Inner Ear Hair Cell Regeneration — Maybe We Can Know More

Yes I mean exo-AAV
I guess they have to approve exo-AAV to use for human without damage. Thats why aaron said 'no'
In addition I couldn't see any detail for how anc80 work. Maybe It is different from AAV virus(Atoh)

 

Yes. And there is already a thread about the Usher syndrome paper which includes links to the actual papers.

György et al. looked at AAV1 and exo-AAV1 with an application to "a mouse model of hereditary deafness (lipoma HMGIC fusion partner-like 5/tetraspan membrane protein of hair cell stereocilia [Lhfpl5/Tmhs−/−])" while Pan et al. looked at Anc80L65 in a mouse model of Usher 1C. György et al. found exo-AAV1 had higher rates of transduction than AAV1 and Pan et al. found Anc80L65 had higher transduction rates than AAV 1, 2, 6, 8, and 9. Pan et al. report rates of 100% for IHC and 95% for OHC. György et al. report rates of 95% for IHC and > 50% for OHC. Most importantly Pan et al. report much, much larger improvements in hearing. In fact, given the improvements in transduction, the György et al. hearing results are somewhat disappointing ("The results also raise some important questions that should be addressed in future studies. First, it is unclear why the high efficiency of hair cell transfection did not yield better functional improvements. More detailed assessment of hair cell structure and function may reveal subtle defects that remain in the treated ear." http://www.cell.com/molecular-therapy-family/molecular-therapy/fulltext/S1525-0016(17)30010-2). So I am tempted to say that this shows Anc80L65 is better than exo-AAV1. However, the applications (genetic defects) are different. It would now be good to see Anc80L65 in the Lhfpl5/Tmhs−/− model and/or exo-AAV1 in the Usher 1C model. (This is assuming that each method could handle the other payload.) All of that being said, the results definitely appear to be in favor of Anc80L65 at the moment.

This is my assumption. György et al. do not provide any information on the use of exo-AAVs in humans. In contrast, Anc80L65 has not been used in humans. According to Pan et al. "Finally, while conventional AAV vectors have a good safety profile, next-generation AAVs, such as Anc80L65, have not yet been evaluated in humans and thus will require careful scrutiny." And according to Landegger et al. "Further validation of Anc80L65 as a gene transfer vector for use in human inner ear gene therapy will require targeting-efficiency studies in large-animal models; additional exploration of the window of opportunity for therapeutic intervention; and pharmacology and toxicology studies to investigate the safety of Anc80L65 upon cochlear administration."

Both are complicated, but in my opinion gene therapy is going to prove less complicated than stem cells but has, when used in the context of genetic deafness, more narrow application. The Novartis trial is a bit odd in that gene therapy is being used in an attempt to regenerate hair cells rather than to fix a specific genetic defect. Momentum is picking up for using gene therapy to deal with specific genetic defects, and I think we are on the way to seeing actual treatments for inherited forms of deafness. We are getting to the point that the main issue is going to be the therapeutic window.

 

One of the people in the novartis study, Jeff, did not see improvement, in fact from what I remember in the end it just made his hearing worst, but it was already pretty bad to begin with. He ended getting a cochlear implant shortly after he figured out it didnt do anything for him. I know another girl Amanda noted some improvement. Rob Gerk gained balance, but no hearing improvement from what I read also. These names are talked about on these forums. I don't know of any other participants that made any comments. I think it is all hush hush.

Penate, I know you are frustrated with the forums and answers, and so is everyone else. No one likes this, but reality is we are pretty far away from any type of cure for hearing loss, T, or H. Reading the research is great for hope, but anything that comes out of there must be analyzed, tested, trials must be done, etc. Your best case from a breakthrough to a cure would be 10 years if everything goes right. There are many positives right now, but as you can see from the posts, we are in clinical trials and newer better vectors are coming out. This would start the entire process all over again. At the end of the day we don't know how these therapies will work on humans and they are very careful now. Someone could end up with monster T or worst from any of these gene therapies so it is going to be slow going. There won't be any miracles, but there will be lots of articles to read of the breakthroughs we are experiencing.

