J Periodontol • September 2004

Locally Applied Isosorbide DecreasesBone Resorption in ExperimentalPeriodontitis in RatsR.F.C. Leitão,* F.A.C. Rocha,† H.V. Chaves,* V. Lima,* F.Q. Cunha,‡ R.A. Ribeiro,* and G.A.C. Brito§

1227

Background: The role of nitric oxide (NO) on bone metabo-lism is controversial, since it can either stimulate bone formationor resorption. We investigated the effect of local administrationof the NO donor isosorbide in an experimental periodontal dis-ease model.

Methods: Wistar rats were subjected to a ligature placementaround the cervix of the right second upper molar and were sac-rificed after 11 days. Alveolar bone loss was measured in onequadrant as the sum of the distances between the cuspid tip andthe alveolar bone along the axis of each molar root, which was sub-tracted from the contralateral side, used as unligated control. Thesemiquantitative histopathological scale of the periodontium wasbased on cell infiltration and alveolar bone and cementum integrity.Groups were treated with a gel containing 1% or 5% isosorbideapplied to the vestibular side of the molar gingiva 1 hour beforethe placement of the ligature and then twice daily until sacrifice.Controls included one group subjected to periodontitis and no treat-ment (NT) and another that received the gel containing just thevehicle (V).

Results: The application of the vehicle gel produced an increaseof the alveolar bone resorption, without altering the inflammatorychanges, compared to the NT group. The 5% isosorbide produceda significant reduction of the alveolar bone resorption, comparedto V and NT. This reduction was confirmed by histological analy-sis, showing less inflammatory cell infiltration and preservation ofthe cementum and the alveolar process.

Conclusion: Local application of isosorbide reduces alveolarbone resorption in experimental periodontal disease in rats, sug-gesting a local anti-inflammatory effect of isosorbide. J Periodontol2004;75:1227-1232.

KEY WORDSAnimal studies; bone and bones/metabolism; boneresorption/prevention and control; isosorbide/therapeuticuse; nitric acid.

* Department of Physiology and Pharmacology, Faculty of Medicine, Federal Universityof Ceará, Ceará, Brazil.

† Department of Clinical Medicine, Faculty of Medicine, Federal University of Ceará.‡ Department of Pharmacology, School of Medicine of Ribeirão Preto, University of São

Paulo, SP, Brazil.§ Department of Morphology, Faculty of Medicine, Federal University of Ceará.

Inflammatory periodontal disease repre-sents the most important cause of toothloss in adults.1 Although multifactorial,

the pathogenesis of periodontitis consistsof a cascade of inflammatory and immu-nological reactions, which have not yetbeen fully elucidated.2,3 Research hasshown that proinflammatory cytokines andprostaglandins are essential componentsof the pathogenesis of this disease.4-7

Cytokines such as interleukin-1, tumornecrosis factor, and interferon-γ have beenshown to stimulate nitric oxide (NO) pro-duction in many cell types 8 and NO hasbeen implicated in a vast array of inflam-matory conditions.9,10

Nitric oxide is a short-lived radical thatplays an important role in several bio-logical processes.11 In vascular endothe-lium and nervous tissues, NO is producedconstitutively by calcium/calmodulin-dependent forms of nitric oxide synthase(NOS). Other cell types, including macro-phages,12 neutrophils,13 chondrocytes,14

osteoblasts,15 osteoclasts,8 and bone mar-row cells16 can be induced to produce NOafter exposure to endotoxin or cytokines.NO production by these cells is catalyzedby an inducible form of NOS (iNOS),which is functionally and structurally dis-tinct from the endothelial constitutiveNOS enzyme11,14,17

Some authors advocate an importantrole for NO in periodontal disease18-23 asan additional mediator of bone resorptionresponsible for disease progression. How-ever, this gas appears to play a detrimentalas well as a beneficial role. Detrimentaleffects may include a cytotoxic action

3119.qxd 9/16/04 3:17 PM Page 1227

1228

Isosorbide Decreases Periodontal Bone Resorption Volume 75 • Number 9

towards the adjacent host tissues including alveolarbone.24 The exact role of NO in bone metabolism is stillcontroversial. Apparently, low NO levels are essential tomaintain homeostasis,25 whereas high NO levels, thatoccur during inflammatory conditions, may induce boneresorption.18 This effect may result from the formationof reactive nitrogen species, such as the peroxynitriteanion, that result from the combination of NO and thesuperoxide anion (O2

