Mast cells and lymphocyte subsets in pulps from healthy andcarious human teethPatrícia Freitas, DDS,a Carolina Purens Novaretti, DDS,b Camila Oliveira Rodini, DDS, MS,c

Aline Carvalho Batista, DDS, MS, PhD,d and Vanessa Soares Lara, DDS, MS, PhD,e Bauruand Goiânia, BrazilUNIVERSITY OF SÃO PAULO AND FEDERAL UNIVERSITY OF GOIÁS

Objective. To evaluate the presence of cytolytic T lymphocytes (CD8�), memory T cells (CD45RO�), helper Tlymphocytes (CD4�), and mast cells in pulps from healthy and carious human teeth.Study design. The teeth were separated into groups: I � unerupted; II � partially erupted, without caries; III �erupted, without caries; IV � erupted with shallow dentine caries; and V � teeth with pulp polyps. Theimmunoperoxidase staining procedure was used to detect CD8, CD45RO, CD4, and tryptase (mast cell marker)antigens. The number of each cell type was obtained by counting the number of cells per mm2.Results. Mast cells were only present in pulp polyps. Pulps from carious teeth contained more CD4� and CD8�cellsthan from noncarious teeth. There was a significant decrease in the number of lymphocytes in pulp polyps incomparison to the other groups.Conclusions. Mast cells probably do not contribute to the early vascular or specific immune responses in the initialdental pulp pathosis, although they may be involved in a chronic phase of pulp inflammation such as pulp polyps. Onthe other hand, CD4� and CD8�T cells participate mainly in initial phenomena of the immune response to incipientcaries and seem not to substantially contribute to the response in pulp polyps. (Oral Surg Oral Med Oral Pathol Oral

Radiol Endod 2007;103:e95-e102)

Dental caries is the most common cause of pulp dis-ease, in which inflammatory and immunologic reac-tions occur in response to products of bacterial metab-olism that penetrate into the pulp through the dentinaltubules.1 In the dental pulp, antigen-presenting cells,which are class II molecule-expressing cells usually ofdendritic morphology, capture protein antigens and mi-grate to regional lymph nodes. After presentation ofpeptide fragments in conjunction with class II majorhistocompatibility complex molecules to antigen-spe-cific T cells (naive T lymphocytes), helper T lympho-cytes (CD4�) clonally expand and differentiate into

Supported by grants from the Fundação de Amparo à Pesquisa doEstado de São Paulo (FAPESP-01/02205-1 and 01/09651-7 [toC.P.N. and P.F., respectively]).aMSc student in Oral Pathology, Department of Stomatology, BauruDental School, University of São Paulo.bGraduate Student, Bauru Dental School, University of São Paulo.cMSc student in Oral Pathology, Department of Stomatology, BaruDental School, University of São Paulo.dProfessor, Department of Stomatology (Oral Pathology), DentalSchool, Federal University of Goiás.eProfessor, Department of Stomatology (Oral Pathology), Baru Den-tal School, University of São Paulo.Received for publication Jul 20, 2006; returned for revision Sep 27,2006; accepted for publication Nov 21, 2006.1079-2104/$ - see front matter© 2007 Mosby, Inc. All rights reserved.

doi:10.1016/j.tripleo.2006.11.031

effector T cells or memory T cells (CD45RO�). Effec-tor T cells and some memory T cells enter into thevasculature and migrate to the pulp.2 Then, residentantigen-presenting cells such as macrophages interactwith and present antigen directly to the T cells, whichare locally activated and trigger the effector phase ofthe immune response against bacterial antigens.2,3 Ac-cording to the profile of cytokine production, CD4�/CD45RO� cells stimulate either the activation of mac-rophages, other CD4� cells and cytolytic Tlymphocytes (CD8�), or the proliferation and differen-tiation of B lymphocytes.4 The CD8� cells recognizeand kill target cells expressing foreign peptide antigensin association with class I major histocompatibilitycomplex molecules.4

Some investigations have detected the presence of Tlymphocytes (CD4�, CD8�, and CD45RO�T cells) inhealthy human dental pulp5-10 and an increase in in-flamed tissue.6,7,9 Regarding T-cell subsets, althoughmost of the studies have found that CD8�T cells out-numbered CD4�T cells in both healthy and inflameddental pulp,5-7,9 Mangkornkarn et al.8 detected moreCD4� than CD8�T cells in normal human dental pulpby using flow cytometric analysis. The possible in-crease in the number of T cells in inflamed compared tonormal pulp tissue may indicate that these cells areimportant in the immune response during pulp inflam-

mation.

