Methods and compositions for prevention and treatment of hearing loss

Abstract

The present disclosure provides preventing and / or treating hearing loss in an individual in need thereof comprising administering a composition comprising one or more viral vectors comprising a nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein. The present disclosure also provides method for preventing and / or treating hearing loss in an individual in need thereof comprising administering an effective amount of COL18A1, SPARC, and IGFBP2 proteins to the inner ear of the individual.

Description

[0001]Docket No.: 33612-20007.40 / 00633-0396WO1 METHODS AND COMPOSITIONS FOR PREVENTION AND TREATMENT OF HEARING LOSS CLAIM OF PRIORITY This application claims the benefit of U.S. Provisional Patent Application Serial No. 63 / 684,369, filed on August 17, 2024. The entire contents of the foregoing are incorporated herein by reference. STATEMENT OF FEDERALLY SPONSORED RESEARCH This invention was made with government support under Grant no.531206 and Grant no.530984, awarded by Department of Defense- Congressionally Directed Medical Research Programs. The government has certain rights in the invention. SEQUENCE LISTING This application contains a Sequence Listing that has been submitted electronically as an XML file named 00633-0396WO1_SL_ST26.xml. The XML file, created on August 15, 2025, is 30,224 bytes in size. The material in the XML file is hereby incorporated by reference in its entirety. FIELD OF INVENTION Provided herein are methods of preventing and / or treating hearing loss in an individual comprising administering a composition comprising one or more viral vectors comprising one or more nucleic acids, together encoding endostatin protein, SPARC protein, and IGFBP2 protein, wherein the one or more viral vectors is administered to the inner ear of the individual. Also provided herein are methods of preventing and / or treating hearing loss in an individual comprising administering an effective amount of endostatin, SPARC, and IGFBP2 proteins to the inner ear of the individual. BACKGROUND OF THE INVENTION Hearing loss (HL) is one of the most common forms of sensory deficit affecting over 36 million people in the US. Noise-induced hearing loss (NIHL) affects over 9 million people whereas age-related hearing loss (ARHL) affects 30% and 50% of 60 to 75-year-olds, respectively. Hearing loss disproportionately affects human patients. Hearing loss, due to the loss of ability to detect sound, is generally permanent. There are different degrees of hearing loss, Docket No.: 33612-20007.40 / 00633-0396WO1 including mild, moderate, severe and profound, measured by ABR (Auditory Brainstem Response) threshold shift. Hearing impairment can also manifest as deficiencies in word recognition, especially in noisy environments. The mechanisms underlying different types of hearing loss are different. The damage to and death of hair cells, the inner ear sensory cells that detect sounds, is one of the major causes of permanent hearing loss. The loss of synapses, termed synaptopathy, between inner hair cells and spiral ganglion neurons (SGN) is likely the underlying cause of word recognition deficiency. Hearing functions through the detection of mechanical vibrations by inner ear sensory hair cells, which release neurotransmitters that are absorbed by the auditory spiral ganglion neurons, which generate electric signals and pass them into the brain. The cells (including hair cells and ganglion neurons) and their structures (stereocilia of hair cells, synapses between hair cells and SGN, and neurites of SGN) along the path of sound detection and transmission can be damaged by noise, ototoxic drugs, viral infection, gene mutations, and aging, which ultimately lead to hearing impairment. Noise-induced hearing loss (NIHL) and age-related hearing loss (ARHL) are the two main categories of hearing loss. Cochlear synaptopathy, characterized by the loss of synaptic connections between hair cells and the peripheral axons of spiral ganglion neurons without hair cell or neuronal death, is believed to be the earliest sign of damage in noise- and age-related hearing impairments. This condition leads to difficulties in understanding speech in noisy environments and contributes to complex phenomena such as tinnitus and hyperacusis. No effective therapies are currently available for cochlear synaptopathy. Accordingly, there is an unmet need for therapies that can treat and prevent hearing loss. Provided herein are methods that meet such needs. Specifically, the invention described in this application focuses on promoting axon outgrowth of the auditory neurons in the injured cochlea to restore neuronal functions. BRIEF SUMMARY Provided herein are methods of preventing and / or treating hearing loss in an individual in need thereof comprising administering a composition comprising one or more viral vectors comprising one or more nucleic acids encoding a COL18A1 protein, a SPARC protein, and an IGFBP2 protein, wherein the one or more viral vectors is administered to the inner ear of the individual. In some embodiments, the one or more viral vectors is selected from the group consisting of a lentivirus vector, an adenovirus vector, an adeno-associated Docket No.: 33612-20007.40 / 00633-0396WO1 virus (AAV) vector, a herpes simplex virus (HSV) vector, and a retrovirus vector. In some embodiments, the one or more viral vectors comprises an AAV vector. In some embodiments, the nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein is each operably linked to an expression control element. In some embodiments, the expression control element is a promoter sequence. In some embodiments, the promoter sequence is selected from the group consisting of a constitutive promoter sequence, an inducible promoter sequence, and a tissue-specific promoter sequence. In some embodiments, the promoter sequence is a constitutive promoter sequence. In some embodiments, the constitutive promoter sequence is selected from the group consisting of MYO15A promoter sequence, a MYO7A promoter sequence, a MYO6 promoter sequence, a STRC promoter sequence, a Pou4f3 promoter sequence, a OTOF promoter sequence, and a SLC26A5 promoter sequence. In some embodiments, the constitutive promoter sequence is a CMV promoter sequence. In some embodiments, the promoter sequence is a tissue-specific promoter sequence. In some embodiments, the tissue-specific promoter sequence is selected from the group consisting of a stria vascularis cell-specific promoter sequence, a hair cell-specific promoter sequence, a supporting cell-specific promoter sequence, and a spiral ganglion neuron-specific promoter sequence. In some embodiments, the expression control element comprises an enhancer sequence; optionally, wherein the enhancer sequence is a CMV enhancer sequence. In some embodiments, the nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein are located on the same viral vector. In some embodiments, the viral vector comprises a nucleic acid encoding a cleavable linker between each of the sequence encoding COL18A1 protein, the sequence encoding SPARC protein, and the sequence encoding IGFBP2 protein. In some embodiments, the cleavable linker is a self-cleaving peptide selected from the group consisting of F2A, P2A, T2A, and E2A. In some embodiments, the cleavable linker is a P2A peptide. In some embodiments, the P2A peptide has 1, 2, or 3 amino acid substitutions or insertions or deletions compared to the amino acid sequence of SEQ ID NO: 14. In some embodiments, three separate proteins are produced from translation of the one or more nucleic acids encoding COL18A1 protein, SPARC protein, and IGFBP2 protein. In some embodiments, the AAV vector comprises an Igκ leader sequence (SEQ ID NO: 13) at the 5’ end of the nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein. Docket No.: 33612-20007.40 / 00633-0396WO1 In some embodiments, the COL18A1, SPARC, and IGFBP2 proteins are secreted by inner hair cells or outer hair cells of the individual. Also provided herein are methods for preventing and / or treating hearing loss in an individual in need thereof comprising administering an effective amount of COL18A1, SPARC, and IGFBP2 proteins or an effective amount of an mRNA encoding the COL18A1 protein, an mRNA encoding the SPARC protein and an mRNA encoding the IGFBP2 protein to the inner ear of the individual. In some embodiments, the effective amount of an mRNA encoding the COL18A1 protein, an mRNA encoding the SPARC protein and an mRNA encoding the IGFBP2 protein are administered in a lipid nanoparticle (LNP). In some embodiments, the COL18A1 protein is endostatin. In some embodiments, the COL18A1 protein has at least 90%, 95%, 97%, or 98% identity to, or comprises, the amino acid sequence of SEQ ID NO: 2 or 4. In some embodiments, the SPARC protein has at least 90%, 95%, 97%, or 98% identity to, or comprises, the amino acid sequence of SEQ ID NO: 6 or 8. In some embodiments, the IGFBP2 protein has at least 90%, 95%, 97%, or 98% identity to, or comprises, the amino acid sequence of SEQ ID NO: 10 or 12. In some embodiments, the COL18A1, SPARC, and IGFBP2 proteins are administered at about a 1:1:1 ratio. In some embodiments, the hearing loss is age-related hearing loss. In some embodiments, the hearing loss is noise-induced hearing loss. In some embodiments, the compositions comprising a viral vector as described herein, or the COL18A1, SPARC, and IGFBP2 proteins or mRNA, are administered directly to the inner ear of the individual. In some embodiments, the compositions comprising a viral vector as described herein, or the COL18A1, SPARC, and IGFBP2 proteins or mRNA, is administered through the round window membrane of an ear of the individual. In some embodiments, the composition or the COL18A1, SPARC, and IGFBP2 proteins is administered to both ears of the individual. In some embodiments, the methods result in regeneration of the axons of spiral ganglion neurons and / or synapses between cochlear hair cells and peripheral nerve fibers, and / or enhances axon outgrowth of spiral ganglion neurons in the individual’s cochlea, thereby restoring hearing function. In some embodiments, the method increases neuronal fiber length in the outer hair cell region of the individual’s cochlea. In some embodiments, the methods enhance survival of spiral ganglion neurons. In some embodiments, the method enhances neurite-inner hair cell binding. In some embodiments, the methods further comprise determining the Wave I amplitude of the individual at a frequency before and after the individual has received the Docket No.: 33612-20007.40 / 00633-0396WO1 composition or the COL18A1, SPARC, and IGFBP2 proteins. In some embodiments, the methods result in a higher Wave I amplitude at a frequency in the individual who is administered the composition or the COL18A1, SPARC, and IGFBP2 proteins as compared to the Wave I amplitude at the same frequency in an individual who is not administered the composition or the COL18A1, SPARC, and IGFBP2 proteins. In some embodiments, the Wave I amplitude in response to 20-100 dB SPL stimulus is higher in the individual who is administered the composition or the COL18A1, SPARC, and IGFBP2 proteins as compared to the Wave I amplitude in response to 20-100 dB SPL stimulus in an individual who is not administered the composition or the COL18A1, SPARC, and IGFBP2 proteins. In some embodiments, the methods further comprise determining speech recognition of the individual before and after the individual has received the composition or the COL18A1, SPARC, and IGFBP2 proteins. In some embodiments, the method improves speech recognition of the individual as compared to speech recognition of the individual prior to administration of the composition or the COL18A1, SPARC, and IGFBP2 proteins. In some embodiments, the method improves hearing function after 4 weeks. In some embodiments, the method improves hearing function after 12 weeks. In some embodiments, the individual is human. Also provided herein are one or more nucleic acids encoding COL18A1, SPARC, and IGFBP2. In some embodiments, the one or more nucleic acids encoding COL18A1 is one or more nucleic acids encoding endostatin. Also provided herein are vectors comprising the nucleic acids described herein. In some embodiments, the vector is a viral vector. In some embodiments, the vector is an AAV vector. Also provided herein are one or more LNPs comprising the one or more nucleic acids as described herein. Also provided herein are pharmaceutical compositions comprising an effective amount of COL18A1, SPARC, and IGFBP2 proteins, or mRNA encoding the COL18A1, SPARC, and IGFBP2 proteins. In some embodiments, the COL18A1 is endostatin. In some embodiments, the COL18A1 protein has at least 90%, 95%, 97%, or 98% identity to, or comprises, the amino acid sequence of SEQ ID NO: 2, 17 or 18. In some embodiments, the SPARC protein has at least 90%, 95%, 97%, or 98% identity to, or comprises, the amino acid sequence of SEQ ID NO: 6 or 8. In some embodiments, the IGFBP2 protein has at least 90%, 95%, 97%, or 98% identity to, or comprises, the amino acid sequence of SEQ ID NO: 10 or 12. Docket No.: 33612-20007.40 / 00633-0396WO1 It is to be understood that one, some, or all of the properties of the various embodiments described herein may be combined to form other embodiments of the present invention. These and other aspects of the invention will become apparent to one of skill in the art. These and other embodiments of the invention are further described by the detailed description that follows. DESCRIPTION OF THE FIGURES The present application can be understood by reference to the following description taken in conjunction with the accompanying figures. FIGs.1A-1F show that single-cell RNA sequencing identified putative cell types across the sensory epithelium from injured and regenerated mouse cochleae. FIG.1A: Pictures illustrating the axon (TUJ) outgrowth to the sensory epithelium from the existing spiral ganglion neurons (SGNs) in the hair cell (PVALB) regeneration model. PVALB: Parvalbumin; TUJ: Tuj1; SE: Sensory Epithelium. FIG.1B: Schematic diagram depicting the experimental setup for collection and processing of cochlear explants using the 10X Genomics Chromium Single Cell 3' Gene Expression workflow. FIGs.1C-1D: UMAP plots showing the putative cell clusters from the integrated t-SNE analysis of the RNAseq data of the Dox and Dox Atoh1 groups pictured in FIG.1B. FIG.1E: Heatmap highlighting key marker genes that were used to infer putative identities compared to all other cell clusters. The rows correspond to the top 20 genes most selectively upregulated in individual clusters (p < 0.01, LogFC > 0.25), and the columns show individual cells ordered by cluster (Kolliker’s organ cells (KO), unclassified-supporting cells (uSC), interdental cells (IdC), Claudius cells / outer sulcus cells (CCOS), hair cells-like cells (HCLC), Reissner’s membrane cells (RMC), Hensen’s cells (HeC), hair cells (HC) and Deiter’s cells (DC)). Scale represents log-transformed and normalized counts scaled to a maximum of 2.5 per row. FIG.1F: Violin plots showing normalized and log-transformed expression values for the two representative marker genes for each cell type by comparison with all other cell clusters. To explore the cellular composition of the samples, principal component analysis was applied on variably expressed genes across all cells and identified 8 distinct clusters: Kolliker’s organ cells (KO); Unclassified cells (UnC); Pillar cells (PC); Claudius cells / outer sulcus cells (CCOS); Endothelial cells (EC); Hensen’s cells (HeC); Hair cells (HC); Deiter’s cells (DC). FIGs.2A-2C show that differentially expressed genes following Atoh1 induction (Dox_Atoh1 vs. Dox) identified putative candidate secreted genes. FIG.2A: Venn diagram depicting global patterns of differentially expressed genes across each sample reveals genes Docket No.: 33612-20007.40 / 00633-0396WO1 that are highly enriched within Dox or Dox_Atoh1-treated samples (MAST analysis, FDR adjusted p value <0.01). The seventy-two shared genes were genes that failed to reach significance and found to be expressed in both samples. FIG.2B: DAVID v6.8 (david.ncifcrf.gov / ) was applied using default settings to identify enriched gene ontology terms based on the 440 up-regulated genes after Atoh1 induction in Dox_Atoh1 sample. The dark grey rectangle points out secreted genes. FIG.2C: Violin plots depicting expression levels of the secreted 30 genes after Atoh1 induction in Dox_Atoh1 vs. Dox samples. FIGs.3A-3D show the process of narrowing down the number of candidate secreted genes for axon outgrowth. FIG.3A: A literature review of the list of secreted genes (see FIG.2C) showed that nine out of 30 genes likely contribute to neurodevelopmental processes such as neuronal migration, regeneration or axon pathfinding. FIG.3B: In order to verify single cell RNA seq data, mRNA expression levels of the 9 secreted genes were analyzed by RT-qPCR in Dox_Atoh1 sample. FIG.3C: Individual treatment with purified secreted proteins in P8 spiral ganglion explant cultures. Here, BDNF (50 ng / ml) was used as positive control along with 7 proteins extracted from the single cell RNA seq data. FIG.3D: Quantitation of neurite regrowth in P8 spiral ganglion explant cultures at day 2 in vitro. Neurite regrowth is quantified by counting NFH-immunofluorescent fibers. HuD antibody stains neuron cell body; DAPI stains DNA / nucleus. None of the test molecules were able to promote axon outgrowth compared to positive control BDNF. Experimental data were presented as mean ± SD. Comparisons between selected groups were made using Student’s t test. ***, p < 0.001 was considered statistically significant. B=BDNF, C=endostatin, S=SPARC, O=OLFML2B, F=FBLN2, L=LGALS1, I=IGFBP-2, G=GPC1. FIGs.4A-4D show co-expression analysis of nine genes by STRING database predicted putative interactions among the molecules. FIG.4A: The STRING analysis highlights the genes found to be co-expressed according to protein–protein association networks. FIG.4B: The co-expression profile among genes as listed Dual, Triple and Quadruple combinations suggest novel, functionally relevant protein-protein interactions. FIG.4C: Coexpression analysis-guided Dual treatment with purified secreted proteins (1mg / ml each) in P8 spiral ganglion explant cultures showed that four different dual combinations did not exhibit significant axon outgrowth compared to BDNF (50 ng / ml) as positive control at day 2 in vitro. FIG.4D: Quantitation of neurite regrowth in P8 spiral ganglion explant cultures. Neurite regrowth is quantified by counting NFH- immunofluorescent fibers. HuD antibody stains neuron cell body. None of the test molecules were able to promote axon outgrowth compared to positive control BDNF. Experimental data Docket No.: 33612-20007.40 / 00633-0396WO1 were presented as mean ± SD. Comparisons between selected groups were made using Student’s t test. **, p < 0.01 was considered statistically significant. FIGs.5A-5C show testing of triple and quadruple combinations of test molecules in P8 SGN explant culture. FIG.5A: STRING coexpression analysis-guided triple treatment with purified secreted proteins (1mg / ml each) in P8 spiral ganglion explant cultures showed that CSI (endostatin + SPARC + IGFBP2) was comparable to positive control BDNF (50 ng / ml) at day 2 in vitro. FIG.5B: Coexpression analysis-guided quadruple treatment with purified secreted proteins (1mg / ml each) in P8 spiral ganglion explant cultures showed that C-I-S-F (endostatin + IGFBP2 + SPARC + FBLN2) was comparable to positive control BDNF (50 ng / ml) at day 2 in vitro. FIG.5C: Quantitation of neurite outgrowth in P8 spiral ganglion explant cultures. Neurite regrowth is quantified by counting NFH- immunofluorescent fibers. HuD antibody stains neuron cell body. CSI showed superior effect over BDNF and other combinations on axon outgrowth by the number of axons per explant. Experimental data were presented as mean ± SD. Comparisons between selected groups were made using Student’s t test. *, p < 0.05; ***, p < 0.001 were considered statistically significant. FIGs.6A-6B show that an antibody blocking assay confirmed the axon-outgrowth effect of CSI molecules on P8 SGN explant cultures. FIG.6A: Concentration-dependent inhibition by a mixture of specific antibodies against triple treatment of endostatin, SPARC, and IGFBP2 (CSI) showed significant decline in the number of axons per SGN explant at day 2 in vitro. Antibodies used in this experiment as follow: Mouse (COL18A1) Endostatin Antibody (R&D System #AF570-SP), Mouse SPARC Antibody (R&D System #AF942-SP), Mouse IGFBP-2 Antibody (R&D System #AF797-SP). FIG.6B: Quantitation of inhibition of neurite outgrowth in P8 spiral ganglion explant cultures. Neurite outgrowth is quantified by counting NFH-immunofluorescent fibers. HuD antibody stains neuron cell body. Experimental data were presented as mean ± SD. Comparisons between selected groups were made using Student’s t test. ***, p < 0.001 was considered statistically significant. FIGs.7A-7B show partial recovery in the number of synapses with CSI treatment following NK-induced injury. FIG.7A: In order to cause excitotoxic injury, the sensory epithelium explants were exposed, after 24 hours in culture (for equilibration), to glutamate receptor agonists NMDA (0.5 mM) and kainic acid (0.5 mM) for 2 hours (NK treatment). After NK treatment, the explants were maintained for an additional 24 hours in culture medium containing BSA (negative control, 1mg / ml) or CSI protein mixture (1mg / ml each). Control samples were cultured in basal culture medium (DMEM / F12 containing 1% FBS and Docket No.: 33612-20007.40 / 00633-0396WO1 N2, B27 supplements) without NK pre-treatment. NK-induced (NMDA and Kainic acid; 0.5 mM each) excitotoxicity for 2 hours caused a significant decline in number of CtBP2 puncta, and resulted in the degeneration of synapses between inner hair cells and peripheral nerve fibers in vitro P14 explant sensory epithelium culture. Cochlear explant cultures were exposed for 2 h in vitro to NK and allowed 24 hours for recovery by BSA (negative control) or CSI. While addition of BSA (1mg / ml) to the culture medium following NK treatment did not improve the synaptic recovery, addition of CSI (1mg / ml each) to the culture medium during the post-NK recovery increased the number of synapses (CtBP2-positive puncta) compared to NK (2h) sample. FIG.7B: Error bars show SEM. Statistically significant difference was found in NK + CSI sample, indicating that addition of CSI protein mixture contributes to synaptic recovery after NK-induced synaptic injury. Experimental data were presented as mean ± SD. Comparisons between selected groups were made using Student’s t test. *, P-value < 0.05; **, P-value < 0.01; ***, P-value < 0.001 were considered statistically significant. FIGs.8A-8B show the construction and characterization of AAV vectors. FIG.8A: Schematic representation of CSI construct flanked by two inverted terminal repeats (ITRs). Igk leader sequence was inserted into 5’ of endostatin to achieve proper secretion of the molecules. In this experiment, the self-cleaving property of the 2A peptide inserted between each coding sequences was used to express all three proteins from a single AAV vector. FIG. 8B: Immunoblot analysis of cell lysate and conditioned culture medium (for 72 hrs) from HEK293 cells transduced with recombinant adeno-associated virus expressing CSI molecules under CMV promoter (AAV2.CMV.Igk.C-S-I.HA). The secreted molecules were analyzed by immunoblots using anti- endostatin antibody, anti-SPARC antibody, anti-IGFBP-2 antibody, and all the target proteins were clearly detected in cell lysate and culture medium (supernatant), while no signal observed in non-infected cell lysate and control medium. FIGs.9A-9C show AAV-mediated triplet inner ear injection rescuing hearing and synaptopathy after noise damage. FIG.9A: A schematic diagram to show the experimental design. Adult CBA / CaJ mice were exposed to noise of 97 dB 1-20 kHz for 2 hours, followed by round window injection of AAV2-Triplet-HA 24 hours later. Hearing was tested 3 weeks after noise exposure. Control was uninjected contralateral ears. FIG.9B: In control ears, ABR (acoustic brain response) thresholds were elevated particularly at 5.66 and 45.24 kHz compared to pre-noise exposed inner ears. In AAV-Triplet-HA injected inner ears, ABR thresholds were significantly reduced at 5.66 and 45.24 kHz, an indication of hearing recovery. FIG.9C: Compared to uninjected contralateral ears, wave 1 amplitudes that Docket No.: 33612-20007.40 / 00633-0396WO1 represent neuronal activities were significantly improved in AAV-Triplet-HA injected ears across all frequencies, an indication of repair of synapses and recovery of neuronal activities. FIGs.10A-10C show CSI protein inner ear injection rescuing synaptopathy after noise damage. FIG.10A: A schematic diagram to show the experimental design. Adult CBA / CaJ mice were exposed to noise of 97 dB 1-20 kHz for 2 hours, followed by round window injection of CSI protein mixture hours later. Hearing was tested 3 weeks after noise exposure. Control was uninjected contralateral ears. FIG.10B: The noise of TTS did not induce ABR or DPOAE threshold shifts in injected or uninjected ears, compared to pre-noise exposed ears. FIG.10C: Compared to uninjected contralateral ears, wave 1 amplitudes that represent neuronal activities were significantly improved in protein injected ears across all frequencies, an indication of repair of synapses and recovery of neuronal activities. FIGs.11A-11B show that AAV2-CSI shows efficacious treatment in restoration of neural activities in mice with age-related hearing loss. FIG.11A: A schematic diagram showing the experimental design. Adult (6-week-old) C57BL / 6J mice were injected with AAV2-CSI with hearing studied 4 and 12 weeks later. FIG.11B: Compared to uninjected contralateral control ears, injected ears showed significantly greater wave 1 amplitudes in all frequencies at 4 weeks post injection and in most frequencies at 12 weeks post injection. FIGs.12A-12C show that individual endostatin does not rescue wave 1 amplitude deficits by noise exposure. FIG.12A: A schematic diagram to show the experimental design. Adult CBA / CaJ mice were exposed to noise of 97 dB 1-20 kHz for 2 hours, followed by round window injection of AAV2-Col18a1-HA 24 hours later. Hearing was tested 3 weeks after noise exposure. Control was uninjected contralateral ears. FIG.12B: The noise of TTS did not induce ABR or DPOAE threshold shifts in injected or uninjected ears, compared to pre-noise exposed ears. FIG.12C: The wave 1 amplitudes were reduced by noise exposure in injected and uninjected ears. These data show the lack of rescue effect by individual endostatin. FIG.13 is a schematic picture illustrating the HC regeneration / neurites outgrowth and the blockade assay by CSI antibodies. The total lench of the TuJ1+ neuronal fibers in the outer hair cell region (OHCr) were counted. FIG.14A is a schematic diagram illustrating the experimental procedure for the adult mouse cochlea culture in vitro. FIG.14B shows that outer hair cells died, and the neuronal fibers retracted from the outer hair cell region in the adult mouse cochlea in vitro culture system. Quantification and comparison of the Tuj1+ neuronal fiber length in the outer hair cell region of the apical turn of the cultured cochleae. Ordinary one-way ANOVA analysis Docket No.: 33612-20007.40 / 00633-0396WO1 with multiple comparisons test, error bar, mean ± SEM; n=4 for each group. OHCr: outer hair cell region; IHCr: inner hair cell region. Scale bars: 20 μm. FIGs.15A-15B show the blockade of the regrowth of the Tuj1+ neuronal fibers to the OHCr after HC regeneration in vitro. FIG.15A: A schematic diagram illustrating the experimental procedure of HC regeneration and neuronal fiber regrowth to the OHCr in vitro. FIG.15B: Upon the Cocktail (VLFsiFsiM) / Ad.Atoh1.mCherry treatment in cultured adult (P30) wildtype mouse cochlea, neuronal fibers were regrowth only in the no CSI antibodies treated samples shown by but not in the CSI treated samples. Quantification and comparison of regenerated HCs and Tuj1+ neuronal fiber lengths in the apical turn of the cultured cochleae among different groups. ****p < 0.001, Ordinary one-way ANOVA analysis with multiple comparisons test, Error bar, mean ± SEM; n=4 for each group. V: VPA; L: LiCl; F: Forskolin; siF: siFIR; siM: siMxi1. OHCr: outer hair cell region; IHCr: inner hair cell region. Scale bars: 20 μm. FIGs.16A-16F show CSI treatment preserving the synapses degradation and triggers neurite-IHC binding in vitro. FIG.16A: A schematic figure of the in vitro culture workflow. Treatment 1 and treatment 2 will be labelled as T1 and T2 afterwards. FIG.16B: The representative confocal figures of the inner hair cell after treatment 1. CSI protein treatment protect the IHC synapses from degenerating. FIG.16C: The quantification of synapse per IHC, represented by CTPB2 dots. BDNF and NT-3 were used as positive controls. About 45- 65 IHCs from at least 5 cochlear were counted. FIG.16D: The representative confocal figures of the inner hair cell after treatment 2. The CSI treatment preserved but could not regenerate the IHC synapse. FIGs.16E-16F: The average quantification of synapse per IHC, represented by CTPB2 dots (FIG.16E) and the neurites connected to each IHC, represented by TUJ1 label (FIG.16F). BDNF and NT-3 used as positive controls. More than 30 IHCs from at least 5 cochlear were counted. The color demonstrated the mean of all individual counts. Scale bar: 10um. Error bars indicate SEM, P-values were derived from two-way ANOVA, ****p < 0.0001. KA=kainic acid. FIGs.17A-17C show CSI preserves synapses in treatment 1. The IHCs come from apical (FIG.17A), middle (FIG.17B) and basal turn (FIG.17C), respectively. FIGs.17D- 17E show quantification