−).26

On the other hand, the beneficial effects of NO in peri-odontal disease may include antimicrobial activity. It hasbeen reported that NO is an important element of thehost defense against Porphyromonas gingivalis, a pri-mary etiological agent of generalized severe periodonti-tis.27 Additionally, studies have suggested that NO inhibitsosteoclast activity.28,29 The inhibitory effects of high NOconcentrations on bone resorption have been demon-strated by other works in various in vitro models;29-31

however, less is known about the bone effects of NO invivo. The aim of this study was to evaluate the effect ofthe local administration of the NO donor isosorbide in anexperimental periodontitis model in rats.

MATERIALS AND METHODSAnimalsFifty-four Wistar rats, weighing 160 to 200 g, from theFederal University of Ceará were housed in temperature-controlled rooms and received water and food ad libitum.Surgical procedures and animal treatments were con-ducted in accordance with the guidelines of InstitutionalAnimal Care and Use of Ceará Federal University, Brazil.

Experimental Periodontal Disease ProtocolAn experimental periodontitis model described previ-ously was used,32,33 with minor modifications of thedescription of the alveolar bone loss measurements asexplained below. Briefly, rats were anesthetized withcloral hydrate (250 mg/kg, i.p.) and a nylon (000)thread ligature was surgically placed around the cervixof the second left maxillary molar. The ligature wasthen knotted on the vestibular side of the tooth, so thatit remained subgingival on the palatal side and supra-gingival on the buccal side. The contralateral right sidewas used as the unligated control.

Measurements of Alveolar Bone LossThe left and right maxillary halves were defleshed andstained with aqueous methylene blue (1%) in order todifferentiate bone from teeth. Bone loss was analyzedusing a stereoscope loupe (4× magnification) to mea-sure the distance between the cusp tip and the alveo-lar bone crest, as a modification of the method describedby Sallay et al.32 and Samejima et al.,33 who measuredbone loss using the distance from the cemento-enameljunction to the alveolar bone crest. In the present study,measurements were made along the long axis of the

vestibular root surfaces of all molar teeth. Three record-ings for the first (three roots) and two recordings forthe second and third molars (two roots each) weremade. Alveolar bone loss was obtained by subtractingthe values of the right maxilla (unligated control) fromthe left. The sum of the distances for each root wasused as a measure of the total alveolar bone loss on thevestibular surface expressed in mm.

Drug TreatmentThere were six animals in each experimental group. Foreach treatment, one group was used for analysis ofalveolar bone resorption and the other for histopathol-ogy. Groups received a gel containing 1% or 5% isosor-bide (dinitrate 2-mononitrate isosorbide kindly providedby Dr. Fernando Cunha, School of Medicine of RibeirãoPreto, University of São Paulo, Brazil) that was locallyapplied with a gentle massage to the vestibular side ofthe molar gingiva, starting 1 hour before the placementof the ligature and then twice daily until sacrifice. Anes-thesia was not used during gel administration since theprocedure was not painful. Control groups consisted ofanimals subjected to the surgical procedure that receivedthe gel vehicle (V), a non-treated (NT) group that wassubjected to the experimental periodontitis but receivedno further manipulation, and a group of animals notsubjected to experimental periodontitis (bilaterally unli-gated) and that received no treatment (naive).

Histopathological AnalysisMaxillae were excised after sacrifice under anesthesia.The specimens were fixed in 10% neutral buffered for-malin and demineralized in 7% nitric acid. These speci-mens were then dehydrated, embedded in paraffin, andsectioned along the molars in a mesio-distal plane forhematoxylin and eosin staining. Three sections of 6 µmeach, which included two roots of the first and two rootsof the second molar, were selected. The area betweenthe first and second molars, where the ligature was placed,was analyzed under light microscopy, using a 0 to 3 scoregrade (described below), considering the inflammatorycell influx and alveolar bone and cementum integrity, andgraded as previously described by Lima et al.;7 i.e., score0: absence of or only discrete cellular infiltration, pre-served alveolar process, and cementum; score 1: mod-erate cellular infiltration, some but minor alveolar processresorption, and intact cementum; score 2: accentuatedcellular infiltration, accentuated degradation of the alve-olar process, and partial destruction of cementum; score3: accentuated cellular infiltrate, complete resorption ofthe alveolar process, and severe destruction of cementum.