e95

OOOOEe96 Freitas et al. May 2007

At the present time, there is an increased awarenessof the potential interactions between mast cells andother components of the immune response contributingto the modulation of humoral and cellular events in hostdefense mechanisms against bacterial infections11,12

and probably participating in the pathogenesis of in-flammatory conditions such as pulp disease. Mast cellsare one of the major effector cells of innate immunitybecause these cells contain several vasoactive sub-stances, including histamine and arachidonic acid me-tabolites.13 Besides, mast cells may significantly influ-ence IgE-independent responses, extending theirpotential from proinflammatory effector cells to regu-latory components of the immune system and contrib-uting to the development of nonspecific as well asspecific inflammatory responses.14-16 The presence andfunction of mast cells in the dental pulp have alsoaroused some controversy. The absence of mast cells inhealthy pulps was observed in previous studies.17-22 Onthe other hand, Farnoush23 and Walsh et al.24 detectedthe presence of mast cells in normal pulp, albeit at alow density24 or without signs of degranulation.23 Still,some studies have demonstrated the presence of mastcells in inflamed pulps.20-24 However, Dockrill18 re-ported that, normally, tooth pulp bound by hard tissuesappears to be devoid of mast cells, but that such cellscan be found when the tooth pulp is able to swell, as inpulp polyps.

Taken together, these findings suggest that the dentalpulp is capable of mounting an immune responseagainst microbial insults from dental caries. However,there are controversial data regarding the presence ofT-lymphocyte subsets and mast cells in inflamed andhealthy human dental pulp. Moreover, published dataabout pulp polyps and the participation of inflammato-ry/immunocompetent cells in pulp diseases are quitescarce. Besides, to our knowledge, there have been nostudies evaluating the changes in these cells duringtooth eruption, which could contribute to a better un-derstanding of the cell population important to maintainhomeostasis in the pulp tissue. So, using immunohis-tochemistry, our study aimed to identify and quantifydifferent types of immunocompetent and inflammatorycells (CD8�, CD4� and CD45RO� T lymphocytes,and mast cells) in human pulps from permanent teeth indifferent clinical conditions.

MATERIAL AND METHODSPulp tissue samples

Samples were either collected from human perma-nent teeth extracted for orthodontic reasons and due toimpaction at private dental clinics and oral surgeryclinics at Bauru Dental School, University of São

Paulo, Brazil (groups I to IV) or selected from paraffin-

embedded sections (group V) from files of the Anatom-ical Pathology Laboratory of the Department of Stoma-tology, Bauru Dental School, University of São Paulo,Brazil. The latter group was composed of human per-manent teeth with chronic hyperplastic pulpitis (pulppolyps) extracted due to nonrestorability. The teethpresenting pulp polyps were only available in filesbecause they are rarely extracted at the present time.Forty samples of human pulp tissues from permanentteeth in different clinical conditions were selected andseparated into 5 groups: unerupted teeth (group I, n �8); partially erupted teeth, no occlusion (group II, n �9); erupted teeth (in occlusion), either without clinicaldentine caries (group III, n � 12) or with clinicalshallow dentine caries (group IV, n � 6); and eruptedteeth with chronic hyperplastic pulpitis (group V, n �5). The age of the patients ranged from 18 to 35 yearsin all groups. This study was approved by the EthicsCommittee of Bauru Dental School, University of SãoPaulo, and informed consent was obtained from allpatients.

Preparation of pulp specimensTo obtain the pulp tissues of groups I, II, III, and IV,

immediately following extraction the apical third of theroots was cut with a water-cooled air turbine to facili-tate penetration of the fixative solution. The teeth wereimmersed in 10% formaldehyde, and 24 hours after thebeginning of fixation, they were longitudinally groovedalong the root and coronal surfaces (tooth splitting)with a carbide round bur using high speed with air-water coolant. A bone chisel was used to crack open thegrooved teeth. The pulp tissues were removed from thehard tissues and then immersed in the same fixative foranother 24 hours. Care was taken to minimize mechan-ical trauma to the dental pulp.

