The DN-CBRb transgenic developed in this study is viable and exhibits no developmental abnormalities. Genomic DNA from wild-type and DN-CBRb positive (DN-CBRb⁺) mice were submitted to standard PCR as described below. PCR products from both genotypes were purified and sequenced using an ABI Prism® 3100 genetic analyzer to confirm the mutation site. The rtTA (B6.Cg-Gt(ROSA)26Sortm1(rtTA*M2)Jae/J: Jaxmice stock number 006965) mouse line was purchased from the Jackson Laboratories. Double positive mice carrying both the DN-CBRb transgene and rtTA inducer allele (DN-CBRb⁺/ROSA-CAG-rtTA⁺) were used for experiments. Experimental negative control groups consisted of mice negative for CBRb and positive for rtTA (DN-CBRb^-/ROSA-CAG-rtTA⁺), as well as mice positive for CBRb and negative for rtTA (DN-CBRb⁺/ROSA-CAG-rtTA^-). Generation of the DN-CBRb transgenic line and all animal care and experiments associated with this study were approved by the Creighton University Institutional Animal Care and Use Committee (IACUC).

To test the possibility of using fusion with a lysosomal protease as an effective means of causing DN inhibition of RB1, we amplified a 1583bp-long Rb1 cDNA fragment corresponding to the amino acid region 369–896 of the RB1 protein (528 amino acids). This fragment was cloned between the EcoR1 and XbaI restriction sites of the pCS2+CB-Myc₆ vector (a gift from Dr. Marshal Horwitz, University of Washington; Li et al., 1996, 2000), to fuse it with the 1012bp long CB-Myc₆ construct (337 amino acids; Figure 1A). The fusion product resulted in a peptide of 865 amino acids with CB- Myc₆ at the N-terminus and Rb1 at the C-terminus. The fusion product has an approximate size of 108 kDa, which is slightly smaller than the endogenous RB1 protein (∼110 kDa). The CB- Myc₆-RB fusion fragment was PCR amplified and cloned into a pTet-splice vector between the SalI and EcoRV restriction sites to obtain the Tet(O)-DN-CBRB under tetracycline promoter (Figure 1B).

To stimulate in vitro overexpression and test the reversibility of the transgene, purified pTet-Splice plasmid containing the DN-CBRb cassette was co-transfected into HEK293 cells, with both DN-CBRb and the tetracycline-controlled transactivator pCMV-Tet3G vector (Clonetech Laboratories, Inc.), for 24 h. After that period, Dox media was removed and a subset of cells were washed from two to three times with PBS and incubated in Dox-free media for an additional 24 h to allow for inactivation of the transgene. Control samples consisted of co-transfected cells not treated with Dox, as well as HEK293 cells transfected only with the transactivator pCMV-Tet3G vector and treated with Dox for a 24-h period. Upon confirmation of effective Dox regulation of the DN-CBRb transgene, the entire cassette from the promoter up to the end of the SV40 polyA terminator element was excised out of the vector, gel purified, and used for the generation of the transgenic founder mice. The founder mice were generated using standard techniques at the Mouse Genome Engineering Core facility, University of Nebraska Medical Center. Pups were genotyped using a primer set specific to the CB-Rb1 fusion region, consisting of forward primer 5′ CTGTGGCATTGAATCAGAAATTGTGGCTGG 3′ and reverse primer 5′ GTACTTCTGCTATATGTGGCCATTACAACC 3′ (Figure 1B), which amplified a product approximately 401 bp (60 bp CB region + 341 Rb1 region; Figure 2). Several independent DN-CBRb founder lines were identified. Five of these lines were confirmed for germline transmission of the transgene. Two of those lines were further bred to establish breeding colonies and to test the transgene expression (i.e., lines 7 and 14). However, the results herein presented correspond to data collected from line 14 only. Reproductive problems with line 7 have precluded us from fully analyzing that line.

