Archives of Oral Biology (2006) 51, 681—688

www.intl.elsevierhealth.com/journals/arob

Requirement of proper occlusal force formorphological maturation of neural componentsof periodontal Ruffini endings of the rat incisor

Lei Shi, Yukako Atsumi, Yumiko Kodama, Shiho Honma, Satoshi Wakisaka *

Department of Oral Anatomy and Developmental Biology, Osaka University Graduate School ofDentistry, 1-8 Yamadaoka, Suita, Osaka 565-0871, Japan

Accepted 15 February 2006

KEYWORDSDevelopment;Periodontal Ruffiniendings;Mechanical stimulation(occlusal force);Immunohistochemistry

Summary The present study examined the effect of reduced occlusal force onmorphological maturation of periodontal Ruffini endings, primary mechanoreceptorsin the periodontal ligament, of the rat incisor. The reduction of occlusal force wasinduced by grinding the cutting edges of unilateral incisors of the rat from postnatalday 14 (PN14d), when periodontal Ruffini endings are immature. Under normaldevelopment, the axon terminals of Ruffini endings gradually ramified with thepassage of time, and showed ruffled outlines having numerous dot-like structuresaround PN28d. When themechanical stimulation was reduced, appearance of dot-likestructures at the axon terminals delayed. Quantitative analysis elucidated that thepercentages of immunoreactive areas for protein gene product 9.5, a marker proteinof neural elements, at ground side were significantly smaller than those at non-groundside 14 days following the initial grinding. The distribution andmorphology of terminalSchwann cells was not apparently affected. The present results indicate that theproper mechanical stimulation to the ligament contributes to the morphologicalmaturation of the periodontal Ruffini endings.# 2006 Elsevier Ltd. All rights reserved.

The periodontal ligament is the dense connectivetissue between the tooth and alveolar bone, sup-porting and anchoring the tooth to the alveolarsocket. There are at least two types of nerve end-ings in the periodontal ligament; nociceptive freenerve endings, and mechanoreceptive specializednerve endings. Despite of the presence of various

* Corresponding author. Tel.: +81 6 6879 2871;fax: +81 6 6879 2875.

E-mail address: wakisaka@dent.osaka-u.ac.jp (S. Wakisaka).

0003–9969/$ — see front matter # 2006 Elsevier Ltd. All rights resedoi:10.1016/j.archoralbio.2006.02.006

kinds of mechanoreceptors in the periodontal liga-ment,1—3 many studies have shown that the Ruffiniending is the primary mechanoreceptor in the liga-ment.4—8 Morphologically, the axonal elements ofthe periodontal Ruffini endings are characterizedby extensive ramification, and the association withspecialized Schwann cells called lamellar or term-inal Schwann cells which possess kidney-shapednuclei.4—6,8 Unlike cutaneous Ruffini endings,periodontal Ruffini endings lack in collagenousencapsulation.4 The periodontal nociceptive nerve

rved.

682 L. Shi et al.

fibres, i.e. free nerve endings, are distributed uni-formly in the ligament, whilst the periodontal Ruf-fini endings are localized unevenly in the ligament,i.e. localized where the periodontal fibres can beeasily extended in response to the occlusion. In theligament of the molar teeth, Ruffini endings areabundantly present in the apical third of the liga-ment, whilst in the incisor of the rodent, they arelocalized in the alveolar half of the lingual period-ontal ligament.9,10 The sensory receptor in theperiodontal ligament is often used as an experi-mental model to study the effect of changes inmechanical stimulation to the ligament on themorphology of sensory receptors since the mechan-ical stimulation to the ligament, i.e. occlusal force,is easily controlled.

