SEARCH

SEARCH BY CITATION

Keywords:

  • syntaxin-1;
  • SNAP-25;
  • VAMP-2;
  • trigeminal ganglion;
  • sensory nerve endings

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

Distribution of three soluble N-ethylmaleimide-sensitive fusion protein attachment protein receptor (SNARE) proteins, syntaxin-1, synaptosomal-associated protein of 25 kDa (SNAP-25), and vesicle-associated membrane protein-2 (VAMP-2), was examined in dental pulp and periodontal ligament of the rat incisor. In the trigeminal ganglion, syntaxin-1 and SNAP-25 immunoreactivity was predominately detected in medium- to large-sized neurons. Most syntaxin-1 immunoreactive neurons expressed SNAP-25. In contrast, VAMP-2 was localized in small- to medium-sized neurons and in slender-shaped cells surrounding SNAP-25-immunopositive neurons. When the inferior alveolar nerve, one of the mandibular nerve branches innervating the dental pulp and periodontal ligament, was ligated, SNARE proteins accumulated at the site proximal to the ligation. In the incisor dental pulp, all nerve fibers displayed immunoreactivity for syntaxin-1, SNAP-25, and VAMP-2. In the periodontal ligament of the incisor, almost all nerve fibers displayed both syntaxin-1 and SNAP-25 immunoreactivity, but lacked VAMP-2 immunoreactivity. SNAP-25 protein expression was localized around the vesicle membranes at the axon terminal of the periodontal mechanoreceptors. These present data suggest that these three SNARE proteins are synthesized at the trigeminal ganglion, transported centrally and peripherally, and expressed in sensory endings where apparent synapses are not present. Because those proteins participate in docking and exocytosis of synapse vesicles in the central nervous system, they might also contribute to vesicle exocytosis at receptive fields where apparent synapses are not present. Anat Rec, 2010. © 2010 Wiley-Liss, Inc.

Chemical neurotransmitters are typically stored and transported in synaptic vesicles. Nerve stimulation activation of voltage-gated Ca2+ channels in the plasma membrane of nerve terminals results in a Ca2+ influx. Increased levels of cytosolic Ca2+ trigger fusion of synaptic vesicle proteins to the presynaptic membrane, resulting in neurotransmitter release at the synapse active zone, where docking and fusion of synaptic vesicles to the presynaptic plasma membrane occur. One of the proposed mechanisms of neurotransmitter release from synaptic vesicles to the presynaptic membrane is the soluble N-ethylmaleimide-sensitive fusion protein attachment protein receptor (SNARE) hypothesis (Rothman, 1994). The protein–protein interaction between SNARE proteins on the vesicle membrane (v-SNARE) and those on the target membrane (t-SNARE) is thought to promote vesicle docking, priming, and membrane fusion. In neuronal cells, vesicle-associated membrane protein-2 (VAMP-2), a v-SNARE, binds specifically to a heterodimeric complex of syntaxin-1 and synaptosomal-associated protein of 25 kDa (SNAP-25). The plasma membrane acts as a target and, therefore, the resulting heterotrimeric complex is the minimal requirement for synaptic vesicle exocytosis (Bennett et al., 1992; Söllner et al., 1993; Südhof, 1995). Previous immunohistochemical studies have demonstrated the presence of these SNARE proteins in brain regions where synapses are present (Elferink et al., 1989; Inoue and Akagawa, 1993; Sesack and Snyder, 1995; Boschert et al., 1996; Chen et al., 1999). These SNARE proteins have also been shown to be expressed in the peripheral nervous system, such as in the retina (Morgans et al., 1996; von Kriegstein et al., 1999), cochlear hair cell synapse (Safieddine and Wenthold, 1999), gustatory system (Yang et al., 2000, 2004, 2007; Ueda et al., 2006), and where synapses are present between axon terminals and specific sensory cells. For example, in the gustatory system, type III taste bud cells have synaptic contact with gustatory nerve terminals and display immunoreactivity for syntaxin-1, SNAP-25, and VAMP-2 (Yang et al., 2000, 2004, 2007; Ueda et al., 2006). This indicates that type III cell synapses use SNARE mechanisms for neurotransmitter release.

Primary sensory neurons are pseudounipolar neurons; their cell bodies extend bilateral processes. The central processes make synaptic contact with secondary neurons, whereas peripheral processes function as sensory receptors without apparent synapses at the axon terminals. Previous immunohistochemical studies have shown the presence of SNARE proteins in cell bodies of the dorsal root and trigeminal ganglia, as well as in the spinal and medulla dorsal horn, where spinal and trigeminal primary neuron central processes terminate (Li et al., 1996a, b; Aguado et al., 1999). These evidences led to the hypothesis that these SNARE proteins are synthesized in cell bodies and are peripherally and centrally transported. Little is known, however, about SNARE protein expression in peripheral axons of primary afferent neurons. In trigeminal primary afferents, Norlin et al. (1999) reported SNAP-25 expression in axonal elements of intradentinal nerve fibers, where no apparent synapse is present. This study was investigated the presence of syntaxin-1, SNAP-25, and VAMP-2 expression in dental pulp and periodontal ligament of the rat incisor. Preliminary results from this study have been published in abstract form (Honma et al., 2004, 2005).