of synapse per IHC, represented by CTPB2 dots on middle (FIG. 17D) and basal turn (FIG.17E), respectively. About 45-65 IHCs from at least 5 cochlear were counted. Scale bar: 10um. Error bars indicate SEM, P-values were derived from two- way ANOVA, ****p < 0.0001. KA=kainic acid. Docket No.: 33612-20007.40 / 00633-0396WO1 FIGs.18A-18B show that CSI preserves synapses in treatment 2 and is more efficacious than other neurotrophic factors. The IHCs come from middle (FIG.18A) and basal turn (FIG.18B), respectively. FIG.18C shows the quantification of synapse per IHC, represented by CTPB2 dots on apical, middle and basal turn, respectively. About 45-65 IHCs from at least 5 cochlear were counted. Scale bar: 10um. Error bars indicate SEM, P-values were derived from two-way ANOVA, ****p < 0.0001, **p < 0.01, *p < 0.05. KA=kainic acid. Plots for each of the apical, middle, and basal groups repeat in the same order as the figure legend: CTRL, KA 24H, KA 48H, KA+CSI T2, KA+BDNF T2, KA+NT-3 T2. FIGs.19A shows representative confocal figures of the inner hair cell after kainic acid (KA) damage, CSI treatment 1 together with antibodies. The IHCs come from middle and basal turn, respectively. The synapse preservation function was dependent on the concentration of antibodies. FIG.19B shows the quantification of synapse per IHC, represented by CTPB2 dots on apical, middle and basal turn, respectively. About 30-40 IHCs from at least 3 cochlear were counted. Scale bar: 10um. Error bars indicate SEM, P-values were derived from two-way ANOVA, ****p < 0.0001, ***p < 0.001. Plots for each of the apical, middle, and basal groups repeat in the same order as listed in the figure legend: CSI + IgG (1μg / ml), CSI + Ab mix (0.125 μg / ml), CSI + Ab mix (0.25 μg / ml), CSI + Ab mix (0.5 μg / ml), CSI + Ab mix (1 μg / ml). FIG.20 shows a schematic diagram showing the procedure for ouabain administration to induce neuropathy in adult mice followed by CSI treatment. FIGs.21A-21B shows results for neurofiber regeneration in ouabain-treated cochlea. FIG.21A shows immunohistochemistry labeling of cochlear apical region tissue with labeling of neurofiber (NF), myosin (MYO7A), and synapsis (CtBP2) for mice 2 weeks after ouabain treatment, 12 weeks after ouabain treatment, 12 weeks after treatment with ouabain then CSI and, wild type (WT). Scale bar: 50um. FIG.21B shows quantification of nerve fibers in mice treated with ouabain 12 weeks post injection (WPI), mice treated with ouabain followed by CSI after two weeks at 10 weeks post injection of the CSI, and wild type mice (WT). ****p < 0.0001. DETAILED DESCRIPTION The present disclosure is based at least in part on the discovery of a combination of three proteins that unexpectedly induces axon outgrowth from existing spiral ganglion neurons to treat cochlear synaptopathy. It is also demonstrated that the combination described Docket No.: 33612-20007.40 / 00633-0396WO1 herein enhance the survival of spiral ganglion neurons and preserves synapses from chemically induced synapse damage. The present application shows for the first time that a combination of three secreted proteins (endostatin, SPARC, IGFBP2; together referred to as “CSI”) induces axon outgrowth. This combination is shown to enhance the survival of spiral ganglion neurons and preserving the synapses in the cochlear hair cell in vitro. To assess the therapeutic efficacy of these molecules in vivo, the secreted proteins were expressed in the existing hair cells via local administration of adeno-associated virus (AAV) expressing CSI mRNA or delivered CSI proteins into the inner ear in a mouse model of noise-induced hearing loss. This approach resulted in the restoration of wave I amplitude, which is indicative of synaptic recovery and functional hearing restoration. In summary, the findings offer novel insights into therapeutic strategies for cochlear synaptopathy associated with both noise-induced and age-related hearing loss. It is contemplated that these findings also inform the development of treatments for other auditory disorders. The following description sets forth exemplary methods, parameters and the like. It should be recognized, however, that such description is not intended as a limitation on the scope of the present disclosure but is instead provided as a description of exemplary embodiments. Definitions The term "conservative amino acid substitution" refers to the interchangeability in proteins of amino acid residues having similar side chains. For example, a group of amino acids having aliphatic side chains consists of glycine, alanine, valine, leucine, and isoleucine; a group of amino acids having aliphatic-hydroxyl side chains consists of serine and threonine; a group of amino acids having amide-containing side chains consists of asparagine and glutamine; a group of amino acids having aromatic side chains consists of phenylalanine, tyrosine, and tryptophan; a group of amino acids having basic side chains consists of lysine, arginine, and histidine; and a group of amino acids having sulfur- containing side chains consists of cysteine and methionine. Exemplary conservative amino acid substitution groups are: valine -leucine -isoleucine, phenylalanine -tyrosine, lysine-arginine, alanine- valine, and asparagine -glutamine. “Encoding” refers to the inherent property of specific sequences of nucleotides in a polynucleotide, such as a gene, a cDNA, or an mRNA, to serve as templates for synthesis of other polymers and macromolecules in biological processes having either a defined sequence of nucleotides (i.e., rRNA, tRNA and mRNA) or a defined sequence of amino acids and the Docket No.: 33612-20007.40 / 00633-0396WO1 biological properties resulting therefrom. Thus, a gene encodes a protein if transcription and translation of mRNA corresponding to that gene produces the protein in a cell or other biological system. The term “exogenous” indicates that the nucleic acid or polypeptide is part of, or encoded by, a recombinant nucleic acid construct, or is not in its natural environment. For example, an exogenous nucleic acid can be a sequence from one species introduced into another species, i.e., a heterologous nucleic acid. Typically, such an exogenous nucleic acid is introduced into the other species via a recombinant nucleic acid construct. An exogenous nucleic acid can also be a sequence that is native to an organism and that has been reintroduced into cells of that organism. In addition, stably transformed exogenous nucleic acids typically are integrated at positions other than the position where the native sequence is found. The term “expression” as used herein is defined as the transcription and / or translation of a particular nucleotide sequence. “Expression vector” refers to a vector comprising a recombinant polynucleotide comprising expression control sequences operatively linked to a nucleotide sequence to be expressed. An expression vector comprises sufficient cis-acting elements for expression; other elements for expression can be supplied by the host cell or in an in vitro expression system. Expression vectors include all those known in the art, such as cosmids, plasmids (e.g., naked or contained in liposomes) and viruses (e.g., lentiviruses, retroviruses, adenoviruses, and adeno-associated viruses) that incorporate the recombinant polynucleotide. As used herein, “COL18A1” refers to any and all COL18A1-associated nucleic acid or protein sequences and includes any sequence that is orthologous or homologous to, or has significant sequence similarity to, a COL18A1 nucleic acid or amino acid sequence derived from any animal including mammals (e.g., humans, mice). The term also includes homologs, orthologs, mutants, variants or fragments thereof. COL18A1 also includes all other synonyms that may be used to refer to the COL18A1 gene or the COL18A1 protein product of the gene (for example, synonyms for the COL18A1 gene include collagen type XVIII alpha 1 chain, KS, KNO, GLCC, and KNO1). The amino acid and nucleic acid sequences of mouse COL18A1 appear, for example, in the NCBI database under Gene ID: 12822 (e.g., NP_001393184.1, NP_001103461.1, NP_034059.2, NP_001393179.1, NP_001393180.1, NP_001393183.1, NP_001393181.1, NP_001393182.1). The amino acid and nucleic acid sequences of human COL18A1 appear, for example, in the NCBI database under Gene ID: 80781 (e.g., NP_085059.2, NP_569711.2, NP_001366429.1). COL18A1 refers to the full Docket No.: 33612-20007.40 / 00633-0396WO1 protein or a fragment thereof. In some embodiments, the active form of COL18A1 is a truncated form. Preferably, the COL18A1 is endostatin. As used herein “endostatin” refers to the biologically active, truncated form of the protein encoded by the COL18A1 gene. As used herein, “SPARC” refers to any and all SPARC -associated nucleic acid or protein sequences and includes any sequence that is orthologous or homologous to, or has significant sequence similarity to, a SPARC nucleic acid or amino acid sequence derived from any animal including mammals (e.g., humans, mice). The term also includes homologs, orthologs, mutants, variants or fragments thereof. SPARC also includes all other synonyms that may be used to refer to the SPARC gene or the SPARC protein product of the gene (for example, synonyms for the SPARC gene include secreted protein acidic and cysteine rich, ON, ONT, OI17, and BM-40). The amino acid and nucleic acid sequences of mouse SPARC appear, for example, in the NCBI database under Gene ID: 20692. The amino acid and nucleic acid sequences of human SPARC appear, for example, in the NCBI database under Gene ID: 6678. As used herein, “IGFBP2” refers to any and all IGFBP2-associated nucleic acid or protein sequences and includes any sequence that is orthologous or homologous to, or has significant sequence similarity to, a IGFBP2 nucleic acid or amino acid sequence derived from any animal including mammals (e.g., humans, mice). The term also includes homologs, orthologs, mutants, variants or fragments thereof. IGFBP2 also includes all other synonyms that may be used to refer to the IGFBP2 gene or the IGFBP2protein product of the gene (for example, synonyms for the IGFBP2 gene insulin like growth factor binding protein 2, IBP2, and IGF-BP53). The amino acid and nucleic acid sequences of mouse IGFBP2 appear, for example, in the NCBI database under Gene ID: 16008. The amino acid and nucleic acid sequences of human IGFBP2 appear, for example, in the NCBI database under Gene ID: 3485. Unless otherwise specified, a “nucleic acid sequence encoding” an amino acid sequence includes all nucleotide sequences that encode the same amino acid sequence. The phrase nucleotide sequence that encodes a protein or an RNA may also include introns. The term “operably linked” refers to positioning of a regulatory region and a sequence to be transcribed in a nucleic acid so as to influence transcription or translation of such a sequence. For example, to bring a coding sequence under the control of a promoter, the translation initiation site of the translational reading frame of the polypeptide is typically positioned between one and about fifty nucleotides downstream of the promoter. A promoter Docket No.: 33612-20007.40 / 00633-0396WO1 can, however, be positioned as much as about 5,000 nucleotides upstream of the translation initiation site or about 2,000 nucleotides upstream of the transcription start site. A promoter typically comprises at least a core (basal) promoter. A promoter also may include at least one control element, such as an enhancer sequence, an upstream element or an upstream activation region (UAR). The choice of promoters to be included depends upon several factors, including, but not limited to, efficiency, selectability, inducibility, desired expression level, and cell- or tissue-preferential expression. It is a routine matter for one of skill in the art to modulate the expression of a coding sequence by appropriately selecting and positioning promoters and other regulatory regions relative to the coding sequence. The terms “patient” or “individual” or “subject” are used interchangeably herein, and refers to a mammalian subject to be treated, such as a human subject or a murine subject. The term “percent sequence identity” or having “a sequence identity” refers to the degree of identity between any given query sequence and a subject sequence. The terms “pharmaceutically acceptable” (or “pharmacologically acceptable”) refer to molecular entities and compositions that do not produce an adverse, allergic or other untoward reaction when administered to an animal or a human, as appropriate. The term “pharmaceutically acceptable carrier,” as used herein, includes any and all solvents, dispersion media, coatings, antibacterial, isotonic and absorption delaying agents, buffers, excipients, hinders, lubricants, gels, surfactants and the like, that may be used as media for a pharmaceutically acceptable substance. The term “polynucleotide” is a chain of nucleotides, also known as a “nucleic acid”. As used herein polynucleotides include, but are not limited to, all nucleic acid sequences which are obtained by any means available in the art, and include both naturally occurring and synthetic nucleic acids. As used herein, the terms “nucleic acid sequence” and “polynucleotide” are used interchangeably throughout the specification and include complementary DNA (cDNA), linear or circular oligomers or polymers of natural and / or modified monomers or linkages, including deoxyribonucleosides, ribonucleosides, substituted and alpha-anomeric forms thereof, peptide nucleic acids (PNA), locked nucleic acids (LNA), phosphorothioate, methylphosphonate, and the like. Polynucleotides include, but are not limited to, all nucleic acid sequences which are obtained by any means available in the art, including, without limitation, recombinant means, i.e., the cloning of nucleic acid sequences from a recombinant library or a cell genome, using ordinary cloning technology and PCR, and the like, and by synthetic means. Docket No.: 33612-20007.40 / 00633-0396WO1 The terms “polypeptide,” “peptide,” and “protein” refer to compounds comprised of amino acid residues covalently linked by peptide bonds. A protein or peptide must contain at least two amino acids. Polypeptides include peptides or proteins comprising two or more amino acids joined to each other by peptide bonds. As used herein, the term refers to both short chains, which also commonly are referred to in the art as peptides, oligopeptides and oligomers, for example, and to longer chains, which generally are referred to in the art as proteins, of which there are many types. “Polypeptides” include, for example, biologically active fragments, substantially homologous polypeptides, oligopeptides, homodimers, heterodimers, variants of polypeptides, modified polypeptides, derivatives, fusion proteins, among others. The polypeptides include natural peptides, recombinant peptides, synthetic peptides, or a combination thereof. The term “transfected” or “transformed” or “transduced” means to a process by which exogenous nucleic acid is transferred or introduced into the host cell. A “transfected” or “transformed” or “transduced” cell is one which has been transfected, transformed or transduced with exogenous nucleic acid. The transfected / transformed / transduced cell includes the primary subject cell and its progeny. To “treat” a disease or disorder as the term is used herein, means to reduce the frequency or severity of at least one sign or symptom associated with the disease or disorder experienced by an individual. Treatment involves an observable beneficial effect of the treating the individual, including a reduction in severity of some or all signs or symptoms associated with the disease or disorder, a slower progression of the disease or disorder, an improvement in the overall health or well-being of the individual, or by other parameters that are specific to the particular disease or disorder. To “prevent” a disease or disorder, as used herein, refers to inhibiting, delaying, or reducing the severity of the onset of a disease or disorder or reducing the likelihood of developing at least one sign or symptom associated with the disease or disorder in an individual. As used herein, “variant” of polypeptides refers to an amino acid sequence that is altered by one or more amino acid residues. The variant may have “conservative” changes, wherein a substituted amino acid has similar structural or chemical properties (e.g., replacement of leucine with isoleucine). More rarely, a variant may have “nonconservative” changes (e.g., replacement of glycine with tryptophan). Analogous minor variations may also include amino acid deletions or insertions, or both. Guidance in determining which amino acid residues may be substituted, inserted, or deleted without abolishing biological activity Docket No.: 33612-20007.40 / 00633-0396WO1 may be found using computer programs known in the art, for example, LASERGENE software (DNASTAR). A “vector” is a composition of matter which comprises an isolated nucleic acid and which can be used to deliver the isolated nucleic acid to the interior of a cell. Examples of vectors include hut are not limited to, linear polynucleotides, polynucleotides associated with ionic or amphiphilic compounds, plasmids, and viruses. Thus, the term “vector” includes an autonomously replicating plasmid or a virus. The term is also construed to include non- plasmid and non-viral compounds which facilitate transfer of nucleic acid into cells, such as, for example, polylysine compounds, liposomes, and the like. Examples of viral vectors include, but are not limited to, adenoviral vectors, adeno-associated virus vectors, retroviral vectors, and the like. The vector can also include a regulatory region. The term “regulatory region” refers to nucleotide sequences that influence transcription or translation initiation and rate, and stability and / or mobility of a transcription or translation product. Regulatory regions include, without limitation, promoter sequences, enhancer sequences, response elements, protein recognition sites, inducible elements, protein binding sequences, 5′ and 3′ untranslated regions (UTRs), transcriptional start sites, termination sequences, polyadenylation sequences, nuclear localization signals, and introns. As used herein, the terms “a”, “an”, and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. Where a term is provided in the singular, the inventors also contemplate aspects of the invention described by the plural of that term. As used here, the term “about” refers to the usual error range for the respective value readily known to the skilled person in this technical field. Reference to “about” a value or parameter herein includes (and describes) embodiments that are directed to that value or parameter per se. As used herein, the term “and / or” is to be taken as specific disclosure of each of the two specified features or components with or without the other. Thus, the term “and / or” as used in a phrase such as “A and / or B” herein is intended to include “A and B,” “A or B,” “A” (alone), and “B” (alone). Likewise, the term “and / or” as used in a phrase such as “A, B, and / or C” is intended to encompass each of the following embodiments: A, B, and C; A, B, or C; A or B; A or C; B or C; A and B; A and C; B and C; A (alone); B (alone); and C (alone). Docket No.: 33612-20007.40 / 00633-0396WO1 As used herein, the terms “comprise,” “comprises,” “comprising,” “include,” “includes,” and “including” can be interchanged and are to be construed as at least having the features to which they refer while not excluding any additional unspecified features. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure is related. Methods of treatment and prevention Provided herein are methods for preventing and / or treating hearing loss in an individual in need thereof comprising administering a composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein, wherein the one or more viral vectors is administered to the inner ear of the individual. Also provided herein are methods for preventing and / or treating hearing loss in an individual in need thereof comprising administering an effective amount of endostatin, SPARC, and IGFBP2 proteins or mRNA encoding the COL18A1, mRNA encoding the SPARC, and mRNA eencoding the IGFBP2 proteins to the inner ear of the individual. Patients with hearing loss can be identified using standard hearing tests known in the art. In some embodiments, the present disclosure can be used to treat and / or prevent inner ear cell damage (e.g., cochlear cell or utricular cell damage) and associated disorders, such as hearing impairments, deafness, vestibular disorders, tinnitus (see, Kaltenbach et al. (2002) J NEUROPHYSIOL, 88(2).699-714s, the contents of which are herein incorporated by reference in their entirety), and hyperacusis (Kujawa et al. (2009) J. NEUROSCI. 29(45):14077-14085, the contents of which are herein incorporated by reference in their entirety). The individual may be deaf or have hearing loss for any reason, or as a result of any type of event. For example, an individual can be deaf or hard-of-hearing as a result of a traumatic event, such as a physical trauma to a structure of the ear. For example, prolonged exposures to concerts, airport runways, and construction areas can cause inner ear damage and subsequent hearing loss. In some embodiments, the hearing loss is noise-induced hearing loss (e.g., a hearing impairment resulting from exposure to loud sounds or noise, either in a single intense burst or over an extended period). In some embodiments, the hearing loss is occupational hearing loss (e.g., exposure a hearing impairment resulting from exposure to loud sounds or noise while working). In some embodiments, the individual is at risk of developing noise induced hearing loss (e.g., a construction worker or military service member). Docket No.: 33612-20007.40 / 00633-0396WO1 In some embodiments, the methods provided herein treat noise-induced hearing loss in an individual in need thereof. In some embodiments, a composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is administered after exposure to a sudden loud noise or prolonged exposure to loud noises. In some embodiments, an effective amount of endostatin, SPARC, and IGFBP2 proteins is administered after exposure to a sudden loud noise or prolonged exposure to loud noises. Any suitable method known in the art can be used to assess hearing (e.g., Wave I amplitude, ABR threshold, and / or DPOAE threshold). In some embodiments, the methods provided herein prevent noise-induced hearing loss in an individual in need thereof. In some embodiments, a composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is administered as a prophylactic against noise damage. In some embodiments, an effective amount of endostatin, SPARC, and IGFBP2 proteins is administered as a prophylactic against noise damage. In some embodiments, the composition or endostatin, SPARC, and IGFBP2 proteins is administered to the individual prior to noise exposure (e.g., a sudden loud noise or prolonged exposure to loud noises). In some embodiments, the method provides protection against noise-induced hearing loss in the individual following administration of the composition. In some embodiments, the hearing loss is age-related hearing loss (also referred to as presbycusis). In some embodiments, the age-related hearing loss is characterized by progressive damage and loss of hair cells (e.g., outer hair cells and / or inner hair cells) in the ear of the individual. In some embodiments, the age-related hearing loss is characterized by progressive loss of synapses between spiral ganglion neuron and the inner hair cells (IHCs) in the ear of the individual (e.g., cochlear synaptopathy). In some embodiments, the methods provided herein treat age-related hearing loss in an individual in need thereof. In some embodiments, a composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is administered to treat age-related hearing loss in an individual in need thereof. In some embodiments, or an effective amount of endostatin, SPARC, and IGFBP2 proteins is administered after exposure to treat age-related hearing loss in an individual in need thereof. In some embodiments, the method recovers hearing in the individual following administration of the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some Docket No.: 33612-20007.40 / 00633-0396WO1 embodiments, the method recovers hearing in the individual following administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some embodiments, the methods provided herein prevent age-related hearing loss in an individual in need thereof. In some embodiments, a composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is administered as a prophylactic against age-related hearing loss. In some embodiments, an effective amount of endostatin, SPARC, and IGFBP2 proteins is administered as a prophylactic against age-related hearing loss. In some embodiments, the method provides protection against age-related hearing loss in the individual following administration of the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the method provides protection against age-related hearing loss in the individual following administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some embodiments, the methods provided herein prevent age-related hearing loss in an individual in need thereof. In some embodiments, the method comprise administering an effective amount of an mRNA encoding a COL18A1 protein, an mRNA encoding a SPARC protein, and an mRNA encoding a IGFBP2 protein. In some embodiments, the mRNA encoding the COL18A1 protein, mRNA encoding SPARC protein, and mRNA encoding IGFBP2 protein are administered in one or more lipid nanoparticles (LNP). In some embodiments, the mRNA encoding the COL18A1 protein, the mRNA encoding SPARC protein, and the mRNA encoding IGFBP2 protein are each administered in one or more LNPs. In some embodiments, the mRNA encoding COL18A1 is an mRNA encoding endostatin. In some embodiments, an effective amount of the mRNA encoding a COL18A1 protein, mRNA encoding a SPARC protein, and mRNA encoding a IGFBP2 protein are administered as a prophylactic against age-related hearing loss. In some embodiments, the method provides protection against age-related hearing loss in the individual following administration of the composition comprising LNPs comprising one or more mRNAs encoding COL18A1 protein, SPARC protein, and IGFBP2 protein. In some embodiments, the method provides protection against age-related hearing loss in the individual following administration of the composition comprising LNPs comprising one or more mRNAs encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the Docket No.: 33612-20007.40 / 00633-0396WO1 method provides protection against age-related hearing loss in the individual following administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some embodiments, the method improves hearing function in the individual in need of treatment for hearing loss following administration of the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the method improves hearing function in the individual in need of treatment for hearing loss following administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some embodiments the method improves hearing function in the individual after 4 weeks. In some embodiments the method improves hearing function in the individual after 12 weeks. Viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein In one aspect, provided herein are methods for preventing and / or treating hearing loss in an individual in need thereof comprising administering a composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein, the vectors can comprise a single nucleic acid encoding all three of the proteins, or can comprise a population of vectors comprising nucleic acids encoding one or two of the proteins, wherein the population includes vectors encoding all three of the proteins. In some embodiments, the one or more viral vectors is administered to the inner ear of the individual. In some embodiments, the one or more viral vectors is an AAV vector. Also provided herein are one or more nucleic acids encoding endostatin, SPARC, and IGFBP2. In some embodiments, the viral vector is selected from the group consisting of a lentivirus vector, an adenovirus vector, an adeno-associated virus (AAV) vector, a vesicular stomatitis virus (VSV) vector, a herpes simplex virus (HSV) vector, a vaccinia virus vector, a pox virus vector, an influenza virus vector, a respiratory syncytial virus vector, a parainfluenza virus vector, a foamy virus vector, oncolytic viruses, and a retrovirus vector. In some embodiments, the one or more viral vectors is selected from the group consisting of a lentivirus vector, an adenovirus vector, an adeno-associated virus (AAV) vector, a herpes simplex virus (HSV) vector, and a retrovirus vector. Docket No.: 33612-20007.40 / 00633-0396WO1 In some embodiments, the viral vector is an adeno-associated viral (AAV) vector. The AAV vector can comprise, for example AAV1, AAV2, AAV2G9, AAV3, AAV5, AAV6, AAV7, AAV8, AAV10, AAV6.2, AAV2.7m8, AAVrh10, AAVrh39, AAVrh43, AAV-DJ, AAVi.e., AAVanc80, AAVPHP.eB, AAVPhP.B, and AAVrh10. In some embodiments, the viral vector is an adeno-associated virus type 2 (AAV2) vector. In some embodiments, the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is located on one or more viral vectors. In some embodiments, the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is located on two different viral vectors. In some embodiments, the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is located on three different viral vectors. In some embodiments, the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is located on the same viral vector. In some embodiments, the method described herein for preventing and / or treating hearing loss