Zheng-Yi Chen - US grants

Sensory hair cells are essential for the transduction of mechanical stimuli into hearing and balance signals in the inner ear. Damage to these hair cells is the major cause of sensorineural hearing impairment, which affects more than 17 million Americans. Recent studies have demonstrated in mammals the capability of damaged hair cells to regenerate in vitro. These observations provide a rationale for the treatment of hearing impairment by regenerating hair cells. This treatment is plausible since several neurotrophic molecules (i.e. TGF-alpha, retinoic acid, and bFGF) have been implicated in the regenerative process. Furthermore, BDNF and NT-3 are believed to maintain the survival of spiral ganglion neurons (which innervate hair cells), and may also be able to aid in hearing restoration. As a first step towards a possible therapy for hearing impairment, this project proposes to use viral vectors to deliver a reporter gene to the inner ear of guinea pigs, and to assess the efficiency of the delivery, the toxicity of the vectors, and the expression pattern of the marker gene within different cell types. Four replication-defective viral vectors (HSV, adenovirus, AAV, and retrovirus), all bearing the marker gene lac Z will be injected into the round window of the inner ear. High titer vectors will be used since the total volume of the cochlear fluid in the guinea pig is small (approx. 3-4 microliters). After injection, the animal will be sacrificed at 2 days or l week, and the expression of the lac Z gene within the cochlea will be evaluated histochemically. A second goal of the project is to assess the capability of the four viral vectors to deliver neurotrophic molecules to the inner ear in the hope that they may be useful in repairing hair cells or spiral ganglion neutrons that have been damaged by ototoxic drugs. Different vectors will contain cDNA constructs for BDNF, NT-3, and bFGF. After injection, the animals will be sacrificed at 2 days or l week, and surviving hair cell and spiral ganglion neurons will be counted. These models have clinical application, as they will lay the groundwork for the development of a gene therapy for hearing loss.

In this application we propose to use microarray technology to study genes controlled by Brn-3.1, a hair-cell- specific transcription factor. The traditional way of studying transcription factor targets has been time consuming and laborious. It also lacks a means of simultaneously isolating multiple genes under the control of, a transcription factor. The microarray approach provides an ideal route for such a task. In addition the combination of human and other genome projects along with microarray technology should shed light on functional pathways controlled by Brn-3.1 and other transcription factors. Three major studies will be carried out. The first is to use three complementary approaches, based on the oligonucleotide array technology, to isolate candidate genes controlled by Brn- 3.1. The tetracycline induction of Brn-3.1 in a human osteosarcoma cell line will provide robust control of production of Brn-3.1, and allow us to survey the greatest number of human genes (35,000) for Brn-3.1 targets. The transfection of the organ of Corti cell line will likely identify the targets which may require the co-factors for Brn-3.1 regulation. The comparison of expression profiles of Brn-3.1 knock-out mouse utricles with control may reveal both direct and in-direct target genes regulated by Brn-3.1. The combination of these approaches will build enough redundancies to ensure the isolation of the candidate genes. The second aim is to systematically characterize the activation and binding on the regulatory regions of candidate genes by Brn- 3.1. Cluster analysis will be used to identify the expression pattern of candidate genes during development. Other genes in the clusters will be examined for the binding site of Brn-3.1 such that more candidate genes will emerge. In addition the 5' regulatory regions of the genes within the same cluster as Brn- 3.1 targets will be examined to identify the shared binding motifs for other important transcription factors. In the third aim of the grant the target genes will be studied in relation to their expression in the hair cell of normal and Brn- 3.1 knock-out mice; in order to provide the casual relation between their expression and the onset of Brn-3.1. Antibodies will be used to localized the proteins in the hair cells. The success of the project will provide information regarding genes and their functional pathways controlled by Brn-3.1, and should establish a model for studying hair cell development controlled by other transcription factors such as Math1.

DESCRIPTION (provided by applicant): One of the major causes for deafness is the loss of inner ear hair cells, the sensory cells that detect sounds. In mammals, hair cells are born during embryonic development and are maintained in quiescence throughout life. Mammalian inner ear, unlike the counterpart in lower vertebrates such as chick or fish, does not regenerate hair cells after damage. Deletion of negative growth genes (Rb1 and p27kip1) led to cell cycle re-entry in embryonic and neonatal inner ear. However, we have shown in adult inner ear, Rb1 deletion is not sufficient to induce proliferation. Further, the proliferating hair cells and supporting cells will ultimately die. Thus the inability to re-enter cell cycle by mature inner ear and apoptosis of proliferating cells present two main challenges to hair cell regeneration. This proposal is designed to specifically address the two issues. We showed that FGF signaling is necessary for hair cell regeneration in zebrafish neuromasts. For the specific aim 1, we will test the hypothesis that, with an inducible mouse model, FGF activation with Rb1 deletion could lead to cell cycle re-entry in adult inner ear in vivo. We have observed such events in vitro. Correlation of FGF activity in the proliferating cells by immunostaining will support the crucial role of FGF in cell cycle re-entry. Using an inducible mouse model to mark supporting cells genetically, we will identify hair cells derived from supporting cell transdifferentiation. In the aim 1b, we will characterize regenerated hair cells in differentiation, synapse formation and function by immunostaining, FM1-43 uptake and transduction current recording. In the second aim, we will evaluate two pathways, IGF1 and p53, for their roles in survival and apoptosis of proliferating hair cells. We have the evidence that IGF1 activation or p53 blockade protect Rb1-/- cochlear hair cells from apoptosis. In the aim 2a, we will determine the necessity of IGF1 in Rb1-/- cochlear hair cell survival by blocking IGF1 function to induce apoptosis. Further, we will specifically block two IGF1 signaling pathways: PI3K/Pdk1/Akt and Raf/Mek/Erk, to assess their respective role in the survival of Rb1-/- cochlear hair cells. In the aim 2b, we will block p53 function by a specific inhibitor and correlate p53 inactivation with Rb1-/- cochlear hair cell survival. We will further study the effects on the p53 pathway after IGF1 activation or inhibition. Inactivation of the p53 pathway after IGF1 activation, or vice versa, is an indication that IGF1 antagonizes p53 function to promote survival. Finally we will induce cell cycle re-entry in the adult inner ear on the p53-null background, and study long-term survival of proliferating hair cells. Cell cycle re-entry in mature inner ear and the survival of proliferating hair cells will make it possible to regenerate functional hair cells.

