Fig 1.
In situ hybridization of Nrp1 and Sema3a.
In situ hybridization showing the expression pattern of Nrp1 (A-C) and Sema3a (D-F) in mice at time points E13.5, E16.5, E18.5, and P1. Nrp1 was not expressed at E16.5 (A), weakly expressed in the SG and SV at E18.5 (B), and more highly expressed in the SG, greater epithelial region (GER), and SV at P1 (C). Sema3a expression pointed by arrows on the abneural side of the cochlear epithelium was observed around E13.5 (D) and continued at E16.5 (E) and E18.5 (F) on the same region of the cochlear epithelium and spiral ganglion neurons. (G, H) Negative controls for Nrp1 (E18.5) and Sema3a (E16.5) using sense probes. (I) Panel I shows the plane of sections. SG: spiral ganglion. SV: stria vascularis. Scale bar = 50 μm.
Fig 2.
Immunostaining of neuropilin-1 in the SGNs/SV and of Sema3a in the organ of Corti and SGNs in mice at postnatal stage (P5).
Mid-modiolar sections of the cochlea were immunostained with antibodies to TUJ1, neuropilin-1, and Sem3a. Blocking peptide (BP) for semaphorin-3A was used as a negative control. Neuropilin-1 receptors were expressed on SGNs and SV in WT mice at P5 (A, B, D, E) and were not expressed on SGNs (C) and SV (F) of Nrp1fl/fl;Pax2Cre mice. Semaphorin-3A expression was detected on organ of Corti (G, H), and SGNs (J, K) at postnatal day 5 (D, E). Panel I and L show semaphorin-3A immunostaining of the organ of Corti (I) and SGNs (L) in presence of blocking peptide for semaphorin-3A. Green in panel E, F, I is background color from TUJ1 staining. Scale bar = 20 μm.
Fig 3.
Presynaptic marker CtBP2 counts of the IHCs and OHCs.
Cochlear whole mounts of WT, Nrp1fl/fl;Pax2Cre mice at P5 (n = 4 per group) and 4 months (n = 5 per group) were immunostained with presynaptic marker CtBP2. Panel A and B shows CtBP2 immunostaining of the 4-month-old mice (A, B). The number of CtBP2 puncta was compared separately to the number of inner and outer hair cells per section. CtBP2 ratios of IHCs did not differ significantly between WT and Nrp1fl/fl;Pax2Cre mutants from P5 to maturity (C). OHC synaptic ribbon ratios of 4-month-old Nrp1fl/fl;Pax2Cre demonstrated a significant decrease in CtBP2 counts in the basal turn (p<0.05). CtBP2 counts of the IHCs showed two cochlea out of five with nearly 50% decreased numbers of the puncta when compared to WT controls; however, the average of the counts (five cochlea) did not show a statistically significant difference (D). *p<0.05. Scale bar = 10 μm.
Fig 4.
SGN density and caspase-3 expression in 4-month-old WT and Nrp1fl/fl;Pax2Cre mutant mice.
Mid-modiolar sections of the cochlea (basal turn) showing lower density of the SGNs in 4-month-old Nrp1fl/fl;Pax2Cre mice compared to WT controls (A, B). Cochlear SGN of 4-month-old WT and Nrp1fl/fl;Pax2Cre mice were immunostained with antibodies to Caspase-3 (Red) to ascertain the fate of the SGN at cochlear maturity (C, D). Caspase-3 positive SGNs pointed by arrows in panel D (red staining) showing apoptosis of neurons in 4-month-old Nrp1fl/fl;Pax2Cre mice compared to WT controls on panel C. (E) There was no statistically significant difference between SGNs count of the Nrp1fl/fl;Pax2Cre mice at P5 when compared to WT controls (n = 4 per group). SGNs count at the apical, middle, and basal (n = 5 per group) turns of the cochlea was consistently lower in the 4-month-old Nrp1fl/fl;Pax2Cre mutant group compared to WT controls (F). *p<0.05. Scale bar = 40 μm for A, and B. Scale bar = 10 μm for C, and D.
Fig 5.
SGN projections in the basal turns of cochleae of WT and Nrp1fl/fl;Pax2Cre mice at P5 and 4-month-old age.
Disorganized outer spiral bundles were evident in cochleae of the Nrp1fl/fl;Pax2Cre mice (P5) in the basal turn (n = 3 per group). The arrow shows disorganized outer spiral bundles (B). TUJ1 immunostaining of the SGN fibers extending into the inner and outer hair cell regions in 4-month-old Nrp1fl/fl;Pax2Cre mutants revealed aberrant axons with abnormal innervation of the OHCs. Arrow shows abnormal innervation of the OHCs (D). WT mice did not show aberrant axon migration or innervation. Scale bar = 20 μm.
Fig 6.
Aberrant innervation of OHCs in Nrp1fl/fl;Pax2Cre mice at 4 months of age.
TUJ1 immunostaining of the mid-modiolar cross-section of cochleae of 4-month-old WT (A) and Nrp1fl/fl;Pax2Cre mice reveals disorganized innervation of the OHCs in Nrp1-CKO compared to WT controls. Arrow shows abnormal innervation of the OHCs (B). Scale bar = 20 μm.
Fig 7.
Loss of sema-mediated activation of neuropilin-1 causes increased sensory domain innervation.
