Genes of Silence
Scientists track down a slew of mutated genes that cause deafness

By JOHN TRAVIS

On the north shore of the island of Bali sits Bengkala, a 700-year-old village
that has earned a measure of fame for its remarkably large number of deaf
people -- currently 48 of the nearly 2,200 inhabitants. These deaf residents have
had a profound influence on Bengkala's culture.

This is a very unusual community. The village has created a unique sign
language, and most people in the village have some facility with it," says John T.
Hinnant of Michigan State University in East Lansing, an anthropologist who has
recently spent time documenting the village's culture on videotape.

Signed names have replaced spoken names. Villagers refer to Hinnant by
making the sign for a videocamera or, he ruefully admits, the sign for a big nose.

Bengkala isn't fertile scientific ground just for anthropologists. Hinnant learned of
the village from geneticists who are working to isolate the mutated gene
responsible for the widespread hearing loss. So far, they've localized the gene
to a small region of chromosome 17.

"We're quite close to having it," says Thomas B. Friedman of the National
Institute on Deafness and Other Communication Disorders in Bethesda, Md.

Once found, the Bengkala gene will join the numerous other genes that
investigators have recently tied to hearing loss. Last year was "quite dizzy with
deafness genes," remarks Karen P. Steel of the Medical Research Council's
Institute of Hearing Research in Nottingham, England.

These new genes are providing some of the first molecular insights into the
workings of the auditory system and into the reasons why about 1 child in 1,000
is born deaf. Mutations in some of the genes produce hearing loss alone. In
other cases, the gene mutations cause a range of other problems as well, such
as blindness.

While the identification of these deafness genes may not help people born with
profound, almost always permanent hearing loss, it may benefit the many whose
deafness strikes later in life. "If you know at a biological level what's going wrong
in the ear, you at least have a chance of intervening, preventing the hearing loss
from getting any worse, or maybe even reversing it," says Steel.
 

 Discerning the origins of deafness, genetic or otherwise, is no easy task.
Indeed, physicians often cannot diagnose the flaw that eliminates a person's
perception of sound.

A person's ability to hear depends on the spiral-shaped cochlea, a pea-size
organ in the inner ear. In each cochlea, a mere 16,000 hair cells, each bearing a
cap of bristles called stereocilia, detect noises. Sound-induced movement of the
stereocilia stimulates the hair cells to generate electric impulses that convey
auditory information to the brain. Physicians suspect that most cases of
deafness stem from problems with the hair cells -- but that's tough to confirm,
since the cochlea is hidden within a bony labyrinth.

Resolving the genetics of deafness is no less challenging. Genetic flaws
account for about half the incidence of congenital deafness, but researchers
estimate that mutations in more than a hundred genes may trigger such hearing
loss. This genetic diversity shows up in the striking fact that the majority of
children born to two deaf parents can hear.

Scientists have been able to find many genes responsible for the syndromic
forms of deafness, in which hearing loss is accompanied by other symptoms. In
1992, for example, investigators found the mutated gene behind Waardenburg's
syndrome, a condition characterized by deafness, widely spaced eyes that are
sometimes mismatched in color, and a white forelock (SN: 5/2/92, p. 296). Last
year, several research groups examining people with deafness, frequent loss of
consciousness, and a dangerous heart arrhythmicity attributed the syndrome to
mutations in various genes that encode components of a protein complex that
allows potassium ions to enter cells.

In contrast, investigators have struggled to find genes behind nonsyndromic
deafness, which is the far more common type, accounting for the majority of
congenital hearing loss.

The hunt has been complicated by a phenomenon to which scientists have given
the unromantic name of assortative mating. "Deaf individuals tend to intermarry,
at least in developed countries," explains Christine Petit of the Pasteur Institute
in Paris.

Since the two members of a deaf couple generally owe their hearing loss to
different genes, assortative mating creates a nightmare for geneticists trying to
track a single deafness gene through a family's history, a useful step in finding
its chromosomal location.