Statistical AnalysisData were described as either means ± SEM or median,as appropriate. Univariate analysis of variance (ANOVA)followed by Bonferroni’s test was used to compare means

3119.qxd 9/16/04 3:17 PM Page 1228

1229

J Periodontol • September 2004 Leitão, Rocha, Chaves, et al.

and Kruskal-Wallis and Mann-Whitney to compare medi-ans; P <0.05 was defined as statistically significant.

RESULTSAnimals treated with 5% isosorbide had significantly lessalveolar bone loss than those treated with the 1% solu-tion. These changes reached statistical significance(P <0.05) when compared to the NT and V groups (Fig. 1).This can be clearly seen in Figure 2A that shows themacroscopic aspects of the contralateral right (unli-gated) side with no resorption of the alveolar bone (Fig.2A-1), severe bone resorption with root exposure in thevehicle group (V) (Fig. 2A-2), and reduction in bone lossin animals subjected to experimental periodontitis andtreated with 5% isosorbide (Fig. 2A-3). The histologi-cal analysis of the region between the first and secondmolars of the naive group (Fig. 2B; Table 1) shows thestructure of the normal periodontium, where gingiva(G), periodontal ligament (PL), alveolar bone (AB),cementum (C), and dentin (D) can be observed. Thehistopathology of animals subjected to periodontitis andreceiving just the vehicle in gel form revealed inflam-matory cell infiltration coupled with severe cementumdestruction and complete alveolar process resorption(Fig. 2C; Table 1), whereas a reduction of inflammatorycell infiltration and a partial preservation of the cemen-tum and the alveolar process were found in the animalssubjected to experimental periodontitis and treated with5% isosorbide (Fig. 2D; Table 1) compared to the vehi-

cle group. These histological changes, as evaluated bythe scoring system used, attained statistical significance(P <0.05; Table 1).

Surprisingly, the animals that received just the gelvehicle displayed a significant increase in the alveolarbone loss, compared to the NT group (P <0.05) (Fig. 1).However, there was no significant difference in the his-tological appearance between the V and NT groups(Table 1).

DISCUSSIONThis study clearly demonstrated that the pharmaco-logical NO donor isosorbide, administered locally (5%gel form), as opposed to 1% isosorbide (gel form),reduced the alveolar bone loss induced by experimentalperiodontal disease in rats.

These data are in agreement with previous studieswhich have reported that NO exerts an inhibitory effecton bone resorption.28-30,34 A biphasic effect of NO onbone has also been reported.31 Experiments with theNO donor S-nitroso-acetyl penicillamine (SNAP) sup-port this view showing generalized suppression of boneresorption at high SNAP concentrations, but potentia-tion of IL-1 induced bone resorption at lower SNAPconcentrations.31 In the present study, however, isosor-bide (1%, gel form) did not alter the bone resorptionseen in periodontal disease. We speculate that this doseof isosorbide was not enough to provide an effectiveconcentration of NO on the periodontal tissue.

Although potential systemic effects of isosorbide havebeen shown,35 we did not observe systemic or con-tralateral effects of the topical isosorbide application onthe left side (unligated side).

Inflammatory diseases, such as rheumatoid arthritisand periodontitis, are associated with increased produc-tion of NO via cytokine activation and increased boneloss, features difficult to reconcile with inhibitory effectsof NO on bone resorption. The cytokine participationin periodontitis is well documented, particularly inter-leukin (IL)-1 and tumor necrosis factor (TNF)-β.32

Previous data from our laboratory showed that chlor-promazine, a TNF synthesis inhibitor, reduced boneloss in experimental periodontitis.7 Indeed, according tothe literature,31 IL-1 and TNF act together to stimulateNO production and bone resorption. However, wheninterferon-γ (IFN) is combined with these cytokines, adramatic stimulation of NO production and inhibitionof bone resorption have been observed.31 Thus, it ispossible that periodontitis is associated with dramaticbone loss because in this condition, IL-1 and TNFproduction are both increased, but IFN production isrelatively deficient, similar to what has been describedin rheumatoid arthrits.36,37 These data are in agree-ment with our findings that clearly show that 5%isosorbide inhibits bone resorption, confirming the anti-resorption effect of NO.