The pulp tissues of group V presented pulp exposureand had been previously demineralized in 5% ethyl-enediaminetetraacetic acid (EDTA) solution (pH 6.0)for approximately 4 months. All pulp specimens weredehydrated in ethanol, immersed in xylene, and embed-ded in paraffin. Sections were cut either at 5-�m thick-ness and stained by hematoxylin and eosin (HE) or at4-�m thickness and immunostained by the immuno-peroxidase (streptavidin-biotin-peroxidase complex)method. All HE-stained samples were microscopicallyevaluated as to the presence of inflammatory infiltrate.

ImmunohistochemistryAfter embedding the tissues in paraffin wax, 4-�m-

thick sections were obtained and collected on silane-coated glass slides (Dako Corp., S3003, Glostrup, Den-mark). The immunohistochemical characterization of

cells was performed using the immunoperoxidase

OOOOEVolume 103, Number 5 Freitas et al. e97

(streptavidin-biotin-peroxidase) method. The sampleswere deparaffinized by immersion in xylene, followedby immersion in alcohol and then incubation with 3%hydrogen peroxide (H2O2) diluted in phosphate-buff-ered saline (PBS; pH � 7.2) for 40 minutes to blockendogenous peroxidase activity. Later, the sectionswere immersed in citrate buffer (pH 6.0; P4809, Sigma-Aldrich Co., St. Louis, Mo) for 20 minutes at 95°C,except for sections destined to be examined with themouse anti-human mast cell tryptase monoclonal anti-body (M7052). Soon afterward, the sections wereblocked by incubation with 3% normal goat serumdiluted in distilled water at room temperature for 20minutes. Then, the slides were incubated with the pri-mary antibodies at 4°C overnight. All antibodies werediluted in 1% PBS bovine serum albumin. Rabbit anti-human CD4 (SC-7219, clone H-370) polyclonal anti-body was purchased from Santa Cruz Biotechnology(Santa Cruz, Calif) and was used at a dilution of 1:300.Mouse anti-human CD8 (M7103, clone C8/144B),CD45RO (M0742, clone UCHL1), and mast celltryptase (M7052, clone AA1) monoclonal antibodieswere purchased from Dako and used at dilutions of1:200, 1:500, and 1:3000, respectively.

Serial tissue sections were used, and each of the 4different antibodies was applied to a different sectionfrom the same block. Following incubation with theprimary antibodies, the sections were washed with PBSand treated with the labeled streptavidin-biotin kit(K0690, Dako, Carpinteria, Calif) before being incu-bated in a solution of 5 mg 3,3�-diaminobenzidine(D4293, Sigma-Aldrich Co.) diluted in 10 mL PBScontaining 180 �L of H2O2 (20 vol), for 20 to 30minutes at room temperature. After washing with dis-tilled water, the slides were counterstained withMayer’s hematoxylin for 5 minutes. Sections from hu-man teeth with periapical lesions previously deminer-alized in 5% EDTA solution, human lymph node tis-sues, lichen planus, and periodontal disease lesionswere used as positive controls and as negative controlsto test the specificity of the immunohistochemical stain-ing. Negative controls were obtained by the omission ofprimary antibodies, which were substituted by 1% PBSbovine serum albumin and by nonimmune rabbit serum(Dako Corp, X0902).

Quantitative analysisThe number of positively stained cells (CD8�,

CD45RO�, CD4�, and mast cell tryptase-positivecells) was determined using 1 section per tooth bymeans of an integration graticule (Carl Zeiss, Göttin-gen, Germany). All cells were counted in 50 consecu-tive microscopic high-power fields (objective, �100) in

a representative section of each specimen, including

both coronal and root pulps, which were microscopi-cally identified. Each field has an area of 0.0176625mm2, obtained from the mathematical expression A �(�d2)/4, where � � 3.14, d � 0.15 mm. The results areexpressed as mean � SD of n observations, per mm2.The statistical analysis was performed using analysis ofvariance (ANOVA), followed by the Tukey test. Due toa nonattendance of the variance homogeneity criteria(homoscedasticity), all data were transformed by theformula x’� ln (x � 0.5) before using ANOVA. Pvalues less than .05 were considered to be statisticallysignificant. The cells positive for mast cell tryptasewere considered to be mast cells.

