Introduction

Envenomations by snakes of the family Viperidae, comprising true vipers and pit vipers, are characterized by a prominent and complex series of pathological alterations at the site of venom injection. These include edema, hemorrhage, necrosis of skeletal muscle, and diverse alterations in the skin, including blistering and dermonecrosis (Warrell, 1996, 2004; Gutiérrez and Lomonte, 2003). Important advances have been made toward understanding the pathogenesis of hemorrhage and myonecrosis in viperid snakebite envenomation (see reviews by Gutiérrez and Ownby (2003), Fox and Serrano (2005), and Gutiérrez et al. (2005)). However, the study of the pathological alterations induced by these venoms in the skin has received little attention, despite its clinical relevance.

Many snake venoms induce dermonecrosis when injected intradermally (i.d.), (Theakston and Reid, 1983). The histological features of dermonecrosis after injection of the venom of the spitting cobra, Naja nigricollis (family Elapidae), were described (Iddon et al., 1987). On the other hand, the main pathological manifestations of skin pathology in viperid snakebite envenomation, that is, dermonecrosis and blistering, were reproduced in a mouse model after the intramuscular injection of a hemorrhagic metalloproteinase isolated from the venom of Bothrops asper (Rucavado et al., 1998). However, the characterization of the pathological events was performed on a qualitative basis only, and a more detailed understanding of these effects induced by venom components is pending.

The study of snake venom-induced skin pathology is relevant from another perspective, that is, as a model to understand dermonecrosis and blister formation from a broader perspective. A variety of diseases, including inherited deficiencies of proteins involved in the dermal–epidermal junctions (Uitto et al., 1997), autoimmune diseases (Schmidt and Zillikens, 2000), and toxicity by natural and synthetic compounds (for example, brown recluse spider venom, sulfur mustard, and lewisite) (King et al., 1994; Tambourgi et al., 2005; Greenberg et al., 2006), affect the skin, causing vesication, blistering, and necrosis. Owing to the well-characterized biochemical and toxicological profiles of snake venom components, they may become useful experimental tools to investigate the pathogenic mechanisms involved in a variety of skin pathologies. In this study, we performed a qualitative and quantitative characterization of the degenerative and reparative/regenerative events occurring in the ear of mice after injection of the hemorrhagic metalloproteinase BaP1 from the venom B. asper.

Results

Development of hemorrhage and histological assessment of skin damage

Tissue samples from control mice injected with phosphate-buffered saline (PBS) showed histological features typical of normal skin (Figure 1a). BaP1, at doses of 6 and 15 g, induced a limited hemorrhagic lesion and histological alterations that reproduced what has been described in the clinical setting of pit viper envenomations, that is, hemorrhage, formation of blisters, and dermonecrosis. A conspicuous edema of rapid onset occurred, evidenced by a significant increment in the thickness of the ear (Figure 2a). Concomitantly, a rapid and prominent hemorrhagic effect developed, as demonstrated by histological analysis (Figure 1b and c) and by the increment in the hemoglobin concentration in ear homogenates (Figure 2b). Qualitatively similar findings were observed with the two toxin doses used.

Figure 1.

Histological alterations induced by BaP1 in the skin. Light micrographs of paraffin-embedded sections of the skin of ears of mice injected i.d. with either (a) 20 l PBS or (b–f) 6 g BaP1 in 20 l PBS. (a) PBS 1 hour after injection, with normal histological pattern. (b) After 1 hour and (c) 6 hours: notice prominent swelling, hemorrhage (arrows), and blister formation (^*). (d) After 24 hours, with loss of epidermis, inflammatory infiltrate (arrow head), proteinaceous exudate, and eschar formation. (e) After 72 hours, with granulation tissue (arrow head) and proliferation of epithelial cells in the borders of denuded epidermis (^*). (f) After 14 days, with normal histological appearance, including sebaceous glands, and a successful re-epithelization. hematoxylin-and-eosin staining. Bars=250 m in (a, b, d–f) and 50 m in (c).

Full figure and legend (311K)

Figure 2.

Edema and hemorrhage induced by BaP1 in the ears. (a) Thickness of the ears of mice injected intradermally with either 20 l PBS or with 15 g BaP1 dissolved in 20 l PBS. At various time intervals, groups of three mice were killed, and the injected ears dissected out and processed for histological observation. The thickness of the ear was determined as described in Materials and Methods. Results are presented as meanSD (n=3). ^*P<0.05 when compared with PBS-injected mice; °P<0.05 when compared with results at 1, 6, and 24 hours. (b) Hemorrhagic effect of BaP1 in the mouse ear skin. Mice were injected intradermally, in the ear, with either 20 l PBS or 15 g BaP1, in 20 l PBS. At various time intervals, animals were killed and ear tissue cut into small pieces and added to Drabkin solution. After centrifugation, absorbance at 540 nm was recorded and hemoglobin concentration was determined. Results are presented as meanSD (n=3). ^*P<0.05 when compared with mice injected with PBS alone.