Figure 1.Effect of isosorbide on alveolar bone loss in experimental periodontitis.Data represent the mean ± SEM of six rats in each group. *P <0.05was considered significantly different compared to rats subjected toexperimental periodontitis which did not receive any treatment (NT);†P <0.05 was considered significantly different compared to theexperimental periodontitis vehicle group (V), which received just thevehicle in gel form (ANOVA; Bonferroni’s test).

3119.qxd 9/16/04 3:17 PM Page 1229

The histopathological analysisperformed in the present study de-monstrated that the treatment with5% isosorbide reduced the inflam-matory cell infiltration, coupled withpreservation of the cementum andthe alveolar process, compared withboth the periodontitis control groupwhich received no treatment (datanot shown) or with the groupwhich received just the vehicle.

The reasons for the effect of thisexogenous NO source on amelio-rating periodontitis are not straight-forward. One possible explanationcould be that isosorbide had ananti-inflammatory effect as it wasgiven prophylactically in this study.In accordance to this hypothesis,the present study showed that themigration of inflammatory cellsto the periodontium was signifi-cantly reduced. In fact it has beenreported that NO inhibits theexpression of adhesion surfacemolecules in the endothelium.38

Another possibility is based in theprevious demonstration of the abil-ity of NO to induce osteoclastapoptosis.28,30,31,34,39 Since peri-odontitis is also associated withincreased numbers of osteoclasts,6

it is possible that bone resorptionby isosorbide could be due to aproapoptotic effect of NO in osteo-clasts, thereby limiting their resorp-tive ability. However, the mechanismby which NO induces apoptosis isnot fully clarified.

A rather surprising observation inthis study was that the mere ap-plication of the gel vehicle provokeda significant increase, around 30%,in the alveolar bone resorption, with-out altering the inflammatory para-meters, when compared to animalssubjected to periodontitis that didnot receive further manipulation. Aplausible explanation would be thatthe local trauma induced by thetwice daily application of the gel, inan inflamed area, provoked an in-crease in tooth displacement there-fore increasing the bone resorption.Therefore, to avoid interpretationbias, we compared the isosorbide-

1230

Isosorbide Decreases Periodontal Bone Resorption Volume 75 • Number 9

Figure 2.A) Macroscopic aspects of the normal maxilla (unligated) (1), teeth subjected to experimentalperiodontitis that received just the vehicle, showing severe bone resorption with root exposure(2), and those subjected to experimental periodontitis and treated with 5% isosorbide, showingreduction in bone loss (3). Photomicrographs of region between the first and second molars ofrats. B) Normal maxilla (naive group), showing cementum (C), alveolar bone (AB), gingiva (G),periodontal ligament (PL), and dentin (D). C) Maxilla after 11 days of experimentalperiodontitis, showing severe inflammatory infiltrate in both the gingiva and periodontal ligament,with extensive cementum destruction and total resorption of the alveolar process. D) Maxillaafter 11 days of experimental periodontitis treated with 5% isosorbide, showing discrete cellinflux and preservation of the alveolar process and cementum (bar = 250 µm; hematoxylin andeosin stain; original magnification ×40).

3119.qxd 9/16/04 3:17 PM Page 1230

1231

J Periodontol • September 2004 Leitão, Rocha, Chaves, et al.

Table 1.

Histological Analysis of Rat Maxillae WithExperimental Periodontitis

Naive NT Vehicle 1% ISB 5% ISB

0 (0-0) 3 (2-3)* 3 (2-3)* 3 (2-3)* 1 (1-2)*†

Data are reported as medians with range in parentheses.* P <0.05 was considered significantly different compared to the naive group.† P <0.05 was considered significantly different compared to the vehicle group

(Kruskal-Wallis and Mann-Whitney).

treated groups to the groups that were just subjected toperiodontitis and to the gel vehicle treated group.

In summary, our data provide evidence that a locallyapplied exogenous NO source reduces local inflamma-tory cell infiltration, cementum resorption, and alveolarbone loss in an experimental periodontitis model. Thelack of an effect of the 1% isosorbide, but a significanteffect with 5% isosorbide, is in accordance with the liter-ature,31 which shows that the effect of NO is concentrationdependent. Thus, an understanding of the indications forisosorbide in treating periodontal disease needs furtherstudies on the mechanisms of NO to the alveolar bone.

ACKNOWLEDGMENTSThe authors gratefully acknowledge José Ivan Rodriguesde Sousa and Maria Silvandira França Pinheiro fromthe Faculty of Medicine, Federal University of Ceará,Brazil, for their technical assistance. This work was sup-ported by Conselho Nacional de Desenvolvimento Cien-tífico e Tecnológico-CNPq, Brazil.

