The crosstalk between the Notch, Wnt, and SHH signaling pathways in regulating the proliferation and regeneration of sensory progenitor cells in the mouse cochlea

Sensory hair cells (HCs) are highly susceptible to damage by noise, ototoxic drugs, and aging. Although HCs cannot be spontaneously regenerated in adult mammals, previous studies have shown that signaling pathways are involved in HC regeneration in the damaged mouse cochlea. Here, we used a Notch antagonist (DAPT), a Wnt agonist (QS11), and recombinant Sonic hedgehog (SHH) protein to investigate their concerted actions underlying HC regeneration in the mouse cochlea after neomycin-induced damage both in vivo and in vitro. With DAPT, the numbers of HCs increased, and supporting cell (SC) proliferation was seen in both the intact and damaged cochlear sensory epithelia, while these numbers were unchanged in the presence of QS11. When simultaneously treated with DAPT and QS11, the number of HCs increased dramatically, and much greater SC proliferation was seen in the cochlear epithelium. In transgenic mice with both Notch1 conditional knockout and β-catenin over-expression, cochlear SC proliferation and HC regeneration were more obvious than in either Notch1 knockout or β-catenin over-expressing mice separately. When cochleae were treated with DAPT, QS11, and SHH together, SC proliferation was even greater, and this proliferation was seen in both the HC region and the greater epithelial ridge. High-throughput RNA sequencing was used to identify the differentially expressed genes between all groups, and the results showed that the SHH and Wnt signaling pathways are involved in SC proliferation. Our study suggests that co-regulation of the Notch, Wnt, and SHH signaling pathways promotes extensive cell proliferation and regeneration in the mouse cochlea. Supplementary Information The online version contains supplementary material available at 10.1007/s00441-021-03493-w.


Introduction
The mammalian cochlea has very limited spontaneous hair cell (HC) regeneration ability, and such regeneration is only seen at embryonic or very early neonatal ages and the quantity and quality of this limited spontaneous HC regeneration cannot restore cochlear function (Jung et al. 2013). When the inner ear sensory HCs are damaged, there is no spontaneous HC regeneration in the adult cochlea, and this leads to permanent hearing loss (Herranen et al. 2020;Nakano et al. 2020). Thus, stimulating sensory precursor cell proliferation and transdifferentiation into new HCs might be an effective method for rescuing hearing dysfunction (Yamoah et al. 2020).
Signaling pathways play fundamental roles in biological activities, and many signaling pathways have crosstalk with each other, which is also the case in the period of inner ear development and during HC regeneration (Jayasena et al. 2008;Petrovic et al. 2015;Roccio and Edge 2019). Jingfang Wu and Wen Li contributed equally to this work.

3
Therefore, regulating these signaling pathways is likely to be an effective way to stimulate sensory progenitor cell proliferation and differentiation into new HCs. The Notch, Wnt, and Sonic hedgehog (SHH) signaling pathways play critical roles in controlling inner ear sensory cell proliferation and differentiation (Roccio and Edge 2019). The Notch signaling pathway induces supporting cell (SC) and HC proliferation and differentiation via lateral inhibition during the early development of the inner ear (Daudet and Zak 2020;Takebayashi et al. 2007), and the Wnt signaling pathway controls sensory progenitor cell polarity, proliferation, specification, and differentiation (van Amerongen and Nusse 2009). The SHH pathway plays an important role in inner ear development (Driver et al. 2008) and regulates cochlear HC differentiation (Benito-Gonzalez and Doetzlhofer 2014; Bok et al. 2013). The Notch and Wnt signaling pathways also crosstalk with each other in regulating and maintaining the balance between cell proliferation and differentiation in many tissues (Fre et al. 2009;Munnamalai and Fekete 2020). SHH and Notch function together to maintain various neuronal progenitor cell populations (Takanaga et al. 2009;Wu et al. 2012), and the interaction between Wnt and SHH signaling is important for the dorsal polarity of otocysts (Ohta and Schoenwolf 2018), and both pathways are required for the formation of the semicircular canals (Hwang et al. 2019). However, the crosstalk between the Notch, Wnt, and SHH signaling pathways in regulating the proliferation and regeneration of sensory progenitor cells in the mouse cochlea has not been reported in detail.
Therefore, we cultured neonatal mouse cochleae with DAPT, QS11, and/or SHH together with the cell proliferation marker EdU for 3 days. Also, we damaged the HCs with neomycin and then treated the cochleae with DAPT, QS11, and/or SHH for 7 days. Furthermore, using transgenic mice, we found that that Notch inhibition and Wnt activation could promote greater proliferation and regeneration of SCs and HCs compared to only inhibiting the Notch pathway. Moreover, simultaneously activating the Wnt and SHH signaling pathways could further collaborate with DAPT to promote the proliferation of sensory progenitor cells. Thus, our work presents novel information about the Notch, Wnt, and SHH signaling pathways and their crosstalk in regulating sensory cell proliferation and regeneration and suggests a new way to regenerate more mammalian HCs after they are damaged.
Sox2-Cre-ER mice were a gift from Dr. Konrad Hochedlinger of Harvard University. The care and use of animals were approved by the Institutional Animal Care and Use Committee of Fudan University in compliance with the NIH guidelines for the care and use of laboratory animals.