For a long time, investigators considered mapping of nonsyndromic deafness
genes an impossible task, notes Richard Smith of the University of Iowa in Iowa
City, who maintains a World Wide Web site that tracks the field's progress (the
hereditary hearing loss home page at http://dnalab-www.uia.ac.be/dnalab/hhh).
 

The first major break in the hunt for nonsyndromic deafness genes came when
Pedro E. León of the University of Costa Rica in San José learned of a large,
local family with many deaf members. The affected members of the family can
hear at birth, but around age 10 they start to go deaf. By age 30, all hearing is
lost.

León traced this trait back through eight generations of the family and obtained
blood samples from nearly 150 living members. He then teamed up with a
research group headed by Mary-Claire King of the University of Washington in
Seattle. In 1992, after tracking genetic markers known to be inherited along with
the deafness, the investigators announced that they had mapped the gene
responsible to a part of chromosome 5.

With that first success in hand, says Smith, many other scientists began to
search around the world for large, inbred populations in which hearing loss is
common. Meanwhile, León and his colleagues continued their quest until, in the
Nov. 14, 1997 Science, they identified the mutated gene behind the Costa
Rican family's deafness.

By examining a library of genes active in the cochlea of a developing fetus,
León's team has shown that the normal version of this gene is turned on in the
inner ear. Yet the brain, heart, lungs, kidney, and many other tissues also seem
to use the gene. That finding poses a difficult question about those with
mutations in the gene.

"Why are these people deaf and otherwise normal, when this gene's product is
used by a vast array of cell types?" asks the University of Washington's Eric D.
Lynch, who led the search for the gene.

The deafness gene resembles a fruit fly gene called diaphanous, which has
offered some clues to the human gene's possible role in hearing. Studies of
diaphanous and a similar mouse gene indicate that their proteins help a
molecule called profilin assemble filaments of another protein, actin.

Such filaments provide a dynamic skeleton for cells. They may be especially
important to the cochlea's hair cells because the stereocilia owe their stiffness
to bundles of actin filaments that are continuously broken apart and rebuilt.

"The hair cells are possibly so reliant on the actin cytoskeleton to perform their
hearing function that we see a [consequence of the gene's mutation] there and
not elsewhere," says Lynch.

To study further the cause of the Costa Rican family's deafness, the researchers
plan to genetically engineer mice to suffer a similar form of hearing loss. "We're
trying to mimic the mutation as closely as possible," says Lynch.

The scientists are also investigating whether some cases of hearing loss stem
from mutations in a second human gene that they have found resembles
diaphanous. "We think it makes a good candidate, considering its sibling's role
in deafness," says Lynch.
 
 
 

The Costa Rican gene wasn't actually the first nonsyndromic deafness gene
isolated. That honor went several months earlier to the gene encoding a protein
called connexin 26. Ironically, this gene came to light during the search for the
cause of an apparently syndromic form of hearing loss.

Investigators were studying a family that had both deafness and a skin disorder.
The researchers focused on a group of channel-forming proteins, or connexins,
that play an important role in the skin. Connexins allow small molecules to pass
between cells.

As suspected, some deaf family members had mutations in the gene for a
connexin. The scientists realized, however, that the family's symptoms had been
mistakenly linked. "Not everyone who had skin disorders was deaf and vice
versa," notes Robert F. Mueller of St. James' University Hospital in Leeds,
England.

In the May 1, 1997 Nature, Mueller and his colleagues reported that mutations in
the gene for connexin 26 produce deafness, but not the family's dermatological
problems.

The investigators also used antibodies that bind to connexin 26 to show that the
protein is present in various regions of the inner ear, but they're still unsure what
role it plays in hearing. Some scientists speculate that flawed connexins may
disrupt the ability of hair cells to take in or expel potassium ions. Both activities
are crucial to the cell's generation of electric impulses.

The gene for connexin 26 illustrates one emerging theme of deafness genetics.
In several cases, different mutations in a single gene can cause either recessive
or dominant forms of inherited hearing loss. In the recessive forms, a deaf child
has inherited a mutant copy of the gene from both parents, and deafness is
usually present at birth. If the gene has a dominant mutation, only one flawed
gene is needed to produce deafness, and the hearing loss often occurs
gradually and later in life.
 
