osteopontin-immunoreactivity in the rat trigeminal ganglion and trigeminal sensory nuclei

Brain Research 919 (2001) 147–154www.elsevier.com/ locate /bres

Research report

Osteopontin-immunoreactivity in the rat trigeminal ganglion andtrigeminal sensory nuclei

a,b , c c*Hiroyuki Ichikawa , Kazuo Yamashita , Teruko Takano-Yamamoto ,a,bTomosada Sugimoto

aDepartment of Oral Function and Anatomy, Okayama University, Graduate School of Medicine and Dentistry, 2-5-1 Shikata-cho,Okayama 700-8525, Japan

bBiodental Research Center, Okayama University, Graduate School of Medicine and Dentistry, 2-5-1 Shikata-cho, Okayama 700-8525, JapancDepartment of Orthodontics, Okayama University, Graduate School of Medicine and Dentistry, 2-5-1 Shikata-cho, Okayama 700-8525, Japan

Accepted 17 August 2001

Abstract

Osteopontin-immunoreactivity (OPN-ir) was examined in the oro-facial tissues and trigeminal sensory nuclei (principal sensory nucleusand spinal trigeminal nucleus) to ascertain the peripheral ending and central projection of OPN-containing primary sensory neurons in thetrigeminal ganglion (TG). No staining was observed using mouse monoclonal anti-OPN antibody preabsorbed with recombinant matureOPN. OPN-immunoreactive (ir) peripheral endings were classified into two types: encapsulated and unencapsulated types. Unencapsu-lated endings were subdivided into two types: simple and complex types. Simple endings were characterized by the thin neurite that wasusually devoid of ramification. These endings were seen in the hard plate and gingiva. The complex type was characterized by the thickramified neurite, and observed in the vibrissa, hard palate, and molar periodontal ligament. Encapsulated endings were found only in thehard palate. The trigeminal sensory nuclei contained OPN-ir cell bodies and neuropil. The neuropil was devoid of ir in laminae I and II ofthe medullary dorsal horn (MDH), and had various staining intensities in other regions of the trigeminal sensory nuclei. Transection of theinfraorbital and inferior alveolar nerves caused an increase of OPN-ir intensity in ipsilateral TG neurons. The staining intensity of theneuropil also increased in the trigeminal sensory nuclei ipsilateral to the neurotomy excepting laminae I and II of the MDH. The presentstudy indicates that OPN-ir primary sensory neurons in the TG innervate encapsulated and unencapsulated corpuscular endings. Suchneurons probably project their central terminals to the trigeminal sensory nuclei except for the superficial laminae of the MDH. 2001Elsevier Science B.V. All rights reserved.

Theme: Sensory systems

Topic: Somatic and visceral afferents

Keywords: Immunohistochemistry; Inferior alveolar nerve; Infraorbital nerve; Osteopontin; Sensory nerve endings; Trigeminal ganglion; Trigeminalsensory nuclei

1. Introduction neurons contain OPN-immunoreactivity (ir). Such neuronssend their myelinated axons to the incisor periodontal

Osteopontin (OPN), a bone matrix protein, has calcium- ligament and palate, and terminate as bush-like endings.binding residues of amino acids [2,10]. This calcium- Therefore, OPN-ir TG neurons are considered to includebinding protein has been shown to be expressed by several low-threshold mechanoreceptors. On the other hand, pri-tissues including, bone and kidney [1,8]. Recently, we mary sensory neurons in the TG supply wide varieties ofreported that OPN was localized to primary sensory tissues in the oro-facial region [3–6,12–15]. These includeneurons in the spinal and trigeminal nervous systems [3]. the facial skin, vibrissal follicles, lip, palate and gingiva.In the trigeminal ganglion (TG), medium-sized to large The tissues have sensory nerve endings such as free nerve

endings, and encapsulated and unencapsulated corpuscularendings [4–6,12–15]. However, little is known about*Corresponding author. Fax: 181-86-235-6612.