Mechanisms controlling the development, main-tenance and regeneration of periodontal Ruffiniendings have not been fully understood. Recentlythe involvement of neurotrophic factors, such asbrain-derived neurotrophic factor (BDNF) has beenproposed for the development, maintenance andregeneration of the periodontal Ruffini endings.11—13 In addition, environmental factors might alsocontribute to the development and maintenance ofthe periodontal Ruffini endings. Developmentalstudies revealed that the morphological matura-tion of periodontal Ruffini endings is closelyrelated to the tooth eruption; the terminal arbor-ization of the periodontal Ruffini ending of theincisor takes place after the eruption of incisors,and completes shortly after the commencement ofthe molar occlusion.14—17 In addition, experimen-tally induced occlusal trauma caused rapid rear-rangement of the periodontal nerve fibres of theadult molar teeth,18,19 indicating that the properocclusal force is important to maintain the mor-phology of the periodontal nerve fibres in the adultanimal. Our previous study also demonstrated thatthe reduction of occlusal force altered the mor-phology of the axon terminals of periodontal Ruf-fini endings of the incisor of the adult rat.20 Thesemorphological data lead us to speculate that theproper mechanical stimulation is also required forthe morphological maturation of periodontal Ruf-fini endings during development. To our knowl-edge, however, there is no report on the effectsof the alteration in occlusal force on the develop-ment of the periodontal mechanoreceptors. Thepresent study, therefore, was conducted to exam-ine the effects of reduced occlusal force on thedevelopment of periodontal Ruffini endings of therat incisor by immunohistochemistry of proteingene product 9.5 (PGP 9.5),21,22 a general neuronalmarker, and S-100 protein, a marker protein forSchwann cell.

Materials and methods

Animals and treatment

Male Sprague—Dawley rats at postnatal day 14(PN14d; n = 27) were divided into nine groups (threeexperimental and six control groups; n = 3 each).The day of birth was designated at PN0d. The pro-tocol for grinding was almost identical to our pre-vious study.20 Briefly, under chloral hydrateanaesthesia (400 mg/kg, i.p., supplemented asnecessary), the cutting edges of right upper andlower incisors were ground unilaterally every secondday by using a diamond bur with continuous waterspray from PN14d. During the grinding, care wastaken not to damage the dental pulp and gingiva ofthe incisor. Experimental animals were allowed tosurvive for 7, 14 and 28 days after initial grinding.Two control experiments were performed; age-matched animals without any treatment served as‘‘normal control’’, and those with anaesthesia everysecond day without grinding served as ‘‘anaesthe-tized control’’. Animals were separated from theirmothers at PN21d. Body weight was measured everysecond day. All animal experiments were reviewedand approved by Intramural Animal Use and CareCommittee at the Osaka University Graduate Schoolof Dentistry.

Tissue preparation andimmunohistochemistry

The protocol for tissue preparation and immunohis-tochemistry was almost identical to our previousstudy.20 Briefly, following appropriate survival per-iod, animals were deeply anaesthetized with chloralhydrate (600 mg/kg, i.p.) and perfused with 0.02 Mphosphate-buffered saline (PBS; pH 7.4) and thenwith a mixture of 4% paraformaldehyde and 0.05%glutaraldehyde in 0.1 M phosphate buffer (PB, pH7.2). Maxillae were removed and further fixed in 4%paraformaldehyde in 0.1 M PB for 2—3 days. Follow-ing decalcification with Plank—Rychlo rapid decal-cifying solution and neutralization with 5% Na2SO4,specimens were cut at a thickness of 20 mm with acryostat, collected in PBS and treated as free-float-ing sections. Sections were treated with PBS con-taining 0.03% H2O2 to inactivate endogenousperoxidase activity for 30 min at room temperature.Following preincubation with PBS containing 3%normal swine serum (Dako, Copenhagen, Denmark)and 1% bovine serum albumin (BSA; Sigma, St. Louis,MO) for 30 min, sections were incubated either withpolyclonal anti PGP 9.5 (1:5000; Chemicon Interna-tional, Temecula, CA) or with polyclonal anti S100(1:5000; Dako) for 16—18 h at room temperature.

Occlusal force and Ruffini endings 683

Following several rinses in PBS, sections were incu-bated with swine biotinylated-anti-rabbit IgG(1:500; Dako), and subsequently with ABC complex(Vector Laboratories, Burlingame, CA) for 90 mineach at room temperature. The horseradish perox-idase activity was visualized by incubation with0.05 M Tris—HCl buffer (pH 7.5) containing 0.04%diamino benzidine (DAB) and 0.003% H2O2 enhancedwith 0.08—0.1% nickel ammonium sulphate. Theimmunostained sections were mounted on gelatin-subbed glass slides, counter-stained with methylgreen, dehydrated through an ascending series ofethanol, cleared with xylene and cover-slipped withPermount (Fisher Scientific Inc., New Jersey).

