Amino acid and acetylcholine chemistry in mountain beaver cochlear nucleus and comparisons to pocket gopher, other rodents, and cat

Abstract

The mountain beaver and pocket gopher are two rodents that live mostly underground in tunnel systems. Previous studies have suggested that their cochlear nucleus structure, particularly that of the dorsal cochlear nucleus (DCN), differs significantly from that of other mammals, that the hearing ability of the pocket gopher is deficient compared to that of other rodents, and that the DCN of the mountain beaver is more responsive to slow oscillations of air pressure than to sounds. We conducted some electrophysiological recordings from mountain beaver DCN and then used microchemical methods to map in mountain beaver cochlear nuclei the distributions of amino acids, including the major neurotransmitters of the brain, and enzyme activities related to the metabolism of the neurotransmitter acetylcholine, which functions in centrifugal pathways to the cochlear nucleus. Similar measurements were made for a pocket gopher cochlear nucleus. Responses to tonal stimuli were found in mountain beaver DCN. Distributions and magnitudes of neurotransmitter and related amino acids within mountain beaver and pocket gopher cochlear nuclei were not very different from those of other rodents and cat. However, the enzyme of synthesis for acetylcholine, choline acetyltransferase, had only low activities in the DCN of both mountain beaver and pocket gopher. The chemical distributions in the mountain beaver DCN support a conclusion that it corresponds to just the superficial DCN portion of other mammals. High correlations between the concentrations of γ-aminobutyrate (GABA) and glycine were found for both mountain beaver and pocket gopher cochlear nuclei, suggesting that their co-localization in cochlear nucleus synapses may be especially prominent in these animals. Previous evidence suggests convergence of somatosensory and auditory information in the DCN, and this may be especially true in animals spending most of their time underground. Our results suggest that the enlarged DCN of the mountain beaver and that of the pocket gopher are not very different from those of other rodents with respect to involvement of amino acid neurotransmitters, but they appear to have reduced cholinergic innervation.

Introduction

Mountain beavers, or sewellels (Aplodontia rufa), and pocket gophers (Geomys bursarius) spend most of their lives underground in tunnel systems (Carraway and Verts, 1993; Nowak, 1999). This should expose them to different auditory environments than animals that spend most of their time above ground. A behavioral study indicated that the pocket gopher has relatively poor hearing (Heffner and Heffner, 1990), and the mountain beaver is considered to have limited hearing ability (Carraway and Verts, 1993). A physiological study of the mountain beaver suggested that much of its first auditory brain center, the cochlear nucleus, specifically the dorsal cochlear nucleus (DCN), was specialized for detection of slow changes in air pressure more than sounds (Merzenich et al., 1973).

In almost all rodents, and most other mammals that have been studied, the DCN is a layered structure, with 3 or 4 distinguishable layers. The second layer from the DCN surface (fusiform soma layer) contains the somata of the main output neurons of the DCN, the fusiform, or pyramidal cells, which have one set of dendrites directed toward the surface (superficial, or apical dendrites) and another set directed away from the surface (deep, or basal dendrites). The superficial dendrites receive input from granule cells and other interneurons, especially cartwheel cells, of the most superficial layer (molecular layer) and the fusiform soma layer. The granule cells relay information descending from higher auditory centers and information from other sensory systems, especially the somatosensory system. The molecular and fusiform soma layers can be referred to collectively as the superficial layers of the DCN. The deeper portion of the DCN, which will be referred to as the deep layer, receives major innervation from auditory nerve fibers, which transmit information about sounds coded by the cochlea. The deep dendrites of the fusiform cells receive input from auditory nerve fibers. Thus, the deep layer of the DCN is closely related to input from the cochlea, whereas the superficial layers are closely related to inputs from other sources (Cohen et al., 1972; Mugnaini et al., 1980; Lorente de Nó, 1981; Wouterlood and Mugnaini, 1984; Osen, 1988; Dunn et al., 1996; Godfrey et al., 1997; Young et al., 1997; Oertel and Young, 2004; Diño and Mugnaini, 2008).

Previous studies have indicated that the cochlear nuclei of the mountain beaver and pocket gopher have some different anatomical characteristics than in other mammals that have been studied, specifically relatively large granular and superficial DCN regions (Merzenich et al., 1973; Godfrey et al., 2016).

Studies of the chemistry underlying the physiological characteristics of the mountain beaver and pocket gopher cochlear nuclei have been lacking, including the chemistry of neurotransmission. Our study was directed at measuring some aspects of neurotransmitter chemistry in the mountain beaver and pocket gopher cochlear nuclei, for comparison to those of other mammals, particularly other rodents. We measured distributions within the cochlear nucleus of amino acids, including the major neurotransmitters glutamate, glycine, and γ-aminobutyrate (GABA), and of enzymes involved in the metabolism of the neurotransmitter acetylcholine. Based on studies of other mammals, especially rodents, glutamate is the likely neurotransmitter of auditory nerve fibers innervating the cochlear nucleus (Wenthold, 1978; Godfrey et al., 1988, 2015) and of cochlear nucleus granule cells (Oliver et al., 1983; Osen et al., 1995; Godfrey et al., 1997), as well as possibly some centrifugal and somatosensory innervation (Wright and Ryugo, 1996; Zeng et al., 2011; Milinkeviciute et al., 2017). Glycine and GABA have major roles as neurotransmitters of cochlear nucleus interneurons (Mugnaini, 1985; Altschuler et al., 1993; Oertel and Wickesberg, 1993; Oertel and Young, 2004; Apostolides and Trussell, 2013) and are also associated with centrifugal (Godfrey et al., 1988; Potashner et al., 1993; Saint Marie et al., 1993) and commissural (Wenthold, 1987) pathways to the cochlear nucleus. Aspartate and glutamine are closely related metabolically and functionally to glutamate (Shank and Aprison, 1981; Dingledine and McBain, 1994; Ross et al., 1995), and taurine is in many cases closely related functionally to GABA and glycine (Ottersen et al., 1988; Albrecht and Schousboe, 2005), including in the anteroventral cochlear nucleus (AVCN) (Song et al., 2012). Involvement of amino acids in other aspects of metabolism, including protein synthesis and energy metabolism (Lehninger, 1975), besides that related to neurotransmission, complicates the conclusions that can be drawn from measurements of their tissue levels. Acetylcholine is especially associated with centrifugal pathways to the cochlear nucleus, including branches from the olivocochlear bundle (Godfrey et al., 1983, 1987a,b, 1990; Sherriff and Henderson, 1994; Kraus and Illing, 2004; Meidinger et al., 2006; Mellott et al., 2011), but it is also associated with a few neuron somata within the cochlear nucleus (Godfrey et al.,1987b, 1990; Godfrey, 1993; Vetter et al., 1993).

Preliminary reports of mountain beaver data have been presented (Godfrey et al., 1996, 1997, 2000b).

Methods

2.1. Animals

Adult male mountain beavers were obtained from the U.S. Department of Agriculture’s Olympia Field Station in Olympia, Washington (courtesy of Dale Nolte). Three were used in the study: MB040595, MB080895, and MB103195. One adult eastern pocket gopher was obtained from Dr. Rickye S. Heffner of the Department of Psychology, University of Toledo. Treatment of animals was in accord with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

2.2. Electrophysiological recordings from mountain beavers

Electrophysiological recordings were attempted from two of the mountain beavers (MB040595 and MB080895) in the laboratory of Dr. Kaltenbach. For this purpose, the animals were each given an intramuscular injection of ketamine/xylazine (117 mg/kg/18 mg/kg) to induce a surgical level of anesthesia, as evidenced by abolition of the toe withdrawal, pinna, and corneal reflexes and muscular contractions during skin pinch. Cerebellar tissue was removed to expose the surface of the DCN, and multiunit recordings were made from the DCN surface, as in previous studies (Kaltenbach and Lazor, 1991; Kaltenbach and McCaslin, 1996; Kaltenbach and Afman, 2000; Kaltenbach and Godfrey, 2019). . Animal MB040595 was maintained under anesthesia for 6 hours for an unsuccessful recording session. Successful recordings were made from the left DCN of MB080895. The animal was maintained under anesthesia, with ketamine/xylazine as needed, for 8 hours for the recording session.

2.3. Preparation of sections of the cochlear nucleus

The methods for the microdissections and assays were the same as we have previously used and described (Godfrey and Matschinsky, 1976; Godfrey et al, 1983, 2000a, 2015; Linker et al., 2018).

The brains of the 3 mountain beavers were frozen in the laboratory of Dr. Kaltenbach. While under ketamine/xylazine anesthesia, as mentioned above, animals MB040595 and MB080895 were quickly decapitated using a guillotine, then the brain removed and transferred to a metal beaker containing Super Friendly Freeze-it, cooled with liquid nitrogen in a surrounding container, within 10 minutes of decapitation. Animal MB103195 was anesthetized with ketamine/xylazine (117 mg/kg/18 mg/kg), decapitated, and the brain removed and frozen within 8 minutes in Freeze-it, as above. After freezing, each brain was transferred to a small vial placed on dry ice filling a medium-sized box. The vial was closed, further surrounded by dry ice, and the box closed and transported to the laboratory of Dr. Godfrey. The brain of the eastern pocket gopher was removed after decapitation and frozen within 16 minutes in Freon (‘Freeze-it’) cooled to its freezing point (−130°C) with liquid nitrogen. Tissues were stored at −80°C until sectioning.