E-mail address: [email protected] (H. Ichikawa). OPN-ir sensory nerve endings other than incisal and

0006-8993/01/$ – see front matter 2001 Elsevier Science B.V. All rights reserved.PI I : S0006-8993( 01 )03019-0

148 H. Ichikawa et al. / Brain Research 919 (2001) 147 –154

palatal bush-like endings. In addition, the central projec- gram was used to measure the optical density of thetion site of OPN-ir TG neurons has never been reported. subnuclei and background (immunonegative area) in each

In the search for peripheral endings and central projec- section. The density of the subnuclei in the ipsilateral andtion of OPN-ir TG neurons, we examine the oro-facial contralateral sides was divided by the background density.tissues and the trigeminal sensory nuclei (principal sensory The quotient was recorded for each subnucleus and will benucleus and spinal trigeminal nucleus) for OPN-ir. The referred to as the ir density, hereafter.change of OPN-ir was also investigated in the TG and The data for the ir density were obtained from 16trigeminal sensory nuclei after transection of the infraorbi- sections of each subnucleus from four animals, and thetal (ION) and inferior alveolar nerves (IAN). difference between the ipsilateral and contralateral sides

were statistically analyzed with a t-test for paired samples.For the control, mouse monoclonal anti-OPN antibody was

2. Materials and methods preabsorbed with recombinant mature OPN (20 mg/ml,R&D Systems, Minneapolis, MN, USA). No staining was

Four TGs, four brainstems containing the trigeminal observed in the control.sensory nuclei, facial skins, vibrissal pads, hard palates, The experiments were carried out under the control ofmaxillae and mandibles were obtained from six male the Animal Research Control Committee in accordanceSprague–Dawley rats (200–300 g). The rats were anes- with The Guidelines for Animal Experiments of Okayamathetized with ether to the level at which respiration was University Medical School, Government Animal Protec-markedly suppressed, and transvascularly perfused with 50 tion and Management Law (No. 105), and Japaneseml of saline followed by 500 ml of 4% formaldehyde in Government Notification on Feeding and Safekeeping of0.1 M phosphate buffer (pH 7.4). The materials were Animals (No. 6). All efforts were made to minimize thedissected and post-fixed with the same fixative for 30 min. number of animals used and their suffering.Maxillae and mandibles were decalcified with 4.13%ethylene diaminetetraacetic acid disodium salt in 0.1 Mphosphate buffer (pH 7.4) for 1 week at room temperature. 3. Results

Materials were soaked in a phosphate-buffered 20%sucrose solution overnight, frozen sectioned at 12 mm, and OPN-ir in the TG and peripheral sensory nerve endings.thaw-mounted on gelatin-coated glass slides. The brain- As described previously [3], the TG contained abundantstem was also cut at 50 mm and treated as free-floating OPN-ir neurons (Fig. 1A). Ir granules were distributedsections. Sections were incubated with mouse monoclonal within the cytoplasm of these neurons. They were mostlyanti-rat OPN antibody (1:20000, American Research Prod- large or medium-sized and small OPN-ir neurons wereucts Inc., Belmont, MA, USA) for 24 h at room tempera- very rare. Their axons were also immunoreactive for OPNture, followed by biotinylated horse anti-mouse IgG and (Fig. 1A).avidin–biotin–horseradish peroxidase complex (Vector OPN-ir nerve endings were observed in the vibrissal padLaboratories Inc., Burlingame, CA, USA). Following and intraoral structures. These endings were classified intonickel ammonium sulfate-intensified diaminobenzidine re- two types: unencapsulated and encapsulated types.action, the sections were dehydrated in a graded series of The unencapsulated endings were subdivided into twoalcohols, cleared in xylene and cover-slipped with Entellan types: complex and simple types. The complex type was(Merck, Darmstadt, Germany). characterized by the bush-like ramification. This type of