Organotypic culture of wild-type mouse cochleae
The mice were euthanized by carbon dioxide asphyxiation and decapitated, and the heads were placed in 75% ethanol and quickly transferred to chilled Hanks' balanced salt solution (HBSS). The temporal bones were dissected out, and the cochleae were isolated from the temporal bones using sterile procedures in ice-cold HBSS. The stria vascular and spiral ganglion were removed with fine forceps.
Explants of the organ of Corti were placed intact on polylysine-coated cover glasses (Sigma, St. Louis, MO, USA) and maintained in four-well culture dishes (Greiner Bio-One, Frickenhausen, Germany) in Dulbecco's modified Eagle's medium (DMEM) and F12 medium supplemented with N2 and B27 (Invitrogen/GIBCO/BRL, Carlsbad, CA, USA) and 50 IU/mL penicillin (Sigma, St. Louis, MO, USA). The tissues were incubated at 37 °C in a humidified atmosphere of 5% CO 2 .
The cochlear explants were treated with 5 μM DAPT and 10 μM QS11 and/or 200 ng/ml SHH for 3 days. Control cultures were incubated in 0.1% DMSO or PBS. The cell proliferation marker EdU (5-ethynyl-2′-deoxyuridine, Invitrogen, Grand Island, NY, USA), which is efficiently incorporated into newly synthesized DNA, was added to the culture medium at a concentration of 10 μM for the entire culture period.
Another group of cochlear explants were treated with 1 mM neomycin (Sigma, St. Louis, MO, USA) for 16 h and then thoroughly rinsed in fresh medium. The cultures 1 3 were then treated with 5 μM DAPT and 10 μM QS11 and/or 200 ng/ml SHH along with 10 μM EdU for another 7 days (Kersigo et al. 2021). Control cultures were cultured in 0.1% DMSO or PBS.

Treatment of Notch1 knockout and β-catenin over-expression mouse cochleae
Sox2-CreER and Notch1-flox (exon1) mice were mated with Catnb-flox (exon3) mice to generate Sox2-CreER, Notch1flox (exon1), and Catnb-flox (exon3) mice. Littermates without Notch1-flox (exon1) or without Catnb-flox (exon3) were used as controls. Tamoxifen (2 mg/25 g body weight, Sigma, St. Louis, MO, USA) was given to the female mice by intraperitoneal injection, and the mother transferred the tamoxifen to the pups via her milk to activate the Cre recombinase. EdU (50 mg/kg) was given to the pups by intraperitoneal injection twice a day for 7 days. The pups were sacrificed at 7 days, and their cochleae were dissected out for subsequent immunohistochemistry processing. The mice were genotyped after being sacrificed.
We also dissected the cochleae from P0 to P1 mice and cultured the cochleae. All of the pups were genotyped after being sacrificed. After 2 h of adhering, the cultured cochleae were exposed to 1.0 mM neomycin in the culture medium for 16 h. The cultures were washed thoroughly with fresh medium and then treated with 1 μM tamoxifen (Sigma, St. Louis, MO, USA) for 2 days and then incubated in culture medium containing 10 μM EdU for another 7 days.