 

[QUOTE="No one likes this, but reality is we are pretty far away from any type of cure for hearing loss, T, or H. Reading the research is great for hope, but anything that comes out of there must be analyzed, tested, trials must be done, etc. Your best case from a breakthrough to a cure would be 10 years if everything goes right. There are many positives right now, but as you can see from the posts, we are in clinical trials and newer better vectors are coming out. This would start the entire process all over again. At the end of the day we don't know how these therapies will work on humans and they are very careful now. Someone could end up with monster T or worst from any of these gene therapies so it is going to be slow going. There won't be any miracles, but there will be lots of articles to read of the breakthroughs we are experiencing.[/QUOTE]

@RB2014 do you really think that a less effective "cure" won't be available say at most 8 years from now. Personally I think a treatment will be available to restore hearing to ~45db levels for a lot of frequencies within that timeframe. It maybe experimental sure and not commercially available but something will exist for some people.

 

@RB2014 do you really think that a less effective "cure" won't be available say at most 8 years from now. Personally I think a treatment will be available to restore hearing to ~45db levels for a lot of frequencies within that timeframe. It maybe experimental sure and not commercially available but something will exist for some people.[/QUOTE]

If Novartis goes to clinical trial 3 we could see a less effective solution for hearing loss in the next 3-5 years. No one except for them really knows what worked and what didnt. Maybe they found the dosage that works best. I'm not sure that will get you 45 db though, but again its all speculation. Everyone else is 18 months away from the first clinical trial which could put you in the 8 year range. Now we have better vectors and understand more. I wouldnt be surprised if they scrap that trial since the results are not going to be what they thought, but just the fact that people gained some hearing and balance is a great breakthrough.

I'm personally hoping for faster than that since these startup companies want to see a return on their investment. I don't think 8 years would be unreasonable to get back 45 db or so. What we aren't going to figure out anytime soon is T and H because that would require almost perfect hearing restored for those that suffer because of hearing loss. Some people get t with no hearing loss or only a 10db hearing loss at 8k. I think even the best current animal research gets us to within 25db of 0.

 

Completely agree. As we gain knowledge, the speed of treatment is exponentially increasing.

 

I have 60dB hearing loss on some frequencies. If you get me 45dB back, I might still have T, but a lot less intrusive...

 

It doesn't have to be, but that is its most natural use, and over time I suspect it will be the more common use. The number of people affected by genetic deafness is relatively small. However, editing a single or a few genes will be easier than doing anything with stem cells. The challenge so far with either technology is the short window of opportunity at least as things are understood now.

 

What is the window of opportunity

 

It's a press release so not clear what to make of it, but Decibel is presenting some new results at the ARO meetings:

https://decibeltx.com/wp-content/up...earch-in-Otolaryngology-ARO-Meeting-FINAL.pdf

 

It’s clear Decibel is working on basic research.

Someone from Decibel was lead and/or senior author on two papers/posters: PD 31 and PS 748. They participated in four other posters/papers: PD 32, PS 349, PS 747, and PS 749. Here are the abstracts.

PD 31

An Atlas of Cell-Type-Specific Transcriptomes in the Newborn Mouse Cochlea

Joseph C. Burns1; Adam Palermo1; Michael C. Kelly2; Matthew W. Kelley3 1Decibel Therapeutics; 2Laboratory of Cochlear Development, NIDCD, NIH; 3Laboratory of Cochlear Development, NIDCD, NIH, Bethesda, Maryland, USA

