Cell Motility and the Cytoskeleton 38:146–162 (1997) Localization of Microtubules Containing Posttranslationally Modified Tubulin in Cochlear Epithelial Cells During Development Joyce Tannenbaum1,2 and Norma B. Slepecky1,3* 1Department of Bioengineering and Neuroscience, Institute for Sensory Research, Syracuse University, Syracuse, New York 2Department of Biology, Syracuse University, Syracuse, New York 3Department of Anatomy and Cell Biology, SUNY Health Science Center, Syracuse, New York In the adult gerbil inner ear, hair cell microtubules contain predominantly tyrosinated tubulin while supporting cell microtubules contain almost exclusively other isoforms. This cell-type specific segregation of tubulin isoforms is unusual, and in this respect the sensory and supporting cells in this sensory organ differ from other cells observed both in vivo and in vitro. Thus, we hypothesized there must be a shift in the presence and location of tubulin isoforms during development, directly associated with the onset of specialized functions of the cells. We describe the appearance and/or disappearance of tubulin isoforms in sensory hair cells and five different supporting cells (inner and outer pillar cells, Deiters cells, cells of Kölliker’s organ, and cells of the tympanic covering layer) during development of the gerbil organ of Corti from birth to 14 days after birth. Tyrosinated tubulin was initially present in all cells and remained predominant in cells that decrease in number (Kölliker’s organ and tympanic covering layer) and exhibit active processes such as secretion and motility (sensory cells). Posttranslational modifications occurred in the supporting cells in a time-dependent manner as the number and length of microtubules increased and development proceeded, but the establishment of elongated cell shape and polarity occurred prior to the appearance of acetylation, detyrosination, and polyglutamylation of tubulin. In the pillar and Deiters cells, posttranslational modifications progressed from cell apex to base in the same direction as microtubule elongation. In the pillar cells, posttranslational modifications occurred first at the apical surfaces. In the pillar cells, the appearance of acetylated tubulin was rapidly followed by the appearance of detyrosinated tubulin. In Deiters cells, the appearance of acetylated tubulin preceded the appearance of detyrosinated tubulin by one or more days. At onset of cochlear function, detyrosinated tubulin and acetylated tubulin had achieved their adult-like pattern, but polyglutamylated tubulin had not. Cell Motil. Cytoskeleton 38:146– 162, 1997. r 1997 Wiley-Liss, Inc. Key words: tubulin; posttranslational modifications; cochlea; organ of Corti; development INTRODUCTION Microtubules are prominent structures in the sensory and supporting cells of the organ of Corti in the inner ear. Based on studies of other cell types, it is thought that r 1997 Wiley-Liss, Inc. *Correspondence to: Norma Slepecky, PhD, Dept. of Bioengineering and Neuroscience, Institute for Sensory Research, Syracuse University, Syracuse, NY 13244-5290. E-mail Norma_Slepecky@isr.syr.edu Received 27 December 1996; accepted 30 April 1997 Modified Tubulin in Cochlear Epithelial Cells their functions include: movement of subcellular organelles [Araki et al., 1993; Ayscough et al., 1993; Baker et al., 1993], establishment of cell polarity [Hyde and Hardham, 1993; Glasgow and Daniele, 1994], generating and maintaining cell shape [Bulinski and Gundersen, 1991; Joshi and Baas, 1993], and provision of structural support [Deanin et al., 1981; Kirschner and Mitchison, 1986; Bulinski and Gundersen, 1991; Audebert et al., 1993]. In these cells, microtubules are divided into at least two subpopulations consisting of tyrosinated tubulin and posttranslationally modified isoforms (detyrosinated, acetylated, polyglutamylated, and nontyrosinatable tubulin). They differ with respect to their kinetics of polymerization and depolymerization and their biochemical composition appears to be predictive of their dynamic properties [Gundersen et al., 1984, 1987; Thompson et al., 1984; Piperno et al., 1987; Schulze and Kirschner, 1987; Baas and Black, 1990]. The determination of cell polarity and shape and the permanence of structures containing microtubules are related to their dynamic properties [Gundersen et al., 1987; Schulze and Kirschner, 1987]. Microtubules form at nucleating centers and elongate when newly synthesized tubulin dimers are preferentially added to the free (non-nucleated) ends. Microtubules shorten by depolymerization, at which time tubulin dimers are removed from the free ends. Dynamic, or ‘‘short-lived,’’ microtubules polymerize and depolymerize suddenly and rapidly as microtubules extend and retract. This is associated with changes in cell shape and movement. Stable, or ‘‘longlived,’’ microtubules display little or no turnover of tubulin subunits, resulting in microtubules that are relatively constant in length. These would have the ability to provide sustained tension and/or support. Dynamic properties may be determined by the biochemical composition of tubulin. In microtubules where depolymerization and polymerization occur rapidly, tubulin is almost entirely in the tyrosinated form [Schulze et al., 1987; Bulinski and Gundersen, 1991] and its presence provides a marker for dynamic microtubules. On the other hand, tubulin present in long-lived microtubules has an increased chance of being modified [Gundersen et al., 1987; Schulze et al., 1987; Bulinski and Gundersen, 1991]. Thus, posttranslational modifications are correlated with the presence of stable microtubule populations and appear in a time-dependent manner [Kirschner and Mitchison, 1986; Gundersen et al., 1987; Piperno et al., 1987; Schulze and Kirschner, 1987; Baas and Black, 1990; Paturle-Lafanechere et al., 1994]. Removal of tyrosine yields detyrosinated tubulin and this modification may be restricted to tubulin in microtubules [Arce et al., 1978; Kumar and Flavin, 1981; Thompson, 1982; Gundersen et al., 1987]. Other modifi- 147 cations that occur primarily to polymerized tubulin include the addition of an acetyl group to lysine residue 40 of a-tubulin to form acetylated tubulin [L’Hernault and Rosenbaum, 1983, 1985] and the addition of one or more glutamyl residues to the glutamic acid residues at the carboxy terminal end of a- and b-tubulin to form polyglutamylated tubulin [Eddé et al., 1990; Wolff et al., 1992]. Removal of the last two amino acids from the carboxy terminal end of a-tubulin prevents tyrosine reattachment and produces tubulin which is nontyrosinatable [Alonso et al., 1993; Paturle-Lafanechere et al., 1994]. In many studies using tissue-culture, the appearance of posttranslational modifications to tubulin is correlated with the loss of dynamic microtubule activity and the stabilization of cell shape. However, few model systems exist to study the role of posttranslational modifications, since most mature cells in vivo, and all cells in culture, contain mixed populations of microtubules which are stable and dynamic, as well as being tyrosinated and posttranslationally modified. The organ of Corti of the mammalian auditory system presents a unique system in which to undertake developmental studies of microtubules in vivo because in the adult organ there are cells which are actively motile (sensory cells) and cells which provide structural support (pillar and Deiters cells). Sensory hair cell microtubules are present in loose networks [Slepecky and Chamberlain, 1985; Steyger et al., 1989; Furness et al., 1990; Slepecky and Ulfendahl, 1992] and contain predominantly tyrosinated tubulin [Slepecky et al., 1995]. In contrast, microtubules in the supporting cells are in discrete, tightly packed bundles [Angelborg and Engstrom, 1972] containing almost exclusively acetylated, detyrosinated, and polyglutamylated tubulin [Slepecky et al., 1995]. Because this cell-type specific segregation is so unusual, we hypothesized that there must be a shift during development, from newly synthesized tyrosinated tubulin to posttranslationally modified isoforms, directly associated with the onset of specialized cell functions. Specifically, we were interested in: 1) the relationship between the appearance of posttranslational modifications to tubulin and the time during which cell polarity is established; 2) the temporal sequence of the posttranslational modifications to tubulin as it relates to changes in cell number, shape, and size; and 3) how the location of microtubules in the sensory and supporting cells changes as development proceeds and how this correlates with the onset of hearing. In the adult inner ear, the sensory epithelium of the cochlea spirals around a central core of nerve fibers. In a cross-section taken