ÐÓ°ÉÔ­´´

‘Deaf given hope by guinea pig,’ ran a newspaper headline last August.
The gist of the story was that guinea pigs in a university laboratory in
Britain had spontaneously recovered their hearing after it had been damaged
by powerful antibiotics. The finding ran counter to all expectations. The
prevailing belief was that inner-ear deafness – the kind of hearing loss
which some people suffer after taking antibiotics, being exposed to loud
noise, or simply as a result of ageing – was permanent in all mammals. And
for good reason: the deafness is caused by the apparently irreparable destruction
of the inner ear’s hair cells, so called because they are equipped with
hair-like antennae for detecting vibrations.

Just a few years ago the prospects for replacing damaged hair cells
in human ears looked bleak. The received wisdom was that the inner ears
of mammals could manufacture these cells once and once only – during development
in the womb. Last summer’s discovery by Carole Hackney and her colleagues
at the University of Keele received widespread attention, but this was just
part of a growing body of evidence to suggest that the inner ears of adults
may be able to produce new hair cells, after all.

Now, seven months later, biologists are celebrating a further breakthrough.
Researchers at the University of Virginia have discovered that tissue cultured
from the balance system of human inner ears can, contrary to expectations,
spontaneously produce new hair cells – a finding that is backed up by British
research which shows recovery in animals that have suffered substantial
hair cell loss . This discovery brings hope that some day there may be a
cure for the most common cause of deafness. ‘I’ve always been optimistic,’
says Jeff Corwin, who directed the research at the University of Virginia.
‘I believe that within our lifetime we are going to be able to help.’

The discovery that hair cells can be grown in the test tube is not the
only reason for optimism. The past decade has seen huge advances in our
general understanding of how the ear works, and in particular of the mechanisms
by which hair cells respond to vibrations deep within the ear’s sound analyser,
the cochlea. Along the way a long-standing puzzle has received the beginnings
of a solution: what makes the human ear so extraordinarily sensitive? The
cochlea can detect sounds softer than a pin dropping and discriminate between
sounds differing by as little as a twentieth of a tone. But how? Over the
past decade it has become clear that the answer has much to do with a subtle
partnership between two kinds of hair cell, the cochlea’s ‘inner’ and ‘outer’
hair cells.

Modern ideas about the role of the cochlea in hearing stretch back to
the 19th century and the theoretical work of the German scientist Hermann
von Helmholtz. It was von Helmholtz who first suggested that when vibrations
reach this fluid-filled helical tube they set in motion a structure known
as the basilar membrane. Bisecting the cochlea from top to bottom, this
membrane was likened by von Helmholtz to a piano’s strings. Shout loudly
into the body of a piano, and certain strings will resonate with your voice;
the basilar membrane, argued Helmholtz, responds similarly.

In the 1940s came the first direct evidence for these ideas. Experimenting
on ears from human corpses, Georg von Bekesy, a Hungarian engineer, showed
that sound energy moves through the cochlea as a travelling wave, being
propelled along the various ‘strings’ of the basilar membrane. Since the
basilar membrane cannot communicate directly with the brain, von Bekesy
reasoned that a second structure called the organ of Corti, which does have
sensory links with the brain, must act to interpret the information encoded
in the vibrating strings.

In the mid-1970s, the first experiments on ears from living animials
confirmed von Bekesy’s theory and started to reveal how the organ of Corti
works. Here, in this narrow strip of cells that runs parallel to the basilar
membrane, the significance of hair cells first came to light. We now know
that each string in the basilar membrane not only vibrates in response to
a particular sound frequency but stimulates a specific subset of ‘inner’
hair cells on the organ of Corti. Tiny electrical currents flow across the
cell’s membrane, and it is these currents which form the basis of the signal
the ear sends to the brain .

So far, so good. But von Bekesy noticed a problem. The structure of
the basilar membrane was such that any sound energy coming into the cochlea
should, by rights, have been rapidly dissipated. His mechanism could not
explain the ability to hear very soft noises and to distinguish between
tones with such accuracy. There seemed to be something else going on in
the cochlea to boost the energy in the basilar membrane and sharpen the
peak of its vibration. But what? Unfortunately, von Bekesy’s experiments
on the ears of corpses couldn’t provide the answer. To show the micromechanical
responses to sound at work, experiments on living tissue from mammals was
essential, and for these researchers had to wait for technical advances
in microscopy and better methods for probing the electrical properties of
cell membranes.

Mystery players

Forty years later the technology had arrived, and with it came a chance
discovery suggesting that the missing players in von Bekesy’s ensemble were
outer hair cells, a type of sensory cell unique to the ears of mammals.
Up until the mid-1980s outer hair cells had been a complete enigma. Unlike
inner hair cells, they were not wired up to deliver sensory information
to the brain. But in 1985 the enigma began to crack. William Brownell, collaborating
initially with researchers at the University of Geneva in Switzerland and
then with one of us (Bechara Kachar), at the National Institutes of Health
in the US, discovered that outer hair cells boast an unusual talent: they
can change length when stimulated by a change in voltage.

