >-
    DFNB116 is not clinically distinguishable from other causes of sloping sensorineural
    loss, so it is found by sequencing. All three families were solved that way - whole-genome
    sequencing in the Turkish family after every known deafness gene had been excluded, and
    exome sequencing of the probands in the other two. Homozygosity mapping is what makes it
    tractable: the Turkish variant sat in the longest of six autozygous runs. Both later
    families were connected through GeneMatcher, which is worth noting for a gene this rare -
    a single family is not enough to publish a novel gene-disease relationship.
  diagnosis_term:
    preferred_term: whole genome sequencing
    term:
      id: NCIT:C101294
      label: Whole Genome Sequencing
  markers: >-
    Biallelic CLDN9 variants in a run of homozygosity, after exclusion of GJB2 and the other
    established nonsyndromic deafness genes.
  evidence:
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "After filtering and excluding variants in all known deafness genes, only one variant remained mapping to an autozygous region: CLDN9 NM_020982.3:c.86delT (p.Leu29ArgfsTer4)."
    explanation: >-
      The diagnostic route in the founding family, and the reason autozygosity mapping is
      the tool that made it work.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "A research collaboration between SN and CP was established using GeneMatcher (Sobreira et al., 2015)."
    explanation: >-
      Records how the two replicating families were brought together, which is the practical
      mechanism by which an ultra-rare gene accumulates enough families to be believed.
treatments:
- name: Hearing Amplification
  description: >-
    Hearing aids are the first-line management for a mild-to-severe sloping sensorineural
    loss, fitted to the audiogram. No reported DFNB116 patient's hearing aid use is described
    in the primary literature, and no disease-modifying therapy exists.
  therapeutic_modality: DEVICE
  treatment_term:
    preferred_term: hearing aid fitting and auditory rehabilitation
    term:
      id: NCIT:C15315
      label: Rehabilitation
    qualifiers:
    - predicate:
        preferred_term: medical device
        term:
          id: NCIT:C16830
          label: Medical Device
      value:
        preferred_term: hearing aid
        term:
          id: NCIT:C183182
          label: Hearing Aid
  notes: >-
    Curated without evidence, deliberately. Neither primary report describes the audiological
    management of any affected individual, so there is no source that says amplification was
    used in DFNB116 specifically. It is curated because it is the management the audiometric findings
    imply, and it is flagged here rather than supported with a quotation from a general
    hearing-loss review, which would attribute to this disease something no source says about
    it. The binding is the clinical action NCIT:C15315 Rehabilitation, not a device term:
    NCIT has no clinical-action term for hearing aid fitting, and NCIT's hearing aid term is a device
    concept outside the NCIT:C25218 Clinical Intervention or Procedure hierarchy that
    TreatmentActionTerm requires. The device concept is therefore carried as a qualifiers
    predicate-value pair (NCIT:C16830 Medical Device / NCIT:C183182 Hearing Aid), the same
    pattern the cochlear implantation entry below uses, so the device stays queryable
    without sitting in a slot it is not reachable for.

    Speech and language therapy (NCIT:C159273) is deliberately not curated as a separate
    treatment. Its yield depends on age at onset relative to language acquisition, and this
    disease's onset is precisely what is unresolved - the entry carries a REFUTE item and an
    open question on whether the loss is prelingual or postlingual. Adding a third uncited
    treatment whose indication turns on the one variable no source settles would assert more
    than the amplification and implantation entries already do.
- name: Cochlear Implantation
  description: >-
    Standard of care for the profound end of the severity range in autosomal recessive
    nonsyndromic hearing loss generally. The 46-year-old mother in the Turkish family was
    profoundly deaf and so would fall within usual candidacy criteria, but no implantation is
    reported in any DFNB116 patient, and there is nothing in the mechanism to predict an
    outcome different from other cochlear causes of deafness.
  therapeutic_modality: DEVICE
  treatment_term:
    preferred_term: cochlear device implantation
    term:
      id: NCIT:C15329
      label: Surgical Procedure
    qualifiers:
    - predicate:
        preferred_term: medical device
        term:
          id: NCIT:C16830
          label: Medical Device
      value:
        preferred_term: cochlear implant
        term:
          id: NCIT:C157820
          label: Cochlear Implant
  notes: >-
    Curated without evidence, on the same basis as the amplification entry above - no
    reported DFNB116 patient's management is described anywhere. Curated because it is the
    standard option at the profound end of this disease's own reported severity range, and marked here so a reader does not mistake it for a documented DFNB116
    outcome. The device term cannot be the treatment_term - NCIT:C157820 Cochlear Implant is
    not reachable from NCIT:C25218 - so it is carried as a qualifier on the surgical action.
- name: Genetic Counselling
  description: >-
    Recurrence risk is 25% per pregnancy for the parents of an affected child. Counselling is
    material in the populations where DFNB116 has been found, all three families being
    consanguineous or from populations where consanguineous union is common, and carriers are
    entirely asymptomatic so there is no clinical way to identify them.
  therapeutic_modality: OTHER
  treatment_term:
    preferred_term: genetic counseling
    term:
      id: NCIT:C15240
      label: Genetic Counseling
  evidence:
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Physical examinations and medical histories in both families did not reveal other clinically relevant phenotypes and carriers of the variants were asymptomatic"
    explanation: >-
      Carriers are clinically silent, which is what makes molecular carrier testing rather
      than examination the only route to identifying them.
animal_models:
- name: nmf329 Cldn9 F35L mouse
  species: Mouse
  genotype: Cldn9 nmf329 (p.Phe35Leu) homozygous, ENU-induced
  publication: PMID:19696885
  description: >-
    An ENU-mutagenesis mouse recovered on a deafness screen and mapped by positional cloning
    to a missense substitution in Cldn9. It is the system in which the entire mechanism of
    this entry was established, and it preceded the human gene discovery by a decade. The
    allele is instructive in its own right: F35L abolishes claudin-9's paracellular ion
    barrier function in transfected epithelial monolayers while leaving the protein correctly
    inserted in the plasma membrane, so the mouse isolates loss of barrier function from loss
    of the protein - which is not what two of the three human alleles do, since those
    mistraffic.
  genes:
  - preferred_term: CLDN9
    term:
      id: hgnc:2051
      label: CLDN9
  modeled_mechanisms:
  - target: Basolateral Potassium Exposure of Cochlear Hair Cells
    relationship: RECAPITULATES
    fidelity: HIGH
    model_scale: TISSUE
    description: >-
      The only system in which the ionic disturbance has been measured at all. Endolymphatic
      and perilymphatic potassium and the endocochlear potential were measured directly by
      double-barrelled microelectrode in vivo, and the causal role of potassium was then
      tested by two independent rescues.
    limitations: >-
      The measurements are murine and have no human counterpart - cochlear fluids are not
      accessible in living patients - so the claim that this ionic disturbance occurs in
      affected humans is inference from a shared gene, not observation. The mouse allele is
      also a missense that preserves membrane targeting, whereas the human truncating and
      in-frame-duplication alleles fail to reach the membrane; the endpoint is likely the
      same but the molecular route differs.
    readouts:
    - name: Perilymphatic potassium concentration
      target: Basolateral Potassium Exposure of Cochlear Hair Cells
      direction: INCREASED
      interpretation: >-
        Raised perilymphatic potassium with normal endolymphatic potassium and normal
        endocochlear potential, localising the leak to the junction rather than the stria.
      evidence:
      - reference: PMID:19696885
        reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        snippet: "The perilymphatic K + concentration was higher in the nmf329/nmf329 cochleas, whereas the endolymphatic K + concentration was similar in cochleas of both genotypes"
        explanation: The measurement itself, with the internal control that makes it interpretable.
    - name: Endocochlear potential
      target: Basolateral Potassium Exposure of Cochlear Hair Cells
      direction: UNCHANGED
      interpretation: >-
        A real negative result. The endocochlear battery is intact, so this is not a strial
        or endolymph-generation defect.
      evidence:
      - reference: PMID:19696885
        reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        snippet: "The EPs were not significantly different in the two groups"
        explanation: Direct in vivo EP measurement at P70-P80, showing no difference from littermate controls.
    evidence:
    - reference: PMID:19696885
      reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      snippet: "Furthermore, the hair-cell loss in the claudin-9-defective cochlea was rescued in vitro when the explanted hearing organs were cultured in a low-K+ milieu and in vivo when the endocochlear K+-driving force was diminished by deletion of the pou3f4 gene."
      explanation: >-
        Two independent rescues, one ex vivo and one genetic, which is what makes this model
        informative about causation rather than only about correlation.
  - target: Basal-Predominant Outer Hair Cell Degeneration
    relationship: RECAPITULATES
    fidelity: MODERATE
    model_scale: TISSUE
    description: >-
      Reproduces the cellular lesion and its base-to-apex gradient with histology that human
      material cannot supply.
    limitations: >-
      Fidelity is graded MODERATE rather than HIGH because the audiometric consequence
      diverges: the mouse is uniformly deaf across 8, 16 and 32 kHz and severely so within
      days of hearing onset, whereas affected humans have normal low-frequency thresholds and
      a range from mild to profound with onset from prelingual to adolescence. The cellular
      lesion is shared; its severity and frequency distribution are not.
    readouts:
    - name: Outer hair cell survival by cochlear turn
      target: Basal-Predominant Outer Hair Cell Degeneration
      direction: DECREASED
      interpretation: >-
        Loss of outer hair cells beginning at the base after the second postnatal week and
        extending apically with age.
      evidence:
      - reference: PMID:19696885
        reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        snippet: "We found a widespread loss of sensory hair cells in the hearing organs of nmf329 mice after the second week of life."
        explanation: The histological readout and its timing.
    - name: Auditory brainstem response threshold
      target: Basal-Predominant Outer Hair Cell Degeneration
      direction: DECREASED
      interpretation: >-
        Severe threshold elevation already present at P16, three to four days after the
        normal onset of hearing, and equally severe at every frequency tested.
      evidence:
      - reference: PMID:19696885
        reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        snippet: "The hearing loss of nmf329/nmf329 mice was equally severe at high- and low-frequency, as indicated by the uniformly high ABR thresholds at 8, 16, and 32 kHz"
        explanation: >-
          The functional readout, and the point at which the model departs from the human
          audiogram, which spares low frequencies.
    evidence:
    - reference: PMID:19696885
      reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      snippet: "We found that the hearing loss of homozygous mutant mice was already severe at P16"
      explanation: >-
        Establishes early, near-congenital severity in the mouse - the timing that the human
        phenotype does not match.
  evidence:
  - reference: PMID:19696885
    reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Positional cloning revealed that the nmf329 strain carries a missense mutation in the claudin-9 gene, which encodes a tight junction protein with unknown biological function."
    explanation: Establishes that the deafness phenotype maps to Cldn9, which is what makes this a model of DFNB116.
  - reference: PMID:19696885
    reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Thus, the ion barrier function of claudin-9 is essential in the cochlea, but appears to be dispensable in other organs."
    explanation: >-
      Organ specificity in the mouse, which is the model's counterpart of the nonsyndromic
      presentation in humans.
  - reference: PMID:19696885
    reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "claudin-9 mutant mice exhibited no signs of vestibular, hepatic, or renal defects"
    explanation: >-
      The specific negative findings behind that organ specificity, including the vestibular
      system where claudin-9 is expressed.
- name: Cldn9 tet-OFF conditional knockdown mouse
  species: Mouse
  genotype: Cldn9+/T tet-OFF transgenic on doxycycline, approximately 8-fold reduced claudin-9 protein
  publication: PMID:37873357
  description: >-
    A dosage model rather than a disease model, and it is included here because of what it
    fails to show. Doxycycline-controlled tet-OFF transgenic mice with roughly eightfold
    reduced cochlear claudin-9 protein are not deaf: their ABR thresholds are elevated by
    only about 5-15 dB, stay constant over months of monitoring, and their outer hair cell
    counts are indistinguishable from wild type at P14 and at fifteen months. Instead they
    develop an extra row of functional, mechanotransducing, long-surviving ectopic inner
    hair cells, apparently because claudin-9 levels participate in lateral inhibition of
    hair cell fate during a postnatal window that has closed by P14.

