“You’ll never be deaf,” Dr. Hoffman said to me years ago. At the time, I thought he meant I’d never lose all my hearing. But what I know now is that technology would take over when my ears no longer worked. Through a cochlear implant, I would continue to hear long after my ears ceased to function.
Research holds the promise that the kind of hearing loss I have may someday be reversible, returning the ear to close to its original pristine condition. Probably not soon and not for me, but most researchers think that within a decade they may have the tools that will eventually allow doctors to stop the progression of sensorineural hearing loss, including age-related hearing loss. Putting those tools into practice will take much longer. (Gene therapy, for people whose hearing loss has a genetic basis, will probably come sooner, possibly in the next decade.) The best guesses for hair cell regeneration—for the much larger group of people whose sensorineural loss is caused by noise or ototoxins or age—range anywhere from twenty to fifty years.
Until recently, scientists focused on the development of devices that would take the place of normal hearing: hearing aids and cochlear implants. The pharmaceutical industry, usually so quick to jump on the opportunity to medicalize a chronic age-related condition—dry eyes and wrinkles, trouble sleeping, lagging sexual function, bladder control, memory loss—has not paid much attention to age-related hearing loss, in terms either of prevention or cure. There are no FDA-approved drugs for the treatment of hearing loss. Demographics alone would suggest they are missing a big opportunity.
In October 2011, the Hearing Health Foundation (formerly the Deafness Research Foundation) held a symposium in New York to kick off its new campaign, called the Hearing Restoration Project, an ambitious program that had enlisted, at that point, fourteen researchers from ten major hearing & loss research centers in the United States. This consortium will share findings, with the goal of developing a biological cure for hearing loss in the next ten years. With a fund-raising target of $50 million, or $5 million a year, the Hearing Restoration Project will tackle the problem of hearing loss with the aim of curing it, not treating it.
The funding is relatively small right now, but there is hope that the foundation will be able to raise more in future years. Individual consortium members may currently receive somewhere between 5 to 20 percent of a laboratory’s annual bud get from the Hearing Health Foundation. But the collaborative nature of the venture is unusual. (A similar consortium exists for the study of myelin diseases—a factor in multiple sclerosis as well as hereditary neurodegenerative diseases.) Under its previous name, the Deafness Research Foundation, funding was limited to early career support to researchers. They’ve now added the Hearing Restoration Project.
The symposium, titled “The Promise of Cell Regeneration,” brought together leading researchers in the field of hearing loss. Dr. George A. Gates, an M.D. and the scientific director of the Hearing Restoration Project, chaired the program. The speakers included Sujana Chandrasekhar, an M.D. and director of New York Otology, who talked from a clinical perspective about the current state of hearing loss research. Ed Rubel, from the University of Washington, discussed the history of hair cell regeneration research and his current work on regenerating hair cells through pharmaceutical applications. Stefan Heller discussed his lab’s announcement in May of 2010 of the first successful attempt at generating mammalian hair cells (of mice) in a laboratory setting from stem cell transplants. Andy Groves, from Baylor, discussed the many still-existing hurdles to hair cell regeneration in humans. Unable to attend was Douglas Cotanche, currently working at Harvard on noise-induced hearing loss in military personnel.
Surrounding the inner and outer hair cells are the so-called supporting cells, which come in all varieties: Deiters’ cells, Claudius’ cells, Hensen’s cells, inner pillar, and outer pillar cells. Supporting cells are the magical cells that instigate regeneration in damaged inner ears of chicks and fish. And they are where someday regeneration may occur in humans.
That limited number of hair cells, as well as their fragility and inaccessibility has hampered research. In his 2010 Cell paper, Stefan Heller noted, “The inner ear shelters the last of our senses for which the molecular basis is unknown.” So little is known about the structure of the inner ear that, as Dr. Gates said, “we have a hard time clinically knowing how much [loss] is outer and how much is inner. That’s why we use the term sensorineural.”
