Development and survival of vertebrate neurons is largely dependent on neurotrophic factors such as nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin-4 (NT4) and neurotrophin-3. These neurotrophins have been shown to promote survival of specific neuronal populations by interactions with the tyrosine kinase receptors trkA, trkB and trkC. NGF signals via trkA. BDNF and NT-4 bind to trkB. NT-3 binds to trkC and, to a lesser degree, trkB. Previous studies have demonstrated that deletion of trkA results in a marked reduction of calcitonin gene-related peptide (CGRP)-containing neurons in the trigeminal ganglion. Because CGRP is expressed in small to medium-sized neurons and because their axons are mainly unmyelinated, trkA is considered to be necessary for normal development of unmyelinated nociceptors. However, in trkB- and trkC-knockout mice, the distribution of CGRP-containing nerve fibers remains unchanged in the intraoral structure. Therefore, it is unlikely that the development of unmyelinated nociceptors depends on trkB or trkC in the TG.

VRL-1, a newly cloned capsaicin-receptor homologue, does not respond to capsaicin and is activated by high temperatures with a threshold of 52 °C. In TGs, VRL-1-containing neurons are medium-sized to large. They are considered to have free nerve endings in their peripheral tissues, and project their myelinated axons to superficial laminae in the medullary dorsal horn. Our previous study demonstrated that VRL-1-immunopositive (ip) TG neurons innervate tooth pulp and facial skin. Forty percent of TG neurons retrogradely labeled from the tooth pulp exhibited VRL-1-immunostaining, whereas VRL-1 staining was detected only in 9% of those labeled from the facial skin. Therefore, it was suggested that myelinated nociceptors are common in the intraoral structure. On the other hand, all nerve fibers disappear in tooth pulp of trkA-knockout mice. However, innervation of tooth pulp has never been found in trkB- or trkC-knockout mice. In addition, little is known about the effects of trk deficiency on myelinated nociceptors in TGs.

In the present study, cell size of TG neurons and distribution paterns of VRL-1-ip neurons were examined in trk-knockout mice to determine the dependency of myelinated nociceptors on expression of the gene. Furthermore, immunohistochemistry of the protein gene product 9.5 (PGP 9.5), a marker for both unmyelinated and myelinated axons, was also performed on palate and tooth of trk-knockout mice to examine effect of its deficiency on peripheral nerve endings. Corpuscular and Merkel endings were excluded from the present analysis, because they are considered to be low-threshold mechanoreceptors.

Mice lacking gene expression of trkA, trkB or trkC were prepared according to the methods of Klein et al., 1993 and Klein et al., 1994 and Smeyne et al. (1994). Homozygous mice, and corresponding wildtype and heterozygous littermates at postnatal day 12–16 were obtained from overnight mating of heterozygous mice (C57BL/6J). Four wildtype mice, 3 trkA?/? mice, 4 trkB?/? mice and 4 trkC?/? mice were used in this study. Animals were anesthetized with pentobarbiturate (0.5 ml/kg, ip) and perfused via the left ventricle with 2 ml saline followed by 50 ml 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). After perfusion the TG and maxilla were dissected, and immersed overnight in the same fixative at room temperature. The maxilla was decalcified with 4.13% ethylene diaminetetraacetic acid disodium salt in 0.1 M phosphate buffer (pH 7.4) for 1 week at room temperature. Subsequently, tissues were stored at 4 °C in phosphate buffered saline (PBS) containing 0.1 mM sodium azide until used. They were immersed in PBS containing 20% sucrose and kept overnight. Cryostat sections were cut horizontally (for the TG) or sagittally (for the maxilla) with a thickness of 10 ?m, mounted on gelatin-coated glass slides and air-dried. For the number and cell size of TG neurons, Nissl staining was performed. The microscopic images of Nissl-stained cell bodies were projected over a digitizer tablet using a drawing tube (×215). Complete series of sections were divided into five subsets so that every fifth section was mounted on the same slides. The number and cross-sectional area of those cell bodies that contained the nucleolus was recorded in one of these subsets. For VRL-1 immunohisotchemistry, sections were incubated with rabbit anti-VRL-1 antibodies followed by incubation with biotinylated goat anti-rabbit IgG and ABC complex (Vector, Burlingame, CA, USA). Data on the proportion of VRL-1-ip neurons of TG neurons were obtained per area of the microscopic field of ganglia from two animals. Cell size of VRL-1-ip cell bodies containing a nucleolus was analyzed in representative sections of ganglia of two animals. For the demonstration of palatal and pulpal innervation, sections of palate and maxilla were incubated with rabbit anti-PGP 9.5 antibodies (dilution, 1:15,000; UltraClone, Rossiters Farmhouse, Wellow, UK) followed by incubation with biotinylated goat anti-rabbit IgG and ABC complex.

Specificity of the primary antibodies used in this study has been described elsewhere.

Experiments were carried out under the control of the Animal Research Control Committee in accordance with The Guidelines for Animal Experiments of Okayama University Medical School, Government Animal Protection and Management Law (No. 105), and Japanese Government Notification on Feeding and Safekeeping of Animals (No. 6). Efforts were made to minimize the number of animals used and their suffering.

TGs contained abundant primary sensory neurons in wildtype and heterozygous mice for trks, but their numbers were strongly decreased in knockout mice (82%, 39% and 48% reduction for trkA, trkB and trkC, respectively; Fig. 1). Cell-size analysis showed that TG neurons had various sizes in wildtype and heterozygous mice (mean±SD, 328.9±184.7 ?m²; range, 25.3–1389.9 ?m²; n, 5257); 28% (1463/5257) of these neurons were smaller than 200 ?m². Approximately, half of them (43% or 2223/5257) had a size of 200–400 ?m² and 30% (1571/5257) had cell bodies that were>400 ?m². In trkA-knockout mice, the number of TG neurons decreased in all cell-size ranges but the reduction was most prominent for small TG neurons (91%, 75%, and 83% reduction in small, medium-sized and large TG neurons, respectively; Fig. 1). As a result, TG neurons were mainly medium-sized or large in knockout mice (mean±SD, 346.7±146.3 ?m²; range, 50.3–1259.8 ?m²; n, 942); 58% of TG neurons (545/942) were medium-sized and 29% (272/942) had large cell bodies. Only 13% of neurons (125/942) were small in the TG. On the other hand, trkB and trkC deficiency caused a decrease in medium-sized TG neurons (39% reduction) and large TG neurons (80% reduction). The number of small TG neurons was not affected by the loss of trkB. However, in trkC-knockout mice, small TG neurons also decreased in number (29% reduction). As a result, TG neurons were mainly small to medium-sized in trkB-knockout mice (mean±SD, 240.0±122.6 ?m²; range, 26.4–815.4 ?m²; n, 3203) and trkC-knockout mice (mean±SD, 254.4±121.5 ?m²; range, 248.7–1028.8 ?m²; n, 2719); 47% (1490/3203) and 38% (1041/2719) of TG neurons were small in trkB- and trkC-knockout mice, respectively. More than 40% were medium-sized in the knockout mice (trkB, 1367/3203; trkC, 1356/2719). Large TG neurons were relatively rare in the mutants (trkB, 11% or 346/3203; trkC, 12% or 322/2719).