RESULTSLight microscopic observations

The pulps from groups I and II presented features ofyoung connective tissue, such as many cells, bloodvessels, and nerves, as well as a limited number ofcollagen fibers. The odontoblastic layer was very evi-dent and well organized. The pulps from group III werecharacterized by a loose connective tissue, with a con-siderable number of spindle-shaped cells, blood ves-sels, and nerves, without focus of inflammation. More-over, some samples in this group presented features ofan older tissue, such as fibrosis, fewer cells, and pres-ence of pulp calcifications. Group IV presented a denseconnective tissue, exhibiting fibrosis, fewer cells, andblood vessels, and a discrete-to-moderate inflammatoryinfiltrate. In pulps from group V, few odontoblastssurvived, and the pulp was replaced by granulationtissue, characterized by many newly formed blood ves-sels, mononuclear cells—especially plasma cells—andpolymorphonuclear cells. These pulp polyps were cov-ered by a layer of well-formed stratified squamousepithelium.

Immunohistochemical observationsThe stained lymphocytes exhibited a dark brown ring

around the cells (Fig. 1), whereas antitryptase antibod-ies reliably identified mast cell granules (Fig. 2). Thenegative controls used for each reaction did not showany marking, confirming the specificity of the proce-dure. The positive controls, including both EDTA-de-mineralized and nondemineralized tissues, presented avariable number of immunostained cells, confirmingthe reactivity and avidity of the antibodies under dif-ferent tissue processing conditions.

Quantitative assessment of T cellsQuantitative analysis of immunostained cells per

mm2 revealed a statistically significant difference in thenumber of each cell type (ANOVA: CD8�, P � .0001;

CD45RO�, P � .0025; CD4�, P � .0009) within the 5

OOOOEe98 Freitas et al. May 2007

groups examined. The Tukey test was used to investi-gate between which groups the difference was signifi-cant.

CD8� cells were found in higher numbers in groupIV (30.01 � 18.05), followed by groups III (16.24 �8.60), I (14.87 � 11.37), II (7.53 � 6.80), and V (0.44� 1.00). There were statistically significant differences

Fig. 1. Immunohistochemical staining of CD8�cells in pulptissue from an erupted noncarious tooth (A). CD45RO�(B)and CD4� (C) cells in pulp tissue from a partially eruptedtooth. Original magnifications �250 (A-C).

in the number of CD8� cells between groups I and V (P

� .0003), II and V (P � .007), and III and V (P �.0002), as well as between IV and V (P � .00012; Fig.3, A). These cells were predominantly distributed in thecoronal pulp, both in the odontoblastic layer region andin the central part of the coronal pulp.

CD45RO� cells were only found in groups I to IV.Human pulps from partially erupted teeth (no occlu-sion) without clinical dentine caries (group II) pre-sented the highest number of CD45RO� cells of allgroups analyzed (9.24 � 5.63). A few cells were ob-served in groups III (4.78 � 5.71), IV (2.91 � 2.42),and I (0.85 � 1.33). There were statistically significantdifferences in the number of CD45RO� cells betweengroups I and II (P � .009), and between II and V (P �.004; Fig. 3, B).

Fig. 2. A, B, Immunohistochemical staining of tryptase-pos-itive cells (mast cells) in chronic hyperplastic pulpitis. Bcorresponds with (*) in A, showing tryptase-positive cells inproximity with reactionary dentine area. Original magnifica-tion �25 (A) and �100 (B).

The frequency of CD4�cells was higher in group IV

OOOOEVolume 103, Number 5 Freitas et al. e99

(19.32 � 16.60), followed, in order, by groups III (7.54� 5.51), II (3.96 � 3.56), I (3.20 �2.67), and V (0.66� 1.00). There were statistically significant differencesin the number of CD4� cells between groups I and IV(P � .049), III and V (P � .037), and between IV andV (P � .0008; Fig. 3, C). Similar to CD45RO� T cells,CD4� T cells were distributed throughout the wholepulp tissue, without a preferred localization.