through each turn, sensory cells (one inner and three outer hair cells) and supporting cells (inner pillar, outer pillar, and three Deiters cells) can be 148 Tannenbaum and Slepecky seen. Cells in the apical turn code for low-frequency sounds and those in the middle and basal turns code for progressively higher frequencies. In the gerbil, as in other species [Romand, 1987; Lim and Rueda, 1992; Roth and Bruns, 1992a, 1992b; Pujol et al., 1996], development of the organ of Corti starts first in the basal turn and progresses toward the apical turn [Ito et al., 1995; Souter et al., 1995; Kuhn and Vater, 1996]. Common to all species is the initial presence of Kölliker’s organ, made up of tall densely packed columnar cells containing irregularly shaped nuclei, and an undifferentiated mass that will become the sensory and supporting cells. Development proceeds (see Fig. 1) until the onset of cochlear function, which coincides with the disappearance of Kölliker’s organ, the formation of the inner spiral sulcus, the maturation of the sensory and supporting cells, the opening of the tunnel of Corti between the inner pillar and outer pillar and the spaces of Nuel between the outer hair cells and Deiters cells, the maturation of the tectorial membrane and basilar membrane, and the formation of synapses with afferent and efferent nerve terminals. In the gerbil, these changes take place from 0–14 days after birth [Finck et al., 1972; Harris and Dallos, 1984; Arjmand et al., 1988; Echteler et al., 1989; Harris et al., 1990; Woolf et al., 1992; Ito et al., 1995; Munyer and Schulte, 1995; Souter et al., 1995; Kuhn and Vater, 1996; Schweitzer et al., 1996]. Thus, the experiments described in this paper were designed to study the appearance and/or disappearance of the different isoforms of tubulin during this time. MATERIALS AND METHODS Dissection Following anesthesia by carbon dioxide inhalation, both temporal bones were removed from gerbils at several hours, one day (24–36 h), seven days, ten days, or fourteen days after birth (dab). Prior to 10 dab, when the middle ear space is not opened and the cochlea not calcified, each cochlea was exposed and the temporal bone was dropped directly into fixative. For gerbils 10 dab and older, holes were made in the apex and base of the cochlea and fixative was perfused through the holes. Both ears from two gerbils of the same age were processed and sections were taken from all ears. The care and use of animals was approved by the Syracuse University IACUC. Fixation Cochleas were fixed in freshly made 3% paraformaldehyde, 0.1% glutaraldehyde, 0.1 M sodium phosphate buffer (PB), pH 7.5, for approximately three days at room temperature. Following one day to one week decalcification in 5% EDTA, cochleas were washed in PB, dehy- Fig. 1. Development of the organ of Corti based on individual tissue sections taken from the basal turn of the cochlea, selected from specimens at different ages during development: dab 5 days after birth; BM 5 basilar membrane; CC 5 Claudius cells; D 5 Deiters cells; IHC 5 inner hair cells; IP 5 inner pillar cells; ISS 5 inner spiral sulcus; KO 5 Kölliker’s organ; OHC 5 outer hair cells; OP 5 outer pillar cells; TM 5 tectorial membrane; TC 5 tunnel of Corti; TCL 5 tympanic covering layer; SV 5 spiral vessel. drated through increasing concentrations of alcohol and propylene oxide, infiltrated in Araldite Durcupan ACM, oriented, and hardened at 45°C for one week. Light Microscopy Representative one-micron midmodiolar sections were cut from all ears. They were placed in drops of water Modified Tubulin in Cochlear Epithelial Cells on gelatin/chrome alum subbed slides, dried down at room temperature, and stained with methylene blue and Azure II. Sections were examined by light microscopy to confirm that preservation by this fixation procedure, which is optimal for immunocytochemistry, was adequate to allow identification of the cell types and structures, and that a progression of developmental stages was visible. Camera lucida drawings from these sections were used to make the diagrams in Figure 1. For comparative purposes, most micrographs in this paper are from the basal turn because it is the only one present at all stages of development. However, since development of the organ of Corti proceeds from the basal to the apical turn, with a lag of 1–2 days in the most apical portion of the cochlea, observation of cells in the middle and apical turns provide information regarding subtle changes in the developmental sequence. Immunocytochemistry For immunocytochemistry, Araldite was removed from additional one-micron midmodiolar sections using a solution of propylene oxide / absolute ethanol / sodium hydroxide pellets. Sections were hydrated from ethanol to water, washed in phosphate buffered saline (PBS) and processed for immunocytochemistry as described previously [Slepecky and Ulfendahl, 1992]. Specimens were treated sequentially with goat serum to block nonspecific activity, primary antibodies diluted 1:100 in PBS, PBS as a wash, goat serum as a blocker, secondary antibodies diluted 1:200 in PBS, and PBS as a wash. Slides were dipped in distilled water to remove salt, air dried, and stored in the dark at room temperature until viewed. In some cases, double-labeled sections were treated sequentially with one primary antibody followed by the secondary antibody and then another primary antibody followed by the secondary antibody. In other cases, simultaneous incubation in a mixture of both primary antibodies was followed by simultaneous incubation in a mixture of the appropriate secondary antibodies (each at the correct final dilution). Control tissue sections were incubated in a primary antibody and an inappropriate secondary antibody to show that there was no cross-reactivity of the secondary antibodies used in the double-labeling experiments. Prior to use, the slides were coverslipped with glycerol:n-propyl gallate used as mounting medium. Tissue sections were photographed with a Zeiss Axioskop microscope using a 1003 oil immersion lens, 100W fluorescent light source, and Kodak TMax 400 film. Primary Antibodies Antibodies for alpha-tubulin (one of the two repeating subunits of tubulin in all microtubules), obtained from Chemicon (El Segundo, CA, AB935) and Sigma ImmunoChemicals (St. Louis, MO, T9026) were used to localize 149 microtubules regardless of the presence of posttranslational modifications. Antibodies to tyrosinated and acetylated alpha-tubulin were obtained from Sigma ImmunoChemicals (St. Louis, MO, T9028 and T6793). Antibody to polyglutamylated tubulin was a gift from Dr. Annie Wolff, College de France, Paris, France [Wolff et al., 1992]. Antibody to detyrosinated tubulin was a gift from Dr. J. C. Bulinski, Columbia University, NY, [Gundersen et al., 1984]. The antibodies were checked by immunoblotting for specificity to tubulin in control tissue (cerebellum) and organ of Corti [Slepecky et al., 1995]. Secondary Antibodies Appropriate secondary antibodies coupled to fluorescent markers were used to localize primary antibody binding. They included antibodies against rabbit immunoglobulins coupled to fluorescein (Accurate Chem. and Sci., Westbury, NY, JGZ1508); antibodies against mouse immunoglobulins coupled to fluorescein (Chemicon, Temecula, CA, AP130F); antibodies against rabbit immunoglobulins coupled to rhodamine (Accurate Chem. and Sci., JGZ2508); antibodies against mouse immunoglobulins, coupled to rhodamine (Accurate Chem. and Sci., JGM2544). RESULTS Type and Location of Microtubules in the Developing Organ of Corti As obvious physical changes occurred in cells during maturation of the organ of Corti, there were concomitant alterations in microtubules—changes in number, cellular location, and biochemical composition. These were observed immunocytochemically using isoform specific antibodies to tubulin on tissue sections of the developing organ of Corti. Not all results described in this section can be seen on the few cross-sections through the organ of Corti presented in the figures because of the complex configuration of the cells (Fig. 2), especially in the later stages of development when the cells have increased in size. If one were to take as a point of reference the apical surface of one outer hair cell and the apical surfaces of its adjacent outer pillar and Deiters cells, one would see that the base of the outer hair cell slants in one direction (toward the apical turn of the cochlea), while the basal portions of the pillar and Deiters cells slant in the opposite direction (toward the base of the cochlea). Depending on the plane of section, which changes subtly as one sections through this spiral organ, it is possible to obtain sections showing the entire length of the outer hair cells (with only parts of pillar and