A Lectin from the Green Seaweed Caulerpa Cupressoides Reduces Mechanical Hyper-Nociception and Inflammation in the Rat Temporoma

International Immunopharmacology 21 (2014) 34–43

Contents lists available at ScienceDirect

International Immunopharmacology

journal homepage: www.elsevier.com/locate/intimp

A lectin from the green seaweed Caulerpa cupressoides reduces mechanical hyper-nociception and inﬂammation in the rat temporomandibular joint during zymosan-induced arthritis

Renata Line da Conceição Rivanor a, Hellíada Vasconcelos Chaves c, Danielle Rocha do Val b, Alice Ramos de Freitas c, Jonas Cavalcante Lemos b, José Ariévilo Gurgel Rodrigues a, Karuza Maria Alves Pereira c, Ianna Wivianne Fernandes de Araújo a, Mirna Marques Bezerra b, Norma Maria Barros Benevides a,⁎ a Department of Biochemistry and Molecular Biology, Federal University of Ceará, Fortaleza, Ceará, Brazil b Faculty of Medicine, Federal University of Ceará, Sobral, Ceará, Brazil c Faculty of Dentistry, Federal University of Ceará, Sobral, Ceará, Brazil article info abstract

Article history: Seaweed lectins have been widely investigated as anti-nociceptive and anti-inflammatory agents. This study Received 14 January 2014 analyzed the anti-nociceptive and anti-inflammatory responses of a lectin from the green seaweed Caulerpa Received in revised form 7 April 2014 cupressoides (CcL) on zymosan-induced arthritis of the rat temporomandibular joint (TMJ). Rats received i.v. Accepted 9 April 2014 CcL 30 min prior to injection of zymosan (2 mg/art.) or 0.9% saline into the left TMJ. Mechanical hyper- Available online 24 April 2014 nociception was measured by the electronic von Frey method at baseline and 4 h after zymosan injection. Animals were euthanized 6 h after zymosan injection and the synovial fluid was collected for leukocyte counting Keywords: fi Marine alga and myeloperoxidase activity assessment. Other animals were treated with ZnPP-IX (3 mg/kg; s.c.), a speci c Hemagglutinin heme oxygenase-1 pathway inhibitor, and naloxone (10 μg/art.), a nonselective opioid receptor antagonist. Mechanical hyper-nociception TMJ tissues were excised to perform histopathological and immunohistochemistry analyses. CcL (0.1, 1 or Inflammatory arthropathies 10 mg/kg) significantly reduced zymosan-induced hyper-nociception (81, 83 and 89.5%, respectively) and inhibited the leukocyte influx (77.3, 80.7 and 98.5%, respectively) compared with the zymosan-only group, as confirmed by myeloperoxidase activity; however, treatment with naloxone or ZnPP-IX did not revert the effects of CcL (10 mg/kg), suggesting that the naloxone-sensitive opioid and heme oxygenase-1 pathways are not involved. CcL also reduced the leukocyte influx and the expression of IL-1β and TNF-α in the TMJ, based on histopathological and immunohistochemistry analyses, respectively. Therefore, CcL reduces TMJ hyper- nociception and inflammation with a mechanism that is partially dependent on TNF-α and IL-1β inhibition. CcL reveals a potentially valuable alternative tool for future studies of TMJ disorders. © 2014 Elsevier B.V. All rights reserved.

1. Introduction agent zymosan [1]. Zymosan is a polysaccharide from yeast cell walls that produces a severe and erosive synovitis [2,3] associated with inflam- Temporomandibular joint (TMJ) disorders are a group of conditions matory pain in animal models of knee arthritis [4]. that result in TMJ pain, which frequently limits talking, chewing, and Increased levels of pro-inflammatory cytokines (IL-1β and TNF-α) other basic daily activities with high levels of pain-related disability. To have been detected in patients with temporomandibular disorders clarify the underlying pathogenesis of temporomandibular disorder- (TMD), suggesting that these cytokines may play a role in the pathogen- related inflammatory pain, we have developed a rat model of TMJ esis of synovitis and degenerative changes of the cartilaginous tissue inflammation using intra-articular injections of the pro-inflammatory and bone of the temporomandibular joint [5,6]. Heme oxygenase 1 (HO-1) plays an important role in the antioxi- dant defense system, and its induction would provide a negative feed- back for cell activation and the production of inflammatory mediators, which could modulate, at least in part, the inflammatory pain process ⁎ Corresponding author at: Universidade Federal do Ceará, Departamento de [7–9]. A role for the endogenous opioid system in the phenomenon of Bioquímica e Biologia Molecular, Bloco 907, Campus do Pici, CEP: 60440-900, Brazil. Tel.: +55 85 3366 94 02; fax: +55 85 3366 97 89. pain [10] and a role for the peripheral opioid receptors in TMJ anti- E-mail address: [email protected] (N.M.B. Benevides). nociception, are well-demonstrated [11,12].

