Effects of Two Serine Proteases from Bothrops Pirajai Snake Venom on the Complement System and the Inﬂammatory Response

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

International Immunopharmacology 15 (2013) 764–771

Contents lists available at SciVerse ScienceDirect

International Immunopharmacology

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

Effects of two serine proteases from Bothrops pirajai snake venom on the complement system and the inﬂammatory response

Danilo L. Menaldo a,⁎, Carolina P. Bernardes a, Juliana C. Pereira a, Denise S.C. Silveira a, Carla C.N. Mamede b,c, Leonilda Stanziola b,c, Fábio de Oliveira b,c, Luciana S. Pereira-Crott a, Lúcia H. Faccioli a, Suely V. Sampaio a,⁎

a Departamento de Análises Clínicas, Toxicológicas e Bromatológicas, Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidade de São Paulo (FCFRP-USP), Ribeirão Preto, SP, Brazil b Instituto de Ciências Biomédicas, Universidade Federal de Uberlândia (ICBIM-UFU), Uberlândia, MG, Brazil c Instituto Nacional de Ciência e Tecnologia em Nano-Biofarmacêutica (N-Biofar), Belo Horizonte, MG, Brazil

article info abstract

Article history: The present study aimed to evaluate the effects of two serine proteases from Bothrops pirajai snake venom, Received 12 October 2012 named BpirSP27 and BpirSP41, on the complement system and the inflammatory response. The effects of Received in revised form 2 February 2013 these enzymes on the human complement system were assessed by kinetic hemolytic assays, evaluating Accepted 27 February 2013 the hemolysis promoted by the classical/lectin (CP/LP) and alternative (AP) pathways after incubation of nor- Available online 15 March 2013 mal human serum with the serine proteases. The results suggested that these enzymes were able to induce modulation of CP/LP and AP at different levels: BpirSP41 showed higher inhibitory effects on the hemolytic Keywords: Snake venoms activity of CP/LP than BpirSP27, with inhibition values close to 40% and 20%, respectively, for the highest con- Bothrops pirajai centration assayed. Regarding AP, both enzymes showed percentages of inhibition of the hemolytic activity Serine proteases around 20% for the highest concentrations tested, indicating similar effects on this complement pathway. Thrombin-like enzymes The proinflammatory effects of B. pirajai serine proteases were evaluated regarding their ability to induce Complement system paw edema, variations in the pain threshold and leukocyte recruitment at the site of injection. Both showed fl In ammation mild effects on these inflammatory processes, leading to low levels of increase of paw volumes and decrease in pain thresholds in rats up to 6 h after injection, and inducing neutrophil recruitment without significant increases in the total number of leukocytes in the inflammatory exudates after 6 and 24 h of administration into mice peritoneal cavity. These results suggest that serine proteases must present a minor role in the inflammation caused by B. pirajai snake venom. © 2013 Elsevier B.V. Open access under the Elsevier OA license.

1. Introduction pain, heat, redness and edema, accompanied or not by loss of function of the affected tissue or organ [2]. Envenomation by Bothrops snakes is characterized by severe local The inflammatory response promoted by Bothrops snake venoms tissue damage, hemorrhage, myonecrosis and prominent inflamma- is strictly related to the progress of tissue damage, resulting in tion. The inflammatory response is an adaptive process in which the edema and pain, leukocyte infiltration and release of cytokines body responds to tissue injury or local infection in order to reestablish and other proinflammatory mediators [3–5]. Although most of the

homeostasis, allowing the healing and restoration of damaged sites available studies indicate a predominant role of phospholipases A2 [1].Inflammation is initiated and conducted by mediators of plasma and metalloproteases in the inflammatory effects promoted by or cellular origin, which promote the typical signs of this response: snake venoms [6,7], the role of serine proteases and other components of the venom on the inflammatory process should not be discarded [1]. The toxic effects promoted by snake venom serine proteases Abbreviations: AP, alternative pathway; CP/LP, classical/lectin pathways; LPS, lipopolysaccharides; NHS, normal human serum; PBS, phosphate buffered saline; SVSPs, (SVSPs) are mainly related to their actions on hemostasis, acting on snake venom serine proteases. several components of the coagulation cascade and the fibrinolytic ⁎ Corresponding authors at: Departamento de Análises Clínicas, Toxicológicas e and kallikrein–kinin systems, which leads to a hemostatic imbalance Bromatológicas, Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidade de on preys [8,9]. The vast majority of SVSPs are able to promote blood São Paulo, FCFRP-USP, Av. do Café, s/nº, CEP 14040-903, Ribeirão Preto-SP, Brazil. clotting and resemble, at least partially, thrombin, a multifunctional Tel.: +55 16 36024286; fax: +55 16 36024725. E-mail addresses: [email protected] (D.L. Menaldo), [email protected] protease with essential role in coagulation, thereby being referred (S.V. Sampaio). to as snake venom thrombin-like enzymes [10,11].

