MICROSCOPY RESEARCH AND TECHNIQUE 41:270–283 (1998) Immunohistochemical Evaluation of Cholinergic Neurons in the Rat Superior Olivary Complex WEIPING YAO AND DONALD A. GODFREY Department of Otolaryngology, Medical College of Ohio, Toledo, Ohio 43699-0008 KEY WORDS: acetylcholinesterase; choline acetyltransferase; vesicular acetylcholine transporter ABSTRACT The cholinergic system in the rat superior olivary complex (SOC) was evaluated by immunohistochemistry for choline acetyltransferase (ChAT) and vesicular acetylcholine transporter (VAChT) and histochemistry for acetylcholinesterase (AChE). ChAT-positive somata were found mostly in the lateral superior olive (LSO) and ventral nucleus of the trapezoid body (VNTB). In the LSO, there were both rostral-caudal and medial-lateral gradients in concentration of ChAT-positive somata; the highest concentration was in the middle of the rostral-caudal extent and the most medial part. The estimated total number of ChAT-positive neurons in the LSO was similar to previous estimates of the total number of lateral olivocochlear neurons. Two groups of ChAT-positive somata were found in the VNTB: a dorsolateral group of larger, multipolar, and more darkly labeled neurons and a ventromedial group of smaller, oval, and more lightly labeled neurons, which was about 5 times as numerous. There was a caudal-to-rostral increase in number of neurons in each group. VAChT immunoreactivity, predominantly localized in puncta, was seen in LSO, VNTB, and LNTB, and, to a lesser extent, in other parts of the SOC. VAChT-positive somata were also found in the VNTB and medial LSO. This distribution pattern of VAChT was generally similar to that of ChAT. AChE labeling had a similar appearance to ChAT labeling in the VNTB but differed in the LSO, where AChE labeling was lighter and associated more with neuropil than with somata. Microsc. Res. Tech. 41:270–283, 1998. r 1998 Wiley-Liss, Inc. INTRODUCTION The superior olivary complex (SOC) is a major auditory center in the lower brain stem, which not only relays ascending auditory information bilaterally from the cochlear nucleus (CN) to higher levels but also provides important descending pathways to the cochlea and CN (Spangler and Warr, 1991). The projections from the SOC to the cochlea are described as the lateral and medial olivocochlear bundle (OCB) systems. The lateral OCB of the rat originates almost entirely from neurons in the ipsilateral lateral superior olive (LSO) and terminates on the peripheral terminals of auditory nerve fibers synapsing with inner hair cells. The medial OCB originates bilaterally, mostly contralaterally, from the ventral nucleus of the trapezoid body (VNTB) and terminates on the outer hair cells (White and Warr, 1983). Projections from the SOC to the CN include some collaterals of the OCB as well as other fibers which project via the trapezoid body (Benson and Brown, 1990; Osen et al., 1984; Sherriff and Henderson, 1994; Spangler and Warr, 1991; Warr, 1992; Warr and Beck, 1996; White and Warr, 1983). Although many neurotransmitters may be involved in descending auditory pathways (Godfrey et al., 1988b; Potashner et al., 1993; Spangler and Warr, 1991; Wenthold, 1991), ample evidence has established that acetylcholine (ACh) is important in the OCB systems and other projections from the SOC to the CN (Altschuler et al., 1985; Godfrey and Ross, 1985; Godfrey et al., 1987a,b; Jasser and Guth, 1973; Osen et al., 1984; Sherriff and Henderson, 1994; Vetter et al., 1991). The exact roles of the r 1998 WILEY-LISS, INC. cholinergic pathways in hearing mechanisms are still unclear. It has been proposed that the OCB system may be important for modulating cochlear sensitivity to sound, that its branches to the CN may be important for resetting neuronal intensity coding to compensate for the adjustments of cochlear sensitivity (Benson and Brown, 1990), and that during cochlear pathology this feedback system might be involved in a mechanism of tinnitus (Levine and Kiang, 1995). Moreover, there is evidence that the OCB system has important influences on the early development of cochlear function (Walsh and McGee, 1997) and that neurons of the LSO, including lateral OCB neurons, show plasticity after cochlear lesions (Illing et al., 1997a,b). There is limited information about the organization of the cholinergic neurons in the SOC and their relation to the OCB. A delineation of this organization is essential to an understanding of the roles of these cholinergic Abbreviations used: ACh, acetylcholine; AChE, acetylcholinesterase; ANOVA, analysis of variance; ChAT, choline acetyltransferase; CN, cochlear nucleus; LNTB, lateral nucleus of the trapezoid body; LSO, lateral superior olive; LSO-b, border of lateral superior olive; LSO-i, intermediate limb of lateral superior olive; LSO-l, lateral limb of lateral superior olive; LSO-m, medial limb of lateral superior olive; MNTB, medial nucleus of the trapezoid body; MSO, medial superior olive; OCB, olivocochlear bundle; SOC, superior olivary complex; SPN, superior paraolivary nucleus; VAChT, vesicular acetylcholine transporter; VNTB, ventral nucleus of the trapezoid body; VNTB-l, group of more lateral and larger VNTB ChAT-positive neurons; VNTB-m, group of more medial and smaller VNTB ChAT-positive neurons. Contract grant sponsor: NIH; Contract grant number: DC 00172. *Correspondence to: Weiping Yao, Department of Otolaryngology, Medical College of Ohio, P.O. Box 10008, Toledo, OH 43699-0008. E-mail: WYAO@Gemini.MCO.EDU Received 15 May 1997; Accepted in revised form 12 September 1997 IMMUNOHISTOCHEMISTRY FOR CHAT AND VACHT IN SOC neurons in the plasticity of the auditory system. The present report represents a step toward this delineation in the rat. We used, as markers of cholinergic structures, antibodies against choline acetyltransferase (ChAT), the synthetase for ACh, and vesicular acetylcholine