University of Washington scientists announced this week that they have developed an in vivo drug toxicity-interaction screen using the lateral line of the larval zebrafish and used it to identify genetic modulators of antibiotic-induced mechanosensory hair cell death of the inner ear, and to identify protective compounds.
According to the researchers, their study, which uses a combination of chemical screening and traditional genetic approaches, represents a new way to identify drugs and drug targets that attenuate hearing and balance disorders.
Writing in the Feb. 29 online edition of Public Library of Science Genetics, the authors reported that they identified five mutations that modulate aminoglycoside susceptibility, and a new class of small molecules — benzothiophene carboxamides — that prevent mechanosensory hair cell death in zebrafish.
They also said that through further characterization of the protective mutant sentinel, or snl, they were able to identify a novel conserved vertebrate gene. The investigators validated the otoprotective effects of the benzothiophene carboximides using a mouse utricle assay.
David Raible, a professor of biological structure at the University of Washington and the corresponding author on the paper, told CBA News this week that working with zebrafish had its advantages over rodent and other in vivo mammalian models.
Raible noted that in humans, the hair cells are inside the inner ear, which is buried in the temporal bone of the skull, which is the hardest bone in the body. Zebrafish, however, have a second set of mechanosensory hair cells, known as the lateral line system, that is found on the surface of their body. Fish use the lateral line system to detect water flow and to interact with other fish during rapid swimming, for example.
These hairs “can be differentially stained by fluorescent vital dyes, which will specifically be taken up by the hair cells,” he said. “The hair cells can be recognized as being there and healthy by their level of fluorescence.”
If drugs or compounds are added to the fish’s environment, its hair cells are immediately exposed to them. “They do not have to go through the animal’s circulation, and we do not have to worry about rates of tissue accumulation, et cetera, which make understanding what concentrations the hair cells in the human inner ear are exposed to difficult to determine,” Raible said.
One class of potentially toxic compounds are aminoglycoside antibiotics such as neomycin, gentamicin, and streptomycin, which at certain doses become ototoxic in humans, according to Raible. In addition, “we found that very precise concentrations of these antibiotics cause the zebrafish lateral line hair cells to die.”
For the assay, the researchers stained the zebrafish with the fluorescent dyes, placed them in the wells of a 96-well plate, and then added small molecules from a 10,960-compound library.
“After we pretreated them with the small molecules for one hour, we added in the aminoglycosides, in this case neomycin, Raible said.
The researchers waited for an hour, and then looked to see if the hair cells had died. “When the hair cells are dead, the fluorescent dye is dispersed and does not give a bright signal,” Raible said.
He added that it is easy to screen through each well of the 96-well plate and identify hits that still have fluorescence because most of the wells will not have fluorescence.
“I thought it was a pretty impressive paper to read through,” Peter Eimon, CSO of zebrafish screening firm Zygogen, told CBA News this week. He said that what he particularly liked about the work was the fact that the authors were able to screen through a very large compound library for in vivo screening purposes. Zygogen did not participate in the current study.
“Screening through that many compounds using a live animal system really illustrates the power of doing screens with zebrafish embryos,” Eimon said.
Eimon said he was also impressed by the fact that the researchers were able to identify hits using their screen, and to some extent validate them using a mouse utricle assay.
Fishing for a Screen
This work could have commercial potential, Raible said. “We have a manuscript that has been submitted as a result of another collaboration that we did where we screened for compounds that might in themselves cause mechanosensory hair cell damage. [Ototoxicity] is a side effect that I believe drug companies would like to avoid.”
The investigators were able to demonstrate using a panel comprising FDA-approved compounds that some of the drugs were ototoxic in the zebrafish and in the mammalian system, said Raible.
He said that the study has been submitted to the Journal of the Association for Research in Otolaryngology.
“We think that in this system, we can do a prescreen, if you will, and look at many different compounds at many different concentrations before we consider testing these compounds in a mammalian system.”
There has been some discussion at UW about commercializing the current assay, Raible said, although he could not provide details. “There has been some work done by the office of technology transfer here. Things are in the works, but I am not sure where it stands.”
There is a big move to use zebrafish, particularly zebrafish in early embryonic stages, for toxicity screening, Eimon said. He added that toxicity screens using zebrafish are much cheaper than those using rodent models or other mammalian models, and they can be done in multiwell formats.
In addition, particularly in the European market, there is an increasing regulatory mandate to reduce, refine, or replace the use of animal models in all aspects of testing, particularly in toxicity testing, said Eimon.
“Zebrafish embryos, prior to the point where they actually start feeding on their own, do not fall within that protected class of lab animals,” Eimon said. “By developing assays that use zebrafish embryos at that stage, I think that there is certainly the sense that you are at least refining the use of animals in testing.”
The market share for an assay like this would be difficult to estimate, said Eimon, though he described as “fairly large” the market for specific classes of antibiotics and certain chemotherapeutics that have ototoxicity as a side effect.
More broadly, age-related hearing loss, which also results from mechanosensory hair cell death, could be a significant market if one considers aging Baby Boomers.
The work is a collaboration between Raible’s group and that of Edwin Rubel at UW, who studies mechanosensory hair cell death and regeneration, while Raible’s lab has been working on zebrafish genetics.
“About five or six years ago, Ed came up with this idea that we should combine our efforts to try to better understand mechanosensory hair cell loss,” Raible explained.
Raible said that the focus of the project has been to look at why mechanosensory hair cells die, and consider ways to prevent it. “Can we use a system that allows us to have access to the hair cells, and look at large numbers of hair cells, so we can get good quantitative data on what genes are involved in modifying the response of hair cells to toxic stimuli? And secondarily, can we use the same system to screen for possible therapeutic compounds that would overcome hair cell death?”
Work from a number of different labs has shown that at the cellular and functional levels, these lateral line cells are the same as those in the human inner ear, said Raible. For example, there is a group of genes that is necessary for the lateral line hair cells in zebrafish to function. These genes are the orthologs of those found in humans that, when mutated, cause a whole series of human congenital hearing and balance disorders, he said.
The authors believed that the lateral line mechanosensory hair cell cells on the zebrafish are going to be quite similar to those of the human inner ear because of that type of background information, but with the advantage that they can look at lots of them, Raible said.
“We think that in this system, we can do a prescreen, if you will, and look at many different compounds at many different concentrations before we consider testing these compounds in a mammalian system,” he said.
However, “one of the interesting characteristics of the mechanosensory hair cells in the lateral line of the zebrafish, compared to those in the human ear, is that they will regenerate. They are generally similar to the hair cells in the human ear, but do they have that one very striking difference,” Eimon said.
Although the UW scientists realized it is likely that they would find compounds that may help to protect zebrafish mechanosensory hair cells, but do nothing to protect mammalian hair cells, “that’s OK,” according to Raible.
“We think that because we can identify robustly things that do work, some subset of them will allow us to move forward,” he said.