Non-steroidal anti-inflammatory drugs are largely used to amelio- 2.3. Marine alga rate TMJ inflammation, although these drugs are associated with several adverse effects (mainly gastrointestinal, cardiovascular and kidney tox- The green seaweed C. cupressoides var. lycopodium (Caulerpaceae, icity) [13,14]. Therefore, there is a great need for scientific studies that Bryopsidales) was collected from the Atlantic coast of Brazil (Pacheco search for new bioactive compounds, such as those derived from plants, Beach, Caucaia, Ceará). After collection, the material was cleaned of epi- with therapeutic action and that confirmed efficacy and safety [15–18]. phytes, washed with distilled water and stored at −20 °C until use. A Seaweeds have lectins, which are proteins or glycoproteins of non- voucher specimen (no. 4977) was deposited in the Herbarium Prisco immune origin containing at least one non-catalytic domain that binds Bezerra in the Department of Biological Sciences, Federal University of reversibly to specific mono- or oligosaccharides [19]. The biotechnolog- Ceará, Brazil. ical importance of these polymers has been described for various fields, including biochemistry [20], agriculture [21] and, more recently, they 2.4. Erythrocytes have been recognized as important tools to control nociception [22] and inflammation [23]. However, to the best of our knowledge [1,24], Blood samples were obtained from healthy adult New Zealand there are no studies about the use of lectins on experimental arthritis albino rabbits and maintained in the Animal Care Unit of the Federal of the rat TMJ. University of Ceará, Fortaleza, Brazil. Caulerpa cupressoides var. lycopodium C. Agardh is a species of green seaweed belonging to the family Caulerpaceae and is widely found 2.5. Isolation of C. cupressoides lectin (CcL) along the coast of northeast Brazil. The isolated lectin from this algal species (CcL) was previously purified and characterized by physi- The CcL was obtained from the fresh algal tissue macerated in the cochemical procedures that indicated a dimer (23–44 kDa) with presence of liquid nitrogen [25], with some modifications that were hemagglutinating activity containing glycine, aspartic acid, glutamic proposed by Vanderlei et al. [26]. Extraction of CcL was performed acid, serine and a low content of basic amino acids. Furthermore, it with Tris–HCl 25 mM, pH 7.5, (TB) at a ratio of 1:4 (w/v, alga:TB). has been reported in the literature that this lectin can be inhibited by After constant stirring for 4 h, the homogenate was filtered through derivatives of galactose and strongly inhibited by the glycoprotein nylon cloth and then centrifuged (6000 ×g, 30 min, 4 °C). The precipi- porcine stomach mucin, a glycoprotein with a terminal GalNAc residue, tate was discarded and the resulting supernatant (crude extract) was and fucose and galactose as internal residues [25]. subjected to ion-exchange chromatography on a DEAE-cellulose col- Our group demonstrated CcL anti-nociceptive and anti-inflammatory umn (1.0 × 27.5 cm), equilibrated and eluted with TB. After elution of effects in classical models of nociception and acute inflammation in vivo the unbound proteins and pigments, the adsorbed proteins were eluted [26], as well as immunomodulatory properties on cell viability in vitro, from the gel using TB containing 0.5 M NaCl. The fraction containing inducing T helper 2 immune responses in mouse splenocytes and hemagglutinating activity was pooled, dialyzed and freeze-dried. producing high levels of IL-10 and IL-6 [27]. Therefore, we analyzed the unexplored anti-nociceptive and anti- 2.6. Experimental protocols inflammatory efficacy of CcL in a model of zymosan-induced TMJ inflammatory hyper-nociception in the rat. Additionally, we investigated 2.6.1. Induction of TMJ arthritis the putative involvement of IL-1β, TNF-α, HO-1 and the endogenous This method was based on Chaves et al. [1]. Thirty minutes before in- opioid system in CcL efficacy. jection with zymosan, rats were pretreated (0.1 ml/100 g body weight) with CcL (0.1, 1 or 10 mg/kg) by intravenous (i.v.) injection or 0.9% sterile saline. Control animals (sham) received the same volume of sterile 2. Materials and methods saline. Rats were briefly anesthetized with inhaled isoflurane and received an intra-articular (i.art.) injection of zymosan (2 mg, 40 μltotal 2.1. Animals volume) dissolved in sterile saline into the left TMJ using a 30-gauge needle and a 1 ml syringe. Sham animals received saline i.art. Male Wistar rats (180 –240 g) from the Animal Care Unit of the Before the zymosan or saline injection, the TMJ skin region was care- Federal University of Ceará in Fortaleza, Brazil, were used throughout fully shaved, the posteroinferior border of the zygomatic arch was pal- the experiments. They were housed in a temperature-controlled room pated, and the needle was inserted inferior to this point and advanced with free access to water and food on a 12/12 h light/dark cycle. For in a medial and anterior direction until the needle made contact with each experiment, groups of six animals were segregated and handled the condyle. This contact was verified by moving the mandible, and separately. All procedures and animal treatments were carried out at the puncture of the needle into the joint space was confirmed by the ambient temperature (20–22 °C), and special care was taken to avoid loss of resistance. The animals were euthanized under anesthesia at 6 h environmental disturbances that might influence animal responses. after zymosan-induced arthritis and inflammatory parameters (total This study was conducted in accordance with the guidelines set forth cell count and myeloperoxidase assay activity) were evaluated. by the U.S. Department of Health and Human Services and with the approval of the Ethics Committee of the Federal University of Ceará, 2.6.2. Evaluation of mechanical hyper-nociception Fortaleza, Brazil (CEPA no. 59/13). Mechanical hyper-nociception in the rat TMJ was evaluated by measuring the threshold intensity of force that, when applied to the TMJ region, would elicit a reflex response in the animal (e.g., head with- 2.2. Drugs and reagents drawal). The measurements consisted of a continuous force applied to the TMJ area, performed by an examiner unaware of the treatments, The following drugs and reagents were used: zymosan, Zinco and using a digital device (Insight, Ribeirão Preto, SP, Brazil) that Protoporfirina-IX (ZnPP-IX), indomethacin, DEAE-cellulose, ο-dianisidine consisted of a rigid filament linked to an electronic device that mea- dihydrochloride, potassium phosphate monobasic, and potassium sured the response threshold in grams (g) when the filament was phosphate dibasic which were purchased from Sigma (St. Louis, MO, applied to the surface of the tested region. The facial areas to be tested USA); hematoxylin, eosin and ethylenediaminetetraacetic acid (EDTA) around the TMJ were carefully shaved, and the animals were placed were purchased from VETEC Química Farm. LTDA, SP (Brazil). The re- into individual plastic cages 45 min before the beginning of the tests. agents and the lectin (CcL) were solubilized in 0.9% sterile NaCl (saline) The animals were submitted to conditioning sessions, consisting or ultrapure water. All chemicals were of analytical grade. of head withdrawal threshold measurements, in the testing room for 36 R.L. da Conceição Rivanor et al. / International Immunopharmacology 21 (2014) 34–43

