Cholinergic neuronal connections in the central auditory system have been known to play essential roles not only in basic neuromodulatory processes (Habbicht and Vater, 1996; Yigit et al., 2003) but also in behavioral arousal or attention (Metherate and Ashe, 1993; Hsieh et al., 2000), plastic response (Ji et al., 2001, 2005; Ji and Suga, 2008), and cognitive functions. In Alzheimer disease, marked degenerative changes are frequently found throughout major sites of the ascending auditory pathway including the primary auditory cortex, the ventral nucleus of the medial geniculate body, and the central nucleus of the inferior colliculus (Ohm and Braak, 1989; Sinha et al., 1993).
At the earlier stages of Alzheimer disease, O'Mahony et al. (1994) found a significant dysfunction in the midlatency response, indicating an impairment of the ascending cholinergic system that originates in the pedunculopontine tegmental nucleus and mostly terminates in the thalamus. Buchwald et al. (1991) also observed a close relation between the auditory and ascending cholinergic systems by pharmacological manipulation with a cholinergic agonist and antagonist.
Morphologic evaluation of the cholinergic fibers and receptors responsible for the function of the central auditory system has been little accomplished except for the cochlear nucleus, where the distribution density of cholinergic fiber–receptor systems has been reported to be regionally variable (Frostholm and Rotter, 1986; Chen et al., 1995; Yao and Godfrey, 1995, 1999; Yao et al., 1996; Jin et al., 2005; Jin and Godfrey, 2006). Tsutsumi et al. (2007) studied the distribution of cholinergic components in the precerebellar nuclei by using vesicular acetylcholine transporter (VAChT) immunohistochemistry and provided evidence that mesopontine cholinergic neurons negatively regulate neocortico-ponto-cerebellar projections at the level of pontine nuclei. Because VAChT is synthesized in major cholinergic neurons and localized to synaptic vesicles for acetylcholine transport (Erickson et al., 1994; Roghani et al., 1994, 1996; Schafer et al., 1994; Usdin et al., 1995), this molecule is used as a cholinergic marker to delineate terminals and preterminal axons (Schafer et al., 1995, 1998; Gilmor et al., 1996; Weihe et al., 1996; Arvidsson et al., 1997; Ichikawa et al., 1997; Roghani et al., 1998).
The aim of this study was to delineate a relative abundance of the fiber–receptor system in specific regions of the central auditory system by means of histochemical detection of cholinergic fibers and receptors. VAChT immunohistochemistry was used to examine cholinergic fibers. Based on the specific nucleotide sequence for m2 and m3 subtypes, digoxigenin-labeled cRNA probes were designed. With these riboprobes, in situ hybridization was conducted to investigate the distribution of their mRNAs. The regional densities of immunoreactive fibers and hybridization-positive neurons were analyzed in the distribution field of the auditory system.
Of the five subtypes of muscarinic receptors (m1–m5), the m2 and m3 subtypes were selected to examine in this study, because these two subtypes have been shown to be abundantly expressed in the cerebral cortex and brainstem regions including some nuclei of the ascending auditory pathway (Buckley et al., 1988; Weiner et al., 1990; Levey et al., 1991, 1994). The m2 and m3 subtypes show different properties in coupling to G-protein and cellular effects (Felder, 1995; Murthy and Makhlouf, 1997; Frazier et al., 2008). The m2 subtypes prefer to be located presynaptic to cholinergic cells and coupled to inhibition of cAMP elevation, whereas m3 subtypes are dominantly located postsynaptic on cholinoceptive cells in forebrain and couple to activation of phosphatidyl inositol turnover.
MATERIALS AND METHODS
All experiments were performed in compliance with the National Institute of Health Guide for the Care and Use of Laboratory Animals (NIH Publications No. 80-23, revised 1996), and all procedures were approved by the Institutional Animal Care and Use Committee. Ten-week-old male C57BL6/J mice (CLEA Japan, Tokyo, Japan) in which Preyer reflex could readily be elicited by hand clap were used in this study.
The brain was sliced into 40-μm coronal sections with Cryotome according to Kase et al. (2007). These sections were immunostained for VAChT with essentially the same method as described previously (Tsutsumi et al., 2007).