All brains were sectioned transversely at 20 μm thickness using a Hacker-Bright microtome-cryostat maintained at −20 °C. Sections were collected in three alternating sets: one set of every third section was collected in aluminum racks (Lowry and Passonneau, 1972) for subsequent freeze drying, a second was melted onto glass slides for staining of Nissl substance with cresyl violet, and a third was melted onto glass slides for staining of acetylcholinesterase (AChE) activity (Karnovsky and Roots, 1964; El-Badawi and Schenk, 1967; Godfrey et al. 1983). The AChE staining of the pocket gopher cochlear nucleus did not turn out well for technical reasons and was not usable. The aluminum racks containing sections for freeze drying were placed into glass vacuum tubes that were put into a −40 °C freezer, attached through rubber tubing and a dry ice trap to a vacuum pump, and evacuated overnight, at pressure less than 10 μm mercury (Lowry and Passonneau, 1972). The freeze drying of the sections from one of the mountain beavers (MB040595) appeared to be slightly incomplete, so that they were not considered suitable for the chemical assays.

2.4. Measurements on the stained sections

The stained sections were studied with a Wild dissecting microscope to determine histological landmarks, including the boundaries of cochlear nucleus subregions, to facilitate the choice of freeze-dried sections for chemical analysis. Volumes of the subregions were measured, as previously reported (Godfrey et al., 2016). To measure neuron soma sizes, stained sections were studied with a Leitz Aristoplan microscope at 540x magnification.

2.5. Isolation of tissue samples for chemical assays

Microdissection of freeze-dried sections to obtain samples for chemical assay was done for the right cochlear nucleus of each animal, in the same way as previously described (Lowry and Passonneau, 1972; Godfrey and Matschinsky, 1976; Linker et al., 2018). The cochlear nucleus portion of each section to be dissected was traced onto paper using a drawing tube attached to the Wild dissecting microscope at 25x magnification, including all discernible regional boundaries around and within the cochlear nucleus. These boundaries were then checked by comparing to the adjacent or nearby stained sections via the drawing tube. As cuts were made through the freeze-dried section to obtain samples, each cut was traced onto the tracing of the section. After the completion of the dissection of samples, therefore, a complete map of all sample locations in the section was retained.

Isolated samples were placed on sample holders, from which they were each individually picked up with a hairpoint and placed on a quartz fiber microbalance of appropriate sensitivity to measure the dry weight (Lowry and Passonneau, 1972). Sample dry weights ranged from 0.2 to 1.6 μg.

2.6. Other sample size measurements

Using the maps of the microdissections, sample volumes were measured by scanning the maps into a computer and then using Neurolucida software to trace the perimeter of each sample, from which the program calculated its area. Each sample’s area was multiplied by the 20 μm section thickness to obtain its volume. Sample volume provided an alternative reference for chemical measurements. Although it could not be measured as accurately as the dry weight, primarily because of possible slight variations in section thickness, volume is less affected by the lipid content of the sample (Godfrey and Matschinsky, 1976). Dry weight per volume values were calculated for samples. They provide estimates of the proportions of the dry weights represented by lipid and are higher for regions containing more myelinated axons (Godfrey and Matschinsky, 1976).

Lipid contents of some samples in each region of the cochlear nucleus of mountain beaver MB080895 were measured. After weighing each sample, it was loaded into a glass tube. The lipid was extracted by adding absolute ethanol, removing almost all the ethanol, and adding and removing hexanes twice. The sample was then removed from the tube and re-weighed (Lowry and Passonneau, 1972; Godfrey and Matschinsky, 1976). The lipid-free dry weights of samples correlated closely with their volumes (Godfrey and Matschinsky, 1976) and can be measured more accurately, but, unlike the volumes, they cannot be measured on the same samples as used for the chemical assays.

2.7. Chemical assays

Choline acetyltransferase (ChAT) activity was measured in the same way as in our previous studies (Fonnum, 1969; Godfrey et al., 1977b, 1983: Linker et al., 2018). Samples were added to 400 μl-capacity polyethylene tubes for the assay. In addition, rat brain homogenate (8 μg wet weight) was added to two tubes to serve as a tissue standard, and two tubes containing no tissue served as a reagent blank. Five μl of incubation medium, containing 100 mM sodium phosphate buffer, pH 7.4, 0.3 M sodium chloride, 80 μM [¹⁴C]acetyl coenzyme A, 6 mM choline chloride, 80 μM neostigmine methyl sulfate, 0.1 % (w/v) bovine serum albumin, and 0.1 % (v/v) Triton X-100, were added to all these tubes, and they were incubated for 30 min at 37.5 °C. The incubation was stopped and [¹⁴C]acetylcholine product extracted by adding, with mixing, 43 μl of ice cold 15 mg/ml sodium tetraphenyl boron in 3-heptanone. After centrifugation to separate the aqueous and heptanone phases, a 38 μl aliquot of the heptanone layer was added, with mixing, to a second 400 μl-capacity polyethylene tube containing 100 μl of 0.5 mg/ml sodium tetraphenyl boron in 10 mM sodium phosphate buffer, pH 7.7, to eliminate ¹⁴C-labeled reagent and tissue blanks not related to ChAT activity (Fonnum, 1969; Ross, 1973; Godfrey et al., 1977b). After centrifugation to separate the aqueous and heptanone phases, a 32 μl aliquot of the heptanone layer was transferred to a scintillation vial containing 5 ml of scintillation fluid (1/3 ‘Scintiverse’ and 2/3 ‘Scintilene’) for counting radioactivity by liquid scintillation counting. Recovery of [¹⁴C]acetylcholine product by the extraction was determined to be 100 % by including, in two additional assay tubes without samples, 48 μM [¹⁴C]acetylcholine in the incubation medium instead of [¹⁴C]acetyl coenzyme A.

Upon reviewing our results for ChAT, we became concerned about the reagent blank and homogenate values, in that the blank values were higher and the homogenate activities lower than had been typical of our past experience. We considered the possibility that there had been some deterioration of the [¹⁴C]acetyl coenzyme A substrate even though it had been stored at −80 °C. To check this possibility, we carried out an additional assay with a newly received lot of [¹⁴C]acetyl coenzyme A. This was done on AVCN and rostral granular region samples microdissected from a section 60 μm rostral to the section of MB103195 that had been assayed for ChAT activity with the old lot of [¹⁴C]acetyl coenzyme A, so that the data obtained with the two lots could be directly compared. The reagent blank in the assay with the new lot was considerably lower than in the assay with the old lot and the ChAT activities higher. The activity of the rat brain homogenate was 50% higher with the new lot, giving a value similar to our past experience with rat brain homogenate (Godfrey and Matschinsky, 1981): approximately 125 μmol/kg wet weight/min. The lower activities, as compared to rat brain homogenate or the cholinergic facial nerve system, in AVCN and rostral granular region samples were 3 times as high with the new substrate lot as with the old one. Based on these results, a correction factor was applied to the mountain beaver cochlear nucleus ChAT activities: they were all multiplied by 3. Based on the rat brain homogenate results, the much higher facial nerve system activities were all multiplied by 1.5.

Acetylcholinesterase (AChE) activity was measured in the same way as in our previous studies (Fonnum, 1969; Godfrey et al., 1977b, 1983). Samples were added to 400 μl-capacity polyethylene tubes for the assay. In addition, rat brain homogenate (2.4 μg wet weight) was added to two tubes to serve as a tissue standard, and two blank tubes containing no tissue served as a reagent blank. All these tubes were incubated for 30 min at 37.5 °C in 11.6 μl incubation medium containing 100 mM potassium phosphate buffer, pH 7.4, 0.7 mM [¹⁴C]acetylcholine, 0.01 mM tetraisopropylpyrophosphoramide (iso-OMPA), 0.1 % (w/v) bovine serum albumin, and 0.1 % (v/v) Triton X-100. The enzyme reaction was terminated by adding, with mixing, 1.5 μl of 1 N hydrochloric acid, then the [¹⁴C]acetate product was extracted into 75 μl of ethyl acetate. After centrifugation to separate the aqueous and ethyl acetate layers, a 69 μl aliquot of the ethyl acetate layer was transferred to a scintillation vial containing 5 ml scintillation fluid (1/3 ‘Scintiverse’ and 2/3 ‘Scintilene’) for measurement of radioactivity by liquid scintillation counting. Recovery of [¹⁴C]acetate product by the extraction was determined for each assay by including, in two assay tubes without samples, 0.2 mM [¹⁴C]acetate in the incubation medium instead of [¹⁴C]acetylcholine. The recovery of product was thereby measured as 82-84 %.