transporter (VAChT), which facilitates packaging of ACh into neurotransmitter vesicles in cholinergic terminals and has been used recently as a marker for localizing especially cholinergic terminals (Gilmor et al., 1996). Because of the consistency and clarity of the soma labeling with the ChAT antibody, we were able to quantitatively estimate the numbers, sizes, and labeling densities of ChAT-immunopositive somata in rat SOC regions. We also compared the results of ChAT and VAChT immunohistochemistry with our preparations of histochemistry for acetylcholinesterase (AChE), the degradative enzyme for ACh, which has long been used as a marker in studies of cholinergic systems. MATERIALS AND METHODS Animals Young adult male rats weighing 250–350 g were used in this study. Their care and use were approved by the National Institutes of Health (NIDCD, DC 00172) and by the Medical College of Ohio Institutional Animal Care and Use Committee. Histochemistry The immunohistochemical procedures basically followed the antibody providers’ recommendations and were similar to those previously reported (Yao and Godfrey, 1995). The animals were, under sodium pentobarbital anesthesia, intracardially perfused with 3–4% paraformaldehyde in phosphate-buffered saline (PBS, pH 7.4). Brain blocks were isolated and cryoprotected with 30% sucrose in PBS. Transverse sections 30-µm thick were cut and collected in cold PBS. Two sets of every fourth section were immunostained, one for ChAT and an adjacent set for VAChT (the other two sets were used for other experiments). Antibody dilutions with PBS containing 5% normal rabbit serum and 0.2% Triton X-100 were 1:4–6 for anti-ChAT antibody (monoclonal, Boehringer Mannheim, Inc., Indianapolis, IN) and 1:1,000 for anti-VAChT antibody (monoclonal, INCSTAR, Stillwater, MN). After one overnight incubation at 4°C with primary antibody, the sections were processed with a Vectastain ABC kit and diaminobenzidine as chromogen. Sections were mounted on slides for light microscopic studies. Controls in which primary antibodies were replaced with normal serum showed no labeling above background. For ChAT immunohistochemistry, we did not include 15% picric acid in the fixative, as recommended by the antibody provider (Boehringer Mannheim, Inc.) for optimal staining of terminals (Yao and Godfrey, 1993). This was because we were using the same animals for VAChT immunohistochemistry as well as ChAT and were not primarily concerned with ChAT-labeled terminals, but rather somata. Nevertheless, there was no substantial difference in terminal labeling for ChAT in these sections from what we have seen with the inclusion of picric acid. Histochemistry for AChE involved preparations in which frozen 10-µm-thick brain sections were incubated for 1–2 hours in a buffered (pH 5.6) medium 271 containing acetylthiocholine iodide, cupric sulfate, potassium ferricyanide, and a blocker of pseudocholinesterase (El-Badawi and Schenk, 1967; Karnovsky and Roots, 1964; Yao and Godfrey, 1993). Imaging and Profile Counts The SOC regions, corresponding to the terminology of White and Warr (1983), including LSO, VNTB, medial superior olive (MSO), medial nucleus of the trapezoid body (MNTB), lateral nucleus of the trapezoid body (LNTB), and superior paraolivary nucleus (SPN), were recognizable in our preparations, after consulting Nisslstained sections from other rats. The boundaries of the structures were traced using a computer program, Neurolucida 2.1 (MicroBrightField, Inc., Colechester, VT), and the locations, sizes (areas in square micrometers), and labeling densities of ChAT-positive soma profiles were recorded on maps. Each staining density, measured as a grey value within the range of 0 to 255 (lightest to darkest), was corrected by subtraction of the average density of the spinal trigeminal tract, an adjacent non-cholinergic structure, as a background control, and compared with those of the facial nucleus, a nearby cholinergic structure. Although these relative staining density estimates do not provide direct quantitation of amounts of cholinergic elements, high correlations have been found between such estimates and their respective assayed enzyme activities for ChAT and AChE in the rat CN (Yao and Godfrey, 1995), and between such measurements and antigen concentrations (Huang et al., 1996). Labeled neurons were grouped, based on their locations and morphological features. The total number of ChAT-positive neurons in each group was estimated by counting labeled profiles. The criteria for counting positively labeled profiles included elimination of small fractions (,5–8 µm in diameter, proportional to soma sizes for the various groups) of labeled structures without somatic features and very lightly labeled profiles (occasional uncertainties, one or two per SOC section, were resolved by comparing measured densities to the background; the required minimal density for counting was the mean background density 1 2 standard deviations). The total count for all sections of each SOC in each rat was multiplied by 4 since every fourth section was counted. The mean from each side of 3 rats (n 5 6) was calculated. Our method of profile counting is the most commonly used method in the literature for estimating numbers of neurons (Coggeshall and Lekan, 1996), and the results should thus be comparable to those of others using similar counting methods in the rat SOC (Aschoff and Ostwald, 1987, 1988; Robertson et al., 1989; Vetter and Mugnaini, 1990, 1992; Vetter et al., 1991; White and Warr, 1983). Although this method can only provide estimates, it was the only reasonable choice for our series of every fourth section in which intracellular details were usually obscured by the immunoreaction. Konigsmark’s equation (4) (Konigsmark, 1970) was applied to correct for overestimation of the total number of neurons because of double-counting