Single-cell transcriptional profiling has emerged as a powerful, unbiased tool for dissecting cellular heterogeneity at the molecular level. qPCR of single cells from the developing inner ear has been used to characterize gene expression patterns critical for lineage specification, differentiation, and subtype identity. RNA-Seq of single cells from the sensory regions of the newborn mouse cochlea and utricle has demonstrated the feasibility of extending these studies to the whole transcriptome level, allowing for more comprehensive identification of genes and pathways. Collecting sufficient numbers of single cells in a cost- and time-efficient manner has been a critical limitation to these techniques, but recent adaptations in droplet microfluidics have improved the throughput 100-fold. Given the relatively small size of the mouse inner ear, this improvement in throughput theoretically permits organ-wide profiling of all the cell types in the cochlea in a single experiment. To test this, we separated the entire cochlear portion of the newborn mouse inner ear, dissociated the tissue, and captured single-cell transcriptomes for RNA-Seq using a droplet microfluidics platform from 10X Genomics. From an input of ~16,000 cells, we captured 4,251 single cells. The capture process took 5 min, demonstrating that tissue dissociation is the only substantial time constraint with this system. Unbiased clustering revealed 23 distinct clusters of cells. Based on expression of known marker genes, 15 of the 23 clusters were identified and included hair cells, supporting cells, neurons, glia, cells of the stria vascularis, interdental cells, and non-sensory epithelial cells of the inner and outer sulcus. Cells within the remaining clusters are likely fibrocytes, mesenchymal cells, immune cells, osteolineage cells, and mesothelial cells; however, their exact localization and identity remains to be determined. At least one cluster of cells could not be ascribed to any known cell type in the cochlea at this stage. The numbers of cells in each cluster were within expected ratios. Differential expression analysis revealed 5,821 genes driving the differences between the 23 cell types, many of which have not previously been localized within the inner ear. This dataset reveals that single-cell RNA-Seq technology has advanced sufficiently that all the major cell types in the cochlea can be profiled in a single experiment. Future applications include comprehensive resolution of cellular fate commitment during development and organ-wide, cell-type-specific responses to insults in adults.

PS 748

Technical Comparison of Four Single-Cell RNA-Seq Methodologies in Newborn Mouse Cochlea

Kathy So1; Matthew Nguyen1; Michael C. Kelly2; Hanna Sherrill3; Joseph C. Mays3; Matthew W. Kelley4; Joseph C. Burns1; Adam Palermo1 1Decibel Therapeutics; 2Laboratory of Cochlear Development, NIDCD, NIH; 3Laboratory of Cochlear Development, NIDCD, Bethesda, Maryland, USA; 4Laboratory of Cochlear Development, NIDCD, NIH, Bethesda, Maryland, USA

A variety of new methodologies are emerging for profiling the levels of transcripts in dissociated single cells, several of which have been applied to inner ear biology. Whole transcriptome profiling of single cells with RNA-Seq is particularly attractive since it allows for unbiased identification of genes that have not been previously implicated in the model under study. There are a number of considerations when evaluating the pros and cons of these methods, including, but not limited to, throughput of single cell capture, fraction of input cells captured, protocol duration, fraction of the transcriptome recovered, and cost per cell. We have systematically compared these parameters for newborn mouse cochlea cells captured on four different platforms: Fluidigm C1 (microfluidics), FACS-based single-cell sorting, Drop-Seq (droplet microfluidics), and 10X Genomics Chromium (droplet microfluidics). The chemistry used for RT-PCR of single-cell mRNA is a major determinant of transcript yield and cost, so we also compared multiple RT-PCR chemistries using the FACS system. RT-PCR was performed according to published protocol with the non- FACS methods. We found that the Fluidigm C1 system has the greatest sensitivity, detecting the highest number of expressed genes per cell on average. However, the Fluidigm system had the lowest throughput, the longest capture duration, and the highest cost. Drop-Seq and the 10X Genomics Chromium systems detected the fewest numbers of genes per cell, but also had the highest throughput and lowest costs per cell. For droplet microfluidics systems, the 10X Genomics Chromium had several important advantages over Drop-Seq, such as higher numbers of genes detected per cell, 10-fold shorter capture durations, and a 4:1 ratio of input cells to captured cells. Interestingly, we found that resolution of cellular heterogeneity appeared to be more dependent upon the number of cells captured than the number of genes detected per cell, suggesting that sequencing large numbers of cells at shallow depth is advantageous for detecting distinct cell types and subtypes. Finally, we found that RT enzyme type and efficient removal of residual primers are critical parameters that should be considered when reverse transcribing and amplifying single-cell mRNA. In summary, we present an in-depth technical guide that should help researchers to select the appropriate methodology for experiments that could be aided by single-cell RNA-Seq.