Free amino acid concentrations were measured in the same way as in our previous studies (Hill et al., 1979; Ross et al., 1995; Godfrey et al., 2000a, 2015, 2017). Samples were added to 300 μL-capacity glass tubes, to which were then added 15 μL of 50 % methanol, which extracted the amino acids from the tissue samples. By means of a Waters WISP autosampler, 8 μL of the methanol was combined with 8 μL of ortho-phthaldialdehyde solution to derivatize the amino acids and make them fluorescent. The fluid containing the derivatized amino acids was then passed through a C8 column and separated by reversed-phase chromatography and gradient elution. The effluent from the column passed through a fluorescence detector connected to a computer that plotted the chromatograms using MAXIMA software (Waters). The amino acids were identified and their amounts calculated by comparing the chromatogram peaks to those of an amino acid standard solution. The standards, at 4 different concentrations, along with blank tubes, were placed before and after the set of sample tubes. The amino acid identities of the peaks in the standard solution were determined by their retention times: in separate experiments, an extra amount of each individual amino acid was added to the standard solution in a given tube to enlarge the chromatogram peak corresponding to that amino acid.

2.8. Materials

For radiometric assays, [1-¹⁴C]acetyl coenzyme A (NEC-313, lot 2483-192, 59.3 mCi/mmol in most cases and NEC-313, lot 3324035, 44.1 mCi/mmol in one additional assay for mountain beaver and NEC313050, lot 3622034, 60.0 mCi/mmol for pocket gopher), [1-¹⁴C]acetylcholine iodide (NEC-350., 3.9 mCi/mmol) and [1-¹⁴C]acetate, sodium salt (NEC-084, 1.9 mCi/mmol), were obtained from New England Nuclear (Perkin Elmer), Boston, MA, USA. [1-¹⁴C]Acetylcholine chloride (CFA.310, 60.0 mCi/mmol, diluted to 6.16 mCi/mmol with unlabeled acetylcholine chloride), used to measure recovery of product in the ChAT assay, was obtained from Amersham-Searle, Arlington Heights, IL, USA.

Absolute ethanol was obtained from Pharmco, Brookfield, CT, USA. Amino acid Standard H, for calibration of amino acid concentrations in the HPLC assays, was obtained from Pierce. Chromatography columns were obtained from Mac-Mod Analytical, Chadds Ford, PA, USA, or Dupont. Other chemicals were obtained from Sigma or Fisher.

Super Friendly Freeze-it (Freon substitute) was obtained from Curtin-Matheson. Vacuum tubes were obtained from Ace Glass, Inc., Vineland, NJ., USA. Aluminum racks for storing freeze-dried sections, microbalances, and tools for microdissection were made in the lab as needed (Lowry and Passonneau, 1972). Polyethylene microtubes (400 μL-capacity) were obtained from Bio-Rad and glass microtubes (300 μL-capacity) from MicroLiter Analytical Supplies, Inc., Suwanee, GA, or Waters.

Neurolucida software was obtained from MicroBrightfield, (MBF Bioscience, Williston, VT).

2.9. Data analysis

Activities of ChAT were expressed as micromoles of acetylcholine product formed per kg dry weight of tissue per min (μmol/kg dry wt/min), activities of AChE as millimoles of acetylcholine hydrolyzed per kg dry weight of tissue per min (μmol/kg dry wt/min), and amino acid concentrations as millimoles per kg dry weight of tissue (mmol/kg dry wt). Data are presented as distributions in maps of individual experiments and as averages across regions. Averaged data are presented as mean ± standard error of the mean (SEM).

Results

3.1. Electrophysiological recordings

Attempts were made in mountain beavers MB040595 and MB080895 to obtain electrophysiological recordings from the DCN. Animal MB040595 was used primarily to work out the surgical procedures for exposing the DCN. This procedure took many hours and resulted in accidental damage to the left DCN. Several attempts were made to record stimulus-driven activity from the right DCN, but no responses to sound were observed. The surgical exposure was more successful in animal MB080895, and recordings of neural activity were collected from multiple locations on the DCN surface. Examples of response areas from medial and lateral aspects of the DCN are shown in Fig. 1. Four recording sites, clustered together at the medialmost aspect of the DCN, yielded well defined response areas with low or moderate thresholds and sharp peaks, having characteristic frequencies (CFs) between 15 and 18 kHz. Some strong responses were also recorded at more lateral locations; that shown in Fig. 1 had a CF of only 600 Hz and a higher threshold than the more medially located activities. These recordings suggest that there may be a tonotopic organization in the mountain beaver DCN, with responses to lower frequencies more laterally located, as in rat (Kaltenbach and Lazor, 1991) and hamster (Kaltenbach and McCaslin, 1996). However, there were other recording sites where no responses to tonal stimulation were seen.

Figure 1.

Spectral response areas recorded from the surface of the left dorsal cochlear nucleus (DCN) of mountain beaver MB080895. The upper left diagram is a tracing of the DCN surface as visualized during the experiment (lateral is to the left and caudal below), with locations of recordings indicated by numbered dots. The response areas recorded at these locations are shown in the other five diagrams, numbered to correspond to the recording locations. The axes are the same for all 5 plots, and the labels are shown in more detail for position 1. The triangle along the abscissa of each plot points to the apparent characteristic frequency of the response area. SPL, sound pressure level in decibels (dB).

3.2. Sections stained for Nissl substance and for AChE activity

As reported previously (Merzenich et al., 1973), the cochlear nucleus of the mountain beaver, when stained for Nissl substance, differs in appearance from those of other rodents in having very large granular regions and DCN (Fig. 2). These are unusually large relative to animal body weight, whereas the ventral cochlear nucleus (VCN) regions are not (Fig. 3 of Godfrey et al., 2016). Observations on our Nissl-stained sections were consistent with those reported by Merzenich et al. (1973) except that we could not confirm their report of a layer of pyramidal, or fusiform, cells in the most dorsomedial part of the DCN. A granular region was seen to extend dorsomedially along the acoustic striae to that location, although with lower density of somata than in the large more ventrally located granular region, and larger somata were seen among the granule cells in the most dorsomedial location, but these somata did not look clearly different from those scattered within the main body of the DCN. Otherwise, in accordance with the report of Merzenich et al. (1973), no obvious layering was seen within the DCN. The entire region appears as lightly stained somata scattered within a lightly stained neuropil. Measurements of soma long axes, as expressions of soma sizes, gave values ranging from about 5 to 30 μm. Measurements of soma long axes in the granular region adjacent to the DCN gave values of about 5-7 μm, with occasional larger somata with long axis as large as 20 μm. The VCN was divided by the entering auditory nerve root, or interstitial nucleus (IN), into AVCN rostral to it and PVCN caudal to it. Soma long axes in AVCN and PVCN, and occasional somata seen in the IN, ranged from about 5 to 30 μm. With the Nissl stain, it could not be determined whether the very small somata were granule or glial cells.

Figure 2.

Transverse sections through the right cochlear nucleus of mountain beaver MB103195 stained for Nissl substance (top) and acetylcholinesterase (AChE) activity (bottom). The 20 μm-thick sections are numbered consecutively from caudal to rostral. The most caudal section for this cochlear nucleus containing DCN is s25 and the most rostral section containing AVCN s217. The 1 mm scale bar applies to all pictures. The indicated dorsal and lateral directions apply to all sections. Oval outlines in sections stained for AChE activity are artifacts, probably resulting from air bubbles trapped under them. Regional abbreviations are A, anteroventral cochlear nucleus (AVCN); Cb, cerebellum (molecular layer, bordered by much darker granular layer); D, dorsal cochlear nucleus (DCN); G, granular region; I, interstitial nucleus (auditory nerve root); P, posteroventral cochlear nucleus (PVCN); RB, restiform body (inferior cerebellar peduncle); S, acoustic striae (dorsal and intermediate); ST, spinal trigeminal tract; T, trapezoid body (ventral acoustic stria).

Figure 3.

Transverse sections through the right cochlear nucleus and adjacent regions of mountain beaver MB103195 stained for acetylcholinesterase (AChE) activity. The arrow in section 123 (s123) points to a darkly stained bundle of parasympathetic salivatory system fibers as they exit the brain. The arrow in section 133 (s133), within the vestibular nerve root, points to the darkly stained centrifugal labyrinthine bundle, which contains olivocochlear and centrifugal vestibular fibers that terminate, respectively, in cochlear and vestibular parts of the inner ear. Regional abbreviations are F, facial nerve root; G, granular region; I, interstitial nucleus (auditory nerve root); P, posteroventral cochlear nucleus (PVCN); RB, restiform body (inferior cerebellar peduncle); ST, spinal trigeminal tract; T, trapezoid body (ventral acoustic stria). Other details as in Figure 2.

We have previously published a picture of pocket gopher cochlear nucleus sections stained for Nissl substance (Fig. 1 of Godfrey et al., 2016). The DCN of the pocket gopher shows 3 clear layers, unlike in the mountain beaver. The DCN and granular regions are large relative to body weight, although not to the same extent as in the mountain beaver, whereas the VCN is not (Fig. 3 of Godfrey et al., 2016). Measurements of soma long axes in pocket gopher cochlear nucleus regions gave values roughly similar to those in mountain beaver: approximately 5-24 μm in DCN molecular and fusiform soma layers, 5-31 μm in DCN deep layer, 15-29 μm in PVCN and AVCN, and 4-9 μm in granular region.

In mountain beaver sections stained for AChE activity (Fig. 2 bottom), the DCN appears lightly and uniformly stained compared to granular regions, and the PVCN and AVCN have intermediate staining densities, which appear to be associated mostly with neuropil. The staining in granular regions is not homogeneous but includes many dark spots of variable size. These extend throughout the granular regions, including the dorsomedial caudal granular region extension. At the dorsal limit of this dorsomedial extension (Figure 1, section 126), the AChE-positive spots could mark the fusiform soma layer described by Merzenich et al. (1973), since the fusiform soma layer of rodents and cats includes AChE-positive patches, or spots (Osen et al., 1984). The dorsomedially located AChE-positive spots were notable in only a small range of AChE-stained sections of MB103195: 119-126. They could, however, also merely represent the dorsomedial limit of the granular region extension, which, as mentioned above, contains larger somata among the granule cells. The exact sizes of the AChE-positive spots were not easily measured because the AChE stain was not as clear at high magnification as the Nissl stain, but the spots appeared to be usually at least twice as large as the granule cell somata measured in the Nissl-stained sections. The acoustic striae, trapezoid body, and IN show little if any AChE staining. Other than a possible stained fiber bundle within the trapezoid body in section 187 of Fig. 2, the darker staining was on the brain surface and dorsomedial to the trapezoid body. Prominent staining was observed in the fiber tracts corresponding in location to the descriptions of the rat parasympathetic salivatory fibers and centrifugal labyrinthine bundle, which includes olivocochlear and centrifugal vestibular fibers (Shute and Lewis, 1960; Brown and Howlett, 1968, 1971; Godfrey et al., 1984: Linker et al., 2018) (Fig. 3).

3.3. Average regional enzyme activities and amino acid concentrations

Regional averages for dry weight per volume, ChAT and AChE activities, and amino acid concentrations are shown in Table 1. For the mountain beavers, the DCN was divided into superficial, middle, and deep portions based only on locations of samples, since there were no clear histological divisions. Our deep portion corresponds approximately to the portion that has been reported to show staining for myelinated fibers (Fig. 3 of Merzenich et al., 1973). For granular regions, rostral samples, near the AVCN, and caudal samples, near the PVCN, were averaged separately, and the caudal dorsomedial extension samples are not included in the averages.

Table 1.

Chemistry of mountain beaver (MB) and pocket gopher (Pkt Gopher) cochlear nucleus regions, as mean ± SEM (n)

	MB080895	MB103195	Pkt Gopher	MB080895
Dry weight per volume (g/L)				% lipid
DCNm or superficial	222 ± 5 (33)	257 ± 7 (31)	220 ± 8 (9)	41 ± 1 (10)
DCNf or middle	243 ± 4 (30)	269 ± 6 (29)	225 ± 15 (5)	42 ± 1 (7)
DCNd or deep	274 ± 6 (31)	288 ± 6 (37)	336 ± 10 (6)	45 ± 1 (13)
Granular-caudal		263 ± 5 (85)		37 ± 1 (5)
Granular-rostral	251 ± 3 (84)	265 ± 4 (67)	296 ± 14 (7)	29 ± 2 (5)
PVCN	435 ± 17 (5)	411 ± 10 (39)	401 ± 12 (6)	60 ± 1 (5)
IN (auditory nerve root)	392 ± 7 (4)	474 ± 25 (13)	557 ± 3 (3)	70 ± 2 (3)
AVCN	353 ± 6 (30)	375 ± 10 (20)	333 ± 6 (9)	54 ± 1 (6)
Acoustic striae	413 ± 10 (21)	452 ± 18 (26)	398 ± 7 (5)	68 ± 2 (5)
Trapezoid body	524 ± 20 (3)		425 (2)	75 ± 1 (3)
Facial nerve root		479 ± 10 (10)	439 ± 16 (4)
Facial nucleus		306 ± 9 (10)
ChAT (μmol/kg dry wt/min)
DCNm or superficial	14 ± 3 (9)	10 ± 3 (10)	12 ± 2 (4)
DCNf or middle	16 ± 2 (10)	14 ± 7 (8)	17 ± 2 (3)
DCNd or deep	22 ± 3 (10)	16 ± 5 (7)	16 ± 8 (3)
Granular-caudal		39 ± 7 (26)
Granular-rostral	31 ± 3 (47)	57 ± 6 (37)	30 ± 8 (5)
PVCN	100 ± 10 (6)	129 ± 34 (11)	18 ± 6 (3)
IN (auditory nerve root)	23 ± 5 (5)	26± 6 (3)
AVCN	117 ± 18 (15)	155 ± 23 (11)	28 ± 4 (5)
Acoustic striae	39 ± 10 (8)	50 ± 7 (12)	15 (2)
Facial nerve root		1504 ± 68 (5)	1745 ± 68 (4)
Facial nucleus		1626 ± 169 (5)
AChE (μmol/kg dry wt/min)
DCNm or superficial	41 ± 2 (9)	47 ± 2 (10)
DCNf or middle	41 ± 2 (7)	48 ± 3 (8)
DCNd or deep	39 ± 1 (8)	45 ± 4 (9)
Granular-caudal		99 ± 4 (30)
Granular-rostral	69 ± 2 (44)	103 ± 4 (20)
PVCN	34 ± 3 (6)	28 ± 2 (12)
IN (auditory nerve root)	8 ± 3 (4)	14 ± 3 (4)
AVCN	41 ± 4 (10)	43 ± 4 (6)
Acoustic striae	31 ± 7 (6)	25 ± 4 (7)
Facial nerve root		21 ± 3 (5)
Facial nucleus		163 ± 13 (5)
Aspartate (mmol/kg dry wt)
DCNm or superficial	7.0 ± 0.4 (14)	9.1 ± 0.3 (16)	10.4 ± 0.5 (6)
DCNf or middle	8.7 ± 0.4 (14)	9.4 ± 0.4 (11)	12.2 ± 0.4 (3)
DCNd or deep	9.3 ± 0.2 (13)	10.5 ± 0.3 (22)	9.4 ± 0.9 (3)
Granular-caudal		11.0 ± 0.4 (32)
Granular-rostral	9.7 ± 0.3 (54)	12.4 ± 0.4 (27)	9.0 ± 0.8 (4)
PVCN		9.9 ± 0.5 (18)	8.8 ± 0.9 (3)
IN (auditory nerve root)		3.1 ± 0.6 (5)	4.2 (2)
AVCN	10.2 ± 0.6 (14)	11.6 ± 0.6 (7)	9.8 ± 0.4 (4)
Acoustic striae	3.9 ± 0.3 (7)	5.9 ± 0.3 (7)	6.6 ± 0.4 (3)
Trapezoid body			7.7 (1)
Glutamate (mmol/kg dry wt)
DCNm or superficial	41 ±2 (14)	49 ± 0.9 (16)	49 ± 0.9 (6)
DCNf or middle	40 ± 2 (14)	47 ± 0.6 (12)	39 ± 4 (3)
DCNd or deep	38 ± 3 (13)	41 ± 1 (22)	26 ± 2 (3)
Granular-caudal		47 ± 1 (32)
Granular-rostral	39 ± 1 (55)	44 ± 1 (27)	35 ± 5 (4)
PVCN		20 ± 0.8 (18)	23 ± 2 (3)
IN (auditory nerve root)		10 ± 0.8 (5)	13 (2)
AVCN	18 ± 0.7 (14)	20 ± 0.9 (7)	27 ± 0.9 (4)
Acoustic striae	16 ± 2 (7)	13 ± 0.8 (7)	19 ± 0.4 (3)
Trapezoid body			20 (1)
Glutamine (mmol/kg dry wt)
DCNm or superficial	52 ± 3 (14)	42 ± 1 (16)	56 ± 2 (6)
DCNf or middle	54 ± 3 (14)	38 ± 0.9 (12)	45 ± 5 (3)
DCNd or deep	51 ± 4 (13)	38 ± 0.7 (22)	31 ± 3 (3)
Granular-caudal		42 ± 0.9 (32)
Granular-rostral	48 ± 1 (55)	36 ± 0.8 (27)	41 ± 4 (4)
PVCN		23 ± 0.7 (18)	25 ± 2 (3)
IN (auditory nerve root)		15 ± 1 (5)	17 (2)
AVCN	24 ± 0.8 (14)	21 ± 0.6 (7)	33 ± 1 (4)
Acoustic striae	24 ± 2 (7)	19 ± 1 (7)	23 ± 1 (3)
Trapezoid body			22 (1)
Glycine (mmol/kg dry wt)
DCNm or superficial	17.4 ± 0.7 (14)	16.0 ± 0.7 (16)	16.0 ± 1.8 (6)
DCNf or middle	18.9 ± 0.8 (14)	16.5 ± 0.8 (11)	18.5 ± 3.0 (3)
DCNd or deep	18.3 ± 0.7 (13)	16.0 ± 0.6 (22)	16.2 ± 2.2 (3)
Granular-caudal		10.2 ± 0.2 (32)
Granular-rostral	8.8 ± 0.3 (54)	8.8 ± 0.2 (27)	11.9 ± 1.5 (4)
PVCN		7.9 ± 0.5 (18)	10.7 ± 1.4 (3)
IN (auditory nerve root)		3.6 ± 0.3 (5)	6.4 (2)
AVCN	9.1 ± 0.7 (14)	9.1 ± 0.5 (7)	15.0 ± 0.3 (4)
Acoustic striae	6.2 ± 0.5 (7)	6.3 ± 1.0 (7)	11.1 ± 1.1 (3)
Trapezoid body			11.2 (1)
GABA (mmol/kg dry wt)
DCNm or superficial	10.5 ± 0.4 (14)	9.6 ± 0.4 (16)	8.2 ± 0.7 (6)
DCNf or middle	11.4 ± 0.4 (14)	9.8 ± 0.5 (12)	8.4 ± 1.5 (3)
DCNd or deep	10.9 ± 0.5 (13)	10.0 ± 0.4 (22)	6.4 ± 1.2 (3)
Granular-caudal		5.9 ± 0.2 (32)
Granular-rostral	5.3 ± 0.1 (55)	6.0 ± 0.2 (27)	5.5 ± 1.0 (4)
PVCN		5.2 ± 0.4 (18)	2.8 ± 0.5 (3)
IN (auditory nerve root)		1.5 ± 0.3 (5)	0.5 (2)
AVCN	5.8 ± 0.4 (14)	6.4 ± 0.3 (7)	5.5 ± 0.5 (4)
Acoustic striae	3.0 ± 0.3 (7)	4.1 ± 0.9 (7)	3.7 ± 0.8 (3)
Trapezoid body			2.9 (1)
Taurine (mmol/kg dry wt)
DCNm or superficial	18.3 ± 0.8 (14)	17.6 ± 0.5 (16)	19.2 ± 1.1 (6)
DCNf or middle	15.0 ± 0.9 (14)	16.8 ± 0.5 (12)	17.5 ± 2.4 (3)
DCNd or deep	12.3 ± 1.0 (13)	13.5 ± 0.5 (22)	13.8 ± 0.7 (3)
Granular-caudal		20.9 ± 0.9 (32)
Granular-rostral	16.7 ± 0.6 (55)	18.3 ± 0.9 (27)	15.1 ± 2.4 (4)
PVCN		4.9 ± 0.3 (18)	11.2 ± 1.4 (3)
IN (auditory nerve root)		4.4 ± 0.5 (5)	9.8 (2)
AVCN	5.0 ± 0.2 (14)	5.3 ± 0.5 (7)	14.1 ± 0.7 (4)
Acoustic striae	3.7 ± 0.4 (7)	3.7 ± 0.2 (7)	10.4 ± 0.3 (3)
Trapezoid body			9.3 (1)

Although MB080895 was anesthetized for a much longer time than MB103195, there were few major consistent differences between their chemistries, and those seen cannot be attributed to the effect of the anesthesia rather than inter-animal variation. Somewhat lower aspartate concentrations in MB080895 resemble a previous finding for cat cochlear nucleus (Godfrey et al., 1977a), but, unlike the report for cat, there was no evidence for reduction of GABA concentrations in MB080895 that might be attributed to an effect of prolonged anesthetization.

Regional averages for dry weight per volume showed similar trends among the mountain beavers and the pocket gopher, with lower values in superficial DCN and granular regions than in VCN, except that there was a higher ratio between DCN deep layer and molecular layer in the pocket gopher (1.5) than between deep and superficial DCN portions of the mountain beavers (1.2 and 1.1). Higher dry weight per volume values, indicating higher lipid content, were measured in fiber tracts, and higher values were measured in the VCN regions than in granular regions and DCN, except for the deep DCN layer of the pocket gopher. Although there was no good reason to discard it, the unexpectedly low value for the IN in MB080895, based on 4 samples in one section, was probably not representative, since it differs from the values in the other two animals and is inconsistent with the % lipid measurements obtained for 3 of the same 4 samples. A low dry weight per volume value could result from the actual section thickness being less than 20 μm. The % lipid values measured for MB080895 generally correlate closely with the dry weight per volume values (correlation coefficient 0.89 with all data included, or 0.91 with all except the IN data), consistent with previous results (Godfrey and Matschinsky, 1976).

Average ChAT activities in both mountain beaver and pocket gopher cochlear nucleus regions were very low compared to the cholinergic facial nerve system, especially in the DCN. Contrastingly, the activities of AChE were relatively high in the mountain beavers, especially in granular regions. There was no clear evidence of any superficial-to-deep gradient in the mountain beaver DCN.

Aspartate concentrations were overall fairly similar among the mountain beavers and pocket gopher, with a trend toward lower values most superficially in the DCN and lower values in fiber tracts (IN, acoustic striae, and trapezoid body).

Glutamate concentrations were higher in DCN and granular regions than in VCN regions of both mountain beaver and pocket gopher, except for pocket gopher DCN deep layer. The almost 2-fold decreasing gradient from superficial to deep portions of the pocket gopher DCN was not seen in the mountain beavers.

Glutamine concentrations showed similar trends as glutamate, with relatively high values in DCN and granular regions and a superficial-to-deep decreasing gradient in the pocket gopher DCN.

Glycine concentrations were similarly high among the mountain beavers and pocket gopher in the DCN, but the pocket gopher had higher concentrations in the VCN regions and acoustic striae than did the mountain beavers.

GABA concentrations tended to be higher in mountain beaver DCN and PVCN than in pocket gopher. There was no superficial-to-deep gradient of GABA concentration across the mountain beaver DCN.

Taurine showed evidence of superficial-to-deep gradients across the DCN of both mountain beaver and pocket gopher. Taurine concentrations in the VCN and acoustic striae were much higher in the pocket gopher than in the mountain beavers, whereas those in DCN and granular regions were fairly similar.

3.4. Mapped distributions of enzyme activities and amino acid concentrations

Figs. 4, 5,6 illustrate the chemical distributions within the right cochlear nucleus of mountain beaver MB103195, in which the sectioning angle was more favorable than in mountain beaver MB080895 for showing all subregions. Dry weight per volume showed some tendency toward higher values in deeper portions of the DCN and caudal granular regions and was especially high in the caudal dorsomedial granular region extension along the acoustic striae (s104 in Fig. 4). Dry weight per volume values were generally higher in the VCN than in the DCN, and, within the VCN, usually higher in more ventral portions of PVCN and more dorsal portions of AVCN. Especially high values were in parts of IN and acoustic striae.

Figure 4.

Distributions of choline acetyltransferase (ChAT) and acetylcholinesterase (AChE) activities and dry weight per volume in transverse sections through the right cochlear nucleus of mountain beaver MB103195. Section numbers are printed at lower right of each section map. Values are color coded as shown. Pieces lacking data are marked with x but are colored to blend with surrounding pieces. Regional abbreviations are: A, anteroventral cochlear nucleus (AVCN); D, dorsal cochlear nucleus (DCN); G, granular region; I, interstitial nucleus (auditory nerve root); P, posteroventral cochlear nucleus (PVCN); S, acoustic striae (dorsal and intermediate). Directional 1 mm scale bar shows dorsal (D), ventral (V), medial (M), and lateral (L) directions in all sections.

Figure 5.

Distributions of glutamate, aspartate, and glutamine concentrations in transverse sections through the right cochlear nucleus of mountain beaver MB103195. Concentrations (mmol/kg dry wt) are color coded as shown. Other details as in Fig. 4.

Figure 6.

Distributions of GABA, glycine, and taurine concentrations in transverse sections through the right cochlear nucleus of mountain beaver MB103195, the same sections as in Fig. 5. Concentrations (mmol/kg dry wt) are color coded as shown. Other details as in Fig. 4.

Lowest ChAT activities were measured within much of the DCN and dorsolateral portions of the caudal granular region. Within the DCN, there was a tendency for higher ChAT activities in more dorsomedial locations. Within the granular regions, there was a tendency toward higher activities in more medial locations, closer to the VCN, where the highest activities were measured. Activities of AChE were strikingly high throughout granular regions and also high throughout DCN as compared to most VCN locations.

Concentrations of glutamate were similarly high in DCN and granular regions (Fig. 5), with a tendency toward lower values in deeper portions and noticeably lower concentrations in the dorsomedial caudal granular region extension (s104), which correlate inversely with the higher dry weight per volume values there (Fig. 4). Lower glutamate concentrations were measured in the VCN and especially in fiber tracts: acoustic striae and IN. Glutamine concentrations were distributed similarly to glutamate except with less contrast between DCN and VCN. Aspartate concentrations were more uniformly distributed, with a tendency toward higher values in deeper portions of the DCN and deeper portions of granular regions.

Concentrations of GABA were highest in the DCN, with a trend toward highest values most dorsomedially and lowest most ventrolaterally (Fig. 6). Within the VCN, GABA concentrations were lowest in rostral PVCN and IN. Concentrations were rather uniformly distributed in granular regions except for lower values in the dorsomedial caudal granular region extension. The distribution of glycine concentrations closely paralleled that of GABA. Taurine concentrations showed superficial-to-deep gradients in both the DCN and caudal granular regions, lower values in the dorsal part of the rostral granular region, and clearly lower values in VCN, dorsomedial caudal granular region extension, and acoustic striae regions.

Figs. 7–8 illustrate the chemical distributions within the cochlear nucleus of the pocket gopher. Unlike in the mountain beaver, the superficial DCN layers – molecular and fusiform soma - had clearly lower dry weight per volume values than deep DCN, which had values similar to those of AVCN (Fig. 7). High values were measured throughout IN. Lowest ChAT activities were in IN and parts of the DCN. Compared to mountain beaver, there was not much variation of ChAT activity throughout most of the pocket gopher cochlear nucleus.

Figure 7.

Distributions of choline acetyltransferase (ChAT) activity and dry weight per volume in transverse sections through the right cochlear nucleus of the pocket gopher. Section numbers are printed at lower right of each section map. Values are color coded, using the same scales as for the mountain beaver (Fig. 4). Pieces lacking data are marked with x but are colored to blend with surrounding pieces. Regional abbreviations are A, anteroventral cochlear nucleus (AVCN); G, granular region; I, interstitial nucleus (auditory nerve root); P, posteroventral cochlear nucleus (PVCN); S, acoustic striae (dorsal and intermediate); T, trapezoid body; d, dorsal cochlear nucleus (DCN) deep layer; f, DCN fusiform soma layer; m, DCN molecular layer. Directional 1 mm scale bar shows dorsal (D), ventral (V), medial (M), and lateral (L) directions in all sections.

Figure 8.

Distributions of glutamate, aspartate, glutamine, GABA, glycine, and taurine concentrations in transverse sections through the right cochlear nucleus of the pocket gopher. Concentrations (mmol/kg dry wt) are color coded as shown, using the same scale as for the mountain beaver (Figs. 5–6). Other details as in Fig. 7.

Concentrations of glutamate showed consistent differences across the DCN layers, with highest values throughout the molecular layer (Fig. 8). As in the mountain beaver, glutamate concentrations were lower in VCN and fiber tracts. Glutamine concentrations were clearly higher throughout the superficial DCN layers than in the deep layer. Aspartate concentrations were more evenly distributed throughout the pocket gopher cochlear nucleus, as in mountain beaver.

Both GABA and glycine showed trends for higher concentrations dorsomedially and lower ventrolaterally in the DCN, as in mountain beaver, and this occurred in all 3 layers. Concentrations of both GABA and glycine were also consistently higher in all AVCN samples than in all PVCN samples and lowest in IN samples. Highest taurine concentrations were measured in superficial DCN layers, but the values throughout VCN regions were not as much lower as in mountain beaver.

3.5. Comparisons with other rodents and with cat

Figs. 9, 10, 11 show comparisons of average regional values for dry weight per volume, ChAT and AChE activities, and amino acid concentrations of mountain beaver and pocket gopher, with available values for 4 other rodents and for cat.

Figure 9.

Bar graphs of average values for dry weight per volume, choline acetyltransferase (ChAT) activity, and acetylcholinesterase (AChE) activity in cochlear nucleus regions, with facial nucleus and nerve root included for comparison, of mountain beaver and pocket gopher compared to those of 4 other rodents and cat. Data for all animals are plotted on the same scale to facilitate magnitude comparisons. Column label abbreviations are, from left: DCNm, dorsal cochlear nucleus molecular layer, except, for mountain beaver, most superficial part of DCN; DCNf, DCN fusiform soma layer, except, for mountain beaver, middle part of DCN; DCNd, DCN deep layer, except, for mountain beaver, deep part of DCN; Gran-c, caudal granular region, near PVCN (not including dorsomedial extension in mountain beaver); Gran-r, rostral granular region, near AVCN; PVCN, posteroventral cochlear nucleus; IN, interstitial nucleus (auditory nerve root); AVCN, anteroventral cochlear nucleus; AS, acoustic striae (dorsal and intermediate); TB, trapezoid body (ventral acoustic stria); FNR, facial nerve root; Fnuc, facial nucleus. ChAT activities are expressed as percentages of FNR activities in the same animal species to put them on scale and because of considerable variation of FNR activity among species. Average ChAT activities for FNR and Fnuc, in μmol/kg dry wt/min, are, for mountain beaver 1504 and 1626, for pocket gopher 1745 and not measured, for rat 6773 and 4588, for chinchilla 1426 and 1748, for hamster 7361 and 5374, and for cat 1077 and 1050. Dry weight per volume data for rat are from Godfrey and Matschinsky (1981) and Godfrey et al. (2017) and also some unpublished data, averaged when two sources have data; for chinchilla are from Godfrey et al. (2015) plus unpublished data; for hamster are from Godfrey et al. (2008 and 2012), averaged when both have data, plus unpublished data; for mouse are unpublished data; and for cat are from Godfrey et al. (1977b). ChAT data for rat are from Godfrey et al. (1983); for chinchilla are unpublished data; for hamster are from Jin et al. (2006) and Godfrey et al. (2013), averaged when both have data; and for cat are from Godfrey et al. (1977b). AChE data for rat are from Godfrey et al. (1983); for hamster are from Godfrey et al. (2008); and for cat are from Godfrey et al. (1977b).

Figure 10.

Bar graphs of average values for glutamate, aspartate, and glutamine concentrations in cochlear nucleus regions, with facial nucleus and nerve root included for comparison, of mountain beaver and pocket gopher compared to those of 4 other rodents and cat. Rat data are from Godfrey et al. (2000a); chinchilla from Godfrey et al. (2015) and unpublished data for Gran-c and AS; hamster from Godfrey et al. (2008 and 2012), averaged when both have data; mouse from Chen et al. (2009); and cat from Godfrey et al. (1977a) and unpublished data, averaged when both have data. Details as in Fig. 9.

Figure 11.

Bar graphs of average values for GABA, glycine, and taurine concentrations in cochlear nucleus regions, with facial nucleus and nerve root included for comparison, of mountain beaver and pocket gopher compared to those of 4 other rodents and cat. Details as in Figs. 9 and 10.

The increasing superficial-to-deep gradient of dry weight per volume in DCN is less in mountain beaver than in any of the other animals, although only slightly less than in mouse, probably related to its lack of clear DCN layering (Fig. 9). The dry weight per volume values throughout the mountain beaver DCN are similar to those in just the superficial DCN layers (DCNm and DCNf) of the other mammals. Otherwise, the regional trends in dry weight per volume are similar among the animals. All of the other animals have higher cochlear nucleus ChAT activities relative to facial nerve root than mountain beaver DCN or pocket gopher all regions, but the ChAT relative activities in AVCN and PVCN are higher in mountain beaver than in any other animal. Because of its relatively high facial nerve root ChAT activity (listed in Fig. 9 legend), the hamster’s DCN and granular region ChAT activities expressed relative to facial nerve root are not much higher than those of mountain beaver and pocket gopher. Activities of AChE are relatively high in mountain beaver cochlear nucleus, similar in most regions to values in rat, except for higher values in granular regions. The steep superficial-to-deep gradient of AChE activity found in cat is not seen in any of the rodents studied.

A decreasing superficial-to-deep gradient of glutamate concentration was measured in the DCN of all the animals, but the gradient is extremely slight in the mountain beaver (Fig. 10). The regional distribution of glutamine is generally similar to that of glutamate in all animals, whereas that of aspartate is not. The higher glutamate and glutamine concentrations throughout the mountain beaver DCN and granular regions, relative to other portions of the cochlear nucleus, are more prominent than in most other animals. The superficial-to-deep gradient of aspartate concentration in the DCN is mostly opposite to that of glutamate and glutamine, although the gradient is very slight in mountain beaver. Concentrations of glutamine in cochlear nucleus regions were generally somewhat higher in mountain beaver and pocket gopher than in other animals, except chinchilla.

The distributions of GABA and glycine among cochlear nucleus regions are fairly similar, except that the relatively lower average GABA concentration in the deep portion of the DCN, as compared to superficial layers, seen in all animals except mountain beaver, was not usually seen for glycine. With the one exception for GABA in mountain beaver, the distributions of both GABA and glycine are quite similar among the animals. Taurine concentration shows a decreasing superficial-to-deep gradient across the DCN for all the animals, including mountain beaver, but the gradient is steeper in rat, hamster, and mouse than in the other animals. The distributions and magnitudes of average taurine concentrations across cochlear nucleus regions show more variation among the animals than do those of any of the other amino acids. All the amino acids tend to have lower concentrations in fiber tracts (IN, TB, and FNR, sometimes AS) than in other regions, inversely related to the higher dry weight per volume of the fiber tracts, although the differences from VCN regions are generally less for taurine.

3.6. Correlations between GABA and glycine distributions

Although GABA and glycine are not closely related metabolically, their distributions seemed quite similar in both mountain beaver and pocket gopher cochlear nuclei. Plots of the GABA concentrations of individual samples vs. their glycine concentrations revealed high correlations (Fig. 12), higher than those between concentrations of metabolically closely related amino acids, except for glutamate vs. glutamine in the pocket gopher (Figs. 13 and 14). The correlations for the metabolically closely related amino acids were higher in pocket gopher than in mountain beaver.

Figure 12.

Scatter plots of GABA concentration vs. glycine concentration for mountain beaver and pocket gopher samples, with linear regression line included. Mountain beaver samples are from both MB80895 and MB103195. Regional locations of samples are indicated by colored letters: acoustic striae, black S; anteroventral cochlear nucleus, red A; for mountain beaver, DCN deep portion, dark blue D, middle portion, medium blue D, and superficial portion, light blue D; for pocket gopher, DCN deep layer, dark blue Dd, fusiform soma layer, medium blue Df, and molecular layer, light blue Dm; granular region, caudal, dark green G, and rostral, light green G; interstitial nucleus, pink I; posteroventral cochlear nucleus, brownish red P; trapezoid body, olive green T. The correlation coefficient r is listed for each plot. Correlations are highly statistically significant (P < 0.001).

Figure 13.

Scatter plots of concentrations of amino acids that are closely related metabolically as precursor and product for samples of mountain beaver cochlear nucleus. The samples are the same ones as represented in Fig. 12, and details are as in Fig. 12. Because of the large numbers of samples, all correlations are statistically significant (P < 0.001).

Figure 14.

Scatter plots of concentrations of amino acids that are closely related metabolically as precursor and product for samples of pocket gopher cochlear nucleus. The samples are the same ones as represented in Fig. 12, and details are as in Fig. 12. All correlations are statistically significant (P < 0.001).

To check the generality of the high GABA vs. glycine correlations in mountain beaver and pocket gopher, we evaluated GABA vs. glycine correlations in the cochlear nuclei of other animals and in other central auditory system regions (Table 2). Correlations were generally high in the cochlear nucleus and usually not as high, although still positive, at higher levels of the auditory system. The high correlations in the medial geniculate were notable, even though glycine concentrations are relatively low there compared to the cochlear nucleus (Godfrey et al., 2012, 2017).

Table 2.

Correlation coefficients for GABA vs. glycine in central auditory regions^a

Animal	CN	SOC	IC	MG	AudCtx
Albino rat	+0.83	+0.76
F344BN rat	+0.77	+0.38	+0.32	+0.71	+0.21
Chinchilla	+0.91
Hamster	+0.79	+0.48	+0.52	+0.72	+0.64
Cat	+0.77

^aAbbreviations are: AudCtx, auditory cortex; CN, cochlear nucleus; F344BN, Fischer 344 x Brown Norway; IC, inferior colliculus; MG, medial geniculate; SOC, superior olivary complex. Data from our previous studies (Godfrey et al., 2000, 2012, 2015, 2017) and some unpublished data were evaluated.

Discussion

4.1. Chemical correlations with unique features of mountain beaver cochlear nucleus

A potential weakness of our study is that the data derive from only two mountain beavers and one pocket gopher. However, our past experience with the microchemical measurements has indicated that, although there are differences among individual animals, they are minor rather than major. This is supported by the comparison of the data for the two mountain beavers in Table 1.

Most of the chemistries measured showed less superficial-to-deep gradients across the DCN in mountain beaver than in pocket gopher or other animals, consistent with lack of clear layers in the mountain beaver DCN. The chemical results for the DCN, together with the huge size of granular regions in the mountain beaver cochlear nucleus, are consistent with the concept that the mountain beaver DCN may be equivalent to just the superficial portion of the DCN (molecular layer and probably also neurons of the fusiform soma layer) of other animals (Godfrey et al., 2016), which receives extensive innervation from granule cells (Mugnaini et al., 1980; Dunn et al., 1996; Godfrey et al., 1997; Oertel and Young, 2004; Diño and Mugnaini, 2008). Although the dry weight per volume values in the entire mountain beaver DCN resemble those for just the superficial part of the DCN in the other mammals, the tendency for slightly higher dry weight per volume values in deeper DCN portions is consistent with the previous observation of more heavily myelinated fibers seen there (Fig. 3 of Merzenich et al., 1973), although their density does not appear to be high. We measured generally somewhat higher dry weight per volume values in the deeper portion of the caudal granular region than in more superficial portions (Figure 4), which would be consistent with the myelinated fibers traversing this granular region portion to reach the DCN.

4.2. Acetylcholine in mountain beaver and pocket gopher cochlear nuclei

The relatively low activities of ChAT, which is a more reliable marker than AChE for the presence of cholinergic elements (Martinez-Murillo and Rodrigo, 1995), in the pocket gopher cochlear nucleus and in the mountain beaver DCN suggest that they receive relatively little centrifugal cholinergic innervation. Relative to centrifugal cholinergic innervation via the trapezoid body route (Godfrey et al., 1987b; Sherriff and Henderson, 1994; Linker et al., 2018), the AChE stain in the mountain beaver has a similar density in the trapezoid body and spinal trigeminal tract (Fig. 3), whereas in rat the density of AChE stain is higher in the trapezoid body than in the spinal trigeminal tract (Fig. 1 of Linker et al., 2018). Based on ChAT activity, the centrifugal cholinergic innervation to the mountain beaver appears to innervate the VCN much more than the DCN or granular regions, unlike in the other animals studied. Unlike ChAT activity, AChE activity is relatively high in the mountain beaver DCN and granular regions, resulting in higher AChE/ChAT ratios than in most other animals. The only other cochlear nucleus location with such high ratios is the superficial DCN of the cat, especially the molecular layer (Godfrey and Matschinsky, 1976; Godfrey et al., 1977b). The cat DCN molecular layer differs from that of the rodents that have been studied both in its very high measured AChE activity and dark staining for AChE activity (Osen and Roth, 1969; Godfrey and Matschinsky, 1976; Osen et al., 1984). The AChE-positive spots in mountain beaver granular regions might be mostly associated with the glomeruli mentioned by Merzenich et al. (1973) and reported for other rodents and cats (McDonald and Rasmussen, 1971; Osen et al., 1984). These glomeruli have been reported in various mammals to represent complex synapses involving mossy fibers from various sources and dendrites of granule cells and unipolar brush cells (Dunn et al., 1996; Mugnaini et al., 1997; Alibardi, 2004). Since AChE is predominantly associated with membranes (Boell and Nachmansohn, 1940; Lewis and Shute, 1966; McDonald and Rasmussen, 1971; Godfrey et al., 1984), the glomeruli, with their abundance of membranes, might contribute considerably to measured AChE activity. In the rat cerebellar vermis, which contains some very high AChE/ChAT ratios, the numerical density of AChE-positive spots, or aggregates, corresponding to cerebellar glomeruli, has been found to correlate much more closely with measured AChE activity than with ChAT activity (Ross et al., 1983). Since AChE tends to be associated with the postsynaptic membranes at cholinergic synapses and in cholinoceptive as well as cholinergic neurons (Lewis and Shute, 1966; Silver, 1967; McDonald and Rasmussen, 1971; Godfrey et al., 1977b; Ross et al., 1983), it is possible that the high AChE/ChAT ratios in the mountain beaver DCN and granular regions are associated with many synapses formed by a few thin cholinergic fibers, but this would need to be tested with methods having higher spatial resolution. There is also evidence for AChE lacking any clear association with cholinergic neurotransmission (Silver, 1967).

4.3. Amino acids in mountain beaver and pocket gopher cochlear nuclei

Despite different habitats and behavioral adaptations of mountain beavers and pocket gophers compared to those of the other mammals studied, there were no major differences in their amino acid distributions. This cannot necessarily be taken to mean that the roles of the amino acids in neurotransmission are similar across the different mammals because of the involvements of amino acids in other functions besides neurotransmission. The most notable differences between mountain beaver and/or pocket gopher and other mammals were the less prominent superficial-to-deep gradients of amino acid concentrations across the mountain beaver DCN, correlating with its lack of obvious layering, and some differences in magnitudes of concentrations. Glutamine concentrations in mountain beaver and pocket gopher tended to be higher than in the other animals and GABA and glycine lower. The lower GABA and glycine concentrations could be associated with lower concentrations in cochlear nucleus interneurons (Mugnaini, 1985; Osen et al., 1990; Oertel and Wickesberg, 1993; Godfrey et al., 1997; Apostolides and Trussell, 2013), less of these inhibitory interneurons, or less innervation from extrinsic GABA and glycine neurons (Davies, 1981; Wenthold, 1987; Godfrey et al., 1988; Potashner et al., 1993; Saint Marie et al., 1993). The presence of a superficial-to-deep gradient of taurine concentration across the mountain beaver DCN cannot be considered as evidence for DCN layering since it also occurred in the caudal granular region (Fig. 6).

