- Email: [email protected]
Kv1.2-immunoreactive primary sensory neurons in the rat trigeminal ganglion
Brain Research 974 (2003) 222–227 www.elsevier.com / locate / brainres
Research report
Kv1.2-immunoreactive primary sensory neurons in the rat trigeminal ganglion Hiroyuki Ichikawa*, Tomosada Sugimoto Department of Oral Anatomy and Function, and Biodental Research Center, Okayama University Graduate School of Medicine and Dentistry, 2 -5 -1 Shikata-cho, Okayama 700 -8525, Japan Accepted 6 March 2003
Abstract Immunohistochemistry for Kv1.2, a subunit of voltage-gated K 1 channels, was performed on the trigeminal ganglion (TG). Immunoreactivity (ir) was detected in half (48%) the TG neurons. These neurons were mostly medium-sized to large (range 137.6–2664.8 mm 2 , mean6S.D. 892.66413.3 mm 2 ). A double immunofluorescence method also revealed co-expression of Kv1.2 and parvalbumin. Half (54%) the Kv1.2-immunoreactive (ir) neurons exhibited parvalbumin-ir, and parvalbumin-ir neurons mostly showed Kv1.2-ir (95%). Kv1.2-ir neurons which co-expressed CGRP-ir were rare in this ganglion. Some 40% of TG neurons retrogradely labeled from the facial skin exhibited Kv1.2-ir, whereas ir was detected in 16% of those labeled from the tooth pulp. The present study indicates that Kv1.2-ir TG neurons include low-threshold mechanoreceptors and nociceptors which innervate the facial skin and tooth pulp, respectively. 2003 Elsevier Science B.V. All rights reserved. Theme: Sensory systems Topic: Somatic and visceral afferents Keywords: Calcitonin gene-related peptide; Facial skin; Immunohistochemistry; Kv1.2; Parvalbumin; Tooth pulp; Trigeminal ganglion
1. Introduction Previous immunohistochemical studies have classified primary sensory neurons into several subpopulations on the basis of their chemical markers [1,5,7,10,11,14,15]. Calcitonin gene-related peptide (CGRP) is considered to be a marker specific to small to medium-sized neurons in the dorsal root (DRG) and trigeminal ganglia (TG) [10,11,14,15]. CGRP-immunoreactive (-ir) primary sensory neurons supply their peripheral receptive field with free nerve endings, and are thought to participate in nociception [9,11,14]. On the other hand, parvalbumin, a member of the calcium-binding protein family, is localized to large neurons in the DRG and TG [1,3–5]. Such neurons are *Corresponding author. Present address: Department of Oral Anatomy and Function, Okayama University Graduate School of Medicine and Dentistry, 2-5-1 Shikata-cho, Okayama 700-8525, Japan. Tel.: 181-86235-5537; fax: 181-86-235-6612. E-mail address: [email protected] (H. Ichikawa).
considered to be proprioceptors in the DRG, because muscle spindles have parvalbumin-ir axons in the spinal nervous system [1]. In the TG, however, parvalbumin-ir neurons supply the oro-facial tissues with encapsulated and unencapsulated corpuscular endings [3,4]. Therefore, lowthreshold mechanoreceptors probably contain this calciumbinding protein in the TG. In addition, a significant number of parvalbumin-ir TG neurons innervate the tooth pulp [6]. Parvalbumin-ir varicose terminals are abundant in the subodontoblastic and odontoblastic layers [8], and are thought to transduce pulpal nociception. Kv1.2, a subunit of voltage-gated K 1 channels, is distributed in the central and peripheral nervous systems [12,13,16,17]. In the dorsal root ganglion (DRG), this ion channel is localized to medium-sized to large neurons [12]. In addition, these neurons are devoid of the vanilloid receptor subtype 1, a marker for small nociceptive neurons [12]. Therefore, Kv1.2-containing neurons are considered to be proprioceptors and low-threshold mechanoreceptors in the DRG. However, little is known about the dis-
0006-8993 / 03 / $ – see front matter 2003 Elsevier Science B.V. All rights reserved. doi:10.1016 / S0006-8993(03)02583-6
H. Ichikawa, T. Sugimoto / Brain Research 974 (2003) 222–227
tribution and function of Kv1.2-containing neurons in the TG. In this study, we examined the distribution of Kv1.2-ir neurons in the TG. The co-expression of this ion channel with parvalbumin or CGRP is also investigated in TG neurons innervating the facial skin and tooth pulp.