PD 32

A Comprehensive Map of Mammalian Auditory Sensory Cell Development Generated Using High-Throughput Transcriptional Profiling

Michael C. Kelly1; Joseph C. Burns2; Adam Palermo2; Joseph C. Mays3; Kathryn L. Ellis4; Robert Morell5; Matthew W. Kelley6 1Laboratory of Cochlear Development, NIDCD, NIH; 2Decibel Therapeutics; 3Laboratory of Cochlear Development, NIDCD, Bethesda, Maryland, USA; 4NIDCD, NIH; 5Genomics and Computational Biology Core, NIDCD, NIH, Bethesda, Maryland, USA; 6Laboratory of Cochlear Development, NIDCD, NIH, Bethesda, Maryland, USA

The exquisite structural and functional architecture of the mammalian organ of Corti is generated through a highly coordinated developmental program that involves cell-type specific gene expression. This program is composed of the dynamic and heterogeneous transcriptional states of diverse cells types within the developing organ of Corti. A characterization of these various states could provide powerful insight into the genes involved in the development of the mammalian auditory end organ. This information has been relatively inaccessible due to technical limitations, but recent advancements in single cell RNA-Seq (scRNA-Seq) techniques have now made analysis on this scale possible.

Through a combination of high-throughput droplet-based and cell-per-well scRNA-Seq approaches, we have built an extensive high-resolution gene expression database of the cells of the organ of Corti and their surrounding epithelial cells from multiple embryonic and early postnatal timepoints. The complete array of hair cells, supporting cells, and surrounding non-sensory cells types are uniquely identified using known marker genes, and additional genes that show cell-type specific expression are revealed. Moreover, the large number of independent observations of each cell type’s transcriptional state over development allow a computational reconstruction of transitions representing cellular differentiation. We use these reconstructed models to identify potential transcriptional cascades that may direct cell fate decisions and changes in plasticity with the organ of Corti. By building branched relationship trees based on transcriptional similarities of cell types representing specific cell types at multiple stages of differentiation, we can also infer lineage relationships and anticipate restrictions. These lineage relationships are supported by preliminary genetic clonal lineage tracing studies, and candidate transcriptional regulators involved in cell fate decisions and differentiation can be evaluated by gain and loss of function studies.

These results further our understanding of the genetic pathways involved in organ of Corti development and patterning, provide a basis of comparison for potential aberrant states in mutant and disease models, and may help identify sets of transcriptional regulators that can direct the differentiation of unique cell types. This work also suggests a pattern of developmental lineage restrictions that provide important developmental context for cell type conversion strategies such as those proposed for auditory hair cell regeneration.

PS 349

Transcriptional Analysis of Heat-Shocked Mouse Utricle: Aligning Transcriptome to Drug Response

Matthew Ryals1; Lindsey May1; Katie Spielbauer1; Michael C. Kelly2; Joseph C. Burns3; Erich Boger1; Matthew W. Kelley4; Ronna Hertzano5; Robert Morell6; Lisa Cunningham1 1National Institute on Deafness and Other Communication Disorders, NIH; 2Laboratory of Cochlear Development, NIDCD, Bethesda, Maryland, USA; 3Decibel Therapeutics; 4Laboratory of Cochlear Development, NIDCD, NIH, Bethesda, Maryland, USA; 5Department of Otorhinolaryngology, Department of Anatomy and Neurobiology, and Institute for Genome Sciences, University of Maryland School of Medicine, Baltimore, MD USA; 6Genomics and Computational Biology Core, NIDCD, NIH, Bethesda, Maryland, USA

Many instances of drug-induced hearing loss are caused by two major classes of ototoxic agents: aminoglycosides and cisplatin. Our lab has shown that heat shock can protect cultured utricles against aminoglycoside-induced hair cell death, and that heat shock results in increased expression of heat shock proteins (HSPs). Here we have used RNA-Seq in order to characterize the full transcriptional response to heat shock beyond the canonical HSP family.