• • •
Ed Rubel’s photo on the University of Washington website shows a balding middle-aged man, elbows on the table, with two yellow chicks. That whimsical photo belies a seriously impressive academic c.v.: Virginia Merrill Bloedel Professor of Hearing Science; Professor of Otolaryngology—Head and Neck Surgery; Professor of Physiology and Biophysics; Adjunct Professor of Psychology. Dr. Gates referred to him as the godfather of hair cell regeneration.
Rubel and his colleagues at the Virginia Bloedel Hearing Research Center see four clinical scenarios that lend themselves to pharmaceutical fixes. The first would reverse sudden sensorineural hearing loss. The second would prevent ototoxic and/or noise-related hearing loss. The third would retard the progression of hearing loss, especially age-related hearing loss. The fourth would restore hearing after it had been lost.
Until 1985 it was thought that no animals could regenerate hair cells once they were destroyed. Rubel, then at the University of Virginia, discovered, quite inadvertently, that some do. The purpose of his research was to determine how long it took for ototoxic drugs to damage hair cells. He and his lab partners chose chicks as their animal model. Chicks have an easily accessible inner ear, and their ears in many ways resemble the human inner ear.
Rubel gave the chicks hair-cell-destroying aminoglycocides—a class of antibiotic known to be ototoxic—and then assigned the new guy at the lab, as Rubel put it in his talk at the conference, a resident named Raul Cruz, to sacrifice the chicks after a certain number of days and study the degree of deterioration in the hair cells. After eight days, Cruz found the chicks had, as expected, lost many cells. But when he studied the slides taken from chicks sacrificed at twenty-two days, instead of more dead cells they showed fewer. Where there had been dead cells, there were now healthy ones. “Raul, you must have mixed up your animals, go back and do it again,” Rubel recounted, adding, to laughs, “Because he was just a resident, he didn’t know what he was doing.”
Around the same time, Doug Cotanche, then a post doc at the University of Pennsylvania, saw the same results in chicks after damage due to intense noise exposure. Rubel and Cotanche published separate papers in different scientific journals, but continued communicating and soon got together to publish dual papers in the prestigious journal Science, showing, as Rubel said, that these were indeed brand-new hair cells “due to new cell division and the creation of new cells in the inner ear.” This was a stunning scientific development. “And wow, we had a new field.”
The next step was to figure out how chickens did it. Studying the cochlea of chicks and other birds, Rubel and others found eventually—over eighteen years!—that bird hair cells do indeed regenerate. Over the same period they discovered many other important molecular and functional details related to this remarkable ability. He showed some slides. The first slide showed the condition of the hair cells shortly after the animals were exposed to noise: “It looks just terrible. All these hair cells are blebbing out and being discarded.” Five days later, however, they could see “baby cells” budding, some of them with the distinctive hair, or microvillus, on top. Then, after a few days, a high-power scanning electron microscope showed that all the hair cells were back. Not perfect, a few small abnormalities, but perfectly functional.
Interestingly, Rubel went on, they found that this happened no matter the age of the bird. Brenda Ryals, a former student of Rubel’s, had a colony of senile quail that, they found, “regenerate cells just as well as a baby chicken does,” Rubel said. New cells are created not only in the cochlea but also in the vestibular epithelium, important for balance. And, perhaps most important, the new cells are appropriately connected to the brain. “The new cells restore near-normal hearing and perfectly normal vestibular reflexes. They restore perception and production of complex vocalizations. Birds lose their song when they lose their hearing, but they gain their song back when they restore their hearing.”
In 2001 Rubel joined forces with David Raible, also at UW, who was using zebrafish, a popular aquarium species, to study development of the nervous system. Eleven years later the two labs are still collaborating on understanding how to prevent and cure hearing loss, and on hair cell regeneration.