Immunohistochemistry
The cultured cochleae were harvested and fixed for 30 min at room temperature with 4% paraformaldehyde in 0.1 M phosphate buffer and then thoroughly rinsed with 0.01 M PBS. After being permeabilized in 0.5% Triton X-100 in PBS (PBST) for 30 min at room temperature, the proliferating cells were labeled with EdU (Click-iT EdU Alexa Fluor 488 Imaging Kit, Invitrogen, Carlsbad, CA, USA) for another 30 min according to the manufacturer's protocol. The tissues were blocked with 10% horse serum in PBST for 30 min, and the HCs were labeled with the rabbit antibody against Myo7a (1:500 dilution; Proteus Biosciences, Ramona, CA, USA) and the goat polyclonal antibody against Sox2 (1:200 dilution; Santa Cruz Biotechnology, Santa Cruz, CA, USA) at 4 °C overnight.
After being washed with PBST to remove the unbound antibodies, the tissues were incubated with donkey antirabbit (1:500 dilution; Invitrogen, Grand Island, NY) and donkey antigoat (1:200 dilution; Invitrogen, Grand Island, NY) diluted in PBST for 1 h at room temperature to visualize Myo7a and Sox2, respectively. The tissues were stained with 4′, 6-diamidino-2-phenylindole dihydrochloride (DAPI) for 5 min at room temperature to visualize the cell nuclei.
To detect functional mechanoelectrical transduction channels, 5 μM FM1-43 (Thermo Fisher Scientific, Eugene, OR, USA) was incubated with the cultured organ of Corti for 30 s and then washed with PBS for three times before fixation.

Image acquisition and quantification
The fluorescent antibody-labeled organ of Corti samples was visualized using a Nikon (Miyagi, Japan) Eclipse 80i microscope, and the high-magnification fluorescent images were obtained with a Leica TCS SP5 microscope (Wetzlar, Germany). Cells were counted manually with the ImageJ software (Wayne Rosband, NIH, USA). The whole-mount cochleae were split into the apical, middle, and basal turns for counting the EdU-labeled cells, Sox2-labeled cells, and Myo7a-labeled cells. All cells were counted per 100-μm length of the cochlea. The length of each region was measured by drawing a line between inner and outer HCs using the ImageJ software. At least five samples in each group from three independent experiments were collected for statistical analysis. The cell counts for the control and treated groups were compared using Student's t-test.

RNA extraction for RNA-Seq analysis
Ten cultured cochleae from independent culture groups were dissolved in RNALater, and RNA-Seq libraries were generated using the SMART-Seq® v4 Ultra® Low Input RNA Kit for Sequencing (Takara Bio USA, Mountain View, CA, USA). Quality control of the raw sequencing data was performed using FastQC (http:// www. bioin forma tics. babra ham. ac. uk/). Adaptors were trimmed and low-quality reads were removed using Trimmomatic (Bolger et al. 2014). Alignment to the mouse reference genome (mm10) was done with Hisat2, and aligned reads were counted with HTseq (https:// doi. org/ 10. 1093/ bioin forma tics/ btu638) according to the UCSC mm10 annotation. Differential expression analysis between conditions was performed using the DESeq2 R package (Love et al. 2014), and genes with an adjusted p-value < 0.05 and fold-change > 1.5 were considered to be differentially expressed. Gene ontology analysis of the significantly differentially expressed genes was done with DAVID, and pheatmap in the R package was used to generate heatmaps.
We also performed quantitative real-time PCR (qRT-PCR) to validate the differentially expressed genes. Three to five cultured cochleae from independent culture groups were pooled and dissolved in 500 μl TRIzol (Invitrogen), and the total RNA was extracted using the RNeasy Micro Kit (QIA-GEN, Hilden, Germany) according to the manufacturer's protocol. The mRNA was reverse-transcribed to synthesize cDNA using the GoScript™ Reverse Transcription System 1 3 (Promega, Maddison, WI, USA). qRT-PCR was performed using the GoTaq® qPCR Master Mix (Promega, Maddison, WI, USA), 1 μl retrotranscribed cDNA, and gene-specific primer sets in a volume of 20 μl in a LightCycle 480 (Roche). GAPDH was used for calibration. Each qRT-PCR was performed in triplicate in a volume of 20 μL using a StepOne Plus Real-Time PCR System (Applied Biosystems, Foster City, CA) for 35 cycles. Gene expression levels in the samples of each group were calculated using the comparative CT method where ΔΔCT = ΔCT sample -ΔCT control and the fold change = 2 −ΔΔCT .

