Archives of Oral Biology 61 (2016) 106–114

Complex cellular responses to tooth wear in rodent molar

A. Mahdeea,b,c,*, A. Alhelala,b,c, J. Easthamc, J. Whitwortha,c, J.I. Gillespiec

aCentre for Oral Health Research, Newcastle University, UKb Institute of Cellular Medicine, Newcastle University, UKc School of Dental Sciences, Newcastle University, UK

A R T I C L E I N F O

Article history:Received 16 July 2015Received in revised form 3 September 2015Accepted 5 October 2015

Keywords:OdontoblastsTooth wearPulp biologyDentinePhysiologyRegeneration

A B S T R A C T

The arrangement and roles of the odontoblast and its process in sensing and responding to injuries suchas tooth wear are incompletely understood. Evidence is presented that dentine exposure by tooth weartriggers structural and functional changes that aim to maintain tooth integrity.Mandibular first molars from freshly culled 8 week Wistar rats were prepared for light microscopy

ground-sections (n = 6), or fixed in 4% paraformaldehyde, decalcified in 17% EDTA, sectioned and stainedwith antibodies to cyto-skeletal proteins (vimentin (vim), a-tubulin (tub) and a-actin), cellularhomeostatic elements (sodium potassium ATPase (NaK-ATPase) and sodium hydrogen exchanger (NHE-1)), and sensory nerve fibres (CGRP) (n = 10) for fluorescence microscopy of worn and unworn regions ofthe mesial cusp.Immunoreactivity (IR) to vim, actin, NaK-ATPase and CGRP was confined to the pulpal third of

odontoblast processes (OPs). IR to tub and nhe-1 was expressed by OPs in full dentine thickness. In areasassociated with dentine exposure, the tubules contained no OPs. In regions with intact dentine,odontoblasts were arranged in a single cell layer and easily distinguished from the sub-odontoblast cells.In regions with open tubules, the odontoblasts were in stratified or pseudo-stratified in arrangement.Differences in structural antibody expression suggest a previously unreported heterogeneity of the

odontoblast population and variations in different regions of the OP. This combined with differences inOPs extension and pulp cellular arrangement in worn and unworn regions suggests active and dynamiccellular responses to the opening of dentinal tubules by tooth wear.

ã 2015 Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

Archives of Oral Biology

journal homepage: www.elsev ier .com/locate /aob

1. Introduction

The dental pulp share intimate spatial and functional relation-ships with dentine, and is commonly described within a pulp-dentine complex (Yu & Abbott, 2007). Odontoblasts (Ods) performa range of formative, supportive, sensory and defensive functionsthroughout life (Ricucci, Loghin, Lin, Spångberg, & Tay, 2014), andthe comparison of regions within the same tooth that have beensubjected to or spared from injury provides one means ofunderstanding important cellular and hard-tissue responses.

Recognised age-related changes in the pulp-dentine complexincluding reduced cellularity, particularly within the odontoblast(Od) and subodontoblast (Sod) layers (Morse, 1991; Murray,

* Corresponding author at: School of Dental Sciences, Faculty of Medicine,Newcastle University, Newcastle upon Tyne NE2 4BW, UK. Fax: +44 1912086137.

E-mail addresses: a.f.mahdee@ncl.ac.uk (A. Mahdee), a.g.m.alhelal@ncl.ac.uk(A. Alhelal), j.e.eastham@ncl.ac.uk (J. Eastham), john.whitworth@ncl.ac.uk(J. Whitworth), james.gillespie@ncl.ac.uk (J.I. Gillespie).

http://dx.doi.org/10.1016/j.archoralbio.2015.10.0040003-9969/ã 2015 Elsevier Ltd. All rights reserved.

Stanley, Matthews, Sloan, & Smith, 2002), and changing in volume,structure and permeability of dentine (Moses, Butler, & Qin, 2006;Tjäderhane, Carrilho, Breschi, Tay, & Pashley, 2009). Muchunderstanding has derived from rodent molar studies, which haveshown a reduction in odontoblast size and cytoplasmic volumewith age (Lovschall, Fejerskov, & Josephsen, 2002), along withphysiological occlusal tooth wear, which is apparent within 4weeks (Schour & Massler, 1949) and consequent hard and softtissue changes (Kawashima, Wongyaofa, Suzuki, Kawanishi, &Suda, 2006; Lovschall et al., 2002), including reactionary dentinedeposition beneath worn cusp tips (Moses et al., 2006).

Dentine exposure by cavity preparation has revealed a reductionof Od numbers, probably resulting from apoptosis (Kitamura,Kimura, Nakayama, Toyoshima, & Terashita, 2001; Murray et al.,2000),butwith newly differentiatedOd-likecellsappearingwithin 7days (Ohshima, 1990). Cellular responses may be age-related, sincethe presence of dendritic like cells in the sub-odontoblast cell layer,with processes extending into the predentine is more prominentafter cavity preparation in aged than young rats (Kawagishi,Nakakura-Ohshima, Nomura, & Ohshima, 2006).