REFERENCES1. Williams RC. Periodontal disease. N Engl J Med 1990;

322:373-376.2. Listgarten MA. Nature of periodontal disease: Patho-

genic mechanism. J Periodontal Res 1987;22:172-178.3. Page RC. The role of inflammatory mediators in the patho-

genesis of periodontal disease. J Periodontal Res 1991;26:230-242.

4. Offenbacher S, Odle BM, Van Dyke TE. The use ofcrevicular fluid prostaglandin E2 levels as a predictor ofperiodontal attachment loss. J Periodontal Res 1986;21:101-112.

5. Agarwal S, Piesco NP, Johns LP, Riccelli AE. Differen-tial expression of IL-1β, TNF-α, IL-6 and IL-8 in humanmonocytes in response to lipopolysaccharides from dif-ferent microbes. J Dent Res 1995;74:1057-1065.

6. Bezerra MM, Lima V, Alencar VBM, et al. Selective cyclo-oxygenase-2 inhibition prevents alveolar bone loss inexperimental periodontitis in rats. J Periodontol 2000;71:1009-1014.

7. Lima V, Bezerra MM, Alencar VBM, et al. Effects of chlor-promazine on alveolar bone loss in experimental peri-odontal disease in rats. Eur J Oral Sci 2000;108:123-129.

8. Ralston SH, Todd D, Helfrich MH, Benjamin N, GrabowskiP. Human osteoblast-like cells produce nitric oxide andexpress inducible nitric oxide synthase. Endocrinology1994;135:330-336.

9. Lappin DF, Kjeldsen M, Sander L, Kinane DF. Induciblenitric oxide synthase expression in periodontitis. J Peri-odontal Res 2000;25:369-373.

10. Lohinai Z, Stachlewitz R, Virág L, Székely AD, Haskó G,Szabó C. Evidence for reactive nitrogen species forma-tion in the gingivomucosal tissue. J Dent Res 2001;80:470-475.

11. Moncada S, Higgs A. The L-arginine-nitric oxide pathway.N Engl J Med 1993;329:2002-2012.

12. Schmidt HHHW, Warner TD, Nakane M, Forstermann U,Murad F. Regulation and subcellular location of nitro-gen oxide synthases in RAW264.7 macrophages. MolPharmacol 1992;41:615-624.

13. McCall TB, Boughton-Smith NK, Palmer RMJ, WhittleBJR, Moncada S. Synthesis of nitric oxide from L-argi-nine by neutrophils. Biochem J 1989;261:293-296.

14. Palmer RMJ, Hickery MS, Charles IG, Moncada S, BaylissMT. Induction of nitric oxide synthase in human chon-drocytes. Biochem Biophys Res Commun 1993;193:398-405.

15. Kukkanem I, Hughes FJ, Buttery LDK, et al. Cytokine-stimulated expression of inducible nitric oxide synthaseby mouse, rat, and human osteoblast-like cells and itsfunctional role in osteoblast metabolic activity. Endocrinol-ogy 1995;136:5445-5453.

16. Punjabi CJ, Laskin DL, Heck DE, Laskin JD. Productionof nitric oxide by murine bone marrow cells. J Immunol1992;149:2179-2184.

17. Moncada S, Palmer RMJ, Higgs AE. Nitric oxide: Phys-iology, pathophysiology and pharmacology. PharmacolRev 1991;43:109-142.

18. Lohinai Z, Benedek P, Fehér E, et al. Protective effectsof mercaptoethylguanidine, a selective inhibitor ofinducible nitric oxide synthase, in ligature-induced peri-odontitis in the rat. Br J Pharmacol 1998;123:353-360.

19. Mantejka M, Partyka L, Ulm C, et al. Nitric oxide syn-thesis is increased in periodontal disease. J PeriodontalRes 1998;33:517-518.

20. Gaspirc B, Masera A, Skaleric U. Immunolocalization ofinducible nitric oxide synthase in localized juvenile peri-odontitis patients. Connect Tissue Res 2002;43:413-418.

21. Shibata K, Warbington ML, Gordon BJ, Kurihara H, VanDyke TE. Nitric oxide synthase activity in neutrophilsfrom patients with localized aggressive periodontitis.J Periodontol 2001;72:1052-1058.