Considering the similar chemical nature of the components of the 2.3. Normal human serum (NHS) coagulation and complement systems, both of which contain series of serine proteases that form proteolytic cascades, one would expect that NHS used in the assays on the complement system was obtained SVSPs could also present significant effects on the complement system. from blood of healthy volunteers of both sexes, aged 20 to 40 years, Also, interactions between both systems have often been proposed collected in the absence of anticoagulants. The blood was allowed to [12,13], which could play an important role after snake envenomations clot at room temperature for 1 h, and the serum was separated from and for subsequent inflammatory reactions and complications. the clot by centrifugation at 500 ×g for 10 minutes at 4 °C. Then, serum The human complement system is composed of more than 30 was pooled, aliquoted and frozen at −80 °C until the experiments. All plasma and cell surface proteins that participate in many important procedures involving humans were approved by the Research Ethics biological processes, including the activation of the inflammatory pro- Committee of FCFRP-USP (CEP/FCFRP nº. 123/2011). cess in response to the generation of proteolytic fragments of its components, and the promotion of phagocytosis and lysis of pathogens and 2.4. Effects on the complement system altered cells by opsonization processes and formation of the membrane attack complex (MAC). Depending on the stimulus, the activation of the The effects of B. pirajai serine proteases on the human comple- complement system can occur by three main paths: the classical (CP), ment system were observed using in vitro assays to assess the hemo- alternative (AP) or lectin (LP) pathways [14].Somestudiesalsoindicate lytic activity promoted by NHS, source of complement components, the existence of additional complement activation paths in the plasma, preincubated for 1 h at 37 °C with different concentrations of the which are dependent of different coagulation factors such as thrombin, enzymes. Prior to these assays, the serine proteases were investigated FXa, FXIa and plasmin [13,15]. for their ability to induce direct hemolysis by incubation with rabbit Although several snake venoms present activity on the complement or sheep erythrocytes in the absence of NHS. system [16–18], studies on the action of isolated toxins, especially from Bothrops venoms, are still scarce. Most snake venom toxins described 2.4.1. Preparation of erythrocyte suspensions with anticomplementary action belong to the class of metalloproteases For CP/LP, blood from three sheep was collected by venipuncture [19–21], and in relation to SVSPs, one of the few published studies is of the jugular vein into plastic tubes containing the same volume of that of Yamamoto and coworkers [22], which showed that flavoxobin Alsever sterile solution (2.05% glucose, 0.8% sodium citrate and from Trimeresurus flavoviridis snakevenomwasabletospecifically 0.42% NaCl). Aliquots of blood were centrifuged at 500 ×g for 10 mi- cleave human complement protein C3, thus acting as a C3 convertase nutes and the pellet of red cells was then washed with triethanolamine −4 −4 enzyme. buffer pH 7.4 containing 1.4 × 10 MCaCl2,5×10 MMgSO4 and Bothrops pirajai serine proteases, called BpirSP27 and BpirSP41, 0.1% gelatin (TEA–Ca2+–Mg2+ buffer), preparing a 5% erythrocyte have recently been characterized by our research group, presenting suspension with approximately 1.2 × 109 cells/mL [24].Then,this significant differences in their molecular masses, N-glycosylation suspension was sensitized with rabbit anti-sheep erythrocyte antibod- levels and N-terminal sequences, besides different fibrin(ogen)olytic, ies (hemolysin), kept at 4 °C for 15 minutes and the absorbance at coagulant and platelet-aggregating activities [23]. Considering the few 700 nm was adjusted to 0.7–0.8 in a spectrophotometer (Beckman data available in the literature regarding the action of SVSPs on the com- Coulter, model DU-70). plement system and the inflammatory response, this study aimed at the For AP, blood from two rabbits was collected by puncture of the evaluation of the effects of B. pirajai serine proteases on the different central ear artery into plastic tubes containing the same volume of human complement pathways, as well as the proinflammatory poten- Alsever sterile solution. Then, blood was incubated with the same vol- tial of these enzymes by the examination of their ability to induce ume of a TEA–EDTA chelating solution for 15 minutes at 37 °C. After edema, pain and recruitment of immune cells. centrifugation, the red cells were washed with TEA–Mg2+ buffer, suspended in Alsever’s solution and stored at 4 °C. For the hemolytic assays, aliquots of this suspension were washed with triethanolamine 2. Materials and methods −3 −3 buffer pH 7.4 containing 8 × 10 M EGTA, 2 × 10 M MgSO4 and 0.1% gelatin (TEA–EGTA–Mg2+ buffer), preparing a 5% erythrocyte 2.1. Materials and toxins suspension and adjusting the absorbance at 700 nm to 0.7–0.8 in a spectrophotometer. The enzymes BpirSP27 and BpirSP41 were isolated from B. pirajai snake venom (purchased from the serpentarium Bioagents Bioactive 2.4.2. Hemolytic assays by kinetic method Proteins Ltd., Batatais-SP, Brazil) by consecutive chromatographic Assays for CP/LP and AP were performed by a kinetic method, using steps on Sephacryl S-200, Benzamidine Sepharose and C2/C18 columns different concentrations of the serine proteases (25–200 μg/mL) incu- (GE Healthcare), as previously described by Menaldo and coworkers bated with NHS at 37 °C for 1 h in 96-well microplates. After the incuba- [23]. Other materials and equipment used were described through- tion, residual complement activity was measured by adding sensitized out the methodology and reagents not specified were of analytical sheep erythrocytes (CP/LP) or rabbit erythrocytes (AP) to each well. grade. The reaction mixtures were homogenized and the absorbance at 700 nm was assayed for 15 minutes in a microplate reader (SpectraMax 2.2. Animals Plus, Molecular Devices). NHS diluted with TEA–Ca2+–Mg2+ buffer or TEA–EGTA–Mg2+ buffer in the absence of the enzymes were employed Male Wistar rats (150–200 g) and female Balb/C mice (20–22 g), as 100% lysis controls for CP/LP and AP, respectively. This methodology which were used for the evaluation of the proinflammatory effects, does not allow distinguishing between CP and LP, since both pathways were bred and provided by the Animal Facilities of University of São share C2, C4 and C3 convertase. Paulo, Campus of Ribeirão Preto-SP. Sheep or adult New Zealand The hemolytic activity was expressed by t½, which corresponds to rabbits were held in the Animal Facilities of University of São Paulo, the time (in seconds) required to occur lysis of 50% of the erythrocytes Campus of Ribeirão Preto-SP, for periodic collection of blood, which present in the reaction mixture, and is directly proportional to the per- was used in the experiments on the complement system. All experi- centage of inhibition of the complement system hemolytic activity [25]. ments involving animals or their blood were approved by the Ethics Values of t½ were calculated from the absorbance results, corresponding Committee on Animal Use of University of São Paulo, Campus of Ribeirão tothetimerequiredfortheabsorbanceofthereactionmixturetobe Preto-SP (CEUA nº. 08.1.192.53.7). reduced by half. The percentage of inhibition of the hemolytic activity 766 D.L. Menaldo et al. / International Immunopharmacology 15 (2013) 764–771 was calculated for each concentration of both serine proteases according endotoxin quantification was done based on a standard curve with to the formula: various concentrations of Escherichia coli endotoxin (0.1 to 1.0 EU/mL).