    The bearing on DFNB116 is that reducing how much claudin-9 there is does not reproduce
    what abolishing its barrier function does. That is a substantive constraint on reading
    the human hypomorphic alleles, and it makes claudin-9 dosage a second, developmental
    axis distinct from the barrier axis this entry's pathophysiology chain describes.
  genes:
  - preferred_term: CLDN9
    term:
      id: hgnc:2051
      label: CLDN9
  modeled_mechanisms:
  - target: Basal-Predominant Outer Hair Cell Degeneration
    relationship: FAILS_TO_RECAPITULATE
    fidelity: LOW
    model_scale: TISSUE
    description: >-
      Substantial reduction of claudin-9 protein does not produce the outer hair cell loss
      or the deafness seen with the barrier-abolishing nmf329 allele.
    limitations: >-
      This is knockdown, not a disease allele: the residual protein is presumably normal
      and still forms functional junctions, whereas F35L and the human variants deliver a
      protein that is dysfunctional or absent from the membrane. So the negative result
      constrains how claudin-9 quantity relates to the phenotype and says nothing directly
      about a mistrafficked or barrier-dead protein. The transgene is also on a CBA-CaJ
      background chosen to avoid age-related hearing loss, and the mild threshold elevation
      that is present is not mechanistically attributed.
    readouts:
    - name: Outer hair cell counts by cochlear turn
      target: Basal-Predominant Outer Hair Cell Degeneration
      direction: UNCHANGED
      interpretation: >-
        No outer hair cell loss at any cochlear location, in contrast to the widespread
        basal loss in the barrier-defective nmf329 mouse.
      evidence:
      - reference: PMID:37873357
        reference_title: "Genetic and pharmacologic alterations of claudin9 levels suffice to induce functional and mature inner hair cells."
        supports: REFUTE
        evidence_source: MODEL_ORGANISM
        snippet: "For OHCs, the numbers along the cochlear contour, apex, middle, and base were not significantly different among the two genotypes at P14"
        explanation: >-
          The negative structural readout. Graded REFUTE against the degeneration node
          because reduced claudin-9 does not produce it.
    - name: Auditory brainstem response threshold
      target: Basal-Predominant Outer Hair Cell Degeneration
      direction: UNCHANGED
      interpretation: >-
        A small, non-progressive threshold elevation instead of the roughly 60 dB deficit
        of the barrier-defective mouse.
      evidence:
      - reference: PMID:37873357
        reference_title: "Genetic and pharmacologic alterations of claudin9 levels suffice to induce functional and mature inner hair cells."
        supports: REFUTE
        evidence_source: MODEL_ORGANISM
        snippet: "with ~5-15 dB threshold elevation in the Cldn9+/T mice"
        explanation: >-
          The functional readout. Graded REFUTE because a mild stable shift is not the
          hearing loss this disease entry describes.
    evidence:
    - reference: PMID:37873357
      reference_title: "Genetic and pharmacologic alterations of claudin9 levels suffice to induce functional and mature inner hair cells."
      supports: REFUTE
      evidence_source: MODEL_ORGANISM
      snippet: "The pattern of hearing threshold remained virtually constant from 2-8 months of monitoring"
      explanation: >-
        Stability over months, against the progressive degeneration that defines the node
        this link points at.
  evidence:
  - reference: PMID:37873357
    reference_title: "Genetic and pharmacologic alterations of claudin9 levels suffice to induce functional and mature inner hair cells."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "the pharmacogenetic downregulation of Cldn9, a tight junction protein, generates robust supernumerary inner HCs (IHCs) in mice"
    explanation: What the model does show - a developmental effect on hair cell number, not degeneration.
  - reference: PMID:37873357
    reference_title: "Genetic and pharmacologic alterations of claudin9 levels suffice to induce functional and mature inner hair cells."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Immunoelectron microscopic analysis showed that Cldn9 levels reduced by ~8-fold in the Cldn9+/T cochlea"
    explanation: >-
      Quantifies the knockdown, which is what makes the absence of degeneration
      interpretable rather than a failure of the manipulation.
discussions:
- discussion_id: cldn9_barrier_mechanism_unmeasured_in_humans
  kind: KNOWLEDGE_GAP
  prompt: >-
    Does the cochlear paracellular potassium leak demonstrated in the Cldn9 mouse actually
    occur in humans with biallelic CLDN9 variants, and is it what deafens them?
  attaches_to:
  - pathophysiology#Failure of the Claudin-9 Paracellular Ion Barrier
  - pathophysiology#Basolateral Potassium Exposure of Cochlear Hair Cells
  rationale: >-
    The barrier mechanism is well demonstrated - in mice and in transfected epithelial
    monolayers. In humans it is entirely inferential. What has been shown in affected people
    is that a CLDN9 allele segregates with hearing loss and that the encoded protein behaves
    abnormally when overexpressed in HEK293, HeLa or MDCK cells. Nobody has measured a human
    perilymphatic potassium concentration, a human endocochlear potential, or the state of a
    human cochlear tight junction, and the tissue required is not obtainable during life.

