Rac1 and Nectin3 are essential for PCP-directed axon guidance in the peripheral auditory system

Abstract

Our sense of hearing is critically dependent on the spiral ganglion neurons (SGNs) that connect the sound receptors in the organ of Corti (OC) to the cochlear nuclei of the hindbrain. Type I SGNs innervate inner hair cells (IHCs) to transmit sound signals, while type II SGNs (SGNIIs) innervate outer hair cells (OHCs) to detect moderate-to-intense sound. During development, SGNII afferents make a characteristic 90-degree turn toward the base of the cochlea and innervate multiple OHCs. It has been shown that the Planar Cell Polarity (PCP) pathway acts non-autonomously to mediate environmental cues in the cochlear epithelium for SGNII afferent turning towards the base. However, the underlying mechanisms are unknown. Here, we present evidence that PCP signaling regulates multiple downstream effectors to influence cell adhesion and the cytoskeleton in cochlear supporting cells (SCs), which serve as intermediate targets of SGNII afferents. We show that the core PCP gene Vangl2 regulates the localization of the small GTPase Rac1 and the cell adhesion molecule Nectin3 at SC-SC junctions through which SGNII afferents travel. Through in vivo genetic analysis, we also show that loss of Rac1 or Nectin3 partially phenocopied SGNII peripheral afferent turning defects in Vangl2 mutants, and that Rac1 plays a non-autonomous role in this process in part by regulating PCP protein localization at the SC-SC junctions. Additionally, epistasis analysis indicates that Nectin3 and Rac1 likely act in the same genetic pathway to control SGNII afferent turning. Together, these experiments identify Nectin3 and Rac1 as novel regulators of PCP-directed SGNII axon guidance in the cochlea.

Keywords: Nectin3; Planar Cell Polarity; Rac1; axon guidance; hair cell; intermediate target; spiral ganglion neuron; supporting cells.

Figure 1:. PCP signaling in the cochlea

A, A cross-sectional view of the cochlea demonstrating Type I and II SGN afferent innervation patterns. The white bar indicates the HC-SC junctional level, while the blue bar indicates the SC-SC junctional level. B, Schematic diagrams showing asymmetric localization of Fzd3/6 and Vangl2 at the HC-SC and SC-SC junctional levels. C, D, Loss of PCP signaling leads to misoriented HBs (C) and a percentage of SGNII afferents to turn errantly toward the cochlear apex (shown in red) (D).

Figure 2:. Rac1 is required in the cochlear epithelium for proper SGNII afferent turning

A, Rac1 localization at the SC-SC junctions in E18.5 control and Vangl2^KO cochleae. White arrowheads mark the IPC row. B, Quantifications of Rac1 staining intensity across apex- and base-facing SC-SC junctions, normalized to the base of the control cochleae in each experiment. Control N = 3; Vangl2^KO N = 4. Total numbers of cell junctions quantified in each group are indicated by n. C, HBs and kinocilia in Rac3^KO and Rac1^cKO; Rac3^KO E18.5 cochlea labeled by F-actin (magenta) and acetylated tubulin (green) staining, respectively. D, SGNII afferents labeled by acetylated tubulin staining in E18.5 control, Rac1^cKO, Rac3^KO, and Rac1^cKO; Rac3^KO cochleae. Afferents turning errantly toward the apex are denoted by cyan arrowheads. E, Quantifications of SGNII afferents turning errantly toward the apex in the indicated genotypes. Number of mice represented in each column is notated. For all graphs: Mean +/− stdev, one-way ANOVA with Tukey’s Post-test, ns: not significant, **: p ≤ 0.01, ****: p ≤ 0.0001. Scale bars: 10 μm.

Figure 3:. Expression of constitutively active Rac1 did not rescue Vangl2^KO HB misorientation or SGNII afferent misturning

A, SGNII afferents labeled by acetylated tubulin staining in P0 cochleae of the indicated genotypes. Afferents turning errantly toward the apex are denoted by cyan arrowheads. B, Quantifications of SGNII afferents turning errantly toward the apex in the indicated genotypes. Number of samples represented in each column is noted. C, HBs marked by F-actin staining in the indicated genotypes. D, Quantifications of HB orientation in the indicated genotypes. Control N = 5; R26-Rac1DA/+ N = 3; Vangl2^cKO N = 4; Vangl2^cKO; R26-Rac1DA/+ N = 6. Total numbers of HCs quantified in each group are indicated by n. E, Localization of the HA-tagged Rac1DA protein in control, Pax2-Cre; R26-Rac1DA/+, and Pax2-Cre; Vangl2^cKO; R26-Rac1DA/+ P0 cochleae at the HC apical surface level (top) and SC-SC junction (bottom). White arrowheads indicate the IPC row. For all graphs: Mean +/− stdev, one-way ANOVA with Tukey’s Post-test, ns: not significant, ****: p ≤ 0.0001. Scale bars: 10 μm (A, E), 6 μm (C).