0 . 1 2.7. Statistical analysis 1 Bt control wells 100 C % inhibition ¼ 100−@ 2 . A t 1 sample wells The experiments were performed in triplicate or quadruplicate and 2 the results were expressed as mean values ± standard deviations (SD). The statistical significance of results was calculated by the program GraphPad Prism 5, using analysis of variance (ANOVA) followed by 2.5. Proinflammatory effects Tukey test and considering p values b0.05 as significant.

2.5.1. Edema-inducing activity 3. Results The edema induced by the enzymes was evaluated in male Wistar rats, with each animal group (n = 4) receiving different doses of the The dosage of bacterial endotoxins in the serine protease samples serine proteases (10, 25 and 50 μg) diluted in 50 μL of sterile phos- showed values below 0.1 EU/mL (results not shown), excluding the phate buffered saline (PBS) by intraplantar injection into the left possibility that the observed effects on the complement system and hind paw of animals. The right hind paw received only PBS as a nega- inflammatory processes could be related to the presence of LPS. In tive control. The volumes of both paws were measured using a addition, B. pirajai serine proteases were not able to promote direct plethysmometer (Ugo Basile, Comerio, Italy, model 7140) before and hemolysis of rabbit or sheep erythrocytes (results not shown), allowing at different time intervals (1, 2, 3, 4, 5 and 6 h) after injection of PBS the use of a kinetic hemolytic method for the determination of their or enzymes, according to Van Arman and colleagues [26].Resultswere effects on the human complement system. calculated as the difference between the values obtained for both paws After incubation of the enzymes with NHS for 1 h at 37 °C, the CP/LP and expressed as percentages of increase in paw volume in relation to hemolytic activity proved to be significantly reduced compared to the the initial measurements (before injection). control without the serine proteases, as shown by the values of t½ (Fig. 1). Both BpirSP27 (Fig. 1A) and BpirSP41 (Fig. 1B) induced a signif- 2.5.2. Hyperalgesia icant increase of t½(p b 0.05) at all the tested concentrations, indicat- The same animals used in the edema-inducing activity were assessed ing modulation of CP/LP by both enzymes. Regarding the inhibition of for the hyperalgesia promoted by B. pirajai serine proteases. The pain the AP hemolytic activity, it was shown that B. pirajai serine proteases threshold was measured with a pressure apparatus (Ugo-Basile, Comerio, promoted significant increases of t½ at most of the tested concentrations Italy) before and at different times (1, 2, 3, 4, 5 and 6 h) after injection of (Fig. 2). The analysis of the percentage of inhibition of the hemolytic PBS or enzymes, according to Randall and Selitto [27]. The weight activity promoted by the complement pathways clearly showed that (in grams) required to cause a nociceptive response (paw bending) BpirSP41 promoted greater modulation of CP/LP compared to BpirSP27, was determined as the nociceptive threshold. A cutoff value of 300 g with inhibition values close to 40% and 20%, respectively, at the concen- was used to avoid damage to the paws. Results were calculated as the tration of 200 μg/mL (Fig. 3A), while the modulation of AP was less pro- difference between the values obtained for both paws and expressed nounced, with both enzymes presenting maximal inhibition around 20% as percentages of decrease in nociceptive threshold in relation to the at the highest analyzed concentrations (Fig. 3B). initial measurements (before injection). The results related to inflammation suggested that B. pirajai serine proteases possess a minor role in the induction of edema (Fig. 4) and 2.5.3. Peritoneal inflammatory exudate analysis pain (Fig. 5) in rats. Regarding the edema-inducing activity, BpirSP27 Balb/C mice (4 animals/group) were injected intraperitoneally apparently showed a dose-dependent behavior, although the increase with 500 μL of PBS (negative control) or solutions of B. pirajai crude in the volume of paw injected with this enzyme did not exceed ~14% at venom (25 μg in 500 μL of PBS) and of the isolated serine proteases the highest tested dose (50 μg) (Fig. 4A). Additionally, all doses showed (25, 50 and 100 μg in 500 μL of PBS). After 6 and 24 h, the animals similar edema profiles, with a significant increase after 1 h of the en- were euthanized by instillation of CO2 and their peritoneal cavities zyme injection and another peak response after 4 h, especially at the were washed with 5 mL of cold PBS, draining the exudate aseptically dose of 50 μg(Fig. 4A). In the case of BpirSP41, the edema profiles pro- with a syringe. The total leukocyte count was performed in a Neubauer moted by the tested doses were more variable (Fig. 4B), with doses of chamber, analyzing each exudate sample diluted in Turk's solution 25 and 50 μg showing very similar responses, with a significant peak re- (1:20). The differential leukocyte count in polymorphonuclear or sponse (p b 0.05) after 4 h of the enzyme injection, while the dose of mononuclear cells was done by concentrating the different exudate 10 μg did not show pronounced peak responses. Similarly to BpirSP27, samples on microscope slides using Cytospin (Shandon Souther Prod- BpirSP41 was unable to induce edema superior to ~12% at the tested ucts Ltd., Manchester, UK), followed by staining with Romanowsky doses (Fig. 4B). Moreover, the decrease in the nociceptive threshold in- stain and examination under an optical microscope (Bioval, São Paulo, duced by both B. pirajai serine proteases was very modest, with few Brazil) at magnification of 400×. Three hundred cells were counted values significantly different from the initial thresholds (p b 0.05) and per slide and the results were expressed as percentages of each cell not exceeding 15% (Fig. 5). Furthermore, the nociceptive profiles in- type observed. duced by different doses of both enzymes varied greatly, with no clear dose-dependent effect observed. At the dose of 50 μg, BpirSP27 induced 2.6. Endotoxin quantification a pain peak response after 3 h of injection (Fig. 5A), while BpirSP41 induced a rather different profile, with a sharp peak response after 1 h of To ensure that the results obtained on the complement system and on injection and another one after 4 h (Fig. 5B). inflammation were not related to the presence of bacterial endotoxins in The evaluation of the peritoneal inflammatory exudates induced theserineproteasesamples,theamountofLPS(lipopolysaccharides) in mice after 6 or 24 h of administration of different doses of B. pirajai was tested using the limulus amebocyte lysate (LAL) QCL-1000® kit serine proteases (Fig. 6) showed that none of the enzymes promoted (Lonza, São Paulo, Brazil), according to the manufacturer's instructions. significant increases (p b 0.05) in the number of leukocytes present in Samples of the serine proteases were evaluated after preincubation at the peritoneal cavity at all tested doses when compared to the negative 37 °C with excess of the inhibitor benzamidine (20 mM) for 1 h to control (PBS). In contrast, B. pirajai crude venom (25 μg) displayed a ensure that their enzymatic activity would not influence the test. The significant increase (p b 0.05) in the total number of leukocytes both D.L. Menaldo et al. / International Immunopharmacology 15 (2013) 764–771 767