The animals were deeply anesthetized with sodium pentobarbital (100 mg kg−1, i.p.). The brain and spinal cord were dissected. Total RNA was extracted from the tissues by the guanidium thiocyanate method. A 10-μg aliquot of the RNA was reverse-transcribed with 5 pmol of oligo(dT)30 as a primer, 1 mM dNTP, and 100 U of ReverTra Ace (Toyobo, Tokyo, Japan) in 20 μL of the reaction mix for 1 hr at 42°C. One microliter aliquot of the RT product was amplified by PCR. The reaction mixture (20 μL) contained 1 μM forward and reverse primers, 200 μM dNTP, and 0.5 U of Ex Taq Hot Start Version (Takara, Otsu, Japan). The PCR conditions were as follows: pretreatment at 94°C for 2 min; 35 cycles of PCR (denaturation, 0.5 min at 94°C; annealing, 0.5 min at 68°C; and extension, 0.5 min at 72°C). The primers mM2 (forward) 5′-atatcccgggcgagcaagagcagaataaag-3′ (30-mer) and mM2 (reverse) 5′-acaggatagccaagattgtcctggtcac-3′ (28-mer) were used to amplify the 556-bp fragment, corresponding to the nucleotide 625–1180 from the translation initiation site of mouse m2 muscarinic receptor cDNA (NM_203491; 1401 bp). The primers mM3 (forward) 5′-gtagcagctatgagctacaacagcaagg-3′ (28-mer) and mM3 (reverse) 5′-cttctggtcttgagagcaaacctcttagcc-3′ (30-mer) were used to amplify the 550-bp fragment, corresponding to the nucleotide 866–1415 from the translation initiation site of mouse m3 muscarinic receptor cDNA (NM_033269; 3168 bp). The PCR-amplified fragments corresponded to the third cytoplasmic loop manifesting a low level of amino acid homology among m1–m5 muscarinic receptors derived from rodent and several other species.
These fragments were subcloned into pGEM-T Easy plasmid (Promega, Madison, WI) and sequenced with ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction Kits (Applied Biosystems, Tokyo, Japan).
The constructs were linearized and subjected to in vitro transcription. The reaction was carried out for 2 hr at 37°C in 20 μL of the transcription buffer, pH 8.0, containing 1 μg of template DNA, 10 mM dithiothreitol, 1 mM GTP, 1 mM ATP, 1 mM CTP, 0.65 mM UTP, 0.35 mM digoxigenin-11-UTP (Roche Diagnostics, Mannheim, Germany), 20 U of ribonuclease inhibitor (Toyobo), and 40 U of T7 RNA polymerase (Stratagene, La Jolla, CA). Then, the DNA template was digested with DNase I for 15 min at 37°C. Riboprobes were precipitated with ethanol and LiCl, resuspended in diethylpyrocarbonate (DEPC)-treated distilled water, and adjusted to a concentration of 200 ng μL−1.
In Situ Hybridization
The method followed those developed by Braissant and Wahli (1998) and was modified as described in detail previously (Trifonov et al., 2009). All solutions used for in situ hybridization were treated with 0.02% DEPC. Mice were perfused with 25 mL of 0.9% saline and 50 mL of a fixative containing 4% paraformaldehyde in 0.1 M sodium phosphate buffer (PB; pH 7.4). The brain and spinal cord were saturated with 30% sucrose in 0.1 M PB overnight at 4°C. They were frozen on a sliding microtome and cut into coronal sections of 40-μm thickness. The tissue sections were stored in cryoprotection buffer (30% sucrose, 30% ethylene glycol, and 50 mM PB) at −20°C. Free-floating sections were rinsed in 0.1% DEPC-activated 0.1 M PB-0.9% saline (PBS, pH 7.4) twice for 15 min. They were equilibrated in 5× standard saline citrate (0.15 M sodium chloride and 0.015 M sodium citrate, pH 7.0; SSC) for 15 min and incubated in hybridization buffer (50% deionized formamide, 40 μg mL−1 salmon sperm DNA, 5× SSC) at 55°C for 2 hr. Hybridization was carried out overnight at 55°C. The hybridization mixture contained 1 ng μL−1 digoxigenin (Dig)-labeled riboprobe in the hybridization buffer. Hybridized sections were then rinsed with 2× SSC at room temperature for 30 min, with 2× SSC at 65°C for 60 min, and with 0.1× SSC at 65°C twice for 30 min. They were equilibrated with Dig buffer 1 (0.1 M Tris hydrochloride, 0.15 M sodium