22. Daghigh F, Borghaei RC, Thornton RD, Bee JH. Humangingival fibroblasts produce nitric oxide in response to pro-inflammatory cytokines. J Periodontol 2002;73:392-400.

23. Batista AC, Silva TA, Chun JH, Lara VS. Nitric oxidesynthesis and severity of human periodontal disease.Oral Dis 2002;8:254-260.

24. Lin SK, Kok SH, Kuo MYP, et al. Nitric oxide promotesinfectious bone resorption by enhancing cytokine-stimu-lated interstitial collagenase synthesis in osteoblasts.J Bone Miner Res 2003;18:39-46.

25. Brandi ML, Hukkanen M, Umeda T, et al. Bidirectionalregulation of osteoclast function by nitric oxide synthaseisoforms. Proc Natl Acad Sci (USA) 1995;92:2954-2958.

26. Murphy MP. Nitric oxide and cell death. Biochim Bio-phys Acta 1999;1411:401-414.

27. Gyurko R, Boustany G, Huang PL, et al. Mice lackinginducible nitric oxide synthase demonstrate impairedkilling of Porphyromonas gingivalis. Infect Immun 2003;71:4917-4924.

28. MacIntyre I, Zaidi M, Towhidul ASM, et al. Osteoclastinhibition: An action of nitric oxide not mediated by cyclicGMP. Proc Natl Acad Sci (USA) 1991;88:2936-2940.

3119.qxd 9/16/04 3:17 PM Page 1231

1232

Isosorbide Decreases Periodontal Bone Resorption Volume 75 • Number 9

29. Stern PH, Diamond J. Sodium nitroprusside increasescyclic GMP in fetal rat long bone cells and inhibits resorp-tion of fetal rat limb bones. Res Commun Chem PatholPharmacol 1992;75:19-28.

30. Lowik CWGM, Nibbering PH, Van Der Ruit M, PapapoulosSE. Inducible production of nitric oxide in osteoblast-likecells and in fetal mouse bone explants is associated withsupression of osteoclastic bone resorption. J Clin Invest1994;93:1465-1472.

31. Ralston SH, Ho LP, Helfrich M, Grabowski PS, JohnstonPW, Benjamin N. Nitric oxide: A cytokine induced regu-lator of bone resorption. J Bone Miner Res 1995;10:1040-1049.

32. Sallay K, Sanavi F, Ring I, Pham P, Behling UH, NowotnyA. Alveolar bone destruction in the immuno-suppressedrat. J Periodontal Res 1982;17:263-274.

33. Samejima Y, Ebisu S, Okada H. Effect of injection withEikenella corrodens on the progression of ligature-induced periodontitis in rats. J Periodontal Res 1990;25:308-315.

34. Kasten TP, Collin-Osdoby P, Patel N, et al. Potentiationof osteoclast bone-resorption activity by inhibition ofnitric oxide synthase. Proc Natl Acad Sci (USA) 1994;91:3569-3573.

35. Gomita Y, Furuno K, Araki Y. Influence of electric footshock on pharmacokinetics of isosorbide dinitrate orallyadministered to rats. Japan J Pharmacol 1989;49:297-299.

36. Galbraith GMP, Hagan C, Steed RB, Sanders JJ, JavedT. Cytokine production by oral and peripheral bloodneutrophils in adult periodontitis. J Periodontol 1997;68:832-838.

37. Sambrook PN, Reeve J. Bone disease in rheumatoidarthritis. Clin Sci 1988;74:225-230.

38. Kubes P, Suzuki M, Granger DN. Nitric oxide: An endoge-nous modulator of leukocyte adhesion. Proc Natl AcadSci (USA) 1991;88:4651-4655.

39. Brune B, Von Knethen A, Sandau KB. Nitric oxide and itsrole in apoptosis. Eur J Pharmacol 1998;351:261-272.

Correspondence: Dr. Gerly Anne de Castro Brito, Departa-mento de Fisiologia e Farmacologia, Faculdade de Medicina,Universidade Federal do Ceará, Rua Coronel Nunes de Melo,1127 Rodolfo Teófilo, 60.430-270 Fortaleza, Ceará, Brazil.Fax: 55-85-288-8333; e-mail: gerlybrito@hotmail.com.

Accepted for publication January 9, 2004.

3119.qxd 9/16/04 3:17 PM Page 1232