    That gap is easy to overlook because the intermediate steps are individually plausible:
    claudin-9 is a tight junction protein, it is expressed in the cochlear epithelia, and the
    mouse phenotype matches. But the entry marks each downstream node PROVISIONAL rather than
    ESTABLISHED, because two of the three human alleles fail in a way the mouse allele does
    not - they never reach the membrane, whereas F35L sits in the membrane and simply stops
    sealing. A mistrafficked claudin could in principle disturb junction assembly more
    broadly than a barrier-dead one that is correctly incorporated.

    The tractable route is not a human measurement. It is a knock-in mouse carrying a human
    allele - which is exactly what the second report proposes - with the same electrode
    measurements repeated. If a p.(Glu159Lys) or p.(Ile124dup) knock-in shows the same
    isolated perilymphatic potassium rise with a preserved endocochlear potential, the
    inference to humans becomes much stronger; if it does not, the human alleles are doing
    something else.
  evidence:
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Identifying additional mutant alleles of CLDN9 and engineering animal models with the corresponding human variants could further our understanding of pathogenic mechanisms that expose therapeutic opportunities."
    explanation: >-
      The authors propose allele-matched animal models, which is the experiment that would
      close this gap.
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: MODEL_ORGANISM
    snippet: "Since CLDN9 is involved in paracellular permeability, this deformity is likely to impair functionality."
    explanation: >-
      Included as an example of the inferential step itself. The founding paper reasons from
      claudin-9's known role to the significance of a murine morphological finding; marked
      INDIRECT because that is a plausibility argument rather than a measurement.
- discussion_id: cldn9_mouse_human_audiometric_mismatch
  kind: HUMAN_MODEL_MISMATCH
  prompt: >-
    Why is the Cldn9 mouse uniformly and severely deaf across all frequencies from the onset
    of hearing, when humans with biallelic CLDN9 variants have normal low-frequency thresholds
    and a loss that can be as mild as moderate and may not progress at all?
  attaches_to:
  - pathophysiology#Basal-Predominant Outer Hair Cell Degeneration
  - phenotypes#High-Frequency Sensorineural Hearing Impairment
  - animal_models#nmf329 Cldn9 F35L mouse
  rationale: >-
    The mismatch is not subtle. ABR thresholds in the nmf329 mouse are uniformly elevated at
    8, 16 and 32 kHz and are already severe at P16, days after hearing normally begins. The
    human phenotype is the opposite shape - normal at 500 Hz, steeply sloping above 1 kHz -
    and spans mild to profound with recognition ranging from prelingual to age 17.

    Two explanations are separable and neither has been tested. The first is allele strength:
    the mouse F35L completely eliminates barrier function in the Ussing chamber assay, while
    two of three human alleles are explicitly described as hypomorphic, so the human cochlea
    may retain enough barrier to protect the apex. The second is that the mouse histology
    itself is not uniform - degeneration begins basally and reaches the apex only with age -
    so a uniformly flat ABR deficit in an animal with a basal-predominant lesion suggests the
    mouse cochlea is more vulnerable overall, or that the ABR at P28 is already reporting a
    later stage of the same gradient the human audiogram catches early.

    This matters for more than tidiness. If the human low-frequency sparing reflects residual
    barrier function rather than a slower version of the same total loss, then the apical
    cochlea in a DFNB116 patient contains living hair cells for decades, and the therapeutic
    question is preservation rather than restoration. A knock-in of a human hypomorphic allele
    with frequency-resolved ABR and turn-by-turn hair cell counts would separate the two
    accounts.
  proposed_experiments:
  - experiment_id: cldn9_human_allele_knockin_frequency_abr
    name: Frequency-resolved audiometry and cochlear histology in a human-allele knock-in mouse
    description: >-
      Generate mice carrying the human p.(Glu159Lys) or p.(Ile124dup) allele, measure ABR
      thresholds by frequency across the lifespan, and count hair cells turn by turn. A
      hypomorphic knock-in that reproduces apical sparing would attribute the human audiogram
      shape to residual barrier function; one that is uniformly deaf like nmf329 would locate
      the difference in species biology instead.
    would_support:
    - pathophysiology#CLDN9 Hypomorphic Missense and In-Frame Alleles
    supporting_outcome:
    - >-
      Knock-in mice carrying a human hypomorphic allele show elevated high-frequency ABR
      thresholds with preserved low-frequency thresholds, and hair cell counts that are
      reduced at the base and preserved at the apex.
    would_refute:
    - pathophysiology#CLDN9 Hypomorphic Missense and In-Frame Alleles
    refuting_outcome:
    - >-
      Knock-in mice are uniformly deaf across frequencies with pan-cochlear hair cell loss,
      indistinguishable from nmf329, indicating that allele strength does not explain the
      human audiogram shape.
  evidence:
  - reference: PMID:19696885
    reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "The hearing loss of nmf329/nmf329 mice was equally severe at high- and low-frequency, as indicated by the uniformly high ABR thresholds at 8, 16, and 32 kHz"
    explanation: The mouse side of the mismatch, measured across three frequencies.
  - reference: PMID:31175426
    reference_title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Hearing thresholds show a normal hearing level at 500 Hz and a steep decline after 1000 Hz in both sisters."
    explanation: The human side of the mismatch, in the family with the strongest (truncating) allele.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "here we report families with two likely pathogenic CLDN9 variants both of which appear to be hypomorphic."
    explanation: >-
      The residual-function hypothesis for the mismatch - the human alleles studied so far
      are not nulls.
- discussion_id: cldn9_dosage_versus_barrier_function
  kind: KNOWLEDGE_GAP
  prompt: >-
    Is DFNB116 caused by having too little claudin-9, or by having claudin-9 that cannot
    seal - and do the hypomorphic human alleles act on the same axis as the mouse
    barrier-dead allele?
  attaches_to:
  - pathophysiology#CLDN9 Hypomorphic Missense and In-Frame Alleles
  - pathophysiology#Failure of the Claudin-9 Paracellular Ion Barrier
  - animal_models#nmf329 Cldn9 F35L mouse
  - animal_models#Cldn9 tet-OFF conditional knockdown mouse
  rationale: >-
    Two mouse manipulations of the same gene give opposite results. The nmf329 allele leaves
    normal amounts of claudin-9 in the membrane but abolishes its ion barrier function, and
    those mice lose their hair cells and go deaf within days of hearing onset. Reducing
    claudin-9 protein roughly eightfold with a tet-OFF transgene leaves outer hair cell
    counts normal at fifteen months and shifts ABR thresholds by only 5-15 dB; what it does
    instead is generate an extra row of functional inner hair cells, implicating claudin-9
    levels in lateral inhibition of hair cell fate during a window that closes by P14.

    So quantity and barrier competence are separable axes, and DFNB116 has been explained
    entirely on the barrier axis. That is very likely right for the truncating allele, which
    delivers no functional protein to the junction at all. It is less obviously right for
    p.(Glu159Lys), which reaches the membrane normally and is predicted to perturb a
    cis-interaction - a lesion that could plausibly reduce effective claudin-9 in strands
    rather than open a leak, which the knockdown data suggest would be relatively well
    tolerated. Nobody has measured paracellular permeability for any human CLDN9 allele.

