Fig 1.
Identification of a missense variant in the candidate DIAPH2 gene.
(A) Pedigree of an Italian family with nonsyndromic hearing loss, showing the segregation of the DIAPH2 variant (NM_006729.4:c.868A>G) within the family. The genotypes of available individuals are indicated below the corresponding symbol and the proband is indicated by a black arrow. W: wild-type allele (A), M: mutant allele (G). ○: obligate carrier, ●: carrier of the variant confirmed by Sanger sequencing. (B) Pure-tone air-conduction thresholds. For each subject the threshold average for the right and left ear and the age at audiometric evaluation is shown. The two affected siblings III1 and III3 are indicated by empty circles and squares, respectively. Their brother (III2, black circles), mother (II2, black triangles), and father (II1, empty triangles) show normal hearing. (C) Schematic representation of DIAPH2 gene, where exons are indicated by rectangles and introns by black lines. Electropherograms showing the sequence surrounding the mutated nucleotide in the proband (right) and his mother (heterozygous carrier, left). The position of the identified variant is indicated by an arrowhead. R: A or G. Exonic nucleotides are indicated with uppercase, while intronic nucleotides are indicated with lowercase letters. (D) Schematic representation of DIAPH2 protein and amino acid sequence alignments of DIAPH2 orthologs in the region surrounding the mutant residue (p.I290). The position of the p.I290V is indicated by an arrowhead. Protein sequences were retrieved from UniProt, and alignments were generated with Clustal Omega. The affected amino acid residue is shadowed. Identical amino acids are marked by an asterisk, while partially conserved residues are indicated by a colon. GBD: GTPase-binding domain, DID: diaphanous inhibitory domain, DD: dimerization domain, FH1: formin homology 1, FH2: formin homology 2, DAD: diaphanous autoregulatory domain, N: N-terminus, C: C-terminus.
Table 1.
Prioritization of variants identified by exome sequencing.
Table 2.
Prioritization of synonymous variants based on their predicted functional impact.
Fig 2.
Diaph2 expression in E17.5 and E18.5 wild-type mouse cochlea.
Brown indicates positive staining. (A-A”) Vibratome cross section of a E17.5 wild-type mouse cochlea showing Diaph2 in green (A) and the hair-cell marker Myosin VI in red (A’). The merged image of the red and green channels shows the co-localization of Diaph2 with the marker in OHCs. (B-B”‘) Cross section of the cochlear turns (B’: apical, B, B”‘: mid, B”: basal turn) of a E18.5 wild-type mouse fetus. At this stage, a different degree of cell differentiation can be observed from the base to the apex. The square brackets indicate OHCs and the arrowheads indicate Kölliker’s organ. Dorsal to the bottom. Scale bars 10 μm.
Fig 3.
Diaph2 expression in P0 wild-type mouse cochlea.
Brown indicates positive staining. (A-D) Cross section of the cochlear turns (B: apical, A, D: mid, C: basal turn) of a P0 wild-type mouse. At this stage, a different degree of cell differentiation can be observed from the base to the apex. The square brackets indicate OHCs (outer hair cells). The arrowheads indicate: Kölliker’s organ (KO), Reissner’s membrane (RM), stria vascularis (SV). Dorsal to the bottom. (E) Higher magnification of the stria vascularis in C. (F-G) Higher magnification of the OHC shown, respectively, in C and D. Scale bar: 10 μm.
Fig 4.
Diaph2 expression in P5 wild-type mouse cochlea.
Brown indicates positive staining. (A) Scala media of the middle turn of cochlea. Scale bar 100 μm. (A’-A”) Higher magnification of the stria vascularis and of the outer sulcus shown in A. Scale bar 10 μm. (B) Scala media of the basal turn of cochlea. Scale bar 100 μm. (B’-B”) Higher magnification of the stria vascularis and of the organ of Corti shown in B. Scale bar 10 μm. Arrowheads point to: organ of Corti (OC), root cells (RC), Reissner’s membrane (RM), stria vascularis (SV).
Fig 5.
Evaluation of DIAPH2 localization and of the length of membrane protrusions in HEK293 under activating stimuli.
DIAPH2 localization and cell morphology was evaluated in HEK293 cells 48 hours after transfection with plasmids coding for HA-tagged isoforms of wild-type or mutant DIAPH2 (DIAPH2-HA WT, DIAPH2-HA MUT), together with a Myc-tagged active form of RhoA (Myc-RhoA). (A) Single middle confocal sections for each channel and the merged image of the blue, red and green channels are shown. Image of colocalized pixels (in white, as generated by Fiji colocalization plugin) superimposed or not on a RG-merge are shown. (B) Single confocal sections of cells at the level of their adhesion to the glass are shown. The merged image of the blue, red and green channels is also shown. On the right, histograms representing the mean length of protrusions measured using Fiji in 20 wt and 20 mut HEK293 cells, with error bars showing the standard deviations. Significance level of unpaired t-test is shown (**: p<0.01). Images were acquired with Leica True Confocal Scanner (TCS) SP8 and are representative of 3 experiments. Scale bar: 10 μm.
Fig 6.
Analysis of auditory function in Diaph2 mutant mice.
(A) Mean ABR thresholds of mice carrying the knock-in variant (Diaph2em3Kcl) and (B) the knock-in variant plus a 19bp deletion (Diaph2em2Kcl) at 4 (left), 8 (center) and 14 (right) weeks of age. Male hemizygotes (empty maroon circles) and female homozygotes (filled red circles) have thresholds comparable to female heterozygotes (filled blue diamonds) and male and female wildtypes (squares, female filled green, male empty gold) at all ages. At 14 weeks old the Diaph2em3Kcl female wildtypes have raised high-frequency thresholds compared to the other genotypes, but since there are only three of them and the seven male wildtypes have normal hearing, we suggest that this is not a relevant biological difference. Numbers of each genotype tested are shown on each plot. Error bars show standard deviation.
Fig 7.
ABR waveform analysis in Diaph2 mutant mice.
ABR waveforms in response to a click stimulus shown at 30dB above threshold (sensation level, SL) at fourteen weeks old. The top panels show individual waveforms, the central panels show the mean waveform, and the bottom panels show the mean ± standard deviation. There is no obvious difference between Diaph2em2Kcl hemizygote and homozygote mice (red, n = 15) and wildtype mice (black, n = 10) (left), or between Diaph2em3Kcl hemizygous and homozygous mice (red, n = 13) and wildtype mice (black, n = 10) (right).