in an individual in need thereof comprising administering a composition comprising one, two, or three viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein, wherein the one, two, or three viral vectors are administered to the inner ear of the individual. Exemplary human and mouse COL18A1, endostatin, SPARC, and IGFBP2 amino acid and nucleic acid sequences are provided in Table 1. COL18A1 / Endostatin COL18A1 encodes the alpha chain of type XVIII collagen. This collagen is one of the multiplexins, extracellular matrix proteins that contain multiple triple-helix domains (collagenous domains) interrupted by non-collagenous domains. A long isoform of the protein has an N-terminal domain that is homologous to the extracellular part of frizzled receptors. Proteolytic processing at several endogenous cleavage sites in the C-terminal domain results in production of endostatin, a potent antiangiogenic protein that is able to inhibit angiogenesis and tumor growth. In the present methods and compositions, either the full-length COL18A1 protein or the endostatin protein can be used. In some embodiments, the COL18A1 protein is a mouse COL18A1 protein. In some embodiments, the mouse COL18A1 protein comprises an amino acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 98%, or at least 99% identity to the amino acid sequence of SEQ ID NO: 16. In some embodiments, the mouse COL18A1 protein comprises an amino acid sequence with at least 90% identity to the amino acid sequence of SEQ ID NO: 16. In some embodiments, the Docket No.: 33612-20007.40 / 00633-0396WO1 mouse COL18A1 protein comprises the amino acid sequence of SEQ ID NO: 16, or a variant thereof having 20 or fewer insertions, deletions, or conservative amino acid substitutions, such as up to 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, or 1 insertions, deletions, or conservative substitutions. In some embodiments, the mouse COL18A1 protein comprises the amino acid sequence of SEQ ID NO: 16. Mouse COL18A1 comprises an endostatin-like domain, a collagen trimerization domain, a LPXTG-anchored collagen-like adhesin domain. In some embodiments, the COL18A protein comprises a domain that has at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or 100% sequence identity with an endostatin-like domain, a collagen trimerization domain, and / or a LPXTG-anchored collagen-like adhesin domain of SEQ ID NO: 16. In some embodiments, the nucleic acid sequence of a wildtype COL18A1 gene is identified by the Genbank accession number NC_000076.7 from nucleotide 76,888,013 to nucleotide 77,002,351 of chromosome 10 according to the GRCm39 assembly of the mouse genome. The wildtype COL18A1 gene comprises 48 exons. One or more mutations (e.g., substitutions, deletions, insertions) of the COL18A1 gene may occur in any one or any combination of the 48 exons, or in any intron or noncoding regions of the COL18A1 gene. In some embodiments, the amino acid sequence of a wildtype COL18A1 protein is identified by the Genbank accession number NP_001103461.1. In some embodiments, the amino acid sequence of a wildtype COL18A1 protein is identified by the Genbank accession number NP_001393179.1. In some embodiments, the amino acid sequence of a wildtype COL18A1 protein is identified by the Genbank accession number NP_001393180.1. In some embodiments, the amino acid sequence of a wildtype COL18A1 protein is identified by the Genbank accession number NP_001393181.1. In some embodiments, the amino acid sequence of a wildtype COL18A1 protein is identified by the Genbank accession number NP_001393182.1. In some embodiments, the amino acid sequence of a wildtype COL18A1 protein is identified by the Genbank accession number NP_001393183.1. In some embodiments, the amino acid sequence of a wildtype COL18A1 protein is identified by the Genbank accession number NP_001393184.1. In some embodiments, the amino acid sequence of a wildtype COL18A1 protein is identified by the Genbank accession number NP_034059.2. In some embodiments, the COL18A1 protein is a human COL18A1 protein. In some embodiments, the endostatin protein is a human endostatin protein. In some embodiments, the endostatin protein is the active form of the human endostatin protein. In some embodiments, the nucleic acid sequence encoding the human COL18A1 protein comprises a Docket No.: 33612-20007.40 / 00633-0396WO1 nucleic acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 98%, or at least 99% identity to the nucleic acid sequence of SEQ ID NO: 3. In some embodiments, the nucleic acid sequence encoding the human COL18A1 protein comprises a nucleic acid sequence with at least 90% identity to the nucleic acid sequence of SEQ ID NO: 3. In some embodiments, the nucleic acid sequence encoding the human COL18A1 protein comprises the nucleic acid sequence of SEQ ID NO: 3. In some embodiments, the human COL18A1 protein comprises an amino acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 98%, or at least 99% identity to the amino acid sequence of SEQ ID NO: 4. In some embodiments, the human COL18A1 protein comprises an amino acid sequence with at least 90% identity to the amino acid sequence of SEQ ID NO: 4. In some embodiments, the human COL18A1 protein comprises the amino acid sequence of SEQ ID NO: 4, or a variant thereof having 18 or fewer conservative amino acid substitutions. In some embodiments, the human COL18A1 protein comprises the amino acid sequence of SEQ ID NO: 4. In some embodiments, an altered (e.g., cleaved, truncated, fragment of COL18A1 protein) COL18A1 protein or biologically active fragments thereof retains at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or at least 95% of the biological activity of full-length, wild type protein. Human COL18A1 comprises an endostatin-like domain, a Laminin G domain, a collagen trimerization domain, a LPXTG-anchored collagen- like adhesin domain. In some embodiments, the COL18A1 comprises a domain that has at least at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or 100% sequence identity with the an endostatin-like domain, a Laminin G domain, a collagen trimerization domain, a LPXTG-anchored collagen-like adhesin domain of SEQ ID NO: 4. In some embodiments, the nucleic acid sequence of a wildtype COL18A1 gene is identified by the Genbank accession number NC_000021.9 from nucleotide 45,405,165 to nucleotide 45,513,720 of chromosome 21 according to the GRCh38.p14 assembly of the human genome. The wildtype COL18A1 gene comprises 43 exons. One or more mutations (e.g., substitutions, deletions, insertions) of the COL18A1 gene may occur in any one or any combination of the 43 exons, or in any intron or noncoding regions of the COL18A1 gene. In some embodiments, the amino acid sequence of a wildtype COL18A1 protein is identified by the Genbank accession number NP_001366429.1. In some embodiments, the amino acid sequence of a wildtype COL18A1 protein is identified by the Genbank accession number NP_085059.2. In some embodiments, the amino acid sequence of a wildtype COL18A1 protein is identified by the Genbank accession number NP_569711.2. Docket No.: 33612-20007.40 / 00633-0396WO1 As an alternative to the full-length COL18A1 protein, the endostatin peptide itself, i.e., the active form, can be used. In some embodiments, the endostatin protein is a mouse endostatin protein. In some embodiments, the nucleic acid sequence encoding the mouse endostatin protein comprises a nucleic acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 98%, or at least 99% identity to the nucleic acid sequence of SEQ ID NO: 1. In some embodiments, the nucleic acid sequence encoding the mouse endostatin protein comprises a nucleic acid sequence with at least 90% identity to the nucleic acid sequence of SEQ ID NO: 1. In some embodiments, the nucleic acid sequence encoding the mouse endostatin protein comprises the nucleic acid sequence of SEQ ID NO: 1. In some embodiments, the mouse endostatin protein comprises an amino acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 98%, or at least 99% identity to the amino acid sequence of SEQ ID NO: 2. In some embodiments, the mouse endostatin protein comprises an amino acid sequence with at least 90% identity to the amino acid sequence of SEQ ID NO: 2. In some embodiments, the mouse endostatin protein comprises the amino acid sequence of SEQ ID NO: 2, or a variant thereof having 20 or fewer insertions, deletions, or conservative amino acid substitutions, such as up to 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, or 1 insertions, deletions, or conservative substitutions. In some embodiments, the mouse endostatin protein comprises the amino acid sequence of SEQ ID NO: 2. In some embodiments, an altered (e.g., cleaved, truncated, fragment of COL18A1 protein) COL18A1 protein or biologically active fragments thereof retains at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or at least 95% of the biological activity of full-length, wild type protein. In some embodiments, the endostatin protein is a human endostatin protein. In some embodiments, the human endostatin is a fragment of the human COL18A1 protein. In some embodiments, the human endostatin protein comprises an amino acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 98%, or at least 99% identity to the amino acid sequence of SEQ ID NO: 17 or 18. In some embodiments, the human endostatin protein comprises an amino acid sequence with at least 90% identity to the amino acid sequence of SEQ ID NO: 17 or 18. In some embodiments, the human endostatin protein comprises the amino acid sequence of SEQ ID NO: 17 or 18, or a variant thereof having 18 or fewer conservative amino acid substitutions. In some embodiments, the human endostatin protein comprises the amino acid sequence of Docket No.: 33612-20007.40 / 00633-0396WO1 SEQ ID NO: 17 or 18. In some embodiments, human endostatin protein may comprise post- translational modification or variations dues to variation in proteolytic processing. In some embodiments, the endostatin protein is a human endostatin protein. In some embodiments, the human endostatin is a fragment of the human COL18A1 protein. In some embodiments, the human endostatin protein comprises an amino acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 98%, or at least 99% identity to the amino acid sequence of SEQ ID NO: 18. In some embodiments, the human endostatin protein comprises an amino acid sequence with at least 90% identity to the amino acid sequence of SEQ ID NO: 18. In some embodiments, the human endostatin protein comprises the amino acid sequence of SEQ ID NO: 18, or a variant thereof having 18 or fewer conservative amino acid substitutions. In some embodiments, the human endostatin protein comprises the amino acid sequence of SEQ ID NO: 18. In some embodiments, human endostatin protein may comprise post-translational modification or variations dues to variation in proteolytic processing. In some embodiments, the endostatin protein is a human endostatin protein. In some embodiments, the human endostatin is a fragment of the human COL18A1 protein. In some embodiments, the endostatin protein is the active form of the human endostatin protein. In some embodiments, the active form of the human endostatin protein comprises an amino acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 98%, or at least 99% identity to the amino acid sequence of SEQ ID NO: 17. In some embodiments, the active form of the human endostatin protein comprises an amino acid sequence with at least 90% identity to the amino acid sequence of SEQ ID NO: 17. In some embodiments, the active form of the human endostatin protein comprises the amino acid sequence of SEQ ID NO: 17, or a variant thereof having 18 or fewer conservative amino acid substitutions. In some embodiments, the active form of the human endostatin protein comprises the amino acid sequence of SEQ ID NO: 17. In some embodiments, active form of the human endostatin protein may comprise post-translational modification or variations dues to variation in proteolytic processing. SPARC SPARC encodes a cysteine-rich acidic matrix-associated protein. The encoded protein is required for the collagen in bone to become calcified but is also involved in extracellular matrix synthesis and promotion of changes to cell shape. Docket No.: 33612-20007.40 / 00633-0396WO1 In some embodiments, the SPARC protein is a mouse SPARC protein. In some embodiments, the nucleic acid sequence encoding the mouse SPARC protein comprises a nucleic acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 98%, or at least 99% identity to the nucleic acid sequence of SEQ ID NO: 5. In some embodiments, the nucleic acid sequence encoding the mouse SPARC protein comprises a nucleic acid sequence with at least 90% identity to the nucleic acid sequence of SEQ ID NO: 5. In some embodiments, the nucleic acid sequence encoding the mouse SPARC protein comprises the nucleic acid sequence of SEQ ID NO: 5. In some embodiments, the mouse SPARC protein comprises an amino acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 98%, or at least 99% identity to the amino acid sequence of SEQ ID NO: 6. In some embodiments, the mouse SPARC protein comprises an amino acid sequence with at least 90% identity to the amino acid sequence of SEQ ID NO: 6. In some embodiments, the mouse SPARC protein comprises the amino acid sequence of SEQ ID NO: 6, or a variant thereof having 20 or fewer conservative amino acid substitutions, such as up to 19, 18, 1716, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, or 1 insertions, deletions, or conservative amino acid substitutions. In some embodiments, the mouse SPARC protein comprises the amino acid sequence of SEQ ID NO: 6. Mouse SPARC comprises an EF-hand extracellular calcium-binding motif, e.g., a follistatin-like SPARC domain. In some embodiments, the SPARC protein comprises a domain that has at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or 100% sequence identity with the EF-hand extracellular calcium-binding motif, and / or follistatin-like SPARC domain of SEQ ID NO:6. In some embodiments, the nucleic acid sequence of a wildtype SPARC gene is identified by the Genbank accession number NC_000076.7 from nucleotide 55,284,985 to nucleotide 55,310,906 of the reverse strand of chromosome 11 according to the GRCm39 assembly of the mouse genome. The wildtype SPARC gene comprises 10 exons. One or more mutations (e.g., substitutions, deletions, insertions) of the SPARC gene may occur in any one or any combination of the 10 exons, or in any intron or noncoding regions of the SPARC gene. In some embodiments, the amino acid sequence of a wildtype SPARC protein is identified by the Genbank accession number NP_001277746.1. In some embodiments, the amino acid sequence of a wildtype SPARC protein is identified by the Genbank accession number NP_033268.1. In some embodiments, the SPARC protein is a human SPARC protein. In some embodiments, the nucleic acid sequence encoding the human SPARC protein comprises a Docket No.: 33612-20007.40 / 00633-0396WO1 nucleic acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 98%, or at least 99% identity to the nucleic acid sequence of SEQ ID NO: 7. In some embodiments, the nucleic acid sequence encoding the human SPARC protein comprises a nucleic acid sequence with at least 90% identity to the nucleic acid sequence of SEQ ID NO: 7. In some embodiments, the nucleic acid sequence encoding the human SPARC protein comprises the nucleic acid sequence of SEQ ID NO: 7. In some embodiments, the human SPARC protein comprises an amino acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 98%, or at least 99% identity to the amino acid sequence of SEQ ID NO: 8. In some embodiments, the human SPARC protein comprises an amino acid sequence with at least 90% identity to the amino acid sequence of SEQ ID NO: 8. In some embodiments, the human SPARC protein comprises the amino acid sequence of SEQ ID NO:8, or a variant thereof having 20 or fewer conservative amino acid substitutions, such as up to 19, 18, 1716, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, or 1 insertions, deletions, or conservative amino acid substitutions. In some embodiments, the human SPARC protein comprises the amino acid sequence of SEQ ID NO: 8. Human SPARC comprises an EF-hand extracellular calcium-binding motif and a follistatin-like SPARC domain. In some embodiments, the SPARC protein comprises a domain that has at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or 100% sequence identity with the EF- hand extracellular calcium-binding motif, and / or follistatin-like SPARC domain of SEQ ID NO:8. In some embodiments, the nucleic acid sequence of a wildtype embodiments SPARC gene is identified by the Genbank accession number NC_000005.10 from nucleotide 151,661,096 to nucleotide 151,686,915 of the reverse strand of chromosome 5 according to the GRCh38.p14 assembly of the human genome. The wildtype SPARC gene comprises 10 exons. One or more mutations (e.g., substitutions, deletions, insertions) of the SPARC gene may occur in any one or any combination of the 10 exons, or in any intron or noncoding regions of the gene. In some embodiments, the amino acid sequence of a wildtype SPARC protein is identified by the Genbank accession number NP_001296372.1. In some embodiments, the amino acid sequence of a wildtype SPARC protein is identified by the Genbank accession number NP_001296373.1. In some embodiments, the amino acid sequence of a wildtype SPARC protein is identified by the Genbank accession number NP_003109.1. Docket No.: 33612-20007.40 / 00633-0396WO1 IGFBP2 The protein encoded by IGFBP2 is one of six similar proteins that bind insulin-like growth factors I and II (IGF-I and IGF-II). The encoded protein can be secreted into the bloodstream, where it binds IGF-I and IGF-II with high affinity, or it can remain intracellular, interacting with many different ligands. In some embodiments, the IGFBP2 protein is a mouse IGFBP2 protein. In some embodiments, the nucleic acid sequence encoding the mouse IGFBP2 protein comprises a nucleic acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 98%, or at least 99% identity to the nucleic acid sequence of SEQ ID NO: 9. In some embodiments, the nucleic acid sequence encoding the mouse IGFBP2 protein comprises a nucleic acid sequence with at least 90% identity to the nucleic acid sequence of SEQ ID NO: 9. In some embodiments, the nucleic acid sequence encoding the mouse IGFBP2 protein comprises the nucleic acid sequence of SEQ ID NO: 9. In some embodiments, the mouse IGFBP2 protein comprises an amino acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 98%, or at least 99% identity to the amino acid sequence of SEQ ID NO: 10. In some embodiments, the mouse IGFBP2 protein comprises an amino acid sequence with at least 90% identity to the amino acid sequence of SEQ ID NO: 10. In some embodiments, the mouse IGFBP2 protein comprises the amino acid sequence of SEQ ID NO: 10, or a variant thereof having 20 or fewer conservative amino acid substitutions, such as up to 19, 18, 1716, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, or 1 insertions, deletions, or conservative amino acid substitutions. In some embodiments, the mouse IGFBP2 protein comprises the amino acid sequence of SEQ ID NO: 10. Mouse IGFBP2 comprises a thyroglobulin type-1 repeat domain. In some embodiments, the IGFBP2 comprises a domain that has at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or 100% sequence identity with the thyroglobulin type-1 repeat domain of SEQ ID NO:10.. In some embodiments, the nucleic acid sequence of a wildtype IGFBP2 gene is identified by the Genbank accession number NC_000067.7 from nucleotide 72,863,650 to nucleotide 72,891,633 of the reverse strand of chromosome 1 according to the GRCm39 assembly of the mouse genome. The wildtype IGFBP2 gene comprises 6 exons. One or more mutations (e.g., substitutions, deletions, insertions) of the IGFBP2 gene may occur in any one or any combination of the 6 exons, or in any intron or noncoding regions of the IGFBP2 gene. In some embodiments, the amino acid sequence of a wildtype IGFBP2 Docket No.: 33612-20007.40 / 00633-0396WO1 protein is identified by the Genbank accession number NP_001297588.1. In some embodiments, the amino acid sequence of a wildtype IGFBP2 protein is identified by the Genbank accession number NP_001297588.1. In some embodiments, the amino acid sequence of a wildtype IGFBP2 protein is identified by the Genbank accession number NP_032368.2. In some embodiments, the IGFBP2 protein is a human IGFBP2 protein. In some embodiments, the nucleic acid sequence encoding the human IGFBP2 protein comprises a nucleic acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 98%, or at least 99% identity to the nucleic acid sequence of SEQ ID NO: 11. In some embodiments, the nucleic acid sequence encoding the human IGFBP2 protein comprises a nucleic acid sequence with at least 90% identity to the nucleic acid sequence of SEQ ID NO: 11. In some embodiments, the nucleic acid sequence encoding the human IGFBP2 protein comprises the nucleic acid sequence of SEQ ID NO: 11. In some embodiments, the human IGFBP2 protein comprises an amino acid sequence with at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 98%, or at least 99% identity to the amino acid sequence of SEQ ID NO: 12. In some embodiments, the human IGFBP2 protein comprises an amino acid sequence with at least 90% identity to the amino acid sequence of SEQ ID NO: 12. In some embodiments, the human IGFBP2 protein comprises the amino acid sequence of SEQ ID NO: 12, or a variant thereof having 20 or fewer conservative amino acid substitutions, such as up to 19, 18, 1716, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, or 1 insertions, deletions, or conservative amino acid substitutions.. In some embodiments, the human IGFBP2 protein comprises the amino acid sequence of SEQ ID NO: 12. Human IGFBP2 comprises a thyroglobulin type-1 repeat domain comprising, an insulin growth factor-binding protein domain In some embodiments, the IGFBP2 comprises a domain that has at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or 100% sequence identity with the insulin growth factor-binding protein domain of SEQ ID NO:12. In some embodiments, the nucleic acid sequence of a wildtype IGFBP2 gene is identified by the Genbank accession number NC_000002.12 from nucleotide 216,632,828 to nucleotide 216,664,436 of chromosome 2 according to the GRCh38.p14 assembly of the human genome. The wildtype IGFBP2 gene comprises 6 exons. One or more mutations (e.g., substitutions, deletions, insertions) of the IGFBP2 gene may occur in any one or any combination of the 6 exons, or in any intron or noncoding regions of the IGFBP2 gene. In some embodiments, the amino acid sequence of a wildtype IGFBP2 protein is identified by the Genbank accession number Docket No.: 33612-20007.40 / 00633-0396WO1 NP_000588.3. In some embodiments, the amino acid sequence of a wildtype IGFBP2 protein is identified by the Genbank accession number NP_001300919.1. In some embodiments, the amino acid sequence of a wildtype IGFBP2 protein is identified by the Genbank accession number NP_001300921.1. In some embodiments, the amino acid sequence of a wildtype IGFBP2 protein is identified by the Genbank accession number NP_001300922.1. Regulatory Elements In some embodiments, the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is operably linked to an expression control element. In some embodiments, the expression control element is a promoter sequence. In some embodiments, the nucleic acid encoding endostatin protein, the nucleic acid encoding SPARC protein, and the nucleic acid encoding IGFBP2 are each linked to a promoter sequence. Any suitable promoter known in the art can be operably linked to the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the promoter sequence is selected from the group consisting of a constitutive promoter sequence, an inducible promoter sequence, and a tissue-specific promoter sequence. In some embodiments, the viral vector is an AAV vector. In some embodiments, the promoter sequence operably linked to the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is a constitutive promoter sequence. In some embodiments, the nucleic acid encoding endostatin protein, the nucleic acid encoding SPARC protein, and the nucleic acid encoding IGFBP2 are each linked to a constitutive promoter. In some embodiments, the constitutive promoter sequence is selected from a group consisting of a human β-actin (ACTB) promoter sequence, a cytomegalovirus (CMV) promoter sequence, a elongation factor-1α (EF1α) promoter sequence, a phosphoglycerate kinase (PGK) promoter sequence, and an uubiquitinC (UbC) promoter sequence. In some embodiments. the constitutive promoter sequence is a CMV promoter sequence. In some embodiments, the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is operably linked to a CMV promoter sequence. In some embodiments, the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is operably linked to a ACTB promoter sequence. In some embodiments, the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is operably linked to a EF1α promoter sequence. In some embodiments, the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is operably linked to a PGK promoter Docket No.: 33612-20007.40 / 00633-0396WO1 sequence. In some embodiments, the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is operably linked to a UbC promoter sequence. In some embodiments, the promoter sequence is located upstream of the transcription site of the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is operably linked to an expression control element, and the expression control element is a tissue-specific promoter sequence. In some embodiments, the nucleic acid encoding endostatin protein, the nucleic acid encoding SPARC protein, and the nucleic acid encoding IGFBP2 are each linked to a tissue-specific promoter sequence. In some embodiments, the tissue specific promoter sequence is selected from the group consisting of MYO15A promoter sequence, a MYO7A promoter sequence, a MYO6 promoter sequence, a STRC promoter sequence, a Pou4f3 promoter sequence, a endostatin promoter sequence, a SPARC promoter sequence, a IGFBP2 promoter sequence, a OTOF promoter sequence, and a SLC26A5 promoter sequence. In some embodiments, the tissue-specific promoter sequence is selected from the group consisting of a stria vascularis cell-specific promoter sequence (e.g, Kcne1), a hair cell-specific promoter sequence (e.g., Myo15), a supporting cell-specific promoter sequence (e.g., Sox2), and a spiral ganglion neuron-specific promoter sequence (e.g., TuJ1). In some embodiments, the promoter sequence is located upstream of the transcription site of the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the viral vector is an AAV vector. In some embodiments, the endostatin, SPARC, and IGFBP2 proteins are expressed from the at least one viral vector at about a 1:1:1 ratio. In some embodiments, the nucleic acid encoding endostatin protein, the nucleic acid encoding SPARC protein, and the nucleic acid encoding IGFBP2 are each linked to the same promoter sequence. In some embodiments, the nucleic acid encoding endostatin protein, the nucleic acid encoding SPARC protein, and the nucleic acid encoding IGFBP2 is each linked to a similar strength promoter, or to separate copies of the same promoter. In some embodiments, the promoter sequence is selected from the group consisting of MYO15A promoter sequence, a MYO7A promoter sequence, a MYO6 promoter sequence, a STRC promoter sequence, a Pou4f3 promoter sequence, a endostatin promoter sequence, a SPARC promoter sequence, a IGFBP2 promoter sequence, a OTOF promoter sequence, and a SLC26A5 promoter sequence. In some embodiments. the promoter sequence is a CMV promoter sequence. In some embodiments, the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein are each operably linked to a CMV promoter sequence. Docket No.: 33612-20007.40 / 00633-0396WO1 In some embodiments, the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is operably linked to an enhancer sequence. In some embodiments, the nucleic acid encoding endostatin protein, the nucleic acid encoding SPARC protein, and the nucleic acid encoding IGFBP2 is each linked to an enhancer sequence. In some embodiments, the enhancer sequence is a CMV enhancer sequence. In some embodiments, the enhancer sequence is a proximal enhancer sequence. In some embodiments, the enhancer sequence is located upstream of the transcription site of the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the viral vector is an AAV vector. In some embodiments, the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein are located on the same viral vector. In some embodiments, the viral vector comprises nucleic acid encoding a cleavable linker between each of the sequence encoding endostatin protein, the sequence encoding SPARC protein, and the sequence encoding IGFBP2 protein. In some embodiments, the cleavable linker is a self-cleaving peptide. The self-cleaving nature of these linker ensures that separate, functional proteins are produced, rather than a fused protein. This is useful when the individual proteins need to function independently. Self-cleavable linkers, like 2A peptides, enable the co-expression of multiple proteins from a single messenger RNA (mRNA) transcript through ribosomal skipping. When the ribosome encounters the 2A peptide sequence, it skips the formation of the peptide bond at a specific site, resulting in the nascent polypeptide chain being separated, releasing the upstream protein. The ribosome can then resume translation of the downstream coding sequence, producing the next protein. In some embodiments, three separate proteins are produced from translation of the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the cleavable linker is a self-cleaving peptide selected from the group consisting of F2A, P2A, T2A, and E2A. In some embodiments, the cleavable