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

All animal experimental protocols were reviewed and approved by the Intramural Animal Use and Care Committee of Osaka University Graduate School of Dentistry prior to onset of experiments. A total of 13 male, Sprague Dawley, adult rats, weighing 200–250 g, were purchased from Nihon Doubutsu (Osaka, Japan).

RT-PCR Analysis

Three animals were sacrificed by an overdose injection of chloral hydrate (600 mg/kg, i.p.), and the trigeminal ganglia were removed. Total RNA was isolated using the SV Total RNA Isolation System (Promega, Madison, WI), according to the manufacture's instructions. First-strand cDNA synthesis was performed by reverse transcription in a final volume of 2 μg total RNA. Primer sequences for each PCR are shown in Table 1.

Table 1. The PCR primers encoding SNARE proteins
Target proteinAccession no.Primer (Forward) Primer (Reverse)
Syntaxin-1AM95734GGATCATCATGGACTCCAGC
TATCCAAAGATGCCCCCGA
Syntaxin-1BM95735ACGACATCAAAATGGACTCGCAG
GGCCTACAAGCCCAGTGTC
SNAP-25AAB003991AGGACGCAGACATGCGTAATGAACTGGAGG
TTGGTTGATATGGTTCATGCCTTCTTCGACACGA
SNAP-25BAB003992AGGACGCAGACATGCGTAATGAACTGGAGG
TTGGTTGATATGGTTCATGCCTTCTTCGACACGA
VAMP-2AM24105CACCGTCCCGCCTGC
CTTCTTAGGCAGGGCAGACTC

PCR amplification was performed using the following conditions; 94°C for 30 sec, 58°C for 30 sec, 72°C for 30 sec, and these steps were repeated for 35 cycles, followed by a final elongation step at 72°C for 7 min. Amplification products were analyzed by 1.5% agarose gels electrophoresis and visualized with ethidium bromide under UV illumination.

Surgery and Tissue Preparation

Five animals were used for the ligation experiment (three animals for experiment and two animals for control). Under chloral hydrate anesthesia (400 mg/kg, i.p), an incision was made in the buccal skin, and the buccal surface of the mandibular bone was exposed. Following removal of a small amount of bone covering the mandibular canal through the use of a dental drill, the inferior alveolar nerve (IAN) was exposed and tightly ligated with a silk thread. The ligated IAN was returned to the mandibular canal. All wounds were then sutured, and no postoperative treatment, such as antibiotics, was administered to the rats. The animals were sacrificed 24 hr after ligation.

Five unoperated rats and five rats for IAN ligation experiment were deeply anesthetized with chloral hydrate (600 mg/kg, i.p.) and transcardially perfused with 0.02 M phosphate-buffered saline (PBS; pH 7.4), followed by 4% paraformaldehyde/0.05% glutaraldehyde in 0.1 M phosphate buffer (PB, pH 7.2). The trigeminal ganglia, mandibles, and maxillae were resected from the unoperated, adult rats and further fixed in 4% paraformaldehyde/0.1 M PB for 2–3 days. The IAN was carefully dissected out from the mandibular canal of the animals used for IAN ligation experiment. Dental pulp from the lower incisor was carefully expired from the untreated animals. The mandibles were decalcified using 7.5% ethylene diamine tetraacetic acid (EDTA) for 4–6 weeks at 4°C under gentle agitation. The decalcifying solution was replaced at least twice a week. All specimens were soaked in 20% sucrose/PBS for cryoprotection. Mandibles were sectioned at a thickness of 30 μm (for light microscopic analysis) or 50 μm (for electron microscopic analysis) with a cryostat, collected in PBS, and treated as floating sections. The trigeminal ganglia were sectioned at a thickness of 20 μm and treated as floating sections. The IAN and dental pulp were cut longitudinally with a thickness of 20 μm with a cryostat and thaw-mounted onto poly-L-lysine-coated glass slides.