Figure 4:. Rac1 is required in the cochlear epithelium for proper core PCP protein localization

A, C, Vangl2 (A) and Dvl3 (C) localization at SC-SC junctions in E17.5 control and Rac1^cKO cochleae. B, D, Quantifications of Vangl2 (B, N=5 each) and Dvl3 (D, N=3 each) staining intensity across apex- and base-facing SC-SC junctions, normalized to the base of the control cochleae in each experiment. Total numbers of cell junctions quantified in each group are indicated by n. Mean +/− stdev, one-way ANOVA with Tukey’s Post-test, ns: not significant, *: p ≤ 0.05, ***: p ≤ 0.001, ****: p ≤ 0.0001. White arrowheads indicate the IPC row. Scale bars: 10 μm.

Figure 5:. Nectin3 localization in SCs is regulated by Vangl2

A, Nectin3 localization to the HC-SC junctions in the basal and mid-apical regions of E18.5 control and Vangl2^KO cochleae. B, Quantifications of Nectin3 staining intensity at the HC-SC junctions in E18.5 control and Vangl2^KO cochleae normalized to the average of the littermate control basal region. For both genotypes, N = 4. C, Nectin3 localization at SC-SC junctions in the same samples and regions as in A. Panels in left column are adjusted to the same brightness as panels showing the HC-SC junctional staining, while the brightness in the middle and right columns is uniformly adjusted to optimize visibility. D, Quantifications of Nectin3 staining intensity across apex- and base-facing SC-SC junctions, normalized to the base of the control cochleae in each experiment. Control N = 5; Vangl2^KO N = 6. For all graphs: Total numbers of cell junctions quantified in each group are indicated by n. Mean +/− stdev, one-way ANOVA with Tukey’s Post-test, ns: not significant, ***: p ≤ 0.001, ****: p ≤ 0.0001. White arrowheads indicate the IPC row. Scale bars: 10 μm.

Figure 6:. Nectin3 is necessary for proper SGNII afferent turning

A-E, SGNII afferents labeled by acetylated tubulin staining in P0-P1 cochleae of the indicated genotypes, with errantly turning afferents denoted by cyan arrowheads. F, G, Quantifications of SGNII afferents errantly turning toward the apex in the indicated Nectin3^del10 (F) and Nectin3^dE2 genotypes (G). Although both alleles are predicted to be null alleles, the Nectin3^del10/del10 had significantly higher SGNII afferent misturning compared with the Nectin3^dE2/dE2 (unpaired Student’s T-test, p=0.0133, comparison not shown). For all graphs: Mean +/− stdev, One-way ANOVA with Tukey’s Post-test, ns: not significant, *: p ≤ 0.05; **: p ≤ 0.01, ****: p ≤ 0.0001. Scale bar: 10 μm.

Figure 7:. Nectin3 and Rac1 likely act in the same genetic pathway to regulate SGNII afferent turning

A-C, SGNII afferents labeled by acetylated tubulin staining in E18.5-P0 Nectin3^KO (A), Rac1^cKO (B), and Nectin3^KO; Rac1^cKO (C) cochleae, with examples of errantly turning afferents denoted by cyan arrowheads. D, Quantification of SGNII afferents errantly turning toward the apex in the indicated genotypes. E, Rac1 localization at the SC-SC junctions of control and Nectin3^KO E18.5 cochleae. White arrowheads indicate the IPC row. F, Quantification of Rac1 staining intensity across apex- and base-facing SC-SC junctions, normalized to the base of the control cochlea in each experiment. N = 5 for both genotypes. G, Nectin3 localization at the SC-SC junctions of Control and Rac1^cKO E18.5 cochleae. White arrowheads indicate the IPC row. H, Quantification of Nectin3 staining intensity across apex- and base-facing SC-SC junctions, normalized to the base of the control cochlea in each experiment. N = 3 for both genotypes. For all graphs: Total numbers of cell junctions quantified in each group are indicated by n. Mean +/− stdev, one-way ANOVA with Tukey’s Post-test, ns: not significant, *: p ≤ 0.05, ***: p ≤ 0.001, ****: p ≤ 0.0001. Scale bar: 10 μm.

Figure 8:. Proposed model of SC-derived SGNII afferent guidance cues

A, A proposed “Goldilocks” model for optimal substrate properties required for SGNII afferent turning. Both increased presence of Rac1 and Nectin3, as occurred in Vangl2^KO, and their absence, impairs substrate-derived axon guidance cues for SGNII afferent turning. B, Genetic evidence supports a regulatory loop for protein localization of Vangl2, Nectin3, and Rac1 at SC-SC junctions to guide SGNII afferents. Rac1 also regulates Dvl3 localization of SC-SC junctions.