350 350 A B 300 300 * * 250 250 * * * * * 200 * * * 200

150 150 t1/2 (seconds) t1/2 (seconds)

100 100

50 50

0 0 C 25 50 100 150 200 C 25 50 100 150 200 BpirSP27 (µg/mL) BpirSP41 (µg/mL)

Fig. 1. Effects of the serine proteases BpirSP27 (A) and BpirSP41 (B) on the hemolytic activity of the classical/lectin pathways (CP/LP) of the human complement system. t½ corresponds to the time (in seconds) required for the occurrence of 50% hemolysis by the complement system. One hundred percent lysis control (C) represents the hemolytic activity of NHS in the absence of the enzymes. Results are expressed as mean values ± SD (n = 3). (*) Values signiﬁcantly different from the control (p b 0.05).

after 6 and 24 h of administration (Fig. 6). In addition, differentiation of 4. Discussion the leukocytes present in these exudates in polymorphonuclear cells (mainly neutrophils) and mononuclear cells (lymphocytes and mono- Studies have shown that compounds of animal venoms can modu- cytes) (Fig. 7) showed that, after 6 h of administration, both serine late the complement system by different mechanisms, acting either by proteases and B. pirajai crude venom promoted signiﬁcant increases direct cleavage of a speciﬁc component or interacting with components (p b 0.05) of neutrophils when compared to the negative control of to form complexes capable of activating and amplifying the comple- PBS (Fig. 7A). After 24 h, the number of neutrophils remained signif- ment cascade [28]. Regarding Bothrops venoms, Pidde-Queiroz and icantly higher for B. pirajai crude venom and for some doses of the colleagues [17] showed that several of them, including that of B. pirajai, isolated enzymes in comparison to that of the negative control (Fig. 7B), were able to activate CP of the complement system even in the absence but the values were less prominent than those observed after 6 h of of sensitizing antibody, and the activation was shown to be at least injection. partially associated to the cleavage of the plasma inhibitor C1-INH by

350 350 A B 300 300

* 250 * * 250 * * * * * * 200 200

150 150 t1/2 (seconds) t1/2 (seconds)

100 100

50 50

0 0 C 25 50 100 150 200 C 25 50100 150 200 BpirSP27 (µg/mL) BpirSP41 (µg/mL)

Fig. 2. Effects of the serine proteases BpirSP27 (A) and BpirSP41 (B) on the hemolytic activity of the alternative pathway (AP) of the human complement system. t½ corresponds to the time (in seconds) required for the occurrence of 50% hemolysis by the complement system. One hundred percent lysis control (C) represents the hemolytic activity of NHS in the absence of the enzymes. Results are expressed as mean values ± SD (n = 3). (*) Values signiﬁcantly different from the control (p b 0.05). 768 D.L. Menaldo et al. / International Immunopharmacology 15 (2013) 764–771

55 55 BpirSP27 BpirSP27 50 BpirSP41 A 50 BpirSP41 B

45 45

40 40

35 35

30 30

25 25

20 20

15 15

10 10 Inhibition of the AP hemolytic activity (%)

Inhibition of the CP/LP hemolytic activity (%) 5 5

0 0 0 25 50 75 100 125 150 175 200 225 0 25 50 75 100 125 150 175 200 225 Enzyme (µg/mL) Enzyme (µg/mL)