Immunohistochemical controls were performedby omission of primary antibody or biotinylatedsecondary antibody or ABC complex; they did notshow any immunoreactions.

Quantitative analysis

For quantification, percentages of immunoreactivearea at the restricted area were measured from fivesections per animal (total 15 sections for one experi-mental group) according to our protocol.16,20

Briefly, the site of the entrance of the thick nervebundle was located just below the centre of theboxed area (1950 mm � 650 mm), because they formRuffini endings immediately after entering the liga-ment. The total areas of periodontal ligamentexcluding tooth, alveolar bone and blood vessels,and immunoreactive area were measured using theScion Image software package (ver. 4.2; ScionImage, CA), and the percentages of the latter tothe former were calculated. The significance ofdifference was assessed by paired Student’s t-testfor comparison between the percentages of immmu-noreactive areas of the ground side and those ofnon-ground side of the same animals, and by one-way analysis of variance (ANOVA) and post hoccomparison (Fisher’s PLSD test) among the percen-tages of immunoreactive areas of control animals ofthose of experimental animals. A p-value less than0.05 was considered a significant difference.

Results

General observations

The time course of changes in the body weight ofanimals in the experimental group was almost iden-tical to that of age-matched control animals andanimals in the anaesthetized control group (data notshown). At the beginning of grinding, i.e. PN14d, theincisors have appeared in the oral cavity. In some

animals, cutting edges of upper and lower incisorsslightly touched each other, whilst other animals didnot. During the entire experimental period, thecutting edges of upper and lower incisors at groundside never touched. The occlusal pattern betweenupper and lower incisors of non-ground side was notthe same among individual animals. In some ani-mals, the tips touched each other as observed incontrol animals, whilst the incisors were elongatedin others.

Development of periodontal Ruffiniendings following grinding from PN14d

Since the rodent incisors erupt continuously, theperiodontal ligament of the incisor has unique his-tological properties. The lingual periodontal liga-ment of rodent incisors is divided into three parts.23

The alveolar half is called alveolus-related part(ARP) is the non-moving areas during the eruption,and contains many blood vessels and nerve fibres.The periodontal fibres in the ARP are loosely dis-tributed. The tooth half is termed tooth-relatedpart (TRP), and the periodontal fibres in this partare tightly arranged. In the TRP, neither vascula-tures nor neural elements are found under normalcondition. The border between the ARP and TRP iscalled as the shear zone.

At PN14d, thick PGP 9.5-immunoreactive nervefibres were detected in the ARP of the normalanimals (Fig. 1A). They ramified slightly, and showedsmooth outline; there was no or very few dot-likestructures at the terminal portions (Fig. 1A and B).The S-100-immunoreactive rounded cells extendinglong processes gathered at the ARP (Fig. 1C).

At PN21d, Ruffini endings of control animals rami-fied repeatedly, but less expanded (Fig. 2A). At theaxon terminals, a few dot-like structures wererecognized (Fig. 2A). At ground side 7days followinggrinding from PN14d, i.e. PN21d, the terminals ofRuffini endings were less expanded their terminalscompared to the age-matched control animals(Fig. 2B). The outlines of the axon terminals showedsmooth surface (Fig. 2B). In contrast, the morphol-ogy of Ruffini endings at non-ground side was almostidentical to that of age-matched control animals(Fig. 2C). At PN28d, Ruffini endings of control ani-mals expanded compared to the previous stage, andthe dot-like structures at the terminal portions wereapparent (Fig. 3A). At ground side 14 days aftertreatment, the Ruffini endings ramified repeatedly,but the terminal expansion was not developed com-pared to the age-matched control animals (Fig. 3B).Morphology of Ruffini ending at non-ground side wasalmost comparable to that of normal control ani-mals (Fig. 3C).

684 L. Shi et al.

Figure 1 PGP 9.5- (A and B) and S-100-immunoreactivity (C) in the lingual periodontal ligament of the upper incisor ofcontrol animals at PN14d. (A) The terminal portions of periodontal Ruffini endings begin to ramify. (B) Highermagnification of the boxed area in (A). The terminal portions show smooth outlines. (C) S-100-immunoreactive roundedcells extending long process gather at the ARP. Asterisks: shear zone. Scale bar in Fig. 3C: 100 mm for (A) and (C) and50 mm in (B).