2. Materials and methods Five TGs were obtained from four male Sprague–Dawley rats (200–300 g). Rats were anesthetized with ether to the level at which respiration was markedly suppressed, and transvascularly perfused with 50 ml of saline followed by 500 ml of 4% formaldehyde in 0.1 M phosphate buffer (pH 7.4). The materials were dissected and post-fixed with the same fixative for 30 min, soaked in a phosphatebuffered 20% sucrose solution overnight, frozen sectioned at 12 mm, and thaw-mounted on gelatin-coated glass slides. For demonstration of Kv1.2-ir, ABC (avidin–biotin– horseradish peroxidase complex) and indirect immunofluorescence methods were performed. Sections were
223
incubated with rabbit anti-Kv1.2 serum (1:5000 for ABC, 1:500 for immunofluorescence; Alomone, Jerusalem, Israel) for 24 h at room temperature. They were then incubated with biotinylated goat anti-rabbit IgG and ABC complex (Vector, Burlingame, CA, USA), or lissamine rhodamine B chloride-conjugated donkey anti-rabbit IgG (1:500; Jackson ImmunoResearch, Bal Harbor, ME, USA). For cell size analysis of ABC-stained neurons, the microscopic image (3215) of the cell bodies was projected over a digitizer tablet using a drawing tube. The cross-sectional area of those cell bodies that contained the nucleolus was recorded. For simultaneous visualization of Kv1.2 with parvalbumin or CGRP, a double immunofluorescence method was used. The sections were incubated for 24 h at room temperature with a mixture of rabbit anti-Kv1.2 serum (1:500) and either monoclonal anti-parvalbumin antibody (1:1000; Sigma, St Louis, MO, USA) or guinea pig antiCGRP serum (1:1000; Milab, Sweden). The sections were then treated with a mixture of lissamine rhodamine B chloride-conjugated donkey anti-rabbit IgG (1:500 for Kv1.2; Jackson ImmunoResearch) and either fluorescein isothiocyanate-conjugated donkey anti-mouse IgG (1:100
Fig. 1. Immunofluorescent microphotographs for Kv1.2 (A, B, D), parvalbumin (C) and CGRP (E) in the TG. The TG contains abundant Kv1.2-ir primary sensory neurons (A). These neurons are medium-sized to large. Panels B and C, and D and E show the same fields of view, respectively. A double immunofluorescence method demonstrates that many Kv1.2-ir neurons (arrows in B) show parvalbumin-ir (arrows in C). Arrows in D and E point to a neuron which co-expresses Kv1.2- and CGRP-ir. Bars represent 50 mm (A, B). Panels B–E are at the same magnification.
224
H. Ichikawa, T. Sugimoto / Brain Research 974 (2003) 222–227
For the control, rabbit anti-Kv1.2 serum was preabsorbed with the protein (20 mg / ml, Alomone Labs.). No staining was observed in the control. The specificities of other antibodies have been described elsewhere [6,8].
3. Results
3.1. Distribution and cell size of Kv1.2 -ir neurons in the TG
Fig. 2. Histogram showing the cell size spectrum of Kv1.2-ir and -immunonegative neurons in the DRG. The data were obtained from 515 TG neurons.
for parvalbumin; Jackson ImmunoResearch) or fluorescein isothiocyanate-conjugated donkey anti-guinea pig IgG (1:100 for CGRP; Jackson ImmunoResearch). For demonstration of Kv1.2-ir TG neurons innervating the facial skin and tooth pulp, four male rats (300–350 g) were used. Under deep anesthesia by i.p. injection with ethyl carbamate (650 mg / kg) and pentobarbital sodium (20 mg / kg), 0.1–0.2 ml of 1% fluorogold (FG; Fluorochrom, Englewood, CO, USA) in distilled water was injected into the infraorbital skin or right first maxillary molar tooth pulp. After 3 days, the animals were reanesthetized with ether, and transvascularly perfused with 4% formaldehyde. The right TGs were frozen-sectioned at 12 mm, mounted on gelatin-coated glass slides, and processed for co-expression of Kv1.2 with parvalbumin or CGRP as described above. The brain stem of the rats which had received tracer application from the tooth pulp, was also cryosectioned at 50 mm and examined with fluorescent illumination for possible FG labeling. All these FG-labeled cells are considered to be primary neurons innervating the tooth pulp and the possibility of periodontal spread of FG through the apical foramen is negligible, because the labeled neuron was not observed in the mesencephalic trigeminal tract nucleus. For cell size analysis of cutaneous and tooth pulp neurons, the cross-sectional area of FGlabeled cell bodies which contained nuclear profiles was measured on glossy prints (3165). Because of the halo surrounding the FG-positive cells, however, accuracy of measurement was compromised.
The TG contained abundant Kv1.2-ir neurons (Fig. 1A). Half (47.7% or 245 / 515) the TG neurons were immunoreactive for the ion channel. These neurons were scattered throughout the ganglion. Immunoreactivity (ir) was detected in the cytoplasm but not in the nucleus. As shown in Fig. 2, Kv1.2-ir TG neurons were mostly medium-sized to large (range 137.6–2664.8 mm 2 , mean6S.D. 892.66413.3 mm 2 ); 73.1% (87 / 119) of large TG neurons (.1000 mm 2 ) expressed Kv1.2-ir and 52.1% (137 / 263) of medium-sized neurons (400–1000 mm 2 ) showed ir. Only 15.8% (21 / 133) of small neurons (,400 mm 2 ) showed Kv1.2-ir.
3.2. Co-expression of Kv1.2 with parvalbumin or CGRP As described previously [1,5,10,14,15], the TG contained many parvalbumin- and CGRP-ir neurons. Parvalbumin-ir neurons were mostly large, while CGRP-ir neurons were small to medium-sized. Kv1.2-ir neurons were more abundant than parvalbumin- and CGRP-ir neurons in the TG. Half (54.2% or 417 / 769) the Kv1.2-ir neurons exhibited parvalbumin-ir (Fig. 1B,C). Conversely, parvalbumin-ir neurons mostly co-expressed Kv1.2-ir (95.2% or 417 / 438). On the other hand, co-expression of Kv1.2 and CGRP was rare in the TG. Only 2.2% (16 / 722) of Kv1.2ir neurons exhibited CGRP-ir, while 3.6% (16 / 443) of CGRP-ir neurons showed Kv1.2-ir (Fig. 1D,E).