To know the effect of axotomy on OPN-ir in the TG and endings consisted of a thick neurite with two to six twigsbrainstem, the right ION and IAN were transected in five in the distal end (Fig. 1C, D). They were observed inanimals under deep anesthesia by i.p. injection with ethyl vibrissal follicles and molar periodontal ligaments. Incarbamate (650 mg/kg) and pentobarbital sodium (20 vibrissal follicles, OPN-ir nerve fibers ran from the base ofmg/kg). After 7 days, the animals were reanesthetized follicles to the upper portion, and ramified toward the hairwith ether, and transvascularly perfused with 4% formalde- shaft beneath sebaceous glands (Fig. 1B, C). The morphol-hyde. The TG and brainstem were frozen-sectioned, and ogy of complex endings in the molar periodontal ligamentstained for OPN-ir as described above. For cell size was similar to that of corpuscular endings which have beenanalysis of OPN-ir neurons in the ipsilateral and contrala- described in the hard palate and incisor periodontal liga-teral TGs, the microscopic image (3215) of the cell bodies ment (Fig. 1D) [3–6]. Simple endings were characterizedwas projected over a digitizer tablet using a drawing tube. by the thin neurite that were usually devoid of ramification.The cross-sectional area of those cell bodies that contained In the hard palate and gingiva, thin OPN-ir fibers formedthe nucleolus was recorded. For the image analysis of the subepithelial nerve plexuses. Some ir fibers separated fromtrigeminal sensory nuclei, the optic image of the nuclei the plexuses and penetrated the mucous epithelium. Withinwas captured with a digital camera (COOLPIX990, Nikon, the epithelium, they formed terminal swellings withoutTokyo, Japan) and stored into a computer (Power Macin- reaching the stratum corneum (Fig. 1E).tosh G4, Apple, Cupertino, USA). The NIH Image pro- Encapsulated endings were found only in the hard

H. Ichikawa et al. / Brain Research 919 (2001) 147 –154 149

Fig. 1. Microphotographs for OPN in the TG (A, G, H) and oro-facial tissues (B–F). In the TG, many large or medium-sized neurons show OPN-ir (A). Irgranules are distributed within the cytoplasm of these neurons. Arrows in (A) point to an ir axon derived from a TG neuron. Many OPN-ir nerve fibers areobserved at the base of the vibrissal follicle (B). They run toward the upper portion of the follicle (arrows in B) and ramify toward the hair shaft (arrows inC). In the molar periodontal ligament, OPN-ir nerve fibers make ramification and form the complex type of endings (arrows in D). The hard palate containssimple (E) and encapsulated nerve endings (F). Transection of the ION and IAN causes the increase of OPN-ir intensity in the ipsilateral TG (H) comparedto the contralateral TG (G). In the ipsilateral TG, the ir is detected throughout the cytoplasm of numerous neurons (arrows in H) and ir axons appear toincrease in number. Bars550 mm (A, C, D, E, F, G, H) and 200 mm (B).


palate. At the most posterior part of the hard palate, similar in the contralateral and normal TGs (Fig. 1A, G).slightly elongated capsules were located within the palatal OPN-ir in the trigeminal sensory nuclei.salivary gland. In the capsules, OPN-ir thick neurites ran In the trigeminal sensory nuclei, OPN-ir was detected instraight and terminal swellings were occasionally observed two distinct structures; neuronal cell bodies including(Fig. 1F). perikarya and proximal dendrites, and the neuropil. The