Full figure and legend (20K)

The extent of edema and hemorrhage was heterogeneous in the two sides of the ear separated by the central cartilage structure, being more intense in the side that was injected with the toxin, thus evidencing that the cartilage layer constitutes a diffusion barrier for the toxin. Formation of blisters was observed as early as 1 hour after injection (Figure 1b and c). They were characterized by a clear separation of dermis and epidermis. The percentage of the length of dermal–epidermal interface that became vesicated showed a maximum at 1 hour, corresponding to 4418% of the total epidermal–dermal interface. At later time intervals, some vesicated areas became denuded of epidermis (not shown). Extravasated erythrocytes were not located in the space left after the separation of epidermis, that is, in the blister fluid, but instead remained confined to the dermis (Figure 1b and c). No overt morphological alterations were observed in the keratinocytes of regions showing blistering at 1 and 6 hours (Figure 1b and c). Some areas were devoid of epidermis, and a proteinaceous fibrinoid exudate was observed at the denuded skin surface at 24 and 72 hours (Figure 1d and e).

An inflammatory reaction occurred after the initial pathological alterations described. It was evidenced by the presence of an abundant leukocyte infiltrate in which neutrophils were the predominant cell type. Leukocytes were present in the dermis, initially around venules and later on throughout the dermis. Inflammatory infiltrate was followed by the formation of granulation tissue. No evident alterations were noticed in the structure of cartilage. Afterward, a re-epithelialization process was evident, characterized by increments in the numbers of keratinocytes at the borders of the denuded lesions (Figure 1e), followed by the migration of these cells underneath the eschar. Re-epithelization was complete at 7 days, and the skin regained its normal appearance at 14 days (Figure 1f), including the presence of skin appendages, that is, sebaceous glands. Scar tissue with collagen was absent after the reparative response, and the thickness of the ear returned to control values at 7 and 14 days (Figure 2a).

The main pathological features described, that is, hemorrhage and blistering, were completely abolished when BaP1 was incubated with the metalloproteinase inhibitor batimastat, demonstrating that these effects depend on the proteolytic activity of the enzyme. Batimastat alone did not induce any pathological effect in the ear. Injection of 10 g of a serine proteinase isolated from B. asper venom did not induce any of the pathological effects described. On the other hand, depletion of neutrophils did not modify the pathological effects induced by BaP1. The extent of hemorrhage and blistering was similar in neutropenic and non-neutropenic mice.

Alterations in the microvascular density

Inmunostaining of the endothelial cell marker vascular endothelial growth factor receptor-2 (VEGFR-2) allowed the assessment of the changes occurring in the microvasculature. A reduction in the MVD was observed as early as 1 hour after injection (Figure 3) evidencing early damage to endothelial cells. A rapid process of endothelial cell replication and revascularization occurred, as revealed by the increase in the MVD within the first 72 hours after BaP1 injection (Figure 3). Then, at 7 and 14 days, there was a slight but significant reduction in MVD (Figure 3).

Figure 3.

Microvascular density in the ear tissue of mice injected intradermally with either 20 l PBS or 15 g BaP1 dissolved in 20 l PBS. Groups of three mice were killed at various time intervals, and the ears were dissected out, embedded in cryopreservation medium, and frozen 5 m sections were prepared and fixed in 100% ethanol. Endothelial cells were visualized by immunostaining with anti-mouse VEGFR-2, as described in Materials and Methods. Capillary vessels were identified as immunostained structures having less than 10 m diameter. Microvascular density (MVD) was estimated as the number of capillaries per mm². Results are presented as meanSD (n=3). ^*P<0.05 when compared with values of mice injected with PBS alone.

Full figure and legend (12K)

Immunohistochemical analysis of basement membrane at the dermal–epidermal junction

Immunostaining for type IV collagen and laminin, key components of the dermal–epidermal junction, was analyzed. Control samples from mice injected with PBS had a continuous staining for these antigens along the dermal–epidermal junction (not shown). In samples of tissue injected with BaP1, the immunostaining for type IV collagen and laminin was conserved in most of the areas observed, suggesting that a widespread degradation of these proteins did not occur. In areas where blisters were present, immunostaining for these proteins was observed predominantly at the base, that is, at the dermal side, of the blisters (Figure 4). In the case of type IV collagen, there was also some staining in the roof of some areas of blisters (Figure 4).

Figure 4.

Immunostaining for laminin and type IV collagen. Cryosections were prepared from the ears of mice injected with 15 g BaP1 in 20 l PBS and killed 6 hours after injection. Sections were stained with anti-laminin -1 chain or anti-type IV collagen; all sections were also stained with Hoechst 33258 for the detection of nuclei. Panels (a–c) and (d–f) correspond to two separate sections. (a) Immunostaining for laminin. (d) Immunostaining for type IV collagen. (b and e) Hoechst staining. (c and f) Merged images of Hoechst and either (c) anti-laminin or (f) anti-type IV collagen. Notice that the immunostaining for both basement membrane components predominate in the base of the blisters, whereas the nuclei of keratinocytes are located in the roof of the blisters. Bar=250 m.