linker is a P2A peptide. In some embodiments, the P2A peptide has 1, 2, or 3 amino acid substitutions or insertions or deletions compared to the amino acid sequence of SEQ ID NO: 14. In some embodiments, the P2A peptide has 1 amino acid substitution or insertion or deletion compared to the amino acid sequence of SEQ ID NO: 14. In some embodiments, the P2A peptide has 2 amino acid substitutions or insertions or deletions compared to the amino acid sequence of SEQ ID NO: 14. In some embodiments, the amino acid substitution is a conservative amino acid substitution. In some embodiments, the P2A peptide has 3 amino acid substitutions or insertions or deletions compared to the amino acid sequence of SEQ ID NO: 14. In some embodiments, the P2A peptide comprises Docket No.: 33612-20007.40 / 00633-0396WO1 the amino acid sequence of SEQ ID NO: 14. In some embodiments, the viral vector is an AAV vector. In some embodiments, the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein are located on the same viral vector, and the viral vector comprises an Igκ leader sequence at the 5’ end of the nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the Igκ leader sequence has 1, 2, or 3 amino acid substitutions or insertions or deletions compared to the amino acid sequence of SEQ ID NO: 13. In some embodiments, the Igκ leader sequence has 1 amino acid substitution or insertion or deletion compared to the amino acid sequence of SEQ ID NO: 13. In some embodiments, the Igκ leader sequence has 2 amino acid substitutions or insertions or deletions compared to the amino acid sequence of SEQ ID NO: 13. In some embodiments, the amino acid substitution is a conservative amino acid substitution. In some embodiments, the Igκ leader sequence comprises the amino acid sequence of SEQ ID NO: 13. In some embodiments, the viral vector is an AAV vector. In some embodiments, the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein, wherein the one or more viral vectors is administered to the inner ear of the individual, and the endostatin, SPARC, and IGFBP2 proteins are secreted by inner hair cells of the individual. In some embodiments, the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein, wherein the one or more viral vectors is administered to the inner ear of the individual, and the endostatin, SPARC, and IGFBP2 proteins are secreted by outer hair cells of the individual. In some embodiments, the viral vector is an AAV vector. In some embodiments, the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is administered directly into the inner ear of the individual in need of treatment for hearing loss. In some embodiments, the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is administered through the round window membrane of the ear of the individual. In some embodiments, the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is administered in one ear of the individual. In some embodiments, the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is administered in both ears of the individual. Docket No.: 33612-20007.40 / 00633-0396WO1 Calculations of sequence homology or identity (the terms are used interchangeably herein) between sequences may be performed as follows. To determine the percent identity of two amino acid sequences, or of two nucleic acid sequences, the sequences are aligned for optimal comparison purposes (e.g., gaps can be introduced in one or both of a first and a second amino acid or nucleic acid sequence for optimal alignment and non-homologous sequences can be disregarded for comparison purposes). In a preferred embodiment, the length of a reference sequence aligned for comparison purposes is at least 30%, preferably at least 40%, more preferably at least 50%, even more preferably at least 60%, and even more preferably at least 70%, 75%, 80%, 82%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% of the length of the reference sequence. The amino acid residues or nucleotides at corresponding amino acid positions or nucleotide positions are then compared. When a position in the first sequence is occupied by the same amino acid residue or nucleotide as the corresponding position in the second sequence, then the molecules are identical at that position (as used herein amino acid or nucleic acid "identity" is equivalent to amino acid or nucleic acid "homology"). The percent identity between the two sequences is a function of the number of identical positions shared by the sequences, taking into account the number of gaps, and the length of each gap, which need to be introduced for optimal alignment of the two sequences. The comparison of sequences and determination of percent identity between two sequences can be accomplished using a mathematical algorithm. In one embodiment, the percent identity between two amino acid sequences is determined using the Needleman et al. ((1970) J. Mol. Biol.48:444-453) algorithm which has been incorporated into the GAP program in the GCG software package (available at gcg.com), using either a BLOSUM 62 matrix or a PAM250 matrix, and a gap weight of 16, 14, 12, 10, 8, 6, or 4 and a length weight of 1, 2, 3, 4, 5, or 6. In yet another embodiment, the percent identity between two nucleotide sequences is determined using the GAP program in the GCG software package (available at gcg.com), using a NWSgapdna.CMP matrix and a gap weight of 40, 50, 60, 70, or 80 and a length weight of 1, 2, 3, 4, 5, or 6. One set of parameters (and the one that can be used if the practitioner is uncertain about what parameters should be applied to determine if a molecule is within a sequence identity or homology limitation of the present disclosure) is a BLOSUM 62 scoring matrix with a gap penalty of 12, a gap extend penalty of 4, and a frameshift gap penalty of 5. Docket No.: 33612-20007.40 / 00633-0396WO1 The percent identity between two amino acid or nucleotide sequences can be determined using the algorithm of Meyers et al. ((1989) CABIOS 4:11-17) which has been incorporated into the ALIGN program (version 2.0), using a PAM120 weight residue table, a gap length penalty of 12 and a gap penalty of 4. As shown in Example 2, the methods described herein have advantages for treating hearing loss in an individual in need thereof, such as inducing axon outgrowth from the existing spiral ganglion neurons and enhancing the survival of spiral ganglion neurons and preserve synapses from chemically induced synapse damage. In some embodiments, the method results in any one, two, three, four, five, or six of (i)-(vi) effects, thereby restoring hearing function: (i) regeneration of the axons of spiral ganglion neurons, (ii) regeneration of synapses between cochlear hair cells and peripheral nerve fibers, (iii) increasing axon outgrowth of spiral ganglion neurons in the individual’s cochlea, (iv) increasing neuronal fiber length in the outer hair cell region of the individual’s cochlea, (v) enhancing survival of spiral ganglion neurons, and (vi) enhancing neurite-inner hair cell binding. In some embodiments, administration of the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein results in increased regeneration and / or increased axon outgrowth of the axons of spiral ganglion neurons (SGNs). SGNs are the primary auditory neurons transmitting sound signals from the cochlea to the brain and have limited regenerative capacity, particularly in humans. This limited regeneration is a major factor in hearing loss, where damage to SGNs can lead to irreversible hearing loss. SGNs are bipolar neurons that send axons into the central nervous system, and are the first neurons in the auditory system to fire an action potential, and supply all of the brain's auditory input. In some embodiments, administration of the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein results in increased regeneration and / or increased axon outgrowth of the axons of spiral ganglion neurons in the individual by at least 20%, at least 30% , at least 40%, at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 100%, at least 200%, or at least 300%. In some embodiments, administration of the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein results in increased regeneration and / or increased axon outgrowth of the axons of spiral ganglion neurons in the individual by about 30%, about 35%, about 40%, about 45%, about 50%, about 55%, about 60%, about 65%, about 70%, about 75%, about 80%, about 85%, about 90%, about 95%, about 100%, about 200%, or about 300%. In some embodiments, administration of the composition results in increased Docket No.: 33612-20007.40 / 00633-0396WO1 regeneration and / or increased axon outgrowth of the axons of spiral ganglion neurons in the individual as compared to regeneration of the axons of spiral ganglion neurons in an individual who did not receive the composition. In some embodiments, the regeneration of the axons of spiral ganglion neurons is evaluated by measuring the number of NFH-positive axonal projections. In some embodiments, the outgrowth of the axons of spiral ganglion neurons is evaluated by measuring the length of NFH-positive axonal projections. In some embodiments, confocal microscopy images are taken of the axonal projections and regeneration of the axons of spiral ganglion neurons is quantified as the average number of NFH-labeled fibers. In some embodiments, confocal microscopy images are taken of the axonal projections and outgrowth of the axons of spiral ganglion neurons is quantified as the average length of NFH-labeled fibers. In some embodiments, administration of the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein results in increased regeneration of synapses between cochlear hair cells and peripheral nerve fibers. Cochlear hair cells, specifically inner hair cells (IHCs), form specialized synapses with spiral ganglion neurons (SGNs). These synapses are crucial for transducing sound vibrations into electrical signals that the brain can interpret. The SGNs, also known as type I auditory nerve fibers, then relay this information to the brain. Regeneration of synapses between cochlear hair cells and peripheral nerve fibers is limited in humans. In some embodiments, administration of the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein results in increased regeneration of synapses between cochlear hair cells and peripheral nerve fibers in the individual by at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, or at least 100%. In some embodiments, administration of the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein results in increased regeneration of synapses between cochlear hair cells and peripheral nerve fibers in the individual by about 50%, about 55%, about 60%, about 65%, about 70%, about 75%, about 80%, about 85%, about 90%, about 95%, or about 100%. In some embodiments, administration of the composition results in increased regeneration of synapses between cochlear hair cells and peripheral nerve fibers in the individual as compared to regeneration of synapses between cochlear hair cells in an individual who did not receive the composition. In some embodiments, the regeneration of synapses between cochlear hair cells and peripheral nerve fibers is evaluated by measuring the number of CtBP2-positive synapses. In some embodiments, confocal microscopy images are taken of the synapses between cochlear Docket No.: 33612-20007.40 / 00633-0396WO1 hair cells and peripheral nerve fibers stained with an anti- CtBP2 antibody and regeneration of synapses between cochlear hair cells and peripheral nerve fibers is quantified as the average number of CtBP2-positive synapses. In some embodiments, the regeneration of synapses between cochlear hair cells and peripheral nerve fibers is measured as the number of CtBP2 puncta per inner hair cell. In some embodiments, administration of the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein results in an increased number of synapses between cochlear hair cells and peripheral nerve fibers. Cochlear hair cells, specifically inner hair cells (IHCs), form specialized synapses with spiral ganglion neurons (SGNs). These synapses are crucial for transducing sound vibrations into electrical signals that the brain can interpret. The SGNs, also known as type I auditory nerve fibers, then relay this information to the brain. Increasing the number of synapses between cochlear hair cells and peripheral nerve fibers is limited in humans. In some embodiments, administration of the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein results in an increased synapse number between cochlear hair cells and peripheral nerve fibers in the individual by at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 100%, at least 150%, at least 200% or at least 250%. In some embodiments, administration of the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein results in increased number of synapses between cochlear hair cells and peripheral nerve fibers in the individual by about 50%, about 55%, about 60%, about 65%, about 70%, about 75%, about 80%, about 85%, about 90%, about 95%, about 100%, about 150%, about 200% or about 250%. In some embodiments, the number of synapses is evaluated quantification of synapse per IHC. In some embodiments, the number of synapses per IHC is represented by CtBP2-positive synapses. In some embodiments, administration of the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein results in enhanced neuronal fiber length in the outer hair cell region of the individual’s cochlea. Neurite outgrowth into the outer hair cell region indicates that new hair cell-like secreting signals to stimulate the growth of retracted neurites, e.g., after hearing loss. In some embodiments, administration of the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein results in enhanced neuronal fiber length in the outer hair cell region of the individual’s cochlea. In some embodiments, administration of the Docket No.: 33612-20007.40 / 00633-0396WO1 composition results in enhanced neuronal fiber length in the outer hair cell region of the individual’s cochlea as compared to neuronal fiber length in the outer hair cell region of an individual who did not receive the composition. In some embodiments, neuronal fiber length in the outer hair cell region is evaluated by measuring the length of TuJ1+ neuronal fibers length from Hensen cell region, Deiters’ cell region, and outer pillar cell region. In some embodiments, confocal microscopy images are taken of neuronal fibers in the individual’s cochlea stained with an anti-TuJ1 antibody and neuronal fiber length in the outer hair cell region is measured as the average length of TuJ1+ neuronal fibers in the Hensen cell region, Deiters' cell region, and outer pillar cell region. In some embodiments, administration of the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein results in enhanced neurite-inner hair cell binding in the individual in need of treatment for hearing loss. The binding of neurites (axonal extensions of neurons) to inner hair cells refers to the synaptic connection formed between the neurites of spiral ganglion neurons (SGNs) and the inner hair cells, which is necessary for transmission of electrical signals, converted from sound vibrations by the inner hair cells, to the brain via the auditory nerve. In some embodiments, administration of the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein results in enhanced neurite-inner hair cell binding in the individual by at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 100%, or at least 150%. In some embodiments, administration of the composition comprising a nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein results in enhanced neurite-inner hair cell binding in the individual by about 5%, about 10%, about 15%, about 20%, about 25%, about 30%, about 35%, about 40%, about 45%, about 50%, about 55%, about 60%, about 65%, about 70%, about 75%, about 80%, about 85%, about 90%, about 95%, about 100%, or about 150%. In some embodiments, administration of the composition results in enhanced neurite-inner hair cell binding in the individual as compared to neurite-inner hair cell binding of an individual who did not receive the composition. In some embodiments, neurite-inner hair cell binding is the average neurite count per inner hair cell. In some embodiments, confocal microscopy images are taken of neurites stained with an anti-TuJ1 antibody and inner hair cells stained with an anti-Myo7A antibody, and the amount of neurite-inner hair cell binding is measured by the average TuJ1-positive neurite count per Myo7A-positive inner hair cell. Docket No.: 33612-20007.40 / 00633-0396WO1 Where appropriate, before and / or following treatment, the hearing of the individual can be assessed. In some embodiments, the hearing of the individual is assessed to determine whether the hearing has improved. Methods for measuring hearing are well-known and include, but are not limited to, pure tone audiometry, air conduction, auditory brainstem response (ABR) and bone conduction tests. These exams can measure the limits of loudness (intensity) and pitch (frequency) that an individual can hear. Hearing tests in humans include behavioral observation audiometry (for infants to seven months), visual reinforcement orientation audiometry (for children 7 months to 3 years) and play audiometry for children older than 3 years. Oto-acoustic emission testing can be used to test the functioning of the cochlea hair cells, and electro-cochleography provides information about the functioning of the cochlea and the first part of the nerve pathway to the brain. In some embodiments, hearing of the individual is assessed by auditory brainstem response (ABR). ABR is a measurement of auditory pathway function from the auditory nerve to the mesencephalon (Young A. et al., Auditory Brainstem Response. [Updated 2023 Jan 12]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2024 Jan., the contents of which is herein incorporated by reference in its entirety). In some embodiments, ABR threshold provides a metric for overt hearing loss. Any suitable method known in the art can be used for recording ABR. In some embodiments, the method further comprises determining the ABR threshold (e.g., a minimum level of sound intensity that elicits a detectable ABR) of the individual at one or more frequencies (e.g., about 5, about 8, about 11, about 16, about 22, and about 32 kHz). In some embodiments, hearing of the individual is assessed by determining the Wave I amplitude of the individual. Wave 1 amplitude represents the summed activity of the auditory nerve fibers contacting inner hair cells (IHCs) and can be used to assess the individual for potential cochlear synaptopathy. Any suitable method known in the art can be used for determining Wave I amplitude. In some embodiments, the method further comprises determining the Wave I amplitude of the individual at one or more frequencies (e.g., about 5, about 8, about 11, about 16, about 22, and about 32 kHz). In some embodiments, the method comprises determining the Wave I amplitude of the individual at a frequency before and after the individual has received the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the method results in a higher Wave I amplitude at a frequency in the individual who is administered the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein as compared to the Wave I amplitude at the same frequency in an individual Docket No.: 33612-20007.40 / 00633-0396WO1 who is not administered the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the Wave I amplitude in response to 20-100 dB SPL stimulus is higher in the individual who is administered the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein as compared to the Wave I amplitude in response to 20-100 dB SPL stimulus in an individual who is not administered the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, administration of the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein results in higher Wave I amplitude by at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 100%, at least 150%, or at least 200%. In some embodiments, administration of the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein results in higher Wave I amplitude by about 50%, about 55%, about 60%, about 65%, about 70%, about 75%, about 80%, about 85%, about 90%, about 95%, about 100%, about 150%, or about 200%. In some embodiments, hearing of the individual is assessed by determining the speech recognition of the individual. Any suitable method known in the art can be used for determining speech recognition. Speech recognition is assessed in hearing tests to evaluate how well a person understands spoken words, which is crucial for diagnosing and managing hearing loss. This assessment typically involves measuring the speech reception threshold (SRT), which is the softest level at which someone can repeat speech, and word recognition scores (WRS), which gauge the ability to understand words at a comfortable listening level. In some embodiments, the method comprises determining the speech recognition of the individual before and after the individual has received the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the method results in improved speech recognition in the individual who is administered the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein as compared to the speech recognition in an individual who is not administered the composition comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the method results in improved speech recognition. In some embodiments, the method improves hearing function in the individual in need of treatment for hearing loss following administration of the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, Docket No.: 33612-20007.40 / 00633-0396WO1 and IGFBP2 protein. In some embodiments, the method improves hearing function in the individual 1 week following administration of the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the method improves hearing function in the individual 4 weeks following administration of the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments the method improves hearing function in the individual 12 weeks following administration of the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the individual is human. Endostatin, SPARC, and IGFBP2 proteins In another aspect, provided herein are methods for preventing and / or treating hearing loss in an individual in need thereof comprising administering an effective amount of endostatin, SPARC, and IGFBP2 proteins or mRNA encoding endostatin, SPARC, and IGFP2 proteins directly to the inner ear of the individual. Also provided herein are pharmaceutical compositions comprising an effective amount of endostatin, SPARC, and IGFBP2 proteins and mRNAs encoding endostatin, SPARC, and IGFP2 proteins. The effective amount of endostatin, SPARC, and IGFBP2 proteins may include full endostatin, SPARC, and IGFBP2 proteins and / or truncations thereof. In some embodiments, the pharmaceutical compositions comprise phosphate-buffering saline (PBS). In some embodiments, one or more of the COL18A1, SPARC, and IGFBP2 proteins are full-length proteins. In some embodiments, an amino acid sequence encoding the COL18A1, SPARC, or IGFBP2 protein is an unmodified wild-type sequence. Alternatively, the amino acid sequence encoding the COL18A1, SPARC, or IGFBP2 protein can be modified using standard techniques. In some embodiments, one or more the COL18A1, SPARC, and IGFBP2 proteins are altered proteins. In some embodiments, one or more the COL18A1, SPARC, and IGFBP2 proteins are a fragment of the wildtype protein or truncated proteins. In some embodiments, the COL18A1 protein is a fragment or truncation of the wildtype COL18A1.In some embodiments, the truncation is a truncation comprising the C- terminal region of COL18A1. In some embodiments, endostatin comprises amino acids 1337- 1519 of the full length COL18A1. In some embodiments, amino acids 1337-1519 comprise Docket No.: 33612-20007.40 / 00633-0396WO1 an amino acid proteolytic fragment. In some embodiments, the COL18A1 protein is an endostatin. In some embodiments, one or more of the endostatin, SPARC, and IGFBP2 proteins are full-length proteins. In some embodiments, an amino acid sequence encoding the endostatin, SPARC, or IGFBP2 protein is an unmodified wild-type sequence. Alternatively, the amino acid sequence encoding the endostatin, SPARC, or IGFBP2 protein can be modified using standard techniques. In some embodiments, one or more the endostatin, SPARC, and IGFBP2 proteins are altered (e.g., cleaved, truncated, fragment) endostatin, SPARC, and IGFBP2 proteins or biologically active forms thereof. In some embodiments, one or more the endostatin, SPARC, and IGFBP2 proteins are a full protein, a fragment of the wildtype protein or truncated proteins. In some embodiments, the full protein, the fragments of the wildtype protein or truncated proteins are transcribed from highly conserved mouse and human genes. In some embodiments, the IGFBP2 mouse and human genes share 87.5% sequence similarity. In some embodiments, the SPARK mouse and human genes share 92.6% sequence similarity. In some embodiments, the genomic region expressed as the mouse and human endostatins share 84.7% sequence similarity. In some embodiments, the endostatin, SPARC, and IGFBP2 proteins are administered at an approximately equimolar ratio. In some embodiments, the endostatin, SPARC, and IGFBP2 proteins are administered at about a 1:1:1 ratio. In some embodiments, the endostatin, SPARC, and IGFBP2 proteins are not administered at an equimolar ratio. In some embodiments, the endostatin, SPARC, and IGFBP2 proteins are administered in a formulation comprising phosphate-buffering saline (PBS). In some embodiments, an effective amount of endostatin, SPARC, and IGFBP2 proteins is administered directly into the inner ear of the individual in need of treatment for hearing loss. In some embodiments, an effective amount of endostatin, SPARC, and IGFBP2 proteins is administered through the round window membrane of the ear of the individual. In some embodiments, an effective amount of endostatin, SPARC, and IGFBP2 proteins is administered in one ear of the individual. In some embodiments, an effective amount of endostatin, SPARC, and IGFBP2 proteins is administered in both ears of the individual. As shown in Example 2, the methods described herein has advantages for treating hearing loss in an individual in need thereof, such as inducing axon outgrowth from the existing spiral ganglion neurons and enhancing the survival of spiral ganglion neurons and preserve synapses from chemically induced synapse damage. In some embodiments, the method results in any one, two, three, four, five, or six of (i)-(vi) effects, thereby restoring Docket No.: 33612-20007.40 / 00633-0396WO1 hearing function: (i) regeneration of the axons of spiral ganglion neurons, (ii) regeneration of synapses between cochlear hair cells and peripheral nerve fibers, (iii) increasing axon outgrowth of spiral ganglion neurons in the individual’s cochlea, (iv) increasing neuronal fiber length in the outer hair cell region of the individual’s cochlea, (v) enhancing survival of spiral ganglion neurons, and (vi) enhancing neurite-inner hair cell binding. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in increased regeneration and / or increased axon outgrowth of the axons of spiral ganglion neurons (SGNs). SGNs are the primary auditory neurons transmitting sound signals from the cochlea to the brain and have limited regenerative capacity, particularly in humans. This limited regeneration is a major factor in hearing loss, where damage to SGNs can lead to irreversible hearing loss. SGNs are bipolar neurons that send axons into the central nervous system, and are the first neurons in the auditory system to fire an action potential, and supply all of the brain's auditory input. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in increased regeneration and / or increased axon outgrowth of the axons of spiral ganglion neurons in the individual by at least 20%, at least 30% , at least 40%, at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 100%, at least 200%, or at least 300%. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in increased regeneration and / or increased axon outgrowth of the axons of spiral ganglion neurons in the individual by about 30%, about 35%, about 40%, about 45%, about 50%, about 55%, about 60%, about 65%, about 70%, about 75%, about 80%, about 85%, about 90%, about 95%, about 100%, about 200%, or about 300%. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in increased regeneration and / or increased axon outgrowth of the axons of spiral ganglion neurons in the individual as compared to regeneration of the axons of spiral ganglion neurons in an individual who did not receive an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some embodiments, the regeneration of the axons of spiral ganglion neurons is evaluated by measuring the number of NFH-positive axonal projections. In some embodiments, the outgrowth of the axons of spiral ganglion neurons is evaluated by measuring the length of NFH-positive axonal projections. In some embodiments, confocal microscopy images are taken of the axonal projections and regeneration of the axons of spiral ganglion neurons is quantified as the average number of NFH-labeled fibers. In some embodiments, confocal microscopy images are taken of the axonal projections and outgrowth Docket No.: 33612-20007.40 / 00633-0396WO1 of the axons of spiral ganglion neurons is quantified as the average length of NFH-labeled fibers. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in increased regeneration of synapses between cochlear hair cells and peripheral nerve fibers. Cochlear hair cells, specifically inner hair cells (IHCs), form specialized synapses with spiral ganglion neurons (SGNs). These synapses are crucial for transducing sound vibrations into electrical signals that the brain can interpret. The SGNs, also known as type I auditory nerve fibers, then relay this information to the brain. Regeneration of synapses between cochlear hair cells and peripheral nerve fibers is limited in humans. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in increased regeneration of synapses between cochlear hair cells and peripheral nerve fibers in the individual by at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, or at least 100%. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in increased regeneration of synapses between cochlear hair cells and peripheral nerve fibers in the individual by about 50%, about 55%, about 60%, about 65%, about 70%, about 75%, about 80%, about 85%, about 90%, about 95%, or about 100%. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in increased regeneration of synapses between cochlear hair cells and peripheral nerve fibers in the individual as compared to regeneration of synapses between cochlear hair cells in an individual who did not receive an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some embodiments, the regeneration of synapses between cochlear hair cells and peripheral nerve fibers is evaluated by measuring the number of CtBP2-positive synapses. In some embodiments, confocal microscopy images are taken of the synapses between cochlear hair cells and peripheral nerve fibers stained with an anti- CtBP2 antibody and regeneration of synapses between cochlear hair cells and peripheral nerve fibers is quantified as the average number of CtBP2-positive synapses. In some embodiments, the regeneration of synapses between cochlear hair cells and peripheral nerve fibers is measured as the number of CtBP2 puncta per inner hair cell. In some embodiments, the method comprises delivering a endostatin protein as described herein to the ear of the individual. In some embodiments, the method comprises delivering a COL18A1 protein as described herein to the ear of the individual. In some embodiment, the endostatin protein is a human or a mouse protein, comprising the amino acid sequence set forth in SEQ ID NO: 2, 17 or 18, or a variant comprising least 50%, at least Docket No.: 33612-20007.40 / 00633-0396WO1 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 98%, or at least 99% identity to the amino acid sequence of SEQ ID NO: 2, 17 or 18. In some embodiments, the endostatin protein is packaged in a lipid nanoparticle. In some embodiments, the method comprises delivering a SPARC protein as described herein to the ear of the individual. In some embodiment, the SPARC protein is a human or a mouse protein, comprising the amino acid sequence set forth in SEQ ID NO: 6 or 8, or a variant comprising least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 98%, or at least 99% identity to the amino acid sequence of SEQ ID NO: 6 or 8. In some embodiments, the SPARC protein is packaged in a lipid nanoparticle. In some embodiments, the method comprises delivering a IGFBP2 protein as described herein to the ear of the individual. In some embodiment, the IGFBP2 protein is a human or a mouse protein, comprising the amino acid sequence set forth in SEQ ID NO: 10 or 12, or a variant comprising least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95%, at least 98%, or at least 99% identity to the amino acid sequence of SEQ ID NO: 10 or 12. In some embodiments, the IGFBP2 protein is packaged in a lipid nanoparticle. In some embodiments, the endostatin protein, the SPARC protein, and the IGFBP2 protein are delivered in the same lipid nanoparticles. In some embodiments, the endostatin protein, the SPARC protein, and the IGFBP2 protein are delivered in different lipid nanoparticles. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in an increased number of synapses between cochlear hair cells and peripheral nerve fibers. Cochlear hair cells, specifically inner hair cells (IHCs), form specialized synapses with spiral ganglion neurons (SGNs). In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in an increased synapse number between cochlear hair cells and peripheral nerve fibers in the individual by at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 100%, at least 150%, at least 200% or at least 250%. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in increased number of synapses between cochlear hair cells and peripheral nerve fibers in the individual by about 50%, about 55%, about 60%, about 65%, about 70%, about 75%, about 80%, about 85%, about 90%, about 95%, about 100%, about 150%, about 200% or about 250%. In some embodiments, the Docket No.: 33612-20007.40 / 00633-0396WO1 number of synapses is evaluated quantification of synapse per IHC. In some embodiments, the number of synapses per IHC is represented by CtBP2-positive synapses. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in enhanced neuronal fiber length in the outer hair cell region of the individual’s cochlea. Neurite outgrowth into the outer hair cell region indicates that new hair cell-like secreting signals to stimulate the growth of retracted neurites, e.g., after hearing loss. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in enhanced neuronal fiber length in the outer hair cell region of the individual’s cochlea. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in enhanced neuronal fiber length in the outer hair cell region of the individual’s cochlea as compared to neuronal fiber length in the outer hair cell region of an individual who did not receive an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some embodiments, neuronal fiber length in the outer hair cell region is evaluated by measuring the length of TuJ1+ neuronal fibers length from Hensen cell region, Deiters’ cell region, and outer pillar cell region. In some embodiments, confocal microscopy images are taken of neuronal fibers in the individual’s cochlea stained with an anti-TuJ1 antibody and neuronal fiber length in the outer hair cell region is measured as the average length of TuJ1+ neuronal fibers in the Hensen cell region, Deiters' cell region, and outer pillar cell region. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in enhanced neurite-inner hair cell binding in the individual in need of treatment for hearing loss. The binding of neurites (axonal extensions of neurons) to inner hair cells refers to the synaptic connection formed between the neurites of spiral ganglion neurons (SGNs) and the inner hair cells, which is necessary for transmission of electrical signals, converted from sound vibrations by the inner hair cells, to the brain via the auditory nerve. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in enhanced neurite-inner hair cell binding in the individual by at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 100%, or at least 150%. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in enhanced neurite-inner hair cell binding in the individual by about 5%, about 10%, about 15%, about 20%, about 25%, about 30%, about 35%, about 40%, about 45%, about 50%, about 55%, about 60%, about 65%, about 70%, about 75%, about 80%, about 85%, about 90%, about 95%, about 100%, or about 150%. In some embodiments, Docket No.: 33612-20007.40 / 00633-0396WO1 administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in enhanced neurite-inner hair cell binding in the individual as compared to neurite-inner hair cell binding of an individual who did not receive an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some embodiments, neurite-inner hair cell binding is the average neurite count per inner hair cell. In some embodiments, confocal microscopy images are taken of neurites stained with an anti-TuJ1 antibody and inner hair cells stained with an anti- Myo7A antibody, and the amount of neurite-inner hair cell binding is measured by the average TuJ1-positive neurite count per Myo7A-positive inner hair cell. In some embodiments, hearing of the individual is assessed by determining the Wave I amplitude of the individual. Wave 1 amplitude represents the summed activity of the auditory nerve fibers contacting inner hair cells (IHCs) and can be used to assess the individual for potential cochlear synaptopathy. Any suitable method known in the art can be used for determining Wave I amplitude. In some embodiments, the method further comprises determining the Wave I amplitude of the individual at one or more frequencies (e.g., about 5, about 8, about 11, about 16, about 22, and about 32 kHz). In some embodiments, the method comprises determining the Wave I amplitude of the individual at a frequency before and after the individual has received an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some embodiments, the method results in a higher Wave I amplitude at a frequency in the individual who is administered an effective amount of endostatin, SPARC, and IGFBP2 proteins as compared to the Wave I amplitude at the same frequency in an individual who is not administered an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some embodiments, the Wave I amplitude in response to 20-100 dB SPL stimulus is higher in the individual who is administered an effective amount of endostatin, SPARC, and IGFBP2 proteins as compared to the Wave I amplitude in response to 20-100 dB SPL stimulus in an individual who is not administered an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some embodiments, administration of an effective amount of COL18A1, SPARC, and IGFBP2 proteins results in higher Wave I amplitude by at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 100%, at least 150%, or at least 200%. In some embodiments, administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins results in higher Wave I amplitude by about 50%, about 55%, about 60%, about 65%, about 70%, about 75%, about 80%, about 85%, about 90%, about 95%, about 100%, about 150%, or about 200%. Docket No.: 33612-20007.40 / 00633-0396WO1 In some embodiments, hearing of the individual is assessed by determining the speech recognition of the individual. Any suitable method known in the art can be used for determining speech recognition. Speech recognition is assessed in hearing tests to evaluate how well a person understands spoken words, which is crucial for diagnosing and managing hearing loss. This assessment typically involves measuring the speech reception threshold (SRT), which is the softest level at which someone can repeat speech, and word recognition scores (WRS), which gauge the ability to understand words at a comfortable listening level. In some embodiments, the method comprises determining the speech recognition of the individual before and after the individual has received an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some embodiments, the method results in improved speech recognition in the individual who is administered an effective amount of endostatin, SPARC, and IGFBP2 proteins as compared to the speech recognition in an individual who is not administered an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some embodiments, the method results in improved speech recognition following administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some embodiments, the method improves hearing function in the individual in need of treatment for hearing loss following administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the method improves hearing function in the individual 1 week following administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the method improves hearing function in the individual 4 weeks following administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments the method improves hearing function in the individual 12 weeks following administration of an effective amount of endostatin, SPARC, and IGFBP2 proteins comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the individual is human. Docket No.: 33612-20007.40 / 00633-0396WO1 Hearing Loss Provided herein are methods for preventing and / or treating hearing loss in an individual in need thereof comprising administering a composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein, wherein the one or more viral vectors is administered to the inner ear of the individual. Also provided herein are methods for preventing and / or treating hearing loss in an individual in need thereof comprising administering an effective amount of endostatin, SPARC, and IGFBP2 proteins to the inner ear of the individual. Also provided herein are methods for preventing and / or treating hearing loss in an individual in need thereof comprising administering an effective amount of an mRNA encoding endostatin proteins, an mRNA encoding SPARC proteins, and an mRNA encoding IGFBP2 proteins to the inner ear of the individual. In some embodiments, the methods provided herein are useful for preventing and treating hearing loss in an individual. Patients with hearing loss can be identified using standard hearing tests known in the art. In some embodiments, the present disclosure can be used to treat and / or prevent inner ear cell damage (e.g., cochlear cell or utricular cell damage) and associated disorders, such as hearing impairments, deafness, vestibular disorders, tinnitus (see, Kaltenbach et al. (2002) J NEUROPHYSIOL, 88(2).699-714s, the contents of which is herein incorporated by reference in its entirety), and hyperacusis (Kujawa et al. (2009) J. NEUROSCI. 29(45):14077-14085, the contents of which is herein incorporated by reference in its entirety). The individual may be deaf or have hearing loss for any reason, or as a result of any type of event. For example, an individual can be deaf or hard-of-hearing as a result of a traumatic event, such as a physical trauma to a structure of the ear. For example, prolonged exposures to concerts, airport runways, and construction areas can cause inner ear damage and subsequent hearing loss. In some embodiments, the hearing loss is cochlear synaptopathy, i.e., “hidden hearing loss.” Cochlear synaptophy is a type of hearing loss characterized by the damage or loss of synapses between the inner hair cells and auditory nerve fibers in the cochlea, the inner ear structure responsible for hearing. This damage can occur due to noise exposure or aging and can lead to difficulties in understanding speech in noisy environments, tinnitus, and hyperacusis, even with normal hearing sensitivity as measured by traditional audiograms. In Docket No.: 33612-20007.40 / 00633-0396WO1 some embodiments, cochlear synaptopathy is a type of noise-induced hearing loss. In some embodiments, cochlear synaptopathy is a type of age-related hearing loss. Sensory epithelia of the inner ear contain two major cell types: hair cells and supporting cells. (G. Wan et al., Semin Cell Dev Biol.2013 May; 24(5): 448-459). Hair cells convert the energy in sound and head movements into neurophysiological signals that are relayed to the brainstem. In mammals, six sensory organs contain hair-cell epithelia. In the cochlear organ, which is specialized for hearing, hair cells reside within the organ of Corti, atop the basilar membrane, which vibrates in response to sound waves. Similarly, each of the five vestibular organs (the utricle, the saccule, and the three canal organs) contains sensory epithelia with hair cells that are activated by head movements and gravitational force. Hair cells are innervated by neurons whose cell bodies sit outside the sensory epithelium, either in a sensory ganglion within the temporal bone (afferent neurons) or in the hindbrain (efferent neurons). There are two types of hair cells in the cochlea: outer and inner hair cells. Outer hair cells are distal from the spiral limbus, and generally there are three to five rows of hair cells that run the length of the cochlear duct (about 20,000 in number in humans). Inner hair cells are proximal to the spiral limbus. There is only one row of inner hair cells that run the length of the cochlear duct (about 3500 in number in humans). Sound is converted into electrical signals in the cochlea by hair cells. These signals are then transmitted to the brain via auditory nerve fibers, with synapses acting as the crucial connections between the two. In cochlear synaptopathy, these synapses are damaged or lost, disrupting the flow of information to the brain. Individuals with cochlear synaptopathy may struggle to understand speech in noisy situations, experience tinnitus (ringing in the ears), or have heightened sensitivity to sound (hyperacusis). This type of hearing loss can be “hidden” because traditional hearing tests (like audiograms) may not detect it, as they primarily assess the function of hair cells and not the synapses. As such, cochlear synaptopathy is difficult to diagnose because traditional audiometry tests often yield normal results even when damage to the auditory nerve synapses has occurred. However, several methods are being explored to identify this condition, including electrophysiological tests like ECochG and behavioral tests like speech- in-noise tests. Docket No.: 33612-20007.40 / 00633-0396WO1 Routes of Administration In some embodiments, a composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is administered to the inner ear of the individual. In some embodiments, an effective amount of endostatin, SPARC, and IGFBP2 proteins is administered to the inner ear of the individual. In some embodiments, an effective amount of mRNA encoding endostatin proteins, mRNA encoding SPARC proteins, and mRNA encodings IGFBP2 proteins is administered to the inner ear of the individual. In some aspects, provided herein are methods of preventing hearing loss in an individual comprising administering to the ear of the individual a composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some aspects, provided herein are methods of treating hearing loss in an individual comprising administering to the ear of the individual a composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein. In some embodiments, the nucleic acid sequence encoding a COL18A1 protein, SPARC protein, and IGFBP2 proteins are comprised by one or more viral vectors, wherein the one or more viral vectors is administered to the inner ear of the individual. In some embodiments, the methods provided herein are useful for preventing and treating hearing loss. In some aspects, provided herein are methods of preventing hearing loss in an individual comprising administering to the ear of the individual an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some aspects, provided herein are methods of treating hearing loss in an individual comprising administering to the ear of the individual an effective amount of endostatin, SPARC, and IGFBP2 proteins. In some embodiments, the methods provided herein are useful for preventing and treating hearing loss. In some embodiments, the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein is formulated as a pharmaceutical composition containing the appropriate carriers and / or excipients. In some embodiments, an effective amount of endostatin, SPARC, and IGFBP2 proteins is formulated as a pharmaceutical composition containing the appropriate carriers and / or excipients. In some embodiments, the agent is solubilized in a carrier, for example, a viscoelastic carrier, that is introduced locally into the inner ear. In some embodiments, the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein Docket No.: 33612-20007.40 / 00633-0396WO1 or an effective amount of endostatin, SPARC, and IGFBP2 proteins is administered to an individual, e.g., an individual being treating for hearing loss, using a local route of administration. Such local routes of administration include administering the composition into the ear of an individual and / or the inner ear of an individual, for example, by injection and / or using a pump. In some embodiments, the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein or an effective amount of endostatin, SPARC, and IGFBP2 proteins is injected into the ear (e.g., auricular administration), such as into the luminae of the cochlea (e.g., the Scala media, Sc vestibulae, and Sc tympani). In some embodiments, the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein or an effective amount of endostatin, SPARC, and IGFBP2 proteins is administered directly into the inner ear of the individual. In some embodiments, the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein or an effective amount of endostatin , SPARC, and IGFBP2 proteins is administered through the round window membrane of the ear of the individual. In some embodiments, the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein or an effective amount of endostatin, SPARC, and IGFBP2 proteins is administered in one ear of the individual. In some embodiments, the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein or an effective amount of endostatin, SPARC, and IGFBP2 proteins is administered in both ears of the individual. Such methods are routinely used in the art, for example, for the administration of steroids and antibiotics into human ears. In some embodiments, the composition is administered to the ear of the individual via a cochleostomy (e.g., via a surgical procedure that involves forming an opening into the cochlea). In some embodiments, the composition is administered to the ear of the individual via cochleostomy, round window injection, or canalostomy. In such embodiments, the composition can be injected into the cochlea with or without a pump. In some embodiments, the composition comprising one or more viral vectors comprising a nucleic acid encoding endostatin protein, SPARC protein, and IGFBP2 protein or an effective amount of endostatin, SPARC, and IGFBP2 proteins is administered to the inner ear using a catheter or pump. A catheter or pump can, for example, direct the agent into the cochlea luminae or the round window of the ear. Exemplary drug delivery systems Docket No.: 33612-20007.40 / 00633-0396WO1 suitable for administering one or more compounds into an ear, e.g., a human ear, are described in U.S. Patent Publication No.2006 / 0030837 and U.S. Pat. No.7,206,639, the contents of which are herein incorporated by reference in their entirety. In certain embodiments, a catheter or pump can be positioned, e.g., in the ear (e.g., the outer, middle, and / or inner ear) of an individual during a surgical procedure. Also provided herein are LNPs comprising mRNA encoding endostatin, SPARC, and / or IGFP2 proteins. In some embodiments, each LNP comprises mRNA encoding endostatin, mRNA encoding SPARC, and mRNA encoding IGFP2. In some embodiments, effective amounts of the LNPs are administered to an ear of an individual. Having generally described the compositions, methods, and processes of this disclosure, the same will be better understood by reference to certain specific examples, which are included herein to further illustrate the disclosure and are not intended to limit the scope of the invention as defined by the claims. Sequence Listing Table 1: Sequences Description N e e Docket No.: 33612-20007.40 / 00633-0396WO1 gttccacatccggccagcca cagagggccc aggggtgctg ttcgccatca cggactcggc gcaggccatggtcttgctgg gcgtgaagct ctctggggtg ca ac c acca acat ctccctctctacacaaac ca tca Docket No.: 33612-20007.40 / 00633-0396WO1 cctccaggat tgccagggaa tcagggccctccaggaccca agggcgccaa aggagaagtg ggcccccccg gaccaccagg gcagtttccg tttgactttc ttcatt a ctaaat aa aa a aacc a tatca Docket No.: 33612-20007.40 / 00633-0396WO1 QGPPGPPGPS FRHDKLTFID MEGSGFGGDL EALRGPRGFP GPPGPPGVPG LPGEPGRFGV NSSDVPGPAG LPGVPGREGP PGFPGLPGPP e Docket No.: 33612-20007.40 / 00633-0396WO1 MRAWIFFLLC LAGRALAAPQ QTEVAEEIVE Mouse SPARC amino EETVVEETGV PVGANPVQVE MGEFEDGAEE TVEEVVADNP CQNHHCKHGK VCELDESNTP acid sequence (302 e o e Docket No.: 33612-20007.40 / 00633-0396WO1 caaaggggccaagcacctgagtcttgaagaaccaaagaagttgcggccacctc ctgccagaactccatgccagcaagaactggaccaagtacttgaaaggattagtac catc ctcccaatatca ccacttaacaccttattctctcacatcc o e Docket No.: 33612-20007.40 / 00633-0396WO1 METDTLLLWVLLLWVPGSTGD Igκ signal peptide ATNFSLLKQAGDVEENPGP P2A linker Docket No.: 33612-20007.40 / 00633-0396WO1 RGDAGQKGER GEPGAPGGGF FSSSVPGPPG PPGYPGIPGP KGESIRGPPG PPGPQGPPGI GYEGRQGPPG PPGPPGPPSF PGPHRQTVSV d m) d EXEMPLARY EMBODIMENTS Embodiment 1. A method for preventing and / or treating hearing loss in an individual in need thereof comprising administering a composition comprising one or more viral vectors comprising a nucleic acid encoding a COL18A1 protein, a SPARC protein, and a IGFBP2 Docket No.: 33612-20007.40 / 00633-0396WO1 protein, wherein the one or more viral vectors is administered to the inner ear of the individual. Embodiment 2. The method according to embodiment 1, wherein the one or more viral vectors is selected from the group consisting of a lentivirus vector, an adenovirus vector, an adeno-associated virus (AAV) vector, a herpes simplex virus (HSV) vector, and a retrovirus vector. Embodiment 3. The method according to embodiment 2, wherein the one or more viral vectors comprises an AAV vector. Embodiment 4. The method according to any one of embodiments 1-3, wherein the nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein is each operably linked to an expression control element. Embodiment 5. The method according to embodiment 4, wherein the expression control element is a promoter sequence. Embodiment 6. The method according to embodiment 5, wherein the promoter sequence is selected from the group consisting of a constitutive promoter sequence, an inducible promoter sequence, and a tissue-specific promoter sequence. Embodiment 7. The method according to embodiment 6, wherein the promoter sequence is a constitutive promoter sequence. Embodiment 8. The method according to embodiment 7, wherein the constitutive promoter sequence is selected from the group consisting of MYO15A promoter sequence, a MYO7A promoter sequence, a MYO6 promoter sequence, a STRC promoter sequence, a Pou4f3 promoter sequence, a OTOF promoter sequence, and a SLC26A5 promoter sequence. Embodiment 9. The method according to embodiment 8, wherein the constitutive promoter sequence is a CMV promoter sequence. Embodiment 10. The method according to embodiment 6, wherein the promoter sequence is a tissue-specific promoter sequence. Embodiment 11. The method according to embodiment 10, wherein the tissue-specific promoter sequence is selected from the group consisting of a stria vascularis cell-specific promoter sequence, a hair cell-specific promoter sequence, a supporting cell-specific promoter sequence, and a spiral ganglion neuron-specific promoter sequence. Embodiment 12. The method according to embodiment 4, wherein the expression control element comprises an enhancer sequence; optionally, wherein the enhancer sequence is a CMV enhancer sequence. Docket No.: 33612-20007.40 / 00633-0396WO1 Embodiment 13. The method according to any one of embodiments 1-12, wherein the nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein are located on the same viral vector. Embodiment 14. The method according to embodiment 13, wherein the viral vector comprises nucleic acid encoding a cleavable linker between each of the sequence encoding COL18A1 protein, the sequence encoding SPARC protein, and the sequence encoding IGFBP2 protein. Embodiment 15. The method according to embodiment 14, wherein the cleavable linker is a self-cleaving peptide selected from the group consisting of F2A, P2A, T2A, and E2A. Embodiment 16. The method according to embodiment 15, wherein the cleavable linker is a P2A peptide. Embodiment 17. The method according to embodiment 16, wherein the P2A peptide has 1, 2, or 3 amino acid substitutions or insertions or deletions compared to the amino acid sequence of SEQ ID NO: 14. Embodiment 18. The method according to embodiment any one of 14-17 wherein three separate proteins are produced from translation of the nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein. Embodiment 19. The method according to any one of embodiments 13-18, wherein the AAV vector comprises an Igκ leader sequence (SEQ ID NO: 13) at the 5’ end of the nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein. Embodiment 20. The method according to any one of embodiments 1-19, wherein the COL18A1, SPARC, and IGFBP2 proteins are secreted by inner hair cells or outer hair cells of the individual. Embodiment 21. A method for preventing and / or treating hearing loss in an individual in need thereof comprising administering an effective amount of COL18A1, SPARC, and IGFBP2 proteins or an effective amount of an mRNA encoding the COL18A1 protein, an mRNA encoding the SPARC protein and an mRNA encoding the IGFBP2 protein to the inner ear of the individual. Embodiment 22. The method of embodiment 21, wherein the effective amount of an mRNA encoding the COL18A1 protein, an mRNA encoding the SPARC protein and an mRNA encoding the IGFBP2 protein are administered in a lipid nanoparticle (LNP). Embodiment 23. The method according to any one of embodiments 1-22, wherein the COL18A1 Docket No.: 33612-20007.40 / 00633-0396WO1 Embodiment 24. The method according to any one of embodiments 1-23, wherein the COL18A1 protein has at least 90% identity to the amino acid sequence of SEQ ID NO: 2 or 4. Embodiment 25. The method according to any one of embodiments 1-24, wherein the SPARC protein has at least 90% identity to the amino acid sequence of SEQ ID NO: 6 or 8. Embodiment 26. The method according to any one of embodiments 1-25, wherein the IGFBP2 protein has at least 90% identity to the amino acid sequence of SEQ ID NO: 10 or 12. Embodiment 27. The method according to any one of embodiments 23-26, wherein the COL18A1, SPARC, and IGFBP2 proteins are administered at about a 1:1:1 ratio. Embodiment 28. The method according to any one of embodiments 1-27, wherein the hearing loss is age-related hearing loss. Embodiment 29. The method according to any one of embodiments 1-28, wherein the hearing loss is noise-induced hearing loss. Embodiment 30. The method according to any one of embodiments 1-29, wherein the composition or the COL18A1, SPARC, and IGFBP2 proteins are administered directly to