I'd like to describe an experimental system that Denny Freeman and I have developed to study cochlear mechanics and then describe some of our preliminary results.
Let me begin by describing our experimental preparation. We isolate the cochlea from an alligator lizard and clamp it over a hole in our experimental chamber. The hole was designed to mimic the bony ring that supports in cochlea in situ. Using a piezo, we modulate the pressure in the fluid, just as the stapes does in vivo. We calibrate the sound source with an underwater microphone built into the chamber directly beneath the basliar membrane. Using a compound microscope and a strobe light, we take stop action pictures of the resulting audio frequency motion.
Our in vitro preparation allows imaging of all parts of the cochlea.
The drawing represents a cross sectional view of the cochlea. The cluster of hair cells rest on the basilar membrane, and are surmounted by a gelatinous tectorial membrane.
In our in vitro preparation, we view the cochlea in the direction indicated by the arrow, and the focal plane of the microscope is indicated by the dashed line. The left panel shows the image seen at this focal plane, which is through the top of the tectorial membrane.
By changing the focal plane of the microscope, different parts of the cochlea become visible, a property called optical sectioning.
still picture (gif): 88k
At this plane of section, one can see the tips of several hair bundles. The edge of the tectorial membrane can also be seen, expecially at the right margin of the image.
still picture (gif): 101k
At this plane of section, the bases of several hair bundles can be seen, as well as the right boundary of the tectorial membrane. This endolymphatic surface is also called the reticular lamina.
Additional structures of the cochlea can be seen as the plane of focus is changed.
movie (mpeg): 606k
Such sets of two-dimensional images characterize the entire three dimensional structure of the cochlea.
Our in vitro preparation provides for hydrodynamic stimulation at audio frequencies.
The sinusoidal stimulus is illustrated here. Using stroboscopic illumination, we obtain stop-action images, which in this case are synchronized to 8 different phases of the stimulus. The arrow in the drawing indicates the phase at which the image on the left was acquired.
By viewing the images in sequence, we can see the audio-frequency movements in slow motion.
We can repeat these measurements at other planes of focus.
movie (mpeg): 135k
Thus, by combining stroboscopic illumination with optical sectioning, we obtain a temporal sequence of three-dimensional images, which can then be analyzed to estimate the motions of any structure in the organ.
We have made measurements of sound-induced motions of cochlear structures in 6 preparations, each from a different alligator lizard.
We have studied motions resulting from tonal stimuli with frequencies between 40 Hz and 2 kHz at moderate sound pressure levels.
We show results from one experiment to illustrate how the motion measurement system can be used to address a spectrum of issues in cochlear mechanics.
These images illustrate motions that result for a 513 Hz, 78 dB SPL tone.
The left panel shows the top of the tectorial membrane. The right panel shows the bases of several hair bundles, the right edge of the tectorial membrane, and the reticular lamina.
Notice that motion of the tectorial membrane is significantly smaller than the motion of the reticular lamina.
Although have had some variability across the preparation, to date we have never seen the tectorial membrane move more than the reticular lamina. Results have been most repeatable above 200 Hz, where the tectorial membrane consistantly moves less than the reticular lamina.
To examine cochlear micromechanics, we can electronically zoom in on one particular hair bundle.
These panels illustrate images of one hair bundle and of the overlying tectorial membrane at six different planes of focus separated by 3 micrometers. The dashed line in the drawing indicates the plane of section for the highlighted image.
These images illustrate the motion that results when the receptor organ is stimulated at 513 Hz.
As discussed earlier, it is clear that motions of the hair bundle are greater than those in the tectorial membrane.
But the new images provide information about motions at the level of the bundle. Specifically, we can identify the kinociliary bulb, and see that the bundle is moving parallel to its excitatory direction. If we could estimate the magnitude of the motion from the video images, then we could calcutate how much the bundle is rotating.
Now, sound-induced motions of cochlear structures are very small --- much smaller than the resolution of either our microscope or video camera. Nevertheless, as you've seen these motions can be detected from the video images --- and the basic idea behind quantitatively estimating that motion is illustrated here. Each square represents a pixel in the CCD camera. The circle represents the image of a specimen as it moves across the face of the camera. The bar plots at the bottom represent the gray value generated at each pixel in the row indicated by the arrow. Although the circle moves less than the size of a pixel, the gray values generated by the pixels near the edge vary.
We've developed an algorithm which takes advantange of these variations in gray value to estimate motions that are smaller than a pixel.
Using optical magnification of 100 X, we can measure motions of cochlear structures as small as 14 nm without averaging. With averaging, even smaller motions can be measured.
So now let's return to the images of the hair bundle and tectorial membrane and quantify the motion.
Above or below each image is a plot of the average horizontal motion in that image. The dots indicate the phase at which the images shown were acquired.
The magnitude of the motion decreases as we go from the base to the top of the tectorial membrane. The base moves .5 micrometers peak-to-peak, the tips, .25 micrometers, and the top of the tectorial membrane moves .18 micrometers.
The phase of the motion relative to the electrical stimulus generally decreases from the base to the tectorial membrane as well. The phase lag starts at 10 degrees and increases to a lag of 82 degrees.
While these absolute motions are interesting, the physiologically relevant motion is the rotation of the hair bundle about its base. Using the quantitative motion estimates, we can shift all of the pictures to compensate for the motion of the base. The result is movies of motion relative to the base.
In this frame of reference, motion of the base of the hair bundle, shown in the lower right, is much reduced. Relative motions of the tips of the stereocilia, shown in the lower left, are large --- showing that the hair bundle is being displaced in its excitatory direction. Relative motions of the tectorial membrane are still larger.
From the quantitative analysis of images like these, we will be able to determine the relation between motion of the tectorial membrane and rotation of the hair bundle --- a relation generally called micromechanics.
We can electronically zoom-in even more to concentrate on the motions of individual stereocilia.
Motions in these images, which are at a focal plane midway between the tip and base of the hair bundle, show that the hair bundle does not rotate as a rigid body. Rather, there is motion between the individual stereocilia.
To focus on the relative motion of stereocilia, we can estimate the motions of stereocilia at the right margin of the hair cell, and electronically shift the pictures to freeze that edge.
The resulting images clearly show motions between stereocilia. Focus for example on any of the stereocilia near the center of the bundle. The distance between it and the stereocilia to its left and right clearly change in synchrony with the hydrodynamic stimulation. Transduction channels in hair cells are thought to be stimulated by stretching "tip-links" that connect neighboring stereocilia.
This video shows that there is a dynamical relation between hair bundle rotation and iter-ciliary distance, a transformation that we call nanomechanics. This nanomechanical relation could have important signal-processing properties including possibly tuning of the hair cell.
However, we cannot yet conclude that this stereociliary motion is typical. At present, we can only assess viability using visual cues at the light microscope level. We are currently developing methods that will allow us to routinely measure the electrical response of the cochlea as an independent viablilty test.
To summarize. We've developed an in vitro preparation of the cochlea of the alligator lizard in which we can deliver calibrated hydrodynamic stimuli. We use stroboscopic illumination and optical sectioning to construct sequences of three-dimensional images that characterize the sound-induced motions of cochlear structures. From these images we can compute the motions of any visible structure. This new method