Gene expression was analyzed at the whole tissue and single cell levels within the utricle epithelium. A subset of the differentially-expressed transcripts identified by RNA-Seq was validated using RT-qPCR. Actively-translated mRNA species from hair cell -specific populations were isolated via affinity-tagged ribosomal (Ribotag) immunoprecipitation. Enrichment for hair cell-derived transcripts was validated from these samples using RT-qPCR. Using the Library of Integrated Cellular Signatures (LINCS), a list of compounds (perturbagens) that induce gene expression profiles that are similar to the utricle heat shock response was determined. Ability of selected perturbagens to elicit heat shock like transcriptional changes was evaluated by RT-qPCR, and perturbagen capacity for otoprotection against aminoglycoside exposure was determined.

A profile of differentially-expressed transcripts, or signature, in the heat-shocked utricle was determined from whole-tissue RNA-Seq data. This signature contained transcripts from multiple gene families including those regulating proteostasis, stress signaling, and protein folding. Single cell cell RNA-Seq transcriptome data clearly delinated hair cells versus supporting cells within the epithelium, and a different expression profile was observed for heat shocked cells versus control cells. Ribotag immunoprecipitation of hair cell transcripts elucidated the actively- translated mRNA species within the hair cells following heat shock. A search of LINCS perturbagens with similar signatures identified several compounds including HSP90 inhibitors and proteasome inhibitors. Expression profiling of utricles exposed to selected perturbagens indicate that these compounds can induce a ‘heat shock-like’ expression within the utricle. Preliminary in vitro testing of selected perturbagenes suggests that some may offer protection against aminoglycoside ototoxicity.

A medium-throughput screen of perturbagens for otoprotective capacity may reveal additional compounds capable of mediating a heat-shock conditioned protective effect against ototoxicity. Analysis of cell-specific gene differences identified in the single-cell and Ribotag experiments will further hone the matching of perturbagen gene expression profiles and reveal cell-type specific responses to stress within the sensory epithelium of the inner ear. This work was supported by the NIDCD Division of Intramural Research.

PS 747

Drop-Seq as a Lower-Cost, High-Throughput Method for Single-Cell Gene Expression Profiling of Cochlear Cells

Joseph C. Mays1; Joseph C. Burns2; Matthew W. Kelley3; Michael C. Kelly1 1Laboratory of Cochlear Development, NIDCD, Bethesda, Maryland, USA; 2Decibel Therapeutics; 3Laboratory of Cochlear Development, NIDCD, NIH, Bethesda, Maryland, USA

The mammalian cochlea is comprised of diverse cell types that vary significantly in structure and function. Single-cell mRNA sequencing (scRNA-seq) is able to provide transcriptomic profiles of individual cochlear cells to help to characterize the genetic heterogeneity between cell types. While this is a powerful method for studying complex tissues, commercial platforms for scRNA-seq can be prohibitively costly and low-throughput, preventing the generation of data sets from a sufficient number of cochlear cells of each type. Here, we utilize the Drop-Seq technique (Macosko et al., 2015) as a lower-cost, high-throughput method for performing single-cell RNA-seq on cochlear tissue. Drop- Seq allows for the generation of transcriptomes from individual dissociated cells by capturing mRNA from each cell with barcoded beads contained in nanoliter droplets. mRNA is reverse-transcribed and unique barcodes are used to map sequenced reads back to individual cells in order to efficiently generate expression data for each cell.

While the depth of coverage is lower than that of commercial platforms, Drop-Seq allows for a broad survey of cells to be processed at once. Utilizing this method, we have generated expression data from cochlear cells at late embryonic and early postnatal time points. We have compared these transcription profiles to existing scRNA-Seq data generated using the Fluidigm C1 platform and find that we can identify the same major cell types. The lower relative cost of droplet-based scRNA-Seq methods, such as Drop-Seq, make it a reasonable gene expression profiling technique to assess transcriptional changes at single-cell resolution in various experimental paradigms. To demonstrate this, we have compared Drop-Seq scRNA-Seq data from control cultured or HDAC-treated cochlear explants, which show cell type-specific morphological and molecular changes. Differential expression analysis shows transcriptional profile changes both at the whole-tissue level, and unique changes within individual cell types following HDAC inhibitor treatment, providing insight into the role that histone dynamics play in cochlear development and maintenance of cell states.