    Two things would resolve it: Ussing-chamber permeability measurements on monolayers
    expressing each human variant, the assay that established the mouse F35L allele as
    barrier-dead; and attention to whether hypomorphic-allele carriers have any of the
    developmental signature - supernumerary inner hair cells are not visible on an
    audiogram, but the reduced afferent synapse counts that accompanied them in the mouse
    would show up as a wave I amplitude deficit disproportionate to threshold.
  evidence:
  - reference: PMID:37873357
    reference_title: "Genetic and pharmacologic alterations of claudin9 levels suffice to induce functional and mature inner hair cells."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "The findings suggest that Cldn9 levels coordinate embryonic and postnatal HC differentiation, making it a viable target for altering IHC development pre- and post-terminal differentiation."
    explanation: >-
      The dosage axis stated by its own authors - claudin-9 level as a developmental
      variable, separate from the barrier role that explains the disease.
  - reference: PMID:37873357
    reference_title: "Genetic and pharmacologic alterations of claudin9 levels suffice to induce functional and mature inner hair cells."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "By contrast, the P14 inner ear injected with Cldn9-shRNA produced no detectable increase in ectopic IHCs as counted by three independent blinded observers"
    explanation: >-
      Bounds the developmental window to before P14, which is what makes it a separate
      phenomenon from the degeneration that begins after the second postnatal week.
  - reference: PMID:19696885
    reference_title: "A claudin-9-based ion permeability barrier is essential for hearing."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "In an epithelial cell line, heterologous expression of wild-type claudin-9 reduced the paracellular permeability to Na+ and K+, and the nmf329 mutation eliminated this ion barrier function without affecting the plasma membrane localization of claudin-9."
    explanation: >-
      The barrier axis, and the assay that would have to be run on the human alleles to
      place them on it.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "The localization of CLDN9 with the substitution of lysine at position 159 was similar to that of the wild-type, when expressed in MDCK-II cells."
    explanation: >-
      The human allele that makes the question live: it is in the membrane, so whether it
      leaks or merely thins the strand has not been distinguished.
- discussion_id: cldn9_cys25trp_allele_uncertain
  kind: KNOWLEDGE_GAP
  prompt: >-
    Is the CLDN9 c.75C>G p.(Cys25Trp) allele reported in a child with fluctuating hearing loss
    actually pathogenic, and does fluctuating hearing loss belong in the DFNB116 phenotype?
  attaches_to:
  - genetic#CLDN9
  - phenotypes#Bilateral Sensorineural Hearing Impairment
  rationale: >-
    A fourth individual has been reported with a CLDN9 missense allele at a highly conserved
    cysteine and a steeply sloping, fluctuating audiogram. It is deliberately excluded from
    this entry's phenotype and prevalence counts. The reasons are specific rather than
    generic caution: the same individual carried variants in FLNA and ANKRD11 and had
    unrelated upper airway findings, no functional work was done on the allele, and the
    subsequent authors said in print that more evidence would be needed.

    Fluctuation is the part worth resolving, because it is not merely an extra severity tier -
    it would be a different mechanism. Nothing in the potassium-leak model predicts a hearing
    threshold that recovers, since hair cell death does not reverse. A genuinely fluctuating
    CLDN9 phenotype would imply a reversible functional disturbance of the barrier rather
    than progressive cell loss, and that would change how the whole entry reads. Establishing
    it needs functional characterisation of p.(Cys25Trp) and at least one more family.
  evidence:
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Recently, a c.75C>G;p.(Cys25Trp) CLDN9 variant was suggested to cause fluctuating hearing loss with a steeply sloping audiogram in one individual"
    explanation: The report of the fourth allele and its distinctive fluctuating course.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The affected individual also had variants in FLNA and ANKRD11 and exhibited turbinate hypertrophy, allergic rhinitis and nasopharyngeal nodule."
    explanation: >-
      The confounding genotype and phenotype that make the attribution to CLDN9 uncertain in
      that individual.
  - reference: PMID:34265170
    reference_title: "Variants of human CLDN9 cause mild to profound hearing loss."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Additional experimental evidence would be useful in evaluating the predicted pathogenicity of the CLDN9 c.75C>G;p.(Cys25Trp) variant."
    explanation: The subsequent authors' own judgement that the allele is not yet established.
references:
- reference: PMID:20301607
  title: "Genetic Hearing Loss Overview."
  tags:
  - GeneReviews
- reference: PMID:19696885
  title: "A claudin-9-based ion permeability barrier is essential for hearing."
- reference: PMID:31175426
  title: "A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss."
- reference: PMID:34265170
  title: "Variants of human CLDN9 cause mild to profound hearing loss."
- reference: PMID:37873357
  title: "Genetic and pharmacologic alterations of claudin9 levels suffice to induce functional and mature inner hair cells."
notes: >-
  Curated from the three primary reports - the founding Turkish family, the two-family
  replication, and the Cldn9 mouse - plus a claude_code deep-research sweep used as a lead
  generator. Scope decisions worth recording:

  The fourth reported individual, with p.(Cys25Trp) and fluctuating hearing loss, is excluded
  from the phenotype and prevalence sections and handled as an open question instead; its
  primary report was available only as an abstract, which contains no CLDN9 sentence, so
  every statement about it here is quoted from the second study's description of it rather
  than from the source.

  Both device treatments are curated without evidence and say so in their own notes. Neither
  primary report describes the audiological management of a single affected patient, so
  citing a general hearing-loss source for them would attribute to DFNB116 something no
  source says about DFNB116. Note this is a plain statement, not the `review_notes` waiver
  sentinel, which applies only to `environmental:` entries.

  The pathophysiology chain deliberately stops short of conforming its final node to
  sensorineural_hair_cell_loss#Progressive Sensorineural Hearing Loss, because progression is
  contradicted in one of the three families.

  It also declines that module's Cochlear Amplification Loss and Spiral Ganglion Neuron
  Degeneration node, for a separate reason: no CLDN9 source reports spiral ganglion
  neurons at all, in the nmf329 mouse or in patients. That is absence of evidence rather
  than a contradicted claim, so the node is left unconformed instead of being asserted
  from the module.

  GeneReviews PMID:20301607 is carried in the top-level references: block as a
  bibliographic tag only. Its cached abstract is a statement of the chapter's purpose
  with no mention of CLDN9 or claudin, so nothing in it is quotable and no evidence item
  cites it.
📚

References & Deep Research

References

5
PMID:20301607
Genetic Hearing Loss Overview.
No top-level findings curated for this source.
PMID:19696885
A claudin-9-based ion permeability barrier is essential for hearing.
No top-level findings curated for this source.
PMID:31175426
A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss.
No top-level findings curated for this source.
PMID:34265170
Variants of human CLDN9 cause mild to profound hearing loss.
No top-level findings curated for this source.
PMID:37873357
Genetic and pharmacologic alterations of claudin9 levels suffice to induce functional and mature inner hair cells.
No top-level findings curated for this source.

Deep Research

1

Deep research results are used as seeds for research; they do not undergo the same validation as the main records and may contain errors. How we use deep research.

Evaluations and curation notes (2)

Record notes

Curated from the three primary reports - the founding Turkish family, the two-family replication, and the Cldn9 mouse - plus a claude_code deep-research sweep used as a lead generator. Scope decisions worth recording: The fourth reported individual, with p.(Cys25Trp) and fluctuating hearing loss, is excluded from the phenotype and prevalence sections and handled as an open question instead; its primary report was available only as an abstract, which contains no CLDN9 sentence, so every statement about it here is quoted from the second study's description of it rather than from the source. Both device treatments are curated without evidence and say so in their own notes. Neither primary report describes the audiological management of a single affected patient, so citing a general hearing-loss source for them would attribute to DFNB116 something no source says about DFNB116. Note this is a plain statement, not the `review_notes` waiver sentinel, which applies only to `environmental:` entries. The pathophysiology chain deliberately stops short of conforming its final node to sensorineural_hair_cell_loss#Progressive Sensorineural Hearing Loss, because progression is contradicted in one of the three families. It also declines that module's Cochlear Amplification Loss and Spiral Ganglion Neuron Degeneration node, for a separate reason: no CLDN9 source reports spiral ganglion neurons at all, in the nmf329 mouse or in patients. That is absence of evidence rather than a contradicted claim, so the node is left unconformed instead of being asserted from the module. GeneReviews PMID:20301607 is carried in the top-level references: block as a bibliographic tag only. Its cached abstract is a statement of the chapter's purpose with no mention of CLDN9 or claudin, so nothing in it is quotable and no evidence item cites it.