chloride, pH 7.5) for 5 min. Subsequently, the tissue sections were reacted with alkaline phosphatase-labeled anti-Dig antibody (Roche Diagnostics; diluted 1:5,000) in Dig buffer 1 containing 1% blocking reagent for 2 hr. Sections were rinsed in Dig buffer 1 twice for 15 min. For alkaline phosphatase histochemistry, the tissue sections were equilibrated with Dig buffer 2 (0.1 M Tris hydrochloride, 0.1 M sodium chloride, 0.05 M magnesium chloride, pH 9.5) for 5 min and incubated overnight in a light-protected condition with a reaction mixture containing 450 μg mL−1 nitroblue tetrazolium and 175 μg mL−1 5-bromo-4-chloro-3-indolylphosphate in Dig buffer 2 at 37°C. The enzyme reaction was terminated by rinsing sections in TE buffer, pH 8.0, for 15 min. Finally, the tissue sections were thoroughly washed in 0.9% saline, mounted onto gelatin-coated glass slides, air dried, equilibrated with 50%, 70%, 95%, 100%, and 100% ethanol, clarified in xylene three times, and coverslipped with Canada balsam. The hybridized sections were examined with Nikon E800M light microscope under brightfield illumination. For the studies of cytoarchitecture, some adjacent sections were stained with 0.1% cresyl violet. The nomenclature of the nuclei and related fiber systems and the approximate locations of primary auditory cortex were according to Paxinos and Franklin (2001).
Levels of VAChT-immunostained material and digoxigenin labeling were quantified by digitized image analysis using Image J software ver. 1.39. Gray levels were converted to optical densities by using standard internal curves. The background signal was determined by the values obtained from the area of the optic tract and subtracted from values obtained in all brain sites harboring hybridized cells. Three to six subregions were measured per site and the mean optical density determined. Data are expressed as the mean of the values determined from three animals and calculated as signal intensity per unit area.
RESULTS AND DISCUSSION
The tissue sections hybridized with riboprobes for m2 and m3 muscarinic receptor subtypes showed distinct signals with differential expression patterns (Fig. 1). Neurons expressing m2 subtypes were predominantly found in the lateral habenular nucleus, superior colliculus, parabigeminal nucleus, and pedunculopontine tegmental nucleus (Fig. 1a,c). The pedunculopontine tegmental nucleus is regarded as a robust cholinergic cell group that supplies axons to the thalamus and many brainstem structures (see Tsutsumi et al., 2007 for further refs.). A vast majority of neurons expressing m2 receptors were also present in the pontine nuclei (Figs. 1c, 2g) and in motor nuclei of the cranial nerves, including the facial nucleus (Fig. 2h). The pontine nuclei have been shown to receive abundant VAChT-immunoreactive fibers traveling from the pedunculopontine tegmental nucleus (Tsutsumi et al., 2007). On the other hand, the hybridization-positive neurons preferring m3 subtypes could be seen principally in the cerebral cortex and hippocampus (Fig. 1b,d). In most of the neocortical areas, m3 receptors were localized to cortical layers II and III (Fig. 2i) and, to a lesser degree, to layers V and VI. The m2 receptors were expressed much more weakly in cortical layers IV and V (Fig. 1a,c). In the hippocampus, m3 receptors were present most prominently in CA1-3 (Fig. 1b), whereas m2 receptors were marked in the dentate gyrus (Fig. 1a).
The m2 and m3 subtypes were expressed in the central auditory system including the cochlear nuclei, inferior colliculus, nucleus of the brachium of the inferior colliculus, ventral nucleus of the medial geniculate body, and primary auditory cortex (Fig. 2d–f,i). These sites were subjected to densitometric analysis (Fig. 3) because considerable amounts of VAChT-immunoreactive varicose fibers and terminal-like structures were detectable in some of these sites (Fig. 2a–c).