3.3. Kv1.2 -ir TG neurons innervating the facial skin and tooth pulp At 3 days after FG application to the infraorbital skin or upper molar tooth pulp, many cell bodies were labeled in the TG. They were predominantly located in the maxillary division of the ganglion. Cutaneous TG neurons were of various sizes (small: 25% or 169 / 676; medium-sized: 52.4% or 354 / 676; and large: 22.6% or 153 / 676), whereas tooth pulp TG neurons were mostly medium-sized to large
Fig. 3. Immunofluorescent microphotographs for FG (A, D, G, J), Kv1.2 (B, E, H, K), parvalbumin (C, I) and CGRP (F, L). Panels A–C, D–F, G–I and J–L show the same fields of view, respectively. A TG neuron retrogradely labeled from the facial skin with FG (arrows in A, D) co-expresses Kv1.2 with parvalbumin (arrows in B, C), but not CGRP (arrowheads in E, F). Arrowheads in A–C point to Kv1.2-ir cutaneous neurons which are devoid of parvalbumin-ir. Co-expression of Kv1.2 with parvalbumin or CGRP was rare in tooth pulp TG neurons. Arrows in G–I point to a tooth pulp TG neuron which is immunoreactive for both Kv1.2 and parvalbumin. Arrowheads in J–L point to a tooth pulp TG neuron which exhibits Kv1.2-ir but not CGRP-ir. Bar represents 50 mm (A). All panels are at the same magnification.
H. Ichikawa, T. Sugimoto / Brain Research 974 (2003) 222–227
225
226
H. Ichikawa, T. Sugimoto / Brain Research 974 (2003) 222–227
that 9.3% (31 / 332) of cutaneous neurons co-expressed Kv1.2- and parvalbumin-ir (Fig. 3A–C). Some 25% (31 / 124) of Kv1.2-ir cutaneous neurons contained parvalbumin-ir, whereas 83.8% (31 / 37) of parvalbumin-ir cutaneous neurons showed Kv1.2-ir. Only 2.9% (10 / 344) of cutaneous neurons co-expressed Kv1.2- and CGRP-ir (Fig. 3D–F). Some (7% 10 / 144) of Kv1.2-ir cutaneous neurons had CGRP-ir and 11.1% (10 / 90) of CGRP-ir cutaneous neurons expressed Kv1.2-ir (Fig. 3G–L). On the other hand, 18.9% (14 / 74) and 52% (26 / 50) of tooth pulp neurons were immunoreactive for parvalbumin and CGRP, respectively. However, co-expression of Kv1.2 with parvalbumin- (1.4% or 1 / 74) or CGRP-ir (0 / 50) was rare in tooth pulp neurons.
Fig. 4. Histogram showing the cell size spectrum of Kv1.2-ir and -immunonegative neurons in cutaneous TG neurons. The data were obtained from 676 cutaneous neurons.
(small: 6.5% or 8 / 124; medium-sized: 60.5% or 75 / 124; and large: 33.1% or 41 / 124) (Fig. 3A,B,D,E,G,H,J,K). Retrograde tracing and immunohistochemical methods revealed that 39.6% (268 / 676) and 16.1% (20 / 124) of TG neurons innervating the facial skin and tooth pulp, respectively, contained Kv1.2-ir. As shown in Fig. 4, 64.1% (98 / 153) of large and 44.1% (156 / 354) of medium-sized cutaneous neurons were immunoreactive for the channel. Only 8.3% (14 / 169) of small cutaneous neurons showed ir. On the other hand, 14.6% (6 / 41) of large and 18.7% (14 / 75) of medium-sized tooth pulp neurons showed Kv1.2-ir. Small tooth pulp neurons were devoid of ir (0 / 8) (Fig. 5). The double immunofluorescence method demonstrated
Fig. 5. A histogram showing the cell size spectrum of Kv1.2-ir and -immunonegative neurons in tooth pulp TG neurons. The data were obtained from 124 tooth pulp neurons.
4. Discussion The present study demonstrated that Kv1.2 was localized to primary sensory neurons in the TG. Such neurons were mostly medium-sized to large. Small Kv1.2-ir neurons were rare in the TG. The cell size finding in the TG is similar to that reported for the DRG [16]. In the DRG, Kv1.2-ir neurons are thought to be proprioceptors and low-threshold mechanoreceptors. However, most Kv1.2-ir TG neurons are unlikely to be proprioceptors because the number of proprioceptors is small in the TG. In addition, the present double immunofluorescence method revealed that Kv1.2-ir neurons were immunoreactive for parvalbumin. Therefore, low-threshold mechanoreceptors probably contain Kv1.2-ir in the TG. The present study also demonstrated that cutaneous and tooth pulp TG neurons showed Kv1.2-ir. Because the tooth pulp is considered to be innervated exclusively by nociceptive afferents in the TG, Kv1.2-ir neurons probably include nociceptors in this ganglion. Tooth pulp primary nociceptors are different from those innervating facial skin. Unlike facial nociceptors, they often contain calcium-binding proteins such as parvalbumin and calbindin D-28k [6–8]. In contrast, those containing substance P are relatively rare [2]. These characteristics probably reflect the abundance of medium-sized to large nociceptors in the tooth pulp. This idea is supported by the present finding that tooth pulp neurons are usually larger than cutaneous neurons. The tooth pulp is enclosed in layers of hard tissues and highly protected from physical insults. Nevertheless, the dentoenamel junction, which does not receive dense innervation, is extremely sensitive to mechanical stimulation. The channel, which is otherwise involved in low-threshold mechanoception, may be involved in dentinal nociception. In conclusion, we have described Kv1.2-ir neurons in the TG. Immunoreactivity was detected in half the TG neurons. These neurons were mostly medium-sized to large. More than half the Kv1.2-ir neurons showed parvalbumin-ir, whereas only 2% of such neurons contained CGRP-ir. On the other hand, the proportion of Kv1.2-ir
H. Ichikawa, T. Sugimoto / Brain Research 974 (2003) 222–227
neurons among cutaneous neurons (40%) was higher than among tooth pulp neurons (16%). Kv1.2-ir TG neurons probably include low-threshold mechanoreceptors and nociceptors which innervate the facial skin and tooth pulp, respectively.