Deiters cell processes) or sections showing the entire length of the outer pillar and Deiters cells (with only parts of outer 150 Tannenbaum and Slepecky Fig. 2. Diagram showing the complex arrangement of the inner hair cells (IHC), outer hair cells (OHC), inner pillar cells (IP), outer pillar cells (OP), and Deiters cells (D) in the mature organ of Corti. OHC1 refers to a first row outer hair cell, and D1 refers to a Deiters cell that supports a first row outer hair cell. There are three rows of outer hair cells and three rows of Deiters cells. The microtubule bundles in each of the supporting cells is identified: the portion of the transcellular bundle in the apical surface process of the IP; the transcellular and basal bundles in the pillar process of the IP; the pillar bundle in the OP; the beam bundle in the OP; the basal bundle in the base of the D cells; the bundle in the phalangeal process, and the apical process of the D cells. hair cells), but never sections showing the entire length of these cells together. a-Tubulin One day after birth, staining with antibodies to a-tubulin (Fig. 3A-1) showed microtubules throughout all cells in the developing organ of Corti. There was diffuse labeling, which is attributed to labeling of individual microtubules, since soluble tubulin should have leached out of these preparations during processing (personal communication, Dr. John Tucker). In the developing inner and outer sensory hair cells, networks of microtubules were seen throughout the cell body, except in the region of the developing cuticular plate at the apical surface of these cells. Developing supporting cells were identified by the position of their nuclei close to the basilar membrane. While there was strong labeling of microtubules in all cells, microtubules in the pillar cells were arranged in parallel bundles rather than in loose networks seen in the sensory cells and the cells of Kölliker’s organ. Microtubules were most abundant (based on the increased intensity of label) in the top half of the inner pillar. Uniform labeling of microtubules was seen in the developing Deiters cells, with staining intensity similar to that seen in the sensory cells. Microtubules were present in the kinocilia on the apical surface of all cells in the developing organ of Corti. Seven days after birth, all antibodies to tubulin and tubulin isoforms used in this study gave weaker labeling of sensory cells at this stage of development than at 1 dab and 10 dab. The decreased intensity was consistently found on tissue sections from all specimens, even though they were fixed, processed. and stained at different times. Antibodies for a-tubulin (Fig. 3A-7) weakly labeled microtubules in most cells. The pillar cells appeared the most obviously changed at this time. The inner pillar cell was extending a process, so that the apical portion of the inner pillar was slightly bent over the outer pillar. A similar process was being extended by the outer pillar, to eventually interdigitate between the outer hair cells in the first row. The microtubule bundle in the inner pillar cell was more extensively labeled than at 1 dab, suggesting an increase in the number and length of the microtubules. While the inner pillar microtubules displayed prominent labeling along their length, the outer pillar cells showed only weak labeling. The basal regions of the inner and outer pillars were enlarging and the basal portions of both showed only diffuse labeling. However, microtubules in these regions and throughout the outer pillar bundle were parallel to each other, unlike the loose networks seen previously. Microtubules in the phalangeal processes of the Deiters cells were labeled, but networks of microtubules in the base of the Deiters cells could not be differentiated from the microtubules in the outer hair cells. Kinocilia could no longer be detected. At 10 dab, antibodies to a–tubulin (Fig. 3A-10) showed that the organ of Corti is approaching adult-like shape. Networks of microtubules remained in what was left of Kölliker’s organ and the tympanic covering layer. The inner and outer hair cells displayed discrete networks of microtubules. Strongly labeled microtubule bundles were found throughout the inner pillar cells (transcellular and basal bundles) and outer pillar cells (beam and pillar bundles). Distinct bundles of microtubules were observed in the phalangeal processes and basal portions of the Deiters cells. At 14 dab, labeling with antibodies to a-tubulin (Fig. 3A-14) showed that the organ of Corti was almost fully mature. Microtubules were present in very diffuse networks in what little remained of Kölliker’s organ and the tympanic covering layer. Hair cells contained networks of microtubules, diffuse throughout the central Modified Tubulin in Cochlear Epithelial Cells Fig. 3. Immunocytochemical localization of tubulin isoforms in the organ of Corti as seen in tissue sections taken from the basal turn of the cochlea at different ages (dab 5 days after birth). (A) Antibodies to a-tubulin; (B) antibodies to tyrosinated tubulin; (C) antibodies to detyrosinated tubulin; (D) antibodies to acetylated tubulin; (E) antibodies to polyglutamylated tubulin. Distribution of the different isoforms shifts as development proceeds on days 1, 7, 10, 14, and 55 after birth. Long arrow, inner pillar cell; short arrow, outer pillar cell; double arrows, Deiters cells; horizontal arrow, kinocilium; * 5 nerves. 151 152 Tannenbaum and Slepecky Fig. 3. (Continued.) Modified Tubulin in Cochlear Epithelial Cells 153 regions and dense just below the cuticular plates and near the synapses. Microtubules were prominent in both bundles in each of the pillar cells, displaying equal intensity labeling in both cells. Tyrosinated Tubulin At 1 dab, antibodies specific for tyrosinated tubulin (Fig. 3B-1) gave a labeling pattern that was almost similar to that seen with antibodies to a-tubulin in most cells of the organ of Corti. However, unlike the labeling pattern seen with antibodies to a-tubulin where the intensity was greater in the apical half of the cell, the labeling for tyrosinated tubulin was intense and uniform along the entire length of the inner pillar cell. At 7 dab, antibodies to tyrosinated tubulin (Fig. 3B-7) also displayed decreased labeling in the sensory cells and nerve fibers. In the base of the inner pillar and throughout the outer pillar, the microtubules labeling for tyrosinated tubulin appeared more in networks than in discrete bundles. The intensity of labeling for tyrosinated tubulin in the apical half of the inner pillar was much less than that seen in the basal half of the cell. Labeling was seen in the phalangeal processes of the Deiters cells near its apical surface. At 10 dab, antibodies to tyrosinated tubulin (Fig. 3B-10) showed diffuse labeling of microtubules in the remains of Kölliker’s organ and the tympanic covering layer, especially along the apical surfaces of the cells. The pillar cells, while larger, were not yet mature. This is based on the presence of tyrosinated tubulin in the basal region of the pillar cells, where microtubules were not yet organized into discrete bundles. Microtubules in Deiters cells could not be differentiated from those in outer hair cells. At 14 dab, antibodies for tyrosinated tubulin (Fig. 3B-14) weakly labeled the microtubules in the few cells remaining in Kölliker’s organ and the tympanic covering layer. Antibodies diffusely labeled the sensory cells and portions of the Deiters cells in the phalangeal processes and basal regions. While there was weak diffuse labeling of tyrosinated tubulin in the inner and outer pillar cells, tyrosinated tubulin was conspicuously lacking in the microtubules forming the bundles. Detyrosinated Tubulin Fig. 3. (Continued.) At 1 dab, antibodies to detyrosinated tubulin (Figs. 3C-1, 4A8-1, 4B8-1, 5A8-1, 5B8-1) weakly labeled all cells in the developing organ of Corti. In Kölliker’s organ, microtubules were present along the lateral edges of the columnar cells. In the developing sensory cells, detyrosinated tubulin weakly labeled microtubules throughout the hair cells and especially the region directly below the cuticular plate. In the developing pillar cells, detyrosinated tubulin was present in microtubules primarily at the 154 Tannenbaum and Slepecky Figs. 4 and 5. In the inner pillar cell (vertical arrow) at 1 dab, the appearance of acetylated tubulin is rapidly followed by the appearance of detyrosinated tubulin. In the Deiters cell (angled arrows), acetylated tubulin precedes detyrosinated tubulin by several days. Using double labeling techniques on individual sections, this pattern can be seen in the middle (Figs. 4A,A8, Figs. 5A,A8) and basal (Figs. 4B,B8; Figs. 5B, B8) turns of the cochlea. apical surface of the inner pillar. The region most