Statistics
All data were analyzed with the GraphPad Prism software using a two-tailed, unpaired Student's t-test when comparing two groups or with a one-way ANOVA followed by a Dunnett's multiple comparisons test when comparing more than two groups. All data are expressed as either a percentage or as the mean ± SEM. p-Values < 0.05 were considered statistically significant.
We also used FM1-43 to label HCs with active mechanotransduction channels. FM1-43 fluorescence could be seen in the EdU + /Myo7a + cells in the cultured cochleae of the DAPT + QS11 group ( Supplementary Fig. 1), suggesting that the mitotically regenerated HCs after DAPT and QS11 co-treatment had active mechanotransduction channels.

Fig. 1
Co-treatment with DAPT and QS11 promotes SC proliferation and HC regeneration in the cochlea in vitro in an age-dependent manner. The P0-P1 cochleae were cultured with 10 μM QS11 a or 5 μM DAPT b or with both QS11 and DAPT c for 3 days. There were significantly greater numbers of EdU + /Sox2 + cells and EdU + / Myo7a + cells in the apex of the cochlea compared to the DAPT-only group and the QS11-only group. The EdU + cells in the DAPT + QS11 co-treated cochleae were seen throughout the sensory cell region in the apical turn. Scale bar = 20 μm. A control cochlea treated with vehicle is shown in d (Scale bar = 50 μm). The histograms show the differences in the numbers of EdU + cells e and Myo7a + cells f between these groups. The cells were counted per 100-μm length along the cochlea. The numbers of EdU + cells were counted between the outer HCs and the inner HCs. (*p < 0.05, **p < 0.01, ***p < 0.001). The P4 cochleae were cultured with 10 μM QS11 g-g'" or 5 μM DAPT h-h'" or with both QS11 and DAPT i-i'" for 3 days. The number of EdU + and Myo7a + cells was greater than in the DAPT-only and QS11-only groups, while the numbers of Myo7a + cells and EdU + cells in the sensory area were fewer compared to P0-P1 cochleae with the same treatment. Scale bar = 20 μm. The histograms show the differences in the numbers of Myo7a + cells j and EdU + cells k in the apical turn of the cochlea between the P0-P1 mice and the P4 mice for the different treatment groups. The cells were counted per 100-μm length along the cochlea. The numbers of EdU + cells were counted between the outer HCs and the inner HCs. (*p < 0.05, **p < 0.01, ***p < 0.001)

Notch inhibition together with Wnt activation promotes SC proliferation and HC regeneration in damaged mouse cochleae in vitro
To investigate the effect of Wnt activation and Notch inhibition after HC ablation, we cultured P0-P1 mouse cochleae with neomycin to damage the sensory HCs. After HC ablation, the cochleae were treated with DAPT and/or QS11 and EdU for 7 days.
These results suggest that inhibiting Notch signaling induces regeneration of new HCs in the damaged cochlea and that simultaneous Notch inhibition and Wnt activation induces even more regenerated HCs in the damaged cochlea.