At PN42d, the terminal portions of Ruffini endingsof control animals were well expanded, and manydot-like structures were recognized at the terminalportions (Fig. 4A and B). At ground side, the terri-tories occupying Ruffini endings were relativelysmaller 28 days following grinding from PN14d(Fig. 4D), and no or very few dot-like structureswere recognized at their terminal portions (Fig. 4E).The morphology of Ruffini endings at non-groundside (Fig. 4G and H) was similar to the age-matchedanimals. The distribution and morphology of S-100-immunoreactive structures was almost identicalamong the control animals (Fig. 4C) and ground side(Fig. 4F) and non-ground side (Fig. 4I) of the experi-mental animals at PN42d; the rounded immunor-eactive cells extended long processes within theARP.

Figures 2 and 3 PGP 9.5-immunoreactivity in the lingual pcontrol animals (A), and at ground (B) and non-ground (C) sides3) following initial grinding at PN14d. Fig. 2(A) The terminacompare to animals at PN14d. (B) Number of Ruffini endings sl(C) The distribution andmorphology of nerve terminals are almterminal portions show dendritic appearance. (B) The numbesmaller. (C) The terminal portions are well expanded as observFig. 3C: 100 mm for all figures.

Quantitative analysis

Fig. 5 shows the changes in the percentages of PGP9.5-immunonoreactive areas in the restricted areasfollowing grinding from PN14d. In control animals,the percentage of immunoreactive areas at PN14dwas 1.5 � 0.1%, and gradually increased, reaching4.2 � 0.2% at the end of experiment, i.e. PN42d. Inthe ground side, the percentage of immunoreactiveareas at 7 days following initial grinding increased to2.2 � 0.2%, and then decreased to 1.8 � 0.1% at 14days after initial treatment, and slightly increasedto 2.6 � 0.1% at 28 days after grinding from PN14d.There were significant differences in the percentageof PGP 9.5-immunoreactive areas between groundside and non-ground side from 14 days afterinitial grinding. The changes in the percentage of

eriodontal ligament of the upper incisor of age-matchedof the experimental animal 7 days (Fig. 2) or 14 days (Fig.

l portions of periodontal Ruffini endings begin to expandightly decreases and their terminals have smooth outline.ost identical to that of age-matched animals. Fig. 3(A) Ther and size of terminal portion of Ruffini endings becomeed in the control animal. Asterisks: shear zone. Scale bar in

Occlusal force and Ruffini endings 685

Figure 4 PGP 9.5- (left and middle columns) and S-100-immunoreactivity (right column) in the lingual periodontalligament of the upper incisor of age-matched control animals (A—C), and ground (D—F) and non-ground (G—I) sides 28days following grinding from PN14d. (A) The terminal portions of periodontal Ruffini endings extensively arborize incontrol animals. (B) Higher view of the boxed area in (A). Many dot-like structures (arrows) are recognized. (C) Many S-100-immunoreactive rounded cells and processes are present in the ARP. (D) Ruffini endings decrease in size and numberat ground side. (E) Higher magnification of the boxed areas in (D). Outlines of axon terminals become smooth. (F) Thedistribution and morphology of S-100-immunoreactive structures at ground side is almost identical to that of age-matched control animal. (G—I) Distribution and morphology of PGP9.5- and S-100-immunoreactive structures at the non-ground side are almost identical to that of age-matched control animals. Asterisks: shear zone. Scale bar in (I): 100 mm for(A), (C), (D), (F), (G) and (I), 50 mm for (B), (E) and (H).

immunoreactive areas at non-ground side weresimilar to those of control animals at the samepostnatal day.

Discussion

The present immunohistochemical study demon-strates that reduction of mechanical stimulationto the ligament, i.e. occlusal force, causes the delayin morphological maturation of periodontal Ruffiniendings during development.