Comparing the predominant T-cell type in each

Fig. 3. Number of CD8� (A), CD45RO� (B), and CD4� (C)T cells per mm2 in human dental pulps, distributed accordingto the different clinical conditions. Data are expressed asmean � SD, analysis of variance, and the Tukey test (groupswith the same letters considering each T-cell type representP � .05).

group, group I presented a significantly higher number

of CD8� cells than CD4� cells (P � .002) andCD45RO� cells (P � .0001). In addition, in this group,CD4� cells were significantly more frequent thanCD45RO� cells (P � .031). In group III, the number ofCD8� T cells was significantly higher than the numberof CD45RO� T cells (P � .011).

Quantitative assessment of tryptase-positive cellsTryptase-positive cells (mast cells) were only found

in chronic hyperplastic pulpitis (group V) and werepreferentially located in the coronal pulp (mean of18.46 � 13.60 mast cells per mm2) compared to theroot pulp (mean of 0.26 � 0.22 mast cells per mm2).Mast cells in pulp polyps were mostly found scatteredthroughout the whole connective tissue, mainly aroundnerve fibers, newly formed blood vessels, and mono-nuclear as well as polymorphonuclear inflammatorycells.

DISCUSSIONIn agreement with previous studies,5-10 our investi-

gation demonstrated the presence of CD4� and CD8�

cells, even in small numbers, in normal human dentalpulp from unerupted teeth as well as partially andtotally erupted teeth, highlighting the probable partici-pation of these cells in the immunosurveillance of thepulp tissue. It is therefore reasonable to assume that thepulp is provided with an immune cell background, evenin the absence of communication with the oral cavityenvironment or a carious stimulus. Although generallywithout a statistically significant difference, we ob-served a trend toward an increase in the number ofCD4� and CD8� cells in pulps from teeth with clinicalshallow dentine caries (initial inflammation, group IV)in comparison to healthy pulps (groups I to III). Thesefindings are supported by other authors,6,7,9 suggestinga specific cellular immune response to an initial cariousstimulus. Nevertheless, a statistically significant de-crease of CD8� cells was detected in teeth with pulppolyps (group V) when compared to the other groups.Likewise, the number of CD4� cells was lower in pulppolyps than in pulps from erupted teeth without clinicaldentine caries (group III) and erupted teeth with clinicalshallow dentine caries (group IV). Based on this, wesuggest that the direct exposure of the dental pulp to theoral cavity with intense proliferation of granulationtissue may have allowed spread of the antigen motifsand consequent decrease of T cells. Besides, HE stain-ing of our pulp polyp samples revealed many plasmacells. Therefore, an increasing humoral response mayoccur in an advanced and chronic phase of pulp inflam-mation. B lymphocytes may migrate to the inner regionof the pulp tissue in response to cytokines released by

T cells and then proliferate and differentiate into im-

OOOOEe100 Freitas et al. May 2007

munoglobulin-secreting plasma cells, which producelocal antibodies against the antigens at a later stage ofpulp inflammation. Moreover, considering the shift ofdominant microorganisms during caries progression,different cytokine profiles might elicit a polarizationtoward a type 1 response in shallow caries and be lesspolarizing in deep carious lesions, allowing appearanceof B cells and plasma cells.25 Accordingly, Hahn et al.6

and Izumi et al.9 reported that T cells predominate indental pulps under shallow caries, and as the cariouslesion enlarges, T cells continue to dominate but B cellsand plasma cells appear in substantial numbers. Similarto other reports in the literature regarding cell popula-tion in oral chronic inflammatory diseases, Seymour etal.26 have related that a change from gingivitis to peri-odontitis involves a shift from a predominantly T-celllesion to a B-cell/plasma cell lesion. Based on ourresults, the CD8� and CD4� T cells were present inpulps from human teeth in different clinical conditionsand probably participate mainly in initial phenomena ofthe inflammatory response to incipient caries and seemnot to substantially contribute to the response in pulppolyps.

The CD45RO�T cells were found in healthy pulps(groups I to III) as well as in small numbers in pulpsfrom teeth with clinical shallow dentine caries, but notin chronic hyperplastic pulpitis. We observed fewCD45RO�T cells in pulps from unerupted teeth (groupI) and a significant increase in the group from partiallyerupted teeth, even without clinical dentine caries(group II). This finding may reflect the entrance ofantigens in pulp tissue, arising from the oral environ-ment, inasmuch as the tooth begins to erupt. This maybe possible because there is communication betweenoral fluids, enamel, and the inner portion of dentinaltubules.27,28 Thus, the presentation of antigens to Tlymphocytes in lymphoid tissues and migration of thesecells to the pulp tissue as memory T cells (CD45RO�)becomes possible. Izumi et al.9 and Sakurai et al.10 alsoobserved the presence of CD45RO� cells in healthy aswell as in inflamed dental pulp. However, the change inthe number of CD45RO� cells during tooth eruptionobserved in our study is a new finding.