Initially nobody knew what to make of this finding. Both inner and
outer hair cells responded to vibrations by producing an electrical signal,
but it seemed that outer hair cells were also equipped to do the opposite:
vibrate in response to electrical signals. But to what effect? An important
additional clue was the location of outer hair cells. The inner hair cells
are located near to the attachment of the basilar membrane, where it hardly
vibrates. But all mammals have at least three rows of outer hair cells,
and these are positioned above the section of the basilar membrane which
moves the most in response to sound.

An idea gradually took hold among researchers. If outer hair cells could
make contact with this section of the basilar membrane as they vibrate,
perhaps the cells could amplify the membrane’s oscillations in the same
way one might push a child on a swing. By sustaining the oscillations in
this way the outer hair cells would, in effect, be sharpening the frequency
resolution of the basilar membrane. In a sense, they would act like miniature
amplifiers inside the cochlea, taking sound energy and feeding it back through
the basilar membrane.

A compelling theory, but could we prove it? The first thing to establish
was that outer cells were capable of vibrating at high enough frequencies,
and with enough force, to influence the movements of the basilar membrane.
The answers came from a powerful and relatively new approach to probing
the electrical properties of cell membranes, the so-called patch-clamp technique.
Developed in Germany in the early 1980s by Nobel prizewinners Erwin Neher
and Bert Sakmann, this involves sealing the tip of a fine glass electrode
onto the surface of a cell to monitor the electrical behaviour of a small
patch of membrane.

The challenge with outer hair cells was to discover how the electrical
behaviour of their membranes linked up with their ability to change length.
Here the patch-clamp technique was a boon. It enabled researchers to alter
the voltage or current across the membrane of an outer hair cell while simultaneously
manipulating it in other ways – injecting chemicals into the cell, sucking
out cellular fluids, and so on.

The patch-clamp technique was used for the first time on outer hair
cells in 1987, and progress was swift. Jonathan Ashmore, an electrophysiologist
at the University of Bristol, quickly showed that the cells could indeed
generate forces big enough to influence the basilar membrane, and at frequencies
of at least 10 kilohertz (the human ear can detect frequencies ranging between
about 20 hertz and 20 kilohertz).

From Ashmore’s experiments the following picture emerged. When the hair
‘bundles’ on the surface of an outer hair cell are stimulated to move, they
trigger a voltage change across the cell’s membrane, which in turn signals
the cell to change its length. Moreover, this length change is proportional
to the voltage change – a crucial finding which meant that the cells would
always vibrate at exactly the same frequency as their hair bundles, and
by implication the basilar membrane.

Dancing to the music

It was at this point that I (Holley) became interested in the problem.
In Bristol, I was urged by Ashmore to look down a microscope at cells taken
from an animal’s cochlea. What I saw was a single outer hair cell dancing
like a human figure to music. Suspecting that these movements were driven
by a very novel biological process, I gave up my work on cilia and began
to search for the mechanism. Meanwhile, in Washington DC, Kachar had already
begun to examine outer hair cells using the latest and most powerful techniques
in microscopy. We joined forces in December 1990, in a collaboration with
Federico Kalinec at the National Institutes of Health in Washington.

It soon became apparent that the way outer hair cells change length
is quite different from any other comparable process in the body. The contractions
of muscle cells and cilia require energy in the form of ATP; not so those
of outer hair cells. Nor do the movements of outer hair cells depend quite
so critically on calcium ions flowing into the cells leading to a change
in membrane potential. Intrigued by these anomalies, we began to probe more
deeply into the workings of hair cells. How do these cells detect voltage
changes across their membranes? How do they convert electrical signals into
mechanical forces? Where are the molecular motors that generate movement?

At first we thought the mechanism was located inside the cell. But by
destroying the intracellular structures without breaking the membrane, and
showing that the outer hair cell could still move, we concluded that the
mechanism must be somewhere in the membrane itself. All cell membranes contain
protein molecules, and outer hair cells are no exception. However, electron
microscopy studies revealed that, in this case, the proteins are unusually
densely packed, giving the cell surface a texture resembling a cobbled yard.
Was this cobbled yard the secret to the cells’ ability to change shape?
Our hunch was confirmed by pictures from high-resolution video cameras showing
the membrane contracting in response to voltage changes across its surface.
So the first puzzle was cracked: protein molecules in the cell membrane
are the motors that drive the vibrations of hair cells. The race is now
on to identify and isolated these proteins so that we can study them.