 
 

Another lesson hearing researchers have learned is that syndromic and
nonsyndromic deafness can represent two sides of the same coin. They have
found that certain mutations in a gene produce both hearing loss and other
symptoms, while other mutations seem to cause just deafness.

"For clinical geneticists, who have spent their lives asking, 'Is this syndromic or
nonsyndromic?' it's mind-boggling to find out it's just an artifact of where the
mutation is," says Mueller.

Take Pendred's syndrome, perhaps the most common syndromic cause of
hereditary hearing loss. Usually born deaf, though the hearing loss sometimes
occurs during childhood, people with Pendred's syndrome also develop goiter,
an enlarged thyroid gland, around puberty.

In 1996, two research groups that had studied large, inbred families exhibiting
Pendred's syndrome reported that they had mapped the responsible gene to
chromosome 7. One group, led by Benjamin Glaser of Hadassah University
Hospital in Jerusalem and Val C. Sheffield of the Howard Hughes Medical
Institute at the University of Iowa in Iowa City, then contacted Eric D. Green of
the National Human Genome Research Institute in Bethesda, Md.

"My lab's passion is chromosome 7," says Green, who has developed a
detailed map of genetic markers on the chromosome.

After narrowing the chromosomal region under suspicion, Green and his
colleagues began sifting through genes there, testing whether deaf family
members had mutations in any of them. As reported in the December 1997
Nature Genetics, the investigators finally struck gold with a gene that appears to
encode a protein, which they call pendrin, that ferries sulfate molecules across
cellular membranes.

Cells attach sulfates to proteins and many other molecules for myriad reasons,
and scientists have recently discovered that mutations in other sulfate
transporter genes can cause dwarfism or constant diarrhea, notes Green.

As expected, the gene encoding pendrin turned out to be active in the adult
thyroid gland. Preliminary experiments suggest that it is also active in the
developing fetal cochlea, which may explain the deafness associated with
Pendred's syndrome.

"We think there's a direct role this protein plays in the development of the inner
ear," says Green. "The cochlea is like a spiral staircase, and [in people with
Pendred's] it just doesn't have the right number of turns in it."

Blurring the definition of Pendred's syndrome, however, the investigators have
also found that certain mutations in the pendrin gene can cause deafness
without thyroid problems. "Not all individuals develop goiters," says Green.

This finding, he adds, may mean that mutations in the pendrin gene account for
many more cases of deafness than researchers had previously suspected.
 
 
 

A similar conclusion now seems likely for a gene tied a few years ago to
Usher's syndrome. In this condition, people born deaf slowly lose their sight. The
gene encodes a protein called myosin VIIa, and in several reports this year,
investigators have described mutations in the gene among deaf people with no
obvious visual problems.

The gene first came to light through studies of mice. Over the years, scientists
have identified many deaf mouse strains. As names like shaker and waltzer
indicate, these strains were often first identified by their head-tossing, odd
circling behavior, or other abnormal movements that seem to reflect balance
difficulties stemming from inner ear problems.

Several years ago, Steel and her colleagues discovered that a mouse strain
called shaker-1 owes its hearing loss to mutations in the gene for myosin VIIa.
This protein and many similar ones can bind to actin filaments and, through the
effort of a built-in motor, move along them, often ferrying some form of molecular
cargo.

Steel and her colleagues then realized that the human version of the gene for
myosin VIIa resides in a part of chromosome 11 already implicated in one form
of Usher's syndrome. Further research by Petit and other investigators
established that mutations in this gene indeed cause the syndrome's visual and
auditory problems.

Several cell types in the eye produce myosin VIIa, and according to studies
conducted by Tama Hasson of Yale University School of Medicine, the protein is
abundant in stereocilia and other specific regions of hair cells. "We know where
the protein is, so now the question is, what does the protein do," says Hasson.

While the search for the protein's natural roles continues, Steel and her
colleagues report in the November 1997 Neuron that myosin VIIa appears to
play a role in the accumulation of a certain antibiotic inside hair cells.
Consequently, the presence of myosin VIIa in the inner ear may explain why
antibiotic-induced deafness is one of the major nongenetic causes of hearing
loss.
 