Transection of the ION and IAN caused the increase of nuclei contained various numbers of OPN-ir cell bodiesOPN-ir intensity in the ipsilateral TG (Fig. 1G, H). In (figure not shown but see Figs. 3A,C and 4A, C for themany ir cell bodies, the intense stain obscured the granular nuclei contralateral to the neurotomy). The medullaryappearance of the ir. In addition, the number of OPN-ir dorsal horn (MDH) was almost devoid of ir cell bodiesthick axons increased (Fig. 1G, H). However, the pro- except for lamina V where some large neurons showed theportions of OPN-ir neurons among all TG neurons in the ir. In the rostral parts of the trigeminal sensory nucleiipsilateral (mean6S.D.540.868.5%, range531.5–54.6%) (subnuclei interpolaris, oralis and principalis), OPN-irand contralateral TGs (mean6S.D.538.567.4%, range5 neurons were abundant and their cell bodies had various26.5–47.2%) were very similar and the difference was not sizes, i.e. those in the ventral parts of subnuclei interpolarissignificant (P.0.95, Mann–Whitney). The cell size spec- and oralis were large while those in the subnuclei prin-trum of OPN-ir TG neurons appeared to remain unchanged cipalis and the dorsal parts of subnuclei interpolrais andafter the neurotomy (Fig. 2). OPN-ir TG neurons were oralis were small. The neuropil was devoid of OPN-ir inmostly large in the ipsilateral (mean6S.D.51191.06454.6 laminae I and II of the MDH, and weakly stained in

2 2mm , range5145.7–3008.8 mm ) and contralateral laminae III and IV. The ir of neuropil was moderate in

2(mean6S.D.51174.86407.1 mm , range5206.4–3115.4 other regions of the trigeminal sensory nuclei. Many axons2

mm ) TGs. In these TGs, most OPN-ir neurons were larger showed the ir in the spinal trigeminal tract and trigeminal2than 800 mm (ipsilateral TG; 84.7% or 266/314, contrala- sensory root.

2teral TG; 81.6% or 252/309) and those ,400 mm were The ir density of the trigeminal sensory nuclei ipsilateralvery rare (ipsilateral TG; 1.9% or 6/309, contralateral TG; to the neurotomy was always higher than that on the1.3% or 4/314). 16.2% (50/309) and 14.0% (44/314) of contralateral side (Table 1). The distribution and stainingOPN-ir neurons in the ipsilateral and contralateral TGs, intensity of OPN-ir cell bodies were very similar between

2respectively, measured 400–800 mm . The distributions of the ipsilateral and contralateral sides (Figs. 3A–D andOPN-ir neurons and the intensity of their ir were very 4A–D). However, the intensity of OPN-ir of neuropil

Fig. 2. Histograms showing the cell size spectrum of OPN-ir TG neurons after the ION and IAN transection. The data were obtained from 309 ipsilateraland 314 contralateral TG neurons.


Fig. 3. Microphotographs for OPN-ir in the MDH (A, B) and subnucleus interpolaris (C, D) contralateral (A, C) and ipsilateral (B, D) to the transection ofthe ION and IAN. On the contralateral side, the neuropil is weakly stained in the deep layer of the MDH and moderately in the subnucleus interpolaris. Thestaining intensity of OPN-ir neuropil increases in the MDH and subnucleus interpolaris ipsilateral to the neurotomy (A–D). However, the neuropil oflaminae I and II of the MDH is devoid of the ir on both sides (asterisks in A and B). OPN-ir cell bodies are rare in the MDH (A) and abundant in thesubnucleus interpolaris (C) contralateral to the neurotomy. The distribution and staining intensity of these neurons are similar between the ipsilateral andcontralateral sides (A–D). Arrows in A and B indicate the dorsal medullary reticular field that contains some OPN-ir neurons. A bar in (A)5200 mm. Allpanels are at the same magnification.


Fig. 4. Microphotographs for OPN-ir in the subnuclei oralis (A, B) and nucleus principalis (C, D) contralateral (A, C) and ipsilateral (B, D) to thetransection of the ION and IAN. The neuropil is moderately stained in the subnuclei oralis (A) and principalis (C) contralateral to the neurotomy. Thestaining intensity of OPN-ir neuropil increases in the MDH and subnucleus interpolaris ipsilateral to the neurotomy (A–D). OPN-ir cell bodies areabundant in these subnuclei contralateral to the neurotomy (A, C). The distribution and staining intensity of these neurons are similar between theipsilateral and contralateral sides (A–D). Arrows point to the facial (A, B) and trigeminal motor nuclei (C, D) that contain many OPN-ir neurons.Arrowheads in (C) and (D) point to OPN-ir axons presumably originated from the trigeminal motor nucleus. A bar in (A)5200 mm. All panels are at thesame magnification.