HEI-OC1 cells, kindly provided by Dr. Fedrico Kalinec (David Geffen School of Medicine, UCLA, Los Angeles, CA, USA), were maintained at 33°C with 10% CO2 (at permissive conditions) with 10% DMEM (#11965-084, GIBCO-BRL). 10,000 cells were plated on a 96-well plate and allowed to incubate overnight. On the following day transient transfection of DN-CBRB and rtTA construct was done using lipid based transfection reagent DharmaFECT Duo (T-2010-03). HEI-OC1 transfection rate was calculated based with pmR-ZsGreen1 (Clontech), wherein the transfected cells exhibited green fluorescence. 2 ug of pmR-ZsGreen1 was transfected in parallel, in a separate well. 1 ug/mL Dox was provided for experimental samples. Plasmid expression was measured 24 h after transfection and transfection rate was calculated by counting the number of GFP positive cells over the total cells labeled with Hoechst. A transfection rate of ∼70% was estimated. Of note, the treatment of untransfected HEI-OC1 cells with Dox did not affect RB1 protein expression or cell proliferation. Cell proliferation was assessed by CyQuant NF cell proliferation kit (Invitrogen, C35007). Cell proliferation assay was performed 48 h after transfection, according to manufacturer’s protocol. Briefly, the cell culture media was removed and 100 μl of 1X dye binding solution was added into each microplate well. The microplate was incubated for additional 1 h at 37°C. The fluorescence intensity for each sample was measured using fluorescence microplate reader (FLUOstar OPTIMA, BMG Labtech), with excitation at 485 nm and emission detection at 530 nm. To quantify any potential cell death following transfection, HEI-OC1 cells were trypsinized and counted. 0.1 × 10⁶ cells were plated on a coverglass in a 12-well plate and incubated overnight at 37°C. On the following day, cells were transfected with the DN-CBRb and rtTA constructs using DharmaFECT Duo (T-2010-03), lipid based transfection reagent. Positive control samples consisted of untransfected HEI-OC1 cells treated with 1 μM Staurosporine for 4 h. After 48 h of incubation, HEI-OC1 cells were analyzed for detection of active caspases according to the manfacturer’s protocol (Millipore, APT523). Briefly, 1X CaspaTag reagent solution was added and the cells were incubated at 37°C for 1 h. 5 ug/ml DAPI was used to label the nuclei of the dying cells for 10 min at room temperature. The cells were washed twice with 1X wash buffer and fixed using the supplier provided fixative. Thereafter, the coverslips were mounted and observed under the epifluorescence microscope (Nikon Eclipse 80i).

Whole-mount ISH detecting DN-CBRb mRNA was performed on Dox-treated (experimental) and untreated (control) P21 DN-CBRb⁺/ROSA-CAG-rtTA⁺ mice cochleae as previously described (Barritt et al., 2012). An approximately 600 nucleotides long riboprobe was generated by in vitro transcription of PCR products derived from DN-CBRb cDNA using primer 5′GGCTGGCAGCCAACTCTTGGAACCTTGACTGG. Reacted tissues were whole mounted in glycerol to be observed under the microscope or embedded on OTC media prior to frozen sectioning on a microtome cryostat. Whole mount and sections were imaged by differential interference contrast microscopy on a Nikon Eclipse 80i microscope. Sections of the cochlea and spiral ganglion neurons were counterstained with Nuclear Fast Red (Vector Laboratories, H3403).

Total RNA was isolated from dissected cochleae using RNeasy kit (Qiagen) according to the of the manufacturer’s instructions. Up to 20 μg of total RNA were treated with RNase-free DNase (Turbo DNfree; Ambion), and RNA concentration and quality were assessed on a Nanodrop spectrophotometer and Agilent Bioanalyzer, respectively. 250 ng of RNA were reverse transcribed using MultiScribe reverse transcriptase and random primers (Applied Biosystems). Semi-quantitative PCR assay was done in triplicate for each sample using SYBR green (Applied Biosystems, StepOne plus system). A corresponding amplification for each sample was performed with template without reverse transcriptase enzyme and used as control to rule out any possible genomic contamination. Amplification was performed using primers set specific to the CB-Rb1 fusion region, consisting of forward primer: 5′ CTGTGGCATTGAATCAGAAATTGTGGCTGG-3′ and reverse primer 5′TACTTCTGCTATATGTGGCCATTACAACC-3′, for 40 cycles. The relative quantitation of mRNA abundance was normalized to endogenous β-Actin using StepOne Software (Applied Biosystems; version 2.1). T-tests were performed on the normalized gene expression values to determine whether differences were statistically significant. P < 0.05 was considered significant.