4.4. GABA vs. glycine correlations

Strong evidence has been reported for co-function of GABA and glycine at some synapses in neonatal rat spinal cord slices (Jonas et al., 1998). The high correlations between GABA and glycine concentrations in the cochlear nucleus are consistent with evidence for colocalization and co-function of these two neurotransmitters in cochlear nucleus regions (Osen et al., 1990; Kolston et al., 1992; Altschuler et al., 1993; Juiz et al., 1996; Moore et al., 1996; Kemmer and Vater, 2001; Apostolides and Trussel, 2013). The very high correlation coefficients in mountain beaver, pocket gopher, and chinchilla suggest that the relative amount of colocalization of GABA and glycine may be particularly high in the cochlear nuclei of these rodents. In the mountain beaver, the correlations between GABA and glycine concentrations are especially high in the DCN (Fig. 12). In cat, guinea pig, baboon, and horseshoe bat, extensive colocalization of GABA and glycine immunoreactivities has been shown for neurons in the superficial DCN, especially cartwheel cells (Osen et al., 1990; Kolston et al., 1992; Moore et al., 1996; Kemmer and Vater, 2001), correlating with the evidence that the mountain beaver DCN corresponds to the superficial part of the DCN of other mammals. The extent to which GABA and glycine may co-function at particular synapses in the mountain beaver and pocket gopher will of course depend on the nature of the transmitter receptors at those synapses (Dugué et al., 2005; Apostolides and Trussell, 2013). In mouse cochlear nucleus slices, the glycine receptors have been found to predominate at cartwheel cell synapses (Golding and Oertel, 1996; Apostolides and Trussell, 2013), but this cannot be assumed to hold true for all mammals.

High correlations between GABA and glycine concentrations are not at all typical of brain regions. They are generally less in higher centers of the auditory system. In the central auditory system of Fischer 344 x Brown Norway (F344BN) rats, there is a clear trend for higher glycine, relative to GABA, in lower auditory centers and higher GABA in higher centers (Godfrey et al., 2017). In other central nervous system regions, a high correlation between GABA and glycine concentrations has been found in rat retina (correlation coefficient +0.91, Ross et al., 1989), but low or negative correlations in other regions: F344BN rat hippocampus stratum oriens (+0.21, unpublished data from the study of Godfrey et al., 2017), albino rat vestibular nuclei (−0.31, Li et al., 1994), albino rat olfactory bulb (−0.37, unpublished data from the study of Godfrey et al., 1980), overall rat brain sensory regions (−0.34, Ross et al., 1995), rabbit lumbar spinal cord (−0.50, Berger et al., 1977), and rabbit hippocampal formation (−0.84, Berger et al., 1977).

4.5. Functional implications

The relatively poor hearing of the pocket gopher has been measured (Heffner and Heffner, 1990). Although the hearing of the mountain beaver has not been measured with behavioral testing, previous observations suggest that it has limited hearing ability but well developed somatosensory and olfactory senses (Carraway and Verts, 1993).

The physiological measurements available previous to the present study suggest that, although VCN activity in the mountain beaver resembles that in other mammals, DCN activity does not (Merzenich et al., 1973). Merzenich et al. (1973) reported predominantly responses to slow pressure variations along with some responses to very low frequency sounds at high intensities. Our electrophysiological recordings demonstrate that neurons in the mountain beaver DCN display strong frequency-tuned responses to sound. These responses were strongest in the medial part of the DCN, where neurons were tuned to the highest frequencies (15-18 kHz), but strong responses were also observed more toward the lateral direction (approximately midway between the medial and lateral extremities of the DCN), where low-frequency (600 Hz)-tuned neurons with somewhat higher thresholds were recorded. The distribution of the high-frequency-tuned neurons medially and low-frequency-tuned neurons laterally is consistent with the gradient of frequencies observed in other rodent species we have studied, such as hamster (Kaltenbach and McCaslin, 1996; Meleca et al., 1997) and rat (Kaltenbach and Lazor, 1991), which declines from high to low frequencies along the medial-lateral axis of the DCN. This suggests that the DCN of the mountain beaver is both strongly responsive to sound and tonotopically organized. However, the presence of neurons tuned to 600 Hz at a locus approximately midway between the medial and lateral extremities of the DCN suggests that the tonotopic range extends to much lower frequencies than we have observed at corresponding locations in hamsters and rats. In the mountain beaver, sites showing responses to sound were interspersed among neighboring sites showing no responses to sound. This absence of responsiveness could explain the unresponsive neurons reported by Merzenich et al. (1973). Overall, our results overlap those of Merzenich et al. (1973) in suggesting that the mountain beaver DCN is specialized for low frequency hearing, although our data indicate that this specialization encompasses a significant part of the audiometric range of other mammalian species and that neurons are more responsive to sound than observed previously, at least between 600 Hz and 18 kHz. We did not examine the responses of the mountain beaver DCN to slow variations of pressure as in the Merzenich study.

Despite differences in lifestyle among the animals compared in our study, the distributions of amino acid concentrations show few notable differences among them. This suggests that amino acid distributions in mammalian cochlear nuclei may not vary greatly among species, so that what has been measured so far in the various mammals may be representative of what would be measured in the human cochlear nucleus.

The distributions of ChAT activity in cochlear nucleus regions vary greatly among the mammals studied. This suggests that acetylcholine neurotransmission in the cochlear nucleus, which appears to be predominantly associated with its centrifugal innervation (Godfrey et al., 1983, 1987a,b, 1990; Sherriff and Henderson, 1994; Kraus and Illing, 2004; Meidinger et al., 2006; Mellott et al., 2011) and affect detection of specific sounds within noisy backgrounds (Pickles and Comis, 1973; Pickles, 1976), may vary greatly among animal species, as has been suggested previously (Moore, 1988; Godfrey et al., 1997). Although VCN ChAT activity is much higher in mountain beaver than in pocket gopher, their DCN ChAT activities are similarly low, suggesting that centrifugal cholinergic pathways to the DCN are not important for their fairly similar lifestyles. In other mammals, cholinergic innervation activates granule cells, which then affect the activity of cartwheel and fusiform cells of the DCN via glutamatergic synapses (Oliver et al., 1983; Osen et al., 1984, 1995; Chen et al., 1994, 1999; Godfrey et al., 1997; Köszeghy et al., 2012; He et al., 2014; Stefanescu and Shore, 2017). This superficial DCN circuitry enables information from other sensory systems, including somatosensory and vestibular (Itoh et al., 1987; Weinberg and Rustioni, 1987; Burian and Gstoettner, 1988; Kevetter and Perachio, 1989; Zhao et al., 1995; Wright and Ryugo, 1996; Godfrey et al., 1997), to affect the processing of auditory information (Koehler and Shore, 2013; Wu et al., 2015). Although not organized into a layer, some of the larger somata in the mountain beaver DCN may correspond to fusiform cells, which send information forward in the auditory system, to the inferior colliculus (Osen, 1988; Berrebi and Mugnaini, 1991; Oertel and Young, 2004). Thus, in the mountain beaver, centrifugal cholinergic influences on multisensory integration in the DCN appear to be reduced, whereas centrifugal cholinergic influences on VCN activity, which may counterbalance cholinergic olivocochlear effects on the cochlea (Kraus and Illing, 2004; Meidinger et al., 2006), by an olivocochlear system that appears to be well represented in mountain beaver (Fig. 3), appear to be increased. In the pocket gopher, both centrifugal cholinergic influences appear to be reduced.

Previous evidence suggests that the structures of the DCN and granular regions vary more among mammals than that of the VCN (Moore, 1980; Godfrey et al., 2016). How differences in DCN and granular region structure may relate to function are not clear. For animals living mostly underground, there tends to often be increased size of cochlear nucleus granular regions, as in mountain beaver, pocket gopher, and mole, but the unlayered DCN structure of the mountain beaver differs from the layered DCN structure of pocket gopher and mole (Kudo et al., 1990; Godfrey et al., 2016).

For animals spending much of their time underground, one might expect that sensitivity to ground vibrations may be important, and physiological results for the mountain beaver suggest that its DCN is important for sensing such vibrations (Merzenich et al., 1973). There is now considerable evidence that the cochlear nucleus granular regions and DCN of mammals receive innervation from somatosensory centers, both trigeminal and cuneate (Itoh et al., 1987; Weinberg and Rustioni, 1987; Young et al., 1995, 1997; Wright and Ryugo, 1996; Shore et al., 2000; Zhou and Shore, 2004; Haenggeli et al., 2005; Zeng et al., 2011), which affect the function of DCN neurons, partly via glutamatergic neurotransmission (Zhou and Shore, 2004; Kanold et al., 2011; Shore, 2011; Zeng et al., 2011; Koehler and Shore, 2013). The high glutamate and glutamine concentrations in the mountain beaver DCN and granular regions may be partly associated with major somatosensory innervation at glutamatergic synapses (Zeng et al., 2011). It would appear that the mountain beaver would be a good animal model for studying somatosensory effects on cochlear nucleus activity.

Highlights.

The mountain beaver dorsal cochlear nucleus (DCN) shows no evidence of layers

Some responses to tones were recorded in mountain beaver DCN

Mountain beaver DCN and granular regions have low choline acetyltransferase activity

Mountain beaver ventral cochlear nucleus has high choline acetyltransferase activity

Amino acid levels in mountain beaver cochlear nucleus resemble those of other rodents

Abbreviations in text

AChE

acetylcholinesterase

AVCN

anteroventral cochlear nucleus

CF

characteristic frequency

ChAT

choline acetyltransferase

DCN

dorsal cochlear nucleus

F344BN

Fischer 344 x Brown Norway

GABA

γ-aminobutyric acid, or γ-aminobutyrate

HPLC

high performance liquid chromatography

IN

interstitial nucleus (auditory nerve root)

PVCN

posteroventral cochlear nucleus

VCN

ventral cochlear nucleus