Fig. 3. Inhibition percentages of the hemolytic activity of CP/LP (A) and AP (B) of the human complement system promoted by B. pirajai serine proteases. The percentages of inhibition were calculated considering the t½ values obtained for the negative control and samples, as described in Materials and methods. Results are expressed as mean values ± SD (n = 3). venom proteases. The cleavage of this inhibitor and some isolated venom decreased the hemolytic activity related to CP and AP due to components (C3 and C4) was evaluated using specific inhibitors of the proteolytic cleavage of the α chain of C3 and C4 components [30]. metalloproteases and serine proteases (1,10-phenanthroline and The metalloprotease atrase B isolated from Naja atra venom inhibited phenylmethylsulfonyl fluoride — PMSF, respectively), and the results the activation of AP and CP of the complement in a dose-time- indicated that these should be the main venom constituents responsible dependent manner, cleaving the components C6, C7, C8 and fac- for the effects observed on the human complement system [17]. tor B [21]. Regarding the class of serine proteases, flavoxobin Some isolated snake venom toxins were already described with from T. flavoviridis venom was able to activate AP, acting as a C3 action on the complement system. The cobra venom factor (CVF) from convertase that independently and specifically cleaves human C3 Naja naja kaouthia venom, for example, directly promoted cleavage of at the Arg726–Ser727 site to form C3a and C3b, initiating the complement proteins, such as C2, C3 and C4, and stabilization of the complement cascade [22]. C3 convertase of AP, generating amplification of the activation cascade In the case of B. pirajai serine proteases, the increase of t½ observed and formation of the anaphylatoxins C3a and C5a, as well as other acti- after incubation of NHS with the enzymes indicates their action on the vation fragments as C3b and C4b [29].AproteasefromVipera lebetina complement, although it is not possible, only by the methodology used

20 20 BpirSP27 (10 µg) BpirSP41 (10 µg) BpirSP27 (25 µg) A BpirSP41 (25 µg) B BpirSP27 (50 µg) BpirSP41 (50 µg)

15 15

* * * * * 10 10 Increase in paw volume (%) Increase in paw volume (%) 5 5

0 0 0 102 3 4 5 6 1 2 3 4 5 6 Time after treatment (h) Time after treatment (h)

Fig. 4. Edema induced by the serine proteases BpirSP27 (A) and BpirSP41 (B). The induction of edema was assessed in rats by the increase in paw volume at different times (1–6h) after injection of PBS (negative control, right hind paw) or enzymes (10, 25 and 50 μg, left hind paw). Results are expressed as mean values ± SD (n = 4), calculated by the difference between the volumes obtained for both paws and expressed as percentage of increase in paw volumes. (*) Values signiﬁcantly different from the initial measurements (p b 0.05). D.L. Menaldo et al. / International Immunopharmacology 15 (2013) 764–771 769

0 0

-5 -5

-10 -10 * * *

-15 -15 * Decrease in pain threshold (%) Decrease in pain threshold (%)

BpirSP27 (10 µg) BpirSP41 (10 µg) BpirSP27 (25 µg) BpirSP41 (25 µg) BpirSP27 (50 µg) A BpirSP41 (50 µg) B -20 -20 0 102 3 4 5 6 1 2 3 4 5 6 Time after treatment (h) Time after treatment (h)

Fig. 5. Hyperalgesia induced by the serine proteases BpirSP27 (A) and BpirSP41 (B). The pain threshold was assessed in rats by the paw bending in response to pressure applied at different times (1–6 h) after injection of PBS (negative control, right hind paw) or enzymes (10, 25 and 50 μg, left hind paw). Results are expressed as mean values ± SD (n = 4), calculated by the difference between the thresholds obtained for both paws and expressed as percentage of decrease in pain thresholds. (*) Values significantly different from the initial measurements (p b 0.05). in this study, to conclude if the effects are due to activation and/or inac- B. pirajai venom, similarly to that reported for Bothrops jararaca venom tivation of components or inhibition of activation pathways and/or of [31]. In fact, the thrombin-like enzyme TLBm isolated from the venom regulatory proteins of the system. Nevertheless, considering that of Bothrops marajoensis showed similar edema-inducing results to BpirSP27 and BpirSP41 are thrombin-like enzymes, it is possible that BpirSP27 and BpirSP41, promoting no significant increases in the paw those SVSPs act directly on different complement components as previ- volume of mice injected with doses up to 20 μg and not inducing the in- ously stated for thrombin, which dose- and time-dependently generated flammatory infiltrate characteristic of Bothrops envenomations [32]. C5a in the absence of C3 and C3a in the presence of C3 [15]. Several studies show the action of Bothrops venoms in the induc- B. pirajai serine proteases presented mild effects on inflammatory tion of edema and pain in rats and mice, with very low doses inducing processes such as the induction of edema and pain, suggesting that these exacerbated effects, often accompanied by hemorrhage [33–38]. This enzymes should possess a minor role in the inflammation promoted by suggests the participation of metalloproteases in these processes, which

12 12 11 * A 11 B 10 10 *

9 9

8 8 cells / cavity) cells / cavity) 6 6 7 7

6 6

5 5

4 4

3 3

2 2 Total leukocyte count (x 10 Total leukocyte count (x 10

1 1

0 0 NCSV 25 50 100 25 50 100 NC SV 25 50 100 25 50 100 BpirSP27 BpirSP41 BpirSP27 BpirSP41 (µg) (µg) (µg) (µg)

Fig. 6. Total leukocyte count of the peritoneal inﬂammatory exudates after 6 h (A) and 24 h (B) of administration of B. pirajai serine proteases. The total number of leukocytes present in the peritoneal cavity of mice after injection of PBS (NC, negative control), B. pirajai crude venom (SV, 25 μgin500 μL of PBS) or enzymes (BpirSP27/BpirSP41, 25, 50 and 100 μgin500 μLof PBS) was counted in a Neubauer chamber. Results are expressed as mean values ± SD (n =4).(*)Valuessigniﬁcantly different from NC (p b 0.05). 770 D.L. Menaldo et al. / International Immunopharmacology 15 (2013) 764–771