To enable transgene expression, the DN-CBRb mice were bred to the ROSA-CAG-rtTA tetracycline inducer line. Genotyping of the rtTA construct was performed as described at the Jackson Laboratories website (http://jaxmice.jax.org/strain/006965.html). Double-positive (DN-CBRb⁺/ROSA-CAG-rtTA⁺) and negative controls (DN-CBRb^-/ROSA-CAG-rtTA⁺, DN-CBRb⁺/ROSA-CAG-rtTA^-) mice were treated with 2 mg/ml Dox in drinking water (post-weaning) or through their lactating mothers (pre-weaning). Alternative Dox doses have been tested (e.g., lower or higher than 2 mg/ml). However, as the method relies on the volume of water consumed by the animals, lower doses proved to be less efficient, resulting on a high variability on the phenotype. Such variability was attributed to a decrease on the volume of water consumed by some animals, and consequently smaller Dox absorption. When higher than 2 mg/ml was used, animals tended to drink less water (and absorb less Dox) due to Dox’s bitter taste. Further experiments with injectable Dox at concentrations higher than 2 mg/mL showed no differences from what we normally observe when providing 2 mg/mL Dox in the drinking water. Hence, we proceeded with Dox in drinking water. At the end of the Dox-treatment, several tissues were harvested (e.g., cochlea, eye, liver, heart, lung, and kidney) from DN-CBRb⁺/ROSA-CAG-rtTA⁺ and DN-CBRb⁺/ROSA-CAG-rtTA^- control mice and either stored in RNAlater® stabilization solution (Ambion) for subsequent RNA or protein extraction or kept in 4% paraformaldehyde (PFA) for histological analysis.

The tissues collected in RNA later and stored at -80°C were homogenized in Ripa lysis buffer (Thermo scientific #89901) with protease inhibitor (Thermo scientific #88664). Homogenization was done using Omni Homogenizers. Lysates obtained from tissues or cells were centrifuged at 14,000 rpm for 20 min at 4°C. The supernatant was used for protein estimation using the Lowrey method (BioRad DC protein assay kit #500–0112). Thereafter, 20 μg of protein was resolved on 10% SDS-PAGE and proteins were then transferred to PVDF membranes in a Bio-Rad TransBlot apparatus according to the manufacturer’s instructions at 100 V for 90 min. The proteins were transferred to PVDF membrane (Millipore Immobilon-P IPVH304FO) at 100 V for 90 min using a Biorad wet-tank apparatus. After incubation in blocking solution, the PVDF membrane was blocked in 5% blocking solution for 2 h at room temperature and probed overnight at 4°C with anti-RB1 (antibody against total RB1 protein; Abcam, ab6075), anti- active caspase 3 (Millipore, AB3623), anti -c-myc (Sigma, c3956) and anti β-actin (Santa Cruz, sc-69879) overnight at 4°C. The membranes were then washed and incubated in appropriate HRP-conjugated secondary antibodies, anti-rabbit (Santa Cruz, sc-2054) and anti-mouse (Santa Cruz, sc-358923) for 1 h at RT. After washing, peroxidase-bound protein bands were visualized by chemiluminescence using ECL substrate (Pierce, Rockford, IL, USA).

Whole mount immunohistochemistry was performed as previously described (Rocha-Sanchez et al., 2011). Immunohistological staining of Myosin VIIa (M7a; Proteus Biosciences) and active capsase 3 (Millipore, AB3623) was performed on DN-CBRb⁺/ROSA-CAG-rtTA⁺ and DN-CBRb^-/ROSA-CAG-rtTA⁺ formalin-fixed whole-mount preparations of dissected cochlear neurosensory epithelia. Tissue pieces (apex, middle, and base) were block/permeabilized with 5% NGS/0.1% Tween 20 at room temperature for 2–3 h, incubated with a primary antibody for 48 h in blocking buffer, and washed three times with PBS. Samples were then labeled with Alexa 568-conjugated secondary antibodies (1:500; Invitrogen), washed with PBS, counterstained with DAPI (5 ug/ml) for 2 h at room temperature. The cochleae were then mounted using Prolong anti-fade (Invitrogen) and imaged using a Leica TCS SP8 MP confocal microscope. Cell counting was performed on 200X images obtained from three different regions (apex, middle, and base) of the DN-CBRb⁺/ROSA-CAG-rtTA⁺ and DN-CBRb^-/ROSA-CAG-rtTA⁺ (control) cochleae. Statistical analysis between the two groups was done using paired t-test. Values are represented as ±SEM. P < 0.05 was considered significant.