Full figure and legend (142K)

TUNEL analysis

Tissue from control mice injected with PBS did not present TUNEL-positive cells in the dermis and epidermis (Figure 5). No positive TUNEL cells were observed in capillary vessels in mice injected with BaP1 at 6 and 24 hours (Figure 5), suggesting that endothelial cells were not affected by apoptosis in this model. In contrast, a prominent staining occurred in many keratinocytes at the epidermis in tissue injected with BaP1 (Figure 5). There was no correlation between apoptosis and the formation of blisters, as TUNEL-positive cells were observed in areas with blisters and in areas devoid of blisters.

Figure 5.

TUNEL staining in sections of mouse ears injected with either 20 l PBS or 15 g BaP1, dissolved in 20 l PBS. Samples were collected 6 hours after injection. Cryosections were prepared and DNA fragmentation was assessed by TUNEL, as described in Materials and Methods. In parallel, sections were stained with anti-VEGFR-2 and Hoechst 33258 for staining of endothelial cells in capillaries and nuclei, respectively. (a–d) Control samples injected with PBS. (e–h) Samples injected with BaP1. (a and e) Hoechst staining. (b and f) Immunostaining for VEGFR-2. (c and g) TUNEL staining. (d and h) Merged images of the three fluorochromes. Notice the predominance of TUNEL-positive cells in the epidermis, and the lack of TUNEL staining in endothelial cells. Bar=250 m.

Full figure and legend (113K)

Discussion

The mouse ear model used in this study constitutes a useful experimental approach to investigate degenerative and reparative/regenerative processes in the skin (Metcalfe et al., 2006). Our observations demonstrate that metalloproteinase BaP1 induces a series of effects that closely mimic what has been described in the skin of patients envenomated by viperid snakebites, that is, edema, inflammation, hemorrhage, blistering, and dermonecrosis (Otero et al., 2002; Gutiérrez and Lomonte, 2003; Warrell, 2004). In contrast, no pathological effects occurred after injection of a serine proteinase isolated from the same venom.

The major pathological findings observed had a very rapid onset and were completely abrogated by preincubation of the enzyme with the metalloproteinase inhibitor batimastat, thus evidencing that the proteolytic activity of BaP1 is critical for its effects. These findings strongly suggest that BaP1-induced pathology is due to the direct hydrolysis of proteins in the dermis and at the dermal–epidermal junctions. As widely demonstrated for other snake venom metalloproteinases (Bjarnason and Fox, 1994), BaP1 is able to digest in vitro various extracellular matrix proteins, including those forming basement membranes, that is, laminin, nidogen, and type IV collagen (Rucavado et al., 1995; Escalante et al., 2006). Such degradative activity has also been described in mouse muscle tissue in vivo (Escalante et al., 2006). It is therefore suggested that the rapid pathological effects on capillary vessels and dermal–epidermal junction hereby described are due to the direct proteolytic activity of BaP1 on the basement membranes of these structures.

Endogenous tissue proteinases are involved in the pathogenesis of skin damage in other toxic models, such as after administration of the alkylating agent sulfur mustard (Chakrabarti et al., 1998; Ray et al., 2002), and in dermonecrosis induced by the sphingomyelinase from the venom of the spider Loxosceles sp., where matrix metalloproteinase (MMP)-9 plays a role in skin damage (Tambourgi et al., 2005; Paixão-Cavalcante et al., 2006, 2007). The possible involvement of inflammatory proteinases in the onset of endothelial cell damage and blistering induced by BaP1 is unlikely on the basis of three experimental observations: (a) Local lesions are of very rapid onset after BaP1 injection, whereas there is a delay in the appearance of these effects in sulfur mustard and Loxosceles-induced damage, which are mediated by endogenous proteinases. (b) Despite the increment in MMP-9 in the skin of mice injected with BaP1, zymographic analysis evidenced only the presence of the latent forms of these MMPs (Rucavado et al., 1998), thus suggesting that at early time intervals, MMPs are not engaged in active proteolysis. (c) Depletion of neutrophils, which contain abundant proteinases including MMPs (Tambourgi et al., 2005), did not modify the extent of hemorrhage and blister formation after BaP1 injection. Previous studies have demonstrated that neutrophils do not participate in the local tissue damage induced by Bothrops sp. venoms in mice (Teixeira et al., 2003, 2005). Therefore, the role of MMPs and other inflammatory proteinases in our model is likely to be related to the processes of tissue remodeling and regeneration, and not to the rapid pathological effects induced by BaP1 in the skin.

Acute capillary vessel damage is one of the most relevant effects induced by snake venom metalloproteinases, an action that leads to hemorrhage and has implications in tissue reparative and regenerative events (Gutiérrez and Rucavado, 2000; Gutiérrez et al., 2005). The mechanism of capillary damage leading to hemorrhage depends on a proteolytic cleavage of structurally relevant components at the basement membrane of capillaries (Bjarnason and Fox, 1994), with the consequent weakening of the capillary stability, followed by the distention of the capillary due to the action of biophysical forces normally operating in the circulation (Gutiérrez et al., 2005), a process that leads to endothelial cell damage, capillary wall disruption, and extravasation (Escalante et al., 2006; this work). By studying samples collected at 24 and 72 hours, and 1 and 2 weeks, we were able to follow the revascularization in the skin. To our knowledge, this is the first quantitative analysis of the process of revascularization in tissue affected by snake venom metalloproteinases. Interestingly, in our model, there was a very rapid restitution of immunostaining to the endothelial marker VEGFR-2, as MVD reached high values at 72 hours, although MVD at 7 and 14 days were still lower than those of control mice. Such early increment in endothelial cell density, followed by a drop, has been described in other experimental models (Schirmer et al., 2004) and is likely to reflect an early endothelial replication phase, stimulated by the release of angiogenic factors by resident as well as inflammatory cells (Lokmic et al., 2006), followed by a regulation of MVD at later time intervals.

Various snake venom metalloproteinases, including BaP1, have been shown to induce apoptosis on endothelial cells in culture (Masuda et al., 2000, 2001; Wu et al., 2001; You et al., 2003; Díaz et al., 2005; Tanjoni et al., 2005); however, the actual role of endothelial cell apoptosis in snakebite envenomation in vivo has not been investigated previously. Our findings in the skin clearly evidence that apoptosis does not occur in endothelial cells after injection of BaP1, as judged by the TUNEL assay. Capillary endothelial cells are rapidly affected after metalloproteinase injection (Ownby et al., 1978; Moreira et al., 1994; Gutiérrez et al., 2006); hence, it is proposed that endothelial cells die predominantly by necrosis (Moreira et al., 1994; Gutiérrez et al., 2006).

The formation of blisters, described in human snakebite envenomations, was reproduced in this experimental model. The rapid appearance of blisters, and their abrogation when BaP1 was incubated with batimastat, suggests that they occur through the direct cleavage of BaP1 of proteins at the dermal–epidermal junction. This complex structure is formed by various proteins and layers, including the hemidesmosomes, the connection between integrin 46 and laminin 5 at the lamina lucida, the lamina densa, with various basement membrane components, and the connection between type VII collagen and fibrillar collagens at the anchoring fibrils (Woodley and Chen, 2001). In the case of BaP1-induced blisters, the presence of immunostaining for type IV collagen and laminin predominantly at the base of the blister strongly suggests that the main site of cleavage is at the lamina lucida, with the consequent separation of epidermis from components of the basement membrane at the lamina densa. This is likely to reflect a different susceptibility of lamina lucida and lamina densa to the action of this metalloproteinase.

A large number of keratinocytes were TUNEL positive, thus suggesting widespread apoptosis in these epithelial cells. Other experimental models of blistering, such as damage induced by sulfur mustard, also concur with epithelial cell apoptosis (Greenberg et al., 2006). In both cases, apoptosis occurred not only in areas of blistering, but also in regions where epidermis was not separated from the dermis. The mechanisms behind such widespread apoptotic process are unknown, but a proteolytic cleavage of epithelial cell integrins or laminin 5 at the lamina lucida may be involved. Such limited proteolysis may occur without an overt separation of dermis and epidermis, but may affect survival signal pathways that depend on the interaction of cell integrins with extracellular matrix components (Grossmann, 2002).

Dermonecrosis is followed by two possible reparative scenarios that follow the inflammatory reaction. In some cases, a regenerative process ensues, with re-epithelization and revascularization of the damaged skin associated with lack of fibrotic scar formation. Alternatively, dermonecrosis might be followed by a deficient regenerative process and by the formation of a permanent fibrotic scar associated with loss of function (Ferguson and O'Kane, 2004; Metcalfe et al., 2006). The predominant outcome depends on the balance between profibrotic and proregenerative inflammatory cytokines and growth factors and may vary with the anatomical site of skin damage and the genetic makeup of the animal strain (Metcalfe et al., 2006). In the case of dermonecrosis due to BaP1, adequate re-epithelization occurred in the ear model and in a study of skin damage in the lower limb of mice (Rucavado et al., 1998; this work), with complete restoration of epidermis and dermis and with a lack of scar formation.

In conclusion, metalloproteinase BaP1 induces a series of pathological events in the mouse ear that closely resemble the pathological alterations described in human viperid snakebite envenomations. The rapid disruptive effects are likely to depend on the direct proteolytic activity of this enzyme on extracellular matrix components. Such drastic pathologic scenario is followed by inflammation, granulation tissue formation, and a successful and complete re-epithelization of the skin. Our observations corroborate the relevance of metalloproteinases in the skin pathology associated with viperid snakebite envenomation and stress the potential usefulness of metalloproteinase inhibitors in the treatment of snake venom-induced local tissue damage (Escalante et al., 2000; Rucavado et al., 2000).