Zebrafish proved to be an even better animal model for studying some aspects of hearing-loss prevention and regeneration than birds. In addition to hair cells in the inner ear, aquatic vertebrates like fish have hair cells on the outside of the body, in something called the lateral line. The lateral line is used for detecting change in water currents and its cells are physiologically very similar to human inner ear cells. At electron microscope level, intracellular structure is similar. It turns out that fish and reptiles, like birds, regenerate hair cells, as do frogs and other animals. “So why can’t we?” Rubel asked.
The Rubel/Raible team subjected the zebrafish larvae to ototoxic screening, again using aminoglycoside antibiotics. They tested drugs and druglike compounds to find ones that inhibit hair cell death in the fish. This work may lead to the development of protective cocktails to preserve hair cells before exposure to antibiotics or ototoxic chemotherapy drugs. They may also be given to humans after ototoxic assaults, which include noise exposure.
In March of 2012, I met with Rubel and a group of younger fellow researchers at the Virginia Merrill Bloedel Center. Rubel is a charismatic leader, but he insisted on referring to these researchers not as part of his lab but as independent scientists, with their own NIH grants, some with their own Hearing Restoration projects. David Raible was out of town. Raible, Jennifer Stone, and Elizabeth Oesterle collaborate on different projects with each other and with Rubel. But, Rubel said, “Having sort of started the hair-cell regeneration field, I feel very comfortable getting out of it and doing other things.”
Jennifer Stone is a cell biologist and neuroanatomist, who works primarily on avian hair cell regeneration. About five or six years ago she started working with mice, with several of the other participants, including Rubel and Elizabeth Oesterle, a cell biologist, and Clifford Hume, an M.D./Ph.D. clinician scientist. Stone led a recent study which found that after virtually all the vestibular hair cells in adult mice are killed, 16 percent of the hair cell population comes back spontaneously.
“It’s a new discovery,” Stone said. “It’s not entirely surprising, but I think we’ve demonstrated it pretty definitively.” Because spontaneous regeneration happens in only certain regions of the vestibular system, it helps the researchers narrow the field. By comparing the tissue in this region to tissue in others, they may understand the factor that allows regeneration in one place but not in another. Once we understand what allows the tissue to make new hair cells in these regions, Stone said, we can determine what would be needed to “release the brake,” as she put it.
The p27 gene, which regulates cell division and helps prevent cancer, is one such molecule. To allow these hair cells to divide, the p27 gene would need to be turned off. Or maybe, she added, “it could be that we need to push the pedal on the gas: add something to promote division. It could be that we have to both put on the brakes and push on the pedal to start this process in mammals.”
Julian Simon is a chemist, a Ph.D. pharmacologist, who got interested in the toxicity of cancer therapeutic drugs when clinicians at the Seattle Cancer Care Alliance, the patient arm of the Fred Hutchinson Cancer Research Center and the University of Washington, complained about the ototoxicity of certain chemotherapy drugs, Cisplatin prominently among them. Simon said that 30 to 40 percent of patients who go on Cisplatin regimens for lung cancer sustain significant and permanent hearing loss. (Rubel told me that some reports suggest an even higher percentage, up to 80 percent.) Simon’s approach is to use small molecules to “perturb” biological systems. “We know what we’d like the cells to do, and in this case we’d like to take cells that would otherwise die and keep them living.” Because the whole process of sensory hair cell death is—“with all due respect to present company” (meaning his fellow researchers)—“poorly understood, by learning how we can protect these cells from dying, maybe we can also learn something about the way cells die. Why they die.”
The team’s approach is multidisciplinary, involving not only research scientists and clinicians but also psychologists, genetic counselors, audiologists, and special education specialists. In adult hearing loss, they are also looking at the role of prescription medications in age-related hearing loss. Many are life-saving medications, but sometimes less toxic substitutes may be available.
The UW group moved on to a lively discussion about how they would advise the parents of a young child getting implants. Should the child get implants in both ears? Cochlear implants cause the destruction of the so-called support cells that might give rise to new hair cells. Hence, should the parents “save” one ear in the hope that cell regeneration technology will eventually enable the child to hear normally out of that ear? Henry Ou said that parents often ask him about a second implant. “Sometimes they ask, ‘Do you think there’s hope that this is going to be fixed?’ I say, ‘Yeah.’ But at the same time, if I don’t think there’s hope, I shouldn’t be doing research on it. I’m a conflicted person to ask.”