F-actin was stained with Alexa 488 conjugated phalloidin. Briefly, formalin-fixed cochleae were washed with PBS three times and permeabilized with 0.25% Triton X-100 for 10 min. Thereafter, the cochlear tissue was labeled with Alexa 488-phalloidin (1: 200; Sigma–Aldrich) for 30 min. After three washes with PBS, the specimens were counterstained with DAPI, and imaged using Leica TCS SP8 MP confocal microscope.

To label mitotically active cells, a single, subcutaneous injection of the thymidine analog 5-ethynyl-2′-deoxyuridine (EdU; 50 mg/kg) in DMSO was administered to DN-CBRb⁺/ROSA-CAG-rtTA⁺ and DN-CBRb⁺/ROSA-CAG-rtTA^- mice 4 h before tissue harvesting. EdU incorporation into DNA of whole-mount cochlea was detected using the Click-iT EdU Alexa 488 Fluor Imaging kit (Invitrogen) and double stained with DAPI following the manufacturer’s instructions and experimental procedures previously described (Kaiser et al., 2009; Rocha-Sanchez et al., 2011). Samples were imaged using a Leica TCS SP8 MP confocal microscope.

To overcome the inherent problems associated with complete and permanent Rb1 deletion in mice, we sought to develop a DN version of the RB1 protein that would not only inhibit RB1 activity, regardless of the presence of functional endogenous RB1, but also bind to and trigger proteasome-mediated degradation of the protein complexes that normally associate with RB1. To this end, we modified a previously published system (Li et al., 1996, 2000) consisting of the fusion of the preprocathepsin B protease associated with most of the entire pocket domain and C-terminal region of Rb1. The resultant CB-myc6-RBf1ΔC (DN-CBRb) cassette has a carboxyl-terminal deletion between residues 369 and 896 (Li et al., 1996, 2000). To avoid the complications associated with constitutive Rb1 elimination, the DN-CBRb cassette was cloned under the tetracycline inducer Tet(O) promoter as described in materials and methods (Figures 1A,B). In vitro testing of the construct was performed in OC derived HEI-OC1 cells by transient transfection. Expression of the fusion protein was detected by Western blot with anti-RB1 antibody, as described below.

To test the effectiveness of Dox-regulated DN-CBRb activation and the reversibility of the system, HEK 293 cells were co-transfected with purified pTet-Splice plasmid containing the CB-Myc₆-Rb1 construct and the pCMV-Tet3G vector. Addition of Dox to the culture media significantly inhibited the expression of the endogenous RB1 protein (Figure 3). Following a 24-h period, Dox treatment was withdrawn and a subset of those Dox-treated HEK 293 cells was provided with a Dox-free media for an additional 24 h. Consistent with the reversible nature of the system, RB1 expression was restored in those cells after Dox removal from the culture media (Figure 3). Noteworthy, no changes in RB1 expression were observed on co-transfected HEK 293 cells not treated with Dox or on Dox-treated HEK 293 cells transfected with the pCMV-Tet3G vector only (Figure 3).