Co-regulation of Notch inhibition and Wnt activation promotes SC proliferation and HC regeneration in transgenic mouse cochleae
We made the Sox2-CreER mice and Notch1-flox (exon1) mice and mated them with Catnb-flox (exon3) mice to generate Sox2-CreER, Notch1-flox (exon1), and Catnb-flox (exon3) mice in which Wnt signaling was activated and Notch signaling was inhibited specifically in Sox2 + SCs ( Fig. 3(a)). In control mice lacking the Sox2-CreER line, the number of HCs was not significantly changed compared to wild-type mice and no Sox2 + /EdU + or Myo7a + / EdU + cells were observed in the sensory epithelium ( Fig. 3(b-b")). In the cochleae of the Notch1-flox (f/f), Catnb-flox (exon3) (f/ +), and Sox2-CreER (+ / −) mice, Fig. 2 Co-treatment with DAPT and QS11 promotes SC proliferation and HC regeneration in neomycin-damaged cochleae in vitro. The P0-P1 cochleae were cultured with 1.0 mM neomycin for 16 h, then treated with either media only a-a", 5 μM DAPT b-b", 10 μM QS11 c-c", or both d-d" for another 7 days. The number of Myo7a + cells was decreased in the apical, middle, and basal turns of the cochlea, especially in the basal turn and middle turns in the control and QS11-only groups. However, the numbers of EdU + / Sox2 + cells and EdU + /Myo7a + cells were increased in the DAPTonly and DAPT + QS11 groups, especially in the apical turn of the cochlea. In addition, there were many more EdU + cells in the sensory area of the cochlea than in the DAPT-only group, especially in the apical turn of the cochlea (scale bar = 20 μm). The histograms show the differences in the numbers of EdU + cells e and Myo7a + cells f in the different turns of the cochlea. The cells were counted per 100-µm length along the cochlea, which was measured between the outer HCs and the inner HCs (**p < 0.01, ***p < 0.001). The P0-P1 Atoh1-eGFP transgenic mouse cochleae were cultured with 2.0 mM neomycin for 16 h then treated with either media only g-g", 5 μM DAPT h-h", 10 μM QS11 i-i", or both j-j" for another 7 days. Myo7a + cells were clearly lost in the basal turn and the middle turn of the cochlea in the control and QS11-only groups, and there were no EdU + /Sox2 + cells or EdU + /Myo7a + cells in the sensory domain in these two groups. However, there were some EdU + / Sox2 + cells and EdU + /Myo7a + cells in the sensory area of the cochleae in the DAPT-only and the combination groups, with more cells seen in the combination group compared to the DAPT-only group, especially in the apical turn of the cochlea. The P0-P1 Atoh1-eGFP mouse cochleae were cultured only with EdU, and there were no obvious EdU + /Sox2 + cells or EdU + /Myo7a + cells in the sensory area k. The histograms show the differences in the numbers of EdU + cells l and Myo7a + cells m in different turns of the cochlea. The cells were counted per 100-µm length along the cochlea, which was measured between the outer hair cells and the inner hair cells (*p < 0.05, **p < 0.01) there were numerous Sox2 + /EdU + cells but only a few Myo7a + /EdU + cells in the sensory epithelium. Most of the EdU + cells were in the apical turn and in the pillar cell region of the cochlea. The number of HCs was increased, especially in the apical turn of the cochlea (Fig. 3(c-c")). The vast majority of the increased HCs were also Sox2 + and EdU + , indicating that they were mitotically regenerated immature HCs.

Fig. 3
Transgenic mouse model for Notch1 inhibition and Wnt activation in SCs. The Sox2-CreER and Notch1-flox (exon1) mice were crossed with Catnb-flox (exon3) mice to generate pups. Tamoxifen (2 mg/25 g) was given to the female mice by intraperitoneal injection, and the mother transferred the tamoxifen to the pups via her milk to activate the Cre recombinase. EdU (50 mg/kg) was given to the pups by intraperitoneal injection twice a day for 7 days a. In the Notch1flox (f/f), Catnb-flox (exon3) (f/ +), and Sox2-CreER (− / −) mouse cochleae, the number of Myo7a + cells was not changed and there were no obvious Sox2 + /EdU + or Myo7a + /EdU + cells in the sensory epithelium in the apical b, the middle b', or the basal b" turns, while in the Notch1-flox (f/f), Catnb-flox (exon3) (f/ +), and Sox2-CreER (+ / −) mouse cochleae there were also numerous Sox2 + / EdU + cells, but few Sox2 + /EdU + cells, in the sensory epithelium. Most of the EdU + cells were in the apical turn and in the pillar cell region of the cochlea including the apical c, the middle c', and the basal c" turns. Scale bar = 20 μm 1 3

The mechanism by which DAPT + QS11 + SHH co-treatment induces SC and HC proliferation and regeneration in neomycin-damaged cochleae in vitro
To systematically investigate the molecular mechanism underlying the effect of DAPT + QS11 + SHH co-treatment, we performed RNA-Seq analyses to assess the genome-wide expression profiles in the cochleae from the neomycin-only control group, the Neo + SHH group, the Neo + DAPT + QS11 group, and the Neo + DAPT + QS11 + SHH group. A total of 3359 DE (differentially expressed) genes between each group were identified (Fig. 6a), and 429 DE genes between the Neo + DAPT + QS11 group and the Neo + DAPT + QS11 + SHH group were identified. Gene ontology analysis suggested the functional roles of these DE genes in sensory system development (Fig. 6b).
Among the DE genes between these groups, we found that the expression of many genes in the SHH, Hippo, Wnt, Notch, and TGF-β pathways as well as many transcription factors was not significantly different between the Neo + SHH group and the control group.