Immunohistochemistry

Sections were incubated for 30 min at room temperature in PBS containing 0.03% H2O2 to inactivate endogenous peroxidase activity. Following preincubation for 30 min with PBS containing 3% normal horse serum (Vector Laboratories, Burlingame, CA) and 1% bovine serum albumin (BSA; Sigma, St. Louis, MO), sections were incubated with one of the following monoclonal primary antibodies for 16–18 hr at room temperature: anti-syntaxin-1 (clone HPC-1, 1:5,000, Sigma), anti-SNAP-25 (clone SP12, 1:5,000, Sigma), and anti-VAMP-2 (1:3,000, clone 69.1, Synapse System, Göttingen, Germany). Following several rinses in PBS, sections were incubated with biotinylated horse anti-mouse IgG preabsorbed with rat IgG (1:100, Vector) and subsequently with ABC complex (Vector). The antigen–antibody binding sites were visualized by incubation with 0.05 M Tris-HCl buffer containing 0.08% diaminobenzidine and 0.003% H2O2. Reactions were enhanced with 0.1% nickel ammonium sulfate. Following the diaminobenzidine reaction, immunostained sections were mounted onto gelatin-coated glass slides and slightly counterstained with methyl green. They were then dehydrated through an ascending series of ethanol, cleared with xylene, and coverslipped with Permount (Fisher Scientific, New Jersey, NJ). IAN sections were incubated with one of the monoclonal antibodies against SNARE protein (1:1,000) for 16–18 hr at room temperature. Following several rinses in PBS, sections were incubated with Cy3-conjugated anti-mouse IgG preabsorbed with rat IgG (1:300, Molecular Probes, Eugene, OR) for 90 min. The sections were then coverslipped with Vectashield (Vector) and viewed with a fluorescent microscope (AxioSkop 2, Carl Zeiss, Germany).

For electron microscopic analysis, maxilla sections from untreated animals were immunostained according to the ABC method as mentioned earlier. However, the nickel ammonium sulfate intensification step was omitted. The immunostained sections were further fixed with 1% OsO4 reduced with 1.5% potassium ferrocyanate for 60 min at 4°C. They were then dehydrated through graded series of ethanol, transferred to propylene oxide, and embedded in Durcapan resin. Ultrathin sections were prepared with a diamond knife and examined with a transmission electron microscope (HT-7000; Hitachi, Tokyo, Japan), following brief staining with uranyl acetate and lead citrate.

For double immunofluorescence, sections were incubated either with a mixture of monoclonal anti-syntaxin-1 (1:1,000) and polyclonal anti-SNAP-25 (1:1,000, Chemicon International, Temecula, CA) or monoclonal anti-VAMP-2 (1:1,000) and polyclonal anti-SNAP-25 (1:1,000). Sections of dental pulp and periodontal ligament were incubated with a mixture of polyclonal anti-PGP 9.5 (1:1,000; Ultraclone, UK) and one of monoclonal antibodies against SNARE proteins at the dilution of 1:1,000 for 16–18 hr at room temperature. Sections were then incubated with a mixture of Alexa Fluor488-conjugated anti-rabbit IgG (1:1,000, Molecular Probes) and Cy3-conjugated anti-mouse IgG preabsorbed with rat IgG (1:300, Molecular Probes) for 90 min at room temperature. Following coverslipping with Vectashield (Vector), the immunostained sections were viewed with a fluorescent microscope (AxioSkop 2; Carl Zeiss), and images were captured using a CCD camera and software (AxioCom Ver. 4.0; Carl Zeiss). The images were transferred to Adobe Photoshop (ver. 5.0)

Immunohistochemical controls were performed by omission of primary antibody, secondary antibody, or ABC complex. They did not display any positive immunoreactions.

Quantification of Cell Size

Cell-size distribution of SNARE-immunopositive neuronal profiles from the trigeminal ganglia was performed. The cross-sectional areas of immunoreactive neurons were measured using Win ROOF software (ver. 5.0, Mitani, Osaka, Japan) from randomly selected images at a magnification of 20×. The areas were at least 50 μm to avoid duplicate counting. Comparison of measured cell size means was assessed using the Mann-Whitney U test. P-values < 0.05 were considered statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

SNARE mRNA and Protein Expression in Trigeminal Ganglia

RT-PCR for mRNA of each SNARE protein isoform of the trigeminal ganglion revealed high levels of syntain-1B, SNAP-25B, and VAMP-2 mRNA expression, but weak expression of syntain-1A and SNAP-25A (Fig. 1).

thumbnail image

Figure 1. RT-PCR analysis of mRNA expression of SNARE protein isoforms in the normal trigeminal ganglion. SNAP-25B, syntaxin 1B, and VAMP-2 mRNA are strongly expressed in the trigeminal ganglion, whereas SNAP-25A and syntaxin 1A mRNA are weakly expressed.