[8]
[9]
Acknowledgements [10]
This work was supported by a grant from the Japanese Ministry of Education, Culture, Sports, Science and Technology to H.I. (No. 14571733). [11]
References [1] M.R. Celio, Calbindin D-28k and parvalbumin in the rat nervous system, Neuroscience 35 (1990) 375–475. [2] K. Fried, J. Arvidsson, B. Robertson, E. Brodin, E. Theodorsson, Combined retrograde tracing and enzyme / immunohistochemistry of trigeminal ganglion cell bodies innervating tooth pulps in the rat, Neuroscience 33 (1989) 101–109. [3] H. Ichikawa, T. Sugimoto, Parvalbumin- and calbindin D-28kimmunoreactive innervation of oro-facial tissues in the rat, Exp. Neurol. 146 (1997) 414–418. [4] H. Ichikawa, D.M. Jacobowitz, T. Sugimoto, Coexpression of calretinin and parvalbumin in Ruffini-like endings in the rat incisor periodontal ligament, Brain Res. 770 (1997) 294–297. [5] H. Ichikawa, T. Deguchi, T. Nakago, D.M. Jacobowitz, T. Sugimoto, Parvalbumin, calretinin and carbonic anhydrase in the trigeminal and spinal primary neurons of the rat, Brain Res. 655 (1994) 241–245. [6] H. Ichikawa, T. Deguchi, T. Nakago, D.M. Jacobowitz, T. Sugimoto, Parvalbumin- and calretinin-immunoreactive trigeminal neurons innervating the rat molar tooth pulp, Brain Res. 679 (1995) 205–211. [7] H. Ichikawa, T. Deguchi, Y. Fujiyoshi, T. Nakago, D.M. Jacobowitz,
[12]
[13]
[14]
[15]
[16]
[17]
227
T. Sugimoto, Calbindin-D28k-immunoreactivity in the trigeminal ganglion neurons and molar tooth pulp of the rat, Brain Res. 715 (1996) 71–78. H. Ichikawa, T. Deguchi, S. Mitani, T. Nakago, D.M. Jacobowitz, T. Yamaai, T. Sugimoto, Neural parvalbumin and calretinin in the tooth pulp, Brain Res. 647 (1994) 124–130. A. Ishida-Yamamoto, E. Senba, M. Tohyama, Distribution and fine structure of calcitonin gene-related peptide-like immunoreactive nerve fibers in the rat skin, Brain Res. 491 (1989) 93–101. ¨ G. Ju, T. Hokfelt, E. Brodin, J. Fahrenkrug, J.A. Fischer, P. Frey, R.P. Elde, J.C. Brown, Primary sensory neurons of the rat showing calcitonin gene-related peptide immunoreactivity and their relation to substance P-, somatostatin-, galanin-, vasoactive intestinal polypeptide- and cholecystokinin-immunoreactive ganglion cells, Cell Tissue Res. 247 (1987) 417–431. L. Kruger, J.D. Silverman, P.W. Mantyh, C. Sternini, N.C. Brecha, Peripheral patterns of calcitonin gene-related peptide general somatic sensory innervation: cutaneous and deep terminations, J. Comp. Neurol. 280 (1989) 291–302. M.N. Rasband, E.W. Park, T.W. Vanderah, J. Lai, F. Porreca, J.S. Trimmer, Distinct potassium channels on pain-sensing neurons, Proc. Natl. Acad. Sci. USA 98 (2001) 13373–13378. M. Sheng, M.L. Tsaur, Y.N. Jan, L.Y. Jan, Contrasting subcellular localization of the Kv1.2 K 1 channel subunit in different neurons of rat brain, J. Neurosci. 14 (1994) 2408–2417. J.D. Silverman, L. Kruger, Calcitonin gene-related-peptide-immunoreactive innervation of the rat head with emphasis on specialized sensory structures, J. Comp. Neurol. 280 (1989) 303–330. G. Skofitsch, D.M. Jacobowitz, Calcitonin gene-related peptide coexists with substance P in capsaicin sensitive neurons and sensory ganglia of the rat, Peptides 6 (1985) 747–754. R.W. Veh, R. Lichtinghagen, S. Sewing, F. Wunder, I.M. Grumbach, O. Pongs, Immunohistochemical localization of five members of the Kv1 channel subunits: contrasting subcellular locations and neuronspecific co-localizations in rat brain, Eur. J. Neurosci. 7 (1995) 2189–2205. S. Yokoyama, H. Takeda, H. Higashida, Expression of Kv1.2 potassium channels in rat sensory ganglia. An immunohistochemical study, Ann. NY Acad. Sci. 868 (1999) 454–457.