Fig. 5
The role of SHH with DAPT and QS11 in the damaged inner ear in vitro. Neonatal P0-P1 cochleae were cultured with 1.0 mM neomycin for 16 h and then were treated with either 200 ng/ml SHH a-a", d-d", 5 μM DAPT, or 10 μM QS11 b-b", e-e" or the combination of all three c-c", f-f" for 7 days. There were some EdU + / Sox2 + cells and EdU + /Myo7a + cells in the sensory area of the cochlea. The numbers of EdU + cells in the DAPT + QS11 + SHH group were greater than those in the DAPT + QS11 group a-a", b-b", c-c", especially in the apical turn of the cochlea d-d", e-e", f-f" (scale bar = 20 μm). The histograms show the numbers of Myo7a + cells d and EdU + cells e in these groups. The cells were counted per 100-μm length along the cochlea, which was measured between the outer hair cells and the inner hair cells (*p < 0.05, ***p < 0.001) Comparing the Neo + DAPT + QS11 + SHH group and the Neo + DAPT + QS11 group, the upregulated genes included Pou4f3 and Atoh1 in the TF pathway (Fig. 6e); Rassf4, Fat1, Fat3, etc., in the Hippo pathway; Cd44 and Fzd4 in the Wnt pathway (Fig. 6g); and Bmper, Serpine1, and Tgfb1 in the TGF-β pathway (Fig. 6c). The downregulated genes included Hes1, Il33, and Fabp7 in the Notch pathway (Fig. 6f), Wif1 in the SHH and Wnt pathways (Fig. 6g, h), and Meis and Tshz3 in the Hippo pathway. These results indicated that treatment with SHH alone could not significantly upregulate the SHH pathway, while DAPT + QS11 co-treatment could. The DAPT + QS11 + SHH co-treatment induced even more SC proliferation, likely via the TF, Hippo, and Wnt signaling pathways in response to neomycin-induced damage in the cochlea (Fig. 7a).
The qRT-PCR also showed that the Jag1, Hes1, and Hes5 genes were downregulated in the Neo + DAPT + QS11 and the Neo + DAPT + QS11 + SHH groups compared to the control group, while the β-catenin, Wnt16, Cd44, Atoh1, Pou4f3, Bmp2, and Disp2 genes were upregulated (Fig. 7b). These results were consistent with the RNA-Seq results, which also indicated that Neo + DAPT + QS11 and Neo + DAPT + QS11 + SHH co-treatment could inhibit the Notch signaling pathway and activate the Wnt and SHH signaling pathways, while these pathways were unchanged in the SHH-only group.

Discussion
Inner ear sensory HCs are required for hearing and balance functions. These highly differentiated mammalian sensory HCs are vulnerable to damage by ototoxic drugs, noise, genetic defects, etc., and they have a very limited ability to spontaneously regenerate new HCs after being damaged (Oesterle et al. 2008;Yamoah et al. 2020). In contrast, damaged HCs can be regenerated through the proliferation and differentiation of SCs in non-mammalian vertebrates (Hawkins et al. 2006). This ability is absent in mammals largely because the SCs cannot directly differentiate into HCs or undergo proliferation followed by differentiation into HCs. Therefore, controlling the genes and signaling pathways associated with HC regeneration in order to activate SC and HC proliferation and differentiation is the ideal way to induce HC regeneration.

