Actin Filaments, Stereocilia, and Hair Cells of the Bird Cochlea. V. How the Staircase Pattern of Stere iliary Lengths Is Generated

The stereocilia on each hair cell are arranged into rows of ascending height, resulting in what we refer to as a "staircase-like" profile. At the proximal end of the cochlea the length of the tallest row of stereocilia in the staircase is 1.5 gm, with the shortest row only 0.3 gm. As one proceeds towards the distal end of the cochlea the length of the stereocilia progressively increases so that at the extreme distal end the length of the tallest row of the staircase is 5.5 ~tm and the shortest row is 2 I.tm. During development hair cells form their staircases in four phases of growth separated from each other by developmental time. First, stereocilia sprout from the apical surfaces of the hair cells (8-10-d embryos). Second (10-12-d embryos), what will be the longest row of the staircase begins to elongate. As the embryo gets older successive rows of stereocilia initiate elongation. Thus the staircase is set up by the sequential initiation of elongation of stereociliary rows located at increased distances from the row that began elongation. Third (12-17-d embryos), all the stereocilia in the newly formed staircase elongate until those located on the first step of the staircase have reached the prescribed length. In the final phase (17-d embryos to hatchlings) there is a progressive cessation of elongation beginning with the shortest step and followed by taller and taller rows with the tallest step stopping last. Thus, to obtain a pattern of stereocilia in rows of increasing height what transpires are progressive go signals followed by a period when all the stereocilia grow and ending with progressive stop signals. We discuss how such a sequence could be controlled. H ow a cell controls the position, direction, and length of its extensions or processes is a question that has interested cell and developmental biologists for many years. As more information has accumulated on the cytoskeletal elements in these extensions the question has gradually been honed down to one dealing with how a cell controls the length, number, and orientation of its major cytoskeletal components, microtubules, and actin filaments. The hair cells in the cochleae of vertebrates are ideal to study in this regard because from their surface are numerous extensions or stereocilia whose length, width, number, and orientation are rigorously defined (see 14 for references). Within the membrane that limits each stereocilium is a cross-bridged bundle of actin filaments which gives the stereocilium its shape and characteristics. Thus if we can discern what regulates the length of the actin filaments, the number of actin filaments per bundle, and the position of the actin filaments in the cytoplasm, we should be able to begin to understand how the cell exerts control over the physical dimensions of its cell extensions. In earlier papers in this series we described how the hair cells form and elongate their stereocilia. We related these macroscopic events to the number and distribution of the actin filaments and cross-bridges that lie within the stereocilia and the apical cytoplasm (cuticular plate), and began to formulate testable predictions about how the physical dimensions of the stereocilia might be controlled (13, 18). Before we test some of these predictions there are additional aspects of stereociliary differentiation that must be described in greater detail. One of these concerns the question of how an individual hair cell grows stereocilia of different, yet predictable lengths, off the same surface. We know that the stereocilia increase in length from the "front" of the hair cell bundle to the "back, but are approximately equal within a row of stereocilia across the hair bundle (7). In short the stereocilia are arranged into rows of increasing height presenting to the viewer a "staircase-like" profile. There are many ways that a staircase could be developed. One is to generate some type of gradient in the cell such that some stereocilia, depending on their location, grow at a progressively greater rate than others. Another is to form different steps of the staircase at different developmental times, and a third is to grow all the stereocilia at the same rate but either start or stop the growth of rows of stereocilia at earlier or later times. What we find is that the development of the staircase oc9 The Rockefeller University Press, 0021-9525/88/02/355/11 $2.00 The Journal of Cell Biology, Volume 106, February 1988 355-365 355 on N ovem er 6, 2017 jcb.rress.org D ow nladed fom curs in four phases separated from each other by developmental time. In short there are progressive go signals followed by a period when all the stereocilia grow and ending with progressive stop signals. "lhus a complex pattern can be built up following simple rules during developmental time. Materials and Methods Fertilized chicken eggs of the White Leghorn variety were obtained from a local supplier and incubated at 37~ Hatching occurred at day 21. Dissection, Fixation, and Mounting of the Cochleae The cochleae of embryos and chicks were dissected as outlined in Tilney and Saunders (14) and Tilney et al. (18). Fixation was carried out by immersion of the cochlea, still surrounded on its basal and lateral surfaces with cartilage, in a freshly made solution of 1% OsO4 in 0.1 M phosphate buffer at pH 6.3. Fixation was carried out at 0~ for 45-60 rain. After fixation the cochleae were dehydrated in acetone to 75 %. The tegmentum vasculosum was then removed and the tectorial membrane lifted free with fine forceps. At this point as much as possible of the cartilage and connective tissue attached to the superior surface of the cochlea was removed as was possible because to see the staircase one must "look across" the superior surface of the cochlea. The cochleae were then dehydrated in pure acetone, critically point dried, oriented on stubs and sputter coated (14, 17). The orientation on the stub is important as the detector of our scanning electron microscope (AMR 1000) is situated at 90* to the column. Thus the greatest signal from the specimen is when the stub is oriented at 45 ~ . Since we want to see the staircase by looking across the surface of the cochlea, the greatest signal is obtained if the superior surface is inclined upward from the stub by •45~ Determination of the Lengths of the Stereocilia Afar the specimen was oriented appropriately so that the stereociliary bundles stand upright in front of the viewer, a series of photographs was taken at a magnification of 8,000-16,000 (depending upon the length of the stereocilia) at five locations along the superior surface of each cochlea. The negatives of these photographs were enlarged 2.5 times to form 8 x 10 inch prints of the bundles. From the prints we could accurately measure the lengths of the stereocilia. At least 10 and usually many more adjacent stereociliary bundles were photographed for each of the five locations. We first photographed the bundles looking toward the front of the staircase or the side on which stereocilia of increasing height were visible, then we rotarsi the cochlea 180 ~ and tilted it appropriately, and took photographs of stereociliary bundles examined from their back side in which only the tallest row was visible. Thus we photographed both the front and back surfaces of the same cells or cells located in the immediate vicinity. Measurements of the length of the stereocilia were then made with a ruler under a lit magnifier. There is little ambiguity as to the length of the tallest row of stereocilia when viewed from the back of the staircase (see references 14 and 17 for a discussion of the accuracy). The average length of the shortest row of stereocilia is more difficult as they are not of uniform length. Sometimes there are tiny microvilli at the base of the first row. These do not seem to be stereocilia as they do not have the same widths as the other stereocilia in the bundle, all of which seem to be remarkably constant in width. We did not include these microvilli in our measurements. On the graphs (Figs. 4, 7, 8, and 9) where we express the length of stereocilia as a function of position, we included error bars for each point to illustrate one standard deviation. Above these we indicate the number of measurements made. We have also included a table to show how accurately biology can determine lengths. In this table we show the largest and smallest value that we measure for each point.