4 consecutive days under controlled temperatures (20–22 °C) and low mounted on glass slides previously prepared with an organosilane- illumination intensity [28]. On the fifth day, the basal force threshold based adhesive (3-aminopropyltriethoxysilane). Briefly, it consisted of value was recorded (in triplicate) before injection of CcL (0.1, 1 or the following steps: the sections went through 2 baths in xylol, each 10 mg/kg, i.v.), zymosan (2 mg/i.art., 40 μl), sterile saline (0.9% NaCl, lasting 10 min. After this, they were immersed in three passages of ab- w/v, i.art.) or indomethacin (5 mg/kg body weight, s.c.) and was record- solute alcohol, then washed in running water, and immediately passed ed again after 4 h. through distilled water. After antigen retrieval, endogenous peroxidase was blocked 2.6.3. Collection of the TMJ synovial fluid and cell counting (15 min) with 3% (v/v) hydrogen peroxide, and the sections were Six hours after zymosan injection (i.art.), the rats were euthanized washed in phosphate-buffered saline (PBS). Sections were incubated under anesthesia and exsanguinated. The superficial tissues were dis- overnight (4 °C) with a primary rabbit anti-TNF-α,IL-1β and HO-1 sected, and the TMJ cavity was washed to collect the synovial fluid by antibodies (ab13243, ABCAM®, England, UK), at a dilution of 1:200, a pumping-and-aspiration technique using 0.05 ml of EDTA in neutral and afterwards washed with a phosphate buffered saline solution, PBS buffered PBS. This procedure was repeated twice. The total number of (phosphate buffered saline). white cells in the synovial lavage fluid was counted using a Neubauer The slides were incubated with a secondary antibody LSAB Kit chamber [1]. (DAKO®, Carpinteria, CA, EUA) for 10 min at ambient temperature. The slides were then incubated with a biotinylated goat anti-rabbit an- 2.6.4. Determination of myeloperoxidase (MPO) activity tibody diluted 1:400 in PBS-BSA. After washing, the slides were incubat- MPO is an enzyme found primarily in azurophilic granules within ed with an avidin–biotin–horseradish peroxidase conjugate (Strep ABC neutrophils and has been used extensively as a biochemical marker of complex, VECTASTAIN ABC Reagent and peroxidase substrate solution) granulocyte infiltration into various tissues. The extent of neutrophil for 30 min according to the VECTASTAIN protocol. IL-1β and TNF-α accumulation in the TMJ synovial fluid was measured using an MPO were visualized with the chromogen 3,3-diaminobenzidine (DAB). Neg- activity assay as previously described [29].Briefly, MPO activity was ative control sections were processed simultaneously as described determined by measuring the change in absorbance at 450 nm using above but with the first antibody being replaced by 5% PBS-BSA. None o-dianisidine dihydrochloride and 1% hydrogen peroxide. A unit of of the negative controls showed IL-1β, TNF-α or HO-1 immunoreactiv- MPO activity was defined as the activity that required the conversion ity. Counter-staining was performed with hematoxylin, and afterwards of 1 μmol of hydrogen peroxide to water within 1 min at 22 °C. The the specimens were dehydrated in alcohol and diaphanized in xylol. results are reported as MPO units/joint. Positive values were assigned to all cells that exhibited brown staining in the cytoplasm, irrespective of the staining intensity. 2.6.5. Histopathological analysis Six hours after zymosan-induced arthritis, the rats were euthanized 2.6.9. Statistical analysis and the TMJ was excised. The specimens were fixed in 10% neutral buff- The data are presented as the mean ± standard errors (S.E.M.) for ered formalin for 24 h, demineralized in 10% EDTA (7% formic acid), six animals per group. Differences between the means were compared embedded in paraffin, and sectioned along the long axis of the TMJ. Sec- using a one-way ANOVA followed by the Bonferroni test. In the histo- tions of 5 μm, which included the condyle, articular cartilage, articular pathological analysis of the TMJ, the Kruskal–Wallis test was used, disk, synovial membrane and periarticular tissue, were evaluated followed by Dunn's test to compare medians. Values of p b 0.05 were under light microscopy. For the specimens processed for routine hema- considered to be statistically significant. toxylin–eosin (H & E) staining, histological analysis was used to assign a graded 0–4 score based on the cell influx into the synovial membrane, 3. Results the periarticular tissue and the musculoskeletal tissue [1]. 3.1. CcL reduces mechanical hyper-nociception, leukocyte cell count and 2.6.6. Evaluation of possible involvement of the opioid system MPO activity on zymosan-induced TMJ acute arthritis in rats To assess the possible participation of the opioid system in the antinociceptive response to CcL, rats were pre-treated with naloxone The i.art. injection of zymosan (2 mg/art., 40 μl) elicited (p b 0.001) a (10 μg/art., 15 μl), 35 min before the injection with zymosan (2 mg/art., mechanical hyper-nociception response (58.4 ± 6.4 g) compared with 35 μl). Five minutes later, the animals received the CcL (10 mg/kg, i.v.), the sham group (118.8 ± 4 g) at 4 h after the inflammatory stimulus, morphine (1000 μg, 0.1 ml, s.c.) or 0.9% sterile saline. The mechanical as measured by a clear decrease in the mechanical threshold for head hyper-nociception responses to these treatments were recorded after 4 withdrawal. The i.v. treatment of rats with CcL (0.1, 1 or 10 mg/kg), h of zymosan (2 mg/art., 35 μl) or sterile saline injected i.art. into the administered 30 min prior to zymosan into the rat left TMJ, caused a left TMJ of the rats. When two drugs were injected in the same TMJ, the reduction in mechanical hyper-nociception at all tested doses by 81% total volume of injection was 50 μl [30]. (107.3 ± 4.4 g), 83% (108.5 ± 5 g) and 89.5% (112.5 ± 3.4 g), respectively, (p b 0.001), as measured at 4 h after zymosan injection. Indo- 2.6.7. Evaluation of possible involvement of the HO-1 pathway in methacin (5 mg/kg, s.c.) also produced an anti-nociception effect of nociceptive and inflammatory responses 69%(100±8.5g,pb 0.01) (Fig. 1A). This assay was based on Freitas et al. [31], Vanderlei et al. [9] and Additionally, the i.art. injection of zymosan into the rat left TMJ re- Grangeiro et al. [8]. Animals (n = 6 per group) were treated with ZnPP- sulted in a significant increase in the number of polymorphonuclear IX (3 mg/kg, s.c.), followed by administration of CcL (10 mg/kg, i.v.) 30 cells (19,500 ± 241 cells/mm3) from the TMJ synovial fluid at 6 h, com- min later. Zymosan (2 mg/art., 40 μl) was injected after 1 h. Mechanical pared to the sham group (31 ± 6 cells/mm3). Pretreatment of rats with hyper-nociception and cell number from the TMJ synovial lavage fluid CcL (0.1, 1 or 10 mg/kg, i.v.) decreased (p b 0.001), in a dose-dependent were measured after 4 and 6 h following the stimulus, respectively. manner, the leukocyte cell count from the TMJ synovial lavage fluid by MPO activity assessment was also performed to confirm the data. The 77.3% (4450 ± 808.5 cells/mm3), 80.7% (3750 ± 1233 cells/mm3)and results were expressed as in Sections 2.6.2, 2.6.3 and 2.6.4. 98.5% (295.5 ± 156 cells/mm3), respectively, in comparison with the zymosan group (Fig. 1B). Indomethacin resulted in 90.7% (1800 ± 2.6.8. Immunohistochemistry for IL-1β,TNF-α and HO-1 638 cells/mm3) inhibition (p b 0.001) of the leukocyte number in the Immunohistochemistry for IL-1β,TNF-α and HO-1 were performed synovial fluid. The MPO activity also decreased in the TMJ synovial using the streptavidin–biotin (Labeled Streptavidin Biotin — LSAB) meth- lavage fluid at all doses tested by 63% (29.5 ± 9 U/joint fluid), 92% od in formalin-fixed, paraffin-embedded tissue sections (4 μm thick), (10 ± 4.5 U/joint fluid) and 98% (2.5 ± 1 U/joint fluid), respectively R.L. da Conceição Rivanor et al. / International Immunopharmacology 21 (2014) 34–43 37 AB

Fig. 1. Effect of CcL on mechanical hyper-nociception (A), leukocyte influx (B) and MPO activity (C) of zymosan (Zy)-induced TMJ arthritis (4 h post-injection). Zy (2 mg/art., 40 μl) or 0.9% saline was injected i.art. into the left TMJ of the rats. Indomethacin (5 mg/kg) was given s.c. 1 h before stimulus. Animals received s.c. saline or CcL (0.1, 1 or 10 mg/kg, i.v.). Data are expressed as the mean ± S.E.M. of six animals for each group (ANOVA, Bonferroni's test). ***p b 0.001 indicates significant difference from the sham group. ###p b 0.001 indicates asignificant difference from the Zy group. +++p b 0.001 indicates a significant difference from the indomethacin group.