Research report
Kv1.2-immunoreactive primary sensory neurons in the rat trigeminal ganglion Hiroyuki Ichikawa*, Tomosada Sugimoto Department of Oral Anatomy and Function, and Biodental Research Center, Okayama University Graduate School of Medicine and Dentistry, 2 -5 -1 Shikata-cho, Okayama 700 -8525, Japan Accepted 6 March 2003
Abstract Immunohistochemistry for Kv1.2, a subunit of voltage-gated K 1 channels, was performed on the trigeminal ganglion (TG). Immunoreactivity (ir) was detected in half (48%) the TG neurons. These neurons were mostly medium-sized to large (range 137.6–2664.8 mm 2 , mean6S.D. 892.66413.3 mm 2 ). A double immunofluorescence method also revealed co-expression of Kv1.2 and parvalbumin. Half (54%) the Kv1.2-immunoreactive (ir) neurons exhibited parvalbumin-ir, and parvalbumin-ir neurons mostly showed Kv1.2-ir (95%). Kv1.2-ir neurons which co-expressed CGRP-ir were rare in this ganglion. Some 40% of TG neurons retrogradely labeled from the facial skin exhibited Kv1.2-ir, whereas ir was detected in 16% of those labeled from the tooth pulp. The present study indicates that Kv1.2-ir TG neurons include low-threshold mechanoreceptors and nociceptors which innervate the facial skin and tooth pulp, respectively. 2003 Elsevier Science B.V. All rights reserved. Theme: Sensory systems Topic: Somatic and visceral afferents Keywords: Calcitonin gene-related peptide; Facial skin; Immunohistochemistry; Kv1.2; Parvalbumin; Tooth pulp; Trigeminal ganglion
1. Introduction Previous immunohistochemical studies have classified primary sensory neurons into several subpopulations on the basis of their chemical markers [1,5,7,10,11,14,15]. Calcitonin gene-related peptide (CGRP) is considered to be a marker specific to small to medium-sized neurons in the dorsal root (DRG) and trigeminal ganglia (TG) [10,11,14,15]. CGRP-immunoreactive (-ir) primary sensory neurons supply their peripheral receptive field with free nerve endings, and are thought to participate in nociception [9,11,14]. On the other hand, parvalbumin, a member of the calcium-binding protein family, is localized to large neurons in the DRG and TG [1,3–5]. Such neurons are *Corresponding author. Present address: Department of Oral Anatomy and Function, Okayama University Graduate School of Medicine and Dentistry, 2-5-1 Shikata-cho, Okayama 700-8525, Japan. Tel.: 181-86235-5537; fax: 181-86-235-6612. E-mail address: [email protected] (H. Ichikawa).
considered to be proprioceptors in the DRG, because muscle spindles have parvalbumin-ir axons in the spinal nervous system [1]. In the TG, however, parvalbumin-ir neurons supply the oro-facial tissues with encapsulated and unencapsulated corpuscular endings [3,4]. Therefore, lowthreshold mechanoreceptors probably contain this calciumbinding protein in the TG. In addition, a significant number of parvalbumin-ir TG neurons innervate the tooth pulp [6]. Parvalbumin-ir varicose terminals are abundant in the subodontoblastic and odontoblastic layers [8], and are thought to transduce pulpal nociception. Kv1.2, a subunit of voltage-gated K 1 channels, is distributed in the central and peripheral nervous systems [12,13,16,17]. In the dorsal root ganglion (DRG), this ion channel is localized to medium-sized to large neurons [12]. In addition, these neurons are devoid of the vanilloid receptor subtype 1, a marker for small nociceptive neurons [12]. Therefore, Kv1.2-containing neurons are considered to be proprioceptors and low-threshold mechanoreceptors in the DRG. However, little is known about the dis-
0006-8993 / 03 / $ – see front matter 2003 Elsevier Science B.V. All rights reserved. doi:10.1016 / S0006-8993(03)02583-6
H. Ichikawa, T. Sugimoto / Brain Research 974 (2003) 222–227
tribution and function of Kv1.2-containing neurons in the TG. In this study, we examined the distribution of Kv1.2-ir neurons in the TG. The co-expression of this ion channel with parvalbumin or CGRP is also investigated in TG neurons innervating the facial skin and tooth pulp.