Create: Hearing_Loss_Autosomal_Recessive_116 · 2026-09-07T22:46:55Z · View source

De novo curation of DFNB116 (CLDN9, MONDO:0033670) from three primary human reports (PMID:31175426 Turkish family, c.86delT; PMID:34265170 Pakistani and Moroccan families, p.Glu159Lys and p.Ile124dup) plus the Cldn9 nmf329 mouse (PMID:19696885) and the Cldn9 dosage/tet-OFF study (PMID:37873357). Lump/split: entry_type DISEASE, consistent with 26 sibling numbered DFNA/DFNB entries; MONDO:0033670 is a leaf with one causal gene and no descendants. Mechanism discipline was the main curation decision. The endolymph-perilymph barrier story is demonstrated in mouse and in transfected epithelial monolayers only: the nmf329 mouse has a raised perilymphatic K+ with a NORMAL endocochlear potential and normal endolymphatic K+, and causality is established by two rescues (low-K+ organ culture, pou3f4 deletion). Nothing equivalent has been measured in a human cochlea, so the three pathophysiology nodes downstream of the protein lesion carry mechanism_confidence: PROVISIONAL, the general claudin-barrier statement is graded directness: INDIRECT, and two discussions record the gap (cldn9_barrier_mechanism_unmeasured_in_humans; cldn9_dosage_versus_barrier_function). A HUMAN_MODEL_MISMATCH discussion records that the mouse is uniformly deaf across frequencies while the human audiogram spares low frequencies. Conformance: three nodes conform to sensorineural_hair_cell_loss (Cochlear Sensory Epithelium Insult, Cochlear Ionic Homeostasis Disruption and Oxidative Stress, Hair Cell Mechanotransduction Failure and Death). The final node deliberately does NOT conform to that module's Progressive Sensorineural Hearing Loss node, because progression is contradicted in one of the three families (stable over 10 years in family HLRBS10). Excluded: the fourth reported individual with p.(Cys25Trp) and fluctuating hearing loss is kept out of the phenotype and prevalence sections and handled as a KNOWLEDGE_GAP; its primary report (PMID:33924653) is abstract-only in cache and contains no CLDN9 sentence, so every statement about it is quoted from PMID:34265170's description. Both device treatments (amplification, cochlear implantation) are curated without evidence and say so in notes, because no primary report describes any patient's audiological management. Deep research: claude_code provider, 250 s, 8/8 references resolved, 0 unresolved; preflight-dr PASS (CLDN9 x33, OMIM 619093 matched). Two factual errors in the DR report were not propagated: it gives the Nakano PMID as 19696886 (correct: 19696885) and describes family HLRBS10 as a Turkish mother and two daughters (correct: Pakistani siblings). None of its suggested ontology CURIEs were used - its Term Validation section flagged UBERON:0000407 called Cochlea (actually sympathetic trunk), UBERON:0004551 called Organ of Corti (obsolete plantar arteries), and CL:0002067 called type I outer hair cell (type A enteroendocrine cell). Validation: just validate, validate-terms, count-verified-snippets (78/78), check-duplicate-keys, check-entity-refs, check-causal-targets, check-qualifier-terms, check-qualifier-terms-online, check-enum-values, check-folded-hyphens, check-snippet-length, check-title-snippets, check-snippet-grading, check-environmental-evidence, check-term-cache-integrity, normalize-cache all pass; just validate-disorders passes (78/78 snippets).

Claude Code ▸
1. Disease Information
claude-haiku-4-5-20251001, claude-sonnet-5 9 citations 2026-09-07T22:32:40.730279

1. Disease Information

Overview. DFNB116 (Deafness, Autosomal Recessive 116) is an ultra-rare form of nonsyndromic sensorineural hearing loss (SNHL) caused by biallelic (homozygous) pathogenic variants in CLDN9, which encodes the tight-junction protein claudin-9. It was first described in 2019 in a consanguineous Turkish family and further characterized in 2021 in two additional consanguineous families (Turkish and Moroccan). The condition is characterized by childhood- or adolescent-onset, asymmetric, slowly progressive moderate-to-profound SNHL, with a steeply sloping (high-frequency-predominant) audiometric configuration in younger patients that broadens to affect all frequencies with age (Sineni et al., 2019, PMID:31175426; Ramzan et al., 2021, PMID:34265170).

Key identifiers:

Resource Identifier
OMIM phenotype #619093 — DEAFNESS, AUTOSOMAL RECESSIVE 116; DFNB116
OMIM gene *615799 — CLAUDIN 9; CLDN9
Gene symbol / HGNC CLDN9 / HGNC:2051
MONDO MONDO:0033670
Locus 16p13.3
Inheritance Autosomal recessive

Synonyms: Deafness, autosomal recessive 116; DFNB116; CLDN9-related nonsyndromic hearing loss; autosomal recessive nonsyndromic hearing loss due to CLDN9 mutation. There is no distinct ICD-10/ICD-11 code beyond the general "hereditary sensorineural hearing loss, bilateral" (ICD-10 H90.3 / ICD-11 code for genetic hearing loss) — DFNB loci are not individually coded in ICD.

Evidence base. All currently published knowledge derives from aggregated case reports of a small number of consanguineous families (a Turkish family reported independently by two groups totaling ~3 affected individuals, and one Moroccan proband) plus one model-organism (mouse) mechanistic study — there is no large EHR-derived or population-cohort dataset. This is one of the rarest and most recently molecularly defined DFNB loci.

2. Etiology

Disease causal factor: Purely monogenic/genetic. DFNB116 is caused by homozygous (or compound heterozygous) loss-of-function or missense variants in CLDN9 (16p13.3), a single-exon gene encoding claudin-9, a bicellular tight-junction protein of the inner ear epithelium.

Genetic risk factors: - Biallelic CLDN9 variants are both necessary and sufficient; heterozygous carriers (e.g., unaffected parents) are asymptomatic, consistent with strict autosomal recessive inheritance. - Consanguinity is a major risk amplifier: all reported families are consanguineous (Turkish and Moroccan pedigrees), and the disease-causing alleles were found in the homozygous state via autozygosity/exome sequencing (Sineni 2019; Ramzan 2021). - No genetic modifier loci have been described for CLDN9-related deafness to date; the extreme rarity of the condition (only a handful of published probands) precludes genotype–phenotype-modifier studies.

Environmental risk factors: None specifically documented for DFNB116. As with other progressive SNHL disorders, generic noise exposure or ototoxic drug exposure could theoretically accelerate residual hair-cell loss, but this has not been studied for CLDN9 patients specifically.

Protective factors: None reported (genetic or environmental). No protective CLDN9 alleles are described.

Gene–environment interaction: Not studied. Given the proposed mechanism (see §6) — loss of a paracellular potassium barrier in the reticular lamina leading to chronic low-grade K⁺ toxicity to hair cells — it is biologically plausible that superimposed noise or ototoxic insult could compound damage, but this remains speculative and unstudied.

3. Phenotypes

Primary phenotype — Sensorineural hearing loss (laboratory/clinical sign, audiometric): - Onset: Childhood to adolescence (not congenital/prelingual in all reported cases — several patients had documented progression from milder loss). - Laterality/symmetry: Notably asymmetric between ears in some patients (Ramzan et al. describe "asymmetric mild to profound hearing loss"). - Severity: Highly variable across and even within families — ranging from moderate to profound. In the Turkish mother-and-two-daughters family: the mother had bilateral symmetric profound SNHL, the proband had moderate SNHL, and her affected older sister had severe SNHL. - Audiometric configuration: In younger/less severely affected patients, a steeply sloping high-frequency loss (near-normal thresholds at 500 Hz with a steep decline above 1000 Hz) is typical; this progresses over time toward a flatter, more severe pattern across all frequencies, consistent with age-related worsening in the mother of the index family. - Progression: Slowly progressive — the disorder is explicitly described as evolving from a high-frequency-predominant loss in youth to more severe/pantonal loss in adulthood. - Frequency among affected individuals: 100% of biallelic CLDN9 carriers reported to date manifest hearing loss (fully penetrant recessive trait), though severity/age of onset is variable (variable expressivity).

No syndromic features. DFNB116 is strictly nonsyndromic — no vestibular, renal, ophthalmologic, cardiac, or other organ-system phenotypes have been reported in any published case, distinguishing it from claudin-related syndromic disease.

Quality of life impact: Not formally studied (no EQ-5D/SF-36 data specific to DFNB116), but as a progressive bilateral SNHL beginning in childhood/adolescence, expected impacts mirror other progressive nonsyndromic hearing loss: speech/language development risk if onset is early and unaided, educational and psychosocial effects, and increasing reliance on amplification/rehabilitation with age.

Suggested HPO terms (to be verified against current HPO release before curation): - HP:0000365 — Hearing impairment - HP:0000407 — Sensorineural hearing impairment - HP:0000408 — Progressive sensorineural hearing impairment - Bilateral vs. asymmetric SNHL descriptors (specific frequency-graded HPO terms, e.g., "Mild/Moderate/Severe/Profound sensorineural hearing impairment," HP:0008625/HP:0008619 family) — exact IDs should be confirmed via HPO browser at curation time.