(p b 0.05) (Fig. 1C). Animals pretreated with indomethacin (5 mg/kg, and the skeletal muscle tissue, compared to the sham group. The cell s.c.) also showed anti-inflammatory effects (55.5%, 53.3 ± 2 cells/mm3, type was predominantly neutrophils, which are known to characterize p b 0.05). acute inflammation. Edema was also observed in the synovium. Pre- treatment with CcL (10 mg/kg, i.v.) reduced the cellular infiltrate and edema in the synovial membrane, periarticular tissue (p b 0.01) and 3.2. CcL reverses tissue alteration in the zymosan-inflamed TMJ as assessed skeletal muscle tissue (p b 0.05), when compared to the zymosan by H & E staining group. The inflammatory process was considered as moderate to severe. These data can be seen in the TMJ photomicrographs (Fig. 2). Table 1 shows the scores attributed to TMJ histopathological analysis and compares the values between the sham and zymosan groups. Groups of animals were treated with zymosan (2 mg/art., 40 μl) and 3.3. Naloxone does not block the anti-nociceptive effect of CcL CcL (10 mg/kg, i.v.). The sham group received injection (i.art.) of saline in the same volume. TMJ tissues were fixed, sectioned and stained with Rat pretreatment with naloxone (10 μg/art., 15 μl), 5 min before CcL H & E. A significant (p b 0.05) increase in the inflammatory parameters (10 mg/kg, i.v.), did not interfere the antinociceptive response of this was observed in the zymosan group. compound on zymosan-induced TMJ acute arthritis in rats. By contrast, Six hours after zymosan-induced TMJ arthritis, an inflammatory cell at this dose, naloxone blocked the antinociceptive action of morphine influx was observed in the synovial membrane, the periarticular tissue (1000 μg, 0.1 ml, s.c.) in this model (Fig. 3).

Table 1 Histopathological analysis of TMJ on zymosan-induced arthritis: the effects of CcL (10 mg/kg, i.v.).

Groups Cell influx in the synovial membrane Cell influx in the periarticular tissue Cell influx in the muscular tissue

Sham 0 (0–0) 0 (0–0) 0 (0–0) ⁎ ⁎ ⁎ Zy 4 (3–4) 4(3–4) 2(2–3) ⁎⁎⁎ ⁎⁎⁎ ⁎⁎ CcL 0 (0–0) 1(1–1) 0(0–2)

0 — uninjured; 1 — low injury; 2 — mild injury; 3 — moderate injury; 4 — severe injury. ⁎ p b 0.05 versus sham group (Kruskal–Wallis, Dunn's). ⁎⁎ p b 0.05 indicate signiﬁcant difference in relation to Zy group (Kruskal-Wallis, Dunn's). ⁎⁎⁎ p b 0.01 indicate signiﬁcant difference in relation to Zy group (Kruskal–Wallis, Dunn's). 38 R.L. da Conceição Rivanor et al. / International Immunopharmacology 21 (2014) 34–43

Fig. 2. Histopathological analysis of zymosan-induced TMJ arthritis in rats. (A) TMJ photomicrographs from the sham group (40×). (B, C and D) Synovial membrane, periarticular tissue and musculoskeletal tissue of the sham group without cell influx (400×). (E) Photomicrographs of TMJ in the Zy group (2 mg/art., 40 μl) (40×). (F, G and H) Synovial membrane, periarticular tissue and musculoskeletal tissue of the zymosan group with intense cell influx (400×). (I) Photomicrographs of the TMJ with CcL (10 mg/kg, i.v.) (40×). (J, L and M) Synovial membrane periarticular tissue and musculoskeletal tissue from the group treated with CcL without cell influx (400×). C: condyle; AC: articular cartilage; AD: articular disk; SM: synovial membrane; PAT: periarticular tissue; and MET: musculoskeletal tissue. H & E staining. Black arrows indicate neutrophils.

3.4. CcL does not have anti-nociceptive and anti-inflammatory actions Table 2 shows the scores attributed to TMJs and compares the values through the HO-1 pathway between the zymosan arthritic TMJ group, CcL (10 mg/kg, i.v.), specific HO-1 inhibitor ZnPP-IX (3 mg/kg), and ZnPP-IX plus CcL. The histopath- To investigate the possible role of the HO-1 pathway in the anti- ological analyses showed that CcL reduced the inflammatory parame- nociceptive and anti-inflammatory effects of CcL, the rats were ters to a normal level with a low inflammatory cell influx in the pretreated (s.c.) with ZnPP-IX (3 mg/kg), which is a specific HO-1 inhib- periarticular tissue (p b 0.05), compared to the zymosan group. The in- itor, resulting in the inhibition of this pathway. It was observed that the flammatory process was considered moderate to severe for the synovial anti-nociceptive and anti-inflammatory effects of the CcL (10 mg/kg, membrane and periarticular tissue, and for the musculoskeletal tissue, i.v.) on the zymosan-induced arthritis in the rat TMJ model were not the inflammatory process was considered medium to moderate. inhibited in the presence of ZnPP-IX (Fig. 4A, B), a result that was further The group treated with ZnPP-IX maintained a moderate to severe confirmed by MPO activity measurement and histopathological analysis inflammatory response for the synovial membrane and periarticular (Fig. 4C). tissue, and a medium to moderate response for the musculoskeletal tissue. The group treated with ZnPP-IX + CcL showed reduced inflammatory cell influx in the synovial membrane and the periarticular tissue; however, only the periarticular tissue was statistically significant. These data can be seen in the TMJ photomicrographs (Fig. 5).