To investigate whether the inhibition in RB1 expression leads to increase in cell proliferation, HEI-OC1 cells were co-transfected with pTet-Splice-DN-CBRb and pCMV-Tet3G vectors and submitted or not (control group) to Dox treatment, as described above. Increase in proliferation was quantified by measuring the DNA content in both experimental and control groups, as well as in untransfected HEI-OC1 cells. Consistent with the activation of the DN-CBRb construct, the results demonstrated that co-transfection with pTet-Splice-DN-CBRb and pCMV-Tet3G vectors, followed by Dox-induced gene expression, led to a modest, but significant increase in DNA content compared to the control groups, which is an indirect measure of increase in cell number (Figure 4). In contrast, no significant changes in DNA content were observed between transfect HEI-OC1 cells not treated with Dox and untransfected HEI-OC1 (Figure 4). Next, to assess the possibility of cell death following activation of the DN-CBRb construct, the protein levels of active caspase 3 were measured by immunoblotting. Positive control consisted of HEI-OC1 cells treated with 1 μM Staurosporine. Overall, no significant changes in active caspase 3 protein levels were observed between untransfected control (treated and untreated with Dox) and experimental (transfected/Dox-untreated and transfected/Dox-treated) cells even after 48 h of transfection (Figure 5A). These results were recapitulated by in situ caspase detection on untransfected controls and experimental HEI-OC1 cells (Figures 5B,C). In sharp contrast, significant cell death was observed on the Staurosporine-treated untransfected HEI-OC1 cells (positive control; Figure 5D).

To investigate the functionality of the CB-myc6-Rb1 construct in vivo, the pTet-splice vector containing the DN-CBRb cassette was digested with the NotI enzyme to release the transgenic cassette, including the promoter and terminator elements. Gel-purified DNA was used to generate the transgenic mice using standard pronuclear injection techniques. DN-CBRb transgenic lines were further bred to the ROSA-CAG-rtTA tetracycline inducer line, which enables ubiquitous expression of DN-CBRb upon Dox administration. Litters from mating of DN-CBRb to ROSA-CAG-rtTA mice were genotyped as described in material and methods. DN-CBRb⁺/ROSA-CAG-rtTA⁺ and DN-CBRb^-/ROSA-CAG-rtTA⁺ were Dox-induced and analyzed for transgene expression. Ten independent DN-CBRb founder lines were identified. Five of these lines were confirmed for germline transmission of the transgene and used as founders for further breeding. Currently four of those lines are cryopreserved. The results presented here correspond to one single line of DN-CBRb mouse. To examine the in vivo feasibility of the method, we bred the DN-CBRb line with the tetracycline inducer ROSA-CAG-rtTA mice. DN-CBRb⁺/ROSA-CAG-rtTA⁺ mice did not exhibit any deficits and were further used for transgene induction experiments. Furthermore, to test for any potential leaky expression of the DN-CBRb transgene, we measured RB1 protein expression in DN-CBRb⁺/ROSA-CAG-rtTA⁺ mice in the absence or presence of Dox. Supporting a tight regulation of the transgene by Dox, a significant decrease in RB1 protein expression was observed in DN-CBRb⁺/ROSA-CAG-rtTA⁺ mice fed with Dox, but not in the untreated DN-CBRb⁺/ROSA-CAG-rtTA⁺ mice (Figure 6). Given the evidences of specific Dox regulation of DN-CBRb⁺/ROSA-CAG-rtTA⁺, added to the fact that no leaky transgene expression was observed on any other genotype combination, either in the presence or absence of Dox, we chose to use DN-CBRb^-/ROSA-CAG-rtTA⁺ (western blotting and histology) and DN-CBRb⁺/ROSA-CAG-rtTA^- (RT-PCR) as negative controls.

To determine the length of time needed to activate the ROSA-CAG-rtTA inducer line and elicit the DN-CBRb activation, DN-CBRb⁺/ROSA-CAG-rtTA⁺ mice and DN-CBRb^-/ROSA-CAG-rtTA⁺ mice were divided into three different groups and administered Dox in drinking water for 3 (Group 1), 7 (Group 2), and 10 (Group 3) consecutive days. Immediately after treatment, protein was extracted from postnatal (P) day 36 mice cochleae and analyzed by western blot using mouse anti-RB1 and c-myc hexameric antibodies (Figures 7A–C). Consistent with an initial accumulation of RB1 protein, quantification of RB1-Dox dose kinetics revealed a more than 20-fold increase in RB1 protein in cochleae of double-positive mice compared to the negative control samples (Figures 7A–D). Such accumulation may reflect the detection of both DN-CBRb and native RB1 subunits, prior to their association and lysosomal degradation. This notion was further supported by the observation that DN-CBRb⁺/ROSA-CAG-rtTA⁺ cochleae from groups 2 and 3 displayed steadily less RB protein than the control samples (Figure 7B,D). Analyses of liver and heart biopsies confirmed the pattern of RB1-Dox dose kinetics described above (data not shown). Of note, no significant differences in RB1 expression were observed in mice treated with Dox for periods longer than 10 consecutive days (data not shown). Hence, all treatments in subsequent experiments were performed on a 10-day Dox treatment.