the inner ear of the individual. Embodiment 31. The method according to any one of embodiments 1-30, wherein the composition or the COL18A1, SPARC, and IGFBP2 proteins is administered through the round window membrane of the ear of the individual. Embodiment 32. The method according to any one of embodiments 1-31, wherein the composition or the COL18A1, SPARC, and IGFBP2 proteins is administered to both ears of the individual. Embodiment 33. The method according to any one of embodiments 1-32, wherein the method results in regeneration of the axons of spiral ganglion neurons and / or synapses between cochlear hair cells and peripheral nerve fibers, and / or enhances axon outgrowth of spiral ganglion neurons in the individual’s cochlea, thereby restoring hearing function. Embodiment 34. The method according to any one of embodiments 1-33, wherein the method increases neuronal fiber length in the outer hair cell region of the individual’s cochlea. Embodiment 35. The method according to any one of embodiments 1-34, wherein the method enhances survival of spiral ganglion neurons. Embodiment 36. The method according to any one of embodiments 1-35, wherein the method enhances neurite-inner hair cell binding. Docket No.: 33612-20007.40 / 00633-0396WO1 Embodiment 37. The method according to any one of embodiments 1-36, further comprising determining the Wave I amplitude of the individual at a frequency before and after the individual has received the composition or the COL18A1, SPARC, and IGFBP2 proteins. Embodiment 38. The method according to embodiment 37, wherein the method results in a higher Wave I amplitude at a frequency in the individual who is administered the composition or the COL18A1, SPARC, and IGFBP2 proteins as compared to the Wave I amplitude at the same frequency in an individual who is not administered the composition or the COL18A1, SPARC, and IGFBP2 proteins. Embodiment 39. The method according to embodiment 38, wherein the Wave I amplitude in response to 20-100 dB SPL stimulus is higher in the individual who is administered the composition or the COL18A1, SPARC, and IGFBP2 proteins as compared to the Wave I amplitude in response to 20-100 dB SPL stimulus in an individual who is not administered the composition or the COL18A1, SPARC, and IGFBP2 proteins. Embodiment 40. The method according to any one of embodiments 1-39, further comprising determining speech recognition of the individual before and after the individual has received the composition or the COL18A1, SPARC, and IGFBP2 proteins. Embodiment 41. The method according to embodiment 40, wherein the method improves speech recognition of the individual as compared to speech recognition of the individual prior to administration of the composition or the COL18A1, SPARC, and IGFBP2 proteins. Embodiment 42. The method according to any one of embodiments 1-41, wherein the method improves hearing function after 4 weeks. Embodiment 43. The method according to any one of embodiments 1-42, wherein the method improves hearing function after 12 weeks. Embodiment 44. The method according to any one of embodiments 1-43, wherein the individual is human. Embodiment 45. One or more nucleic acids encoding COL18A1, SPARC, and IGFBP2. Embodiment 46. The one or more nucleic acid of embodiment 45, wherein the one or more nucleic acids encoding COL18A1 is one or more nucleic acids encoding endostatin. Embodiment 47. One of more LNPs comprising the one or more nucleic acids of embodiment 45 or 46. Docket No.: 33612-20007.40 / 00633-0396WO1 Embodiment 48. The one or more nucleic acids according to embodiment 45 or 46, wherein the nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein is each operably linked to an expression control element. Embodiment 49. The one or more nucleic acids according to embodiment 48, wherein the expression control element is a promoter sequence. Embodiment 50. The one or more nucleic acids according to embodiment 48, wherein the promoter sequence is selected from the group consisting of a constitutive promoter sequence, an inducible promoter sequence, and a tissue-specific promoter sequence. Embodiment 51. The one or more nucleic acids according to embodiment 50, wherein the promoter sequence is a constitutive promoter sequence. Embodiment 52. The one or more nucleic acids according to embodiment 51, wherein the constitutive promoter sequence is selected from the group consisting of MYO15A promoter sequence, a MYO7A promoter sequence, a MYO6 promoter sequence, a STRC promoter sequence, a Pou4f3 promoter sequence, a OTOF promoter sequence, and a SLC26A5 promoter sequence. Embodiment 53. The one or more nucleic acids according to embodiment 52, wherein the constitutive promoter sequence is a CMV promoter sequence. Embodiment 54. The one or more nucleic acids according to embodiment 50, wherein the promoter sequence is a tissue-specific promoter sequence. Embodiment 55. The one or more nucleic acids according to embodiment 54, wherein the tissue-specific promoter sequence is selected from the group consisting of a stria vascularis cell-specific promoter sequence, a hair cell-specific promoter sequence, a supporting cell-specific promoter sequence, and a spiral ganglion neuron-specific promoter sequence. Embodiment 56. The one or more nucleic acids according to embodiment 48, wherein the expression control element comprises an enhancer sequence; optionally, wherein the enhancer sequence is a CMV enhancer sequence. Embodiment 57. The one or more nucleic acids according to embodiment 46-56, wherein the nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein are located on the same viral vector. Embodiment 58. The one or more nucleic acids according to embodiment 57, wherein the viral vector comprises nucleic acid encoding a cleavable linker between each of the sequence encoding COL18A1 protein, the sequence encoding SPARC protein, and the sequence encoding IGFBP2 protein. Docket No.: 33612-20007.40 / 00633-0396WO1 Embodiment 59. The one or more nucleic acids according to embodiment 58, wherein the cleavable linker is a self-cleaving peptide selected from the group consisting of F2A, P2A, T2A, and E2A. Embodiment 60. The one or more nucleic acids according to embodiment 59, wherein the cleavable linker is a P2A peptide. Embodiment 61. The one or more nucleic acids according to embodiment 60, wherein the P2A peptide has 1, 2, or 3 amino acid substitutions or insertions or deletions compared to the amino acid sequence of SEQ ID NO: 14. Embodiment 62. The one or more nucleic acids according to embodiments 58-61 wherein three separate proteins are produced from translation of the nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein. Embodiment 63. The one or more nucleic acids according to embodiments 57-62, wherein the AAV vector comprises an Igκ leader sequence (SEQ ID NO: 13) at the 5’ end of the nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein. Embodiment 64. A vector comprising the nucleic acids according to any one of embodiments 45-63. Embodiment 65. The vector according to embodiment 64, wherein the vector is a viral vector. Embodiment 66. The vector according to embodiment 64 or 65, wherein the vector is an AAV vector. Embodiment 67. A pharmaceutical composition comprising an effective amount of COL18A1, SPARC, and IGFBP2 proteins. Embodiment 68. The pharmaceutical composition according to embodiment 67, wherein the COL18A1 protein has at least 90% identity to the amino acid sequence of SEQ ID NO: 2 or 4. Embodiment 69. The pharmaceutical composition according to embodiment 67 or 68, wherein the SPARC protein has at least 90% identity to the amino acid sequence of SEQ ID NO: 6 or 8. Embodiment 70. The pharmaceutical composition according to any one of embodiments 67-69, wherein the IGFBP2 protein has at least 90% identity to the amino acid sequence of SEQ ID NO: 10 or 12. Docket No.: 33612-20007.40 / 00633-0396WO1 EXAMPLES The presently disclosed subject matter will be better understood by reference to the following Examples, which are provided as exemplary of the invention, and not by way of limitation. Abbreviations: AAV2 (adeno-associated virus type 2); ABR (auditory brainstem response); ARHL (age-related hearing loss); dB (decibel); IHC(s) (inner hair cell(s)); NIHL (noise- induced hearing loss); OHC(s) (outer hair cell(s)). Example 1: Endostatin, SPARC, and IGFBP2 (“CSI”) for Protection and Treatment of Noise Induced Hearing Loss and Age-Related Hearing Loss: Methods and Materials Example 1 describe the material and methods for a study showing that a combination of three secreted proteins (endostatin, SPARC, IGFBP2; together referred to as “CSI”) induced axon outgrowth and enhanced the survival of spiral ganglion neurons and preserving the synapses in the cochlear hair cell, which resulted in the restoration of wave I amplitude in a mouse model. The results of the study are described in Example 2. Single-cell RNA library preparation and sequencing The sensory epithelium from cultured whole cochlea explants was dissociated into single cells using 0.25% trypsin-EDTA for a 15-minute incubation at 37°C. The tissues were gently pipetted up and down in complete medium (DMEM / F12 with 10% FBS) to singularize the cells. The cell suspension was strained twice through a 40 μm mesh into 5 ml of medium and centrifuged at 300 g for 5 minutes to pellet the cells. The pellet was resuspended in 50 μl of complete medium and loaded onto a 10X Chip-G for a target recovery of approximately 2000 cells. Single cells from cochlea explant cultures of the Dox (n=8) and Dox_Atoh1 (n=5) groups were separately captured to prepare two libraries. The single-cell suspensions were immediately transferred to individual channels of a Chromium Next GEM Chip G to generate gel beads in emulsion (GEMs) together with the reverse transcription master mix. This was performed using the droplet-based high-throughput Chromium Controller (10X Genomics), facilitating cell lysis, individual cell barcoding, and reverse transcription using the Chromium Next GEM Single Cell 3’ Kit v3.1 (10X Genomics, PN-1000128). Single-cell library preparation followed the manufacturer's protocol, which included the following steps: emulsion breakage, cDNA amplification, fragmentation, end repair, A-tailing, oligo adapter ligation, Illumina sample index addition, and double-sided size selection. The quality of the single-cell-derived cDNA libraries was assessed with a High Sensitivity DNA Kit on an Agilent 2100 Bioanalyzer. Subsequent to library preparation, next-generation sequencing was Docket No.: 33612-20007.40 / 00633-0396WO1 conducted using paired-end 150 bp reads on the Illumina HiSeqXTen instrument (Novogene Inc.). Computational analysis of single-cell RNA-seq data Raw reads were processed using the Cell Ranger pipeline (version 5.0.0; 10X Genomics) following the manufacturer’s guidelines. This process includes alignment, filtering, barcode counting, unique molecular identifier (UMI) quantification for determining gene transcript counts per cell, quality control, demultiplexing, and mapping to the mouse reference genome (mm10, GRCm38) using the Spliced Transcripts Alignment to a Reference (STAR) algorithm (PMID: 23104886). The count matrices generated by Cell Ranger were loaded into Seurat v3.2.2 (PMID: 31178118) for downstream analysis. Outlier cells were identified based on three metrics: library size, number of expressed genes, and mitochondrial content. Cells with fewer than 200 UMIs or mitochondrial content higher than 20% were filtered out. Datasets from each sample were integrated into a single matrix for comparative analysis using the {Merge} function from Seurat’s integration workflow. Global normalization was performed using the {SCTransform} function in Seurat, which normalizes and scales UMIs and mitochondrial content. The SCTransform method identifies cross- dataset pairs of cells in a matched biological state (‘anchors’), which are used to correct technical variabilities between samples (batch effect correction) and perform comparative single-cell RNA-seq analysis across experimental conditions. For unsupervised clustering, 0.5 was chosen as the resolution parameter. Downstream analysis steps on the integrated dataset included {SelectIntegrationFeatures}, {PrepSCTIntegration}, {FindIntegrationAnchors}, {IntegrateData}, {RunPCA}, {RunUMAP}, {FindNeighbors}, and {FindClusters} functions. Differentially expressed cell type-specific markers were identified using the {FindAllMarkers} function in Seurat. Cells from each cluster were compared to detect uniquely expressed genes. Only genes enriched and expressed in a minimum of 10% of each population (min.pct = 0.1) with a log fold change larger than 0.25 (logfc.threshold = 0.25) were considered. A Wilcoxon rank sum test was used for this analysis. A differentially expressed gene was defined as any gene expressed in at least 10% of cells, with a p-value <0.001 and a 1.5x average fold change compared to other clusters. Additionally, cell clusters were required to have at least 10 uniquely differentially expressed genes compared to other clusters. Differentially expressed genes per cluster were visualized using the {DimPlot} and {FeaturePlot} functions in Seurat, displaying results in a tSNE plot. Violin plots were generated using the {VlnPlot} function to examine gene expression across Docket No.: 33612-20007.40 / 00633-0396WO1 clusters, and a heatmap of the top 20 most differentially expressed genes was created using the {DoHeatmap} function. Cell types were annotated through extensive literature review and analysis of specific gene expression patterns. Gene Ontology analysis GO analysis and identification of differentially expressed genes belonging to the class of secreted genes was performed by using DAVID v6.8 (david.ncifcrf.gov / ) (PMID: 19131956, PMID: 19033363). Venn diagrams were generated by intersection of gene lists obtained from a publicly available online webtool provided by the department of Bioinformatics & Evolutionary Genomics at University of Ghent (bioinformatics.psb.ugent.be / webtools / Venn / ). Real-time PCR The cochlear tissues including the organ of Corti, spiral limbus, and lateral wall were dissected out from the temporal bones of cultured adult cochleae. Total RNA was extracted from cochleae using RNeasy Mini Kits from Qiagen (Valencia, CA). cDNA was synthesized using reverse transcriptase (Takara) with random hexamers. Real-time PCR amplification reactions were carried out using QuantiTeck SYBR Green PCR kit (Takara) on an ABI StepOnePlus system (Applied Biosystem) with StepOne software V2.3. All reactions were carried out in duplicates with the expression of the gene normalized using Gapdh as the endogenous housekeeping control gene. Expression of the target gene mRNA relative to that of the housekeeping mRNA was analyzed by a Relative Expression Software Tool (REST 2009, v.2.0.13). Spiral ganglion neuron explant culture All experimental procedures were approved by the Institutional Animal Care and Use Committee of Massachusetts Eye and Ear and conducted in accordance with the NIH Guide for the Care and Use of Laboratory Animals. CD-1 pups of both sexes at postnatal day 5 to 7 (P5-7) were euthanized and decapitated, and cochleae were extracted from temporal bones and otic capsules in ice-cold Hank’s Balanced Salt Solution (HBSS, Life Technologies). The stria vascularis and organ of Corti were carefully removed and the modiolus containing SGNs was kept intact after removing bony wall, if any. Only the apex and middle turns were used for each culture in the studies to avoid differing region-specific effects of test molecules, which were then cut into equal portions of four pieces in ice-cold Docket No.: 33612-20007.40 / 00633-0396WO1 HBSS. The SGN explants were placed in a 24-well on sterilized round glass coverslips precoated with Poly-D-Lysine / Laminin (Corning #354087), to which they adhere. The divided modiolus pieces containing SGNs were allowed to attach to glass coverslips overnight in 50 ml of attachment medium containing DMEM / F12 – HEPES (ThermoFisher #11330032), GlutaMax (ThermoFisher # 35050061), 5% FBS (ThermoFisher #A3160602), 1X N2 (Gibco #17502048), 1X Pen / Strep (Gibco #15070063) at 37oC, 5% CO2. The cultures were evaluated the following day using light microscope for adhesion to PDL / Laminin substrate. Explants that failed to adhere to the coverslip were discarded; all others were used for these experiments. After tissue attachment was confirmed under the microscope, SGNs were treated with recombinant proteins in 500 ml of growth medium containing Neurobasal (Gibco # 21103049), GlutaMax, Pen / Strep, 1X N2, 1X B27 (Gibco # 17504044). Recombinant BDNF (50 ng / ml; 1.85 nM; ABCAM #ab9794) was used as positive control to compare the effect of the test molecules on neurite outgrowth. Recombinant test proteins were reconstituted in sterile PBS supplemented with 0.1% BSA at 0.1 mg / ml (wt / vol) concentration that were considered as 100X stock solutions, and further diluted in neuronal growth medium at a final concentration of 1 mg / ml during treatments. Recombinant proteins used for in vitro treatment of SGN explants as follows: Recombinant mouse SPARC (R&D Systems #942-SP-050), recombinant mouse Endostatin (cleaved-COL18A; R&D Systems #570-ES-050), recombinant mouse IGFBP-2 (R&D Systems #797-B2-025), recombinant mouse Galectin-1 (LGALS1; R&D Systems #1245-GA-050), recombinant human Fibulin-2 (FBLN2; R&D Systems #9559-FB-050), recombinant human Olfactomedin-like protein 2B (OLFML2B; R&D Systems #5614-NL-050), and recombinant mouse Glypican-1 (GPC1; R&D Systems #4520-GP-050). Cultures were kept in a humidified incubator at 5% CO2and 37 °C for additional 48 hrs after adding test molecules until the day of fixation. Recombinant DNA constructs In this study, pAAV2.CMV.EGFP.WPRE.bGH (plasmid #105530; Addgene), which contains the entire sequence of WT AAV2, was used as host vector in which EGFP cassette was replaced with transgenes. To construct recombinant AAV2 plasmid, a synthetic gBlock (2640 bp, IDT) containing the target sequences in the following order was designed: Igκ signal peptide (METDTLLLWVLLLWVPGSTGD; SEQ ID NO:13), mouse endostatin (C- terminal 184 residues of mouse collagen 18a1; 1344-1527 aa, Gene ID #12822, CCDS #48604.1), P2A-I (ATNFSLLKQAGDVEENPG↓P; SEQ ID NO:14), full length mouse Igfbp2 (305 aa, Gene ID: # 16008, CCDS #15036.1), P2A-II, full length mouse Sparc (302 Docket No.: 33612-20007.40 / 00633-0396WO1 aa, Gene ID #20692, CCDS #24712.1) and HA-tag (YPYDVPDYA; SEQ ID NO:15). gBlock was inserted into the downstream of CMV promoter in the host plasmid linearized by NotI and BamHI enzymes via In-Fusion HD enzyme (Takara), which fuses gBlock DNA fragment and linearized vector precisely by recognizing a 15 bp overlap at their ends. The resultant vector, named as AAV2-CMV-Igk.Col18a1.Igfbp2.Sparc.HA (AAV.C-S-I), harboring CSI expression cassettes driven by the cytomegalovirus (CMV) promoter, consisting of an intron in upstream, and the woodchuck hepatitis virus posttranscriptional regulatory element (WPRE) followed by the bovine growth hormone (bGH) polyA termination sequence in the downstream of the open reading frame. Gene cassettes of C, S and I were separated by inserting a P2A self-cleavage sequence to facilitate individual protein expression. All constructs were fully sequenced prior to transfections to ensure no mutations were introduced during cloning. All the plasmids were propagated in DH5a E. coli cells and plasmids were extracted using endo-free plasmid purification kits (Qiagen). AAV production For recombinant AAV production, transgene vectors were packaged at the University of Massachusetts Medical School, Viral Vector Core, as previously described (PMID: 32457039). In brief, HEK293 cells maintained in DMEM with GlutaMAX, penicillin / streptomycin and 10% FBS were co-transfected with packaging plasmid (pRep2 / Cap2; Agilent Technologies), adenovirus helper plasmid (pHelper; Agilent Technologies), and rAAV plasmid carrying mouse CSI expression cassette flanked by AAV2 ITRs. The pRep2 / Cap2 expresses regulatory and capsid proteins of AAV2 serotype, which excises the recombinant genome from the rAAV vector plasmid, replicate the viral ssDNA genome, and package the genome into AAV virions. Adenovirus E2A, E4 and VA RNAs expressed from the pHelper provides helper functions essential for rAAV rescue, replication, and packaging (PMID: 34695545). HEK293 cells were harvested 72 hours post-transfection, suspended in a lysis buffer containing 50 mM Tris-HCl pH 8.0, 150 mM NaCl, and 1 mM MgCl2, and lysed by three freeze-thaw cycles. Unencapsidated plasmid DNA was digested using 250 U Benzonase (Sigma-Aldrich), and virions were purified by ultracentrifugation on standard CsCl gradient and desalted by dialysis. The vectors are quality control tested by ddPCR titration for DNase-resistant vector genome (vg) concentration using probe and primers targeting the bGH region, and AAV purity was assessed by 4%–12% SDS- acrylamide gel electrophoresis followed by silver staining (Invitrogen). Viral genomic titer for AAV.C-S-I was calculated as 2.10E+12 GC / ml. Docket No.: 33612-20007.40 / 00633-0396WO1 Cell culture and transient expression HEK293 cells (Human embryonic kidney; ATCC #CRL-1573) were cultured in DMEM including high glucose containing 10% FBS (ThermoFisher Scientific), 2 mM GlutaMAX and Pen / Strep (ThermoFisher Scientific) according to the manufacturer’s instructions. Cultures were used for overexpression of test molecules followed by coimmunoprecipitation, immunoblotting, or immunostaining. Briefly, HEK293 cells were plated at the density of 1x106per 10-cm petri dishes and transduced at a MOI of 1x105with AAV.EGFP or AAV.C-S-I. Once needed, cells were transfected with 5 mg of expression plasmid using 10 ml of Lipofectamine-2000 (Invitrogen). Forty eight hours after transfection or transduction, cells were harvested for the purposes indicated above. Preparation of conditioned medium (CM) Day before transduction, HEK293 cells at the density of 1x105were seeded onto 35- mm dishes in culture medium containing DMEM supplemented with 10% FBS, 2 mM GlutaMAX and Pen / Strep. Then, cells were transduced with AAV.C-S-I (MOI: 100K) diluted in fresh culture medium. Twenty-four hours after transduction, culture medium was replaced with neuronal growth medium containing Neurobasal, GlutaMax, Pen / Strep, 1X N2, 1X B27, and cells were cultured for 48-hour more. Supernatants collected from the culture of transduced HEK293 cells were centrifuged, passed through 0.22 mM syringe filters, aliquoted, snap frozen in liquid nitrogen, and stored at -80oC until use. At the time of experiment, CM aliquots were thawed on ice, and mixed 1-to-1 (v / v) with neuronal growth medium to supply fresh medium components, and warmed up to 37oC in water bath before treating SGN explants in 24-well plates. For the immunoblot analysis of target molecules, CM aliquots were concentrated ten times using Amicon Ultra centrifugal filters (MWCO 10kDa) before loading to polyacrylamide gels. Co-immunoprecipitation and immunoblotting To determine whether protein-protein interactions occur among C, S and I proteins, pAAV.C-S-I plasmid was transfected into HEK293 cells cultured in 10-cm dishes. Forty- eight hours after transfection, cells transiently expressing endostatin, IGFBP2, and SPARC- HA molecules were lysed in 1 ml normal buffer (NB) containing 10 mM Tris.HCl pH 7.4, 150 mM NaCl, 0.25% Triton X-100, Protease / Phosphatase inhibitor cocktails (Roche), 1 mM PMSF, and 100U / ml Turbonuclease; or stringent buffer (SB) containing the same Docket No.: 33612-20007.40 / 00633-0396WO1 components, but 500 mM NaCl, 0.5% Triton X-100. Freshly prepared and centrifuge-cleared cell lysates were incubated with 25 ml anti-HA magnetic beads (Pierce, #88836) for 1 h at 4oC with gentle rotation. Beads then were washed six times with 1 ml NB, or SB to remove unbound proteins. Bead-bound proteins were denatured by boiling in 40 ml 1X NuPAGE LDS sample buffer (Invitrogen #NP0007) at 70oC for 10 minutes before SDS-PAGE. Ten ml of the boiled-sample was loaded for probing with target antibodies. The total protein lysate (input; 0.5%) used as expression control and the IP fractions were analyzed by immunoblotting. Proteins were resolved on NuPAGE 4-12%, Bis-Tris polyacrylamide gels (Invitrogen #NP0326BOX) in NuPAGE™ MOPS SDS running buffer (Invitrogen #NP0001), transferred to Invitrolon PVDF membrane (0.45 μm; Invitrogen #LC2005), and probed with primary antibodies against mouse monoclonal anti-HA.11 epitope tag antibody (1:1000; BioLegend #901501), goat polyclonal COL18A1 (endostatin) antibody (1:1000; R&D Systems #AF570), goat polyclonal SPARC antibody (1:1000; R&D Systems #AF942), goat polyclonal IGFBP2 antibody (1:1000; R&D Systems #AF797), mouse monoclonal anti-N- Cadherin (1:500; BD #610920) and mouse monoclonal GAPDH antibody (1:1000; Thermo Fisher #MA5-15738). Primary antibodies were detected by appropriate HRP conjugated secondary antibodies (from Cell Signaling Technology, or Santa Cruz; 1:1000). Blots were visualized by ChemiDoc Imaging System (Bio-Rad). Immunostaining for SGN explants After fixation with 4 % paraformaldehyde for 10 min at room temperature and two washes with ice-cold PBS, the explants were permeabilized with 0.25 % Triton X-100 (Sigma-Aldrich) for 5 min, washed twice with PBS and blocked for nonspecific antibody binding with 5% normal donkey serum (Sigma-Aldrich # D9663) and 0.25% Triton X-100 in 1X PBS for 1 hour at RT. Neurites were labeled for neurofilament using chicken polyclonal anti-neurofilament heavy chain (NFH) primary antibody (1:100; Aves Lab #NFH), mouse monoclonal TUJ-1 (Tubb3) primary antibody (1:100, BioLegend #801201), and neuronal cell bodies (soma) were stained with mouse monoclonal HuD (ELAVL4) primary antibody (1:200, Santa Cruz #sc-48421). After primary antibody incubation overnight at 4 °C, followed by two PBS washes, the neurites stained by 1 h of incubation with Alexa fluorophore-conjugated secondary antibodies (1:500; Invitrogen) against the species of the respective primary antibody. After another washing step, coverslips containing each explant fluorescently labeled to detect SGN somata and axons were mounted on glass slides. Images were taken with Leica SP8 confocal microscope via 10X lens. Neurite outgrowth from the Docket No.: 33612-20007.40 / 00633-0396WO1 SGN was evaluated by measuring the number and / or length of NFH-positive axonal projections. First, images were imported to FIJI software (NIH, Bethesda, MD, USA) and axonal growth is quantified as the average number of NFH-labeled fibers, and only neurites with visible start and endpoints were traced. Average number of neurites, neurite lengths and neuronal cell bodies were compared between treatments. Morphological analyses for SGN explants Neurite evaluations was carried out as previously described (PMID: 19266211) with slight modifications. Neurite outgrowth of immunostained SGN explants was assessed by counting the number and length of the processes or the number of somas in individual explants. A process being longer than the diameter of the explant to be counted. The number of surviving neurons was counted as HuD-labeled cells. When morphology could be clearly assessed, neuronal soma was then scored as monopolar or bipolar. The percentages of different morphologies were calculated from the total number of neurons that were scored unambiguously for each explant as previously described (PMID: 17331652). Confocal images of the explants taken using a Leica SP8 microscope were exported as lif. files. The numbers and length of fluorophore-labeled neurites were quantified using ImageJ Fiji software. By means of a computer mouse, every neurite was traced by a series of straight lines leading from the edge of the explant to the axon tip. The length of each of these lines was determined by the curvatures of the neurite such that the line always overlaid the neurite. The number of branches on an individual neurite was not counted due to their inconsistence appearance across experimental replicates. Almost all neurites from each explant were evaluated, which were compared based on only their number or length. Image editing was limited to sizing, orientation, minor adjustments to contrast and lighting. Statistical analysis for SGN elongation All experiments were performed at least three times. The Prism 9 statistical package (GraphPad Software, Inc) was used in data processing. Statistical analyses were performed with pairwise comparisons using a nonparametric 2-tailed Mann-Whitney test. Values were expressed as mean ± SEM. To be considered statistically significant, p values had to be less than 0.05. Docket No.: 33612-20007.40 / 00633-0396WO1 Data availability Raw and pre-processed single-cell RNA-seq data that support the findings of this study are available through the National Center for Biotechnology Information Gene Expression Omnibus (NCBI GEO) (Dox: GSE205187; Dox_Atoh1: GSE218325). R scripts for data processing are available through github.com / ZYChenLab / CSI. There is no restriction on the use of the code or data for non-profit academic organizations. Noise exposure Mice were exposed to free-field noise, awake and unrestrained, in a small reverberant chamber. An 1-20 kHz octave-band noise was presented for 2 h at 97 dB SPL. The exposure stimulus was generated by a custom white-noise source, filtered (Brickwall Filter with a 60 dB / octave slope), amplified (Crown power amplifier), and delivered (JBL compression driver) through an exponential horn fitted securely to a hole in the top of a reverberant box. Acoustic testing Mice were anaesthetized with xylazine (10 mg / kg, intraperitoneally (i.p.)) and ketamine (100 mg / kg, i.p.). Acoustic stimuli were delivered through a custom acoustic assembly consisting of two miniature dynamic electrostatic earphones (CDMG15008-03A, CUI) to generate primary tones and a miniature