3.5. CcL decreases IL-1β and TNF-α abundance in zymosan-induced TMJ arthritis in rats

Immunohistochemical analysis showed an increase in TNF-α, IL-1β and HO-1 expression, which was characterized by brown-colored cells in the arthritic TMJ of rats injected with zymosan (i.art.), when compared with the sham group (Figs. 6, 7, 8). The sham control group showed expression of TNF-α,IL-1β and HO-1 only in synovial cells. The negative control group sections were composed of arthritic TMJs of rats injected with zymosan that were not treated with anti-TNF-α, anti-IL-1β or anti-HO-1 antibodies. A significant decrease in TNF-α expression was observed in the Fig. 3. Evaluation of the possible involvement of endogenous opioids in the anti- CcL-treated group compared with the zymosan group because the nociceptive effect of CcL on zymosan (Zy)-induced mechanical hyper-nociception and TMJ arthritis in the rat. Zy (2 mg/art., 40 μl) or 0.9% saline was injected i.art. into the left immunohistochemical analysis did not reveal expression of this TMJ of the rats. Animals received s.c. saline or CcL (10 mg/kg, i.v.) with or without nalox- pro-inflammatory cytokine in the synovial membrane or in the neutro- one (10 μg/art.) and morphine (1000 μg, 0.1 μl, s.c.) with or without naloxone. Data are phils. These profiles suggest an immunocellular response by CcL during expressed as the mean ± S.E.M. of six animals for each group (ANOVA, Bonferroni's the acute phase. For IL-1β, there was an intense staining in the synovial test). ***p b 0.01 indicates significant difference from the sham group. ###p b 0.001 indicates significant difference from the Zy group. +++p b 0.001 indicates significant cells and neutrophils in the zymosan group, and this staining was re- difference from the group morphine. &&&p b 0.001 indicates significant difference from duced in the CcL group, similar to the level of the sham controls. This the morphine group with CcL. basal expression of IL-1β and TNF-α in the sham group may be due to R.L. da Conceição Rivanor et al. / International Immunopharmacology 21 (2014) 34–43 39 AB p < 0.001

C p < 0.001

Fig. 4. Effect of CcL in combination with ZnPP-IX on mechanical hyper-nociception (A), leukocyte influx (B) and MPO activity (C) in the rat zymosan (Zy)-induced TMJ arthritis model. Zy (2 mg, 40 μl) or 0.9% saline was injected i.art. into the left TMJ of the rats. Animals received s.c. saline or CcL (0.1, 1 or 10 mg/kg, i.v.). Data are expressed as the mean ± S.E.M. of six animals for each group (ANOVA, Bonferroni's test). ***p b 0.001 indicates a significant difference from the sham group. ###p b 0.001 indicates a significant difference from the Zy group. +++p b 0.001 indicates a significant difference in relation to group ZnPP-IX. + and ++ indicate significant difference (p b 0.05 and p b 0.01, respectively) in relation to the ZnPP-IX group. an injury caused by i.art. injection, nonetheless, the zymosan group demonstrated for the first time the anti-nociceptive and anti- showed intense staining for both cytokines from the synovial cells and inflammatory efficacy of C. cupressoides lectin (CcL) in a model of neutrophils (Figs. 6, 7). zymosan-induced TMJ arthritis in rats. Furthermore, we also evaluat- Finally, immunohistochemical analysis showed intense expression ed the putative involvement of leukocyte influx, cytokine expression of HO-1 in synovial cells and neutrophils from the synovial membrane (IL-1β and TNF-α), the endogenous opioid system and HO-1 path- of rats pretreated with zymosan, and expression of HO-1 in synovial ways in CcL efficacy. cells in the CcL group was similar to that of the zymosan group. In addi- Experimental animal models of TMJ inflammatory hyper-nociception tion, a basal constitutive level of HO-1 in the synovial cells was noted in and arthritis have been proposed for the study of painful inflammatory the sham group (Fig. 8). conditions. Some authors have suggested inducing TMJ arthritis surgical- ly [33] or mechanically [34], using systemic [35,36],submandibular[37], 4. Discussion or intra-articular (i.art) injections [38]. Recently, our group set up a new model of zymosan-induced arthri- Previous studies evaluating biopolymers derived from seaweeds tis in the rat TMJ [1]. In this model, zymosan, a polysaccharide from have suggested lectins as research agents that can allow researchers yeast cell walls, elicits both hyper-nociception and inflammatory to comprehensively investigate events related to both nociception markers including an increase in vascular permeability, edema and and inflammation in animal models [22,23,26,32]. In this work, we cell migration associated with severe synovitis [39,3].

Table 2 Histopathological analysis of TMJ on zymosan-induced arthritis: the effect of CcL (10 mg/kg, i.v.) in the presence of ZnPP-XI.

Groups Cell influx in the synovial membrane Cell influx in the periarticular tissue Cell influx in the muscular tissue

Sham 0 (0–0) 0 (0–0) 0 (0–0) ⁎ ⁎ ⁎ Zy 4 (3–4) 4(3–4) 2(2–3) ⁎ ⁎ ⁎ ZnPP-IX 3 (1–3) 4(2–4) 2(1–3) ⁎⁎ ⁎⁎ ⁎⁎ CcL 0 (0–0) 1(1–1) 0(0–2) ⁎⁎ ZnPP-IX + CcL 1 (1–1) 1 (1–1) 2(0–2)

0 — uninjured; 1 — low injury; 2 — mild injury; 3 — moderate injury; and 4 — severe injury. ⁎ p b 0.05 versus sham group (Kruskal–Wallis, Dunn's). ⁎⁎ p b 0.05 versus Zy group (Kruskal–Wallis, Dunn's). 40 R.L. da Conceição Rivanor et al. / International Immunopharmacology 21 (2014) 34–43

Fig. 5. Histopathological analysis of zymosan-induced TMJ arthritis in rats. (A) Photomicrographs of TMJ the sham group (40×). (B, C and D) Synovial membrane, periarticular tissue and musculoskeletal tissue of the sham group without cell influx (400×). (E) Photomicrographs of the TMJ from the Zy group (2 mg/art., 40 μl) (40×). (F, G and H) Synovial membrane, periarticular tissue and musculoskeletal tissue of the zymosan group with intense cell influx (400×). (I) Photomicrographs of TMJ the ZnPP-IX group (40×). (J, L and M) Synovial membrane periarticular tissue and musculoskeletal tissue of the group treated ZnPP-IX with intense cell influx (400×). (N) Photomicrographs of the TMJ following exposure to ZnPP-IX + CcL (40×). (O, P and Q) The synovial membrane, periarticular tissue and musculoskeletal tissue of the group treated with ZnPP-IX + CcL with a lower inflammatory cell influx (400×). (R) Photomicrographs of the TMJ from the CcL group (40×). (S, T and U) Synovial membrane, periarticular tissue and musculoskeletal tissue of the group treated with CcL without cell influx (400×). C: condyle; AC: articular cartilage; AD: articular disk; SM: synovial membrane; PAT: periarticular tissue; and MET: musculoskeletal tissue. Hematoxylin and eosin staining. Black arrows indicate neutrophils.

Using zymosan-induced arthritis in the rat TMJ, some authors have expression in the synovial cells and neutrophils of the zymosan- recently suggested electroacupuncture as an important therapeutic induced TMJ inflammatory hyper-nociception. strategy for treating TMJ arthritis [24]. Although it was demonstrated Cytokines are produced by various cell types in response to a variety that some products derived from seaweeds, such as sulfated polysac- of stimuli and constitute a link between cellular injury or recognition of charides, inhibit the zymosan-induced inflammatory response in rats non-self and the development of signs and symptoms of inflammation [17,40], there is no evidence for an effect of lectins. [41,42]. Intra-plantar injection of both TNF-α and IL-1β induced In the present study, i.art. injection with 2 mg of zymosan decreased hyper-nociception in mice [43]. Furthermore, writhing responses to the mechanical nociceptive thresholds, which in turn was increased by zymosan and acetic acid were shown to be mediated by TNF-α and IL- CcL treatment. Regarding the inflammatory parameters, CcL administra- 1β [44]. Moreover, in mechanical (von Frey filaments) and thermal tion decreased leukocyte number, which was certified by a decrease in (hot plate) hyper-nociception induced by complete Freund's adjuvant, MPO activity in the synovial lavage. Furthermore, histopathological there was a sequential release of TNF-α and IL-1β [45,46]. analysis from rat TMJs treated with CcL showed a correlated decrease It was also revealed that TNF-α activates IL-1β, leading to the induc- in both leukocyte influx and MPO activity, likely because this treatment tion of connective tissue disorders in chronic inflammatory condi- produced a lower inflammatory cell influx in the synovial membrane. tions [47], and TNF-α was addressed as a biochemical marker of pain These data are in accordance with our previous results showing that and clinical outcome in temporomandibular joints [48,49]. The resolu- CcL has anti-nociceptive and anti-inflammatory functions in classical tion of the inflammatory response during arthritis might be attributed models of nociception in mice and acute inflammation in rats [26]. to the inhibition of pro-inflammatory cytokines, such IL-1β and TNF- Additionally, in the present study, histopathological analyses of immu- α. These two factors are assumed to be the most important cytokines nohistochemical stains showed a decrease in both IL-1β and TNF-α in TMJ disorders [5,47]. In fact, it seems that these cytokines induce R.L. da Conceição Rivanor et al. / International Immunopharmacology 21 (2014) 34–43 41

Fig. 6. Immunohistochemistry analysis for TNF-α of zymosan-induced TMJ arthritis in rats. (A) Negative control sections (absence of anti- TNF-α antibody) from arthritic rats (400×). (B) The synovial membrane from the sham group showed light expression of TNF-α. (C) The synovial membrane from the zymosan group (2 mg/art.) with an intense TNF-α reaction (400×). (D) CcL (10 mg/kg, i.v.) showing no TNF-α reaction in the synovial membrane after zymosan-induced arthritis (400×). Black arrows indicate synoviocytes. the production of metalloproteinases that irreversibly degrade the hyper-nociceptive mediators. This concept allows us to understand extracellular matrix components [5], including articular cartilage, as why the inhibition of cytokines causes analgesia [51]. well as causing bone destruction and cell proliferation [50]. Taken to- According to Gondim et al. [24], electroacupuncture inhibited gether, these experiments strongly suggest that the release of cytokines zymosan-induced TMJ inﬂammatory hyper-nociception, as well as neu- constitutes a link between the TMJ injuries and the release of primary trophil migration, vascular permeability, and TNF-α levels in synovial

Fig. 7. Immunohistochemistry analysis for IL-1β of zymosan-induced TMJ arthritis in rats. (A) Negative control sections (absence of the anti- IL-1β antibody) from arthritic rats (400×). (B) The synovial membrane of the sham group revealed light expression of IL-1β (400×). (C) The synovial membrane of the zymosan group (2 mg/art.) with an intense IL-1β reaction (400×). (D) CcL (10 mg/kg, i.v.) showing no expression of IL-1β in the synovial membrane after zymosan-induced arthritis (400×). Black arrows indicate synoviocytes. 42 R.L. da Conceição Rivanor et al. / International Immunopharmacology 21 (2014) 34–43

Fig. 8. Immunohistochemistry analysis for HO-1 in zymosan-induced TMJ arthritis in rats. (A) Negative control sections (absence of the anti- HO-1 antibody) from arthritic rats (400×). (B) Synovial membrane from the sham group revealed the basal constitutive levels of HO-1 (400×). (C) Synovial membrane of the zymosan group (2 mg/art.), intense expression of HO-1 (400×). (D) CcL (10 mg/kg, i.v.) showing no expression of HO-1 in the synovial membrane after zymosan-induced arthritis (400×). Black arrows indicate synoviocytes.