A. Mahdee et al. / Archives of Oral Biology 61 (2016) 106–114 107

The degree of odontoblast process (OP) extension withindentine remain contentious (Luukko, Kettunen, Fristad, & Berg-green, 2011), with suggestion their limitation to the inner third ofdentine (Byers & Sugaya, 1995; Carda & Peydro, 2006; Goracci,Mori, & Baldi, 1999; Yoshiba, Yoshiba, Ejiri, Iwaku, & Ozawa, 2002),whilst others have demonstrated OPs reaching the dentino-enameljunction (DEJ) (Gunji & Kobayashi, 1983; Kagayama et al., 1999;Sigal, Aubin, & Ten Cate,1985; Sigal, Aubin, Ten Cate, & Pitaru,1984;Tsuchlya, Sasano, Kagayama, & Watanabe, 2002). In the rat molar, ithas been reported that the OPs fully traverse the dentinethroughout life in cusp areas, while in the cervical regions, theyretract to the inner third as teeth mature (Tsuchlya et al., 2002). Inprevious work, we have demonstrated the extension of OPs to therodent incisor DEJ with regional variations and complex branchingalong their path, especially near the DEJ (Mahdee, Alhelal,Eastham, Whitworth, & Gillespie, 2015).

The arrangement and roles of the Ob and its process in sensingand responding to injuries such as tooth wear are incompletelyunderstood. This report presents evidence from the rat molar thatdentine-exposing tooth wear triggers a discrete and thithertoundescribed series of structural and functional changes in that aimto protect deep connective tissues from the oral environment.

2. Methods

In order to investigate tooth changes associated with dentineexposure due to the wear process, this study used ground sectionsto characterise hard tissue changes and decalcified sections withimmunohistochemistry to identify pulp cellular differences.

Ground sections: six lower first molars were extracted fromfreshly culled male Wistar rats (age 8 weeks; weight 240–300 g).Ground sections (100–105 mm thickness) of the largest, mesialcusp, were examined by light microscopy using objectives �20 and�40 to identify variations in tubular arrangement associated withwear process.

Immuno-fluorescence technique: ten lower first molars wereextracted as previously, with immediate fixation of the crowns in4% paraformaldehyde in PBS for 24 h at 4 �C, demineralisation in17% EDTA (pH 7.4) for 4–6 weeks at 37 �C, washing and incubationin graded sucrose solutions (10%, 20%, 30%) for 24 h each at 4 �C forcryoprotection. Specimens were snap frozen in liquid nitrogen, andstored at �80 �C. During sectioning, sagittal sections were takenparallel to the long axis of the tooth until the three buccal cuspswere visible. From this point 20–30 slides with 10 mm thicknesswere obtained from each tooth in order to standardize the sectionsto be within the orientation of the odontoblast processes.

The immunofluorescence staining procedure (Eastham, Ste-phenson, Korstanje, & Gillespie, 2015) was performed as follows:sections were divided randomly into 4 staining groups (approxi-mately 50 slides for each group) and stained by the followingcombination: monoclonal anti-vimentin structural protein (vim)(mouse 1:5000, BioGenex cat# MU074-UC, UK), accompanied withone of the following:: monoclonal anti-a smooth muscle actin(actin) (rabbit 1:200, Abcam cat# ab32575, UK) detectingmicrofilament structural protein, monoclonal anti-NaK-ATPaseenzyme (rabbit 1:500, Abcam cat# ab76020, UK), polyclonal anti-atubulin (tub) (rabbit, 1:1000, Gene Tex, cat# GTX102078, UK) todetect the tubulin structural protein, and polyclonal sodiumhydrogen exchanger-1 (NHE-1) (rabbit, 1:200, Santa Cruz Biotech-nology,cat# sc-28758, UK). Both NaK-ATPase and NHE-1 areenzymes present in the cell membrane and they were used in thepresent study to detect the cell membrane of the odontoblastprocesses. One extra staining group used the following antibodycombination: monoclonal anti-calcitonin gene-related peptide(CGRP) (mouse 1:500, Santa Cruz Biotechnology, cat# sc-57053,UK) for afferent sensory nerve fibre detection combined with the

polyclonal anti-a tubulin. These antibodies have been charac-terised by the manufacturer and have been used in publishedstudies (see respective manufacturer’s data sheets). The primaryantibody combination was applied to each section beforeincubating slides in a humid atmosphere at 4 �C for 24 h. Thefollowing day, slides were washed in a three stage cycle (TBS, TBS-T,and TBS) for 20 min each. The secondary antibodies; donkey anti-mouse Alexa Fluor 488 (Molecular Probes1, Invitrogen) anddonkey anti-rabbit IgG, Alexa Fluor 594 (Molecular Probes1,Invitrogen), were applied in accordance to the species of primaryantibody that had been used and incubated in humidifier for onehour at room temperature. The slides were then washed again in athree stage cycle (TBS, TBS-T, and TBS) for 20 min each, beforeapplying Vectashield hard set mounting medium with DAPI(nucleic acid molecular probe stain) (Vector Laboratories Inc,Burlingame) and placing cover slips.

Control samples were included. The slides were eitherincubated with PBS instead of the primary antibodies, beforestaining with the secondary antibodies, or incubated with PBS only.

In order to standardize the region of interest for all the teeth,this study examined only the mesial cusp of the first mandibularmolar.The stained sections were examined through �10, �20, and�60 magnification with an Olympus BX61 microscope (OlympusCorporation, Tokyo Japan), using Alexa Fluor 488 and 594 fluo-rochromes detected via the microscope light source and dichroicmirror, to split excitation and emission light wavelengths.Representative images were captured with a microscope-mountedOlympus XM10 monochrome camera and examined using ImageJsoftware (Java-based image processing program—National Insti-tute of Health (USA)). For each of the objectives the absolutemagnification was determined using a stage graticule. Thecalibrations using this method were shown in each image.Approximately 50 slides were examined for each group to confirmthe accuracy and consistency of the staining technique and toreveal consistent staining phenomena (Gillespie, Markerink-vanIttersum, & De Vente, 2006).

3. Results

In order to identify time dependent changes in the mesial cuspas a model for tooth wear the following descriptions of tooth wearwere used (Fig. 1):

Stage 0 (S0) tooth regions under unworn dentine surface (lateraland central walls of the mesial cusp).

Stage 1 (S1) tooth regions close to worn dentine surface (at theangle of the cusp from the unworn cusp side).

Stage 2 (S2) tooth regions of minor worn dentine surface (nearthe angle of the cusp from the worn cusp side).

Stage 3 (S3) tooth regions of severe worn dentine surface.All the ground sections observed in this study showed worn

occlusal dentine associated with an underlying area of tertiarydentine close to the pulp space (Fig. 1 big panel). In S0 regions, thecomplexity of the dentinal tubules could easily be recognized inthe outer dentine with abundant lateral and terminal branching(Fig. 1S0). The terminal branching seemed to terminate within theDEJ. A similar pattern of the dentinal tubules is also illustrated inS1 region which is located close to the dentine worn surface at thecusp angle (Fig. 1S1–S2). In the same image it is apparent that in S2region, the dentinal tubules were worn and had lost their outerbranching part. In the S3 region (Fig. 1S3) the dentinal tubulesappeared shorter with the proximity of the worn dentine surface tothe pulp space becoming apparent. Considering the region of thetertiary dentine near the pulp (Fig. 1R), there was no continuity ofthe dentinal tubules between the primary dentine and the newlytissue, which appeared to present tubules with meanderingpattern.

Fig. 1. Ground section of rat first molar mesial cusp. The following structures were identified, P; pulp space, De; dentine, En; enamel, DEJ; dentine-enamel junction. The bigpanel shows the worn occlusal dentine surface to the left, the lateral wall downward, the inner wall of the cusp upward, and an area of tertiary dentine (*). Four regions ofinterest are highlighted (S0, S1–S2, S3 and R) Image S0 shows a region of intact tooth surface with complex lateral branches (arrows), and terminal branches (*) of the dentinaltubules. S1–S2 image shows the cusp angle with S1 region on the top and in which the dentinal tubules still possess complex terminal branches (*) near DEJ. In S2 regions thedentinal tubules appear to be worn to a minor degree. The S3 image identify the severely worn dentine surface. In R panel, an area of tertiary dentine with meandering tubules(arrows) is illustrated. Calibration bars are shown; 100 mm in big panel, 25 mm in S0, S1–S2, S3 and 20 mm in R.

108 A. Mahdee et al. / Archives of Oral Biology 61 (2016) 106–114

A. Mahdee et al. / Archives of Oral Biology 61 (2016) 106–114 109

In decalcified sections, the extension of the OPs to the outerregion of the dentine was demonstrated with NHE-1 (Fig. 2 firstpanel). The intensity of labelling was relatively uniform through-out the full dentine thickness in S0, S1, and S2 regions and thislabelling also followed a linear course along the primary curvatureof the dentinal tubules which was observed in Fig. 1. In S3 regionthe odontoblast processes were limited to the inner dentine whilethe rest appeared to be devoid of any specific fluorescence labelling(Fig. 2A). The OPs showed vim immunoreactivity (IR) in the regionbetween the cell body and into inner dentine (Fig. 2B). This IRchanged to be more NHE-1 within the rest of dentine, indicating aheterogeneity of antibody expression within the same OP.

Fig. 2. Decalcified section of the rat mesial cusp. All images were stained for NHE-1 in rdentine surface in S0, S1, and S2 regions while in S3 region the processes are limited to theouter dentine in regions S1 and S2 is apparent, while in S3 the outer dentine is devoid of NHsub-odontoblast cells (SOd). The OPs show vim-IR near the odontoblast cell body. This IR

calibration bars are shown; 100 mm in big panel, 20 mm in A, and 50 mm in B.

The antibody to a-tub, also labelled the OPs throughout the fullthickness of the dentine. The OPs showed very complex lateral andterminal branches in the outer region of the dentine near the DEJ inS0 regions (Fig. 3A and a). Some of these processes showed morethan 4 terminal branches (Fig. 3a) which indicated the complexityof these branches in the outer dentine. In region S3, tub-IR (Fig. 3Band b) was also absent in the outer dentine and it was limited to theinner third of the dentine similar to the pattern observed withNHE-1. In S2 region the OPs that reached to the outer dentineappeared to be thicker than the ones observed in S0 regions, andwithout lateral branching (Fig. 3b).

ed, vim in green and dapi in blue. The big panel shows the extension of OPs to the inner third of the dentine. In A panel, the continuity of NHE-1-IR for the OPs in theE-1-IR. B image shows a region of the pulp-dentine area with odontoblasts (Od) andgives way NHE-1 when the process extends further within the dentine (arrows). The

Fig. 3. Decalcified sections from S0 compared to S2 and S3 regions of rat first molar mesial cusp. All images were stained for a-tub. A shows the extension of OPs in the outerdentine in an area of intact enamel (En) on the lateral wall of the cusp (an S0 region). The region of interest in A is delineated by box which clearly shows complex terminalbranching of the OPs (arrow). Panel B shows no OPs extending to outer dentine in an S3 region and some processes extending to outer dentine in an S2 region (*). The highermagnification image in panel (b) shows some of the OPs labelled with a-tub extending close to the outer dentine surface in S2 region but these processes have no branches.The calibration bars; 50 mm in A, 100 mm in B, 24 in (a) and (b).

110 A. Mahdee et al. / Archives of Oral Biology 61 (2016) 106–114

IR to a-actin and NaK-ATPase was limited in the inner third ofthe dentine with no immunofluorescent labelling detected in theexternal dentine in all regions of the tooth, as shown in Fig. 4images A and B respectively. The interesting observation for thesetwo antibodies was the cellular heterogeneity of the odontoblastcells expressing these antibodies in comparison to vim (Fig. 4C andD).

The dentinal tubules and OPs showed differences in theirpattern between S0 in one side and S2 and S3 regions on the otherside. There were other differences related to the cellular elementswithin the pulp. In S0 region the odontoblasts appeared as a singlecell layer, with their cellular processes extending into the dentineand could easily be distinguished in their morphology from thesub-odontoblast cells (Fig. 5A). Cellular heterogeneity was alsoidentified within the odontoblasts between vim and tub antibodies(Fig. 5a1 and a2). Moving to S2–S3 regions, the odontoblasts werearranged in a pseudostratified or stratified layer of 2–3 cellsthickness (Fig. 5B). Some cells from the second line of odontoblastswere also observed sending processes into the predentine(Fig. 5b1). Some sub-odontoblast cells in this region showedmodification in their cellular morphology to be similar to theodontoblasts. However, no cellular processes were recognized inthese cells (Fig. 5b2).

Finally the distribution of afferent sensory nerve fibres wasobserved with antibody to CGRP. These fibres appeared withpunctate staining and did not extend further than the innerdentine (Fig. 5C). The varicosities of CGRP-IR fibres gave rise to thispunctate staining, whilst the axons between these varicositieswere very weakly detectable. These sensory nerves showed acomplex distribution in between the sub-odontoblasts and the

odontoblasts and they ran with the odontoblast process(Fig. 5c1 and c2).

None of the control groups showed any specific fluorescentlabelling within the pulp, dentinal tubules, or DEJ.

4. Discussion

This study is among the first to present evidence from the ratmolar that dentine-exposing tooth wear triggers a discrete andhitherto undescribed series of structural and functional changesthat aim to protect deep connective tissues from the oralenvironment. Although the rat molars have enamel free areason their cusps, they are still covered with bone and gingival tissueuntil tooth wear commences soon after eruption, certainly within 4weeks for the rat mandibular first molar (Schour & Massler, 1949)The largest cusp (the mesial) was examined in all cases forconsistency.

Antibodies for structural proteins; vim, a-actin, and a-tub wereused since intermediate filaments, microfilaments and micro-tubules have been reported as major components of theodontoblasts cytoskeleton and have not been shown to existextracellularly in viable cells (Lesot, Meyer, Ruch, Weber, & Osborn,1982; Thomas & Carella, 1983; Thomas & Payne, 1983). The otherantibodies used were for cellular homeostatic elements indicativeof living cells; NaK-ATPase and NHE-1. NaK-ATPase is expressedwidely in the cellular systems of the dental pulp (Duan, 2014),where it is instrumental in maintaining the ionic gradient cross cellmembranes. NHE-1 is an isoform of the sodium/hydrogenexchanger, which is a membrane protein primarily responsiblefor maintaining the intracellular pH. Josephsen et al. (2010) found

Fig. 4. Decalcified sections from S0 region of rat first molar dentine/pulp. Sections were either stained for vim (green), a-actin (red), dapi (blue) in panels A and C, or vim(green), NaK-ATPase (red), dapi (blue) in panels B and D. The following structures were identified: Od; odontoblast cell layer, SOd; sub-odontoblast cells, BV; blood vessel, P;pulp, De; dentine, DEJ; dentino-enamel junction. In panels A and B, the OPs show high IR for vim, a-actin, and NaK-ATPase only in the inner third of the dentine. C and D arehigh power images for Od and SOd in S0 regions. In C and its component images (actin and vim) some Od cells show actin-IR (*) with no vim-IR and others have only vim-IR (+)with no actin-IR. Panels D and its component images (vim and NaK-ATPase) show most of the Od cells labelled both vim and NaK-ATPase-IR but some have no vim-IR (~). Thecalibration bars; 100 mm in A, 200 mm in B, and 15 mm for other images. (For interpretation of the references to colour in this figure legend, the reader is referred to the webversion of this article.)

A. Mahdee et al. / Archives of Oral Biology 61 (2016) 106–114 111

both NaK-ATPase and NHE-1 in cells of the rodent enamel organand linked them with cellular pumping activities. Pulp cells,especially odontoblasts, have also been shown to contain both ofNaK-ATPase and NHE-1 in their cell membrane (Duan, 2014; Linde& Lundgren, 1995). In our previous work with rat incisors, it wasshown that the OPs expressed NaK-ATPase-IR along their entirelength (Mahdee et al., 2015). Another antibody, CGRP, was used as amarker for sensory nerve fibers (Mori, Ishida-Yamamoto, Senba,Ueda, & Tohyama, 1989).

In normal physiology, OP and their complex branches occupythe full tubular space (Carda & Peydro, 2006). The current studyconfirms that the OPs extend the full thickness of the dentine and

terminate within an intact DEJ. This is the first report to identifyNHE-1-IR within the OP (Fig. 3D). Also preliminary observationswith an antibody to Na/Ca exchanger suggest that this transporteris also present predominantly in this region (in preparation). Ifthese proteins are functional, it may reflect the importance ofregulation of intra and extra cellular pH and Ca levels in this part ofthe OP. However, the other antibodies including vim, a-actin, andNaK-ATPase were labelled only in the inner third of the dentine.This could reflect regional differences in the structure of the OPs i.e.certain protein filaments may be present only in defined regionsdepending on the function of the OPs in these regions. Previousstudies have also reported the a-tub-IR of OPs throughout the

Fig. 5. Decalcified sections of rat first molar to compare between S0 and S2–S3 pulp regions. Images A, a1, a2, B, b1, and b2 were stained for vim (green), a-tub (red), and dapi(blue), and images C, c1 and c2 for CGRP (green), a-tub (red), and dapi (blue). A shows the single layer arrangement of the Od cells in S0 region. The two panels below a1 anda2 represent region of interest in image A and identify three types of Od cells: vim-IR (+), tub-IR (*), and vim-tub-IR cells (~). Panel B shows the arrangement of odontoblastsin a pseudo-stratified layer of 2 to 3 cell thickness in S2 and S3 regions of the pulp. Two regions of interest are shown in image B. The first one appeared in image B1 with cells (*)in the second line of the thick odontoblasts cell layer sending processes (arrows) to the predentine (Pd). The second region is identified in image b2 which shows the sub-odontoblast cells heading toward the proximal cells of the odontoblasts. Image C from S2–S3 region shows nerves stained with CGRP running from the pulp and terminating inthe inner part of the De (arrows). A region of interest is shown in image c1 and the component image c2, which clearly identify nerve fibres running between SOd and Ods (*)and then into the inner De (small arrows). The calibration bars; 30 mm in A, B and C, 10 mm in a1, a2, b1 and b2, and 15 mm in c1 and c2. (For interpretation of the references tocolour in this figure legend, the reader is referred to the web version of this article.)

112 A. Mahdee et al. / Archives of Oral Biology 61 (2016) 106–114

thickness of the dentine, whilst vim and a-actin-IR were limited tothe inner dentine in human (Sigal et al., 1985) and rat molars (Sigalet al., 1984). These findings are supported by the current study. Bycontrast, our previous work on the rat incisor found a-actin-IR inOPs throughout the entire thickness of the dentine (Mahdee et al.,2015), and this may reflect the greater concentration of thatprotein in continuously growing teeth compared with rat molarsand human teeth. Therefore, it would appear that OPs in intactmolar surfaces are complex and show regional variations in proteinexpression that may reflect structural and functional specializa-tion.

An illustration of the changes from normal undamaged surfacesto a situation of significant wear is shown in Fig. 6. In normalphysiology, OPs terminated with complex branches within the DEJ.The function of this complex branching is not known, but maysuggest its involvement in detecting the integrity of that region,acting as a receptor field. Any stimulation, mechanical or chemical,could be transmitted along OPs to the cell body of the odontoblasts.The stimulated Od triggering a cascade of events that includesretraction of OPs, recruitment of new cells from sub-odontoblastlayer and possible signaling to afferent nerve fibres underpinningsensation or pain.

It was observed that after exposing of the dentine surface, therewas an initial loss of side processes in S2 regions. However, in areasof minor or perhaps slower wear, the OPs still extended to the endof the dentinal tubules. The processes persisting in this regioncould have more time to occlude the tubules by increasing thethickness of peritubular dentine. This allows these processes to

remain near outer dentine safely. At this stage, changes were seenin the odontoblast and sub-odontoblast layers with an increase incell density. It may therefore be argued that changes in the pulpcells can be arise by the transmission of information from the outerdentine down by the OPs.

In regions of more severe or faster wear (S3 regions), the OPsretracted into the inner third of the tubules, whist most tubulesappeared devoid of processes. Since a-actin was found in the lowerthird of the process (Fig. 2), it is conceivable that the activation ofactin could be responsible for the active retraction of the process.This retraction may prevent the processes from further damageand at the same time allow more space for the deposition ofprotective peritubular and/or intratubular dentine (Linde &Goldberg, 1993) Tsuchlya et al. (2002) also reported regionaldifferences in the extension of the OPs between cervical and cuspareas. They suggested the persistence of OPs in the outer dentine ofworn rat molar cusps, even after aging. They did not note theapparent responses of the OPs to surface wear observed in thisstudy, which could be due to artefactual labelling of someextracellular components.

Changes in the pulp cellular population were also noticed inregions undergoing dentine exposure i.e. S2 and S3 regions in thepresent study (Fig. 5). The first change was the increase inthickness of the odontoblast layer to 2-3 cells which could reflectmore activity of the odontoblasts in this region as they are involvedin the formation of reactionary dentine. However, this could be thenormal distribution of the odontoblasts in this region as these cellstend to be more crowded in areas under the cusps (Lovschall et al.,

Fig. 6. Schematic illustration represent the hypothesis of this study. The following abbreviations have been used: OP is odontoblast process, PD is predentine, NF is nerve fibre,Od is odontoblast, SOd is sub-odontoblast and (*) is the covering tissue. Reading from the left, It illustrates the presentation of an odontoblast within a healthy unworn dentinesurface, followed by early to late pathological changes resulting from the opening of dentinal tubules, and finally the development of defensive reactionary and reparativeresponses.

A. Mahdee et al. / Archives of Oral Biology 61 (2016) 106–114 113

2002). At the same time, differences were also noticed in thepopulation of the sub-odontoblast cells (Fig. 5) as the most distalcells in this layer become more odontoblast-like. These cells couldbe supportive to the remaining odontoblasts or recruited to replaceextensively damaged odontoblasts. However, and under theconditions of this study, all the cells present at the border of thepulp-dentine complex contained processes. Considering the age ofrats used in this study, almost all of these cells are still likely to bethe original odontoblasts.

Additionally, the presence of meandering tubules in theground section (Fig. 1) in the region of tertiary dentine could alsosupport the persistence of the original odontoblasts. However,these meandering tubules are not continuous with the originalprimary tubules and this needs further investigation to identifythe nature of this reactionary dentine and the sequences for sucha defensive mechanism. Finally, examining the teeth of older ratswith more advanced wear progression or possibly with moreaggressive types of stimulus such as cavity preparation (Ohshima,1990), could show the presence of the odontoblast like cellssecreting reparative dentine matrix, which has been suggested inthe last panel in Fig. 6.

Another important observation from the current study is theheterogeneity of the odontoblasts, both in intact and worn regionsof the tooth (Figs. 2 and 4). Functional differences amongodontoblasts may include the deposition and mineralization ofdentine matrix and maintenance of vital functions within the pulpdentine complex, with roles including sensing, nutrition andinflammatory reactions (Luukko et al., 2011).

In regions of tooth wear, the retraction of OPs, could leave morespace within the tubules for intratubular fluid movement. Thishydrodynamic fluid movement could stimulate either the

remaining part of the OPs or persisting afferent sensory nerveswithin the tubules as suggested by the hydrodynamic theory(Ciucchi, Bouillaguet, Holz, & Pashley, 1995). This may alsocontribute to altered sensation including pain in damaged orworn regions of the teeth. This hyper-stimulation of the exposedtubules due to hydrodynamic fluid movement, response tomicrobial or other external noxious stimuli could all stimulatethe odontoblasts to deposit reactionary dentine matrix to increasethe distance between the pulp tissue and the exposed area. Inaddition, afferent fibres in the pulp express CGRP (Mori et al.,1989), and in other systems where CGRP afferent fibers are found,it can be released during nerve activation and have effects on theadjacent cells (Assas, Pennock, & Miyan, 2014). This raises thepossibility that the nerves may feedback to stimulate theodontoblasts. Collateral branches of such axons almost certainlyexist, in addition to the electrical coupling between odontoblasts(Ikeda & Suda, 2013). This has an effect on triggering odontoblasts adistance away (Kimberly & Byers, 1988). Thus the effect of damagemay be spread to odontoblasts over a considerable area.

In summary, this study provides novel observations about thepulp-dentine complex, and suggests a new explanation for the roleof OPs in maintaining tooth integrity. New insights have also beengiven about the roles of cellular complexity within the pulp-dentine complex and possible roles in sensing, defense and repair.Furthermore, two possible sensing approaches for the pulp-dentine complex have been suggested depending on the integrityof outer dentine surface. The examination of rats of varying agesand degrees of tooth wear will shed further light on the pulp-dentine complex reactions associated with exposing dentine, helpto advance understanding of pulp physiology in health and disease,and opportunities for therapeutic intervention.

114 A. Mahdee et al. / Archives of Oral Biology 61 (2016) 106–114

Acknowledgements

The authors acknowledge the Oral Biology Lab in the School ofDental Sciences/Newcastle University for their role in theexperimental work.

This research is funded by the Iraqi Ministry of HigherEducation as a part of PhD scholarship program..

References

Assas, B. M., Pennock, J. I., & Miyan, J. A. (2014). Calcitonin gene-related peptide is akey neurotransmitter in the neuro-immune axis. Frontiers in Neuroscience, 8, 23.

Byers, M. R., & Sugaya, A. (1995). Odontoblast processes in dentin revealed byfluorescent Di-I. Journal of Histochemistry and Cytochemistry, 43(2), 159–168.

Carda, C., & Peydro, A. (2006). Ultrastructural patterns of human dentinal tubules,odontoblasts processes and nerve fibres. Tissue and Cell, 38(2), 141–150.

Ciucchi, B., Bouillaguet, S., Holz, J., & Pashley, D. (1995). Dentinal fluid dynamics inhuman teeth, in vivo. Journal of Endodontics, 21(4), 191–194.

Duan, X. (2014). Ion channels, channelopathies, and tooth formation. Journal ofDental Research, 93(2), 117–125.

Eastham, J., Stephenson, C., Korstanje, K., & Gillespie, J. (2015). The expression ofb3-adrenoceptor and muscarinic type 3 receptor immuno-reactivity in themajor pelvic ganglion of the rat. Naunyn-Schmiedeberg’s Archives ofPharmacology 1–14.

Gillespie, J. I., Markerink-van Ittersum, M., & De Vente, J. (2006). Interstitial cells andcholinergic signalling in the outer muscle layers of the guinea-pig bladder. BJUInternational, 97(2), 379–385.

Goracci, G., Mori, G., & Baldi, M. (1999). Terminal end of the human odontoblastprocess: a study using SEM and confocal microscopy. Clinical Oral Investigations,3, 126–132.

Gunji, T., & Kobayashi, S. (1983). Distribution and organization of odontoblast inhuman dentin. Archivum Histologicum Japonicum, 46(2), 213–219.

Ikeda, H., & Suda, H. (2013). Odontoblastic syncytium through electrical coupling inthe human dental pulp. Journal of Dental Research, 92(4), 371–375.

Josephsen, K., Takano, Y., Frische, S., Praetorius, J., Nielsen, S., Aoba, T., et al. (2010).Ion transporters in secretory and cyclically modulating ameloblasts: a newhypothesis for cellular control of preeruptive enamel maturation. AmericanJournal of Physiology—Cell Physiology, 299(6), C1299–C1307.

Kagayama, M., Sasano, Y., Sato, H., Kamakura, S., Motegi, K., & Mizoguchi, I. (1999).Confocal microscopy of dentinal tubules in human tooth stained with alizarinred. Anatomy and Embryology, 199(3), 233–238.

Kawagishi, E., Nakakura-Ohshima, K., Nomura, S., & Ohshima, H. (2006). Pulpalresponses to cavity preparation in aged rat molars. Cell and Tissue Research, 326(1), 111–122.

Kawashima, N., Wongyaofa, I., Suzuki, N., Kawanishi, H. N., & Suda, H. (2006). NK andNKT cells in the rat dental pulp tissues. Oral Surgery, Oral Medicine, OralPathology, Oral Radiology, and Endodontics, 102(4), 558–563.

Kimberly, C., & Byers, M. (1988). Inflammation of rat molar pulp and periodontiumcauses increased calcitonin gene-related peptide and axonal sprouting.Anatomical Record, 222(3), 289–300.

Kitamura, C., Kimura, K., Nakayama, T., Toyoshima, K., & Terashita, M. (2001).Primary and secondary induction of apoptosis in odontoblasts after cavitypreparation of rat molars. Journal of Dental Research, 80(6), 1530–1534.

Lesot, H., Meyer, J., Ruch, J., Weber, K., & Osborn, M. (1982). Immunofluorescentlocalization of vimentin, prekeratin and actin during odontoblast andameloblast differentiation. Differentiation, 21(1–3), 133–137.

Linde, A., & Goldberg, M. (1993). Dentinogenesis. Critical Reviews in Oral Biology &Medicine, 4(5), 679–728.

Linde, A., & Lundgren, T. (1995). From serum to the mineral phase: the role of theodontoblast in calcium transport and mineral formation. International Journal ofDevelopmental Biology, 39, 213.

Lovschall, H., Fejerskov, O., & Josephsen, K. (2002). Age-related and site-specificchanges in the pulpodentinal morphology of rat molars. Archives of Oral Biology,47(5), 361–367.

Luukko, K., Kettunen, P., Fristad, I., & Berggreen, E. (2011). Structure and functions ofthe dentin-pulp complex, Cohen’s pathways ofthe pulp [Internet]. 10th ed. St.Louis, Missouri, USA: Mosby Elsevier452–503.

Mahdee, A., Alhelal, A., Eastham, J., Whitworth, J., & Gillespie, J. I. (2015). Theextension and complexity of the odontoblast processes in the rat lower incisors.Archives of Oral Biology (in press).

Mori, H., Ishida-Yamamoto, A., Senba, E., Ueda, Y., & Tohyama, M. (1989). Calcitoningene-related peptide containing sensory neurons innervating tooth pulp andbuccal mucosa of the rat: an immunohistochemical analysis. Journal of ChemicalNeuroanatomy, 3(3), 155–163.

Morse, D. R. (1991). Age-related changes of the dental pulp complex and theirrelationship to systemic aging. Oral Surgery, Oral Medicine, Oral Pathology, OralRadiology, and Endodontics, 72(6), 721–745.

Moses, K. D., Butler, W. T., & Qin, C. (2006). Immunohistochemical study of smallintegrin-binding ligand, N-linked glycoproteins in reactionary dentin of ratmolars at different ages. European Journal of Oral Sciences, 114(3), 216–222.

Murray, P., About, I., Lumley, P., Franquin, J.-C., Remusat, M., & Smith, A. (2000).Human odontoblast cell numbers after dental injury. Journal of Dentistry, 28(4),277–285.

Murray, P. E., Stanley, H. R., Matthews, J. B., Sloan, A. J., & Smith, A. J. (2002).Age-related odontometric changes of human teeth. Oral Surgery, Oral Medicine,Oral Pathology, Oral Radiology, and Endodontics, 93(4), 474–482.

Ohshima, H. (1990). Ultrastructural changes in odontoblasts and pulp capillariesfollowing cavity preparation in rat molars. Archives of Histology and Cytology, 53(4), 423–438.

Ricucci, D., Loghin, S., Lin, L. M., Spångberg, L. S., & Tay, F. R. (2014). Is hard tissueformation in the dental pulp after the death of the primary odontoblasts aregenerative or a reparative process? Journal of Dentistry, 42(9), 1156–1170.

Schour, I., & Massler, M. (1949). The teeth. In E. J. Farris, & J. Q. Griffith (Eds.), The Ratin Laboratory Investigations (pp. 102–165). New York: Hafner Press, A Division ofMacmillan Publishing Co, Inc..

Sigal, M., Aubin, J., Ten Cate, A., & Pitaru, S. (1984). The odontoblast process extendsto the dentinoenamel junction: an immunocytochemical study of rat dentine.Journal of Histochemistry and Cytochemistry, 32(8), 872–877.

Sigal, M., Aubin, J., & Ten Cate, A. (1985). An immunocytochemical study of thehuman odontoblast process using antibodies against tubulin, actin, andvimentin. Journal of Dental Research, 64(12), 1348–1355.

Thomas, H., & Carella, P. (1983). A scanning electron microscope study of dentinaltubules from un-erupted human teeth. Archives of Oral Biology, 28(12),1125–1130.

Thomas, H., & Payne, R. (1983). The ultrastructure of dentinal tubules from eruptedhuman premolar teeth. Journal of Dental Research, 62(5), 532–536.

Tjäderhane, L., Carrilho, M. R., Breschi, L., Tay, F. R., & Pashley, D. H. (2009). Dentinbasic structure and composition—an overview. Endodontic Topics, 20(1), 3–29.

Tsuchlya, M., Sasano, Y., Kagayama, M., & Watanabe, M. (2002). The extent ofodontoblast processes in the dentin is distinct between cusp and cervicalregions during development and aging. Archives of Histology and Cytology, 65(2),179–188.

Yoshiba, K., Yoshiba, N., Ejiri, S., Iwaku, M., & Ozawa, H. (2002). Odontoblastprocesses in human dentin revealed by fluorescence labeling and transmissionelectron microscopy. Histochemistry and Cell Biology, 118(3), 205–212.

Yu, C., & Abbott, P. V. (2007). An overview of the dental pulp, its functions andresponses to injury. Australian Dental Journal, 52(Suppl. 1), S4–S16.