intensely labeled was increased in length from that seen several hours after birth, and the microtubules appeared to be more tightly packed. This correlated with the decrease in labeling seen in the inner pillar cells with antibodies to tyrosinated tubulin. Labeling of microtubules in the outer pillar and Deiters cells was not greater than that seen in the outer hair cells at this time. At 7 dab, antibodies to detyrosinated tubulin (Fig. 3C-7) showed weak labeling along the apical surface of Modified Tubulin in Cochlear Epithelial Cells cells in Kölliker’s organ and little labeling of microtubules in the sensory hair cells. However, the microtubules in the inner pillar appeared more numerous than at 1 dab, based on the increased intensity of labeling. Microtubules were present both in the bundle along the apical surface and the bundle in the pillar process. Labeling was uniform along most of the inner pillar length, where little labeling for tyrosinated tubulin was found. Labeling for detyrosinated tubulin was strong but more patchy than that seen with antibodies to acetylated tubulin. The outer pillar showed weak and patchy labeling for detyrosinated tubulin, and the phalangeal processes of the Deiters cells were labeled. At 10 dab, antibodies to detyrosinated tubulin (Fig. 3C-10) labeled predominantly microtubule bundles in the pillar and Deiters cells. Labeling of the inner pillar cells was more intense than labeling of the outer pillar cells. However, labeling was more intense in the mid-portion of the inner pillar cell than it was in the apical process, while in the outer pillar cell it was more intense in the beam bundle than the pillar bundle. Deiters cell microtubule bundles were intensely labeled in the phalangeal processes, with some labeling of the basal bundles under the outer hair cells. At 14 dab, antibodies to detyrosinated tubulin (Fig. 3C-14) labeled predominantly microtubules in the pillar and Deiters cells. Labeling was equally intense and distributed uniformly along the entire length of the microtubules in the pillar processes in these cells; however, the beam bundle in the outer pillar was still more intensely labeled than the portion of the transcellular bundle at the apical surface of the inner pillar cells. Acetylated Tubulin At 1 dab, antibodies to acetylated tubulin (Figs. 3-D1, 4A-1, 4B-1, 5A-1, 5B-1) labeled only a limited number of microtubules present. Microtubules were seen in the columnar cells of Kölliker’s organ, where staining appeared bright only along the lateral margins. In the sensory cells, there was little if any labeling. There was strong labeling of the kinocilia on the apical surfaces of all cells. Bundles of microtubules containing acetylated tubulin appeared prominently in the upper third of the inner pillar cells; there was no acetylated tubulin in the lower portion of the inner pillar cells or in any region of the outer pillar. Although detyrosinated and acetylated tubulin appeared almost simultaneously in the inner pillar, close examination of double-labeled sections (Figs. 4, 5) showed that the extent of labeling of detyrosinated tubulin was always less than that for acetylated tubulin. Acetylated tubulin appeared along the length of the Deiters cells and in the processes interdigitated between the apical surfaces of the outer hair cells, at a time where 155 there was only weak, diffuse labeling for detyrosinated tubulin (Figs. 4, 5). At 7 dab, antibodies to acetylated tubulin (Fig. 3D-7) labeled a few microtubules along the apical surface of the remaining cells of Kölliker’s organ. Little if any labeling was seen in the sensory cells. In the inner pillar, labeling for acetylated tubulin was most intense in the transcellular bundle along the apical surface of the cell, and strong continuous labeling was found along most of the length of the pillar process. There was labeling in the apical region of the outer pillar, especially throughout the beam bundle interdigitating between the outer hair cells, but labeling in the pillar bundle was patchy and similar to that seen for detyrosinated tubulin. Labeling for acetylated tubulin was seen in the apical processes of the Deiters cells, as well as along the phalangeal processes. At 10 dab, antibodies to acetylated tubulin (Fig. 3D-10) continued to weakly label the apical surface of the few remaining cells in Kölliker’s organ, but showed no labeling of the cells in the tympanic covering layer. The inner and outer hair cells lacked acetylated microtubules. In sections where both inner and outer pillar cells were cut through the middle of the microtubule bundles, labeling for acetylated tubulin was equally intense in both the inner and outer pillar cells. It was also of equal intensity in the transcellular and basal bundles in the inner pillar and the beam and pillar bundles in the outer pillar cells, unlike the labeling seen for detyrosinated tubulin. Microtubules throughout the Deiters cells were also prominent, and labeling was more intense than that seen for detyrosinated tubulin. At 14 dab, antibodies to acetylated tubulin (Fig. 3D-14) labeled the organ of Corti with a pattern similar to that seen at 10 dab. There was weak labeling in the few cells remaining in Kölliker’s organ (which become the inner hair cell supporting cells) and no labeling in the few cells remaining in the tympanic covering layer. Sensory cells also lacked labeling for acetylated microtubules. In both the inner and outer pillar, labeling for acetylated tubulin was equally intense and distributed along the entire length of all the microtubule bundles. Labeling for acetylated tubulin was also intense in the basal bundle and phalangeal process bundle of the Deiters cells, but intensity of labeling for acetylated tubulin remained greater than that seen for antibodies to detyrosinated tubulin. Polyglutamylated Tubulin At 1 dab, antibodies specific for polyglutamylated tubulin (Fig. 3E-1) showed weak but diffuse labeling in most cells and kinocilia of the developing organ of Corti. Both pillar cells were diffusely labeled, with intense labeling in the upper quarter of the inner pillar. When single sections were double-labeled, even at this early 156 Tannenbaum and Slepecky stage labeling for polyglutamylated tubulin was less intensive along the length of the pillar cell than labeling for acetylated tubulin. At 7 dab, antibodies to polyglutamylated tubulin (Fig. 3E-7) showed no apparent labeling in the cells of Kölliker’s organ and the tympanic covering layer, although there was weak, diffuse labeling in the apical portion of the sensory cells. The antibodies intensely labeled microtubule bundles in the apical region of the inner pillars. No labeling was seen in the outer pillars and little labeling was seen in the Deiters cell processes. At 10 dab, antibodies to polyglutamylated tubulin (Fig. 3E-10) showed only weak labeling of microtubules in the remains of Kölliker’s organ and the tympanic covering layer. Weak diffuse labeling for polyglutamylated tubulin was seen in the networks of microtubules in the inner and outer hair cells, their supporting cells, and in nerve fibers. In the supporting cells, the appearance of polyglutamylated tubulin lagged behind acetylated tubulin; it was present only in the upper third of the inner pillar, upper quarter of the outer pillar, and small spots at the apical surfaces of the Deiters cells. At 14 dab, antibodies to polyglutamylated tubulin (Fig. 3E-14) weakly labeled the few remaining cells of Kölliker’s organ and the tympanic covering layer. There was still diffuse labeling of the sensory cells and supporting cells, and stronger labeling along the entire length of the microtubule bundles in the pillar and Deiters cells. While the labeling of the microtubule bundles in the supporting cells had attained its adult-like pattern, diffuse labeling was increased in all these cells when compared to that seen in the adult at 55 dab (Fig. 3E-55). DISCUSSION Summary of Results Figure 6 presents a diagram summarizing our findings. A few hours after birth, the organ of Corti appears as a mass of tightly packed, undifferentiated cells in a small region of the basal turn of the cochlea. Microtubules, arranged in networks parallel to the long axis of the cells, are present in Kölliker’s organ, the tympanic covering layer, and all sensory and supporting cells. Our results differ from those seen by other investigators using transmission electron microscopy [Ito et al., 1995] in that we see many more microtubules in the sensory and supporting cells at this early stage of development. However, at the later stages our results confirm the sequence of microtubule assembly and bundle formation seen in the gerbil [Ito et al., 1995] and mouse [Tucker et al., 1992, 1995; Henderson et al., 1994; Henderson et al., 