Simon added: “Parents don’t want to find out when their kid is eighteen that there is something better.” He cited the substantial evidence that children do better in school when they’re implanted earlier, and bilaterally. Rubel agreed with the basic premise that early intervention is enormously important and that cochlear implants in children have become an essential therapeutic option, but expressed skepticism about the value of always doing bilateral cochlear implant surgery. Referring to one study in particular, he said, “The little known fact about this work is that it includes only the top 20 percent of single implant users.” Another study found different results. “So I think it’s still up in the air,” Rubel said.
We just don’t have enough information yet to know the impact that implants make at that critical learning period for language and speech comprehension. But, as Jenny Stone said, the same question could be asked about regenerated hair cells. “The big elephant in the room, I think,” she said, “is that we don’t know whether regenerated hair cells will result in better hearing—appreciation of music, noise, speech—than a cochlear implant can. And I think it’s a huge jump to assume that in twenty years we’ll be there.”
“Well but in fifty years?” Rubel interjected.
“Maybe in fifty years,” Stone replied.
“I keep going back to the bird,” Rubel said, “and we absolutely know that the bird gets great hearing back. They can recognize their own songs, they can learn new songs, not only speech but song recognition!”
“He loves birds,” Jenny Stone said. “I’m not trying to be pessimistic. But it’s going to take a lot of time to really get concrete evidence for what the best type of repair is going to look like.”
Why would this evolutionary adaptation have occurred? Groves speculates that mammals made a trade-off: in the course of developing high-frequency hearing, their hair cells became more specialized, and in the process they lost the ability to regenerate. Although we humans have devised many ways of inflicting hearing loss on ourselves (such as rock concerts, iPods, and heavy machinery), one of the few naturally occurring things that kills hair cells is the wear and tear of old age. (Unless it turns out that even that is the result of accumulated noise exposure.)
“From an evolutionary point of view,” Groves said, “and this sounds rather brutal, but evolution doesn’t care about old age, as long as you live long enough to have kids.” Once your reproductive years are over, your body has done its evolutionary job. As a result, mammals would not suffer a selective disadvantage by losing the capacity to regenerate their hair cells.
Bruce Tempel, at the University of Washington, echoed that Darwinian opinion. For the past twenty to twenty-five years he has been looking at the genes implicated at one or another level in hearing loss. “Truth be told,” he said in an interview, “the reason that I got really interested in the auditory system is because you don’t need it. From a geneticist’s point of view, basically, this is great. This system can be completely nonfunctional and the animal still survives.” He added that stress and hormonal influences on hearing loss are part of the reason the auditory system is so useful to geneticists: “You’re able to identify the genes, the proteins, and from studying the protein itself find out whether there’s a hormone or an influence on the expression of that protein. You can find out if there are interacting proteins that become a cascade linking the different individual proteins and the genes. And what’s really cool about the auditory system is that we can do all that and still have a viable animal.”
Andy Groves also studies the genetics of hearing loss. One of the mammalian genes whose function is to stop cells dividing (necessary, to regulate the size of organs and protect against cancer) is the p27 gene, which Jenny Stone talked about at the UW group meeting. Figuring out how to switch off that gene is one of the biggest obstacles researchers face.
After a great deal of work in cell-culture dishes and looking through microscopes, Groves and his colleague Neil Segil discovered that when they isolated mouse supporting cells from the newborn cochlea, the action triggered the p27 gene to switch off and the supporting cells to start dividing. They don’t know why. Unlike humans, mice cannot hear when they are born. By the time they begin to hear—at about two weeks after birth—mouse supporting cells stubbornly refuse to divide even when isolated from the cochlea. Groves, Segil, and their colleagues are now trying to understand what happens to the aging supporting cells that makes them unable to divide.