Project Summary and Relevance Application of genome editing based therapy requires efficient delivery of editing agents into disease-relevant tissues and cells. Identification of novel delivery materials targeting somatic cells will greatly facilitate the advance of editing based therapy strategies to clinic. We propose to screen a large library of novel lipid nanoparticles for RNP (ribonucleoprotein) delivery of editing agents into the mammalian sensory organ inner ear. Inner ear is an ideal sensory organ to develop new delivery strategies. It consists of multiple differentiated somatic cell types without effective delivery options. Gene mutations in the major inner ear cell types have been associated with genetic hearing loss, which affects one in 500 newborns and currently has no effective therapies. Lipid-based nanoparticle carriers have emerged as one of the most promising materials for delivery and have been successfully used in clinical applications. We have developed a combinatorial library approach to synthesize degradable lipid-like nanoparticles under reductive intracellular environments, and capable of delivering biomolecules with high efficiency and low toxicity. The new bioreducible lipid nanoparticles (bLNPs) have been used to deliver genome editing agents with high efficiency and low toxicity in vivo. We have delivered genome editing RNP by cationic liposomes into mammalian inner ear in vivo, and rescued hearing in mouse models of human genetic hearing loss. To develop editing based therapies to treat diverse forms of genetic hearing loss, it is essential to develop a delivery strategy to target multiple inner ear cell types simultaneously. The mammalian inner has a complicated structure with multiple cell types in small numbers, making it particularly challenging to screen a delivery technology by conventional high-throughput strategies. The lack of a method to detect editing at the level of the individual cell type further hinders our ability to apply this technology in wildtype large animal models that are essential for development of this therapy for clinical application. By combining our strategies to screen nanoparticles for delivery of editing materials to X-linked genes in the male mouse inner ears in vivo, we will overcome these hurdles for effective delivery and editing in diverse inner ear cell types. The study of the human inner ear tissues ex vivo will provide evidence of the relevance of nanoparticle delivery in human disease-relevant tissues. Expansion of our work to large animal models will be a major step towards clinical application of this technology. Our approach with nanoparticles can be applied to the study in other organs requiring somatic cell type editing and in wildtype large animals.

Project Summary and Relevance Genetic hearing loss affects one in 1000 newborns and contributes significantly to general populations with hearing loss. Over 130 deafness loci have been mapped and more than 80 have been cloned. Despite the tremendous progress in deafness gene discovery, there is no medical treatment for genetic hearing loss. There is an urgent and unmet medical need to develop treatment for genetic hearing loss. CRISPR/Cas9-mediated genome editing is transforming biomedical research and with a promise of becoming new treatment of disease. It enables the application of nuclease with guide RNA to pair with and modify DNA permanently, which can be developed into new therapies for wide range of diseases. We have successfully used transient and in vivo local delivery of editing agents to treat a mouse model, Beethoven, of human dominant hearing loss DFNA36. In this application, we propose two broad aims to further develop CRISPR technology for inner ear editing and to apply it to treat genetic hearing loss. First, we will characterize CRISPR nuclease variants SaCas9 and Cpf1 for hair cell editing by direct RNP (ribonucleoprotein) delivery. This study will expand deafness mutations to be targeted by genome editing due to additional PAM (protospacer adjacent motif) sequences and nuclease activities, with a possibility of improvement in editing efficiency and specificity. We will apply whole-genome application (WGA) using purified hair cells for high- throughput sequencing (HTS) and to identify insertions and deletions (indels) in hair cells, and will correlate hair cell editing efficiency with the outcome of hearing rescue. We will evaluate editing in mature hair cells by RNP delivery, which is relevant to treatment in patients as human inner ears are fully mature. We will screen new nanoparticle based liposomes to provide the options for efficient inner ear delivery and editing. In second aim, we will perform hearing rescue studies by RNP delivery of editing agents to disrupt mutations in mouse models of human dominant deafness, including 1). Hearing rescue by editing Oblivion (Obl) mutation in the Pmca2 gene that affects outer hair cells; 2). Hearing rescue by editing a mutation in the microRNA 96 (Mir96) that results in delayed onset progressive hearing loss. Hearing rescue in those models will demonstrate general therapeutic application of genome editing targeting dominant mutations of hair cell origin. The study will delineate the relationship between editing efficiency, specificity and off-target effect with the extent of hearing rescue and evaluate long-term rescue effect and the outcome of intervention at late stages. The proposal has potential to be developed as new platform for genome editing based therapy for genetic deafness. Genetic hearing loss affects large number of children and contributes to adult deafness with no treatment available. We are developing CRISPR/Cas9-mediated genome editing as a new treatment platform to target dominant mutations by transient local delivery into inner ear, with the potential to treat diverse forms of genetic hearing loss.