microphone (FG-23329-PO7, Knowles) to record ear-canal sound pressure near the eardrum. Custom LabVIEW software controlling National Instruments 24-bit soundcards (6052E) generated all ABR and DPOAE stimuli and recorded all responses. For ABR measurements, needle electrodes were inserted at the vertex and ventral edge of the pinna, with a ground reference near the tail. ABR potentials were evoked with 5- ms tone pips (0.5-ms rise–fall with a cos2 onset, delivered at 35 / s). The response was amplified 10,000-fold, filtered (100 Hz–3 kHz passband), digitized, and averaged (1,024 responses) at each SPL. The sound level was raised in 10 dB steps from 20 dB below threshold up to 100 dB SPL at frequencies from 5.66-45.24 kHz (in half-octave steps). Following visual inspection of stacked waveforms, “threshold” was defined as the lowest sound pressure level (SPL) at which any wave could be detected. In general, thresholds were defined by three independent observers. Wave 1 amplitude was defined as the difference Docket No.: 33612-20007.40 / 00633-0396WO1 between the average of the 1-ms pre-stimulus baseline and the wave 1 peak (P1), after additional high-pass filtering to remove low-frequency baseline shifts. For DPOAE measurements, the cubic distortion product was measured in response to primaries f1 and f2. The primary tones were set so that the frequency ratio (f2 / f1) was 1.2 and so that the f2 level was 10 dB below the f1 level. For each f2 / f1 primary pair, primaries were swept in 5-dB steps from 20 dB SPL to 80 dB SPL (for f2). At each level, the amplitude of the DPOAE at 2f1–f2 was extracted from the averaged spectra, along with the noise floor. Threshold was computed by interpolation as the f2 level required to produce a DPOAE at 5 dB SPL. Induction of CSI in vivo. All surgical procedures were done in a clean, dedicated space. Instruments were thoroughly cleaned with 70% ethanol and autoclaved before surgery. Mice were anaesthetized by intraperitoneal injection of xylazine (10 mg / kg) and ketamine (100 mg / kg). Before surgery, the postauricular region of the mice was gently shaved and disinfected by 10% povidone-iodine. With round window membrane injection method, a post-auricular incision was made using small scissors. The soft tissues were bluntly dissected to expose the bulla. A small hole was created in the bulla with a 25-gauge needle and enlarged with forceps to expose the round window (RW) membrane. A Nanoliter Microinjection System (Nanoliter2000; World Precision Instruments, Sarasota, FL, USA) was used in conjunction with a glass micropipette to load AAV-CSI / or CSI protein into the glass micropipette. The total volume of virus delivered was 1 μl for each injection at the speed of 300 nl / min. After injection, the round window membrane was sealed by tissue adhesive. After injection, the tubing was cut and sealed by tissue adhesive. The skin was closed with 7 / 0 suture (Surgical Suture, B0117B72). Adult cochlear culture and viral infection in vitro Different from the neonatal cochlear culture method, in which the cochleae were disassociated from the bone, adult mouse whole cochleae (4-6 weeks old) were dissected with the bone attached. The bulla was first removed from the skull and dipped in 75% ethanol for 5 mins before being placed in HBSS. The vestibular region was also removed. Under a dissecting microscope, the middle ear, vessels and the debris were removed from the bulla. The bone covering the apical turn was removed, and round window and oval window membranes were opened to allow media exchange with the cochlear fluids. The ligament Docket No.: 33612-20007.40 / 00633-0396WO1 portion and Reissner’s membrane at each end of the cochlea were also removed to facilitate the access of medium to the sensory epithelial region. The cochleae were maintained in DMEM / F12 (Invitrogen) supplemented with N2 and B27 (both from Invitrogen). Ad.Atoh1- mCherry adenoviruses was purchased from the SignaGen Laboratories, Rockville, MD, with atiter of 6 × 1010pfu / ml.The blockade of the neuronal fiber regrowth after HC regeneration in vitro. Briefly, after 4 day culturing, cochlea were treated with VPA (1.5 mM), LiCl (8mM), FSK (20μM), siFIR (0.02 μM), and siMxi1 (0.02 μM) for 4 days, followed by Ad.Atoh1.mCherry (4 × 108pfu / ml) infection overnight for the HC regeneration andwere delivered following the manufacturer’s instructions . To block the regrowth of the neuronal fibers to the OHCr, the CSI antibodies were added 2 days after Ad.Atoh1mCherry infection at the concentration (each antibody) of 0.1μg / ml; 0.4μg / ml and 0.8μg / ml for 7-9 days. No CSI antibody treated, and IgG (2.4μg / ml) treated groups were used as positive controls. Cochleae were harvested and decalcified before immunohistochemistry. Organotypic culture of the neonatal cochlea and the Synaptic injury model (Excitotoxic trauma model) (Kainic acid administration) CD-1 pups of both sexes at postnatal day 5 to 7 (P5-7) were euthanized and decapitated, and cochleae were extracted from temporal bones and otic capsules in ice-cold HBSS supplemented with 1x HEPES. The cochlear nerve trunks were removed, and the cochlear explants were dissected into three parts, apical, middle and basal, and placed in 10mm slides positioned in 35mm dish with 4 rings, which had already been coated with Matrigel Matrix (#354248, Corning). The cochlear explants were maintained in culture medium containing DMEM / F12 containing GlutaMax, 1x N2 and 1x B27 overnight at 37℃ with 5% CO2. Explants that failed to adhere to the coverslip were discarded on the second day; all others were used for these experiments. To build synaptic injury model, cochlear explants were treated with 1mM kainic acid (K0250, Sigma) in DMEM / F12 for 2 hours. Then the cochlear explants were treated with / without different recombinant proteins right after or 24 hours after kainic acid treatment, in culture medium. The recombinant proteins include Recombinant mouse SPARC (1μg / ml), recombinant mouse Endostatin (1μg / ml), recombinant mouse IGFBP-2 (1μg / ml), Recombinant BDNF (50 ng / ml), Recombinant NT-3 (50 ng / ml; R&D Systems #267-N3-025). Cultures were kept in a humidified incubator at 5% Docket No.: 33612-20007.40 / 00633-0396WO1 CO2 and 37 °C for additional 24 or 48 hrs. Next the cochleae were fixed in 4% paraformaldehyde at room temperature for 30 min, followed by immunofluorescent staining. Immunohistochemistry for neonatal and adult mouse cochlea For adult mouse cochlea tissue: Mouse cochleae was fixed in 4% paraformaldehyde at 4°C overnight, followed by decalcification in 120 mM EDTA for 24 hours. The decalcified cochlea was used for whole mount immunohistochemistry following a standard procedure. All specimens (adult mouse cochlea / and neonatal mouse cochlea) were incubated at room temperature for 30 min in 10% donkey / goat serum with 0.5% Triton X-100 for blocking. Primary antibodies used in this research were Parvalbumin (1:100, Cell Signaling) and Tuj1 (1:100, BioLegend); and the synapses were stained with mouse (IgG1) anti-Ctbp2 (1:300, BD Biosciences #612044). Species- specific Alexa Fluor-conjugated secondary antibodies were used for detection. Confocal microscopy and Statistical analysis Confocal microscopy was performed using a Leica TCS SP8 with Leica Application Suite Advanced Fluorescence (LAS AF) software V2.6.0. Sequential scanning with different laser channels was used for image acquisitions. Confocal images were processed using ImageJ package (imagej.nih.gov\ij). For Z-stacks, equal numbers of images of adjacent optical sections, 0.1 mm in thickness, were used for processing with identical parameters, including median filtering and adjustment of brightness and contrast, between experimental and control groups. The Prism 10 statistical package (GraphPad Software, Inc) was used in data processing. To count HC-like cells regenerated in the OHCr in vitro, outer hair cell related regions from the apex (80- 100% of the length of cochlear duct from the hook) were included. Briefly, new HC-like cells generated from Hensen cell region, Deiters' cell region, outer pillar cell region were included in the "OHCr" region. TuJ1+ neuronal fibers length from Hensen cell region, Deiters' cell region, outer pillar cell region were included in the "OHCr" region as well. ANOVA analysis with Turkey’s multiple comparisons test was used to compare three or more groups (p<0.05 was considered significant). To measure the ctbp2 number per IHC in the Synaptic injury model, the confocal images were taken from apical, middle and basal parts of each cochlear explant. Prior to quantification, a mean filter (radius = 1 pixels) was applied to each image using Fiji (ImageJ) Docket No.: 33612-20007.40 / 00633-0396WO1 to reduce high-frequency noise and enhance puncta detection. This preprocessing step helped improve the accuracy of subsequent particle counting. The “pathological” temporary threshold shift (TTS), as hearing shift would typically resolve within 12-14 hours after noise exposure (or blast), has an underlying pathology of synaptopathy with reduced connections between inner hair cells (IHC) and auditory nerve fibers (Kujawa, S.G. & Liberman, M.C. J Neurosci 29, 14077-14085 (2009); Furman, A.C., Kujawa, S.G. & Liberman, M.C. J Neurophysiol 110, 577-586 (2013)), also referred to as “hidden hearing loss” due to normal audiometric thresholds. Guest, H., Munro, K.J. & Plack, C.J. Hear Res 356, 116-117 (2017); Hickman, T.T., Smalt, C., Bobrow, J., Quatieri, T. & Liberman, M.C. Sci Rep 8, 10740 (2018). Further, synaptopathy was also detected after noise exposures that caused significant permanent threshold shift (PTS) and HC damage. Cho, S.I. et al. PLoS One 8, e67618 (2013); Kurioka, T. et al. Characteristics of laser-induced shock wave injury to the inner ear of rats. J Biomed Opt 19, 125001 (2014). Deficiencies in speech perception in noise and reduced wave 1 amplitudes, the hallmark of synaptopathy, have been observed in military personnel after noise exposure. Pinsonnault-Skvarenina, A. et al. J Acoust Soc Am 152, 2419 (2022). It has been suggested that synaptopathy can be comorbid or interact with age-related hearing loss (ARHL) and auditory pathologies. Kamerer, A.M. et al. Examining physiological and perceptual consequences of noise exposure. J Acoust Soc Am 146, 3947 (2019). Also see, Cochlear synaptopathy in acquired sensorineural hearing loss: Manifestations and mechanisms. doi.org / 10.1016 / j.heares.2017.01.003; Cochlear Synaptopathy and Noise-Induced Hidden Hearing Loss dx.doi.org / 10.1155 / 2016 / 6143164 Example 2: Endostatin, SPARC, and IGFBP2 (“CSI”) for Protection and Treatment of Noise Induced Hearing Loss and Age-Related Hearing Loss: Results Example 2 describe the results for a study showing that a combination of three secreted proteins (endostatin, SPARC, IGFBP2; together referred to as “CSI”) induced axon outgrowth and enhanced the survival of spiral ganglion neurons and preserving the synapses in the cochlear hair cell, which resulted in the restoration of wave I amplitude in a mouse model. The materials and methods for the study are described in Example 1. Identification of cellular heterogeneity in Atoh1-induced hair cell regeneration. The ability to regenerate hair cells (HCs) in the adult cochlea is a prerequisite for potential development of treatment for sensorineural hearing loss. Previous studies Docket No.: 33612-20007.40 / 00633-0396WO1 demonstrated that co-activation of Myc and Notch is sufficient to reprogram fully differentiated adult cochlear cells including supporting cells (SCs), which efficiently transdifferentiate to hair cells in response to induction signals by overexpression of Atoh1. See, Shu, Y. et al. Nat Commun 10, 5530 (2019) and Quan, Y.Z. et al. Reprogramming by drug-like molecules leads to regeneration of cochlear hair cell-like cells in adult mice. Proc Natl Acad Sci U S A 120, e2215253120 (2023). It was observed that hair cell regeneration coincides with peripheral axon outgrowth from the spiral ganglion neurons (SGNs), (Li, W. et al. A Novel in vitro Model Delineating Hair Cell Regeneration and Neural Reinnervation in Adult Mouse Cochlea. Front Mol Neurosci 14, 757831 (2021); Shu 2019 and Quan 2023) indicating that the preexisting SGNs react to signals from the newly regenerated hair cells (FIG.1A). In the previous study, Dox treatment followed by Atoh1 (Dox_Atoh1) induction results in substantial conversion of SCs to HCs and the formation of new peripheral axons, whereas Dox induction alone fails to reproduce the phenotype1. To identify secreted molecules produced by regenerated HCs, single-cell RNA-seq (scRNAseq) was performed to capture transcriptomic changes between ex vivo cultured cochleae with and without HC regeneration at the single cell resolution (Dox vs. Dox+Atoh1) (FIGs.1B-1C). By bioinformatics analysis, nine distinct clusters were identified including Kolliker’s organ cells (KO), unclassified-supporting cells (uSC), interdental cells (IdC), Claudius cells / outer sulcus cells (CCOS), hair cells-like cells (HCLC), Reissner’s membrane cells (RMC), Hensen’s cells (HeC), hair cells (HC) and Deiter’s cells (DC) (FIGs.1D-1F). Identification of secreted molecules expressed during hair cell regeneration To identify putative secreted molecules, differential gene expression analysis was performed using default parameters in Seurat, and pathway enrichment analysis to characterize the molecular signature of these sample groups. By comparing the changes in the transcriptome of the Dox_Atoh1 versus Dox sample, more than 1,000 genes were observed to exhibit variable expression across samples (FIG.2A), suggesting that Atoh1 induction greatly affects overall gene expression profiles in either way. Additionally, the analysis identified thirty genes annotated as secreted. (FIG.2B). To gain better insight into these genes, the expression profiles of more than 30 genes coding for secreted proteins across sample groups were visualized using a violin plot, confirming higher expression levels in Dox_Atoh1 compared to the Dox group (FIG.2C). To narrow down the list, the following criteria were applied: (i) transcriptional regulation by Atoh1 (PMID: 32211215), (ii) Docket No.: 33612-20007.40 / 00633-0396WO1 supporting literature on their neurobiological functions (FIG.3A), (iii) average fold changes in mRNA expression with Dox_Atoh1 / Dox > 1.2 as assessed by real-time PCR (FIG.3B), and (iv) commercial availability as recombinant proteins. In vitro ganglion culture was used to study the function of the candidates in neurite outgrowth. For the initial screening, postnatal SGN explants were treated in vitro with 30 recombinant proteins individually at a concentration of 1 μg / mL, following the manufacturer's instructions. BDNF was used as a positive control. Forty-eight hours post-treatment, immunohistochemistry for NFH-positive neurites was performed. NFH-positive neurites per explant were quantified using ImageJ-Fiji software (FIG.3C). Two secreted molecules named GAL (Galanin) and CCK (Cholecystokinin) were excluded due to the lack of commercial availability. Each molecule reconstituted at 100 μg / mL in sterile PBS containing 0.1% BSA as carrier. Based on manufacturer’s recommendation the ED50, effective dose of treatment for a desired pharmacologic effect of each protein was used as 1 μg / ml. All recombinant proteins were purchased from R&D Systems. Details are described as follow: - Recombinant Mouse SPARC Protein, #942-SP-050 - Recombinant Mouse Endostatin Protein, #570-ES-050 - Recombinant Mouse IGFBP-2 Protein, #797-B2-025 - Recombinant Mouse Galectin-1 (LGALS1) Protein #1245-GA-050 - Recombinant Human Fibulin 2 (FBLN2) Protein, #9559-FB-050 - Recombinant Human Olfactomedin-2 (OLFML2B) Protein #5614-NL-050 - Recombinant Mouse Glypican 1 (GPC1) Protein, #4520-GP-050 The experiment showed that, despite the presence of HuD+ surviving neurons, none of the tested molecules individually promoted spiral ganglion neurite outgrowth to the same extent as BDNF (FIG.3D). The lack of outgrowth of spiral ganglion neurites by single protein suggested that more than one secreted molecule may work in a concerted manner to promote the outgrowth. Computational analysis using the STRING database was performed to evaluate co-expression profiles of the initial set of nine genes, as supported by literature studies (FIG.4A). The analysis suggested the potential co-acting genes in combinations including: Dual combinations: endostatin + IGFBP2, endostatin + SPARC, endostatin + FBLN2, SPARC + LGALS1; Triple combinations: endostatin + SPARC + IGFBP2, endostatin + SPARC + FBLN2, COL18A1 + IGFBP2 + FBLN2; and Quadruple combination: endostatin + SPARC + FBLN2 + IGFBP2 (FIG.4B). The co-expression analysis-guided dual combinations were tested in SGN explant cultures. No effect on neurite outgrowth was detected (FIGs.4C, 4D). Docket No.: 33612-20007.40 / 00633-0396WO1 The combinations of three (triple) and four (quadruple) proteins were further tested. Triple proteins endostatin, SPARC, and IGFBP2 proteins (CSI) showed a superior neurotrophic effect, resulting in improved axon outgrowth, while other combinations did not have a significant effect (FIGs.5A-5C). To determine if the effect on neurite outgrowth is CSI- specific, an antibody blocking assay was conducted by pre-incubating the CSI proteins with serial dilutions of their respective antibodies. The concentration-dependent blocking assay, using a mixture of antibodies against endostatin, SPARC, and IGFBP2, showed a significant reduction in the number of axons per explant at day 2 in vitro. No change was observed when CSI was preincubated with a non-specific IgG antibody (FIGs.6A, 6B). Further investigation in the P14 mouse cochlear explants indicated that the CSI treatment preserved CtBP2+ synaptic ribbons in the hair cell against excitotoxic injury (FIGs.7A, 7B). In FIG7B, the control groups is “NK” and “NK+BSA,” while the test group is “NK+CSI.” Approximately double the amount of CtBP2 (staining for synapse) in the “NK+CSI” group was identified compared to the “NK” and “NK+BSA” groups. If the CtBP2 level in the “NK” group is considered as 100%, the level in the CSI-treated group can be estimated as around 200%. Taken together, the study identified a novel combination of three secreted molecules CSI by regenerated HCs with potent neurotrophic effect on promoting axon outgrowth, outperforming other combinations. AAV-mediated CSI gene therapy for synaptopathy and hidden hearing loss. Given the effect on neurite outgrowth in vitro stimulated by CSI, the roles of CSI in vivo in mouse models with hearing impairment were further investigated. First, an AAV plasmid encoding the expression cassettes for endostatin, SPARC, and IGFBP2, driven by the CMV promoter, and packaged in AAV2 (AAV2-CMV-CSI) was constructed. This construction is possible because the combined size of three genes is less than 4.7 kb required for AAV packaging (FIG.8A). First, the production of each of the CSI proteins by AAV transduction was evaluated in vitro. HEK293 cells were infected with AAV2-CMV-CSI and performed Western blot analysis using the proteins collected from the conditioned medium (CM). Western study demonstrated that all three proteins were detected by the specific antibody against each protein, confirming that all three proteins are secreted (FIG.8B). Inner ear injection of AAV2-CMV-CSI was performed in mice exposed to noise and studied the effect. Adult CBA / Caj mice (6-week-old) were exposed to the noise of 97 dBoctave band of 1-20 kHz for 2 hours and injected with 1 l of AAV2-CMV-CSI one day later. Docket No.: 33612-20007.40 / 00633-0396WO1 Hearing was tested 3 weeks after injection. Under the noise exposure condition, Temporary Threshold Shift (TTS) is produced with synaptic defect known as synaptopathy, and wave 1 amplitudes are significantly reduced. In the ears injected with AAV2-CMV-CSI, wave 1 amplitudes were significantly higher across all frequencies tested than non-injected control ears (FIGs.9A-9C), demonstrating the treatment effect of AAV2-CMV-CSI. In some frequencies (5.66 and 8 kHz), wave 1 amplitudes were restored to the same level as pre-noise exposed ears. These results demonstrate that locally delivered AAV-Triplet recovered hearing and neuronal activities, in part of restoring synapses. AAV-mediated expression of transgenes is detectable at least one week after injection, indicating an effective treatment by CSI one week after noise exposure. Protein-based CSI treatment for noise-induced synaptopathy To further investigate the effect of CSI, a study by CSI protein delivery in mice exposed to noise that produces synaptopathy was performed (FIG.10A-10C). The CSIproteins, in a mixture of individual proteins at a concentration of 0.3 g / ml for each protein,were injected into the inner ear through the round window with canal fenestration, one day after noise exposure. Significant recovery of the wave 1 amplitudes in all frequencies was detected except for 32 kHz three weeks after injection. This result supports the development of CSI protein therapy to treat synaptopathy and hidden hearing loss in patients with hearing loss caused by acoustic trauma. AAV-mediated CSI gene therapy for age-related hearing loss (ARHL). The potent treatment effect of synaptopathy by the combination of CSI suggests that these proteins may be effective as treatment for age-related hearing loss. Inner ear injection by AAV2-CMV-CSI was performed in the C57 / Bl6 mice (an age-dependent hearing lossmouse model). 1 l of AAV2-CMV-CSI was injected into the adult (6-week-old) C57 / Bl6.Hearing was studied 4 and 12 weeks after injection. In the ears injected with AAV2-CMV- CSI, wave 1 amplitudes were significantly higher across almost all frequencies tested than noninjected control ears (FIGs.11A-11B), demonstrating the treatment effect of AAV2- CMV-CSI. CSI therefore function to rescue wave 1 amplitudes in mice with ARHL. The decreased wave 1 amplitudes are considered to represent word recognition deficiencies, which are most frequently associated with patients with ARHL. Docket No.: 33612-20007.40 / 00633-0396WO1 AAV-mediated endostatin gene expression alone for treatment of synaptopathy Whether the endostatin gene expression alone is sufficient for the treatment of synaptopathy caused by noise exposure was further investigated. Inner ear injection with AAV2-endostatin (AAV2-C) in mice exposed to noise was performed. Adult (6-week-old) CBA / Caj mice were exposed to the noise of 97 dB octave band of 1-20 kHz for 2 hours andinjected 1 l AAV2-C one day later. Hearing was studied 3 weeks after injection. Under thenoise exposure, TTS (transient threshold shift) is produced with synaptopathy, and wave 1 amplitudes are significantly reduced. In the ears injected with AAV2-C, wave 1 amplitudes did not show significantly difference across all the frequencies comparing to the noninjected control ears (FIGs.12A-12C), demonstrating the endostatin expression alone is not sufficient for the treatment of synaptopathy. Thus, the rescue of wave 1 amplitudes seen in the CSI combination (by gene therapy or protein delivery) demonstrates the synergistic effect by CSI, which cannot be achieved by endostatin expression alone. Confirming the CSI signal for neurite outgrowth upon Atoh1-induced hair cell regeneration. This study has developed a reproducible in vitro model to enable research of the biological mechanism of hair cell regeneration on ganglion neurites outgrowth (Shu 2019; Li 2021; Quan 2023) (FIG.13). Specifically, all outer hair cells (OHCs) die, and the neuronal fibers retract from the OHC region in this adult mouse cochlea explant culture system (FIGs. 14A-14B). To further confirm that the CSI molecules secreted by the new HC-like cells are responsible for the neurite outgrowth, the CSI antibody blockade assay in the adult HC regeneration model was performed. Upon the Cocktail (VLFsiFsiM) / Ad.Atoh1.mCherry treatment (Quan 2023) , the HC regeneration and the neurites outgrowth in the OHC region were detected, indicating that the new HC-like cells secrete signals to stimulate the outgrowth of the retracted neurites. To confirm the hypothesis that the CSI proteins represent the exact signal from the HC-like cells, the blockade assay with the CSI specific antibodies was performed. The experiment showed that in the presence of CSI specific antibodies, the neuronal fibers could no longer regrow even after HC regeneration, demonstrating that the CSI signal is essential for neurite outgrowth in the adult cochlear explant culture (FIGs.15A-15C). Docket No.: 33612-20007.40 / 00633-0396WO1 Preservation of synapse survival by CSI treatment Further investigation in the P5 mouse cochlear explants indicated that the CSI treatment preserved CtBP2+synaptic ribbons and the neurites-hair cell binding against excitotoxic injury (FIGs.16A-19B). To build a synaptic injury model, cochlear explants were treated with 1mM kainic acid (KA) (K0250, Sigma) in DMEM / F12 for 2 hours. The cochlear explants were treated with or without recombinant CSI proteins immediately after or 24 hours after kainic acid treatment. The recombinant CSI proteins include recombinant mouse SPARC (1μg / ml), recombinant mouse Endostatin (1μg / ml), recombinant mouse IGFBP-2 (1μg / ml), Recombinant BDNF (50 ng / ml), Recombinant NT-3 (50 ng / ml; R&D Systems #267-N3-025). Cultures were kept in a humidified incubator at 5% CO2and 37 °C for additional 24 or 48 hrs (FIG.16A). The CSI treatment for 48 hrs preserved the inner hair cell (IHC) synapse (CtBP2+ dots) (FIGs.16B and 16C; FIGs.17A-17E); The CSI treatment for 24 hrs preserved the IHC synapse (CtBP2+ dots) and the neurites connected to each IHC (TUJ1+ label) (FIGs.16D-16F; FIGs.18A-18C). According to FIG.16C, the negative control group is “KA 48H,” and the positive control group is “KA+CSI.” Double the amount of CtBP2 in the “KA+CSI” group compared to the “KA” group. By considering the CtBP2 level in the “KA” group as 100%, the level in the CSI-treated group can be approximated as 200%. Further CSI antibody blockade assay confirmed the preservation effect by CSI against the KA induced excitotoxic injury in the P5 neonatal mouse cochlear explant model (FIGs. 19A-19B). Taken together, these results demonstrate that the combination of CSI proteins (endostatin, SPARC, and IGFBP2) provided robust protection of inner hair cell synapses against excitotoxic injury in an in vitro model of auditory neuropathy. Example 3: Endostatin, SPARC, and IGFBP2 (“CSI”) in Oubain-Induced Hearing Loss Model This example shows that a combination of three secreted proteins (endostatin, SPARC, IGFBP2; together referred to as “CSI”) cause neurofiber regeneration in vivo following cochlear denervation in a mouse model. The materials and methods for the preparation of the CSI treatment are described in Example 1. Ouabain treatment was performed on CBA / CaJ mice aged 6–8 weeks. Unilateral cochlear denervation was achieved by applying ouabain solution to the round window niche, See Lang et al. (2011). Sox2 up-regulation and glial cell proliferation following degeneration of spiral ganglion neurons in the adult mouse inner ear. J Assoc Res Otolaryngol 12, 151- Docket No.: 33612-20007.40 / 00633-0396WO1 171. and Lang et al. (2005). Ouabain induces apoptotic cell death in type I spiral ganglion neurons, but not type II neurons. J Assoc Res Otolaryngol 6, 63-74. Oubain-induced hearing loss was used as a model for age realated hearing loss. Mice were anesthetized with intraperitoneal ketamine (100 mg / kg) and xylazine (10 mg / kg). A postauricular skin incision was made behind one ear, and the underlying muscles and facial nerve were separated by blunt dissection to expose the middle compartment of the bulla. A small opening was made to access the round window niche. Ouabain (1-2 μL of 1 mM in distilled water) was applied to the round window membrane using a 10 μL Hamilton syringe. The solution was wicked off and replaced with fresh ouabain every 10 minutes for a total of 1 hour. The bulla was then sealed with dental cement, and the skin incision was closed using 7-0 nonabsorbable sutures. Mice were placed on a 37 °C homeothermic blanket for recovery. Postsurgical analgesia with buprenorphine and meloxicam, along with subcutaneous saline, was administered every 20–24 hours for 3 days. The contralateral ear served as an untreated control. All procedures were approved by the Massachusetts Eye and Ear Institutional Animal Care and Use Committee (IACUC, Protocol #2022N000065). The schematic diagram showing the procedures can be seen in FIG.20. AAV-CSI was delivered (as described in Example 1) to the mouse inner ear two weeks post ouabain treatment, and 10 weeks later, robust neurofibers were seen extending from the modiolus into the cochlear sensory region, with prominent NF labeling. Neurofiber was stained with Chicken anti-Neurofilament H Antibody (1:500, Millipore ab5539), synapses were stained with mouse (IgG1) anti-Ctbp2 (1:300, BD Biosciences #612044), and myosin was stained with Rabbit anti-MYO7A Antibody (1:400, Proteus Biosciences, 25-6790) (MYO7A). As shown in FIG.21A, AAV-CSI treatment results in robust neurofiber regeneration in ouabain-treated cochlea in vivo. In untreated mice (Wild type, WT) the results showed lower neurofibers at much lower intensity than in CSI treated cochlea. In WT, NF-labeled neurofibers were numerous but with very low NF expression. After CSI treatment, many neurofibers were regenerated with intense NF, indicating that nerve regeneration had taken place. However, the actual neurofiber number in the CSI- treated cochlea was lower than that of the WT. FIG.21B shows a quantification of neurofiber number. The quantification showed significantly more neurofibers in the CSI delivered cochlea after ouabain treatment. Docket No.: 33612-20007.40 / 00633-0396WO1 OTHER EMBODIMENTS It is to be understood that while the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims.

Claims

Docket No.: 33612-20007.40 / 00633-0396WO1 CLAIMS What is claimed is:

1. A method for preventing and / or treating hearing loss in an individual in need thereof comprising administering a composition comprising one or more viral vectors comprising a nucleic acid encoding a COL18A1 protein, a SPARC protein, and an IGFBP2 protein, wherein the one or more viral vectors is administered to the inner ear of the individual.

2. The method of claim 1, wherein the one or more viral vectors is selected from the group consisting of a lentivirus vector, an adenovirus vector, an adeno-associated virus (AAV) vector, a herpes simplex virus (HSV) vector, and a retrovirus vector.

3. The method of claim 2, wherein the one or more viral vectors comprises an AAV vector.

4. The method of any one of claims 1-3, wherein the nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein is each operably linked to an expression control element.

5. The method of claim 4, wherein the expression control element is a promoter sequence.

6. The method of claim 5, wherein the promoter sequence is selected from the group consisting of a constitutive promoter sequence, an inducible promoter sequence, and a tissue-specific promoter sequence.

7. The method of claim 6, wherein the promoter sequence is a constitutive promoter sequence.

8. The method of claim 7, wherein the constitutive promoter sequence is selected from the group consisting of MYO15A promoter sequence, a MYO7A promoter sequence, a MYO6 promoter sequence, a STRC promoter sequence, a Pou4f3 promoterDocket No.: 33612-20007.40 / 00633-0396WO1 sequence, a OTOF promoter sequence, and a SLC26A5 promoter sequence.

9. The method of claim 8, wherein the constitutive promoter sequence is a CMV promoter sequence.

10. The method of claim 6, wherein the promoter sequence is a tissue-specific promoter sequence.

11. The method of claim 10, wherein the tissue-specific promoter sequence is selected from the group consisting of a stria vascularis cell-specific promoter sequence, a hair cell-specific promoter sequence, a supporting cell-specific promoter sequence, and a spiral ganglion neuron-specific promoter sequence.

12. The method of claim 4, wherein the expression control element comprises an enhancer sequence; optionally, wherein the enhancer sequence is a CMV enhancer sequence.

13. The method of any one of claims 1-12, wherein the nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein are located on the same viral vector.

14. The method of claim 13, wherein the viral vector comprises nucleic acid encoding a cleavable linker between each of the sequence encoding COL18A1 protein, the sequence encoding SPARC protein, and the sequence encoding IGFBP2 protein.

15. The method of claim 14, wherein the cleavable linker is a self-cleaving peptide selected from the group consisting of F2A, P2A, T2A, and E2A.

16. The method of claim 15, wherein the cleavable linker is a P2A peptide.

17. The method of claim 16, wherein the P2A peptide has 1, 2, or 3 amino acid substitutions or insertions or deletions compared to the amino acid sequence of SEQ ID NO: 14.Docket No.: 33612-20007.40 / 00633-0396WO1 18. The method of claim any one of 14-17, wherein three separate proteins are produced from translation of the nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein.

19. The method of any one of claims 13-18, wherein the AAV vector comprises an Igκ leader sequence, optionally comprising SEQ ID NO: 13, at the 5’ end of the nucleic acid encoding COL18A1 protein, SPARC protein, and IGFBP2 protein.

20. The method of any one of claims 1-19, wherein the COL18A1, SPARC, and IGFBP2 proteins are secreted by inner hair cells or outer hair cells of the individual.

21. A method for preventing and / or treating hearing loss in an individual in need thereof comprising administering an effective amount of COL18A1, SPARC, and IGFBP2 proteins or an effective amount of an mRNA encoding the COL18A1 protein, an mRNA encoding the SPARC protein and an mRNA encoding the IGFBP2 protein to the inner ear of the individual.

22. The method of claim 21, wherein the effective amount of an mRNA encoding the COL18A1 protein, an mRNA encoding the SPARC protein and an mRNA encoding the IGFBP2 protein are administered in a lipid nanoparticle (LNP).

23. The method of any one of claims 1-22, wherein the COL18A1 protein is endostatin.

24. The method of any one of claims 1-23, wherein the COL18A1 protein has at least 90% identity to the amino acid sequence of SEQ ID NO: 2 or 4.

25. The method of any one of claims 1-24, wherein the SPARC protein has at least 90% identity to the amino acid sequence of SEQ ID NO: 6 or 8.

26. The method of any one of claims 1-25, wherein the IGFBP2 protein has at least 90% identity to the amino acid sequence of SEQ ID NO: 10 or 12.

27. The method of any one of claims 23-26, wherein the COL18A1, SPARC, and IGFBP2 proteins are administered at about a 1:1:1 ratio.Docket No.: 33612-20007.40 / 00633-0396WO1 28. The method of any one of claims 1-27, wherein the hearing loss is age-related hearing loss.

29. The method of any one of claims 1-28, wherein the hearing loss is noise-induced hearing loss.

30. The method of any one of claims 1-29, wherein the composition or the COL18A1, SPARC, and IGFBP2 proteins are administered directly to the inner ear of the individual.

31. The method of any one of claims 1-30, wherein the composition or the COL18A1, SPARC, and IGFBP2 proteins is administered through the round window membrane of the ear of the individual.

32. The method of any one of claims 1-31, wherein the composition or the COL18A1, SPARC, and IGFBP2 proteins is administered to both ears of the individual.

33. The method of any one of claims 1-32, wherein the method results in regeneration of the axons of spiral ganglion neurons and / or synapses between cochlear hair cells and peripheral nerve fibers, and / or enhances axon outgrowth of spiral ganglion neurons in the individual’s cochlea, thereby restoring hearing function.

34. The method of any one of claims 1-33, wherein the method increases neuronal fiber length in the outer hair cell region of the individual’s cochlea.

35. The method of any one of claims 1-34, wherein the method enhances survival of spiral ganglion neurons.

36. The method of any one of claims 1-35, wherein the method enhances neurite-inner hair cell binding.Docket No.: 33612-20007.40 / 00633-0396WO1 37. The method of any one of claims 1-36, further comprising determining the Wave I amplitude of the individual at a frequency before and after the individual has received the composition or the COL18A1, SPARC, and IGFBP2 proteins.

38. The method of claim 37, wherein the method results in a higher Wave I amplitude at a frequency in the individual who is administered the composition or the COL18A1, SPARC, and IGFBP2 proteins as compared to the Wave I amplitude at the same frequency in an individual who is not administered the composition or the COL18A1, SPARC, and IGFBP2 proteins.

39. The method of claim 37, wherein the Wave I amplitude in response to 20-100 dB SPL stimulus is higher in the individual who is administered the composition or the COL18A1, SPARC, and IGFBP2 proteins as compared to the Wave I amplitude in response to 20-100 dB SPL stimulus in an individual who is not administered the composition or the COL18A1, SPARC, and IGFBP2 proteins.

40. The method of any one of claims 1-39, further comprising determining speech recognition of the individual before and after the individual has received the composition or the COL18A1, SPARC, and IGFBP2 proteins.

41. The method of claim 40, wherein the method improves speech recognition of the individual as compared to speech recognition of the individual prior to administration of the composition or the COL18A1, SPARC, and IGFBP2 proteins.

42. The method of any one of claims 1-41, wherein the method improves hearing function after 4 weeks.

43. The method of any one of claims 1-42, wherein the method improves hearing function after 12 weeks.

44. The method of any one of claims 1-43, wherein the individual is human.

45. One or more nucleic acids encoding COL18A1, SPARC, and IGFBP2.Docket No.: 33612-20007.40 / 00633-0396WO1 46. The one or more nucleic acids of claim 45, wherein the one or more nucleic acids encoding COL18A1 is one or more nucleic acids encoding endostatin.

47. One or more LNPs comprising the one or more nucleic acids of claim 45 or claim 46.

48. A vector comprising the nucleic acids of claim 45 or claim 46.

49. The vector of claim 48, wherein the vector is a viral vector.

50. The vector of claim 48 or 49, wherein the vector is an AAV vector.

51. A pharmaceutical composition comprising an effective amount of COL18A1, SPARC, and IGFBP2 proteins or mRNA encoding the COL18A1, SPARC, and IGFBP2 proteins.

52. The pharmaceutical composition of claim 51, wherein the COL18A1 as endostatin.

53. The pharmaceutical composition of claim 51 or 52, wherein the COL18A1 protein has at least 90% identity to the amino acid sequence of SEQ ID NO: 2, 17 or 18.

54. The pharmaceutical composition of any of claims 51-53, wherein the SPARC protein has at least 90% identity to the amino acid sequence of SEQ ID NO: 6 or 8.

55. The pharmaceutical composition of any one of claims 51-54, wherein the IGFBP2 protein has at least 90% identity to the amino acid sequence of SEQ ID NO: 10 or 12.

Images