fluid. Additionally, Paiva et al. [17] showed that sulfated polysaccharides effects of CcL (10 mg/kg, i.v.) were observed in the presence of ZnPP from the brown Lobophora variegata decreased both cell influx and TNF- IX, suggesting that the HO-1 pathway is not required for CcL efficacy. α expression in knee joints. In conclusion, we demonstrated the anti-nociceptive and anti- Because it has been largely demonstrated that cytokines play an inflammatory efficacy of CcL in a model of zymosan-induced TMJ in- important role in inflammatory hyper-nociception, our results sug- flammatory hyper-nociception in rats. Additionally, our results strongly gested that CcL efficacy in zymosan-induced TMJ inflammatory suggest that CcL efficacy involves IL-1β and TNF-α inhibition. Given the hyper-nociception might be related, at least in part, to IL-1β and well-demonstrated anti-nociceptive and anti-inflammatory efficacy of TNF-α inhibition. CcL, the design of novel compounds is highly encouraged with the The endogenous opioid system is largely involved in the phenome- hope of defining new pharmacological targets for the treatment of non of the experience of pain [10], as well as in the action of analgesic inflammatory TMJ pain. opiate compounds [18]. In the TMJ, a role of the peripheral opioid receptors in the TMJ anti-nociception has been established [11,12]. In the present study, the antinociceptive response of the lectin C. cupressoides Acknowledgments was not reversed by i.art. injection of naloxone, a nonselective opioid receptor antagonist, in this experimental model, although it was capable This study was supported by the Conselho Nacional de fi of reversing the effects of morphine. This result suggests that endoge- Desenvolvimento Cientí co e tecnológico (CNPq), Ministério da Ciência nous opioids do not mediate the antinociceptive effect of CcL, which e Tecnologia (MCT), Ministério da Saúde (MS), Programa Rede Nordeste modulates TMJ nociception at the peripheral level, but CcL does not act de Biotecnologia (RENORBIO) and Coordenação de Aperfeiçoamento through peripheral opioid receptors. The possibility that the lectin de Pessoal de Nível Superior (CAPES). The authors thank Adalberto C. cupressoides can exert its antinociceptive effects through central Nascimento de Lima Júnior for technical assistance. BENEVIDES, N.M.B. involvement of opioid receptors has not been excluded. However, and BEZERRA, M.M. are senior investigators of CNPq/Brazil. Vanderlei et al. [26] demonstrated that the antinociceptive response of CcL, using classical models of nociception (formalin and hot- References plate tests) in mice, occurred predominantly through a peripheral mechanism. [1] Chaves HV, Ribeiro RA, de Souza AMB, e Silva AAR, Gomes AS, Vale ML, et al. HO-1 is a relevant target for discovering compounds exhibiting Experimental model of zymosan-induced arthritis in the rat tempromandibular joint: role of nitric oxide and neutrophils. J Biomed Biotechnol 2011;2011:1–11. antinociceptive [8] and anti-inflammatory [9] mechanisms. Agents [2] Gegout P, Gillet P, Chevrier D, Guingamp C, Terlain B, Netter P. Charactertzation of that elicit HO-1 also have great therapeutic importance against diseases zymosan-induced arthritis in the rat: effects on joint inflammation and cartilage – associated with inflammatory processes. HO-1 expression may reduce metabolism. Life Sci 1994;55:321 6. [3] Bezerra MM, Brain SD, Greenacre S, Jeronimo SM, de Melo B, Keeble J, et al. Reactive edema, leukocyte adhesion, migration and the production of cytokines nitrogen species scavenging, rather than nitric oxide inhibition, protects from [52]. Vicente et al. [50] reported the induction of HO-1 in a zymosan- articular cartilage damage in rat zymosan-induced arthritis. Br J Pharmacol induced air pouch inflammation model. In the present study, the pre- 2004;141:172–82. fi [4] Bezerra MM, Brain SD, Girão VCC, Greenacre S, Keeble J, Rocha FAC. Neutrophils- treatment of animals with ZnPP-IX, a speci c HO-1 inhibitor, decreased derived peroxynitrite contributes to acute hyperalgesia and cell influx in zymosan the nociceptive thresholds. However, analgesic and anti-inflammatory arthritis. Naunyn Schmiedebergs Arch Pharmacol 2007;374:265–73. R.L. da Conceição Rivanor et al. / International Immunopharmacology 21 (2014) 34–43 43