Download figure to PowerPoint

SNARE protein expression was examined by immunohistochemistry. Syntaxin-1 and SNAP-25 immunoreactivity was observed, respectively, in the cytoplasm of trigeminal ganglion neurons. The nuclei were devoid of expression. Approximately 86.8% (751/865) and 93.4% (640/685) trigeminal ganglion neurons expressed syntaxin-1 and SNAP-25 immunoreactivity, respectively. Colocalization of syntaxin-1 and SNAP-25 was present in many ganglion neurons. Some axons within the ganglion were also immunopositive for syntaxin-1 and SNAP-25 (Fig. 2A,B). The size range (and mean ± SD) of syntaxin-1 and SNAP-25 immunoreactive trigeminal neurons (from 20 randomly selected double-labeled images) was 232.5–2156.4 μm2 (802.2 ± 343.3 μm2; N = 366) for syntaxin-1 and 240.0–2156.4 μm2 (792.9 ± 341.6 μm2; N = 390) for SNAP-25. Approximately 98.9% (362/366) of syntaxin-1 immunoreactive neurons also displayed SNAP-25 immunoreactivity, and ∼92.8% (362/390) of SNAP-25 immunoreactive neurons exhibited immunoreactivity for syntaxin-1. The size range (and mean ± SD) of neurons double labeled with syntaxin-1 and SNAP-25 was 240.0–156.4 μm2 (807.3 ± 341.5 μm2; N = 362). There was no statistical difference in cross-sectional area distribution between each group (Fig. 2D; Table 2).

thumbnail image

Figure 2. (A–C) Double labeling for SNAP-25 (A), syntaxin-1 (B), and merged image in the trigeminal ganglion (C). Some medium- to large-sized neurons express both SNAP-25 and syntaxin-1. Scale bar = 50 μm. (D) Frequency histogram of cross-sectional areas of neurons expressing SNAP-25 (closed circles), syntaxin-1 (open circles), and SNA-25 and syntaxin-1 (gray circles) in the trigeminal ganglion (D).

Download figure to PowerPoint

Table 2. Frequency of cross-sectional areas of SNAP-25, syntaxin-1, and SNAP-25/syntain-1 neurons in trigeminal ganglion
Range (μm2)SNAP-25Syntaxin-1SNAP-25/Syntaxin-1
Number%1%2Number%1%2Number%1
  1. %1: Percentage of neurons per all immunoreactive neurons.

  2. %2: Percentage of double-labeled neurons.

−30051.380.061.666.741.1
300–60012732.686.611230.698.211030.4
600–90013735.194.913035.6100.013035.8
900–1,2007318.798.67219.7100.07219.9
1,200–1,500328.293.8308.2100.0308.3
1,500–1,800102.6100.0102.7100.0102.8
1,800–61.5100.061.6100.061.7
Total39010092.836610098.9362100
Mean792.9  802.2  807.3 
SD341.6  343.3  341.5 

VAMP-2 immunoreactivity was detected in cytoplasm of ∼37.8% (319/844) trigeminal ganglion neurons. Double labeling of SNAP-25 and VAMP-2 immunoreactivity showed partial colocalization of these two proteins in trigeminal ganglion neurons (Fig. 3A,B). The size range (and mean ± SD) of SNAP-25- and VAMP-2-immunopositive neurons was 170.9–2015.1 μm2 (789.3 ± 327.5 μm2; N = 510) for SNAP-25 and 152.8–1833.3 μm2 (582.5 ± 291.0 μm2; N = 369) for VAMP-2. There was significant difference between the two groups. Approximately 48.2% (246/510) of SNAP-25 immunoreactive neurons exhibited VAMP-2 immunoreactivity, and ∼66.7% (246/369) of VAMP-2 immunoreactive neurons showed immunoreactivity for SNAP-25. The size range (and mean ± SD) of neurons expressing both SNAP-25 and VAMP-2 was 170.8–1833.3 μm2 (674.6 ± 300.6 μm2; N = 246) (Fig. 3D, Table 3). The size range (and mean ± SD) of SNAP-25-positive neurons without VAMP-2 expression, and VAMP-2-positive neurons without SNAP-25 expression, was 217.9–2015.1 μm2 (896.1 ± 315.7 μm2; N = 264) and 152.8–1004.7 μm2 (398.1 ± 150.4 μm2; N = 123), respectively.

thumbnail image

Figure 3. (A–C) Double labeling for SNAP-25 (A), VAMP-2 (B), and merged image (C) in the trigeminal ganglion. Some SNAP-25 immunoreactive neurons display neuronal VAMP-2 expression, and other SNAP-25-positive neurons (asterisk in A, C) are surrounded by VAMP-2-positive slender-shaped cells (arrow in B, C; A–C). Scale bar = 50 μm. (D) Frequency histogram of the cross-sectional neuronal areas with expression of SNAP-25 (closed circles), VAMP-2 (open circles), and SNAP-25 and VAMP-2 (grey circles). (E) Frequency histogram of the cross-sectional areas of SNAP-25-positive neurons (closed circles), VAMP-2-positive slender-shaped cells (open circles), and SNAP-25-positive neurons surrounded by VAMP-2-positive slender-shaped cells (gray circle).