 
 
 
 

As investigators continue to catalog new deafness genes, they need to firm up
their estimates of how many cases of hearing loss can be attributed to each
gene. At the moment, most of the excitement centers around connexin 26. The
most dramatic evidence concerning its gene comes from a study by Petit and
her colleagues in the November 1997 Human Molecular Genetics. Among 65
families with histories of hearing loss, most of them from Tunisia, France, New
Zealand, and the United Kingdom, the researchers found that about half had
mutations in the gene for connexin 26.

"It looks like connexin 26 accounts for a lot [of deafness] in certain areas of the
world, but it's not clear how much it accounts for hereditary deafness in the
United States," says Friedman.

As the genetic data accumulate, physicians may begin to screen for mutations
that might cause late-onset hearing loss or counsel people about potential
outcomes of future pregnancies.

The latter issue is particularly important, since only half of congenital hearing
loss stems from inherited mutations. "The main question of a family with a deaf
child is whether they will have another one," notes Petit.

Outside the clinic, scientists will continue to examine how the proteins encoded
by these deafness genes make hearing possible. They will investigate primarily
the mouse inner ear, especially in mice bred to have mutant copies of the
various deafness genes.

Mice will also serve as the initial testing ground for any treatments that might
emerge from the discovery of deafness genes. Last year, for example, Anil K.
Lalwani of the University of California, San Francisco reported at a meeting that
he and his colleagues had for the first time successfully introduced a foreign
gene into the mouse cochlea. With the growing number of identified deafness
genes, physicians should soon know which ones will be useful in gene therapy
for preventing or reversing hearing loss.

References

Denoyelle, F. . . . C. Petit. 1997. Prelingual deafness: High prevalence of a 30delG mutation in
the connexin 26 gene. Human Molecular Genetics 6:2173.

Everett, L.A. . . . E.D. Green. 1997. Pendred syndrome is caused by mutations in a putative
sulphate transporter gene (PDS). Nature Genetics 17(December).

Kelsell, D.P. . . . R.F. Mueller, and I.M. Leigh. 1997. Connexin 26 mutations in hereditary
non-syndromic sensorineural deafness. Nature 387(May 1):80.

Lawani, A.K. 1997. Gene therapy for hearing disorders. Association for Research
Otolaryngology meeting. St. Pete Beach.

Lynch, E.D., et al. 1997. Nonsyndromic deafness DFNA1 associated with mutation of a human
homolog of the Drosophila gene diaphanous. Science 278(Nov. 14):1315.

Furtherther Readings:

Weil, D. . . . K.P. Steel, et al. 1995. Defective myosin VIIA gene responsible for Usher
syndrome type 1B. Nature 374(March 2):60

Winata, S. . . . J.T. Hinnant, et al. 1995. Congenital non-syndromal autosomal recessive
deafness in Bengkala, an isolated Balinese village. Journal of Medical Genetics 32:336.

Sources:

Thomas B. Friedman
Michigan State University
Department of Zoology
203 Natural Science Building
East Lansing, MI 48824

Eric D. Green
National Institutes of Health
National Human Genome Research Institute
Genome Technology Branch
Bethesda, MD 20892

John T. Hinnant
Michigan State University
Department of Anthropology
East Lansing, MI 48824

Mary-Claire King
University of Washington
Departments of Medicine and Genetics
Seattle, WA 98195

Pedro E. Leon
University of Costa Rica School of Medicine
Center for Research in Cellular and Molecular Biology
San Jose 506 (GMT -6)
Costa Rica

Eric D. Lynch
University of Washington
Departments of Medicine and Genetics
Seattle, WA 98195

Robert F. Mueller
St. James' University Hospital
Molecular Medicine Unit
Leeds LS9 7TF
England

Christine Petit
Institut Pasteur
Unite de Genetique des Deficits Sensoriels
CNRS URA 1968
25 rue du Dr. Roux
75724 Paris Cedex 15
France

Karen P. Steel
Medical Research Council
Institute of Hearing Research
University Park
Nottingham NG7 2RD
England

                      copyright 1998 Science Service