1995]. While the adult organ of Corti contains sensory and supporting cells which display cell-type specific localization of modified and unmodified tubulins, this is not the Fig. 6. Diagram summarizing the temporal and spatial pattern of the appearance of tyrosinated, acetylated, and detyrosinated tubulin during development of the organ of Corti. case early in development. Several hours after birth, labeling with antibodies to tyrosinated tubulin displays a pattern similar to that for antibodies to a-tubulin; strong labeling of all cells with increased intensity of labeling in the apical third of the inner pillar cell. Labeling for posttranslationally modified tubulin is restricted to small areas along the apical surface of all cells, and a larger region near the apical surface of the inner pillar cell. One explanation for the few microtubules seen by transmission electron microscopy at this time might be that the more dynamic microtubules, composed of tyrosinated tubulin, are not preserved during processing. Some of these steps (including incubation with cold solutions) could induce dynamic microtubules to depolymerize. As the microtubules are stabilized, reflected in the increased labeling for posttranslational modifications to tubulin, the microtubules remain and the results seen at the ultrastructural level are more similar to the immunofluorescent Modified Tubulin in Cochlear Epithelial Cells results presented here. Another is that tissue sections for light microscopy are thicker than those used for electron microscopy, so that microtubules are easier to detect. At 7 dab, we consistently see a general decrease in the labeling for tubulin (a-tubulin as well as tyrosinated and posttranslationally modified tubulins). A similar decrease in intensity of labeling for tubulin has been seen at a comparable stage of development in isolated outer hair cells in the rat [Vago et al., 1996]. This decrease might reflect depolymerization of microtubules, and the decrease in the number or length of microtubules could destabilize the rigid cytoskeleton and allow the remodeling necessary for cells to attain their mature location, position, shape, and size. In supporting cells, the depolymerization could be associated with the reorganization of microtubules into specific bundles; there is evidence for the loss of almost 2,000 microtubules in the transcellular bundle of the inner pillar cell prior to the formation of the basal bundle [Tucker et al., 1993]. In sensory cells, depolymerization could be associated with the release of microtubule-associated proteins; these could become available for interaction with actin and thus provide a mechanism by which microtubules could help align actin filaments in the cuticular plate [Sobkowicz et al., 1995]. However, the decreased labeling could instead reflect alterations to, rather than depolymerization of, microtubules; interactions of microtubules with proteins which may block antigenic sites would prevent antibody binding. Alternatively, microtubules may remain unaltered but cells increase in size so rapidly that microtubule formation cannot keep up. This would result in a temporary decrease in microtubule density, which recovers as synthesis and polymerization continue. Tyrosinated Tubulin in Cells That Decrease in Number or That Will be Involved With Active Processes Such as Secretion and Motility In the developing organ of Corti, Kölliker’s organ and the tympanic covering layer are initially large, with numerous cells that play a role in the synthesis and secretion of the tectorial and basilar membranes, respectively. While they initially contain microtubules composed of tyrosinated, detyrosinated, acetylated, and polyglutamylated tubulin, at later stages they are enriched primarily with tyrosinated tubulin. Concomitant with the decrease in labeling for microtubules containing modified isoforms of tubulin, the cells diminish in size and number; by 14 dab, Kölliker’s organ and the tympanic covering layer have almost completely disappeared. The presence of posttranslationally modified tubulin in these cells during the early stages of development might provide the initial mechanism for stabilizing shape as the cells in the organ of Corti differentiate. In the later stages, when the levels of the posttranslational modifica- 157 tions change, the continued presence of predominantly tyrosinated tubulin in Kölliker’s organ and the tympanic covering layer might be related to secretion, based on the role of microtubules in organelle movement in other cell types [Araki et al., 1993; Ayscough et al., 1993]. It may also be related to the decrease in cell size and number in both structures, since microtubules containing tyrosinated tubulin play important roles in cells undergoing rapid changes in morphology [Robson and Burgoyne, 1989; Alfa and Hyams, 1991; Pagh-Roehl et al., 1991]. However, the continued presence of tyrosinated tubulin cannot be totally responsible for the dramatic change in the cells of these accessory structures. The sensory hair cells also lack detyrosinated, acetylated, and polyglutamylated tubulin in the developing and adult organ of Corti [Slepecky et al., 1995], yet sensory cells do not decrease in size or number. Posttranslational Modifications to Tubulin Do Not Correlate With the Development of Cell Polarity and Shape From study of cells in culture, it has been suggested that the appearance of posttranslationally modified tubulin correlates with stabilization of microtubules. Selective stabilization would provide the basis for asymmetry, resulting in the establishment of cell polarity and shape during morphogenesis [Kirschner and Mitchison, 1986]. Our study of posttranslational modifications to tubulin in the developing organ of Corti does not support this mechanism. At 1 dab, sensory and most supporting cells contain microtubules composed of predominantly tyrosinated tubulin, with labeling for posttranslationally modified tubulins found only in restricted regions near the apical surface. The outer pillar cells lack detectable microtubules at this stage. Yet all these cells are already asymmetric and elongated. One interpretation of our results is that microtubules are stabilized prior to the onset of posttranslational modifications, supporting observations by others that the appearance of posttranslational modifications to tubulin are neither associated with nor required for the onset of microtubule stability [Gundersen and Bulinski, 1986; Schulze et al., 1987; Khawaja et al., 1988; Pepperkok et al., 1990; Xiang and MacRae, 1995]. Another is that the establishment of cell polarity does not require the presence of stable microtubules, or even any microtubules at all [Alfa and Hyams, 1991; Bre et al., 1991; Euteneuer and Schliwa, 1992]. Posttranslational Modifications to Tubulin in Cells That Provide Mechanical Support to the Organ of Corti While our results do not suggest a correlation between the appearance of posttranslationally modified microtubules and the establishment of cell polarity, they 158 Tannenbaum and Slepecky do suggest that the appearance of posttranslationally modified microtubules is correlated with the elongation of supporting cell microtubules, seen first in the inner pillar and Deiters cells and then in the outer pillar cells [Tucker et al., 1992, 1995; Ito et al., 1995] Over time, the posttranslational modifications to tubulin may provide sites for binding of specific microtubule-associated proteins that further stabilize the cell and cytoskeleton. Actin and intermediate filaments have been shown to associate with stable microtubules [Griffith and Pollard, 1978; Gurland and Gundersen, 1995] and both actin and intermediate filaments are prominent components of the pillar cells, associated with microtubules in the mature cochlea [Slepecky and Chamberlain, 1983; Raphael et al., 1987; Schulte and Adams, 1989]. While nothing is known about the appearance of intermediate filaments during development of the gerbil cochlea, actin is recruited into the microtubule bundles of the pillar and Deiters cells at a stage late in development [Kuhn and Vater, 1996] when we find detyrosination and acetylation of tubulin have already occurred. The abundance of all three of these cytoskeletal elements, as well as the highly ordered organization of the actin filaments with microtubules, suggests they play a role in providing mechanical support to the organ of Corti, required for proper amplitude and frequency of vibration in response to sound stimulus. Pillar cells. In the pillar cell microtubules, a decrease in tyrosinated tubulin labeling is paralleled by an almost simultaneous increase in acetylated and detyrosinated tubulin. The timing and pattern of appearance of these posttranslational modifications correlates with an increase