Comparing the T lymphocyte subsets studied in thepresent work, our results are in accordance with otherreports that showed a higher number of CD8� thanCD4� cells in noninflamed dental pulp,5-7,9 but contrastwith the findings of Mangkornkarn et al.,8 who detectedmore CD4� than CD8�T cells in normal human dentalpulp. We also observed a significantly smaller numberof CD45RO� cells than CD4� and CD8� cells inhuman pulps from unerupted teeth (group I), and thanCD8� cells in human pulps from erupted teeth without

clinical dentine caries (group III). Some studies29-31

have shown that some memory T cells, especiallyCD8� cells, can revert from CD45RO to the CD45RAphenotype, typical of naive T cells. So, it is question-able whether CD45RO is a reliable marker of memoryT cells. Besides, the monoclonal antibody used to stainCD45RO�memory cells in the present work (CD45RO,clone UCHL1) reacts with about 70% of helper T cellsand 35% of cytolytic/suppressor T cells,32 which couldexplain the lower number of CD45RO� cells thanCD4� and CD8� cells observed in our study. Unlikeour findings, Izumi et al.9 observed that CD45RO�

cells were more common than CD4� cells but lessfrequent than CD8� cells.

In our study, mast cells were only present in chronichyperplastic pulpitis (pulp polyps). This finding sup-ports the data reported by Dockrill,18 which showedpresence of these cells only when the tooth pulp wasable to swell, as in pulp polyps. The absence of mastcells in noninflamed pulps (groups I, II, and III) foundin the present study is also consistent with some pre-vious reports,17-22 but contradicts the findings ofFarnoush23 and Walsh et al.24 Moreover, in contrast tothe absence of tryptase-positive cells (mast cells) ingroup IV in our work, several authors detected thepresence of mast cells in inflamed pulp tissues,20-24

especially in chronic inflammation.23

The immunohistochemical localization of mast cellsby antitryptase antibodies in formalin-fixed paraffin-embedded tissue has been shown to be highly selectiveand specific. Such method, used in this report, offersseveral advantages when compared with other tech-niques used to identify mast cells.33 Furthermore, themethod used to obtain the pulp tissues (tooth splitting)in our study appears to be the method of choice forpreservation of mast cell integrity.23 Strictly accordingto our results, the possible absence of mast cells innoninflamed pulps as well as in early pulpitis mayindicate that the involvement of mast cell–mediatedphenomena in the immune response during closed earlypulpitis is unlikely to occur. The lack of these cells mayhave a physiological role, since the release of vasoac-tive substances would be detrimental to the pulp be-cause of an immense increase in tissue pressure, sincethe dental pulp is enclosed by a rigid structure ofmineralized tissues.2 Our findings may suggest the par-ticipation of neuropeptides, such as substance P andcalcitonin gene–related peptide, in neurogenic inflam-mation in the pulp, regulating the blood flow and thepermeability of microvessels. This means that the neu-ropeptides in the pulp may cause a vascular responsedirectly because of the lack of mast cells.34 Theseneuropeptides are present in considerable quantities innerve fibers in dental pulps,35-37 and further studies are

needed to clarify their participation in dental pulp in-

OOOOEVolume 103, Number 5 Freitas et al. e101

flammation. On the other hand, the presence of mastcells in chronic hyperplastic pulpitis observed in ourstudy, as well as the presence of these cells in periapicallesions and chronic periodontitis reported by previousstudies in our laboratory,38,39 showed the involvementof mast cells in a chronic phase of inflammation, con-tributing to the specific immune response and the pro-gression of pulp, periapical and periodontal pathoses. Inthese cases, mast cells may migrate to the inflamedtissue in response to the release of specific chemokinesof the chronic inflammatory response.

It is clear that the communication between foreignantigens, the immune system, and other components ofthe dentin/pulp complex is intricate, and further atten-tion to these aspects will provide a better understandingof tissue responses to caries and a more effective andconservative therapy for the disease. In conclusion, thepresent work revealed an interesting perspective onpulp behavior to long-standing exposure to bacterialelements that characterize pulp polyps. Although anincreased number of mast cells were observed in pulppolyps, only a few T cells were found. The absence ofmast cells in dental pulps without direct exposure wasalso demonstrated, which suggests that mast cells donot contribute to the early vascular response by hista-mine release or to the specific immune response in theinitial dental pulp pathosis. On the other hand, ourresults suggest that CD4� and CD8�T cells participatemainly in initial phenomena of the immune response toincipient caries.

We thank José Roberto Pereira Lauris (Department ofPediatric Dentistry, Orthodontic Treatment and PublicHealth, Bauru Dental School) for his help withstatistical analysis, Fátima Aparecida Silveira (Depart-ment of Stomatology, Bauru Dental School) for hertechnical support, and Gareth Paul Cuttle and Gisele daSilva Dalben for assistance with the manuscript.

REFERENCES1. Bergenholtz G. Inflammatory response of the dental pulp to

bacterial irritation. J Endod 1981;7:100-4.2. Jontell M, Okiji T, Dahlgren U, Bergenholtz G. Immune defense

mechanisms of the dental pulp. Crit Rev Oral Biol Med1998;9:179-200.

3. Nakanishi T, Takahashi K, Hosokawa Y, Adachi T, Nakae H,Matsuo T. Expression of macrophage inflammatory protein 3� inhuman inflamed dental pulp tissue. J Endod 2005;31:84-7.

4. Abbas AK, Lichtman AH. Effector mechanisms of cell-mediatedimmunity. In: Abbas AK, Lichtman AH, editors. Cellular andmolecular immunology. 5th ed. Philadelphia: Saunders; 2003. p.298-317.

5. Jontell M, Gunraj MN, Bergenholtz G. Immunocompetent cellsin the normal dental pulp. J Dent Res 1987;66:1149-53.

6. Hahn CL, Falkler WA Jr, Siegel MA. A study of T and B cellsin pulpal pathosis. J Endod 1989;15:20-6.

7. Gregoire G, Terrie B. Identification of lymphocyte antigens in

human dental pulps. J Oral Pathol Med 1990;19:246-50.

8. Mangkornkarn C, Steiner JC, Bohman R, Lindemann RA. Flowcytometric analysis of human dental pulp tissue. J Endod1991;17:49-53.

9. Izumi T, Kobayashi I, Okamura K, Sakai H. Immunohistochem-ical study on the immunocompetent cells of the pulp in humannon-carious and carious teeth. Arch Oral Biol 1995;40:609-14.

10. Sakurai K, Okiji T, Suda H. Co-increase of nerve fibers andHLA-DR- and/or factor-XIIIa-expressing dendritic cells in den-tinal caries-affected regions of the human dental pulp: an immu-nohistochemical study. J Dent Res 1999;78:1596-608.

11. Welle M. Development, significance, and heterogeneity of mastcells with particular regard to the mast cell-specific proteaseschymase and tryptase. J Leukoc Biol 1997;61:233-45.

12. Villa I, Skokos D, Tkaczyk C, Peronet R, David B, Huerre M, etal. Capacity of mouse mast cells to prime T cells and to inducespecific antibody response in vivo. Immunol 2001;102:165-72.

13. Metcalfe DD, Baram D, Mekori YA. Mast cell. Physiol Rev1997;77:1033-79.

14. Gordon JR, Burd PR, Galli SJ. Mast cells as a source of multi-functional cytokines. Immunol Today 1990;11:458-64.

15. Galli SJ. New concepts about the mast cell. N Engl J Med1993;328:257-65.

16. Skokos D, Le Panse S, Villa I, Rousselle JC, Peronet R, David B,et al. Mast cell-dependent B and T lymphocyte activation ismediated by the secretion of immunologically active exosomes.J Immunol 2001;166:868-76.

17. Wislocki GB, Sognnaes RF. Histochemical reactions of normalteeth. Am J Anat 1950;87:239-76.

18. Dockrill TE. Tissue mast cells in the oral cavity. Aust Dent J1961;6:210-4.

19. Photo P, Antila R. Assay of histamine in dental pulps. ActaOdontol Scand 1970;28:691-9.

20. Zachrisson BU. Mast cells in human dental pulp. Arch Oral Biol1971;16:555-6.

21. Skogedal O. Intracellular complex carbohydrates in the normaland inflamed dental pulp. Scand J Dent Res 1973;81:558-62.