Although we could not measure the exact occlusalforce, the occlusal force at ground side was appar-ently smaller than that of the control animals sincethe upper and lower incisors at the ground side didnot touch each other. In contrast, it is difficult todeterminewhether occlusal force at non-ground sidewas same as the normal animals or not. However, theupper and lower incisors of non-ground side elon-gated and their tips did not touch each other in someanimals, and in other animals the tips touched each

other as observed as in normal animals, indicatingthat the occlusal force at non-ground side was notidentical to that of the normal animals.

As the periodontal ligament contains both freenerve endings and specialized mechanoreceptiveendings,4—7 the present quantitative analysis repre-sents the changes in the densities of both types ofendings. However, as free nerve endings are verythin, the percentages of PGP 9.5-immunoreactivefree nerve endings were negligible from the totalpercentages of immunoreactive area. Thus, changesin percentages of immunoreactive areas seem likelyto represent the changes in immunoreactive areasof periodontal Ruffini endings.

Previous studies have shown that the develop-ment and maturation of nerve fibres in the period-ontal ligament are closely related with the teetheruption; the axonal elements of periodontal Ruffiniendings of the incisors begin to ramify after theeruption of incisors, and showed extensive arbori-zation after the commencement of molar occlu-sion.14—16 The morphological maturation of the

686 L. Shi et al.

Figure 5 Temporal changes in the percentages of theimmunoreactive area for PGP 9.5 in the restricted areas ofcontrol animals (0) and age-matched control animals (greycolumns), grounded side (open columns) and non-groundside (closed column) of the experimental animals. Per-centages of immunoreactive areas at the ground side arestatistically smaller than those of the non-ground sideafter 14 days following initial grinding. *p < 0.05,**p < 0.01 (ANOVA, comparison with age-matched controlanimals), #p < 0.05, ##p < 0.01 (Student’s t-test, compar-ison between ground side and non-ground side at the sameexperimental period).

terminal Schwann cells takes place earlier than thatof the axonal elements in the periodontal Ruffiniendings; distribution of terminal Schwann cellsbecomes identical to that of adult animals aroundPN14—21d.17,24 The present results of the develop-ment of both axonal elements and terminal Schwanncells of periodontal Ruffini endings of control ani-mals are in agreement with those results. Thedevelopment of periodontal Ruffini endings atnon-ground side was also comparable to those ofcontrol animals, indicating that occlusal force atnon-ground side of the experimental animals hadlow effect, if any, on alteration in morphologicalmaturation of the periodontal Ruffini endings.

One of the possible factors affecting the devel-opment of periodontal Ruffini endings was nutrition.As the body weight of the experimental animals wasalmost identical to that of control animals, andbecause the values of immunoreactive areas ofnon-ground side of experimental animals were com-parable to those of normal animals, it is safe to saythat the nutrition does not affect the developmentof periodontal Ruffini endings. One claimed that theanaesthesia have some effects on the developmentof periodontal Ruffini endings. However, the changein body weight in animals at anaesthetized controlgroup was almost identical to those in normal con-trol animals, suggesting that the effects of anaes-thesia may be excluded. Thus, we believe thatchanges in percentages in PGP 9.5-immunoreactive

areas might be mostly due to the reduction of theocclusal force.

The present quantitative analysis showed signifi-cant decrease in the percentage of PGP 9.5-immu-noreactive areas within the restrict areas of groundside from 14 days following initial grinding. Weobserved dot-like structures near the terminal por-tions of the Ruffini endings in control animals and atnon-ground side of the experimental animals atPN21d, and more apparently on PN42d. In develop-ing animals with reduced occlusal force, the appear-ance of PGP 9.5-immunoreactive dot-like structuresdelayed. As previous ultrastructural observationsrevealed that the axon terminals of Ruffini endingsare partially exposed to the surrounding collagenoustissues through slits between the Schwannsheaths,4,8 and appeared as the dot-like structuresnear the axon terminals at the light microscopicallevel.14,15 Thus, decrease in the percentage ofimmunoreactive areas might be partly due to thelack or lower number of PGP 9.5-immunoreactivedot-like structures at the axon terminals duringdevelopment. However, as the immunoreactive areaof dot-like structures was very small, particularly atthe early stage, other possibilities should be con-sidered. The most probable account for thedecrease in the percentages of immunoreactiveareas is the decrease in number and size of period-ontal nerve fibres, although we did not carry out themorphometric analysis. This speculation is sup-ported by our previous study on the adult animal.20