in microtubule number and length. This sequence is commonly found in developing cells [Thompson, 1982; Cummings et al., 1984; Alonso et al., 1988; Beltramo et al., 1989; Gundersen et al., 1989; Bre et al., 1991; Bulinski and Gundersen, 1991] and is similar to the pattern seen in fibroblasts, following cell division or recovery from drug treatment [Bulinski et al., 1988]. The difference between the intensity of labeling in the inner and outer pillar cells at each of the early stages studied can be explained if the developmental changes in the number and location of gerbil pillar cell microtubules are similar to those seen in the mouse [Tucker et al., 1992; Henderson et al., 1995; Mogensen et al., 1997]. Ultrastructural analysis has shown that microtubule nucleation occurs first in the inner pillar cell, near the centrioles at the apical surface. Pericentriolar material migrates to the periphery of the cell and the newly formed microtubules also migrate and are anchored there. Microtubules begin to elongate toward the base of the columnar cell, and additional microtubules are added to the forming bundle. As development proceeds, the cell increases in size and changes in shape. The pericentriolar nucleating and anchoring site migrates laterally, forming a thin apical surface process, filled with microtubules, which extends over the outer pillar cell (see Fig. 2). The microtubules continue to elongate at their free ends, bend toward the base of the cell, and elongate further so that many long microtubules span the distance between the apical surface and the base of the cell, forming the transcellular bundle. At the same time that this bundle is being formed, a large number of shorter microtubules (2,000 out of the 3,000 present in each inner pillar cell) are released from their apical anchoring site and migrate to the base of the cell, where their ends are captured at a basal anchoring site. At this location, the microtubules elongate toward the midregion of the cell to form the basal bundle. Our labeling of microtubules with antibodies to tubulin during development reflects this pattern. The inner pillar microtubules accumulate first in large numbers at the apical surface. As these microtubules elongate toward the base of the cell, they contribute to the transcellular bundle. The inner pillar labels first in the apical third of the cell with antibodies to a-tubulin and tyrosinated tubulin. As the microtubules in the apical surface bundle increase in number and length, posttranslationally modified isoforms appear. The addition and elongation of microtubules is correlated with the appearance of acetylated and detyrosinated tubulin, with labeling increasing in width, length, and intensity. Early in development, labeling for posttranslationally modified isoforms of tubulin shows that microtubules in the transcellular bundle span the distance between the apical and basal surfaces of the cell. The appearance of detyrosinated, acetylated, or polyglutamylated tubulin is not directly responsible for the bending of the transcellular bundle because: 1) the posttranslational modifications precede bending, and 2) a similar pattern of posttranslational modifications is observed at the apical surface of outer pillar cells where the microtubule bundles do not bend, but remain short and parallel to the apical surface of the cell [Tucker et al., 1995]. At the early stages of development, the outer pillar cell lacks detectable microtubules even when large numbers of long and posttranslationally modified microtubules are present in the inner pillar cell. Only at several days after birth are outer pillar cell microtubules observed, nucleated and anchored near the centriole at the apical surface of the columnar cell. As development proceeds, the pericentriolar material divides in two and each portion moves laterally, one toward the inner pillar cells and one toward the outer hair cells. As the outer pillar cell apical process extends to interdigitate between the first row of outer hair cells, the nucleating site and anchoring site (centrioles and pericentriolar material) moves to the lateral-most edge of this process (see Fig. 2) near the outer hair cell [Tucker et al., 1995]. Unlike microtubules in the apical process of the inner pillar cell, Modified Tubulin in Cochlear Epithelial Cells the microtubules nucleated and anchored in the apical processes of the outer pillar cell do not continue to elongate and bend. Most remain relatively short (around 10 µm in the mouse) and run parallel to the apical surface of the outer pillar cell, forming the beam bundle. This is the first region of the outer pillar cell where we found intense labeling of posttranslationally modified microtubules. Other short microtubules are released from the apical anchoring site and migrate down toward the basal surface of the outer pillar cell, where their ends are captured. They then elongate upward to form the pillar bundle, which becomes anchored at the pericentriolar material that was established at the apical surface of the outer pillar, near the inner pillar cell, earlier in development. Our immunocytochemical results confirm this sequence. Since most of the microtubules that eventually comprise the pillar bundle in the outer pillar cell do not accumulate at the apical surface of the cell or elongate as a group, labeling with antibodies to a-tubulin is initially weak and remains weak in this region. Moreover, since all the microtubules at the apical surface are newly formed, most tubulin in the microtubules is tyrosinated. While the appearance of acetylated tubulin is rapidly followed by detyrosinated tubulin in the outer pillar cell, as it is in the inner pillar cell, our results suggest that posttranslational modifications occur as the microtubules are migrating, as reflected in the relatively weak and patchy labeling for acetylated and detyrosinated tubulin seen along the length of the pillar bundle even up to 10 dab. Deiters cells. Labeling for acetylated tubulin allows us to detect the Deiters cells between the developing outer hair cells even at 1 dab, and indicates that microtubules are present in these cells several days before they can be detected by transmission microscopy [Ito et al., 1995]. Unlike the sequence of appearance of the posttranslational modifications in the pillar cells, acetylated tubulin precedes detyrosinated tubulin by several days. A similar sequence has been observed in developing Drosophila embryos, where acetylated tubulin appears first and detyrosinated tubulin appears at a later stage [Warn et al., 1990]. It should be noted that when there is a difference between the timing of appearance of these tubulin isoforms, acetylation does not always precede detyrosination. In developing myoblasts, detyrosinated tubulin is detected first in elongating portions of the cell; acetylated tubulin increases occur much later [Gundersen et al., 1989]. SUMMARY Our results show that during development: 159 1) Tyrosinated tubulin is initially the predominant form of tubulin in all cells. 2) Posttranslational modifications occur in the supporting cells in a time-dependent manner as development proceeds; progression from cell apex to base is in the same direction as microtubule elongation. 3) Establishment of cell polarity appears to precede posttranslational modifications to tubulin, especially as shown in the development of the outer pillar cells. 4) Tyrosinated tubulin remains the predominant form of tubulin in cells that decrease in number and that retain active processes such as secretion and motility. 