Table 1 TG are the candidates for the origin of the ir neuropil inMean numbers and standard errors of the OPN-ir density in the trigeminal the trigeminal sensory nuclei. The increase in ir intensitysensory nuclei after transection of the ION and IAN

of the trigeminal sensory nuclei neuropil followingIpsilateral side Contralateral side peripheral neurotomy reflected the increase in staining(range) (range) intensity of neurons in the TG and in number of ir axons in

aMDH 1.6060.12 1.3460.07 the spinal trigeminal tract. Moreover, the ir in the neuropil(1.37–1.75) (1.21–1.45) was seen throughout the trigeminal sensory nuclei except-aSubnucleus interpolaris 1.7860.08 1.5360.08

ing laminae I and II of the MDH. The distribution is(1.61–1.90) (1.43–1.70)a similar to the known distribution of central terminals ofSubnucleus oralis 1.7960.08 1.6360.09

(1.67–1.93) (1.44–1.77) large myelinated TG neurons labeled with cholera toxin BaSubnucleus principalis 1.8860.19 1.6360.15 subunit [11], and is consistent with the morphology of ir

(1.69–2.36) (1.43–1.97) nerve endings in the periphery that suggested the OPN-irDifference between ipsilateral and contralateral sides was highly signifi- TG neurons were low-threshold mechanoreceptors.cant. The data were obtained from 16 sections of each subnucleus from In conclusion, we have described the peripheral endingfour animals. and central projection of OPN-ir primary sensory neuronsa P,10–7, a t-test for paired samples.

in the TG. These neurons innervate encapsulated andunencapsulated corpuscular endings. They probably project

increased in laminae III and IV of the MDH and the rostral their central terminals to the trigeminal sensory nucleiparts of the trigeminal sensory nuclei ipsilateral to the outside the superficial laminae of the MDH.injury (Figs. 3A–D and 4A–D). The difference of themean density of OPN-ir between the ipsilateral andcontralateral sides for each subnucleus was highly signifi- Referencescant (P,10–7, a t-test for paired samples). The neuropil oflaminae I and II of the MDH were devoid of OPN-ir even [1] W.T. Butler, The nature and significance of osteopontin, Connect.after the neurotomy (Fig. 3B). The number of ir axons Tissue Res. 23 (1989) 123–136.

[2] D.T. Denhardt, X. Guo, Osteopontin; a protein with diverse func-increased after the neurotomy in the spinal trigeminal tracttions, FASEB J. 7 (1993) 1475–1482.and trigeminal sensory root ipsilateral to the neurotomy.

[3] H. Ichikawa, T. Itota, Y. Nishitani, Y. Torii, K. Inoue, T. Sugimoto,Osteopontin-immunoreactive primary sensory neurons in the ratspinal and trigeminal nervous systems, Brain Res. 863 (2000)

4. Discussion 276–281.[4] H. Ichikawa, D.M. Jacobowitz, T. Sugimoto, Coexpression of

calretinin and parvalbumin in Ruffini-like endings in the rat incisorThe present study demonstrated OPN-ir nerve endings inperiodontal ligament, Brain Res. 770 (1997) 294–297.the oro-facial tissues. In vibrissal follicles and intraoral

[5] H. Ichikawa, T. Sugimoto, Parvalbumin- and calbindin D-28k-structures, encapsulated and unencapsulated complex types immunoreactive innervation of oro-facial tissues in the rat, Exp.of endings contained the ir. These endings resembled the Neurol. 146 (1997) 414–418.so-called corpuscular endings. Previous physiological [6] H. Ichikawa, C. Xiao, Y.-F. He, T. Sugimoto, Parvalbumin-immuno-

reactive nerve endings in the periodontal ligaments of rat teeth,studies have demonstrated that encapsulated and unen-Arch. Oral Biol. 41 (1996) 1087–1090.capsulated corpuscular endings are low-threshold mech-