The role of the Notch pathway in regulating the activity of the p27 gene is controversial. Groves mentioned work that Amy Kiernan, currently on the faculty at the University of Rochester, carried out when she was a postdoc with Tom Gridley at the Jackson Laboratory in Bar Harbor, Maine. She managed to inactivate the Notch signaling pathway in mice genetically. Her mice produced extra hair cells and showed some precocious cell division in the cochlea. Another researcher working with Groves and Segil, Angelika Doetzlhofer, did the same, using a pharmacological approach with drugs that blocked Notch signaling. When they blocked the signaling in newborn mice, they saw a 50 percent increase in hair cells and fewer supporting cells. These findings are preliminary, Groves cautioned, and the role of the Notch pathway is still being studied.
Following up on this, Groves and colleagues repeated their Notch blocking experiments in older mice. By the time the mice were three days old, the increase in hair cells had dropped to 30 percent. In six-day-old mice, new hair cells were no longer produced. Although extrapolating this timetable to humans is tricky, the current data suggests that the human cochlea may no longer respond to Notch inhibitors by the time the fetus is five to six months old.
“So here is the take-home message,” Groves concluded. “Our challenge—if you want to set a ten-year challenge—is to understand these roadblocks and then devise methods to get rid of them, and ultimately to apply these methods in a clinical setting.” A clinical setting populated by humans. As Groves said at the beginning of his talk, “We’re not here to treat hearing loss in birds.”
Stefan Heller and his colleagues are taking a different approach to regenerating hair cells. They are attempting to get stem cells—undifferentiated cells that can develop into various specialized cells—to turn into hair cells, by mimicking the naturally occurring developmental processes that lead to formation of the inner ear. They do this in a culture dish and in a laboratory setting, which allows them to learn a lot about the process, such as what it actually takes to make sensory hair cells from scratch.
In March 2012, I visited Heller’s lab at Stanford in Palo Alto. We literally ran into each other as I was looking for his office. Heller is formidably smart but completely unimposing in manner. He was wearing a well-worn T-shirt with a coffee cup on it (half full? half empty? “Definitely half full,” he said), jeans, and sneakers. We talked in his office with a huge humming fish tank taking up about a sixth of the office. I asked if he had zebrafish. He said he didn’t, but Dr. Robert Jackler, the chair of the Stanford Otolaryngology Department and the force behind the accumulation of brain power that makes Stanford’s one of the most important hearing research departments in the world, told me that Heller raises anemones to get novel fluorochromes for his research.
The second type are the induced pluripotent stem cells (iPSCs) referred to in the title of the 2010 article. These are, according to the NIH website, “adult cells that have been genetically reprogrammed to an embryonic cell-like state.” The NIH definition goes on: “It is not known if iPSCs and embryonic stem cells differ in clinically significant ways.” That Heller and his lab were able to produce sensory hair cells in mice using both these kinds of stem cells is significant. Further, that they were “mechanosensitive” means that they were responsive to mechanical stimulation, and that these responses were similar to those in immature hair cells.
The third type are somatic stem cells, cells isolated from a specific organ—like the human ear. As attractive as these cells are to religious conservatives who oppose embryonic stem cell use, up until now they have not seemed to be a viable option because, as Heller said, “these cells are very rare.”
Embryonic stem cells and pluripotent stem cells share an unfortunate feature: they can generate tumors. Heller said that he’s received many e-mails from patients offering to be subjects for human trials. He showed the audience at the hearing regeneration conference a slide of a mouse that had been injected with a small number of these cells: “After one month, this mouse grows an enormous tumor.” Before they can be used to regenerate hair cells, these stem cells will have to be rendered non-tumorigenic.
Somatic stem cells don’t cause tumors, but there aren’t enough of them. Scientists have not been able to isolate enough of these cells from the ear to study their advantages and disadvantages over the more abundant but problematic embryonic and pluripotent cells. Induced pluripotent stem cells appear to be the perfect compromise. These cells can be generated from virtually any cell of someone’s body, and Heller’s lab has been working with somatic cells derived from skin biopsies, usually from a patient’s arm, a human patient with hearing loss.