Download figure to PowerPoint

Table 3. Frequency of cross-sectional areas of SNAP-25, VAMP-2, and SNAP-25/VAMP-2 neurons in trigeminal ganglion
Range (μm2)SNAP-25VAMP-2SNAP-25/VAMP-2
Number%1%2Number%1%2Number%1
  1. %1: Percentage of neurons per all immunoreactive neurons.

  2. %2: Percentage of double-labeled neurons.

−300152.986.74813.027.1135.3
300–60014328.069.917346.857.810141.1
600–90017935.016.49726.385.68333.7
900–1,20011222.030.4359.597.13413.8
1,200–1,500469.024.4113.0100.0114.5
1,500–1,800132.530.851.480.041.6
1,800–30.6000000
Total51010048.236910066.7246100
Mean789.3  582.5  674.6 
SD327.5  291.0  300.6 

In addition to neuronal VAMP-2 expression, VAMP-2 was also expressed in the slender-shaped cells encircling the neurons. The neurons that were surrounded with VAMP-2 immunoreactive slender-shaped cells were 240.6–1851.2 μm2, with a mean ± SD of 897.2 ± 314.1 μm2 (N = 295). Approximately 95.3% (281/295) of VAMP-2-positive slender-shaped cells surrounded the SNAP-25 immunoreactive neurons, and SNAP-25 neurons with VAMP-2 slender-shaped cells were within the size range of 240.6–1851.2 μm2 (mean ± SD = 914.2 ± 311.2 μm2; N = 281). Approximately 33.1% (98/296) of the VAMP-2-positive slender-shaped cells encircled VAMP-2 immunoreactive neurons, and the cross-sectional areas of VAMP-2 neurons with VAMP-2 slender-shaped cells were fallen in the range (and mean ± SD) of 338.2–1833.3 μm2 (818.9 ± 317.7 μm2; N = 98). Approximately 29.1% (86/296) of neurons surrounded by VAMP-2-positive slender-shaped cells showed immunoreactivity for SNAP-25 and VAMP-2, and those double-labeled neurons with VAMP-2 slender-shaped cells had a size range (and mean ± SD) of 338.2–1833.3 μm2 (859.5 ± 316.8 μm2; N = 86). Only two neurons that were surrounded by VAMP-2-positive slender-shaped cells did not express SNAP-25 and VAMP-2 immunoreactivity. (Fig. 3C,E; Table 4).

Table 4. Frequency of cross-sectional areas of SNAP-25 neurons, VAMP-2 satellite cells, and SNAP-25 neurons encircled with VAMP-2 satellite cells in trigeminal ganglion
Range (μm2)SNAP-25 neuronsVAMP-2 satellite cellsSNAP-25/VAMP-2
Number%1%2Number%1%2Number%1
  1. %1: Percentage of neurons per all immunoreactive neurons.

  2. %2: Percentage of double-labeled neurons.

−300152.986.710.3100.010.4
300–60014328.069.95016.984.04214.9
600–90017935.016.411037.494.510437.0
900–1,20011222.030.48227.8100.08229.2
1,200–1,500469.024.43913.2100.03913.9
1,500–1,800132.530.8113.7100.0113.9
1,800–30.6020.7100.020.7
Total51010048.229510095.3281100
Mean789.3  897.2  914.2 
SD327.5  314.1  311.2 

SNARE Proteins at the Site of Ligation

To confirm whether synaptic proteins, which are synthesized in trigeminal ganglion cell bodies, are transported to the periphery, the IAN was ligated within the mandibular canal. In animals with a normal IAN, many axons had immunoreactivity for syntaxin-1 (Fig. 4A), and few axons exhibited immunoreactivity for SNAP-25 and VAMP-2 (Fig. 4B,C). Following ligation, syntaxin-1, SNAP-25, and VAMP-2 protein expression accumulated at the site proximal to nerve ligation (Fig. 4A′–C′).

thumbnail image

Figure 4. Syntaxin-1 (A, A′), SNAP-25 (B, B′), and VAMP-2 (C, C′) expression in normal inferior alveolar nerve (A–C) and site proximal to the ligation (A′–C′). In normal IAN, many axons express immunoreactivity for syntaxin-1 (A), and few axons show immunoreactivity for SNAP-25 (B) and VAMP-2 (C). Following ligation, all proteins are accumulated at the site proximal to the ligation (arrow in A–C). Scale bar = 50 μm.