[7] R.W. Linden, B.J. Millar, Z. Halata, A comparative physiologicalanorecetors in the TG [7,16]. Therefore, the complex type and morphological study of periodontal ligament mechanoreceptorsof endings are probbaly low-threshold mechanoreceptors represented in the trigeminal ganglion and the mesencephalic[3–6,13,14]. In addition, the simple type of unencapsulated nucleus of the cat, Anat. Embryol. 190 (1994) 127–135.

[8] C.A. Loez, E.S. Olson, J.C. Adams, K. Mou, D.T. Denhardt, R.L.endings was observed in palatal and gingival mucosae.Davis, Osteopontin expression detected in adult cochleae and innerBoth this type of endings and the so-called free nerveear fluids, Hearing Res. 85 (1995) 210–222.

endings are characterized by thin neurites [12,14,15]. [9] S.L. Shin, J.H. Cha, M.H. Chun, J.W. Chung, M.Y. Lee, ExpressionUnlike free nerve endings, however, the simple endings of osteopontin mRNA in the adult rat brain, Neurosci. Lett. 273were mostly devoid of ramification. Simple endings and (1999) 73–76.

[10] K. Singh, D. Deonarine, V. Shanmugam, D.R. Senger, A.B. Mukher-free nerve endings may have different sensory modalities.jee, P.L. Chang, C.W. Prince, B.B. Mukherjee, Calcium-bindingIf this was the case, it is unlikely that OPN-ir neurons areproperties of osteopontin derived from non-osteogenic sources, J.

nociceptors in the TG. Biochem. 114 (1993) 702–707.We have also demonstrated that neuronal cell bodies and [11] T. Sugimoto, Y. Fujiyoshi, Y.F. He, C. Xiao, H. Ichikawa, Trigemi-

neuropil showed OPN-ir in the trigeminal sensory nuclei. nal primary projection to the rat brain stem sensory trigeminalnuclear complex and surrounding structures revealed by anterogradeThe distribution of OPN-ir cell bodies is consistent with atransport of cholera toxin B subunit-conjugated and Bandeiraeaprevious observation by in situ hybridization histochemis-simplicifolia isolectin B4-conjugated horseradish peroxidase, Neuro-

try for OPN mRNA [9]. It is unlikely, however, that the sci. Res. 28 (1997) 361–371.neuropil ir is solely derived from the local ir neurons [12] L. Kruger, J.D. Silverman, P.W. Mantyh, C. Sternini, N.C. Brecha,within the nuclei. OPN-ir primary sensory neurons in the Peripheral patterns of calcitonin gene-related peptide general


somatic sensory innervation: cutaneous and deep terminations, J. [15] J.D. Silverman, L. Kruger, Calcitonin gene-related-peptide-immuno-Comp. Neurol. 280 (1989) 291–302. reactive innervation of the rat head with emphasis on specialized

[13] F.L. Rice, B.T. Fundin, J. Arvidsson, H. Aldskogius, O. Johansson, sensory structures, J. Comp. Neurol. 280 (1989) 303–330.Comprehensive immunofluorescence and lectin binding analysis of [16] T. Yamamoto, S. Ozono, K. Watanabe, S. Nagasaki, M. Onozuka,vibrissal follicle sinus complex innervation in the mystacial pad of Structural and functional maturation of the buccal stretch receptorsthe rat, J. Comp. Neurol. 385 (1997) 149–184. in rats, Exp. Brain Res. 111 (1996) 169–177.

[14] E. Schulze, M. Witt, T. Fink, A. Hofer, R.H. Funk, Immuno-histochemical detection of human skin nerve fibers, Acta Histochem.99 (1997) 301–309.

osteopontin-immunoreactivity in the rat trigeminal ganglion and trigeminal sensory nuclei

Documents