The Schwann cells play important roles during thedevelopment.25,26 In the present study, the distri-bution of terminal Schwann cells was examinedusing S-100-immunohistochemistry, and we foundno apparent difference in the distribution of S-100-immunoreactive structures between the groundand non-ground sides of the experimental animals,and between the non-ground side of experimentalanimals and control animals. Muramoto et al.reported the alterations in the morphology of term-inal Schwann cells in the periodontal Ruffini endingsof rat molar teeth following loss of occlusal sti-muli.27 They also examined the staining patternof growth-associated protein-43 (GAP-43), whichis present in the Schwann sheath of periodontalRuffini endings under normal condition,28,29 andfound that staining pattern of GAP-43 altered fol-lowing the loss of occulsal force. Thus, furtherinvestigation is needed to examine the morpholo-gical and histochemical changes of the terminalSchwann cells after reduction of the occlusal forceduring the development.

Molecular mechanism controlling morphologicalmaturation of periodontal Ruffini endings has notbeen fully elucidated. The present study clearly

Occlusal force and Ruffini endings 687

showed the delay of morphological maturation ofperiodontal Ruffini endings following the reductionof occlusal force. It is known that the application ofmechanical stress causes the changes in DNA synth-esis, secretion and expression of various mole-cules.30—32 Similarly, we believe that the changesin mechanical stimulation to the periodontal liga-ment also altered in the synthesis of moleculescontributing the morphological maturation of peri-odontal Ruffini endings; possible molecules beingBDNF and trkB, a high-affinity receptor for BDNF.Maeda and his research group reported that a mal-formation and 18% reduction of the periodontalRuffini endings in BDNF +/� mice,11 and a delay inthe maturation of periodontal Ruffini ending in BDNFnull mutant mice.13 Furthermore, Matsuo et al.reported the absence of periodontal Ruffini endingsin trkB�/� mice.33 These evidences suggest thatBDNF/trkB signaling plays an important role in themaintenance and development of periodontal Ruf-fini endings in rats. We do not knowwhether changesof mechanical stress cause the morphologicalalterations of periodontal Ruffini ending directlyor indirectly. Our previous study demonstrated thatthe expression of trkB immunoreactivity in axonelements and terminal Schwann cells as well as inperiodontal fibroblasts of normal adult animals.34

Thus, it is easily speculated that reduced mechan-ical stimulation of the ligament changes the proper-ties of various types of cells in the periodontalligament, particularly fibroblasts. Further experi-ments are required to elucidate whether changes inocclusal force might affect the expression of BDNF/trkB and other molecules in the ligament.

Clinically abnormal sensations occur immediatelyafter bite-raising, and then adaptation takes place.Previous morphological studies showed that raisingocclusal height caused immediate morphologicalchanges in periodontal Ruffini endings,18,19 suggest-ing the possible correlation of morphological altera-tions of periodontal Ruffini endings and abnormalsensation following bite-raising. However, it has notbeen shown that abnormal sensations occur and thatsubsequent sensory adaptation takes place afterreduced occlusal force. It is proposed that axonalspines, dot-like structures at the axon terminal,play important roles in the perception of the move-ment of collagen fibres in the ligament.35 Thus, thedecreased number of axon spines caused by reducedocclusal force during development might causehypo-sensitivity against mechanical stimulation tothe ligament.

In conclusion, the present study clearly demon-strates that the reduction of mechanical stimuli tothe ligament, i.e. occlusal force, affects the mor-phological maturation of periodontal Ruffini ending

during development. Taken together with previousstudies using developmental rats14—17 and adultrats,18,19,22 we conclude that proper mechanicalstimulation is one of the important environmentalfactors of the morphological maturation of the per-iodontal Ruffini ending during development as wellas maintenance of those mechanoreceptors.

Acknowledgements

Authors thank Prof. Takeyasu Maeda of Niigata Uni-versity for discussion about the results of prelimin-ary experiments. This study was partly supported bygrants-in-aid for the 21st Century COE program‘‘Origination of Frontier BioDentistry’’ from Ministryof Education, Culture, Sports, Science and Technol-ogy of Japan, and for Scientific Research (#16390526to SW) from Japan Society for the Promotion ofScience (JSPS).