Download figure to PowerPoint

SNARE Proteins in Dental Pulp

The majority of dental pulp nerve fibers from the incisor were smooth-surfaced nerve fibers (Fig. 5A,D,G). Most nerve fibers were positive for syntaxin-1 (Fig. 5B,C), SNAP-25 (Fig. 5E,F), and VAMP-2 (Fig. 5H,I) protein expression; but few PGP 9.5 immunoreactive nerve fibers did not show immunoreactivity for SNARE proteins.

thumbnail image

Figure 5. Double labeling for PGP 9.5 (A, D, G), syntaxin-1 (B), SNAP-25 (E), VAMP-2 (H), and merged images (C, F, I) in the dental pulp of the rat incisor. The majority of PGP 9.5 expressing nerve fibers exhibit expression for syntaxin-1, SNAP-25, and VAMP-2 as well. The images in each row were taken from the same section. Scale bar in I represents 50 μm and is applicable to all images.

Download figure to PowerPoint

SNARE Proteins in Periodontal Ligament

PGP 9.5 immunoreactivity was detected in the alveolus-related part (ARP) of the lingual periodontal ligament. They ramified extensively in the ARP, forming Ruffini endings. Nerve bundles entering the periodontal ligament were also immunopositive for PGP 9.5 (Figs. 6A, 8A). Double labeling with PGP 9.5 and syntaxin-1 showed that most PGP 9.5 immunoreactive nerve fibers also expressed syntaxin-1 in the periodontal ligament, with the exception of nerve bundles entering the periodontal ligament (Fig. 6A,B).

thumbnail image

Figure 6. Paired images of PGP 9.5 (A) and syntaxin-1 (B) expression in the lingual periodontal ligament of the adult rat incisor. (A) PGP 9.5 immunoreactive nerve fibers form Ruffini endings at the alveolus-related part (ARP) of the ligament. TRP, tooth-related part. An asterisk indicates the shear zone, the border between ARP and TRP. (B) Most syntaxin-1-positive nerve fibers exhibit PGP 9.5 expression. Some PGP 9.5-positive nerve fibers within nerve bundles (arrow in A, B) do not exhibit syntaxin-1 expression. Scale bar = 100 μm.

Download figure to PowerPoint

SNAP-25 immunoreactive nerve fibers arborized at the ARP, forming Ruffini endings (Fig. 7A). Under electron microscopy, immunoreactive products were observed as electron-dense deposits. Expression was detected in the terminal axoplasm, which were filled with numerous mitochondria. However, the Schwann sheath did not express SNAP-25 protein. Between the immunonegative Schwann sheath space, SNAP-25 immunoreactive axoplasm extended to the surrounding periodontal tissues (Fig. 7B). Under higher magnification, vesicle-like structures were observed within the axon terminal, and immunoreactivity for SNAP-25 was detected surrounding the vesicle membranes (Fig. 7C).

thumbnail image

Figure 7. SNAP-25 immunoreactivity in the lingual periodontal ligament of the adult rat incisor. (A) SNAP-25-positive nerve fibers form Ruffini endings at the ARP of the lingual ligament of the rat incisor. (B) Ultrastructure of SNAP-25-positive axon terminal. Immunoreactive axon terminal (AT) is filled with numerous mitochondria and is surrounded with immunonegative Schwann sheaths (SS). Arrows indicate axonal spines. (C) SNAP-25 immunoreactivity is concentrated around the membrane of vesicle-like structures (arrowheads). Arrow indicates the axonal spine. Scale bars = 100 μm (A), 2 μm (B), and 0.5 μm (C).

Download figure to PowerPoint

VAMP-2 immunoreactivity was not detected in the periodontal ligament even apparent Ruffini endings were recognized (Fig. 8A,B)

thumbnail image

Figure 8. Paired images of PGP 9.5 (A) and VAMP-2 (B) expression in the lingual periodontal ligament of the adult rat incisor. (A) PGP 9.5 immunoreactive nerve fibers showed dendritic appearance at the alveolus-related part (ARP) of the ligament. TRP, tooth-related part. An asterisk indicates the shear zone, the border between ARP and TRP. (B) No immunoreactivity for VAMP-2 is observed. Scale bar = 100 μm.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

The present immunohistochemical analyses revealed that three SNARE proteins, syntaxin-1, SNAP-25, and VAMP-2, were expressed in the cytoplasm of trigeminal neurons. In the dental pulp of the lower incisor, almost all nerve fibers displayed immunoreactivity for syntaxin-1, SNAP-25, and VAMP-2. Moreover, in the periodontal ligament of the incisor, almost all nerve fibers displayed both syntaxin-1 and SNAP-25 immunoreactivity, but did not express VAMP-2 protein.