of somata cut by the sectioning (Coggeshall and Lekan, 1996). Both total and corrected counts were recorded. The corrected ones should be more accurate representations of the actual numbers of neurons, although they may still be overes- 272 W. YAO AND D.A. GODFREY Fig. 1. Photomicrographs of ChAT immunoreactivity in SOC regions and the facial nucleus (FN). The approximate locations of the SOC regions are indicated in the center drawing. Scale bar (50 µm) applies to all photographs. timates (Coggeshall et al., 1990). However, the uncorrected total numbers are also reported because they should be more comparable to previous counts of OCB neurons, for which correction factors were not applied (Vetter and Mugnaini, 1992). Statistical comparisons were made across groups using analysis of variance (ANOVA, one-sided) and t-tests. RESULTS Appearance of Histochemistry ChAT-positive (presumably cholinergic) somata were mostly found in the LSO, the VNTB, and, to a much lesser extent, in the LNTB. No ChAT-positive somata were found in the MSO, SPN, or MNTB. ChAT-positive somata varied in size, shape, and labeling density in different parts of the SOC. Locations of ChAT-positive somata were grouped under lateral, intermediate, and medial limbs of LSO (LSO-l, LSO-i, LSO-m), the border of LSO (LSO-b), which represents the capsular structure surrounding the LSO proper, the LNTB, the lateral, larger-cell group of VNTB (VNTB-l), and the medial, smaller-cell group of VNTB (VNTB-m). Figure 1 shows examples of ChAT-positive somata from these designated regions or groups. Examples of ChAT- IMMUNOHISTOCHEMISTRY FOR CHAT AND VACHT IN SOC 273 Fig. 2. Photomicrographs of VAChT immunoreactivity in SOC regions and the FN. The approximate locations of the SOC regions are indicated in the center drawing. Scale bar (50 µm) applies to all photographs. positive somata of the facial nucleus (FN), a well-known group of cholinergic neurons close to the SOC in the brain stem, are included for comparison. From similar locations, examples of VAChT immunohistochemistry and AChE histochemistry are shown in Figures 2 and 3, respectively. ChAT immunoreactivity was primarily localized in the cytoplasm of somata and major proximal dendrites. There was also diffuse neuropil labeling for ChAT in the SOC, more prominent in the lateral aspect of the LSO and in the VNTB-m. The VAChT immunoreactivity was predominantly localized in puncta (presumably axon terminals) throughout the SOC, more in the LSO, VNTB, and LNTB than in the MNTB, MSO, and SPN. The VAChT-positive puncta were mostly found near somatic profiles, in the LSO, LNTB, VNTB-l, and MNTB, and sparsely along processes, in the LSO, MSO, LNTB, and VNTB. VChAT immunoreactivity was also found in the cytoplasm of some of the somata in the SOC, especially in the VNTB-l, LSO-m, and LSO-i (Fig. 2). Darkly labeled VAChT-positive puncta and somata were both seen in the FN. Histochemical labeling for AChE was more 274 W. YAO AND D.A. GODFREY Fig. 3. Photomicrographs of AChE histochemistry in SOC regions and the FN. The approximate locations of the SOC regions are indicated in the center drawing. The association of AChE with membranes and extracellular spaces (Yao and Godfrey, 1996) probably contributes to the less clearly defined somatic profiles than in the immunohistochemical preparations. Scale bar (50 µm) applies to all photographs. Fig. 4. Soma profiles of ChAT-positive neurons in the rat SOC. The approximate locations of the SOC regions are indicated in the center drawing. Scale bar in VNTB-l applies to all soma profile tracings. Bottom left: Comparison of soma profile areas. Analysis of variance (ANOVA) indicates significant differences among SOC regions. Group vs. group comparisons (t-tests) indicate significant differences between LSO-b and each LSO limb and VNTB-l, and between each LSO limb and VNTB-l, VNTB-m, and LNTB. Bottom right: Comparison of labeling densities with those of facial motoneurons; ANOVA indicates significant differences among them. The average density of ChAT-negative somata in LSO, MNTB, MSO, and SPN was 11% of that of facial motoneurons, less than half the values in SOC ChAT-positive somata. Comparisons (t-tests) between groups indicate significant differences (P , 0.01), except between LSO-l and LSO-b, LSO-l and VNTB-m, and LSO-b and VNTB-m. The lower densities in smaller somata may partly reflect the smaller proportions of the section thickness which they occupy, and darker neuropil labeling in LSO may contribute to the measured densities of somata there. The data were measured on sections from four rats. Numbers of profiles measured were 55 for LSO-l, 79 for LSO-i, 152 for LSO-m, 17 for LSO-b, 20 for LNTB, 81 for VNTB-l, 140 for VNTB-m, and 30 for FN. Fig. 4. Fig. 5. Legend on page 278. Fig. 5. (Continued.) 278 W. YAO AND D.A. GODFREY diffuse and primarily found in the neuropil of the SOC. Relatively dark labeling for AChE activity was seen in somata in the VNTB-l and the FN, closely associated with the somatic membrane. Thus, as expected, ChAT appeared to be a more definitive marker than VAChT or AChE for identifying cholinergic somata. Therefore, the results from ChAT immunohistochemistry were used for quantitative analysis of cholinergic somata in the SOC. To provide direct visual comparison of the shapes and sizes of ChAT-positive neuronal somata in the LSO, LNTB, VNTB, and FN, soma profiles were traced (Schofield and Cant, 1996) using Neurolucida and shown in Figure 4. These tracings provide an impression of the amounts of heterogeneity in soma appearance within and among groups. Within all limbs of the LSO, ChAT-positive somata were small, oval to elongated. In LSO-b, labeled somata were more variable in shape and size, but generally larger than those within the LSO. The labeled somata in the LNTB and VNTB-l were generally relatively large and multiangular in shape, whereas those in VNTB-m were more similar to those within the LSO. The areas and labeling densities of ChAT-positive soma profiles, compared to the values for facial motoneurons as an internal standard, are shown in Figure 4. ANOVA indicated significant differences among the regions for both soma profile area and ChAT labeling density. The distributions of ChAT-positive somata in the LSO and VNTB showed gradients along both lateralmedial and caudal-rostral axes (Fig. 5). The distributional pattern of VAChT immunoreactivity in the SOC generally resembled that of ChAT, but the apparent labeling of somata was less certain when observed at higher magnification (Fig. 2). Histochemical labeling for AChE activity generally corresponded to the ChAT and VAChT immunoreactivities in the VNTB but not in the LSO, where AChE labeling was less prominent. Quantitative Analysis The LSO is the origin of the lateral OCB in the rat (White and Warr, 1983) and is tonotopically organized (Friauf, 1992; Schwartz, 1992; Warr and Beck, 1996). We arbitrarily divided the S-shaped LSO into 9 lateralto-medial sectors in order to quantitate the distribution of ChAT-positive somata along its tonotopic axis. Figure 6 (top) shows increasing concentrations (number of Fig. 5. Photomicrographs of ChAT immunohistochemistry, VAChT immunohistochemistry, and AChE histochemistry in the rat SOC across a caudal-rostral range of 480 µm, along with Neurolucida tracings illustrating the locations of ChAT-positive somata. For each column (series), top is caudal, bottom is rostral, and sections are at 120-µm intervals. Sections beyond each end, caudally and rostrally, contained no recognizable LSO segments. Symbols in the tracing indicate different sizes, shapes, and labeling densities of ChATpositive somata: solid circles are darkly labeled somata in LSO and LSO-b; open diamonds are less darkly labeled somata in LNTB; solid triangles are dorsolaterally located, larger, multipolar, and more darkly labeled somata (VNTB-l) in VNTB; open circles are ventromedially located, smaller, round or oval, and less darkly labeled somata (VNTB-m) in VNTB. Two cholinergic fiber tracts are included in the sections: the bundle of ChAT-positive fibers lateral to the LSO is the facial motor root, and the medial ChAT-positive bundles of fibers traversing the MNTB are abducens nerve rootlets. For each figure, top is dorsal and left is lateral, as indicated in top tracing. Scale bar (500 µm) applies to all photographs. Fig. 6. Distribution of ChAT-positive somata in rat LSO. For lateral-medial comparisons (top), the LSO is divided into nine sectors by straight lines perpendicular to its S-shaped axis, and the numbers 1–9 in the inset correspond to the abscissa labels, from lateral to medial. Sectors 1–3, 4–6, and 7–9 approximately correspond to the lateral, intermediate, and medial limbs, respectively. Sector areas ranged from 0.017–0.030 mm2 with a mode close to 0.025 mm2/sector. Therefore, the spatial densities of ChAT-positive somata are expressed as number per 0.025 mm2. The means and standard errors were obtained from 30 LSO sections in 3 animals. ANOVA indicates significant differences across the sectors (P , 0.001). For caudalrostral comparisons (bottom), the numbers of ChAT-positive somata are plotted at five levels 120 µm apart. ANOVA indicates significant differences across levels (P , 0.001). somata/area) of ChAT-positive somata from lateral to intermediate to medial LSO limbs. There is also a lateral-to-medial increase in concentrations of ChATpositive somata within the intermediate and medial limbs (4 to 6, and 7 to 9). In the caudal-to-rostral direction, a symmetric distribution of cholinergic somata in the LSO was found (Fig. 6, bottom). ANOVA indicated that the variations in number of ChATpositive somata along both lateral-medial (P , 0.001) and caudal-rostral (P , 0.001) axes were statistically significant. ChAT-positive somata within the LSO, the IMMUNOHISTOCHEMISTRY FOR CHAT AND VACHT IN SOC TABLE 1. Estimated numbers of ChAT-positive somata in rat SOC regions (mean 6 SEM for 6 SOCs of 3 rats) Within LSO LSO-b LNTB Total counted soma profiles Corrected counts1 VNTB-l VNTB-m 440 6 33 54 6 9 767 461 6 14 2235 6 85 360 6 27 42 6 7 666 346 6 10 1744 6 64 1Counts were corrected by applying equation (4) of Konigsmark (1970) to somata. The average soma radius was calculated from the average soma profile area (Fig. 4), and the diameter of uncounted fragments was estimated as 5–8 µm, proportional to soma sizes of the different groups. 279 ergic neurons; (4) Within the VNTB, there are two subpopulations of cholinergic neurons: a more dorsolaterally located group with larger, multipolar somata, more darkly labeled for ChAT, and a more numerous ventromedial group with smaller, round/oval somata, more lightly labeled for ChAT; and (5) Cholinergic terminals, as revealed by VAChT-positive puncta, are most concentrated in the LSO and VNTB, similar to the distribution of ChAT-positive somata. Whether these terminals derive from the ChAT-positive neurons in these nuclei or from elsewhere is not known. region immediately surrounding it (LSO-b), and the LNTB ventral to it were counted (Table 1). Of those within LSO, 16% were in the lateral, 33% in the intermediate, and 51% in the medial limb. Of those in LSO-b, 50% were next to the lateral limb of the LSO, 17% next to the intermediate limb, and 33% next to the medial limb. Compared to the LSO, the VNTB extends for a much longer distance along the caudal-rostral axis: 1.8 mm as measured in our preparations. The distribution of ChAT-positive somata along a 1.4-mm length of the VNTB is shown in Figure 7. The larger ChAT-positive somata were clustered more laterally (VNTB-l) and dorsally, while the smaller ChAT-positive somata were spread out more medially (VNTB-m) and ventrally. The numbers of both groups of ChAT-positive somata were counted (Table 1). From caudal to rostral in the VNTB, the ChAT-positive somata in both groups showed significant (P , 0.01), although nonlinear, increases in number (Fig. 8). The larger, more lateral ChAT-positive VNTB neurons are considered to give rise to the medial OCB (Sherriff and Henderson, 1994; Vetter et al., 1991), which provides a nonuniform innervation along the basal-to-apical tonotopic axis in the outer hair cell region of the rat cochlea (Dannhof and Bruns, 1993). Since the VNTB has been reported to have a lateral (low frequency) to medial (high frequency) tonotopic organization (Friauf, 1992; Warr and Beck, 1996), we wanted to examine the medial-lateral distribution of the larger, darker ChAT-positive somata. To do this, we measured their distances from the midline of the brain stem (Fig. 9). The group of these neurons gradually shifted laterally at increasingly rostral locations. As would be predicted from the increasing number of larger, darker ChAT-positive somata at successively more rostral levels (Fig. 8), there were more located laterally than medially within their medial-lateral range. Comparisons to Previous Studies Our observations of ChAT immunohistochemistry in the rat SOC are generally in agreement with those in the literature. For example, the results of Vetter et al. (1991) suggested that ChAT-positive neurons were not homogeneously distributed in rat LSO, with 14, 33, and 53% in the lateral, intermediate, and medial limbs, respectively. Our data are similar (16, 33, 51%). Our results further revealed a trend for a lateral-to-medial increase in concentration of cholinergic neurons within the intermediate and medial limbs of the LSO (Fig. 6), and for a higher density of ChAT immunolabeling in more medially located somata (Fig. 4). In addition, the larger ChAT-positive somata at the border of the LSO may correspond, at least in part, to the shell neurons of Vetter and Mugnaini (1992) and are likely to represent a group of neurons distinct from those inside the LSO (Warr et al., 1997). Nevertheless, since these neurons appear to project with the lateral OCB (Warr et al., 1997), we included them in our total counts of LSO neurons. In the VNTB, which corresponds to the combination of medioventral (MVPO) and rostral (RPO) periolivary regions of Vetter et al. (1991), our results agree with those of Vetter et al. (1991) and Sherriff and Henderson (1994) in finding lateral ChAT-positive neurons that are larger and more darkly labeled than those more medially located. Our results reveal, moreover, that the lateral group is actually located dorsolaterally and the medial group ventromedially. It can be noted from Figure 4 that the smaller, medial somata (VNTBm), although they appear in sections to have lighter ChAT labeling, do not have much lower measured densities than the larger, lateral somata (VNTB-l). The lighter appearance may merely reflect the smaller proportion of the section thickness that they occupy because of their smaller size. The caudal-to-rostral variations in numbers of ChAT-positive neurons in the LSO and VNTB have not been reported previously, although a greater dorsal-ventral size of the VNTB rostrally was noted by Vetter et al. (1991). DISCUSSION Cholinergic Neurons in the SOC Based on our immunohistochemical results, the characteristics of cholinergic neurons in the rat SOC can be proposed to be as follows: (1) Almost all cholinergic somata are in the LSO and VNTB; (2) Within the LSO, small cholinergic somata are distributed across all limbs, but there is a distinct lateral-to-medial increase in number and labeling density for ChAT; (3) At the border of the LSO, a small group of cholinergic somata, larger and more variable in shape than within the LSO proper, may represent a distinct subpopulation of cholin- ChAT-Positive LSO Neurons It is difficult to compare our ChAT-positive neuron counts with counts of OCB neurons to conclude definitely whether or not all lateral or medial OCB neurons may be cholinergic. Differences in tracers, which are probably not all taken up equally well (Aschoff and Ostwald, 1988; Schofield and Cant, 1996), as well as differences in section thicknesses and in assumptions used for counting soma profiles, may lead to significant differences in total counts (Coggeshall and Lekan, 1996). The estimated numbers of lateral and medial OCB neurons for adult rats vary considerably across 280 W. YAO AND D.A. GODFREY studies: lateral 238–720; medial 213–583 (Aschoff and Ostwald, 1987, 1988; Robertson et al., 1989; Vetter and Mugnaini, 1990, 1992; Vetter et al., 1991; White and Warr, 1983). Since these counts were not corrected for double counting, our uncorrected counts should be more comparable to them than our corrected counts. Some previous studies suggested that as few as half the lateral OCB neurons may be cholinergic (Vetter et al., 1991), while others suggested that all are (Godfrey et al., 1984). Our uncorrected count of LSO-region ChATpositive neurons (within LSO 1 LSO-b: 494) is similar to those of Aschoff and Ostwald (1988, 540) and Robertson et al. (1989; 573) for lateral OCB neurons. This similarity suggests that at least most, if not all, lateral OCB neurons are cholinergic. The study of Vetter et al. (1991), which estimated that about half the lateral OCB neurons are ChAT-positive, found a much smaller number of ChAT-positive LSO neurons than we found. This could have resulted from poor penetration of antibody (Altschuler et al., 1984; Vetter et al., 1991), particularly after the sections had been reacted for HRP activity, and/or use of a polyclonal antibody, which, in our preparations, gives less reliable labeling of LSO somata than that used here. Additional work will be needed to more confidently determine what proportion of lateral OCB neurons is cholinergic. The LSO is known to be tonotopically organized (Friauf, 1992; Schwartz, 1992). From lateral-to-medial, the characteristic frequencies of LSO neurons are arranged from low to high (Friauf, 1992; Guinan et al., 1972; Tsuchitani and Boudreau, 1966). The lateral-tomedial increasing gradient of ChAT-positive neurons in the LSO is consistent with a stronger cholinergic influence on higher frequency auditory processing in more basal parts of the cochlea. Cholinergic innervation of the cochlea by the lateral OCB has been documented as ChAT-positive puncta under the inner hair cells (Altschuler et al., 1985), as measured ChAT and AChE enzyme activities in the inner hair cell region (Godfrey and Ross, 1985; Godfrey et al., 1986), and as size of the AChE-positive inner spiral bundle (Firbas, 1978). The latter two measures correlate well with each other and with the estimated distribution of lateral OCB fibers to the inner hair cell region (Dannhof and Bruns, 1993). In the rat inner hair cell region, measured ChAT enzyme activity and size of the inner spiral bundle were much lower in the apical turn than in the middle and basal turns. This is consistent with the lower number of ChAT-positive neurons in the lateral limb of the LSO. However, the finding that the highest density of ChAT-positive somata is in the most medial part of the LSO does not agree with the decreased size of the inner spiral bundle in the most basal part of the cochlea. Also, the distribution of CGRP-positive, presumed cholinergic, terminals in the inner hair cell region does not correspond well with the medial-lateral Fig. 7. Neurolucida tracing of ChAT-positive somata in a series of sections spaced at 120-µm intervals through the rat VNTB, from a different rat than in Figure 5. Top-to-bottom represents caudal-torostral. From top downward, the third through seventh sections contain S-shaped LSO. Solid triangles represent larger and more darkly labeled somata; open circles represent smaller and more lightly labeled somata. IMMUNOHISTOCHEMISTRY FOR CHAT AND VACHT IN SOC 281 Fig. 8. Distribution of ChAT-positive somata along the caudalrostral axis of the VNTB. The S-shaped LSO is present at 480–960 µm. VNTB-l represents larger, more darkly labeled, and more dorsolaterally located ChAT-positive somata (solid triangles in Figs. 5 and 7). VNTB-m represents smaller, more lightly labeled, and more ventrome- dially located somata (open circles in Figs. 5 and 7). The means and standard errors were calculated from 6 series of VNTB sections in 3 animals. The increasing trend from caudal to rostral is statistically significant by analysis of variance (P , 0.01) for each group of somata. distribution of ChAT-positive somata in the LSO (Vetter et al., 1991). frey et al., 1987a,b). To the extent that these fibers derive from the SOC, our results suggest that there are few potential sources available besides the smaller ChAT-positive VNTB neurons. Their large number is sufficient to account for the amount of ChAT activity that disappears from the CN when the trapezoid body is completely cut, and their light staining for AChE activity may correlate with its smaller decline in the CN (Godfrey et al., 1983; Osen et al., 1984) than in the cochlear inner hair cell region (Godfrey and Ross, 1985) after transection of their centrifugal innervations. The larger number rostrally of the smaller ChAT-positive VNTB neurons could be consistent with the relatively larger effects on CN ChAT activity of lesions including more rostral damage (Godfrey et al., 1987b) and correlates with evidence for higher ChAT activities rostrally than caudally in the trapezoid body (Godfrey et al., 1988a). The rat VNTB has been reported to be tonotopically organized in a lateral-to-medial low-to-high frequency organization (Friauf, 1992; Warr and Beck, 1996). A lateral-to-medial increasing gradient in number of larger, darker ChAT-positive neurons (VNTB-l) might, therefore, be expected to correlate with cochlear outer hair cell region gradients of ChAT enzyme activities (Godfrey and Ross, 1985; Godfrey et al., 1986) and ChAT immunoreactivities (Altschuler et al., 1985; Dannhof et al., 1991; Eybalin, 1993), more prominent basally than apically. Since only a few larger, darker ChAT-positive somata are clustered in each transverse section (Figs. 5,7), there is only a small range of medial-lateral locations at any given rostral-caudal level. Therefore, medial-to-lateral variations cannot be easily assessed at one coronal plane. However, a wider ChAT-Positive VNTB Neurons In the VNTB, the counts of ChAT-positive neurons are more involved because the somata are spread over a larger volume and exist as two groups. In our preparations, the uncorrected number of darkly ChAT-labeled large somata (461) is within the range of the counts of medial OCB neurons, to which they correspond (Sherriff and Henderson, 1994; Vetter et al., 1991), similar to those of Aschoff and Ostwald (1988; 556) and Robertson et al. (1989; 474), the studies for which the LSO count was also similar. This similarity is consistent with a conclusion that most or all medial OCB neurons are cholinergic (Dannhof et al., 1991; Vetter et al., 1991). The medial OCB neurons are considered to send some cholinergic collaterals to the CN (Benson and Brown, 1990; Godfrey et al., 1987a; Osen et al., 1984; Sherriff and Henderson, 1994). The smaller, more lightly labeled ChAT-positive neurons in the VNTB (VNTB-m, Figs. 7,8) are about five times as numerous as the larger neurons. It is believed that they are, at least in part, responsible for the cholinergic projections to the CN via the trapezoid body route (Godfrey et al., 1987b) and do not project to the cochlea (Sherriff and Henderson, 1994). These neurons do not stain darkly for AChE (Fig. 3) and may correspond to the more lightly AChElabeled neurons noted in cat by Adams (1989), who also suggested that they contribute cholinergic descending inputs to the CN. In the rat, approximately 65% of the ChAT activity in the CN appears to be related to centrifugal fibers entering from the trapezoid body, as compared to only 20% related to OCB branches (God- 282 W. YAO AND D.A. GODFREY VNTB-m neurons, which are much more numerous, may be important in cochlear nucleus plasticity. The VNTB neuronal groups, with their widespread connections, may have particularly significant influences on plasticity in the auditory system. Further, the VAChTpositive puncta in SOC regions suggest some intimate cholinergic connections among SOC neurons, which may be important for coordinating their activities. ACKNOWLEDGMENTS We thank Dr. William Gunning and Mr. Edward Calomeni of the Department of Pathology, Medical College of Ohio, for their generous provision of photographic facilities and expert assistance. We also thank Dr. W. Bruce Warr, Boys Town National Research Hospital, for valuable consultations on olivocochlear neurons. This work was supported by NIH grant DC 00172. REFERENCES Fig. 9. Medial-lateral distribution of large ChAT-pozsitive VNTB somata. Top: The mean 6 SD of soma locations at each rostral-caudal distance. Bottom: The mean (6 SEM) numbers of somata (per series of 12 transverse sections spaced at 120-µm intervals) located at different distances from the midline are plotted. Each value on the abscissa represents a 100-µm range centered at that distance. Data are from 6 series of VNTB sections in 3 animals. ANOVA indicates significant variance across distances (P , 0.001). By t-test, the number of ChAT-positive somata at 1,800–1,899 µm from the midline is larger than that at 1,600–1,699 µm (P , 0.05). range exists across the entire rostral-caudal extent of the VNTB because of the lateral shift of the group of these somata at increasingly rostral levels. In our analysis, the distribution of the larger, darker ChATpositive somata showed more of them laterally than medially, corresponding to lower and higher frequencies, respectively (Friauf, 1992; Warr and Beck, 1996), and opposite to the expected distribution. This discrepancy suggests that our current knowledge of the tonotopic mapping in the VNTB is probably inadequate, especially considering the several neuronal groups present there (Warr and Beck, 1996) and the differences among studies regarding the definition of the VNTB. Our results suggest that, if the cholinergic OCB neurons are important in cochlear plasticity, the Adams, J.C. (1989) Non-olivocochlear cholinergic periolivary cells. Soc. Neurosci. Abstr., 15:1114. Altschuler, R.A., Fex, J., Parakkal, M., and Eckenstein, F. (1984) Colocalization of enkephalin-like and choline acetyltransferase-like immunoreactivities in olivocochlear neurons of the guinea pig. J. Histochem. Cytochem., 32:839–843. Altschuler, R.A., Kachar, B., Rubio, J.A., Parakkal, M., and Fex, J. (1985) Immunocytochemical localization of choline acetyltransferaselike immunoreactivity in the guinea pig cochlea. Brain Res., 338:1–11. Aschoff, A., and Ostwald, J. (1987) Different origins of cochlear efferents in some bat species, rats, and guinea pigs. J. Comp. Neurol., 264:56–72. Aschoff, A., and Ostwald, J. (1988) Distribution of cochlear efferents and olivo-collicular neurons in the brainstem of rat and guinea pig. Exp. Brain Res., 71:241–251. Benson, T.E., and Brown, M.C. (1990) Synapses formed by olivocochlear axon branches in the mouse cochlear nucleus. J. Comp. Neurol., 295:52–70. Coggeshall, R.E., and Lekan, H.A. (1996) Methods for determining numbers of cells and synapses: A case for more uniform standards of review. J. Comp. Neurol., 364:6–15. Coggeshall, R.E., La Forte, R., and Klein, C.M. (1990) Calibration of methods for determining numbers of dorsal root ganglion cells. J. Neurosci. Methods, 35:187–194. Dannhof, B.J., and Bruns, V. (1993) The innervation of the organ of Corti in the rat. Hearing Res., 66:8–22. Dannhof, B.J., Roth, B., and Bruns, V. (1991) Anatomical mapping of choline acetyltransferase (ChAT)-like and glutamate decarboxylase (GAD)-like immunoreactivity in outer hair cell efferents in adult rats. Cell Tissue Res., 266:89–95. El-Badawi, A., and Schenk, E.A. (1967) Histochemical methods for separate, consecutive and simultaneous demonstration of acetylcholinesterase and norepinephrine in cryostat sections. J. Histochem. Cytochem., 15:580–588. Eybalin, M. (1993) Neurotransmitters and neuromodulators of the mammalian cochlea. Physiol. Rev., 73:309–373. Firbas, W. (1978) The efferent innervation in the region of inner hair cells in the organ of Corti. Acta Otolaryngol., 86:309–313. Friauf, E. (1992) Tonotopic order in the adult and developing auditory system of the rat as shown by c-fos immunocytochemistry. Eur. J. Neurosci., 4:798–812. Gilmor, M.L., Nash, N.R., Roghani, A., Edwards, R.H., Yi, H., Hersch, S.M., and Levey, A.I. (1996) Expression of the putative vesicular acetylcholine transporter in rat brain and localization in cholinergic synaptic vesicles. J. Neurosci., 16:2179–2190. Godfrey, D.A., and Ross, C.D. (1985) Enzymes of acetylcholine metabolism in the rat cochlea. Ann. Otol. Rhinol. Laryngol., 94:409–414. Godfrey, D.A., Park, J.L., Rabe, J.R., Dunn, J.D., and Ross, C.D. (1983) Effects of large brain stem lesions on the cholinergic system in the rat cochlear nucleus. Hearing Res., 11:133–156. Godfrey, D.A., Park, J.L., and Ross, C.D. (1984) Choline acetyltransferase and acetylcholinesterase in centrifugal labyrinthine bundles of rat. Hearing Res., 14:93–106. Godfrey, D.A., Wiet, G.J., and Ross, C.D. (1986) Quantitative histochemistry of the cochlea. In: Neurobiology of Hearing: The Cochlea. R.A. IMMUNOHISTOCHEMISTRY FOR CHAT AND VACHT IN SOC Altschuler, R.P. Bobbin, and D.W. Hoffman, eds. Raven Press, New York, pp. 149–160. Godfrey, D.A., Park-Hellendall, J.L., Dunn, J.D., and Ross, C.D. (1987a) Effects of olivocochlear bundle transection on choline acetyltransferase activity in the rat cochlear nucleus. Hearing Res., 28:237–251. Godfrey, D.A., Park-Hellendall, J.L., Dunn, J.D., and Ross, C.D. (1987b) Effects of trapezoid body and superior olive lesions on choline acetyltransferase in the rat cochlear nucleus. Hearing Res., 28:253–270. Godfrey, D.A., Carlson, L., Parli, J.A., and Ross, C.D. (1988a) Distribution of choline acetyltransferase activity in the trapezoid body of the rat. Soc. Neurosci. Abstr., 14:490. Godfrey, D.A., Parli, J.A., Dunn, J.D., and Ross, C.D. (1988b) Neurotransmitter microchemistry of the cochlear nucleus and superior olivary complex. In: Auditory Pathway. J. Syka and R.B. Masterton, eds. Plenum Publishing Corp., New York, pp. 107–121. Guinan, J.J., Norris, B.E., and Guinan, S.S. (1972) Single auditory units in the superior olivary complex. II: Locations of unit categories and tonotopic organization. Int. J. Neurosci., 4:147–166. Huang, X., Chen, S., and Tietz, E.I. (1996) Immunocytochemical detection of regional protein changes in rat brain sections using computer-assisted image analysis. J. Histochem. Cytochem., 44:981– 987. Illing, R.-B., Cao, Q.-L., and Förster, C.R. (1997a) Lesion-induced modulations of GAP-43 expression suggest a dominant role for the superior olive in auditory brainstem plasticity. Soc. Neurosci. Abstr. 23:184. Illing, R.-B., Horváth, M., and Laszig, R. (1997b) Plasticity of the auditory brainstem: Effects of cochlear ablation on GAP-43 immunoreactivity in the rat. J. Comp. Neurol., 382:116–138. Jasser, A., and Guth, P.S. (1973) The synthesis of acetylcholine by the olivo-cochlear bundle. J. Neurochem., 20:45–53. Karnovsky, M.J., and Roots, L. (1964) A ‘‘direct coloring’’ thiocholine method for cholinesterase. J. Histochem. Cytochem., 12:219–221. Konigsmark, B.W. (1970) Methods for the counting of neurons. In: Contemporary Research Methods in Neuroanatomy. W.J.H. Nauta and S.O.E. Ebbesson, eds. Springer, Heidelberg, pp. 315–338. Levine, R.A., and Kiang, N.Y.S. (1995) A conversation about tinnitus. In: Mechanisms of Tinnitus. J.A. Vernon and A.R. Moller, eds. Allyn and Bacon, Needham Heights, pp. 149–161. Osen, K.K., Mugnaini, E., Dahl, A.L., and Christiansen, A.H. (1984) Histochemical localization of acetylcholinesterase in the cochlear and superior olivary nuclei. A reappraisal with emphasis on the cochlear granule cell system. Arch. Ital. Biol., 122:169–212. Potashner, S.J., Benson, C.G., Ostapoff, E.M., Lindberg, N., and Morest, D.K. (1993) Glycine and GABA: Transmitter candidates of projections descending to the cochlear nucleus. In: The Mammalian Cochlear Nuclei: Organization and Function. M.A. Merchán, J.M. Juiz, D.A. Godfrey, and E. Mugnaini, eds. Plenum Press, New York, pp. 195–210. Robertson, D., Harvey, A.R., and Cole, K.S. (1989) Postnatal development of the efferent innervation of the rat cochlea. Dev. Brain Res., 47:197–207. 283 Schofield, B.R., and Cant, N.B. (1996) Projections from the ventral cochlear nucleus to the inferior colliculus and the contralateral cochlear nucleus in guinea pigs. Hearing Res., 102:1–14. Schwartz, I.R. (1992) The superior olivary complex and lateral lemniscal nuclei. In: The Mammalian Auditory Pathway: Neuroanatomy. D.B. Webster, A.N. Poppe, and R.R. Fay, eds. Springer-Verlag, New York, pp. 117–167. Sherriff, F.E., and Henderson, E. (1994) Cholinergic neurons in the ventral trapezoid nucleus project to the cochlear nuclei in the rat. Neuroscience, 58:627–633. Spangler, K.M., and Warr, W.B. (1991) The descending auditory systems. In: Neurobiology of Hearing: The Central Auditory System. R.A. Altschuler, R.P. Bobbin, and D.W. Hoffman, eds. Raven Press, New York, pp. 27–45. Tsuchitani, C., and Boudreau, J.C. (1966) Single unit analysis of cat superior olive S-segment with tone stimuli. J. Neurophysiol., 29:684– 697. Vetter, D.E., and Mugnaini, E. (1990) An evaluation of retrograde tracing methods for the identification of chemically distinct cochlear efferent neurons. Arch. Ital. Biol., 128:331–353. Vetter, D.E., and Mugnaini, E. (1992) Distribution and dendritic features of three groups of rat olivocochlear neurons: A study with two retrograde cholera toxin tracers. Anat. Embryol., 185:1–16. Vetter, D.E., Adams, J.C., and Mugnaini, E. (1991) Chemically distinct rat olivocochlear neurons. Synapse, 7:21–43. Walsh, E.J., and McGee, J. (1997) Does activity in the olivocochlear bundle affect development of the auditory periphery? In: Proceedings of the International Symposium on Diversity in Auditory Mechanics. E.R. Lewis, G.R. Long, R.F. Lyon, P.M. Narins, C.B. Steele, and E. Hecht-Poinar, eds. River Edge, NJ: World Scientific, pp. 376–385. Warr, W.B. (1992) Organization of olivocochlear efferent systems in mammals. In: The Mammalian Auditory Pathway: Neuroanatomy. D.B. Webster, A.N. Popper, and R.R. Fay, eds. Springer-Verlag, New York, pp. 410–448. Warr, W.B., and Beck, J.E. (1996) Multiple projections from the ventral nucleus of the trapezoid body in the rat. Hearing Res., 93:83–101. Warr, W.B., Boche, J.E.B., and Neely, S.T. (1997) Efferent innervation of the inner hair cell region: Origins and terminations of two lateral olivocochlear systems. Hearing Res., 108:89–111. Wenthold, R.J. (1991) Neurotransmitters of brainstem auditory nuclei. In: Neurobiology of Hearing: The Central Auditory System. R.A. Altschuler, R.P., Bobbin, and D.W. Hoffman, eds. Raven Press, New York, pp. 121–139. White, J.S., and Warr, W.B. (1983) The dual origins of the olivocochlear bundle in the albino rat. J. Comp. Neurol., 219:203–214. Yao, W., and Godfrey, D.A. (1993) Choline acetyltransferase in cochlear root neurons. Hearing Res., 69:76–82. Yao, W., and Godfrey, D.A. (1995) Immunohistochemistry of muscarinic acetylcholine receptors in rat cochlear nucleus. Hearing Res., 89:76–85. Yao, W., and Godfrey, D.A. (1996) Electron microscopic histochemistry for acetylcholinesterase in the dorsal cochlear nucleus of rat. J. Acoust. Soc. Am., 100:2629.