2. Materials and methods Five TGs were obtained from four male Sprague–Dawley rats (200–300 g). Rats were anesthetized with ether to the level at which respiration was markedly suppressed, and transvascularly perfused with 50 ml of saline followed by 500 ml of 4% formaldehyde in 0.1 M phosphate buffer (pH 7.4). The materials were dissected and post-fixed with the same fixative for 30 min, soaked in a phosphatebuffered 20% sucrose solution overnight, frozen sectioned at 12 mm, and thaw-mounted on gelatin-coated glass slides. For demonstration of Kv1.2-ir, ABC (avidin–biotin– horseradish peroxidase complex) and indirect immunofluorescence methods were performed. Sections were
223
incubated with rabbit anti-Kv1.2 serum (1:5000 for ABC, 1:500 for immunofluorescence; Alomone, Jerusalem, Israel) for 24 h at room temperature. They were then incubated with biotinylated goat anti-rabbit IgG and ABC complex (Vector, Burlingame, CA, USA), or lissamine rhodamine B chloride-conjugated donkey anti-rabbit IgG (1:500; Jackson ImmunoResearch, Bal Harbor, ME, USA). For cell size analysis of ABC-stained neurons, the microscopic image (3215) of the cell bodies was projected over a digitizer tablet using a drawing tube. The cross-sectional area of those cell bodies that contained the nucleolus was recorded. For simultaneous visualization of Kv1.2 with parvalbumin or CGRP, a double immunofluorescence method was used. The sections were incubated for 24 h at room temperature with a mixture of rabbit anti-Kv1.2 serum (1:500) and either monoclonal anti-parvalbumin antibody (1:1000; Sigma, St Louis, MO, USA) or guinea pig antiCGRP serum (1:1000; Milab, Sweden). The sections were then treated with a mixture of lissamine rhodamine B chloride-conjugated donkey anti-rabbit IgG (1:500 for Kv1.2; Jackson ImmunoResearch) and either fluorescein isothiocyanate-conjugated donkey anti-mouse IgG (1:100
Fig. 1. Immunofluorescent microphotographs for Kv1.2 (A, B, D), parvalbumin (C) and CGRP (E) in the TG. The TG contains abundant Kv1.2-ir primary sensory neurons (A). These neurons are medium-sized to large. Panels B and C, and D and E show the same fields of view, respectively. A double immunofluorescence method demonstrates that many Kv1.2-ir neurons (arrows in B) show parvalbumin-ir (arrows in C). Arrows in D and E point to a neuron which co-expresses Kv1.2- and CGRP-ir. Bars represent 50 mm (A, B). Panels B–E are at the same magnification.
224
H. Ichikawa, T. Sugimoto / Brain Research 974 (2003) 222–227
For the control, rabbit anti-Kv1.2 serum was preabsorbed with the protein (20 mg / ml, Alomone Labs.). No staining was observed in the control. The specificities of other antibodies have been described elsewhere [6,8].
3. Results
3.1. Distribution and cell size of Kv1.2 -ir neurons in the TG
Fig. 2. Histogram showing the cell size spectrum of Kv1.2-ir and -immunonegative neurons in the DRG. The data were obtained from 515 TG neurons.
for parvalbumin; Jackson ImmunoResearch) or fluorescein isothiocyanate-conjugated donkey anti-guinea pig IgG (1:100 for CGRP; Jackson ImmunoResearch). For demonstration of Kv1.2-ir TG neurons innervating the facial skin and tooth pulp, four male rats (300–350 g) were used. Under deep anesthesia by i.p. injection with ethyl carbamate (650 mg / kg) and pentobarbital sodium (20 mg / kg), 0.1–0.2 ml of 1% fluorogold (FG; Fluorochrom, Englewood, CO, USA) in distilled water was injected into the infraorbital skin or right first maxillary molar tooth pulp. After 3 days, the animals were reanesthetized with ether, and transvascularly perfused with 4% formaldehyde. The right TGs were frozen-sectioned at 12 mm, mounted on gelatin-coated glass slides, and processed for co-expression of Kv1.2 with parvalbumin or CGRP as described above. The brain stem of the rats which had received tracer application from the tooth pulp, was also cryosectioned at 50 mm and examined with fluorescent illumination for possible FG labeling. All these FG-labeled cells are considered to be primary neurons innervating the tooth pulp and the possibility of periodontal spread of FG through the apical foramen is negligible, because the labeled neuron was not observed in the mesencephalic trigeminal tract nucleus. For cell size analysis of cutaneous and tooth pulp neurons, the cross-sectional area of FGlabeled cell bodies which contained nuclear profiles was measured on glossy prints (3165). Because of the halo surrounding the FG-positive cells, however, accuracy of measurement was compromised.
The TG contained abundant Kv1.2-ir neurons (Fig. 1A). Half (47.7% or 245 / 515) the TG neurons were immunoreactive for the ion channel. These neurons were scattered throughout the ganglion. Immunoreactivity (ir) was detected in the cytoplasm but not in the nucleus. As shown in Fig. 2, Kv1.2-ir TG neurons were mostly medium-sized to large (range 137.6–2664.8 mm 2 , mean6S.D. 892.66413.3 mm 2 ); 73.1% (87 / 119) of large TG neurons (.1000 mm 2 ) expressed Kv1.2-ir and 52.1% (137 / 263) of medium-sized neurons (400–1000 mm 2 ) showed ir. Only 15.8% (21 / 133) of small neurons (,400 mm 2 ) showed Kv1.2-ir.
3.2. Co-expression of Kv1.2 with parvalbumin or CGRP As described previously [1,5,10,14,15], the TG contained many parvalbumin- and CGRP-ir neurons. Parvalbumin-ir neurons were mostly large, while CGRP-ir neurons were small to medium-sized. Kv1.2-ir neurons were more abundant than parvalbumin- and CGRP-ir neurons in the TG. Half (54.2% or 417 / 769) the Kv1.2-ir neurons exhibited parvalbumin-ir (Fig. 1B,C). Conversely, parvalbumin-ir neurons mostly co-expressed Kv1.2-ir (95.2% or 417 / 438). On the other hand, co-expression of Kv1.2 and CGRP was rare in the TG. Only 2.2% (16 / 722) of Kv1.2ir neurons exhibited CGRP-ir, while 3.6% (16 / 443) of CGRP-ir neurons showed Kv1.2-ir (Fig. 1D,E).
3.3. Kv1.2 -ir TG neurons innervating the facial skin and tooth pulp At 3 days after FG application to the infraorbital skin or upper molar tooth pulp, many cell bodies were labeled in the TG. They were predominantly located in the maxillary division of the ganglion. Cutaneous TG neurons were of various sizes (small: 25% or 169 / 676; medium-sized: 52.4% or 354 / 676; and large: 22.6% or 153 / 676), whereas tooth pulp TG neurons were mostly medium-sized to large
Fig. 3. Immunofluorescent microphotographs for FG (A, D, G, J), Kv1.2 (B, E, H, K), parvalbumin (C, I) and CGRP (F, L). Panels A–C, D–F, G–I and J–L show the same fields of view, respectively. A TG neuron retrogradely labeled from the facial skin with FG (arrows in A, D) co-expresses Kv1.2 with parvalbumin (arrows in B, C), but not CGRP (arrowheads in E, F). Arrowheads in A–C point to Kv1.2-ir cutaneous neurons which are devoid of parvalbumin-ir. Co-expression of Kv1.2 with parvalbumin or CGRP was rare in tooth pulp TG neurons. Arrows in G–I point to a tooth pulp TG neuron which is immunoreactive for both Kv1.2 and parvalbumin. Arrowheads in J–L point to a tooth pulp TG neuron which exhibits Kv1.2-ir but not CGRP-ir. Bar represents 50 mm (A). All panels are at the same magnification.
H. Ichikawa, T. Sugimoto / Brain Research 974 (2003) 222–227
225
226
H. Ichikawa, T. Sugimoto / Brain Research 974 (2003) 222–227
that 9.3% (31 / 332) of cutaneous neurons co-expressed Kv1.2- and parvalbumin-ir (Fig. 3A–C). Some 25% (31 / 124) of Kv1.2-ir cutaneous neurons contained parvalbumin-ir, whereas 83.8% (31 / 37) of parvalbumin-ir cutaneous neurons showed Kv1.2-ir. Only 2.9% (10 / 344) of cutaneous neurons co-expressed Kv1.2- and CGRP-ir (Fig. 3D–F). Some (7% 10 / 144) of Kv1.2-ir cutaneous neurons had CGRP-ir and 11.1% (10 / 90) of CGRP-ir cutaneous neurons expressed Kv1.2-ir (Fig. 3G–L). On the other hand, 18.9% (14 / 74) and 52% (26 / 50) of tooth pulp neurons were immunoreactive for parvalbumin and CGRP, respectively. However, co-expression of Kv1.2 with parvalbumin- (1.4% or 1 / 74) or CGRP-ir (0 / 50) was rare in tooth pulp neurons.
Fig. 4. Histogram showing the cell size spectrum of Kv1.2-ir and -immunonegative neurons in cutaneous TG neurons. The data were obtained from 676 cutaneous neurons.
(small: 6.5% or 8 / 124; medium-sized: 60.5% or 75 / 124; and large: 33.1% or 41 / 124) (Fig. 3A,B,D,E,G,H,J,K). Retrograde tracing and immunohistochemical methods revealed that 39.6% (268 / 676) and 16.1% (20 / 124) of TG neurons innervating the facial skin and tooth pulp, respectively, contained Kv1.2-ir. As shown in Fig. 4, 64.1% (98 / 153) of large and 44.1% (156 / 354) of medium-sized cutaneous neurons were immunoreactive for the channel. Only 8.3% (14 / 169) of small cutaneous neurons showed ir. On the other hand, 14.6% (6 / 41) of large and 18.7% (14 / 75) of medium-sized tooth pulp neurons showed Kv1.2-ir. Small tooth pulp neurons were devoid of ir (0 / 8) (Fig. 5). The double immunofluorescence method demonstrated
Fig. 5. A histogram showing the cell size spectrum of Kv1.2-ir and -immunonegative neurons in tooth pulp TG neurons. The data were obtained from 124 tooth pulp neurons.
4. Discussion The present study demonstrated that Kv1.2 was localized to primary sensory neurons in the TG. Such neurons were mostly medium-sized to large. Small Kv1.2-ir neurons were rare in the TG. The cell size finding in the TG is similar to that reported for the DRG [16]. In the DRG, Kv1.2-ir neurons are thought to be proprioceptors and low-threshold mechanoreceptors. However, most Kv1.2-ir TG neurons are unlikely to be proprioceptors because the number of proprioceptors is small in the TG. In addition, the present double immunofluorescence method revealed that Kv1.2-ir neurons were immunoreactive for parvalbumin. Therefore, low-threshold mechanoreceptors probably contain Kv1.2-ir in the TG. The present study also demonstrated that cutaneous and tooth pulp TG neurons showed Kv1.2-ir. Because the tooth pulp is considered to be innervated exclusively by nociceptive afferents in the TG, Kv1.2-ir neurons probably include nociceptors in this ganglion. Tooth pulp primary nociceptors are different from those innervating facial skin. Unlike facial nociceptors, they often contain calcium-binding proteins such as parvalbumin and calbindin D-28k [6–8]. In contrast, those containing substance P are relatively rare [2]. These characteristics probably reflect the abundance of medium-sized to large nociceptors in the tooth pulp. This idea is supported by the present finding that tooth pulp neurons are usually larger than cutaneous neurons. The tooth pulp is enclosed in layers of hard tissues and highly protected from physical insults. Nevertheless, the dentoenamel junction, which does not receive dense innervation, is extremely sensitive to mechanical stimulation. The channel, which is otherwise involved in low-threshold mechanoception, may be involved in dentinal nociception. In conclusion, we have described Kv1.2-ir neurons in the TG. Immunoreactivity was detected in half the TG neurons. These neurons were mostly medium-sized to large. More than half the Kv1.2-ir neurons showed parvalbumin-ir, whereas only 2% of such neurons contained CGRP-ir. On the other hand, the proportion of Kv1.2-ir
H. Ichikawa, T. Sugimoto / Brain Research 974 (2003) 222–227
neurons among cutaneous neurons (40%) was higher than among tooth pulp neurons (16%). Kv1.2-ir TG neurons probably include low-threshold mechanoreceptors and nociceptors which innervate the facial skin and tooth pulp, respectively.
[8]
[9]
Acknowledgements [10]
This work was supported by a grant from the Japanese Ministry of Education, Culture, Sports, Science and Technology to H.I. (No. 14571733). [11]
References [1] M.R. Celio, Calbindin D-28k and parvalbumin in the rat nervous system, Neuroscience 35 (1990) 375–475. [2] K. Fried, J. Arvidsson, B. Robertson, E. Brodin, E. Theodorsson, Combined retrograde tracing and enzyme / immunohistochemistry of trigeminal ganglion cell bodies innervating tooth pulps in the rat, Neuroscience 33 (1989) 101–109. [3] H. Ichikawa, T. Sugimoto, Parvalbumin- and calbindin D-28kimmunoreactive innervation of oro-facial tissues in the rat, Exp. Neurol. 146 (1997) 414–418. [4] H. Ichikawa, D.M. Jacobowitz, T. Sugimoto, Coexpression of calretinin and parvalbumin in Ruffini-like endings in the rat incisor periodontal ligament, Brain Res. 770 (1997) 294–297. [5] H. Ichikawa, T. Deguchi, T. Nakago, D.M. Jacobowitz, T. Sugimoto, Parvalbumin, calretinin and carbonic anhydrase in the trigeminal and spinal primary neurons of the rat, Brain Res. 655 (1994) 241–245. [6] H. Ichikawa, T. Deguchi, T. Nakago, D.M. Jacobowitz, T. Sugimoto, Parvalbumin- and calretinin-immunoreactive trigeminal neurons innervating the rat molar tooth pulp, Brain Res. 679 (1995) 205–211. [7] H. Ichikawa, T. Deguchi, Y. Fujiyoshi, T. Nakago, D.M. Jacobowitz,
[12]
[13]
[14]
[15]
[16]
[17]
227
T. Sugimoto, Calbindin-D28k-immunoreactivity in the trigeminal ganglion neurons and molar tooth pulp of the rat, Brain Res. 715 (1996) 71–78. H. Ichikawa, T. Deguchi, S. Mitani, T. Nakago, D.M. Jacobowitz, T. Yamaai, T. Sugimoto, Neural parvalbumin and calretinin in the tooth pulp, Brain Res. 647 (1994) 124–130. A. Ishida-Yamamoto, E. Senba, M. Tohyama, Distribution and fine structure of calcitonin gene-related peptide-like immunoreactive nerve fibers in the rat skin, Brain Res. 491 (1989) 93–101. ¨ G. Ju, T. Hokfelt, E. Brodin, J. Fahrenkrug, J.A. Fischer, P. Frey, R.P. Elde, J.C. Brown, Primary sensory neurons of the rat showing calcitonin gene-related peptide immunoreactivity and their relation to substance P-, somatostatin-, galanin-, vasoactive intestinal polypeptide- and cholecystokinin-immunoreactive ganglion cells, Cell Tissue Res. 247 (1987) 417–431. L. Kruger, J.D. Silverman, P.W. Mantyh, C. Sternini, N.C. Brecha, Peripheral patterns of calcitonin gene-related peptide general somatic sensory innervation: cutaneous and deep terminations, J. Comp. Neurol. 280 (1989) 291–302. M.N. Rasband, E.W. Park, T.W. Vanderah, J. Lai, F. Porreca, J.S. Trimmer, Distinct potassium channels on pain-sensing neurons, Proc. Natl. Acad. Sci. USA 98 (2001) 13373–13378. M. Sheng, M.L. Tsaur, Y.N. Jan, L.Y. Jan, Contrasting subcellular localization of the Kv1.2 K 1 channel subunit in different neurons of rat brain, J. Neurosci. 14 (1994) 2408–2417. J.D. Silverman, L. Kruger, Calcitonin gene-related-peptide-immunoreactive innervation of the rat head with emphasis on specialized sensory structures, J. Comp. Neurol. 280 (1989) 303–330. G. Skofitsch, D.M. Jacobowitz, Calcitonin gene-related peptide coexists with substance P in capsaicin sensitive neurons and sensory ganglia of the rat, Peptides 6 (1985) 747–754. R.W. Veh, R. Lichtinghagen, S. Sewing, F. Wunder, I.M. Grumbach, O. Pongs, Immunohistochemical localization of five members of the Kv1 channel subunits: contrasting subcellular locations and neuronspecific co-localizations in rat brain, Eur. J. Neurosci. 7 (1995) 2189–2205. S. Yokoyama, H. Takeda, H. Higashida, Expression of Kv1.2 potassium channels in rat sensory ganglia. An immunohistochemical study, Ann. NY Acad. Sci. 868 (1999) 454–457.