SNARE Proteins in Trigeminal Ganglia

Previous biochemical and immunohistochemical studies have shown differential distribution patterns of different SNARE isoforms (Elferink et al., 1989; Inoue and Akagawa, 1993; Sesack and Snyder, 1995; Boschert et al., 1996; Chen et al., 1999). Because the primary antibodies used in this study did not recognize these specific isoforms, RT-PCR analysis was performed. Among the many syntaxin subunits, syntaxin-1 has been shown to be predominantly expressed in neuronal tissues. Syntaxin-1 has two isoforms, syntaxin-1A and syntaxin-1B. Aguado et al. (1999) reported differential distribution of syntaxin-1A and syntaxin-1B in the peripheral nervous system. Although both syntaxin-1A and syntaxin-1B are localized in dorsal root ganglia, syntaxin-1A is localized in the boundary of the neuronal cell bodies, and syntaxin-1B is located within the neuronal cell bodies. The present RT-PCR analyses confirmed mRNA expression of both syntaxin-1A and syntaxin-1B in the trigeminal ganglia, although mRNA expression of syntaxin-1B was more predominate than syntaxin-1A. Previous studies have shown that SNAP-25 isoform expression is developmentally regulated. SNAP-25A is expressed early in development, whereas SNAP-25B is expressed during adulthood (Bark et al., 1995; Boschert et al., 1996). These results demonstrated stronger mRNA expression of SNAP-25B than SNAP-25A in adult trigeminal ganglia, which was consistent with previous reports (Bark et al., 1995; Boschert et al., 1996).

The cell-size analysis of cell bodies in trigeminal ganglia demonstrated that both SNAP-25 and syntaxin-1 immunoreactive cells displayed similar distribution patterns in cross-sectional areas. Furthermore, double labeling showed that most syntaxin-1 immunoreactive neurons also expressed SNAP-25 protein. In contrast, the cross-sectional areas of VAMP-2-positive neurons were smaller than from the SNAP-25-positive cells. Double labeling also demonstrated that nearly 50% of VAMP-2-positive neurons were also SNAP-25-positive—primarily the larger neurons. More than 90% of SNAP-25-positive neurons coexpressed syntaxin-1, while nearly 50% of SNAP-25-positive neurons also expressed syntaxin-1 and VAMP-2. Harper and Lawson (1985) reported a linear correlation between conduction velocity and neuronal diameter. If this theory was applied to the trigeminal ganglia, both SNAP-25 and syntaxin-1 immunoreactive neurons would be related to Aβ and/or Aδ fibers, and VAMP-2 would be closely related to Aδ and/or C fibers.

In addition to VAMP-2 expression in neurons, VAMP-2 was also expressed in slender-shaped cells surrounding neurons. Moreover, >95% of those cells surrounded the SNAP-25-positive neurons. According to morphology, the slender-shaped cells appeared to be satellite cells, although ultrastructural analysis was not performed. The relationship between syntaxin-1 and VAMP-2 was not analyzed. However, it is safe to say that VAMP-2-positive satellite cells also surrounded the syntaxin-1-positive neurons, because >90% of SNAP-25-positive neurons also expressed syntaxin-1. Expression of VAMP-2 in satellite cells has not been previously reported. In the central nervous system, some glial cells, such as astrocytes and oligodendrocytes, express VAMP-2 immunoreactivity (Madison et al., 1999; Crippa et al., 2006), which suggests a role for SNARE proteins in glial–neuronal interactions. Therefore, the same mechanism(s) could also take place in satellite cell–neuronal interactions in the trigeminal ganglia.

Primary sensory neurons are pseudounipolar neurons–neuronal cell bodies extend both centrally and peripherally. Because nerve terminals lack the biosynthesis machinery necessary to produce proteins, all SNARE proteins examined in this study were synthesized in neuronal cell bodies of trigeminal ganglia, as demonstrated by RT-PCR analysis. These proteins are peripherally and centrally transported. When the IAN was ligated, these proteins accumulated proximal to the ligation site, indicating that SNARE proteins are transported to the peripheral nerve terminals by anterograde fast axonal transport.

SNARE Proteins in Dental Pulp and Periodontal Ligament

Double labeling with PGP 9.5 and SNARE proteins in the dental pulp showed that most nerve fibers labeled with PGP 9.5 also expressed syntaxin-1, SNAP-25, and VAMP-2. These results indicated that these three SNARE proteins were colocalized in the same nerve fibers.

The periodontal ligament contains at least two types of sensory nerve endings: nociceptive-free nerve endings and mechanoreceptive-specialized nerve endings. Although various types of mechanoreceptors have been reported, the Ruffini ending is the essential mechanoreceptor. Morphologically, periodontal Ruffini endings are characterized as expanded axon terminals with an association to special Schwann cells, called lamellar or terminal Schwann cells (Byers, 1984; for reviews, Byers and Maeda, 1997; Maeda et al., 1989, 1999). Ultrastructurally, a part of the axon terminal is exposed to surrounding collagenous tissues through slits in the Schwann sheaths, and the axoplasm extension plays an important role in mechanoreception (Kannari, 1990; Kannari et al., 1991). Periodontal mechanoreceptors originate from both trigeminal ganglion and trigeminal mesencephalic nuclei. A previous experimental study, however, revealed that periodontal Ruffini endings of the rodent incisor are only innervated from trigeminal ganglia (Byers and Dong, 1989).