120 120 Neutrophils Neutrophils 110 Mononuclear cells A 110 Mononuclear cells B 100 100

90 90

80 80 * * 70 70 * * * 60 * 60 50 * * 50 * 40 * 40 * * 30 30 Differential leukocyte count (%) Differential leukocyte count (%)

20 20

10 10

0 0 NCSV 25 50 100 25 50 100 NCSV 25 50 100 25 50 100 BpirSP27 BpirSP41 BpirSP27 BpirSP41 (µg) (µg) (µg) (µg)

Fig. 7. Differential leukocyte count of the peritoneal inflammatory exudates after 6 h (A) and 24 h (B) of administration of B. pirajai serine proteases. The leukocytes present in the peritoneal cavity of mice after injection of PBS (NC, negative control), B. pirajai crude venom (SV, 25 μg in 500 μL of PBS) or enzymes (BpirSP27/BpirSP41, 25, 50 and 100 μgin 500 μL of PBS) were differentiated in neutrophils and mononuclear cells by analysis under an optical microscope at magnification of 400×. Results are expressed as mean percentages of each cell type ± SD (n = 4). (*) Values for both cell types significantly different from NC (p b 0.05). promote these effects due to the hydrolysis of vessel proteins, inducing of the anaphylatoxins C3a, C4a and C5a due to the action of these SVSPs an inflammatory process accompanied by neutrophil migration and re- on the respective components of the complement system, though other lease of inflammatory mediators, with increased vascular diameter and studies are necessary to confirm or refute this hypothesis. As a matter of plasma extravasation [1]. fact, C5a exhibits a broad spectrum of inflammatory activities, including The edema and pain induced by Bothrops venoms are usually re- neutrophil chemoattraction and hypernociception by a mechanism lated to the formation of arachidonic acid metabolites such as prosta- independent of cytokine release, but dependent on the presence of neu- glandins and leukotrienes, formed by the action of cyclooxygenases trophils [46]. and lipoxygenases, respectively. Other mediators commonly associated Comparing the proinflammatory activities of BpirSP27 and BpirSP41 with these effects are histamine, serotonin, bradykinin, nitric oxide and with those of a P-I class metalloprotease (BpirMP) (unpublished results) platelet activating factor (PAF) [1,38–40].AsB. pirajai serine proteases and a Lys49 phospholipase A2 (piratoxin-I) [47] also from B. pirajai did not present substantial edema-inducing and nociceptive effects, it venom, it is possible to suggest that the induction of pain and edema is was not possible to assess which are the mediators involved in the not directly related to these serine proteases. Some studies showed that low levels of edema and pain induced by these enzymes. this class of enzymes could be indirectly involved in the local inflamma- Corroborating the proinflammatory effects described so far, the tion caused by snake venoms by promoting blood pressure disturbances assessment of the inflammatory exudates induced by B. pirajai serine [31] or releasing endogenous metalloproteases [48]. Also, considering the proteases showed that these enzymes promoted no significant increases correlation between coagulation and complement components, and their (p b 0.05) in the total number of leukocytes present in the peritoneal subsequent effects on inflammation [12,13],itispossiblethattheeffects cavity of mice after 6 and 24 h of administration, differently from that reported for B. pirajai serine proteases on the inflammatory processes in observed for B. pirajai crude venom. Although the enzymes BpirSP27 vivo are somehow related to their actions on the hemostatic system. and BpirSP41 did not induce a significant quantitative increase, a prominent infiltration of neutrophils was observed in the exudates after 6 h of 5. Conclusions administration, indicating that these enzymes should contribute, even if to a lesser extent, to the acute inflammatory response promoted by The results regarding the complement system suggested that B. pirajai B. pirajai venom. After 24 h of administration, an increased neutrophil serine proteases were able to induce modulation of CP/LP and AP at count remained at some doses, but the majority of leukocytes present in different levels, with BpirSP41 causing greater inhibition of the CP/LP the exudates were mononuclear cells. hemolytic activity compared to BpirSP27, and both presenting similar Leukocytes are key elements in the inflammatory responses due to inhibitory effects on the hemolytic activity of AP. Further studies using their phagocytic and secretory activities, migrating to the affected isolated complement components such as C2, C3, C4 and C5, as well as area in response to various inflammatory stimuli. In the acute inflam- the investigation of the generation of cleavage fragments, as C3a and matory response, neutrophils rapidly migrate to the injured site, being C5a, should be performed in order to elucidate the mechanisms of action the main responsible for the initial phagocytosis of infectious agents of these enzymes on the human complement system. and the release of cytokines [1,41]. Several Bothrops venoms have In relation to the proinflammatory effects of the serine proteases, been described as able to induce different levels of leukocyte migration both presented mild effects on processes such as the induction of up to 72 h after their administration [31,42–44], and though some stud- edema and pain, promoting a pronounced neutrophil recruitment with- ies attributed this ability mainly to metalloproteases [31,45], our results out inducing significant increases in the total number of leukocytes in the indicated that SVSPs could also be involved in this process. peritoneal inflammatory exudate after 6 and 24 h of their administra- It is possible that the edema, pain and the increase of neutrophils tion. These results suggest that these enzymes should present a minor after injection of B. pirajai serine proteases is related to the formation role in the inflammation caused by the venom of B. pirajai. D.L. Menaldo et al. / International Immunopharmacology 15 (2013) 764–771 771

Considering that the serine proteases BpirSP27 and BpirSP41 [23] Menaldo DL, Bernardes CP, Santos-Filho NA, Moura LA, Fuly AL, Arantes EC, et al. fi Biochemical characterization and comparative analysis of two distinct serine pro- presented signi cant differences in their amino acid sequences and teases from Bothrops pirajai snake venom. Biochimie 2012;94:2545–58. functional activities [23], the fact that their effects on the complement [24] Almeida JO, Fife Jr EH. Quantitatively standardized complement-fixation methods system and inflammation were mostly similar and mild indicates that for critical evaluation of antigens prepared from Trypanosoma cruzi. Pan Am Health Org 1976;319:86. those immunological responses do not seem to be key targets of this [25] Crott LS, Lucisano-Valim YM, Silva CL, Barbosa JE. Interactions of F1 fractions from class of snake venom enzymes. different strains of Paracoccidioides brasiliensis with human complement and with human neutrophils. Mycopathologia 1997;140:19–27. [26] Van Arman CG, Begany AJ, Miller JM, Pless HH. Some details of the inflammation Acknowledgments caused by yeast and carragenin. J Pharmacol Exp Ther 1965;150:328–33. [27] Randall LO, Selitto JJ. A method for measurement of analgesic activity on inflamed – fi tissue. Arch Int Pharmacodyn Ther 1957;111:409 19. The authors would like to thank the nancial support provided by [28] Dias da Silva W, Tambourgi DV, Campos ACMR, Magnoli F, Petricevich VL, Kipnis TL. CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior), Complement activation by animal venoms. J Toxicol Toxin Rev 1995;14:375–400. FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo) and [29] Müller-Eberhard HJ, Fjellström KE. Isolation of the anticomplementary protein – fi from cobra venom and its mode of action on C3. J Immunol 1971;107:1666 72. CNPq (Conselho Nacional de Desenvolvimento Cientí co e Tecnológico), [30] Gasmi A, Louzir H, Karoui H, Ayeb M, Dellagi K. Purification from Vipera lebetina and also the technical assistance of Tânia M. C. Ogasawara (FCFRP-USP). (desert adder) venom of a protein that depletes human complement. Nat Toxins 1994;2:44–8. [31] Zychar BC, Dale CS, Demarchi DS, Gonçalves LR. Contribution of metalloproteases, fl References serine proteases and phospholipases A2 to the in ammatory reaction induced by Bothrops jararaca crude venom in mice. Toxicon 2010;55:227–34. [1] Teixeira C, Cury Y, Moreira V, Picolo G, Chaves F. Inflammation induced by [32] Vilca-Quispe A, Ponce-Soto LA, Winck FV, Marangoni S. Isolation and characteriza- Bothrops asper venom. Toxicon 2009;154:988–97. tion of a new serine protease with thrombin-like activity (TLBm) from the venom [2] Rocha e Silva M. A brief survey of the history of inflammation. Agents Actions of the snake Bothrops marajoensis. Toxicon 2010;55:745–53. 1978;8:45–9. [33] Trebien HA, Calixto JB. Pharmacological evaluation of rat paw ooedema induced [3] Gutiérrez JM, Rucavado A. Snake venom metalloproteinases: their role in the by Bothrops jararaca venom. Agents Actions 1989;26:292–300. pathogenesis of local tissue damage. Biochimie 2000;82:841–50. [34] Perales J, Amorim CZ, Rocha SLG, Domont GB, Moussatche H. Neutralization of the [4] Moura-da-Silva AM, Butera D, Tanjoni I. Importance of snake venom metalloproteinases oedematogenic activity of Bothrops jararaca venom on the mouse paw by an in cell biology: effects on platelets, inflammatory and endothelial cells. Curr Pharm Des antibothropic fraction isolated from opossum (Didelphis marsupialis)serum.Agents 2007;13:2893–905. Actions 1992;37:250–9. [5] Lopes DS, Baldo C, Oliveira CF, de Alcântara TM, Oliveira JD, Gourlart LR, et al. Character- [35] Lomonte B, Tarkowski A, Hanson LA. Host response to Bothrops asper snake ization of inflammatory reaction induced by neuwiedase, a P-I metalloproteinase venom. Analysis of edema formation, inflammatory cells, and cytokine release isolated from Bothrops neuwiedi venom. Toxicon 2009;54:42–9. in a mouse model. Inflammation 1993;17:93–105. [6] Teixeira CF, Landucci EC, Antunes E, Chacur M, Cury Y. Inflammatory effects of [36] Chaves F, Barbosa M, Gutiérrez JM. Pharmacological study of oedema induced by – snake venom myotoxic phospholipases A2. Toxicon 2003;42:947–62. venom of the snake Bothrops asper (Terciopelo) in mice. Toxicon 1995;33:31 9. [7] Teixeira CF, Fernandes CM, Zuliani JP, Zamuner SF. Inflammatory effects of snake [37] Lôbo de Araújo A, Souza AO, Cruz-Hofling MA, Flores CA, Bon C. Bothrops venom metalloproteinases. Mem Inst Oswaldo Cruz 2005;100:181–4. lanceolatus (fer de lance) venom induces oedema formation and increases vascu- [8] Braud S, Bon C, Wisner A. Snake venom acting on hemostasis. Biochimie 2000;82: lar permeability in the mouse hind paw. Toxicon 2000;38:209–21. 