To investigate the reversibility of the DN-CBRb transgene in vivo, P10 DN-CBRb⁺/ROSA-CAG-rtTA⁺ mice were fed with Dox in drinking water for 10 consecutive days. Controls consisted of untreated DN-CBRb⁺/ROSA-CAG-rtTA⁺ mice (Group 1; -Dox). Following that period, animals were euthanized either immediately (Group 2; +Dox) or 5 days after Dox withdraw (Group 3; ±Dox). Cochleae were dissected and analyzed for RB1 expression (Figure 8). Compared to the control untreated group, RB1 expression was visibly downregulated on group 1 (Figure 8). On another hand, supporting the reversibility of the system, RB1 expression on group 3 was increased compared to group 2. These results are in agreement with previously shown in vitro analysis (Figure 3).

To assess any potential tissue-specific variation on the efficiency of the Dox-mediated DN-CBRb activation, DN-CBRb⁺/ROSA-CAG-rtTA⁺ and age-matched DN-CBRb⁺/ROSA-CAG-rtTA^- mice were treated with Dox for 10 consecutive days. Cochleae, lungs, heart, and eye biopsies were dissected from experimental and control groups at P18 and subjected to western blotting with antibody against RB1 (Figure 9A). As expected, an inter-tissue variation on RB1 expression levels was observed in the DN-CBRb⁺/ROSA-CAG-rtTA^- group, with retina and cochleae showing higher RB1 concentrations (Figure 9A). Consistent with the notion of a possible interaction between the wild-type RB1 and the DN-CBRb transgene, endogenous RB1 expression was consistently decreased in every one of the DN-CBRb⁺/ROSA-CAG-rtTA⁺ mice tissues analyzed (Figure 8A). In spite of some inter-individual variations on the residual amount of RB1 protein after transgene activation, RB1 expression was visibly and significantly downregulated compared to the control animals (Figure 8A,B). These results were in agreement with our quantitative RT-PCR analyses in eye, heart and cochleae of Dox-treated DN-CBRb⁺/ROSA-CAG-rtTA⁺ and age-matched Dox-treated DN-CBRb⁺/ROSA-CAG-rtTA^- mice, which showed significant upregulation of the DN-CBRb transcript in DN-CBRb⁺/ROSA-CAG-rtTA⁺ tissues, but not in DN-CBRb⁺/ROSA-CAG-rtTA^- (Figure 9C). Moreover, RT-PCR analyses not only confirmed the inducibility of the system, by showing no transgene upregulation in the DN-CBRb⁺/ROSA-CAG-rtTA^- mice, but also demonstrated an increased concentration of the transgene in CBRb⁺/ROSA-CAG- rtTA⁺ animals.

Endogenous Rb1 expression is subjected to temporal and spatial variations (Hamel et al., 1993; Moore et al., 1996; Mantela et al., 2005). The present results collected from four different tissues from P18 mice combined with our previous set of experiments (Figure 7A–D) on P36 mice cochleae, support both the temporal and spatial efficiency of the DN-CBRb transgene activation, as well as the utilization of the DN-CBRb mouse model in a variety of studies looking into manipulating RB1 at different postnatal time points.