References

1. Berkovitz B, Shore R, Moxham B. The occurrence of a lamel-lated nerve terminal in the peirodontal ligament of the ratincisor. Arch Oral Biol 1983;28:99—101.

2. Everts V, Beertsen W, van den Hooff A. Fine structure of anend organ in the periodontal ligament of the mouse incisor.Anat Rec 1977;189:73—89.

3. Lewinsky W, Stewart D. The innervation of the periodontalmembrane of the cat with some observation on the functionof the end organs found in that structure. J Anat 1937;71:232—5.

4. Byers MR. Sensory innervation of periodontal ligament of ratmolars consists of unencapsulated Ruffini-like mechanore-ceptors and free nerve endings. J Comp Neurol1985;231:500—18.

5. Byers MR, Maeda T. Periodontal innervation: regional specia-lizations, ultrastructure, cytochemistry and tissue interac-tions. Acta Med Dent Helv 1997;2:116—33.

6. Maeda T, Ochi K, Nakakuma-Ohshima K, Youn SH, Wakisaka S.Ruffini ending as an essential mechanoreceptor in period-ontal ligament–—its morphology, cytochemical feature,regeneration and development. Crit Rev Oral Biol Med1999;10:307—27.

7. Wakisaka S, Atsumi Y, Youn TH, Maeda T. Morphological andcytochemical characteristics of periodontal Ruffini endingunder normal and regeneration processes. Arch Histol Cytol2000;63:91—113.

8. Maeda T, Sato O, Kobayashi S, Iwanaga T, Fujita T. Theultrastructure of Ruffini endings in the periodontal ligamentof rat incisor with special reference to the terminal Schwanncells (K-cells). Anat Rec 1989;223:95—103.

9. Byers MR, O’Connor TA, Martin RF, Dong WK. Mesencephalictrigeminal sensory neurons of cat: axon pathways and struc-ture of mechanoreceptive endings in periodontal ligament. JComp Neurol 1986;250:181—91.

10. Byers MR, Dong WK. Comparison of trigeminal receptor loca-tion and structure in the periodontal ligament of differenttypes of teeth from the rat, cat, and monkey. J Comp Neurol1989;279:117—27.

688 L. Shi et al.

11. Alkhamrah BA, Hoshino N, Kawano Y, Harada F, Hanada K,Maeda T. The periodontal Ruffini endings in brain derived-neurotrophic factor (BDNF) deficient mice. Arch Histol Cytol2003;66:73—81.

12. Harada F, Hoshino N, Hanada K, Kawano Y, Atsumi Y, WakisakaS, Maeda T. The involvement of brain-derived neurotrophicfactor (BDNF) in the regeneration of periodontal Ruffini end-ings following transection of the inferior alveolar nerve. ArchHistol Cytol 2003;66:183—94.

13. Hoshino N, Harada F, Alkhamrah BA, Aita M, Kawano Y,Hanada Y, Maeda T. Involvement of brain-derived neuro-trophic factor (BDNF) in the development of periodontalRuffini endings. Anat Rec 2003;274A:807—16.

14. Nakakura-Ohshima K, Maeda T, Sato O, Takano Y. Postnataldevelopment of periodontal innervation of rat incisors: animmunohistochemical study using protein gene product 9.5antibody. Arch Histol Cytol 1993;56:385—98.

15. Nakakura-Ohshima K, Maeda T, Ohshima H, Noda T, Takano Y.Postnatal development of periodontal Ruffini endings in ratincisor: an immunoelectron microscopic study using proteingene product 9.5 (PGP 9. 5) antibody. J Comp Neurol 1995;362:551—64.

16. Asahito T, Ohshima H, Hanada K, Wakisaka S, Maeda T.Postnatal expression of calretinin-immunoreactivity in per-iodontal Ruffini endings in the rat incisor: a comparison withprotein gene product 9.5 (PGP 9. 5)-immunoreactivity. ArchHistol Cytol 1999;62:57—69.