Overlapping expression of syntaxin-1 and PGP 9.5 in periodontal nerve fibers, with the exception of nerve bundles entering the periodontal ligament, indicated that syntaxin-1 was present in both nociceptive-free nerve endings and mechanoreceptive-specialized endings in the periodontal ligament. Moreover, because most, if not all, syntaxin-1-positive nerve fibers also expressed SNAP-25, this suggested that SNAP-25 was present in free nerve endings and Ruffini endings of the ligament.

It is obvious that VAMP-2 was synthesized in the cell bodies of trigeminal ganglia, and transported peripherally, as well as centrally, because IAN ligation resulted in VAMP-2 accumulation in almost all nerve fibers proximal to the ligation site. Nerve fibers expressed VAMP-2 in the dental pulp of the incisor, but not in the periodontal ligament. There were two possible explanations for these results. First, cell size analysis revealed that VAMP-2 was mainly expressed in small- to medium-sized neurons in the trigeminal ganglion, which suggested that VAMP-2 was mostly associated with nociceptors. The periodontal ligament contains many mechanoreceptors, for example, Ruffini endings, and a few nociceptors, for example, free nerve endings. Although periodontal nociceptors express VAMP-2, the concentration of VAMP-2 was too low in this study to detect with the present immunohistochemical methods. Second, VAMP-2 was not expressed in the periodontal mechanoreceptors, and other v-SNARE proteins might be expressed in the periodontal Ruffini ending.

Functional Implications of SNARE Proteins

In the central nervous system, synaptic vesicle docking and exocytosis require interactions between synaptic vesicle proteins, such as VAMP-2, and the presynaptic plasma membrane proteins, syntaxin-1 and SNAP-25. In the peripheral nervous system, SNARE proteins have been detected in the retina (Morgans et al., 1996; von Kriegstein et al., 1999), cochlear hair cell synapses (Safieddine and Wenthold, 1999), and the gustatory system (Yang et al., 2000), where apparent synapses are present. The SNARE proteins are thought to participate in synapse vesicle docking and exocytosis at the synapse site, as has been shown in the central nervous system. This study, however, demonstrated expression of SNARE proteins in the sensory nerve endings, where no synapse is present. Therefore, other functions should be considered for the SNARE proteins.

Ultrastructural studies of periodontal Ruffini endings have indicated the presence of various sized vesicles in adult and developing animals (Byers, 1985; Maeda et al., 1989; Kannari, 1990; Nakakura-Ohshima et al., 1995). This study demonstrated that SNAP-25 expression was associated with vesicles in axon terminals of periodontal Ruffini endings. Norlin et al. (1999) showed that exocytosis-regulating proteins, such as SNAP-25, Rab 3, synaptotagmin, and synapsin, are expressed in rat molar dentinal tubules, most likely within the intradentinal nerves. The intradentinal nerve fibers have varicosities that contain many various sized vesicles, such as small, clear bodies resembling synaptic vesicles and larger, electron-dense vesicles. However, the typical synapse is not apparent in intradentinal nerve fibers (for review Byers, 1984). Previous studies have speculated that dentinal nerve fibers are capable of exocytosis of transmitter(s) stored in vesicles that resemble synaptic vesicles (Norlin et al., 1999).

This study observed expression of SNAP-25 and syntaxin-1, but lack of VAMP-2 expression, in the periodontal Ruffini endings. Several reports have shown that SNARE proteins are also required for exocytosis in nonneuronal cells, with various combinations of v-SNARE and t-SNARE (Ravichandran et al., 1996; Guo et al., 1998; Chen et al., 2000). Therefore, SNAP-25 and syntaxin-1 could also contribute to exocytosis from vesicles in the periodontal Ruffini endings other than VAMP-2 as a v-SNARE.

In conclusion, this study demonstrated that SNARE proteins were expressed in sensory nerve endings of the dental pulp and periodontal ligament in the rat incisor. Although the functional significance at primary afferent neuronal axon terminals remains to be elucidated, these results suggested that SNARE proteins are expressed at peripheral processes of primary afferent neurons, where there are no apparent synapses. These SNARE proteins might contribute to vesicle exocytosis.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

The authors thank Kohki Kadono and Misato Sakai of Osaka University Faculty of Dentistry for their excellent technical assistance.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED