ARTICLE IN PRESS

0041-0101/$ - see

doi:10.1016/j.tox

�Correspondifax: +5511 372

E-mail addre

(M. Lopes-Ferr

Toxicon 49 (2007) 909–919

www.elsevier.com/locate/toxicon

Analysis of the inflammatory reaction induced by the catfish(Cathorops spixii) venoms

Marcos Emerson Pinheiro Junqueiraa,b, Lidiane Zito Grundb, Noemia M. Oriic,Tania Cristina Saraivab, Carlos Alberto de Magalhaes Lopesa,

Carla Limab, Monica Lopes-Ferreirab,�

aSchool of Medicine, Unesp, Botucatu, Sao Paulo, BrazilbSpecial Laboratory of Applied Toxinology and Immunopathology, Butantan Institute, Sao Paulo, Brazil

cTropical Medicine Laboratory, University of Sao Paulo, Sao Paulo, Brazil

Received 14 August 2006; received in revised form 3 January 2007; accepted 11 January 2007

Available online 23 January 2007

Abstract

Cathorops spixii is one of the most abundant venomous fish of the southeastern coast of the State of Sao Paulo, and

consequently causes a great part of the accidents seen there. The accidents affect mainly fishermen, swimmers and tourists

and are characterized by punctiform or wide wounds, erythema, edema, pain, sudoresis, indisposition, fever, nausea,

vomiting and secondary infection. The objective of this work was to characterize the inflammatory response induced in

mice by both venoms (mucus and sting) of the catfish C. spixii. Our results demonstrated that both venoms induced a great

number of rolling and adherent leukocytes in the post-capillary venules of cremaster muscle of mice, and an increase in the

vascular permeability in peritoneal cavity. Mucus induced the recruitment of neutrophils immediately after injection

followed later by macrophage infiltration. In contrast, the cellular infiltration elicited by sting venom was rapidly resolved.

The peritonitis reaction provoked by venoms was characterized by cytokine (IL-6), chemokines (MCP-1 and KC) or lipid

mediator (LTB4) production in the peritoneal cavity. The macrophages from 7-day mucus venom-induced exudates upon

in vitro mucus venom stimulation, expressed CD11c � MHC class II and release bioactive IL-12p70. On the other hand,

sting venom-elicited peritoneal macrophages lost the ability to differentiate into dendritic cells, following re-stimulation

in vitro with sting venom, they do not express CD11c, nor do they exhibit sufficient levels of MHC class II. In conclusion,

both types of venoms (mucus or sting) promote inflammatory reaction with different profiles, and the inflammatory

reaction induced by the first was characterized by antigen persistence in peritoneal cavity that allowed the activation of

phagocytic cells with capacity of antigenic presentation.

r 2007 Elsevier Ltd. All rights reserved.

Keywords: Catfish venoms; Cathorops spixii; Innate immunity; Macrophage activation; Antigen presentation

front matter r 2007 Elsevier Ltd. All rights reserved

icon.2007.01.004

ng author. Tel.: +5511 3726 7222;

6 1505.

ss: lopesferreira@butantan.gov.br

eira).

1. Introduction

Aquatic animals show the attack and defensebehavior that include the production of substancesexpressing repellent, paralytic or other biologicalactions. In most instances, these substances show a

.

ARTICLE IN PRESSM.E.P. Junqueira et al. / Toxicon 49 (2007) 909–919910

great variety of toxins that are responsible forsymptoms observed following envenomings and forthe complex ecological relationships among organ-isms. The production of toxins by aquatic animals isan important strategy that guarantees its survival ina highly competitive ecosystem. In this way, theseanimals to defend themselves or their territories,produce a significant number of metabolites, whichin combination, result in a great variety of chemicalstructures and complex molecules, as alkaloids,steroids, peptides and proteins with chemical andpharmacological properties, different from those invenoms of terrestrial animals (Russell, 1971).

In this context, Brazil with an extensive coast(approximately 7400 km) shows a wide diversity offauna comprising animals of temperate and tropicalwaters. Many of these have been consideredpotentially dangerous and frequently associatedwith occurrence of accidents in humans because ofthe great affluence of swimmers to the beaches andto the increase of activities related with commercialand sporting fishing, especially the autonomousdivers and underwater fishing (Haddad Jr., 2000).

Among the venomous fish recognized in Brazil,the catfish possess medical importance in conse-quence of accidents provoked in humans, most ofthem resulting in incapability (Haddad and Mar-tins, 2006). The Ariidae family (sea catfish) consistsof 20 genera and 153 species. The most representa-tive genera in the South Atlantic are Arius,Cathorops, Hexanematichthyes, Bagre and Genidens

(Froese and Pauly, 2005). In general, they seek theoutlet of the rivers and lagoons at the time ofspawning and show long and robust stings withsawing edges in the front, each one placed, withinthe dorsal and lateral fins. This venomous apparatusis constituted of quite rigid bone structure wrappedup for a slight tegument membrane carrying threedifferent venoms (a) venom found in the glandularepithelium which covers the sting; (b) venom foundin the glands located in the base of the lateral sting;and (c) venom found in the body mucus producedby cells denominated cell-club (Figueiredo andMenezes, 1978).

Cathorops spixii, one of the most abundantspecies of catfish in the southeastern coast of Brazil,has been incriminated as the main cause of humanaccidents characterized by punctiform or widewounds, erythema, edema, pain, sudoresis, indis-position, fever, nausea and secondary infection(Haddad and Martins, 2006). Taking in viewof the frequency of accidents provoked by catfish

C. spixii in Brazil, the objective of this work was tocharacterize the inflammatory response in miceinduced by two types of venoms: (a) venom foundin the glandular epithelium which covers the sting(sting venom) and (b) venom found in the bodymucus (mucus venom).

2. Material and methods

2.1. Animals and venom

Swiss male, weighing 18–22 g were housed in theanimal care facility at the Butantan Institute andused in accordance with the guidelines provided bythe Brazilian College of Animal Experimentation,and were authorized by the Ethics Committee forAnimal Research of the Butantan Institute (002/2001). Specimens of adult, female and male C. spixii

(Figueiredo and Menezes, 1978) fish were collectedin the Brazilian state of Sao Paulo. The mucusvenom was obtained through scratching of the skinwith a slide glass, being immediately conditioned inice, then it was diluted in sterile saline, homogenized,and centrifuged for collection of the supernatant.The sting venom extraction was accomplished withtrituration and centrifugation. The supernatant wascollected and stored at �70 1C. Protein concentra-tions were determined by the colorimetric method ofBradford (1976). Standard curves were constructedusing bovine serum albumin (Sigma Chemicals, StLouis, MO) diluted in duplicate.

2.2. Sodium dodecyl sulfate-polyacrylamide gel

electrophoresis (SDS-PAGE)

SDS-PAGE was carried out according to themethod of Laemmli (1970). Thirty micrograms ofmucus or sting venoms were analyzed by 12% SDS-PAGE gels. Prior to electrophoresis, the sampleswere mixed 1:1 (v/v) with sample buffer. The gel wasstained with the Coomassie R-250.

2.3. Microcirculatory alterations

Observations of leukocyte interactions in venulesof the mouse cremaster muscle were performed asdescribed by Norman et al. (2000) and Sperandioet al. (2003). Mice were anesthetized with an i.p.injection of sodium pentobarbital (20mg/kg bodyweight), placed on a water-heated bed (37 1C), andthe cremaster muscle was exposed for topicalapplication of venom (25mg diluted in 20ml of sterile

ARTICLE IN PRESSM.E.P. Junqueira et al. / Toxicon 49 (2007) 909–919 911

saline). Control experiments were performed byapplying 20ml saline under otherwise identical condi-tions. Muscle preparations were observed in atriocular microscope (Axioskope, Carl-Zeiss, Ger-many) and analyzed with image analyzer software(KS 300, Kontron, Germany). The images wereobtained using a � 10/025 longitudinal distanceobjective/numeric aperture and 1.6 optovar. Fiveminutes of observation were recorded before applica-tion of the venoms to analyze the dynamics in controltissue. Experiments were carried out for up to 30min.

2.4. Evaluation of the vascular permeability

For permeability analysis, the Evans blue dye,20mg/kg in 200ml of saline was i.v. administered20min before the venoms (12.5; 25; 50 or 100mgdiluted in 200ml of sterile saline) or 200ml of saline i.p.administration. After 2 h, mice were sacrificed, andtheir peritoneal cavity was washed with 2ml of ice-cold phosphate-buffered saline (PBS) plus 0.1%bovine serum albumin (BSA). The cells were spundown and the optical density (OD) of the supernatantwas measured at 620nm as an indicator of Evans blueleakage into the peritoneal cavity (Sirois et al., 1988).The results were expressed in mg of Evans blue/ml andthe concentration of Evans blue was calculated froma standard curve of a known concentration.

2.5. C. spixii-induced peritonitis

Different groups of mice were injected i.p. with100mg of both venoms (mucus or sting) diluted in500ml of sterile saline. Mice only injected with salinewere used as control. At time points indicated (2, 24,48 h, and 7 days) after venoms injection, animals weresacrificed by CO2 asphyxiation, peritoneal cells wererecovered by peritoneal lavage using 5ml of ice-coldsterile PBS plus 0.1% EDTA (ethylenediaminetetraa-cetic acid). Typically, peritoneal exudate lavage fluidis free of red coloration, indicating the lack of redblood cell contamination. If present, red blood cells inperitoneal lavage were lysed in Tris–buffered ammo-nium chloride (pH 7.2) buffer. After centrifugation,the supernatant from cell suspension was collected forcytokine and chemokine analyses.

2.6. Quantification of peritoneal cavity cell

infiltration

The leukocyte cell counts from the peritonealexudate lavage fluid were performed using a

hemocytometer and cytocentrifuge slides were pre-pared, air dried, fixed in methanol, and stained(Wright–Giemsa, Scientific Products, Chicago, IL).For differential cell counts, 300 leukocytes wereenumerated and identified as macrophages orpolymorphonuclear neutrophils, on the basis ofstaining and morphologic characteristics.

2.7. Flow cytometric analysis

Peritoneal cells (5� 105ml) from 48-h stingvenom exudates or 7-day mucus exudates wereseeded to plate substratum at 37 1C for 18 h.Nonadherent cells were removed by washing withwarm PBS and adhered macrophages were re-stimulated in vitro with sting or mucus venoms(1 mg/ml). After 4 h, the supernatants were stored forIL-12p70 determination, and adhered macrophageswere washed, counted, and resuspended in FACSbuffer (1% BSA in PBS containing 0.01% NaN3).For phenotypic analysis, cells (0.2–1� 106 cells/stain) were initially incubated with either 10%mouse serum or CD16/CD32 (Fc block) for20min at 4 1C. Subsequently, cells were incubatedwith RPE anti-mouse CD11b, FITC anti-mouseCD11c, and RPE anti-mouse MHC class II (majorhistocompatibility complex class II). All incubationswere performed on ice for 20min and were followedby three washes with FACS buffer. Appropriateisotype controls were used in all cases. For flowcytometric analysis, a typical forward and sidescatter gate was set to exclude dead cells andaggregates; a total of 104 events in the gate wereanalyzed using a FACScalibur and Cell Quest Prosoftware (BD Biosciences, San Jose, CA).

2.8. Eicosanoid assays

Concentrations of LTB4 (Leukotriene B4) weremeasured in the peritoneal exudate lavage fluidcollected for 2 h after venoms or saline injection, bya specific enzymatic immunoassay, using a com-mercial kit (Cayman Chemicals, MI, USA). In brief,100 ml aliquots of each sample were incubated withthe eicosanoid conjugated with acetylcholinesteraseand the specific rabbit antiserum in 96-well micro-titration plates, coated with anti-rabbit IgG mousemonoclonal antibody. After addition of the sub-strate, the absorbances of the samples were recordedat 412 nm in a microplate reader, and concentrationof the eicosanoid was estimated from standardcurve.

ARTICLE IN PRESS

97.0 -

66.0 -

45.0 -

30.0 -

20.1 -

14.4 -

Venoms

MucusSting

Mw

(kDa)

Fig. 1. Eletrophoretical profile of Cathorops spixii mucus or sting

venoms. Venoms of C. spixii were analyzed by SDS-PAGE using

polyacrylamide resolution gel 12% under no-reduction condi-

tions, and revealed by Coomassie Blue. Left lane, Mw markers

and respective molecular weights.

µg of venoms diluted in 200 µl of sterile saline

*

*

*

* *

**

0

50

100

150

200

Saline 12.5 25 50 100

µg o

f E

vans

blu

e/m

l

**

Mucus Venom

Sting Venom

Fig. 2. Evaluation of the vascular permeability in peritoneal

cavity after C. spixii venoms injection. Mice were injected i.p.

with different doses of venom (12.5; 25; 50 and 100mg diluted in

200ml of sterile saline) or 200ml of saline and 20min before

received a i.v. injection of the Evans blue. The supernatant of

peritoneal wash was measured 2 h after by espectophotometry at

620 nm. The vascular permeability was expressed as mg of Evans

blue/ml. The bars represent the mean7SEM. *po0.05 compared

with control group.

M.E.P. Junqueira et al. / Toxicon 49 (2007) 909–919912

2.9. Quantification of cytokines and chemokines

Cytokines and chemokines were measured in thesupernatant of the peritoneal exudate lavage fluid orof the macrophage cultures by a specific two-sitesandwich ELISA, using the OpteIA for Interleukin-1 beta (IL-1b), tumor necrosis factor-alpha (TNF-a), and Interleukin–6 (IL-6), IL-12p70, KC (Che-mokine family with homology to human IL-8), andMonocyte chemoattractant protein-1 (MCP-1) ac-cording to the manufacturer’s instructions (B&DPharmingen, Oxford, UK). Binding of biotinylatedmonoclonal antibodies was detected using strepta-vidin–biotinylated horseradish peroxidase complexand 3,30,5,50-tetramethylbenzidine (B&D Pharmin-gen, Oxford, UK). Samples were quantified bycomparison with standard curves of recombinantmice cytokines and chemokines. The results wereexpressed as the arithmetic mean7SEM for tripli-cate samples. Detection limits were 7.8 pg/ml foreach cytokine and chemokine.

2.10. Statistical analysis

All results were presented as means7SEM of atleast six animals in each group. Parametric datawere evaluated using analysis of variance, followedby the Tukey test for multiple comparisons.Non-parametric data were assessed using theMann–Whitney test. Differences were consideredstatistically significant at po0.05. The SPSS statis-tical package (Release 13.0, Evaluation Version,2004) was employed.

3. Results

3.1. Eletrophoretical profile of mucus and sting

venoms

Venoms were submitted to 12% SDS–PAGE(30 mg of protein/well) and after running severalbands were visualized (Fig. 1). Sting venompresented bands located mainly between 66 to97 kDa, around 45 kDa, and at 14.4 kDa. The bandaround 45 kDa was intensively observed in mucusvenom that shows more one band around 14.4 kDa.

3.2. Induction of an Ag-specific inflammatory

response in the peritoneal cavity

Alterations in vascular permeability were deter-mined by quantifying the amount of Evans blue in

ARTICLE IN PRESSM.E.P. Junqueira et al. / Toxicon 49 (2007) 909–919 913

the peritoneal lavage after i.p. venoms injection.The Evans blue dye binds to serum proteins andthus can be used to quantify alterations in vascularpermeability. The result depicted in Fig. 2 showsthat injection of both venoms in all doses caused anincrease in vascular permeability into the peritonealcavity 2 h after injection.

To investigate the potential for C. spixii venoms(mucus or sting) in leukocyte rolling and adhesionto endothelial cells under the conditions that prevailin living microvessels, the cremaster muscle of micewas used for topical application of venom (25 mgdiluted in 20 ml saline), and the experiments werecarried out for up to 30min (Fig. 3). A few rollingleukocytes (velocity 430 mm/s), but essentially not

Mucus Venom

Sting Venom

Sterile Saline Muc

min after topical application

00

20

40

60

80

100

10 20 30

Rol

ling

Leu

kocy

te(p

er m

in)

*

*

*

*

**

RollingA B

Fig. 3. Analysis of alterations in microcirculation induced by C. spixii

sterile saline were topically applied in the cremaster muscle of anesthe

observed for up to 30min and each 10min the rolling (A) and the adh

control group.

firmly adherent cells, were observed in the post-capillary venules of control mice (data not shown).The average rolling and adherent leukocyte numberwas higher in the venoms-injected mice than incontrol mice at any of the time points. Numerousleukocytes interacted with the endothelium in thecremaster of mucus or sting venoms mice, andthe vast majority of these cells adhered firmly to thevessel walls until 30min after venoms injection.Analysis of the recorded videotapes did not showany evidence of accumulated platelets in postcapil-lary cremaster muscle venules of venoms- or salinemice.

The inflammatory reaction in the peritonealcavity following mucus injection was characterized

Mucus Venom

Sting Venom

us Venom Sting Venom

min after topical application

*

*

*

*

*

0

5

10

15

20

25

30

0 10 20 30

Adh

eren

t Leu

kocy

te(p

er 1

00 u

m)

Adherence

venoms. Samples of 25 mg of different venoms diluted in 20 ml oftized mice. The aspect of the pre- and post-capillary venules was

esion (B) were registered during 1min. *po0.05 compared with

ARTICLE IN PRESS

Saline

Mucus Venom

0

100

200

300

400

*

Tot

al c

ells

(x

105 )

*

2 h 24 h 48 h 7 day

*

Neu

trop

hils

(x

105 )

0

50

100

150

*

*

Saline 2 h 24 h 48 h 7 day

0

100

200

300

400

Mac

roph

ages

cel

ls (

x 10

5 )

Saline 2 h 24 h 48 h 7 day

*

Time after injection

*

0

100

200

300

400

**

Tot

al c

ells

(x

105 )

Saline

Sting Venom

2 h 24 h 48 h 7 day

*

0

50

100

150

Saline 2 h 24 h 48 h 7 day

Neu

trop

hils

(x

105 )

*

Saline 2 h 24 h 48 h 7 day

0

100

200

300

400

Mac

roph

ages

cel

ls (

x 10

5 )

*

Time after injection

A

B

C

D

E

F

Fig. 4. Induction of peritonitis by C. spixii venoms. At different time points (2, 24, 48 h and 7 days) after i.p. injection of 100 mg of the

mucus or sting venom diluted in 500ml sterile saline animals were sacrificed and peritoneal cavities were washed for total (A, D),

neutrophils (B, E), and macrophages (C, F) cell count. Mice only injected with saline were considered as control group. The results

represent the mean7SEM. *po0.05 compared with control group.

M.E.P. Junqueira et al. / Toxicon 49 (2007) 909–919914

ARTICLE IN PRESSM.E.P. Junqueira et al. / Toxicon 49 (2007) 909–919 915

by a typical �2-fold increase in total cell number,remaining to 7 days (Fig. 4A). To attempt tocompare the inflammatory reaction induced by bothtype of venoms from C. spixii, mice were injectedwith sting venom. The sting-injected mice showedan increase in cell number in the peritoneal cavity,remainig to 48 h, with an �1.9-fold increase,followed by a decrease back to normal resident cellnumbers by day 7 (Fig. 4D). The composition ofcells present in the peritoneal cavity was analyzedmorphologically at each time point following theinduction of mucus- or sting venom peritonitis. Insaline-injected mice few polymorphonuclear neu-trophils were present, with macrophages being thepredominant cell type (Fig. 4). Injection of mucusvenom into the peritoneal cavity caused a rapidinflux of neutrophils, reaching a peak at 24 h, butthen dropping rapidly at 48 h (Fig. 4B); this wasfollowed later by macrophage infiltration into thecavity, on day 7 (Fig. 4C). However, in mice injectedwith the sting venom, there was a significantrecruitment of neutrophil only at 2 h (Fig. 4E) andsubsequent infiltration of macrophages at 24 and48 h (Fig. 4F).

3.3. Inflammatory mediators in the peritoneal cavity

induced by venoms

The peritonitis reaction is characterized by acuteinflammation that involves the migration of leuko-cytes, vascular leakage, and cytokine, chemokinesor lipid mediators production. Thus, the release ofIL-1b, TNF-a, IL-6, KC, MCP-1, and LTB4 in ourvenom-induced peritonitis model in mice wascompared. In the Fig. 5 it is seen that both venomswere able to induce a significant release of LTB4 inthe peritoneal cavity 2 h after injection. In this time,significant levels of IL-6 was also seen in peritonealcavity of mice injected mainly with mucus venom(Fig. 5B), and both venoms elicited elevated KC andMCP-1 chemokines production (Fig. 6). IL-1b andTNF-a could not be detected in peritoneal exudatelavage fluid following either mucus- or sting venom-induced inflammation (data not shown).

3.4. Characterization of cell populations in the

peritonitis models by surface marker analysis

As previously showed (Fig. 4) the injection ofmucus or sting venoms in mice caused a macro-phage infiltration into the peritoneal cavity on day 7or 48 h, respectively. These periods of time were

chosen for evaluating the role of mucus or stingvenoms on macrophage stimulation, because themacrophage response was maximal. For then,adhered macrophages were re-stimulated in vitro

with sting or mucus venoms (1 mg/ml, each one) for4 h. Following stimulation, it was observed anincrease in the levels of bioactive IL-12p70 inculture supernatants from 7-day mucus or 48-hsting venoms-induced exudates compared withsupernatants of macrophages culture from miceinjected with saline (Fig. 7A).

Adherent macrophages from saline injected miceexpresses high level of CD11b (82.3472.9%) andlow number of CD11c positive cells (3.7970.1%).The MHC class II expression in CD11c positivemacrophages was also low (4.1370.1%) (Fig. 7B).Following in vitro re-stimulation with mucus venomof 7-day mucus vemons-induced exudates, thenumber of CD11b positive cells remained high(71.6672.6%), but the number of CD11c or CD11c� MHC class II positive macrophages wassignificantly increased (5.8970.2% and 6.5770.2%, respectively, Fig. 7C). By contrast, only46.5571.7% of adherent macrophages from 48-hsting-induced peritonitis were positive forCD11b, and the expression levels of CD11c orMHC class II in CD11c positive cells weresignificantly lower (Fig. 7D).

4. Discussion

The catfish C. spixii are broadly distributed alongthe whole coast of Brazilian sea and river ecosys-tems, and provoke frequent accidents in swimmers,tourists, and mainly in fishermen. In this study, theinflammatory reaction induced by the mainlyvenoms (mucus and sting) of C. spixii wasinvestigated, allowing the examination of thekinectics of leukocyte recruitment into peritonealcavity and the mediators production that takingplace during this type of response.

Increased vascular permeability leading to vascu-lar leakage is a central feature of all inflammatoryreactions and is critical for the formation of aninflammatory exudate. We have shown that mucusor sting venoms of C. spixii induce an increase invasopermeability in the peritoneal cavity. Becauseleukotrienes are the most products to exert directeffects on vascular tone and permeability (Brain andWilliams, 1985), the presence of LTB4 induced byboth venoms was likely to account for these effectsobserved.

ARTICLE IN PRESS

0.0

0.5

1.0

1.5

2.0

*

KC

(ng

/mL

)

*

Saline MucusVenom

StingVenom

0

50

100

150

*

*

MC

P-1

(pg

/mL

)

Saline MucusVenom

StingVenom

A B

Fig. 6. Quantification of chemokines in supernatant of peritoneal washes from mice injected with C. spixii venoms. Two hours after i.p.

injection of 100mg of the mucus or sting venom diluted in 500ml sterile saline, animals were sacrificed and peritoneal cavities were washed

for KC (A) and MCP-1 (B) determinations by specific ELISA. Values represent the mean7SEM. *po0.05 compared with control group.

Saline MucusVenom

StingVenom

0

100

200

Leu

kotr

iene

B4

(pg/

mL

)* *

0

50

100

150

*

IL-6

(pg

/mL

)

#

*

Saline MucusVenom

StingVenom

A B

Fig. 5. Leukotriene B4 and IL-6 concentrations in the peritoneal fluid after C. spixii venoms injection. Two hours after i.p. injection of

100mg of the mucus or sting venom diluted in 500ml sterile saline, animals were sacrificed and peritoneal cavities were washed for LTB4

(A) and IL-6 (B) determinations by specific ELISA. Each bar represents the mean7SEM. *po0.05 compared with control group;

#po0.05 compared with sting group.

M.E.P. Junqueira et al. / Toxicon 49 (2007) 909–919916

In addition, this result is similar to that observedwith the venoms of another Brazilian fish, Thalasso-

phryne nattereri or rays Potamotrygon cf. scobina

and Potamotrygon gr. orbygnyi which are alsocapable of inducing augmented vascular permeabil-ity in mice. It can be suggested that the presence of asimilar toxin with 14–15 kDa in these venoms couldbe related with this effect (Lopes-Ferreira et al.,1998; Lima et al., 2003; Magalhaes et al., 2006).However, the presence of homologous toxinsamong these venoms will only be confirmed aftersequencing determination.

The extravasation of immune cells from theperipheral blood through the vascular endotheliuminto the extracellular matrix is a common event in

inflammatory manifestations (Cid, 1996). Futher-more, the cellular infiltration induced by bothvenoms was evaluated. The results using intravitalmicroscopy showed that the mucus and stingvenoms applied topically in cremaster muscle actdirectly on endothelial cells of post-capillary venulescreating an adhesive surface for rolling a greatnumber of leukocytes. In contrast, the augmentedrolling and adhesiveness of leukocytes to theendothelium induced by of P. cf. scobina andP. gr. orbygnyi venoms was only observed aftersubcutaneous injection of these venoms, indicatingthat the alterations in microcirculatory net wassubsequent to a systemic inflammatory effect ofthese venoms (Magalhaes et al., 2006). Then, these

ARTICLE IN PRESS

FITC

71.66 %

CD

11b

PE

CD11c FITC

5.89%

PE

6.57%

CD11c FITC

46.55

MH

C I

I PE

%

FITC

CD

11b

PE

CD11c FITC

2.28 %

PE

2,82 %

CD11c FITC

PE

MH

C I

I PE

3.79 %

CD11c FITC

4.13 %

CD11c FITC

MH

C I

I PE

CD

11b

PE

82.34 %

FITC

Saline

Mucus Venom (7-day)

Sting Venom (48-hours)

IL-1

2p70

(pg

/ml)

Mucus Venom7-day

Sting Venom48-hours

0

10

20

30

40

50

60*

Saline Saline

Mucus Venom Sting Venom

*

104

103

102

101

100

104

103

102

101

100

104

103

102

101

100

104

103

102

101

100

104

103

102

101

100

104

103

102

101

100

104

103

102

101

100

104

103

102

101

100

104

103

102

101

100

100 101 102 103 104 100 101 102 103 104 100 101 102 103 104

100 101 102 103 104 100 101 102 103 104 100 101 102 103 104

100 101 102 103 104 100 101 102 103 104 100 101 102 103 104

A

B

C

D

Fig. 7. Activation markers in macrophage population. Peritoneal cells (5� 105ml) from 48-h sting venom exudates or 7-day mucus

exudates after adherence were re-stimulated in vitro with sting or mucus venoms (1mg/ml). After 4 h, the supernatants were stored for

IL-12p70 determination by ELISA (A), and cells from mice injected with saline (B), mucus (C) or sting venoms (D) were analyzed by the

expression of CD11b, CD11c, or CD11c � MHC class II. Data are the mean7SEM of positive cells from two experiments.

M.E.P. Junqueira et al. / Toxicon 49 (2007) 909–919 917

ARTICLE IN PRESSM.E.P. Junqueira et al. / Toxicon 49 (2007) 909–919918

results suggest that both venoms of C. spixii (mucusor sting) elicited a remarkable adhesion moleculesengagement among leukocytes and the endotheliumand significant levels of chemokines that arethought to integrate inflammatory signals fortransmigration. This was confirmed by the analysisof high levels of KC and MCP-1. KC (CXC ora-chemokines) could mediate recruitment of neu-trophils from the bone marrow through the ligationin CXC chemokine ligand 8 (CXCL8) receptors(Terashima et al., 1998), and MCP-1, a CC orb-chemokine acts especially in monocytes (Rollins,1996).

In addition to the marked and sustained inflam-matory reaction in mucus venom injected mice, highlevels of IL-6 was observed in the peritoneal exudatelavage fluid of these mice, although significant IL-6levels and fast resolution of the leukocyte inflam-mation in sting venom mice were detected. It isinteresting to note that IL-6, which throughdifferential control of leukocyte recruitment, activa-tion, and apoptosis has recently emerged as aregulator of the immunological switch from innateto acquired immunity (Diehl and Rincon, 2002;Jones, 2005). High levels of IL-6 are secreted byantigen presenting cells (APC, Rincon and Flavell,1997), and a series of in vivo studies indicate the dualeffect of IL-6 on T cell polarization: IL-6 deficient-mice produce low levels of bioactive IL-12p70(Romani et al., 1996) and the differentiation intoTh1 cells by IL-12, can be impaired in the presenceof IL-6 (Rincon et al., 1997).

Dendritic cells (DCs) are professionally adaptedantigen-presenting cells that induce and coordinateimmune responses (Banchereau and Palucka, 2005).Peritoneal macrophages can be induced to differ-entiate in vitro into cells exhibiting typical DCmorphology, phenotype, and function (Rezzaniet al., 1999). These DC express MHC class II andthe integrin CD11c, a marker found predominantly,although not exclusively, on dendritic cells in themouse (Makala et al., 2002). In this view, the role ofmucus venom on macrophage differentiation wasanalyzed. The macrophages from 7-day mucusvenom-induced exudates upon in vitro stimulationwith mucus venom, expressed CD11c � MHC classII and release bioactive IL-12p70. The presence ofaugmented expression of MHC class II in this cells,confirming their maturity, but their activation statusremains to be determined. On the other hand, stingvenom-elicited peritoneal macrophages lost theability to differentiate into dendritic cells, following

re-stimulation in vitro with sting venom, they do notexpress CD11c, nor do they exhibit sufficient levelsof MHC class II.

Again, this experiments confirm the differentpattern of inflammatory reaction elicited by bothtypes of C. spixii venoms (mucus or sting), andsuggest that the marked presence of toxins with�45 kDa in mucus venom can be involved withimmunogenic properties.

In conclusion, both types of venoms (mucus orsting) promote inflammatory reaction with differentprofiles, and the inflammatory reaction induced bythe first was characterized by antigen persistence inperitoneal cavity that allowed the activation ofphagocytic cells with capacity of antigenic presenta-tion. Furthermore, our finding showed that mucusvenom can affect the phenotype of macrophages,inducing a maturation of this cells through theincrease of the expression of molecules responsiblefor the antigen presentation as MHC class II.

Acknowledgments

The authors wish to thank the PhysiopathologyLaboratory of Butantan Institute for the use of theinstrument for intravital microscopy. Supported byfunds provided by FAPESP and CNPq.

References

Banchereau, J., Palucka, A.K., 2005. Dendritic cells as ther-

apeutic vaccines against cancer. Nat. Rev. Immunol. (5),

296–306.

Bradford, M.M., 1976. A rapid and sensitive method for the

quantitation of microgram quantities of protein utilizing the

principle of protein-dye binding. Anal. Biochem. (72),

248–254.

Brain, S.D., Williams, T.J., 1985. Inflammatory edema induced

by synergism between calcitonin gene-related peptide (CGRP)

and mediators of increased vascular permeability. Br. J.

Pharmacol. 86 (4), 855–860.

Cid, M.C., 1996. New developments in the pathogenesis of

systemic vasculitis. Curr. Opin. Rheumatol. (8), 1–10.

Diehl, S., Rincon, M., 2002. The two faces of IL-6 on Th1/Th2

differentiation. Mol. Immunol. 39 (9), 531–536.

Figueiredo, J.L., Menezes, N.A., 1978. Manual de Peixes

Marinhos do Brasil—II. Teleostei (1). Muzeu de Zoologia—

Universidade de Sao Paulo. Sao Paulo, pp. 34–95.