17. Hayashi S, Nakakura-Ohshima K, Ohshima H, Noda T, HonmaS, Wakisaka S, Maeda T. The development of terminalSchwann cells associated with periodontal Ruffini endingsin the rat incisor ligament. Brain Res 2000;858:167—71.

18. Kvinnsland I, Heyeraas KJ. Effect of traumatic occlusion onCGRP and SP immunoreactive nerve fiber morphology inrat molar pulp and periodontium. Histochemistry 1992;97:111—20.

19. Sodeyama T, Maeda T, Takano Y, Hara K. Responses of period-ontal nerve terminals to experimentally induced occlusaltrauma in rat molars: an immunohistochemical study usingPGP 9.5 antibody. J Periodont Res 1996;31:235—48.

20. Shi L, Kodama Y, Atsumi Y, Honma S, Wakisaka S. Requirementof occlusal force for maintenance of terminal morphology ofthe periodontal Ruffini endings. Arch Histol Cytol 2005;68:289—99.

21. Gulbenkian S, Wharton J, Polak J. The visualisation of car-diovascular innervation in the guinea pig using an antiserumto protein gene product 9.5 (PGP 9. 5). J Auton Nerv Syst1987;18:235—47.

22. Thompson RJ, Doran JF, Jackson P, Dhillon AP, Rode J. PGP9.5–—a new marker for vertebrate neurons and neuroendo-crine cells. Brain Res 1983;278:224—8.

23. Shore R, Berkovitz B. Model to explain differential movementof periodontal fibroblasts. Arch Oral Biol 1978;23:507—9.

24. Atsumi Y, Imai T, Matsumoto K, Sakuda M, Kurisu K, WakisakaS. Effects of neonatal injury of the inferior alveolar nerve onthe development and regeneration of periodontal nervefibers in the rat incisor. Brain Res 2000;871:201—9.

25. Mirsky R, Jessen KR. Schwann cell development, differentia-tion and myelination. Curr Opin Neurobiol 1996;6:89—96.

26. Mirsky R, Jessen KR. The neurobiology of Schwann cells. BrainPathol 1999;9:293—311.

27. Muramoto T, Takano Y, Soma K. Time-related changes inperiodontal mechanoreceptors in rat molars after the lossof occlusal stimuli. Arch Histol Cytol 2000;63:369—80.

28. Maeda T, Byers MR. Different localization of growth-asso-ciated protein (GAP-43) in mechanoreceptors and free nerveendings of adult rat periodontal ligament, dental pulp andskin. Arch Histol Cytol 1996;59:291—304.

29. Youn SH, Maeda T, Kurisu K, Wakisaka S. Growth-associatedprotein-43 (GAP-43) in the regenerating periodontal Ruffiniendings of the rat incisor following injury to the inferioralveolar nerve. Brain Res 1998;787:41—8.

30. Kletsas D, Basdra EK, Papavassiliou AG. Mechanical stressinduces DNA synthesis in PDL fibroblasts by a mechanismunrelated to autocrine growth factor action. FEBS Lett1998;430:358—62.

31. Yoshino H, Morita I, Murota SI, Ishikawa I. Mechanical stressinduces production of angiogenic regulators in culturedhuman gingival and periodontal ligament fibroblasts. J Per-iodont Res 2003;38:405—10.

32. Yamashiro T, Fukunaga T, Kobashi N, Kamioka H, Nakanishi T,Takigawa M, Takano-Yamamoto T. Mechanical stimulationinduces CTGF expression in rat osteocytes. J Dent Res2001;80:461—5.

33. Matsuo S, Ichikawa H, Silos-Santiago I, Kiyomiya K, Kurebe M,Arends J, Jacquin M. Ruffini endings are absent from theperiodontal ligament of trkB knockout mice. SomatosensMotor Res 2002;19:213—7.

34. Atsumi Y, Hayashi S, Nakakura-Ohshima K, Maeda T, Kurisu K,Wakisaka S. Heterogeneous localizations of Trk B amongindividual periodontal Ruffini endings in the rat incisor. ArchHistol Cytol 1999;62:435—40.

35. Chambers M, Andres K, von During M, Iggo A. The structureand function of the slowly adapting type II mechanoreceptorin hairy skin. Quart J Exp Physiol 1972;57:417—45.