Materials and Methods

Metalloproteinase

BaP1, a 23 kDa hemorrhagic metalloproteinase, was isolated from the venom of adult specimens of Bothrops asper collected at the Pacific region of Costa Rica, as described previously (Gutiérrez et al., 1995; Rucavado et al., 1998). Homogeneity of the preparation was demonstrated by SDS-PAGE run under reducing conditions (Laemmli, 1970). A serine proteinase was isolated from this venom as described by Pérez et al. (2008).

Animals

CD-1 mice (18–20 g body weight) were used throughout the study. Animals were maintained at 12:12 hours light/dark cycle and received water and food ad libitum. The experimental protocols used were approved by the Committee for the Care and Use of Laboratory Animals of Universidad de Costa Rica.

Histological assessment of tissue alterations

Mice were anesthetized with a mixture of ketamine and xylazine. Animals were then injected i.d. in the ear with either 6 or 15 g BaP1, dissolved in 20 l of 0.12 M NaCl, 0.04 M sodium phosphate buffer, pH 7.2 (PBS). In some experiments, mice were injected with 10 g of a serine proteinase isolated from the same venom. Control animals were injected with PBS, under otherwise identical conditions. Groups of three mice were killed at various time intervals (1, 6, 24, and 72 hours, and 7 and 14 days). The ears of both controls and treated animals were dissected out and embedded in cryopreservation medium (ThermoShandon, Pittsburgh, PA) on dry ice; samples were stored at -70°C until used. Frozen sections of 5 m thickness were prepared with a cryostat (Leica CM1850, Wetzlar, Germany). Three sections were placed onto positively charged slides (Erie Scientific Company, Portsmouth, NH), followed by fixation in 100% ethanol, for at least 20 minutes at room temperature, and then they were stored at -70°C. In some experiments, injected ears were dissected out and fixed in 10% formalin solution, prepared using PBS as diluent, and processed routinely for embedding in paraffin. Sections were stained with hematoxylin and eosin for histological observation.

Morphometric analyses

Tissue sections were analyzed for the following parameters: (a) thickness of the ear, determined with the Image Pro-Plus program (Media Cybernetics, Silver Spring, MD); in each section, thickness in mm was determined in three different regions of the ear for each sample analyzed. (b) Percentage of blistering, corresponding to the length of the dermal–epidermal interface presenting blisters, divided by the total length of dermal–epidermal interface analyzed, and multiplied by 100.

Quantification of hemorrhage

Groups of three mice were injected with 15 g BaP1, dissolved in 20 l PBS, as described. Controls received 20 l PBS. At 1, 6, and 24 hours, mice were killed and the injected ears were dissected out and cut into small pieces with a razor blade. All tissue sections were then added to 5 ml of Drabkin reagent and left undisturbed for 24 hours at 4°C. Then, samples were centrifuged at 600 g for 5 minutes, the absorbance of the supernatants at 540 nm was recorded, and hemoglobin concentration was calculated from a standard curve as a quantitative index of hemorrhage (Escalante et al., 2000).

Inhibition of metalloproteinase activity

BaP1 was incubated with the peptidomimetic hydroxamate metalloproteinase inhibitor batimastat (80 M), as described previously (Escalante et al., 2000). Controls included BaP1 without batimastat, and batimastat alone. Aliquots of 20 l of the mixtures, containing 15 g of BaP1, were injected i.d. in the ears of mice, as described. Animals were killed at 1 and 6 hours, and tissues processed routinely for paraffin embedding and hematoxylin-and-eosin staining, as described. Sections were observed for the appearance of pathological alterations (hemorrhage, edema, and blistering).

Effect of neutrophil depletion on pathological changes

The protocol described by Teixeira et al. (2003) for neutrophil depletion, based on the administration of a mAb that recognizes a surface marker in mature granulocytes, was followed. Neutrophil depletion was demonstrated by performing differential leukocyte blood counts. Control and neutropenic mice were injected i.d. in the ear with either 15 g BaP1, dissolved in 20 l PBS, or with 20 l PBS alone. Animals were killed at 1 and 6 hours, and tissues routinely processed for paraffin embedding. Sections were stained with hematoxylin and eosin and analyzed for the presence of edema, hemorrhage, and blistering.

Immunostaining for endothelial cell markers

To identify endothelial cells in capillaries, frozen cross sections were immunostained with a rat mAb anti-mouse VEGFR-2 (BD Biosciences Pharmigen, San Diego, CA), at a 1:20 dilution in TNB buffer, using the TSA Biotin System commercial kit (Perkin Elmer, Boston, MA). Afterward, a biotinylated rabbit anti-rat polyclonal antibody, diluted 1:200 in TNB buffer, was applied. This step was followed by the application of streptavidin peroxidase (1:100 in TNB) and tyramide (1:50 in amplification solution). Finally, sections were incubated with the fluorochrome Cy3-streptavidin and the tissue was co-stained with Hoechst 33258 (Sigma-Aldrich, St Louis, MO) to identify cell nuclei. The immunostained sections were observed in a fluorescence microscope (Olympus, Tokyo, Japan) and the images were captured at a magnification of 20, using a camera Cool SNAP-Pro (Media Cybernetics, Bethesda, MD). For morphometric analysis, three non-overlapping optic fields were captured in each section. The total area of each optic field was determined in mm² and the number of capillaries immunostained for VEGFR-2 was counted using the Image Pro-Plus program (Media Cybernetics). Vessels with a diameter less than 10 m were considered as capillaries. The MVD was calculated according to the definition of Vermeulen et al. (2002), in which the MVD corresponds to the number of capillaries per mm² of tissue.

Immunostaining for basement membrane components

To evaluate the effects of BaP1 on basement membrane proteins, cryosections of injected mouse ears were incubated with a rabbit anti-human collagen IV polyclonal antibody (Fitzgerald Industries International, Concord, MA) and a goat anti-rabbit biotinylated polyclonal antibody (DakoCytomation, Glostrup, Denmark). For the immunodetection of laminin, cryosections were incubated with goat anti-human laminin -1 (Santa Cruz Biotechnology, Santa Cruz, CA) and a rabbit anti-goat biotinylated polyclonal antibody (DakoCytomation). Afterward, sections were incubated with streptavidin-Cy3 (Zymed Laboratories Inc., South San Francisco, CA). In addition, all sections were also stained with Hoechst 33258 for the location of nuclei.

Determination of apoptosis by TUNEL staining

DNA fragmentation, as determined by the TUNEL assay, was assessed in cryosections of injected mouse ears, using the Apotag kit (Chemicon, Temecula, CA). For each time interval, nine tissue sections were stained, following the protocol provided by the manufacturer (www.chemicon.com). To identify TUNEL-positive cells, sections were co-stained with anti-VEGFR-2 and Hoechst 33258 (Sigma-Aldrich). Images of three different, nonoverlapping fields were obtained in each section and images of each fluorochrome were merged using the Image Pro-Plus program.

Statistical analysis

The significance of the differences between the mean values of various quantitative analyses was determined by analysis of variance using the program Statistica 6.0, considering P<0.05 as significant. The significance of differences between pairs of means was analyzed by the Tukey test.

Conflict of Interest

The authors state no conflict of interest.

References

1. Bjarnason JB, Fox JW (1994) Hemorrhagic metalloproteinases from snake venoms. Pharmac Ther 62:325–372 | Article | ChemPort |
2. Chakrabarti AK, Ray P, Broomfield CA, Ray R (1998) Purification and characterization of protease activated by sulfur mustard in normal human epidermal keratinocytes. Biochem Pharmacol 56:467–472 | Article | PubMed | ChemPort |
3. Díaz C, Valverde L, Brenes O, Rucavado A, Gutiérrez JM (2005) Characterization of events associated with apoptosis/anoikis induced by snake venom metalloproteinase BaP1 on human endothelial cells. J Cell Biochem 94:520–528 | Article | PubMed | ChemPort |
4. Escalante T, Franceschi A, Rucavado A, Gutiérrez JM (2000) Effectiveness of batimastat, a synthetic inhibitor of matrix metalloproteinases, in neutralizing local tissue damage induced by BaP1, a hemorrhagic metalloproteinase from the venom of the snake Bothrops asper. Biochem Pharmacol 60:269–274 | Article | PubMed | ChemPort |
5. Escalante T, Shannon J, Moura-da-Silva AM, Gutiérrez JM, Fox JW (2006) Novel insights into capillary vessel basement membrane damage by snake venom hemorrhagic metalloproteinases: a biochemical and immunohistochemical study. Arch Biochem Biophys 455:144–153 | Article | PubMed | ChemPort |
6. Ferguson MW, O'Kane S (2004) Scar-free healing: from embryonic mechanisms to adult therapeutic intervention. Philos Trans R Soc Lond B Biol Sci 359:839–850 | Article | PubMed | ChemPort |
7. Fox JW, Serrano SMT (2005) Structural considerations of the snake venom metalloproteinases, key members of the M12 reprolysin family of metalloproteinases. Toxicon 45:969–985 | Article | PubMed | ISI | ChemPort |
8. Greenberg S, Kamath P, Petrali J, Hamilton T, Garfield J, Garlick JA (2006) Characterization of the initial response of engineered human skin to sulfur mustard. Toxicol Sci 90:549–557 | Article | PubMed | ChemPort |
9. Grossmann J (2002) Molecular mechanisms of "detachment-induced apoptosis-anoikis". Apoptosis 7:247–260 | Article | PubMed | ISI | ChemPort |
10. Gutiérrez JM, Lomonte B (2003) Efectos locales en el envenenamiento ofídico en América Latina. In: Animais Peçonhentos no Brasil. Biología, Clínica e Terapêutica dos Acidentes (Cardoso JLC, França FOS, Wen FH, Málaque CM, Haddad V, eds), São Paulo: Sarvier, 310–323
11. Gutiérrez JM, Núñez J, Escalante T, Rucavado A (2006) Blood flow is required for rapid endothelial cell damage induced by a snake venom hemorrhagic metalloproteinase. Microv Res 71:55–63 | Article | ChemPort |
12. Gutiérrez JM, Ownby CL (2003) Skeletal muscle degeneration induced by venom phospholipases A₂: insights into the mechanisms of local and systemic myotoxicity. Toxicon 42:915–931 | Article | PubMed | ChemPort |
13. Gutiérrez JM, Romero M, Díaz C, Borkow G, Ovadia M (1995) Isolation and characterization of a metalloproteinase with weak hemorrhagic activity from the venom of the snake Bothrops asper (terciopelo). Toxicon 33:19–29 | Article | PubMed |
14. Gutiérrez JM, Rucavado A (2000) Snake venom metalloproteinases: their role in the pathogenesis of local tissue damage. Biochimie 82:841–850 | Article | PubMed |
15. Gutiérrez JM, Rucavado A, Escalante T, Díaz C (2005) Hemorrhage induced by snake venom metalloproteinases: biochemical and biophysical mechanisms involved in microvessel damage. Toxicon 45:997–1011 | Article | PubMed | ChemPort |
16. Iddon D, Theakston RDG, Ownby CL (1987) A study of the pathogenesis of local skin necrosis induced by Naja nigricollis (spitting cobra) venom using simple histological staining techniques. Toxicon 25:665–672 | Article | PubMed | ChemPort |
17. King JR, Peters BP, Monteiro-Riviere NA (1994) Laminin in the cutaneous basement membrane as a potential target in lewisite vesication. Toxicol Appl Pharmacol 126:164–173 | Article | PubMed | ChemPort |
18. Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685 | Article | PubMed | ISI | ChemPort |
19. Lokmic Z, Darby IA, Thomson EW, Mitchel GM (2006) Time course analysis of hypoxia, granulation tissue and blood vessel growth, and remodeling in healing rat cutaneous incisional primary intention wounds. Wound Rep Reg 14:277–288 | Article |
20. Masuda S, Ohta T, Kaji K, Fox JW, Hayashi H, Araki S (2000) cDNA cloning and characterization of vascular apoptosis-inducing protein 1. Biochem Biophys Res Comm 278:197–204 | Article | PubMed | ChemPort |
21. Masuda S, Hayashi H, Atoda H, Morita T, Araki S (2001) Purification, cDNA cloning and characterization of the vascular apoptosis-inducing protein, HV1, from Trimeresurus flavoviridis. Eur J Biochem 268:3339–3345 | Article | PubMed | ChemPort |
22. Metcalfe AD, Willis H, Beare A, Ferguson MW (2006) Characterizing regeneration in the mammalian external ear. J Anat 209:439–446 | Article | PubMed |
23. Moreira L, Borkow G, Ovadia M, Gutiérrez JM (1994) Pathological changes induced by BaH1, a hemorrhagic proteinase isolated from Bothrops asper (terciopelo) snake venom, on mouse capillary blood vessels. Toxicon 32:977–987 | Article | ChemPort |
24. Otero R, Gutiérrez J, Mesa MB, Duque E, Rodríguez O, Arango JL et al. (2002) Complications of Bothrops, Porthidium and Bothriechis snakebites in Colombia. A clinical and epidemiological study of 39 cases attended in a university hospital. Toxicon 40:1107–1114 | Article | PubMed | ChemPort |
25. Ownby CL, Bjarnason JB, Tu AT (1978) Hemorrhagic toxins from rattlesnake (Crotalus atrox) venom. Pathogenesis of hemorrhage induced by three purified toxins. Am J Pathol 93:201–218 | PubMed | ChemPort |
26. Paixão-Cavalcante D, van den Berg CW, Fernandes-Pedrosa MF, Gonçalves de Andrade RM, Tambourgi DV (2006) Role of matrix metalloproteinases in HaCaT keratinocytes apoptosis induced by Loxosceles venom sphingomyelinase D. J Invest Dermatol 126:61–68 | Article | PubMed | ChemPort |
27. Paixão-Cavalcante D, van den Berg CW, Gonçalves de Andrade RM, Fernandes-Pedrosa MF, Okamoto CK, Tambourgi DV (2007) Tetracycline protects against dermonecrosis induced by Loxosceles spider venom. J Invest Dermatol 127:1410–1418 | Article | PubMed | ChemPort |
28. Pérez AV, Rucavado A, Sanz L, Calvete JJ, Gutiérrez JM (2008) Isolation and characterization of a serine proteinase with thrombin-like activity from the venom of the snake Bothrops asper. Braz J Med Biol Res 41:12–17 | PubMed |
29. Ray P, Chakrabarti AK, Broomfield CA, Ray R (2002) Sulfur mustard-stimulated protease: a target for antivesicant drugs. J Appl Toxicol 22:139–140 | Article | PubMed | ChemPort |
30. Rucavado A, Escalante T, Franceschi A, Chaves F, León G, Cury Y et al. (2000) Inhibition of local hemorrhage and dermonecrosis induced by Bothrops asper snake venom: effectiveness of early in situ administration of the peptidomimetic metalloproteinase inhibitor batimastat and the chelating agent CaNa₂EDTA. Am J Trop Med Hyg 63:313–319 | PubMed | ChemPort |
31. Rucavado A, Lomonte B, Ovadia M, Gutiérrez JM (1995) Local tissue damage induced by BaP1, a metalloproteinase isolated from Bothrops asper (terciopelo) snake venom. Exp Mol Pathol 63:186–199 | Article | PubMed | ISI | ChemPort |
32. Rucavado A, Núñez J, Gutiérrez JM (1998) Blister formation and skin damage induced by BaP1, a hemorrhagic metalloproteinase from the venom of the snake Bothrops asper. Int J Exp Pathol 79:245–254 | PubMed | ChemPort |
33. Schmidt E, Zillikens D (2000) Autoimmune and inherited subepidermal blistering diseases: advances in the clinic and the laboratory. Adv Dermatol 16:113–157 | PubMed | ChemPort |
34. Schirmer S, Hoefer I, Buschmann I (2004) Peripheral hind limb ischemia models. In: Methods in Endothelial Cell Biology (Augustin HG, ed), Berlin: Springer-Verlag, 197–206
35. Tambourgi DV, Paixão-Cavalcante D, Gonçalves de Andrade RM, Fernandes-Pedrosa MF, Magnoli FC, Morgan P et al. (2005) Loxosceles sphingomyelinase induces complement-dependent dermonecrosis, neutrophil infiltration, and endogenous gelatinase expression. J Invest Dermatol 124:725–731 | Article | PubMed | ISI | ChemPort |
36. Tanjoni I, Weinlich R, Della Casa MS, Clissa PB, Saldanha-Gama RF, de Freitas MS et al. (2005) Jararhagin, a snake venom metalloproteinase, induces a specialized form of apoptosis (anoikis) selective to endothelial cells. Apoptosis 10:851–861 | Article | PubMed | ChemPort |
37. Teixeira CFP, Chaves F, Zamunér SR, Fernandes CM, Zuliani JP, Cruz-Hofling MA et al. (2005) Effects of neutrophil depletion in the local pathological alterations and muscle regeneration in mice injected with Bothrops jararaca snake venom. Int J Exp Pathol 86:107–115 | Article | PubMed |
38. Teixeira CFP, Zamunér SR, Zuliani JP, Fernandes CM, Cruz-Hofling MA, Fernandes I et al. (2003) Neutrophils do not contribute to local tissue damage, but play a key role in skeletal muscle regeneration, in mice injected with Bothrops asper snake venom. Muscle Nerve 28:449–459 | Article | PubMed | ChemPort |
39. Theakston RDG, Reid HA (1983) Development of simple standard assay procedures for the characterization of snake venoms. Bull World Health Org 61:949–956 | PubMed | ChemPort |
40. Uitto J, Pulkkinen L, McLean WH (1997) Epidermolysis bullosa: a spectrum of clinical phenotypes explained by molecular heterogeneity. Mol Med Today 3:457–465 | Article | PubMed | ISI | ChemPort |
41. Vermeulen PB, Gasparini G, Fox SB, Colpaert C, Marson LP, Gion M et al. (2002) Second international consensus on the methodology and criteria of evaluation of angiogenesis quantification in solid human tumors. Eur J Cancer 38:1564–1579 | Article | PubMed | ISI | ChemPort |
42. Warrell DA (1996) Clinical features of envenoming by snake bites. In: Envenomings and Their Treatments (Bon C, Goyffon M, eds), Lyon: Fondation Marcel Mérieux, 63–76
43. Warrell DA (2004) Snakebites in Central and South America: epidemiology, clinical features and clinical management. In: The Venomous Reptiles of the Western Hemisphere (Campbell JA, Lamar WW, eds), vol II. Ithaca: Comstock, 709–761
44. Woodley DT, Chen M (2001) The basement membrane zone. In: The Biology of the Skin (Freinkel RK, Woodley DT, eds), New York: The Parthenon Publishing Group, 133–152
45. Wu WB, Chang SC, Liau MY, Huang TF (2001) Purification, molecular cloning and mechanism of action of graminelysin I, a snake-venom-derived metalloproteinase that induces apoptosis in human endothelial cells. Biochem J 357:719–728 | Article | PubMed | ChemPort |
46. You WK, Seo HJ, Chung KH, Kim DS (2003) A novel metalloprotease from Gloydius halys venom induces endothelial cell apoptosis through its protease and disintegrin-like domains. J Biochem 134:739–749 | Article | PubMed | ChemPort |

Acknowledgments

We thank Dr Catarina F.P. Teixeira (Laboratorio de Farmacologia, Instituto Butantan, Sao Paulo, Brazil) for providing the antigranulocyte mAbs, as well as the collaboration of Carmen Corella (Instituto Clodomiro Picado) in other aspects of the work. This study was supported by Vicerrectoría de Investigación (projects 741-A7-604 and 741-A7-502) and by the International Foundation for Science (IFS) (project F/4096-1). This work was presented by N. Jiménez in partial fulfillment of the requirements for the M.Sc. degree at Universidad de Costa Rica.