(A-F) Whole mount samples from E16.5 cochleae from WT and Nrp1sema-/sema- mice. Neurons were stained with anti-TUJ1 antibodies (green) and hair cells were stained with anti-Myo6 antibodies (red). The arrows in D and F point to excessive nerve fibers innervating the hair cell region. Panels C and D show TUJ1 labeling with Sox2 immunolabeling, which identifies the different supporting cell types. OHC, outer hair cell; IHC, inner hair cell; dc, Deiters’ cell; opc, outer pillar cell; ipc, inner pillar cell. (G-N) Analysis of SGN afferents (Syt1 staining; magenta) and olivocochlear efferents (Gap43 staining; green) from E18.5 cochleae from WT and Nrp1sema-/sema- mice. The arrows in panels M and N point to an aberrant nerve fiber in the OHC region of Nrp1sema-/sema- cochleae. The arrowheads in M and N point to a cluster of nerve fibers that appears to be traversing across the cochlear radial fibers in an unusual manner. (O) Panel O showing that loss of Sema-mediated activation of neuropilin-1 can lead to a significant increase in Syt+ SGN processes projecting into the OHC region.
Fig 8.
Nrp 1 mediates the semaphorin-3A-induced inhibition of neurite outgrowth.
SGN explants were transfected with either 50nM Nrp1 siRNA or scrambled siRNA, and treated with semaphorin-3A (250 ng/ml). Neurons were stained with β-tubulin III monoclonal antibodies. Control (A); Semaphorin-3A alone (B); Nrp1 siRNA alone (C); D. Nrp1 siRNA and semaphorin-3A (D); Scrambled siRNA alone (E); Scrambled siRNA and semaphorin-3A (F). Scale bar = 200 μm. (G,H, J) Recombinant Sema3a inhibits SGN outgrowth in whole cochlear cultures. (G, H) Three-dimensional confocal z-stacks from E17.5 cochlear cultures treated with 20nM of either control IgG-Fc or Sema3a-Fc. Tissue samples were stained with anti TUJ1 to mark SGNs (white) and Myo6 for HCs (blue). Note the dramatically diminished presence of fibers in the sample treated with Sema3a-Fc (arrowheads). Scale bar = 10um. (I) Average length of neurites grown from the SGN explants showing statistically significant decreased neurite outgrowth for Sama3a, Nrp1 siRNA, and scrambled siRNA/Sema3a when compared to the control group. * p<0.001. (J) Panel J showing average percent neurite coverage of sensory epithelium. Error bars, SEM. * p<0.0001. (K) Western immunoblots of Nrp1 protein from SGN explants after transfection with either Nrp1 siRNA or scrambled siRNA control as indicated. β- Actin was used as a loading control.
Fig 9.
ABR for WT, Nrp1fl/+;Pax2Cre, and Nrp1fl/fl;Pax2Cre mice at 2 and 4 months of age.
ABR results of the 2-month-old mice (Panel A) showed that the hearing thresholds of the Nrp1fl/fl;Pax2Cre group (n = 8) were significantly higher than WT controls (n = 11) at 4kHz, 8kHz, 16kHz, 24kHz, and 32kHz (A). At 4 months of age (Panel B), Nrp1fl/fl;Pax2Cre mice (n = 8) showed elevated hearing thresholds at all tested frequencies except 12 kHz when compared to WT mice (n = 11) (B).
Fig 10.
Wave I P-N amplitude for WT, Nrp1fl/+;Pax2Cre, and Nrp1fl/fl;Pax2Cre mice at 2 and 4 months of age.
Analysis of the wave I P-N amplitudes at 80dB suprathreshold did not show significant difference between Nrp1fl/+Pax2Cre mice and WT controls at 2-month-old (A); however, statistically significant higher amplitudes at 8kHz, 12kHz, and 24kHz were shown in 4-month-old Nrp1fl/fl;Pax2Cre mutants compared to controls (B). *p<0.05.
Fig 11.
Stria vascularis abnormalities in Nrp1fl/fl;Pax2Cre mutant mice at P5 and 4 months of age.
Panel A showing neuropilin-1 expression (green) on vascular tissue of the cochlea. Arrows show neuropilin-1 stained vessels (A). Lectin immunostaining of WT and Nrp1fl/fl;Pax2Cre cochleae at P5 and 4 months of age were performed to assess the morphology of the micro-vessels of the SV at the basal turn of the cochlea (n = 3 per group) (B-G). Grossly enlarged SV microvessels were apparent in both ages of Nrp1fl/fl;Pax2Cre mutants. The minimum and maximum microvessel diameter of the Nrp1fl/fl;Pax2Cre mice were 4.17μm and 43.67μm at P5, and 3.22 and 95μm in 4-month-old mice, respectively. The minimum and maximum microvessel diameters of the WT mice were 5.32μm and 23.90μm at P5, and 5.2μm and 26.15μm in 4-month-old mice, respectively. Panel B and C showing normal morphology of the microvessels of the WT mice at P5 and 4-month-old (B). Nrp1fl/fl;Pax2Cre mutants at P5 showed some enlarged microvessels when compared to WT controls (D). Panel E-G showing vascular abnormalities of 3 different 4-month-old Nrp1fl/fl;Pax2Cre mutants (E, G). Overall, the maximum microvessel diameter in 4-month-old Nrp1fl/fl;Pax2Cre mice was 3.6 fold higher than WT controls. Scale bar = 30 μm.