The orchestrated signaling pathways in cochlear development and HC regeneration
The Notch, Wnt, and SHH signaling pathways all participate in regulating the development of the inner ear. During inner ear development, the Notch signaling pathway regulates the directional differentiation of cells and the formation of chimeras (Daudet and Zak 2020;Yamamoto et al. 2006) and plays important roles in the process of HC regeneration. Inhibition of the Notch signaling pathway by γ-secretase inhibitors in newborn chickens and in neonatal and adult mice induces significant numbers of newly regenerated HCs that derive from sensory progenitor cells or SCs (Lin et al. 2011;Mizutari et al. 2013).
The Wnt signaling pathway also plays critical roles in inner ear development and is involved in otic placode directional differentiation, cochlear duct extension, sensory cell directional differentiation, and cell polarity regulation (Qian Fig. 7 Differential expression analysis between groups. Gene ontology analysis was performed with the DE genes between the Neo + DAPT + QS11 and Neo + DAPT + QS11 + SHH groups a. The results of qRT-PCR of some of the DE genes showed dif-ferential expression in the Neomycin control, Neo + SHH, Neo + DAPT + QS11, and Neo + DAPT + QS11 + SHH groups, and these differences were consistent with the RNA-Seq results b van Amerongen and Nusse 2009). Activating the Wnt signaling pathway promotes the proliferation and colony formation of Lgr5 + sensory progenitor cells in the mouse cochlea (Chai et al. 2012).
SHH regulates the differentiation of HCs by influencing the cell cycle of the progenitor cells located in the cochlea (Bok et al. 2013), and it plays an important role in HC differentiation by regulating the expression of Atoh1 (Hu et al. 2010).

Crosstalk between signaling pathways
The Notch and Wnt signaling pathways crosstalk with each other on all different levels in cells in a variety of tissues (Fre et al. 2009;Qian et al. 2015). β-catenin has a central position in the Wnt signaling pathway and is the most active component participating in the crosstalk with other pathways. Studies have shown that, in the inner ear, the Wnt signaling pathway can upregulate the expression of the Jagged1, Notch1, and Hes1 genes, and these genes work through the classical Wnt/β-catenin signaling pathway to regulate the formation and size of the otic placode (Jayasena et al. 2008).
The Notch and SHH signaling pathways interact antagonistically to define the position and size of the prosensory domain in the developing cochlea (Driver et al. 2008), while the two pathways cooperate to maintain various neuronal stem and progenitor cell populations (Munnamalai and Fekete 2020;Wall et al. 2009). The balance between Wnt and SHH signaling activities is key in distinguishing between vestibular and auditory cell types (Hwang et al. 2019;Riccomagno et al. 2005).
The role of the crosstalk between all three of the Notch, Wnt, and SHH signaling pathways in sensory cell regeneration in the mammalian cochlea has not been reported before. In this study, we report the effect of the combination of Notch inhibition and Wnt activation and/or SHH signaling activation in sensory cell proliferation and regeneration in the damaged mouse cochlea. Notch pathway inhibition promotes progenitor cell differentiation into HCs in the inner ear (Jeon et al. 2011), while Wnt pathway activation promotes progenitor cell proliferation in the neonatal mouse cochlea (Chai et al. 2012). Our previous study found that inhibiting the Notch signaling pathway could induce SC proliferation and HC mitotic regeneration, while inhibiting the Wnt signaling pathway decreased SC proliferation and HC regeneration (Li et al. 2015). This means that the SC proliferation and HC regeneration induced by Notch inhibition are facilitated by activation of the Wnt signaling pathway.
In this study, we simultaneously inhibited the Notch signaling pathway and upregulated the Wnt signaling pathway in the mouse cochlea, and this induced much more SC proliferation compared to Notch inhibition alone, and the mitotically regenerated HCs with DAPT + QS11 co-treatment had active mechanotransduction channels. The same results were obtained in transgenic mice in which the Notch pathway was knocked out and the Wnt pathway was activated in Sox2 + SCs. With increased age, however, the proliferation and regeneration ability of the cochlear HCs and SCs induced by inhibition of Notch and activation of Wnt decreased, which may be due to the decreased expression of the Notch signaling pathway along with sensory HC differentiation (Lanford et al. 1999).
Notch signaling is upregulated after HC ablation in the mouse cochlea (Batts et al. 2009). Many previous studies have shown that Notch inhibition can generate new HCs by transdifferentiation of SCs in response to ototoxic damage (Lin et al. 2011). In our study, after HC ablation, simultaneous inhibition of Notch and activation of Wnt also promoted greater SC proliferation and HC regeneration compared to Notch inhibition alone. The experiments in the Atoh1-eGFP mouse cochlea further confirmed this result.