[5] Suzuki T, Segami N, Nishimura M, Nojima T. Co-expression of interleukin-1β and [30] Eisenberg E, Vos BP, Strassman AM. The peripheral antinociceptive effect of tumor necrosis factor α in synovial tissues and synovial fluids of temporomandibu- morphine in a rat model of facial pain. Neuroscience 1996;72:519–25. lar joint with internal derangement: comparison with histological grading of [31] Freitas A, Alves-Filho JC, Secco DD, Neto AF, Ferreira SH, Barja-Fidalgo C, et al. synovial inflammation. J Oral Pathol Med 2002;31:549–57. Heme oxygenase/carbon monoxide-biliverdin pathway down regulates neutrophil [6] Takahashi T, Kondoh T, Fukuda M, Yamazaki Y, Toyosaki T, Suzuki R. Proinflamma- rolling, adhesion and migration in acute inflammation. Br J Pharmacol 2006;149: tory cytokines detectable in synovial fluids from patients with temporomandibular 345–54. disorders. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1998;85:135–41. [32] Bitencourt FS, Figueiredo JG, Mota MRL, Bezerra CCR, Silvestre PP, Vale MR, et al. [7] Alcaraz MJ, Fernandez P, Guillen MI. Anti-inflammatory actions of the heme Antinociceptive and anti-inflammatory effects of a mucin-binding agglutinin isolat- oxygenase-1 pathway. Curr Pharm Des 2003;9:2541–51. ed from the red marine alga Hypnea cervicornis. Naunyn Schmiedebergs Arch [8] Grangeiro NMG, Aguiar JA, Chaves HV, Silva AAR, Lima V, Benevides NMB, et al. Pharmacol 2008;377:139–48. Heme oxygenase/carbon monoxide-biliverdin pathway may be involved in the [33] Lekkas C. Experimental degenerative temporomandibular joint disease. Int J Oral antinociceptive activity of etoricoxib, a selective COX-2 inhibitor. Pharmacol Rep Maxillofac Surg 1994;23:423–4. 2011;63:112–9. [34] Imai H, Sakamoto I, Yoda T, Yamashita Y. A model for internal derangement and [9] Vanderlei ESO, de Araújo IWF, Quinderé ALG, Fonte BP, Eloy YRG, Rodrigues JAG, osteoarthritis of the temporomandibular joint with experimental traction of the et al. The involvement of the HO-1 pathway in the anti-inflammatory action of a sul- mandibular ramus in rabbit. Oral Dis 2001;7:185–91. fated polysaccharide isolated from the red seaweed Gracilaria birdiae.Inflamm Res [35] Zamma T. Adjuvant-induced arthritis in the temporomandibular joint of rats. Infect 2011;60:1121–30. Immun 1983;39:1291–9. [10] Bodnar RJ, Klein GE. Endogenous opiates and behavior. Peptides 2005;26:2629–711. [36] Al-Mobireek AF, Darwazeh AMG, Hassanin MB. Experimental induction of rheuma- [11] Hayashi K, Sugisaiki M, Ota S, Tanabe H. mu-Opioid receptor mRNA expression and toid arthritis in temporomandibular joint of the guinea pig: a clinical and radio- immunohistochemical localization in the rat temporomandibular joint. Peptides graphic study. Dentomaxillofac Radiol 2000;29:286–90. 2002;23:889–93. [37] Goulart AC, Correia FAS, Sousa SCOM, Luz JGC. Study of the inflammatory process in- [12] Chicre-Alcântara TC, Torres-Chávez KE, Fischer L, Clemente-Napimoga JT, Melo V, duced by injection of carrageenan or formalin in the rat temporomandibular joint. Parada CA, et al. Local kappa opioid receptor activation decreases temporomandib- Braz Oral Res 2005;19:99–105. ular joint inflammation. Inflammation 2012;35:371–6. [38] Barretto SR, de Melo GC, dos Santos JC, de Oliveira MGB, Pereira-Filho RN, Alves AVF, [13] Kulinsky VI. Biochemical aspects of inflammation. Biochemistry 2007;72:595–607. et al. Evaluation of anti-nociceptive and anti-inflammatory activity of low-level laser [14] Fosslien E. Review: cardiovascular complications of non-steroidal anti-inflammatory therapy on temporomandibular joint inflammation in rodents. J Photochem drugs. Ann Clin Lab Sci 2005;35:347–85. Photobiol B 2013;129:135–42. [15] Iwalewa EO, McGaw LJ, Naidoo V, Eloff JN. Inflammation: the foundation of diseases [39] Rocha FAC, Aragão JR, de Oliveira RC, Vale MR, Ribeiro RA. Periarthritis promotes and disorders. A review of phytomedicines of South African origin used to treat pain articular incapacitation in zymosan-induced arthritis in rats. Inflamm Res and inflammatory conditions. Afr J Biotechnol 2007;6:2868–85. 1999;48:485–90. [16] Young E. The anti-inflammatory effects of heparin and related compounds. Thromb [40] Siqueira RCL, Silva MSJ, Alencar DB, Pires AF, Alencar NMN, Pereira MG, et al. In vivo Res 2008;122:743–52. anti-inflammatory effect of a sulfated polysaccharide isolated from the marine [17] Paiva AAO, Castro AJG, Nascimento MS, Will LSEP, Santos ND, Araújo RM, et al. An- brown algae Lobophora variegata. Pharm Biol 2011;49:167–74. tioxidant and anti-inflammatory effect of polysaccharides from Lobophora variegata [41] Cunha FQ, Poole S, Lorenzetti BB, Ferreira SH. The pivotal role of tumor necrosis on zymosan-induced arthritis in rats. Int Immunopharmacol 2011;11:1241–50. factor α in the development of inflammatory hyperalgesia Br. J Pharmacol [18] Rodrigues JAG, Vanderlei ESO, Silva LMCM, De Araújo IWF, De Queiroz INL, De Paula 1992;107:660–4. GA, et al. Antinociceptive and anti-inflammatory activities of a sulfated polysaccha- [42] Cunha FQ, Ferreira SH. Peripheral hyperalgesic cytokines. Adv Exp Med Biol ride isolated from the green seaweed Caulerpa cupressoides. Pharmacol Rep 2003;521:22–39. 2012;64:282–92. [43] Cunha TM, Verri Jr WA, Silva JS, Poole S, Cunha FQ, Ferreira SH. A cascade of cyto- [19] Peumans WJ, Van Damme EJN. Lectin as plant defense proteins. Plant Physiol kines mediates mechanical inflammatory hypernociception in mice. Proc Natl 1995;109:347–52. Acad Sci U S A 2005;102:1755–60. [20] Liener IE, Sharon N, Goldstein IJ. The lectins: properties, functions, and applications [44] Ribeiro RA, Vale ML, Thomazzi SM, Paschoalato AB, Poole S, Ferreira SH, et al. in biology and medicine. Orlando: Academic Press; 1986. Involvement of resident macrophages and mast cells in the writhing nociceptive [21] Rüdiger H, Gabius Hans-J. Plant lectins: occurrence, biochemistry, functions and response induced by zymosan and acetic acid in mice. Eur J Pharmacol 2000;387: applications. Glycoconj J 2001;18:589–613. 111–8. [22] Neves SA, Freitas ALP, Souza BWS, Rocha MLA, Correia MVO, Sampaio DA, et al. [45] Safieh-Garabedian B, Poole S, Allchorne A, Winter J, Woolf CJ. Contribution of Antinociceptive properties in mice of a lectin isolated from the marine alga Amansia interleukin-1 beta to the inflammation-induced increase in nerve growth factor multifida Lamouroux. Braz J Med Biol Res 2007;40:127–34. levels and inflammatory hyperalgesia. Br J Pharmacol 1995;115:1265–75. [23] Silva LMCM, Lima V, Holanda ML, Pinheiro PG, Rodrigues JAG, Lima MEP, et al. [46] Woolf CJ, Allchorne A, Safieh-Garabedian B, Poole S. Cytokines, nerve growth factor Antinociceptive and anti-inflammatory activities of lectin from marine red alga and inflammatory hyperalgesia: the contribution of tumour necrosis factor alpha. Br Pterocladiella capillacea. Biol Pharm Bull 2010;33:830–5. J Pharmacol 1997;121:417–24. [24] Gondim DV, Costa JL, Rocha SS, Brito GAC, Ribeiro RA, Vale ML. Antinociceptive and [47] Nordahl S, Alstergren P, Kopp S. Tumor necrosis factor-alpha in synovial fluid anti-inflammatory effects of electroacupuncture on experimental arthritis of the rat and plasma from patients with chronic connective tissue disease and its relation temporomandibular joint. Can J Physiol Pharmacol 2012;90:395–405. to temporomandibular joint pain. J Oral Maxillofac Surg 2000;58:525–30. [25] Benevides NMB, Holanda ML, Melo FR, Pereira MG, Monteiro ACO, Freitas ALP. [48] Shafer DM, Assael L, White LB, Rossomando EF. Tumor necrosis factor-alpha as a bio- Purification and partial characterization of the lectin from the marine green alga chemical marker of pain and outcome in temporomandibular joints with internal Caulerpa cupressoides (Vahl) C. Agardh. Bot Mar 2001;44:17–22. derangements. J Oral Maxillofac Surg 1994;52:786–91. [26] Vanderlei ESO, Patoilo KKNR, Lima NA, Lima APS, Rodrigues JAG, Silva LMCM, et al. [49] Emshoff R, Puffer P, Strobl H, Gassner R. Effect of temporomandibular joint Antinociceptive and anti-inflammatory activities of lectin from the marine green arthrocentesis on synovial fluid mediator level of tumor necrosis factor-alpha: alga Caulerpa cupressoides. Int Immunopharmacol 2010;10:1113–8. implications for treatment outcome. Int J Oral Maxillofac Surg 2000;29:176–82. [27] Abreu TM, Silva LMCM, Vanderlei ESO, De Melo CML, Pereira VRA, Benevides NMB. [50] Vicente AM, Guillen MI, Habib A, Alcaraz MJ. Beneficial effects of heme oxygenase-1 Cytokine production induced by marine algae lectins in BALB/c mice splenocytes. up-regulation in the development of experimental inflammation induced by zymo- Protein Pept Lett 2012;19:975–81. san. J Pharmacol Exp Ther 2003;307:1030–7. [28] Denadai-Souza A, Camargo LDL, Ribela MTCP, Keeble JE, Costa SKP, Muscara MN. [51] Ferreira SH, Cunha FQ, Lorenzetti BB, Michelin MA, Perretti M, Flower RJ, et al. Role Participation of peripheral tachykinin NK1 receptors in the carrageenan-induced of lipocortin-1 in the anti-hyperalgesic actions of dexamethasone. Br J Pharmacol inflammation of the rat temporomandibular joint. Eur J Pain 2009;13:812–9. 1997;121:883–8. [29] Bradley PP, Christensen RD, Rothstein G. Cellular and extracellular myeloperoxidase [52] Abraham NG, Kappas A. Pharmacological and clinical aspects of heme oxygenase. in pyogenic inflammation. Blood 1982;60:618–22. Pharmacol Rev 2008;60:79–127.