These results illustrate the potential use of droplet-based scRNA-Seq methods for surveying normal and perturbed gene expression profiles of cells within the mammalian cochlea. Though some trade-offs are made in sensitivity and ease of use, the lower cost and higher-throughput make Drop-Seq attractive for a variety of applications. In addition to using these methods to characterize transcriptional changes in mutant and drug-treated tissue, we also hope to utilize transcriptional profiles to help optimize culture conditions so that our in vitro models best represent in vivo conditions.

PS 749

Single-cell RNA-Seq Reveals Transcriptional Diversity in the Spiral Ganglion

Hanna Sherrill1; Michael C. Kelly1; Tessa Sanders1; Joseph C. Burns2; Robert Morell3; Matthew W. Kelley4 1Laboratory of Cochlear Development, NIDCD, Bethesda, Maryland, USA; 2Decibel Therapeutics; 3Genomics and Computational Biology Core, NIDCD, NIH, Bethesda, Maryland, USA; 4Laboratory of Cochlear Development, NIDCD, NIH, Bethesda, Maryland, USA

Spiral ganglion (SG) neurons convey auditory information from the cochlea to the brainstem. Despite the importance of this structure for normal hearing and cochlear implant function, our understanding of the development and diversity of neuronal phenotypes within the SG is still limited, in part because the total number of neurons within the ganglia is relatively limited. The development of methods to isolate and transcriptionally profile single cells using RNA sequencing has led to the identification and characterization of cellular heterogeneity in a variety of tissues including the sensory epithelia of the inner ear. Based on these studies, it seems likely that a similar approach could be used to address questions related to cellular diversity and development within the SG.

As a first step, fluorescent SG neurons from MapT-EGFP mice were dissected and isolated at E16 or P1. Individual neurons were then captured using two different platforms, the microfluidics-based Fluidigm C1 system and FACS-based single cell sorting. A minimum of 40 cells were profiled at each age. Initial results indicated wide transcriptional heterogeneity at E16 stemming from cellular differentiation and maturational processes, and more discrete transcriptional heterogeneity at P1. At P1, three transcriptionally distinct groups of neurons were identified by unbiased clustering. The most transcriptionally distinct group corresponds to ~15% of SGN cells collected, but these cells do not express Prph, indicating that they are not type II neurons. The three groups are thus far not easily defined transcriptionally by known type I and type II markers such as Prph or EphA4.

 

Do these papers give any hints as to what they plan on doing? Does this hint that they are aiming to cure hearing loss that is lost via loud noise exposure or hearing loss in general? Just at loss of what to make of these papers. Anyone know if they have put a timeline out on when they feel they will have treatment in the clinic?

 

The race for a cure is already launched and the successful winner will earn ££££$$$$€€€€, near 2035 in the worst case or near 2025 in the best case.

 

When I see what Decibel Therapeutics is doing, I think of Stanford university. They also use the same approach if I am not mistaken.
I suppose if processes are so complex because of all the interaction between genes, this is the way to do it.

 

@Samir
I have been reading this topic lately and I personally thank you for your dedicated effort here about the information collected!
Very useful and more easy to understand and follow....(at least for me!)

Absolutely true!

 

That was what I was referring to (-;
http://www.the-scientist.com/?articles.view/articleNo/43804/title/Inner-Ear-Cartography/
When I read the following in this article: "Within the mammalian cochlea, apical cells retain regenerative capacity for a few weeks after birth, but basal cells do not."
This was in 2015. By now researchers should have the information what the difference is between these cells?
Most inner ear hearing loss is in this area.
What makes it such a challenge?