To investigate cell-specific expression of the DN-CBRb transgene, Dox-treated and untreated (control) P21 DN-CBRb⁺/ROSA-CAG-rtTA⁺ mice cochleae were submitted to ISH with transgene-specific riboprobes. Supporting efficient activation and tight regulation of the transgene, strong cytoplasmic and nuclear signal was detected at the IHCs region, including inner pillar cells and border cells, throughout the length of the cochlea (Figure 10A–E). Likewise, hybridization signal was observed on supernumerary cells along the length of the cochlea (Figure 10B,C–E), suggesting efficient transmission of the transgene to the daughter cells. No hybridization signal was observed on the control samples (data not shown). Light hybridization signal was also observed in the cytoplasma of outer HCs, Deiters’ cells, outer pillar cells, and spiral ganglion neurons (Figure 10D,F). Cytoplasmic detection of the transgene further evidences efficient transport of the DN-CBRb mRNA from the cells’ nuclei to be translated into active proteins. Noteworthy, mosaic transgene expression, which is part of the intrinsic nature of transgenic constructs, was also observed on DN-CBRb expression (Figure 10A–F). Such phenomenon is likely the result of incomplete recombination, a common occurrence associated with transgenic constructs. Therefore, incomplete recombination associated with the inherent mosaic expression of rtTA inducible systems may explain, at least in part, the patchy pattern of supernumerary cells observed on DN-CBRb⁺/ROSA-CAG-rtTA⁺ mouse.

To better understand the possible consequences of temporally regulated RB1 inhibition in the inner ear, cochleae of DN-CBRb⁺/ROSA-CAG-rtTA⁺ and DN-CBRb^-/ROSA-CAG-rtTA⁺ mice at P10 and P28 were dissected and submitted to immunohistochemistry with antibody against the HC marker Myosin VIIa (M7a). To date, a total of 40 ears (five mice per genotype × 2 time points) have been analyzed. While no changes have been observed in the cochleae of DN-CBRb^-/ROSA-CAG-rtTA⁺ mice, supernumerary M7a-positive cells were observed at the IHCs’ region of DN-CBRb⁺/ROSA-CAG-rtTA⁺ cochleae at the two time points analyzed (Figures 11A–K). Counting of those supernumerary cells was performed and the results compared between the different turns of the cochlea (i.e., apex, middle, and base) and time points by t-test (Figure 11L). No statistical differences were observed on the concentration of supernumerary cells between the middle and basal turns at either time points. Nevertheless, comparison between the numbers of supernumerary cells within the combined middle and basal turns and the apical turn was found to be significant (p < 0.05). Likewise, DN-CBRb⁺/ROSA-CAG-rtTA⁺ mouse at P10 showed significantly more supernumerary cells (p < 0.05) than at P28. These results are in agreement with the expression pattern of the endogenous Rb1 gene in the mouse cochlea (Mantela et al., 2005), showing a base-to-apex expression gradient, which is downregulated during postnatal development. Of note, no significant changes were observed in OHCs. Next, to start gathering information on the morphology of those supernumerary cells, we looked for the presence and morphology of stereocilia by using phalloidin labeling on DN-CBRb⁺/ROSA-CAG-rtTA⁺ cochlea at P10. As whole, no visible differences were observed on stereocilia morphology, density, and orientation between supernumerary and regular IHCs at P10 (Figure 12A–C). Additional analyses are currently underway to assess functionality and maturation of those newly generated HCs.

Unscheduled proliferation of inner ear cells upon complete or conditional RB1 deletion generally leads to massive cell death (Mantela et al., 2005; Sage et al., 2005, 2006). In the absence of any detectable apoptotic-like features in Dox-treated DN-CBRb⁺/ROSA-CAG-rtTA⁺ cochlea, we performed immunohistochemical analysis using antibody specific to detect any potential active caspase 3 activity in those mice cochleae at P10. Cochleae treated with Staurosporine were used as a positive control. No positive caspase 3 labeling was observed, except for a single positive nucleus at the basal turn of one of the ten cochleae analyzed (data not shown). Likewise, EdU labeling of DN-CBRb⁺/ROSA-CAG-rtTA⁺ cochlea did not reveal any detectable signs of proliferation within the OC (Figure 12D). However EdU-positive cells were still observed in the greater epithelial ridge (Figure 12D), suggesting that proliferation of the cells giving rise to the supernumerary HCs took place earlier during the 10 days of Dox treatments. Additional analyses of DN-CBRb⁺/ROSA-CAG-rtTA⁺ mice cochleae submitted to shorter periods of Dox treatment is likely to shed light on the timing when proliferation is taking place. No EdU-positive cells were observed in the control DN-CBRb^-/ROSA-CAG-rtTA⁺ mice ears (data not shown).