4. Genetic/Molecular Information

Causal gene: CLDN9 (claudin 9), OMIM 615799, HGNC:2051, chromosome 16p13.3. CLDN9 is a single-exon (intronless) gene*, typical of the claudin family.

Reported pathogenic variants:

Variant (cDNA) Protein Type Family/Origin Zygosity Source
c.86delT p.(Leu29ArgfsTer4) Frameshift/truncating Turkish consanguineous family (3 affected) Homozygous Sineni et al. 2019, PMID:31175426
c.475G>A p.(Glu159Lys) Missense Turkish mother + 2 daughters Homozygous Ramzan et al. 2021, PMID:34265170
c.370_372dupATC p.(Ile124dup) In-frame duplication Moroccan proband (35-year-old woman) Homozygous Ramzan et al. 2021, PMID:34265170

ACMG classification: All three variants were classified as pathogenic/likely pathogenic for autosomal recessive nonsyndromic hearing loss in their respective publications, based on: absence from population databases (gnomAD, and >1,000 in-house Turkish control chromosomes for c.86delT), segregation with disease in the family, homozygosity in affected/heterozygosity in unaffected relatives, and (for the frameshift variant) predicted truncation of the single-exon protein.

Population frequency: All three variants are described as absent or vanishingly rare in population databases (gnomAD); DFNB116 has not been assigned a population carrier frequency, consistent with an extremely rare, largely private/founder-type allele profile in the reported consanguineous pedigrees.

Functional consequence: Claudin-9 is an integral tetraspan membrane protein of bicellular tight junctions. The truncating c.86delT variant is predicted to produce a severely truncated, non-functional protein (loss of function). The missense (p.Glu159Lys) and in-frame duplication (p.Ile124dup) variants are hypothesized to disrupt claudin-9's paracellular ion-barrier function without necessarily abolishing protein expression — consistent with the mouse nmf329 model, where a missense change (F35L) in the first extracellular loop "eliminated the ion-barrier function" of claudin-9 while the protein remained properly localized to the tight junction (Nakano et al., 2009, PLOS Genetics, PMC2720454).

Modifier genes: None identified; too few cases to assess.

Epigenetic/chromosomal data: No epigenetic or copy-number/structural mechanisms have been reported for DFNB116; all known cases are point/small indel variants.

Somatic vs. germline: Germline only (Mendelian inheritance) — not applicable to somatic/cancer mechanisms (note: CLDN9 overexpression has been separately implicated as a driver in gastric cancer progression via glycolysis/PD-L1 pathways [PMID:40458308], but this is an unrelated somatic oncology context, not part of the DFNB116 germline disease mechanism).

5. Environmental Information

No environmental, toxic, occupational, dietary, or infectious contributors have been described for DFNB116 in the literature — it is a purely monogenic disorder. There is no known infectious trigger. (As an aside with no established relevance to hearing-loss pathogenesis, claudin-9 is separately known to act as a co-receptor/entry cofactor for hepatitis C virus in hepatocytes — an unrelated tissue-specific role of the same protein, not implicated in the ear phenotype.)

6. Mechanism / Pathophysiology

Causal chain (inferred primarily from the mouse claudin-9 model, extrapolated to human disease):

  1. Biallelic loss-of-function or ion-barrier-disrupting missense/indel variant in CLDN9 → loss or dysfunction of claudin-9 protein in cochlear epithelial bicellular tight junctions [demonstrated in human probands and directly in the mouse F35L model].
  2. Claudin-9 normally forms a subdomain of the bicellular tight-junction strand at the reticular lamina — the apical junctional network sealing the basolateral compartment of hair cells and supporting cells away from the K⁺-rich endolymph — localizing specifically beneath more apical strands formed by other claudins (Nakano et al. 2009, PMC2720454). Loss of claudin-9 leads to a breach in this paracellular ion-permeability barrier.
  3. This breach results in abnormal paracellular leakage of K⁺ (and Na⁺) across the reticular lamina, specifically elevating the K⁺ concentration in the perilymph surrounding the outer hair cells — this is inferred/demonstrated in the mouse model (measured directly) and inferred by analogy in human disease.
  4. Critically, the endocochlear potential (EP) and endolymphatic K⁺ concentration remain normal in the mouse model — distinguishing this mechanism from EP-collapse mechanisms seen in other forms of hereditary hearing loss (e.g., some connexin/GJB2-related or stria vascularis disorders). The defect is localized to the reticular-lamina barrier, not the endolymph-generating machinery.
  5. Chronic exposure of hair cells (initially outer hair cells) to elevated perilymphatic K⁺ causes a toxic ionic microenvironment, which leads to progressive hair-cell degeneration beginning in the late postnatal/early auditory-maturation period (after the second postnatal week in mice).
  6. Experimentally, hair-cell loss in claudin-9–deficient mouse cochleae was rescued when the endocochlear K⁺-driving force was pharmacologically/experimentally diminished, directly demonstrating (not merely inferring) that K⁺ toxicity — rather than a structural or developmental hair-cell defect — is the proximate cause of degeneration (Nakano et al. 2009, PMC2720454).
  7. Progressive loss of outer (and eventually additional) hair cells culminates in the clinical phenotype: slowly progressive, often asymmetric, sensorineural hearing loss with a high-frequency-predominant onset (reflecting the basal-to-apical vulnerability gradient typical of cochlear hair-cell pathology) that broadens with age as hair-cell loss extends — this final step (mouse-to-human correlation) is inferred by mechanistic analogy, not directly demonstrated in human temporal-bone or biopsy tissue, since no human cochlear pathology specimens from DFNB116 patients have been reported.

Molecular pathway/process involvement: - Tight-junction assembly and paracellular barrier formation (claudin-based bicellular tight junction; GO:0120193 "bicellular tight junction assembly," GO:0016338 "calcium-independent cell-cell adhesion via plasma membrane cell-adhesion molecules") - Ion homeostasis in the inner ear (paracellular K⁺/Na⁺ transport regulation) - Hair-cell survival/degeneration signaling (downstream of ionic stress) — specific apoptotic pathway not yet characterized for CLDN9-related hair-cell death.

Cell types involved: - Cochlear outer hair cells (primary site of initial degeneration) — CL:0002067 (type I outer hair cell) or general "auditory hair cell" CL:0000201/CL:0002267 (verify exact CL ID at curation) - Inner hair cells (later/more severe involvement) — CL:0002261 or similar - Supporting cells and other epithelial cell types lining the endolymphatic space, where claudin-9 is broadly expressed (Deiters' cells, Hensen cells, etc.)

Interesting mechanistic corollary — hair-cell fate regulation: A more recent study (2023) found that pharmacologic/genetic downregulation of Cldn9 (rather than loss of ion-barrier function per se) in the neonatal mouse cochlea induces formation of supernumerary, functional, long-surviving inner hair cells, implicating claudin-9–dependent junctional signaling in lateral inhibition of hair-cell fate during a critical postnatal developmental window (P2–P7) (Kelley lab et al., 2023, eLife/PMC10592694, PMID:37873357). This is mechanistically distinct from the degenerative K⁺-toxicity pathway above and suggests claudin-9 may play dual roles: (a) a structural ion-barrier role whose loss causes late, progressive hair-cell death (the DFNB116 disease mechanism), and (b) a developmental signaling role in hair-cell number specification (a finding with regenerative-medicine implications, not yet linked to human disease modifier effects).

Anatomical structures affected (UBERON, provisional — verify at curation): - Cochlea (UBERON:0000407) - Organ of Corti (UBERON:0004551) - Reticular lamina / cochlear epithelium - Outer and inner hair cells within the organ of Corti

No syndromic organ involvement — the disease is confined to the inner ear.

7. Anatomical Structures Affected

  • Organ level: Inner ear (cochlea) only; no other organ system is affected. Auditory system (special sense organ), not cardiovascular/renal/neurologic.
  • Tissue/cell level: Cochlear sensory epithelium (organ of Corti) — outer hair cells (earliest/most affected), inner hair cells, and supporting cells of the reticular lamina, all of which normally express claudin-9 at their apical bicellular tight junctions.
  • Subcellular level: Apical plasma membrane / bicellular tight-junction complex (GO Cellular Component: "bicellular tight junction," GO:0005923); claudin-9 localizes to a specific subdomain beneath the main tight-junction strand.
  • Localization: Bilateral, though asymmetric severity between ears has been documented in at least one family — an unusual feature for a purely genetic disorder, suggesting stochastic or environmental modifying factors on an otherwise symmetric genetic lesion.