“The work is very exciting,” he told me. “Treating the cells from the biopsy with reprogramming factors, they can turn a somatic cell into an induced pluripotent stem cell (iPS cell). They can then grow them in a culture much the way they do embryonic cells, but without the religious or ethical controversy.
“We are basically making hair cells from human skin cells,” he said. “These cells are not from the ear, so making the claim that these are hair cells is a difficult one. But they do have all the features of hair cells. They look like hair cells, they express genes that one would expect to find in hair cells, and they are functional, and moreover, we are approaching the point where we can generate human hair cells.” Many steps remain before this becomes anything like a clinical reality, however, and each step takes a long time and a lot of money.
It’s a long way from mouse to man, but Heller said at the Hearing Restoration symposium that despite the challenge, “we’re getting close.”
One of the major findings of the last five or ten years, Heller said, is coming to understand the roadblocks. Once they know what obstacles stand in the way of transplantation, they can begin to figure out how to get around them. The first roadblock is the fact that these cells cause tumors. Looking ahead, Heller said, scientists need another five to ten years to solve that problem—a knotty one that involves learning how to generate pure cells and cells that are not tumorigenic. Once they solve that one, they will encounter new roadblocks: how to deliver the stem cells into the ear, determining the appropriate site for integration of cells, how to ensure their long-term survival, how to block immune system responses, how to make sure the cells function—“and, of course, whether the cells improve hearing.”
As a young assistant professor, Heller told the Hearing Restoration symposium, if he had been asked how long it would take to cure hearing loss, he would have said, “You know, in five years we’ll have a solution for certain things.” Over time, he went on, “I got a sense of the difficulty of the problem and of all the roadblocks and the issues we have to deal with. And I’m getting frustrated myself, how long it takes to overcome a single one of these roadblocks. And then you’re over one hill and there’s another one.” The difference now, he said, is that “we know where we have to go, and what we have to do. It’s difficult to assess whether it will take ten years or twenty years or even fifty years.”
Later, he went back to the time line again. “I think for transplantation we need another five to ten years before we are at the point where we can generate pure cells and cells that are not tumorigenic, to start doing experiments with animals.”
Ed Rubel, in our interview, also gave a time line for his lab’s work: “I think with proper funding, we can, in ten years, develop ways to get sufficient numbers of hair cells in a laboratory mammal cochlea as a model. We [meaning researchers in the field] will then go on to optimize the drug or drugs in all the ways needed to use them safely in humans, and only then go to clinical trials.” He pointed out that they already know some genes and some compounds that facilitate the production of new hair cells in some conditions, but they don’t have the lead compound. Even once they find that lead compound, he added, “all the safety trials, in vitro trials and small animal trials, all that preclinical work, usually takes eight to ten years.”
As for gene therapy, for those whose hearing loss has a genetic basis, Stefan Heller cites what happened with research on vision and blindness: “Twenty years ago this was an open field, and now it has evolved into a flourishing clinical field and a very lively biotechnology field” with a market for drugs and procedures. “I think we can use the vectors and tools they’ve developed and bring them into our field. So I think five to seven years is probably a reasonable time frame for seeing results in animal studies. Gene therapy would be used on people whose deafness is caused by a mutation in a certain gene. If you could deliver the correct gene into the inner ear, you might be able to repair hearing loss before it progresses too far for repair.
As for the development of prophylactic drugs, the use of “high throughput methods” will help make the time line a little shorter. High throughput methods—also called high content screens—use multiple cell culture dishes testing hundreds or thousands of compounds. Robots may also be used to speed the testing process. This requires work “with big pharma—because we cannot do this in our lab,” Heller said. High throughput screening and the backing of big pharma would increase efficiency, allow the earlier use of screens directly with human cells so they don’t have to go through mice first and then on to humans.