851–9. [38] Barbosa AM, do Amaral RO, Teixeira CF, Hyslop S, Cogo JC. Pharmacological char- [9] Serrano SM, Maroun RC. Snake venom serine proteinases: sequence homology vs. acterization of mouse hind paw oedema induced by Bothrops insularis (jararaca substrate specificity, a paradox to be solved. Toxicon 2005;45:1115–32. ilhoa) snake venom. Toxicon 2003;42:515–23. [10] Pirkle H. Thrombin-like enzymes from snake venoms. Thromb Haemost 1998;79: [39] Teixeira CFP, Cury Y, Oga S, Jancar S. Hyperalgesia induced by Bothrops jararaca 675–83. venom in rats: role of eicosanoids and platelet activating factor (PAF). Toxicon [11] Castro HC, Zingali RB, Albuquerque MG, Pujol-Luz M, Rodrigues CR. Snake venom 1994;32:419–26. thrombin-like enzymes: from reptilase to now. Cell Mol Life Sci 2004;61:843–56. [40] Chacur M, Picolo G, Gutiérrez JM, Teixeira CFP, Cury Y. Pharmacological modula- [12] Markiewski MM, Nilsson B, Ekdahl KN, Mollnes TE, Lambris JD. Complement and tion of hyperalgesia induced by Bothrops asper (terciopelo) snake venom. Toxicon coagulation: strangers or partners in crime? Trends Immunol 2007;28:184–92. 2001;39:1173–81. [13] Amara U, Rittirsch D, Flierl M, Bruckner U, Klos A, Gebhard F, et al. Interaction be- [41] Ulbrich H, Eriksson EE, Lindbom L. Leukocyte and endothelial cell adhesion mole- tween the coagulation and complement system. Adv Exp Med Biol 2008;632:71–9. cules as targets for therapeutic interventions in inflammatory disease. Trends [14] Wagner E, Frank MM. Therapeutic potential of complement modulation. Nat Rev Pharmacol Sci 2003;24:640–7. Drug Discov 2010;9:43–56. [42] Gutiérrez JM, Chaves F, Cerdas L. Inflammatory infiltrate in skeletal muscle [15] Huber-Lang M, Sarma JV, Zetoune FS, Rittirsch D, Neff TA, McGuire SR, et al. Gen- injected with Bothrops asper venom. Rev Biol Trop 1986;34:209–14. eration of C5a in the absence of C3: a new complement activation pathway. Nat [43] Flores CA, Zappellini A, Prado-Franceschi J. Lipoxygenase-derived mediators may Med 2006;12:682–7. be involved in in vivo neutrophil migration induced by Bothrops erytromelas and [16] Eggertsen G, Fohlman J, Sjöquist J. In vitro studies on complement inactivation by Bothrops alternatus venoms. Toxicon 1993;31:1551–9. snake venoms. Toxicon 1980;18:87–96. [44] Moreira V, Dos-Santos MC, Nascimento NG, da Silva HB, Fernandes CM, D'Império [17] Pidde-Queiroz G, Furtado MF, Filgueiras CF, Pessoa LA, Spadafora-Ferreira M, van Lima MR, et al. Local inflammatory events induced by Bothrops atrox snake venom den Berg CW, et al. Human complement activation and anaphylatoxins generation in- and the release of distinct classes of inflammatory mediators. Toxicon 2012;60: duced by snake venom toxins from Bothrops genus. Mol Immunol 2010;47:2537–44. 12–20. [18] Tanaka GD, Queiroz GP, Furtado MD, van den Berg CW, Tambourgi DV. Micrurus [45] Fernandes CM, Zamuner SR, Zuliani JP, Rucavado A, Gutiérrez JM, Teixeira CF. snake venoms activate human complement system and generate anaphylatoxins. Inflammatory effects of BaP1, a metalloproteinase isolated from Bothrops asper BMC Immunol 2012;13:4–11. snake venom: leukocyte recruitment and release of cytokines. Toxicon 2006;47: [19] Farsky SH, Gonçalves LR, Gutiérrez JM, Correa AP, Rucavado A, Gasque P, et al. 549–59. Bothrops asper snake venom and its metalloproteinase BaP-1 activate the comple- [46] Ting E, Guerrero AT, Cunha TM, Verri Jr WA, Taylor SM, Woodruff TM, et al. Role of ment system. Role in leucocyte recruitment. Mediators Inflamm 2000;9:213–21. complement C5a in mechanical inflammatory hypernociception: potential use of [20] Shoibonov BB, Osipov AV, Kryukova EV, Zinchenko AA, Lakhtin VM, Tsetlin VI, C5a receptor antagonists to control inflammatory pain. Br J Pharmacol 2008;153: et al. Oxiagin from the Naja oxiana cobra venom is the first reprolysin inhibiting 1043–53. the classical pathway of complement. Mol Immunol 2005;42:1141–53. [47] Landucci EC, de Castro RC, Toyama M, Giglio JR, Marangoni S, de Nucci G, et al. Inflam- [21] Sun QY, Bao J. Purification, cloning and characterization of a metalloproteinase matory oedema induced by the Lys-49 phospholipase A2 homologue piratoxin-I in the from Naja atra venom. Toxicon 2010;56:1459–69. rat and rabbit. Effect of polyanions and p-bromophenacyl bromide. Biochem Pharmacol [22] Yamamoto C, Tsuru D, Oda-Ueda N, Ohno M, Hattori S, Kim ST. Flavoxobin, a serine 2000;59:1289–94. protease from Trimeresurus flavoviridis (habu snake) venom, independently cleaves [48] Saravia-Otten P, Frisan T, Thelestam M, Gutiérrez JM. Membrane independent ac- Arg726–Ser727 of human C3 and acts as a novel, heterologous C3 convertase. Immu- tivation of fibroblast proMMP-2 by snake venom: novel roles for venom protein- nology 2002;107:111–7. ases. Toxicon 2004;44:749–64.