VAChT-Immunoreactive Varicose Fibers
The densest accumulation of VAChT-immunoreactive varicose fibers was seen in the cochlear nuclei, especially on the margin of the anteroventral cochlear nucleus (Fig. 2c). In the anteroventral cochlear nucleus and in the posteroventral cochlear nucleus, however, a vast majority of the nuclear territories were virtually free from such varicose fibers (Fig. 3a). Thus, the VAChT immunoreactivity in the entire cochlear nucleus was presumed to be less intense than that indicated by densitometry in the cochlear nuclear areas containing the densest accumulation of VAChT-immunoreactive fibers. Immunolabeled fibers and terminal-like structures were present at moderate levels in the primary auditory cortex, the ventral nucleus of the medial geniculate body (Fig. 2a), and the nucleus of the brachium of the inferior colliculus (Fig. 2b). VAChT-immunoreactive varicose fibers were seen at low levels in the remaining sites.
m2 Hybridization-Positive Cells
The highest density of distribution of neurons expressing m2 subtypes was observed in the dorsal cochlear nucleus in accordance with earlier observations in the cochlear nucleus (Yao et al., 1996; Jin and Godfrey, 2006) (Fig. 3b). The distribution density of cells with m2 hybridization signal was moderate in the primary auditory cortex, the ventral nucleus of the medial geniculate body (Fig. 2d), the nucleus of the brachium of the inferior colliculus (Fig. 2e), and the dorsal and external cortex of the inferior colliculus (Fig. 2f). In the remaining sites, the distribution density of neurons expressing m2 subtype was low.
m3 Hybridization-Positive Cells
Many neurons expressing m3 muscarinic receptor mRNA were seen in the primary auditory cortex and mainly localized to the layers II and III (Fig. 2i). The neurons expressing m3 receptors were only seen at low levels in all the other sites of the central auditory system (Fig. 3c).
In view of the fact that the density of distribution of cholinergic terminal-like structures and neurons with m2 hybridization signal was moderate in layers IV and V of the primary auditory cortex, the ventral nucleus of the medial geniculate body, and the nucleus of brachium of the inferior colliculus, it was assumed that the signaling in these regions might be implemented by m2-mediated cholinergic transmission. In the inferior colliculus and the dorsal cochlear nucleus, the distribution density of cells with m2 hybridization signal was moderate and cholinergic axon terminals were small in number. Thus, it was supposed that the m2-mediated signaling in these regions might not be dominant.
The nucleus of the brachium of the inferior colliculus contains auditory neurons subserving the coding for auditory space in association with neurons located in the deep layers of the superior colliculus (Kudo et al., 1984; Schnupp and King, 1997; King et al., 1998; Doubell et al., 2000; Nodal et al, 2005). Nevertheless, the chemical architecture of the nucleus of the brachium of the inferior colliculus has been little elucidated. Our findings revealed the relative abundance of m2-receptor subtype and cholinergic terminal-like elements in the nucleus of the brachium of the inferior colliculus (Fig. 2e) and m2 subtype-expressing neurons in the deep layer of superior colliculus (Fig. 1c), thus implicating an important contribution of m2 muscarinic receptors for the coding of auditory space information.
The cerebral cortex and thalamus have been shown to express multiple types of muscarinic receptors including m2 and m3 subtypes (Buckley et al., 1988; Levey et al., 1991, 1994). Our results suggest that the m3 muscarinic receptor is the predominant subtype in the cortex, expressed in the neurons of the superficial cortical layers. Neurons of the auditory cortex and medial geniculate nucleus are vulnerable to neurodegeneration characteristic to Alzheimer disease (Sinha et al., 1993). Important question is whether the neurons expressing m3 subtypes might be one of the most vulnerable types. In normal cortices, stimulation of the neurons of the nucleus basalis facilitates auditory thalamocortical synaptic transmission (Metherate and Ashe, 1993; Hsieh et al., 2000). Activation of m3 receptors in the superficial layers of the cerebral cortex might be involved in the process of the synaptic transmission in the auditory thalamocortical system.
The authors thank Fumio Yamashita and Tetsuji Yamamoto for technical assistance and Yuki Okada for expert secretarial work. Stefan Trifonov is supported by the Japanese Government Monbukagakusho (MEXT) Scholarship Program.