Exactly how these proteins cause the membranes in which they are embedded
to contract is still a mystery. But evidence from other cells suggests that
the protein molecules might do so simply by altering their shape. There
are many examples in nature of membrane proteins changing shape in response
to a voltage change. The best known are so-called ion channels that alter
their shape to open or close a pore which lets ions flow across the membrane.
If all the protein molecules in the membrane of a hair cell were to change
their shape by only a few per cent this would change the shape of the cell.

By the end of last year enough pieces of the puzzle had fallen into
place for us to explain how outer hair cells might act like tiny amplifiers
within the ear. There are several stages. In the first, sound sets in motion
the basilar membrane, which in turn stimulates the bundles on outer hair
cells to move. These vibrating bundles then lead to voltage changes across
the cell membranes which cause the cells themselves to vibrate. Their vibrations
feed directly back to the basilar membrane, whose modified oscillations
are detected by inner hair cells. Finally, inner hair cells transmit electrical
signals to nerves that carry information about the frequency of the sound
to the brain.

Without outer hair cells our hearing would be 100 times less sensitive
to sound. What noise we could hear would be distorted and fuzzy. A greater
understanding of their mechanism, so crucial to hearing, must surely bring
hope to the 6 per cent of people who suffer debilitating deafness.

Matthew Holley is a Royal Society research fellow in the Department
of Physiology in the School of Medical Sciences, University of Bristol.
Bechara Kachar is acting chief of Structural Cell Biology at the National
Institute of Deafness and Other Communicative Disorders in Washington DC.

* * *

1: SOUNDS PROMISING

The hair cells you are born with must last a lifetime. Should any of
them become damaged, that damage is permanent – as any rifleman will testify.
Fish and amphibians are luckier: their ears continue to produce new hair
cells throughout development. This encouraged the somewhat pessimistic view
that perhaps only growing ears can regenerate hair cells.

But in the late 1980s experiments on young chicks, whose ears stop growing
before birth, suggested otherwise. After being exposed to loud noises,
chicks’ ears generate new hair cells to replace those that have been damaged.
At the same time Doug Cotanche and Jeff Corwin, at the Universities of Hawaii
and Boston respectively, discovered that these new hair cells are supplied
by division of the ‘supporting’ cells that surround hair cells in the ears
of chicks.

Now Corwin and his colleagues report in Science (12 March) that human
hair cells may be similarly regenerated in the test tube. The key was to
culture both hair cells and supporting cells artificially in the same dish,
a technique pioneered by Guy Richardson at the University of Sussex. The
cultures were exposed to antibiotics that killed the hair cells but not
their supporting cells. Within two days new hair cells began to appear,
and after a week regeneration was well under way. As in birds, the new cells
were produced by division of their supporting cells.

Criticism that these results could be attributed to the artificial conditions
of the cultured tissue are countered in the same issue of Science. Andrew
Forge and his colleagues at the University of London Institute of Laryngology
and Otology, describe a similar recovery in damaged ears of living guinea
pigs.

Unfortunately, there is a catch. The hair cells regenerated by the researchers
come not from the hearing system but from the balance system. Although both
systems are located in the inner ear and use hair cells to detect vibrations,
their structural organisation is completely different.

Experience tells us that the hearing system in humans cannot regenerate
spontaneously. But experiments do indicate that repair is physically possible.
Corwin’s team is already testing a chemical derived from vitamin A which
stimulates production of new hair cells in damaged tissue from mouse embryos.

* * *

2: THE INNER SECRETS OF HEARING

Hair cells are so named because of the small bundle of fine hairs that
projects vertically from the cell surface. It is here that sound vibration
is converted into nerve signals. Seen through a scanning electron microscope
the hair bundles stand proud on top of their cells like sets of organ pipes.
Each hair is between 1 and 5 micrometres long, straight and cylindrical
with a tapered base. About 100 hairs are arranged on each cell, in several
rows with the tallest at the back and the shortest in front. Jim Pickles
at the University of Birmingham showed that the tips of the shorter hairs
are connected to the sides of their longer neighbours by thin cables called
tip-links, each composed of just a few protein molecules.

Measurements on bullfrog hair cells by Jim Hudspeth at the University
of California at San Francisco showed that tiny electrical currents occured
in the region of the tip-links when the bundles were moved. Such experiments
have led to the present theory of how mechnical changes lead to electrical
changes within hair cells.

When the longer hairs are tilted away from the shorter ones the tip-links
are stretched like elastic bands and each one pulls open a tiny molecular
pore embedded in the surrounding cell membrane. The inner surface of the
cell membrane is normally negatively charged compared with the outer surface,
but this polarisation changes dramatically when the pores are opened and
positively charged potassium ions rush into the cell. The electrical signal
thus generated is transmitted, via the nerve terminal at the base of the
cell, to the brain.