5) The appearance of acetylated tubulin is rapidly followed by the appearance of detyrosinated tubulin in the pillar cells, but the appearance of acetylated tubulin precedes the appearance of detyrosinated tubulin in the Deiters cells by several days. 6) The cell-type specific pattern of posttranslational modifications appears to be established at the time of onset of cochlear function (12 dab); however, the intensity of labeling for polyglutamylated tubulin does not appear adult-like until cochlear function is also mature (30 dab). In the cochlea, sensory and supporting cells play important but differing functional roles. In ears from adult animals, microtubules in the inner and outer sensory hair cells are arranged in loose networks found throughout the cell, while those in the supporting cells are arranged almost exclusively in tightly packed bundles which run the length of each cell [Angelborg and Engstrom, 1972; Steyger et al., 1989; Furness et al., 1990; Slepecky and Ulfendahl, 1992; Raphael et al., 1994; Kuhn and Vater, 1995]. In outer hair cells, microtubules may be involved in remodeling of the cytoskeleton following shape changes related to outer hair cell motility. In both inner and outer hair cells, microtubules are found in areas where organelles are being transported; in the apex of the cell adjacent to the cuticular plate and in the base of the cell near the synaptic vesicles. Thus, microtubules in sensory cells are thought to be dynamic and involved in active processes. The pillar cells possess microtubules in bundles, larger than any others described so far for mammalian cells [Tucker et al., 1992]. It has been assumed that these microtubules provide rigidity and mechanical support to the organ of Corti. Microtubules in supporting cells of the cochlea also differ biochemically and structurally from those found in the sensory cells. They are enriched with microtubules containing posttranslationally modified isoforms of tubulin, suggesting that they are stable and long-lived; micro- 160 Tannenbaum and Slepecky tubules in the sensory cells are predominantly tyrosinated, suggesting that they are dynamic and short-lived [Slepecky et al., 1995]. The supporting cell microtubules differ from those in the sensory cells in that they are composed of 15 protofilaments [Saito and Hama, 1982; Kikuchi et al., 1991a], instead of the more commonly found 13 protofilaments. It is interesting that supporting cells in the sensory epithelia of the vestibular system contain microtubules with 13 protofilaments and both tyrosinated and detyrosinated tubulin [Kikuchi et al., 1991b; Ogata and Slepecky, 1995]. Since the organ of Corti vibrates during stimulation while the vestibular sensory epithelium is stationary, the differences in the microtubules may be correlated with the mechanical properties of the supporting cells. One can further speculate that alterations in the pattern and location of posttranslationally modified tubulin isoforms might change the mechanics of this vibratory organ and subsequently alter the frequency selectivity and sensitivity of this sensory organ. ACKNOWLEDGMENTS We thank Molly Watts for help with the diagrams based on the tissue sections, Tammy Stage for help with photography, and the Deafness Research Foundation for funding. REFERENCES Alfa, C.E., and Hyams, J.S. (1991): Microtubules in the fission yeast Schizosaccharomyces pombe contain only the tyrosinated form of alpha-tubulin. Cell Motil. Cytoskeleton 18:86–93. Alonso, A.d.C., Arce, C.A., and Barra, H.S. (1988): Relationship between the tyrosination state of tubulin and the activities of tubulin: Tyrosine ligase and tubulin carboxypeptidase in rat muscle during development. Eur. J. Biochem. 177:517–522. Alonso, A.d.C., Arce, C.A., and Barra, H.S. (1993): Tyrosinatable and nontyrosinatable tubulin subpopulations in rat muscle in comparison to those in brain. Biochim. Biophys. Acta 1163:26–30. Angelborg, C., and Engstrom, H. (1972): Supporting elements in the organ of Corti. Acta Otolaryngol. Suppl. 301:49–60. Araki, N., Ohono, J., Lee, T., Takashima, Y., and Ogawa, K. (1993): Nematolysosomes (elongated lysosomes) in rat hepatocytes: Their distribution, microtubule dependence, and role in endocytic transport pathways. Exp. Cell Res. 204:181–191. Arce, C.A., Hallak, M.E., Rodriguez, J.A., Barra, H.S., and Caputto, R. (1978): Capability of tubulin and microtubules to incorporate and to release tyrosine and phenylalanine and the effect of those amino acids on tubulin assembly. J. Neurochem. 31:205–210. Arjmand, E., Harris, D., and Dallos, P. (1988): Developmental changes in frequency mapping of the gerbil cochlea: Comparison of two cochlear locations. Hearing Res. 32:93–96. Audebert, S., Desbruyeres, E., Gruszczynski, C., Koulakoff, A., Gros, F., Denoulet, P., and Eddé, B. (1993): Reversible polyglutamylation of a- and b- tubulin and microtubule dynamics in mouse brain neurons. Mol. Biol. Cell 4:615–626. Ayscough, K., Hajibagheri, N.M.A., Watson, R., and Warren, G. (1993): Stacking of golgi cisternae in Schizosaccharomyces pombe requires intact microtubules. J. Cell Sci. 106:1227–1237. Baas, P.W., and Black, M.M. (1990): Individual microtubules in the axon consist of domains that differ in both composition and stability. J. Cell Biol. 111:495–509. Baker, J., Theurkauf, W.E., and Schubiger, G. (1993): Dynamic changes in microtubule configuration correlate with nuclear migration in the preblastoderm Drosophila embryo. J. Cell Biol. 122:113–121. Beltramo, D.M., Arce, C.A., and Barra, H. (1989): Tyrosinationdetyrosination of tubulin and microtubules during the development of chick erythrocytes. Mol. Cell. Biochem. 89:47–56. Bre, M.-H., Pepperkok, R., Kreis, T.E., and Karsenti, E. (1991): Cellular interactions and tubulin detyrosination in fibroblastic and epithelial cells. Biol. Cell 71:149–160. Bulinski, J.C., and Gundersen, G.G. (1991): Stabilization and posttranslational modification of microtubules during cellular morphogenesis. BioEssays 13:285–293. Bulinski, J.C., Richards, J.E., and Piperno, G. (1988): Posttranslational modification of alpha-tubulin: Detyrosination and acetylation differentiate populations of interphase microtubules in cultured cells. J. Cell Biol. 106:1213–1220. Cummings, R., Burgoyne, R.D., and Lytton, N.A. (1984): Immunocytochemical demonstration of alpha-tubulin modification during axonal maturation in the cerebellar cortex. J. Cell Biol. 98:347– 351. Deanin, G.G., Preston, S.F., and Gordon, M.W. (1981): Carboxyl terminal tyrosine metabolism of alpha-tubulin and changes in cell shape: Chinese hamster ovary cells. Biochem. Biophys. Res. Commun. 100:1642–1650. Echteler, S.M., Arjmand, E., and Dallos, P. (1989): Developmental alterations in the frequency map of the mammalian cochlea. Nature 341:147–149. Eddé, B., Rossier, J., LeCaer, J., Desbruyeres, E., Gros, F., and Denoulet, P. (1990): Posttranslational glutamylation of a-tubulin. Science 247:83–85. Euteneuer, U., and Schliwa, M. (1992): Mechanism of centrosome positioning during the wound response in BSC-1 cells. J. Cell Biol. 116:1157–1166. Finck, A., Schneck, C.S., and Hartman, A.F. (1972): Development of cochlear function in the neonate Mongolian gerbil (Meriones unguiculatus). J. Comp. Physiol. Psych. 78:375–380. Furness, D.N., Hackney, C.M., and Steyger, P.S. (1990): Organization of microtubules in cochlear hair cells. J. Electron Microsc. Tech. 15:261–279. Glasgow, J.E., and Daniele, R.P. (1994): Role of microtubules in random cell migration: Stabilization of cell polarity. Cell Motil. Cytoskeleton 27:88–96. Griffith, L.M., and Pollard, T.D. (1978): Evidence for actin filamentmicrotubule interaction mediated by microtubule-associated proteins. J. Cell Biol. 86:958–965. Gundersen, G.G., and Bulinski, J.C. (1986): Microtubule arrays in differentiated cells contain elevated levels of a post-translationally modified form of tubulin. Eur. J. Cell Biol. 42:288–294. Gundersen, G.G., Kalnoski, M.H., and Bulinski, J.C. (1984): Distinct populations of microtubules: Tyrosinated and non-tyrosinated alpha-tubulin are distributed differently in vivo. Cell 78:779– 789. Gundersen, G.G., Khawaja, S., and Bulinski, J.C. (1987): Postpolymerization detyrosination of alpha-tubulin. A mechanism for subcellular differentiation of microtubules. J. Cell Biol. 105:251–264. Gundersen, G.G., Khawaja, S., and Bulinski, J.C. (1989): Generation of a stable, posttranslationally modified microtubule array is an early event in myogenic differentiation. J. Cell Biol, 109:2275– 2288. Modified Tubulin in Cochlear Epithelial Cells Gurland, G., and Gundersen, G.G. (1995): Stable, detyrosinated microtubules function to localize vimentin intermediate filaments in fibroblasts. J. Cell Biol. 131:1275–1290. Harris, D.M., and Dallos, P. (1984): Ontogenetic changes in frequency mapping of a mammalian ear. Science 225:741–743. Harris, D.M., Rotche, R., and Freedom, T. (1990): Postnatal growth of the cochlear spiral in Mongolian gerbil. Hear. Res. 50:1–6. Henderson, C.G., Tucker, J.B., Chaplin, M.A., Mackie, J.B., Maidment, S.N., Mogensen, M.M., and Paton, C.C. (1994): Reorganization of the centrosome and associated microtubules during the morphogenesis of a mouse cochlear epithelial cell. J. Cell Sci. 107:589–600. Henderson, C.G., Tucker, J.B., Mogensen, M.M., Mackie, J.B., Chaplin, M.A., Slepecky, N.B., and Leckie, L.M. (1995): Three microtubule-organizing centres collaborate in a mouse cochlear epithelial cell during supracellularly coordinated control of microtubule positioning. J. Cell Sci. 108:37–50. Hyde, G.J., and Hardham, A.R. (1993): Microtubules regulate the generation of polarity in zoospores of Phytophthora cinnamomi. Eur. J. Cell Biol. 62:75–85. Ito, M., Spicer, S.S., and Schulte, B.A. (1995): Cytological changes related to maturation of the organ of Corti and opening of Corti’s tunnel. Hear. Res. 88:107–123. Joshi, H.C., and Baas, P.W. (1993): A new perspective on microtubules and axon growth. J. Cell Biol. 121:1191–1196. Khawaja, S., Gundersen, G.G., and Bulinski, J.C. (1988): Enhanced stability of microtubules enriched in detyrosinated tubulin is not a direct function of detyrosination level. J. Cell Biol. 106:141– 149. Kikuchi, T., Takasaka, T., Tonosaki, A., Katori, Y., and Shinkawa, H. (1991a): Microtubules of guinea pig cochlear epithelial cells. Acta Otolaryngol. (Stockh.) 