The mechanism underlying the orchestrated signaling and gene expression
We performed RNA-Seq and analyzed the gene expression between the SHH-treated group, the DAPT + QS11treated group, and the DAPT + QS11 + SHH-treated group in order to further explore the mechanism behind SC and HC proliferation and regeneration. When comparing the SHH-treated group with the control group, we found that gene expression between these two groups showed almost no differences. In contrast, 429 DE genes were identified when comparing the Neo + DAPT + QS11 group with the Neo + DAPT + QS11 + SHH group, including transcription factor genes, Hippo pathway genes, Wnt pathway genes, and TGF-β pathway genes. Thus, treatment with SHH could not significantly upregulate the SHH pathway, but co-treatment with DAPT + QS11 could upregulate the SHH pathway and induce SC proliferation and HC regeneration. Co-treatment with DAPT + QS11 + SHH induced even more SC proliferation via the transcription factor, Hippo, and Wnt signaling pathways.
These results suggest that the Wnt signaling pathway has synergistic effects with the Hippo signaling pathway in promoting SC proliferation and HC regeneration in the mouse cochlea, and simultaneous Notch inhibition and Wnt activation induced significant SC proliferation and led to the regeneration of numerous new HCs in the neonatal mouse cochlea. The new HCs were mostly regenerated from transdifferentiation of SCs, while some were mitotically regenerated from transdifferentiation of proliferating SCs and only a few were regenerated directly from mitotic division of surviving HCs. With the greater dose of neomycin and greater damage to the HCs of the cochlea, we observed more newly regenerated HCs by co-regulating the Notch and Wnt 1 3 pathways compared to Notch inhibition alone. The sensory cells in the apical turn of the cochlea have a greater capacity for proliferation and mitotic generation compared to the middle and basal turns. Thus, the SC proliferation and HC regeneration were mostly seen in the apical turn of the mouse cochlea.
When co-regulating the Notch, Wnt, and SHH signaling pathways, the SHH pathway enhances the synergistic effects of the Notch and Wnt pathways to induce greater sensory progenitor cell proliferation not only in the HC area, but also in the GER of the cochlea. Some genes are involved in mediating the crosstalk between these signaling pathways (Chen et al. 2017), and the interaction of these signaling pathways promotes sensory progenitor cell proliferation and HC regeneration.
Some important genes involved in HC differentiation and regeneration, such as p27, Hes1, and Hes5, are expressed in the GER of the cochlea at different stages (Zheng et al. 2000;Zine et al. 2001). Hes1 and Hes5 are negative regulatory genes of HC differentiation and might participate in HC differentiation by antagonizing Atoh1. Hes1 was mainly expressed in the GER and lesser epithelial ridge, while Hes5 was mainly expressed in the lesser epithelial ridge and in only a narrow band of cells in the GER (Gemmell and Nelson 1992). Hes5-deficient mice have extra HCs, and the extra HCs also express Atoh1 (Zine et al. 2001). In our study, co-regulating the Notch, Wnt, and SHH pathways downregulated Hes1 and upregulated Atoh1 more so than in the mice in which only Notch and Wnt pathways were co-regulated, and this might be the reason why there was greater sensory progenitor cell proliferation in the GER of the cochlea. The inner HCs might be derived from the cells on the farthest side of the GER (Gemmell and Nelson 1992). Taken together, our results suggest that co-regulating the Notch, Wnt, and SHH pathways is an effective way to induce more sensory progenitor cell proliferation that might lead to the regeneration of new inner HCs.

Conclusions
In summary, we report that inhibition of the Notch signaling pathway and activation of the Wnt signaling pathway can induce SC proliferation and new HC regeneration after the HCs have been damaged and that this effect is further promoted by upregulation of the SHH signaling pathway. Thus, co-regulation of the Notch, Wnt, and SHH signaling pathways might provide a new and effective method for inducing sensory progenitor cell proliferation and HC regeneration after HC damage in the mammalian inner ear and for restoring hearing and balance functions. Data availability The datasets generated during or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations
Ethical approval The care and use of animals were approved by the Institutional Animal Care and Use Committee of Fudan University in compliance with the NIH guidelines for the care and use of laboratory animals.

Conflict of interest The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.