22. Miller GS, Sternberg RN, Piliero SJ, Rosenberg PA. Histologicidentification of mast cells in human dental pulp. Oral Surg OralMed Oral Pathol 1978;46:559-66.

23. Farnoush A. Mast cells in human dental pulp. J Endod1984;10:250-2.

24. Walsh LJ, Davis MF, Savage NW. Relationship between mastcell degranulation and inflammation in the oral cavity. J OralPathol Med 1995;24:266-72.

25. Hahn CL, Best AM, Tew JG. Cytokine induction by Streptococ-cus mutans and pulpal pathogenesis. Infect Immun2000;68:6785-9.

26. Seymour GJ, Powell RN, Davies WI. Conversion of a stableT-cell lesion to a progressive B-cell lesion in the pathogenesis ofchronic inflammatory disease: an hypothesis. J Clin Periodontol1979;6:267-77.

27. Ten Cate AR. Structure of the oral tissues. In: Ten Cate AR,editor. Oral histology: development, structure and function. 5thed. St. Louis: Mosby; 1998. p. 1-9.

28. Torneck CD. Dentin-pulp complex. In: Ten Cate AR, editor. Oralhistology: development, structure and function. 5th ed. St. Louis:Mosby; 1998. p. 150-96.

29. Michie CA, McLean A, Alcock C, Beverley PCL. Lifespan ofhuman lymphocyte subsets defined by CD45 isoforms. Nature1992;360:264-5.

30. Bell EB, Sparshott SM, Bunce C. CD4� T-cell memory, CD45Rsubsets and the persistence of antigen–a unifying concept. Im-munol Today 1998;19:60-4.

31. Wills MR, Carmichael AJ, Weekes MP, Mynard K, Okecha G,

Hicks R, et al. Human virus-specific CD8� CTL clones revert

OOOOEe102 Freitas et al. May 2007

from CD45ROhigh to CD45RAhigh in vivo: CD8� T cellscomprise both naive and memory cells. J Immunol1999;162:7080-7.

32. Smith SH, Brown MH, Rowe D, Callard RE, Beverley PCL.Functional subsets of human helper-inducer cells defined by anew monoclonal antibody, UCHL1. Immunol 1986;58:63-70.

33. Walls AF, Jones DB, Williams JH, Church MK, Holgate ST.Immunohistochemical identification of mast cells in formalde-hyde-fixed tissue using monoclonal antibodies specific fortryptase. J Pathol 1990;162:119-26.

34. Ohkubo T, Shibata M, Yamada Y, Kaya H, Takahashi H. Role ofsubstance P in neurogenic inflammation in the rat incisor pulpand the lower lip. Arch Oral Biol 1993;38:151-8.

35. Wakisaka S. Neuropeptides in the dental pulp: distribution, ori-gins, and correlation. J Endod 1990;16:67-9.

36. Goodis H, Saeki K. Identification of bradykinin, substance P, andneurokinin A in human dental pulp. J Endod 1997;23:201-4.

37. Caviedes-Bucheli J, Lombana N, Azuero-Holguin MM, MunozHR. Quantification of neuropeptides (calcitonin gene-related

peptide, substance P, neurokinin A, neuropeptide Y and vasoac-

tive intestinal polypeptide) expressed in healthy and inflamedhuman dental pulp. Int Endod J 2006;39:394-400.

38. Rodini CO, Batista AC, Lara VS. Comparative immunohisto-chemical study of the presence of mast cells in apical granulomaand periapical cysts: possible role of mast cells in the course ofhuman periapical lesions. Oral Surg Oral Med Oral Pathol OralRadiol Endod 2004;97:59-63.

39. Batista AC, Rodini CO, Lara VS. Quantification of mast cells indifferent stages of human periodontal diseases. Oral Dis2005;11:249-54.

Reprint requests:

Vanessa Soares Lara, DDS, MS, PhDRua Caetano Sampieri, 4-25Apto 72, Edifício BarcelonaBairro: Vila UniversitáriaBauru-SP, Brazil 17012-460

vanessa@fob.usp.br