111:286–290. Kikuchi, T., Takasaka, T., Tonosaki, A., Watanabe, H., Hozawa, K., Shinkawa, H., and Wada, H. (1991b): Microtubule subunits of guinea pig vestibular epithelial cells. Acta Otolaryngol. Suppl. 481:107–111. Kirschner, M., and Mitchison, T. (1986): Beyond self-assembly: From microtubules to morphogenesis. Cell 45:329–342. Kuhn, B., and Vater, M. (1995): The arrangements of F-actin, tubulin and fodrin in the organ of Corti of the horseshoe bat (Rhinolophus rouxi) and the gerbil (Meriones unguiculatus). Hear. Res. 84:139–156. Kuhn, B., and Vater, M. (1996): The early postnatal development of F-actin patterns in the organ of Corti of the gerbil (Meriones unguiculatus) and the horseshoe bat (Rhinolophus rouxi). Hear. Res. 99:47–70. Kumar, N., and Flavin, M. (1981): Preferential action of a brain detyrosinating carboxypepdidase on polymerized tubulin. J. Biol. Chem. 256:7678–7686. L’Hernault, S.W., and Rosenbaum, J.L. (1983): Chlamydomonas alpha-tubulin is post-translationally modified in the flagella during flagellar assembly. J. Cell Biol. 97:258–263. L’Hernault, S.W., and Rosenbaum, J.L. (1985): Chlamydomonas alpha-tubulin is post-translationally modified by acetylation on the epsilon-amino group of a lysine. Biochemistry 24:473–478. Lim, D.J., and Rueda, J. (1992): Structural development of the cochlea. In Romand, R. (ed.): ‘‘Development of Auditory and Vestibular Systems.’’ Amsterdam: Elsevier, pp. 33–58. Mogensen, M.M., Mackie, J.B., Doxsey, S.J., Stearns, T., and Tucker, J.B. (1997): Centrosomal deployment of g-tubulin and pericentrin: Evidence for a microtubule nucleating domain and a minus end docking domain in certain mouse epithelial cells. Cell Motil. Cytoskeleton 36:276–290. Munyer, P.D., and Schulte, B.A. (1995): Developmental expression of 161 proteoglycans in the tectorial and basilar membrane of the gerbil cochlea. Hear. Res. 85:85–94. Ogata, Y., and Slepecky, N.B. (1995): Immunocytochemical comparison of posttranslationally modified forms of tubulin in the vestibular end-organs of the gerbil: Tyrosinated, acetylated and polyglutamylated tubulin. Hear. Res. 86:125–131. Pagh-Roehl, K., Wang, E., and Burnside, B. (1991): Posttranslational modifications of tubulin in teleost photoreceptor cytoskeletons. Cell. Mol. Neurobiol. 11:593–610. Paturle-Lafanechere, L., Manier, M., Trigault, N., Pirollet, F., Mazarguil, H., and Job, D. (1994): Accumulation of delta 2-tubulin, a major tubulin variant that cannot be tyrosinated, in neuronal tissues and stable microtubule assemblies. J. Cell Sci. 107:1529– 1543. Pepperkok, R., Bre, M.H., Davoust, J., and Kries, T.E. (1990): Microtubules are stabilized in confluent epithelial cells but not in fibroblasts. J. Cell Biol. 111:3003–3012. Piperno, G., LeDizet, M., and Chang, X. (1987): Microtubules containing acetylated a-tubulin in mammalian cells in culture. J. Cell Biol. 104:289–302. Pujol, R., Lavigne-Rebillard, M., and Lenoir, M. (1996): Development of sensory and neural structures in the mammalian cochlea. In Rubel, E.W., Popper, A.N., and Fay, R.R. (eds.): ‘‘Development of the Auditory System. Handbook of Auditory Research.’’ New York: Springer, vol. 5. Raphael, Y., Marshak, G.H., Barash, A., and Geiger, B. (1987): Modulation of intermediate-filament expression in developing cochlear epithelium. Differentiation. 35:151–162. Raphael, Y., Athey, B.D., Wang, Y., Lee, M.K., and Altschuler, R.A. (1994): F-actin, tubulin and spectrin in the organ of Corti: Comparative distribution in different cell types and mammalian species. Hear. Res. 76:173–187. Robson, S.J., and Burgoyne, R.D. (1989): Differential localisation of tyrosinated, detyrosinated, and acetylated alpha-tubulins in neurites and growth cones of dorsal root ganglion cells. Cell Motil. Cytoskeleton 12:273–282. Romand, R. (1987): Tonotopic evolution during development. Hear. Res. 28:117–123. Roth, B., and Bruns, V. (1992a): Postnatal development of the rat organ of Corti. I. General morphology, basilar membrane, tectorial membrane and border cells. Anat. Embryol. 185:559–569. Roth, B., and Bruns, V. (1992b): Postnatal development of the rat organ of Corti. II. Hair cell receptors and their supporting elements. Anat. Embryol. 185:571–581. Saito, K., and Hama, K. (1982): Structural diversity of microtubules in the supporting cells of the sensory epithelium of guinea pig organ of Corti. J. Electron Microsc. 311:278–281. Schulte, B.A., and Adams, J.C. (1989): Immunohistochemical localization of vimentin in the gerbil inner ear. J. Histochem. Cytochem. 37:1787–1797. Schulze, E., and Kirschner, M. (1987): Dynamic and stable microtubule populations in cells. J. Cell Biol. 104:277–290. Schulze, E., Asai, D.J., Bulinski, J.C., and Kirshner, M. (1987): Posttranslational modification and microtubule stability. J. Cell Biol. 105:2167–2177. Schweitzer, L., Lutz, C., Hobbs, M., and Weaver, S.P. (1996): Anatomical correlates of the passive properties underlying the developmental shift in the frequency map of the mammalian cochlea. Hear. Res. 97:84–94. Slepecky, N.B., and Chamberlain, S.C. (1983): Distribution and polarity of actin in inner ear supporting cells. Hear. Res. 10:359–370. 162 Tannenbaum and Slepecky Slepecky, N.B., and Chamberlain, S.C. (1985): Immunoelectronmicroscopic and immunofluorescent localization of cytoskeletal and muscle-like contractile proteins in inner ear sensory hair cells. Hear. Res. 20:245–260. Slepecky, N.B., and Ulfendahl, M. (1992): Actin-binding and microtubule associated proteins in the organ of Corti. Hear. Res. 57:201–215. Slepecky, N.B., Henderson, C.G., and Saha, C. (1995): Posttranslational modifications of tubulin in the cochlea suggest microtubules in hair cells are dynamic while microtubules in supporting cells are stable. Hear. Res. 91:136–147. Sobkowicz, H.M., Slapnick, S.M., and August, B.K. (1995): The kinocilium of auditory hair cells and evidence for its morphogenetic role during the regeneration of stereocilia and cuticular plates. J. Neurocytol. 24:633–653. Souter, M., Nevill, G., and Forge, A. (1995): Postnatal development of membrane specialisations of gerbil outer hair cells. Hear. Res. 91:43–62. Steyger, P.S., Furness, D.N., Hackney, C.M., and Richardson, G.P. (1989): Tubulin and microtubules in cochlear hair cells: Comparative immunocytochemistry and ultrastructure. Hear. Res. 42:1–16. Thompson, W.C. (1982): The Cyclic Tyrosination/Detyrosination of Alpha Tubulin. In: L. Wilson (ed.) ‘‘Methods in Cell Biology.’’ Chapter 15, pp. 235–255. Thompson, W.C., Asai, D.J., and Carney, D.H. (1984): Heterogeneity among microtubules of the cytoplasmic microtubule complex detected by a monoclonal antibody to alpha-tubulin. J. Cell Biol. 98:1017–1025. Tucker, J.B., Paton, C.C., Richardson, G.P., Mogensen, M.M., and Russell, I.J. (1992): A cell surface-associated centrosomal layer of microtubule-organizing material in the inner pillar cell of the mouse cochlea. J. Cell Sci. 102:215–226. Tucker, J.B., Paton, C.C., Henderson, C.G., and Mogensen, M.M. (1993): Microtubule rearrangement and bending during assembly of large curved microtubule bundles in mouse cochlear epithelial cells. Cell Motil. Cytoskeleton 25:49–58. Tucker, J.B., Mogensen, M.M., Paton, C.C., Mackie, J.B., and Henderson, G. (1995): Formation of two microtubule-nucleating sites which perform differently during centrosomal reorganization in mouse cochlear epithelial cells. J. Cell Sci. 108:1333–1345. Vago, P., Ripoll, C., Tournebize, R., and Lenoir, M. (1996): Distribution of actin and tubulin in outer hair cells isolated from developing rat cochlea: A quantitative study. Eur. J. Cell Biol. 69:308–315. Warn, R.M., Harrison, A., Planques, V., Robert-Nicoud, N., and Wehland, J. (1990): Distribution of microtubules containing posttranslationally modified alpha-tubulin during Drosophila embryogenesis. Cell Motil. Cytoskeleton 17:34–45. Wolff, A., deNechaud, B., Chillet, D., Mazarguil, H., Desbruyeres, E., Audebert, S., Edde, B., Gros, F., and Denoulet, P. (1992): Distribution of glutamylated a and b-tubulin in mouse tissues using a specific monoclonal antibody, GT335. Eur. J. Cell Biol. 59:425–432. Woolf, N.K., Koehrn, F.J., and Ryan, A.F. (1992): Immunohistochemical localization of fibronectin-like protein in the inner ear of the developing gerbil and rat. Dev. Brain Res. 65:21–33. Xiang, H., and MacRae, T.H. (1995): Production and utilization of detyrosinated tubulin in developing Artemia larvae: Evidence for a tubulin-reactive carboxypeptidase. Biochem. Cell Biol. 73:673–685.