8. Temporal Development

  • Onset: Childhood or adolescent onset in most reported cases (not universally congenital/prelingual), though severity and exact onset age vary between and within families.
  • Onset pattern: Insidious/gradual rather than acute.
  • Progression: Documented slow progression — audiometric configuration evolves from a steep high-frequency-predominant loss in younger patients to a more severe, broader-frequency loss in adulthood (illustrated by the mother vs. daughters comparison in the index Turkish family, where the older, more distantly examined individual had profound loss while the younger probands had moderate-to-severe loss).
  • Disease course pattern: Progressive, not episodic or relapsing-remitting; no spontaneous remission reported.
  • Duration: Chronic, lifelong, non-fluctuating.
  • Critical periods (from mouse model): Hair-cell degeneration in the claudin-9-deficient mouse model begins after the second postnatal week — i.e., after the onset of hearing function/EP maturation — consistent with a "vulnerability window" once the ionic gradients driving mechanotransduction are established. In the separate hair-cell-fate study, the critical developmental window for claudin-9-related supernumerary hair-cell induction was P2–P7, closing by P14.

9. Inheritance and Population

Epidemiology: No formal prevalence or incidence estimates exist. DFNB116 is one of the rarest molecularly defined nonsyndromic hearing-loss loci, with only three published families/probands worldwide to date (one Turkish family reported twice/independently confirmed, one additional Turkish family, and one Moroccan proband) — likely reflecting both true rarity and recent discovery (2019/2021) rather than a stable population estimate. It should be considered ultra-rare among the >120 known DFNB loci (in contrast to common causes such as GJB2/DFNB1, which accounts for a large fraction of prelingual recessive deafness in many populations).

Inheritance pattern: Autosomal recessive, fully penetrant in the homozygous state based on all reported pedigrees.

Penetrance: Complete (100%) in reported homozygotes, though expressivity is variable (moderate to profound severity; variable onset age).

Genetic anticipation: Not observed/not applicable (not a repeat-expansion disorder).

Germline mosaicism: Not reported.

Founder effects: Each reported allele (c.86delT, c.475G>A, c.370_372dupATC) appears distinct and family-specific; no shared founder haplotype across the Turkish and Moroccan families has been described, consistent with independent private/founder mutations arising in different consanguineous lineages rather than a single ancestral founder allele.

Consanguinity: A defining feature of every reported family — all probands arose from consanguineous unions, which is the expected ascertainment pattern for an ultra-rare autosomal recessive condition uncovered by homozygosity mapping/exome sequencing in inbred pedigrees.

Carrier frequency: Not established; presumably very low to negligible in outbred populations given absence from gnomAD.

Population demographics: All reported cases are of Turkish or Moroccan ancestry; no data exist for other ethnic groups, though this likely reflects ascertainment (research groups working with these consanguineous cohorts) rather than a true ethnic restriction.

Sex ratio: No sex predilection is apparent — reported affected individuals include both males and females (the index Turkish family reported was mother + 2 daughters, i.e., all female, but this is a single pedigree, not evidence of sex-linkage — the condition is autosomal, not X-linked).

Age distribution: Reported affected individuals range from childhood (probands) to at least their 30s–40s (Moroccan proband, age 35; mother in Turkish pedigree).

10. Diagnostics

Clinical/audiological tests: - Pure-tone audiometry — the primary diagnostic modality, showing bilateral (often asymmetric) SNHL with a steeply sloping high-frequency configuration in younger patients, progressing to broader-frequency involvement. - Standard newborn hearing screening (otoacoustic emissions [OAE] and/or automated auditory brainstem response [ABR]) would be expected to detect this condition if onset is early enough, though several reported cases had childhood/adolescent (not neonatal) identification. - No syndromic features on physical exam (renal ultrasound, ophthalmologic exam, ECG, etc. would be expected to be normal, consistent with nonsyndromic classification), though such syndromic work-up is a standard part of ruling out mimics (see differential diagnosis below).

Genetic testing: - Exome sequencing was the diagnostic method in all three published families (via homozygosity/autozygosity mapping combined with exome sequencing in consanguineous pedigrees). - Hereditary hearing loss gene panels: CLDN9 is included in some commercial/national hearing-loss NGS panels (e.g., Genomics England PanelApp "Monogenic hearing loss" panel lists CLDN9), making panel-based or exome-based testing the practical diagnostic route for a new patient today, since single-gene Sanger testing would rarely be first-line given how rare this locus is. - Because CLDN9 is a single-exon gene, both point-variant detection (exome/panel sequencing) and coverage-based deletion/duplication analysis are straightforward technically. - Chromosomal microarray, karyotype, FISH, and mitochondrial DNA testing are not relevant for this Mendelian single-gene disorder and would only be used to exclude alternative diagnoses.

Differential diagnosis: Other autosomal recessive nonsyndromic SNHL loci must be excluded, especially: - DFNB1 (GJB2/GJB6) — the most common cause of recessive nonsyndromic hearing loss worldwide. - DFNB29 (CLDN14) — another claudin-family gene causing recessive hearing loss, mechanistically related (also a tight-junction/paracellular-barrier claudin), useful as a direct comparator. - Other progressive recessive SNHL genes (e.g., SLC26A4/Pendred if goiter present, MYO15A, TMC1, OTOF, CDH23, etc.) depending on audiometric/clinical pattern. - Syndromic hearing loss (Usher syndrome, Pendred syndrome, etc.) should be excluded by absence of retinal, vestibular, or thyroid findings.

Screening: No disease-specific population screening program exists; detection occurs through standard universal newborn hearing screening (if congenital/early-onset) or through audiology referral when progressive loss is noticed in childhood/adolescence, followed by genetic confirmation.

11. Outcome/Prognosis

  • Mortality: None — DFNB116 is not associated with any increased mortality; it is a purely audiologic condition.
  • Morbidity: Progressive bilateral hearing impairment with functional impact on speech/language development (if early-onset and unaided), communication, and educational/psychosocial outcomes; severity trajectory trends toward profound loss with age based on the available (very limited) longitudinal family data.
  • Disease course: Slowly progressive rather than static; audiometric worsening over years to decades, from a high-frequency-predominant pattern to a flatter, more severe pattern.
  • Complications: None beyond the hearing loss itself and its downstream communicative/developmental impact — no reported vestibular, balance, or other complications.
  • Recovery potential: None spontaneously; amplification (hearing aids) or cochlear implantation (for severe-to-profound cases) restores functional hearing but does not reverse the underlying hair-cell loss.
  • Prognostic factors: Too few cases exist to correlate specific variant type (truncating vs. missense/in-frame duplication) with severity or progression rate; the truncating c.86delT variant and the missense/duplication variants have all been associated with moderate-to-profound loss, without a clear genotype-severity correlation established yet.

12. Treatment

There is no CLDN9-specific or gene-targeted therapy for DFNB116. Management follows standard-of-care for progressive nonsyndromic sensorineural hearing loss:

  • Amplification: Hearing aids (NCIT:C120533 or general "Hearing Aid Usage") are the mainstay for mild-to-severe loss.
  • Cochlear implantation: Indicated for patients who progress to severe-to-profound bilateral loss with insufficient benefit from hearing aids (NCIT:C15329 Surgical Procedure as the treatment_term action, with a device qualifier for the cochlear implant device itself per dismech convention).
  • Aural (re)habilitation: Speech-language therapy, auditory training, and educational support, particularly important given the childhood/adolescent onset (NCIT:C159273 Speech Therapy).
  • Genetic counseling: Recommended for affected families given the autosomal recessive inheritance and high consanguinity background, to inform recurrence risk (25% for future offspring of carrier parents) and reproductive options (NCIT:C15240 Genetic Counseling).
  • No pharmacotherapy, gene therapy, or RNA-based therapy has been developed or trialed for DFNB116 specifically. There are no registered CLDN9-targeted clinical trials on ClinicalTrials.gov as of current literature.
  • Research-stage relevance (not a clinical treatment): The 2023 finding that claudin-9 downregulation can induce supernumerary functional inner hair cells in mice (PMC10592694) is of interest for future hair-cell regeneration strategies broadly, but is not a therapeutic approach for restoring claudin-9 function in DFNB116 patients (whose disease mechanism is loss of the ion-barrier, not excess claudin-9) and should not be conflated with a treatment for this specific disorder.