Froese, R., Pauly, D., 2005. FishBase. World Wide Web

electronic publication.

Haddad Jr., V., 2000. Atlas of Dangerous Aquatic Animals of

Brazil: A Guide to Identification and Treatment, Editora

Roca, Sao Paulo 145pp.

Haddad Jr., V., Martins, I.A., 2006. Frequency and gravity of

human envenomations caused by marine catfish (suborder

ARTICLE IN PRESSM.E.P. Junqueira et al. / Toxicon 49 (2007) 909–919 919

siluroidei): a clinical and epidemiological study. Toxicon 47

(8), 838–843.

Jones, S.A., 2005. Directing transition from innate to acquired

immunity: defining a role for IL-6. J. Immunol. 175 (6),

3463–3468.

Laemmli, U.K., 1970. Cleavage of structural proteins during

assembly of the head of bacteriophage T4. Nature (227),

680–685.

Lima, C., Clissa, P.B., Piran-Soares, A.A., Tanjoni, I., Moura-

da-Silva, A.M., Lopes-Ferreira, M., 2003. Characterization of

local inflammatory response induced by Thalassophryne

nattereri fish venom in a mouse model of tissue injury.

Toxicon 42, 499–507.

Lopes-Ferreira, M., Barbaro, K.C., Cardoso, D.F., Moura-da-

Silva, A.M., Mota, I., 1998. Thalassophryne nattereri fish

venom: biological and biochemical characterization and

serum neutralization of its toxic activities. Toxicon (36),

405–410.

Magalhaes, K.W., Lima, C., Piran-Soares, A.A., Marques, E.E.,

Hiruma-Lima, C.A., Lopes-Ferreira, M., 2006. Biological and

biochemical properties of the Brazilian Potamotrygon stin-

grays: Potamotrygon cf. scobina and Potamotrygon gr.

orbignyi. Toxicon 47 (5), 575–583.

Makala, L.H., Nishikawa, Y., Mishima, M., Inoue, N., Xuan, X.,

Suzuki, H., Fujisaki, K., Mikami, T., Nagasawa, H., 2002.

Phenotype and function of murine peritoneal cavity macro-

phage derived-dendritic cells. J. Vet. Med. Sci. 64 (9),

813–820.

Norman, K.E., Katopodis, A.G., Thoma, G., Kolbinger, F.,

Hicks, A.E., Cotter, M.J., Pockley, A.G., Hellewell, P.G.,

2000. P-Selectin glycoprotein ligand-1 supports rolling on E-

and P-selectin in vivo. Blood (96), 3585–3591.

Rezzani, R., Rodella, L., Zauli, G., Caimi, L., Vitale, M., 1999.

Mouse peritoneal cells as a reservoir of late dendritic cell

progenitors. Br. J. Haematol. 104 (1), 111–118.

Rincon, M., Flavell, R.A., 1997. Transcriptional control of Th1/

Th2 polarization. Curr. Biol. (7), 729–732.

Rincon, M., Anguita, J., Nakamura, T., Fikrig, E., Flavell, R.A.,

1997. Interleukin (IL)-6 directs the differentiation of IL-4-

producing CD4+ T cells. J. Exp. Med. 185 (3), 461–469.

Rollins, B.J., 1996. Monocyte chemoattractant protein 1: a

potential regulator of monocyte recruitment in inflammatory

disease. Mol. Med. Today (5), 198–204.

Romani, L., Mencacci, A., Cenci, E., Spaccapelo, R., Toniatti,

C., Puccetti, P., Bistoni, F., Poli, V., 1996. Impaired

neutrophil response and CD4+ T helper cell 1 development

in interleukin 6-deficient mice infected with Candida albicans.

J. Exp. Med. 183 (4), 1345–1355.

Russell, F.E., 1971. Poisonous Marine Animals. TFH Publica-

tions, New York.

Sirois, M.G., Jancar, S., Braquet, P., Plante, G.E., Sirois, P., 1988.

PAF increases vascular permeability in selected tissues: effect

of BN-52021 and L-655,240. Prostaglandins 36 (5), 631–644.

Sperandio, M., Smith, M.L., Forlow, S.B., Olson, T.S., Xia, L.,

McEver, R.P., Ley, K., 2003. P-selectin glycoprotein ligand-1

mediates L-selectin-dependent leukocyte rolling in venules.

J. Exp. Med. 197 (10), 1355–1363.

Terashima, T., English, D., Hogg, J.C., van Eeden, S.F., 1998.

Release of polymorphonuclear leukocytes from the bone

marrow by interleukin-8. Blood (92), 1062–1069.