13. Prevention

  • Primary prevention: Not applicable in the traditional sense (no modifiable environmental cause); the only "primary prevention" avenue is reproductive genetic counseling and carrier/prenatal testing in families with a known CLDN9 pathogenic variant, particularly relevant given the strong consanguinity association in all reported pedigrees. Preimplantation genetic testing (PGT-M) would be technically feasible once a familial variant is known.
  • Secondary prevention: Early identification via universal newborn hearing screening (OAE/ABR) followed by prompt audiological and genetic diagnosis to enable early intervention (hearing aids, early language exposure) and mitigate developmental impact — standard practice for any pediatric SNHL, not DFNB116-specific.
  • Tertiary prevention: Ongoing audiological monitoring given the progressive nature of the disease, to trigger timely escalation from hearing aids to cochlear implantation as loss worsens.
  • Immunization: Not applicable (non-infectious etiology).
  • Public health/environmental interventions: Not applicable.

14. Other Species / Natural Disease

  • Taxonomy: No naturally occurring CLDN9-related hearing loss has been reported in domestic animals, livestock, or wildlife (no OMIA entry identified for spontaneous claudin-9 deafness).
  • Orthologous gene: Mouse Cldn9 (chromosome 17), NCBI Gene — direct ortholog of human CLDN9, extensively studied in the ENU-mutagenesis mouse model below.
  • Comparative biology: The tight-junction/paracellular-barrier function of claudin-9 in the reticular lamina appears evolutionarily conserved between mouse and human, based on concordant mechanistic and phenotypic findings (progressive hair-cell loss, preserved endocochlear potential) between the mouse model and the limited human clinical data.
  • Zoonotic potential: Not applicable (not an infectious disease).

15. Model Organisms

Primary model — mouse nmf329 strain (Jackson Laboratory Neuroscience Mutagenesis Facility line): - Type: Induced (ENU/N-ethyl-N-nitrosourea chemical mutagenesis), germline point mutation, homozygous recessive mouse model. - Genotype: Missense mutation F35L in the first extracellular loop of claudin-9 (Cldn9), NCBI Taxon:10090 (Mus musculus). - Phenotype recapitulation: Recessive deafness with progressive outer-hair-cell loss beginning after the second postnatal week — closely mirrors the human progressive, hair-cell-degeneration-driven SNHL phenotype. Loss of ion-barrier function (elevated perilymphatic K⁺) with preserved endocochlear potential was directly demonstrated, and hair-cell loss was experimentally rescued by reducing the K⁺-driving force, providing strong mechanistic (not just phenotypic) validation of the model (Nakano et al., 2009, PLOS Genetics 5(8):e1000610, PMID:19696886 (inferred from PMC ID PMC2720454), full text at PMC2720454). - Limitations: The mouse missense allele (F35L) differs from all three reported human alleles (a frameshift and two distinct missense/in-frame variants at different residues), so it is a model of the general claudin-9 ion-barrier-loss mechanism rather than an exact genocopy of any specific human DFNB116 allele. Whether the human truncating variant (c.86delT) behaves identically (complete loss of protein vs. a stable dysfunctional protein as in F35L) has not been directly tested in vivo. - Research applications: This model has been used to establish the K⁺-toxicity mechanism of hair-cell degeneration and, more recently, to explore claudin-9's separate role in hair-cell-fate specification via shRNA/pharmacologic knockdown, revealing a developmental (lateral-inhibition-like) function distinct from its structural ion-barrier role (2023 study, PMC10592694, PMID:37873357).

No other model systems (zebrafish, Drosophila, C. elegans, iPSC-derived organoid, or cell-line models) specific to claudin-9-related hearing loss were identified in this search; the mouse nmf329 line and related Cldn9 knockdown/knockout constructs are the sole reported models.

Comparative note: The paralogous claudin gene CLDN14 (DFNB29) — also expressed in cochlear tight junctions — provides a useful comparative model/disease pair, since both claudins were independently shown to matter for hearing via distinct but related paracellular-barrier mechanisms, though CLDN14 knockout mice show a different pattern (endocochlear potential collapse in some contexts) — a nuance to weigh in any cross-claudin mechanistic comparison during KB curation.

Summary of Key Evidence Citations

Citation Content
Sineni et al., 2019, PMID:31175426, Hum Genet First description of DFNB116; c.86delT frameshift in Turkish consanguineous family
Ramzan et al., 2021, PMID:34265170, PMC8435009, Hum Mutat 42(10):1321-1335 Two additional families; c.475G>A (p.Glu159Lys) and c.370_372dupATC (p.Ile124dup)
Nakano et al., 2009, PMC2720454, PLOS Genetics Mouse nmf329 model; mechanistic basis (K⁺-toxicity, ion-barrier loss, EP preserved)
2023 study, PMID:37873357, PMC10592694, eLife (reviewed preprint) Claudin-9 downregulation induces supernumerary functional inner hair cells; developmental role
OMIM #619093 Clinical synopsis, phenotype/gene relationship
OMIM *615799 CLDN9 gene entry

Note on evidence completeness: This is a very sparsely studied disease (fewer than 10 published affected individuals across 3 families as of this search), so several standard knowledge-base categories (formal prevalence, natural disease in other species, treatment trial data, quality-of-life instruments, biomarkers, imaging findings, histopathology) have no published data — this should be recorded as an absence of evidence, not curated as a negative finding.

Sources: - 619093 - DEAFNESS, AUTOSOMAL RECESSIVE 116 - OMIM - Entry - *615799 - CLAUDIN 9; CLDN9 - OMIM - A truncating CLDN9 variant is associated with autosomal recessive nonsyndromic hearing loss - PubMed (PMID:31175426) - Variants of human CLDN9 cause mild to profound hearing loss - PMC (PMID:34265170) - A Claudin-9–Based Ion Permeability Barrier Is Essential for Hearing - PMC - Genetic and pharmacologic alterations of claudin9 levels suffice to induce functional and mature inner hair cells - PMC (PMID:37873357) - CLDN9 - Wikipedia - CLDN9 gene - GenCC - Gene: CLDN9 (Monogenic hearing loss) - Genomics England PanelApp

Reference Validation

Checked with linkml-reference-validator 0.2.1.

Outcome Count
References checked 8
Resolved 8
Unresolved (possible confabulation) 0
Unverifiable 0
References weighed for topical relevance 8
On topic 4
Off topic 0

All extracted references resolved successfully.

Term Validation

Checked with linkml-term-validator 0.4.5, through the ols: adapter.

Outcome Count
Terms checked 21
Resolved 17
Unresolved (possible confabulation) 0
Obsolete 2
Unverifiable 2
Terms whose name was checked 8
Terms named correctly 3
Terms named as a different term 4
Terms whose name is worth a second look 1

Terms the report names something else

These identifiers resolve, so nothing about them looks wrong, and the ontology calls them something unrelated to what the report calls them. That usually means the identifier is not the one the sentence needs:

  • MONDO:0033670 (1 mention) - the report calls it "MONDO"; MONDO calls it hearing loss, autosomal recessive 116
  • CL:0002267 (1 mention) - the report calls it "verify exact CL ID at curation"; CL calls it type D cell of stomach
  • UBERON:0000407 (1 mention) - the report calls it "Cochlea"; UBERON calls it sympathetic trunk
  • UBERON:0004551 (1 mention) - the report calls it "Organ of Corti"; UBERON calls it obsolete set of proper plantar digital arteries

Obsolete terms

These terms are real but deprecated. Citing one is not a fabrication; it does mean the report is naming something the ontology has retired:

  • CL:0000201 (CL_0000201) (1 mention) - replaced by CL:0000202
  • UBERON:0004551 (obsolete set of proper plantar digital arteries) (1 mention)

Terms whose name is worth a second look

The report's name for these is recognisably related to the term's own name without being one of them. A loose paraphrase reads the same way as a citation of the wrong sibling term - and so does a related synonym, which the ontology records precisely because it names something adjacent rather than the same thing - so these are listed rather than judged:

  • CL:0002067 (1 mention) - the report calls it "type I outer hair cell"; CL calls it type A enteroendocrine cell

Prefixes with no resolver

Terms carrying these prefixes were not checked either way, because no configured ontology covers them. An unrecognised prefix may name an ontology this run could not reach as easily as one that does not exist, so nothing here is evidence of fabrication: Taxon.