The venom of South American rattlesnakes inhibits macrophage functions and is endowed with anti-inflammatory properties

The injection of Crotalus durissus terrificus venom into the foot pad of mice did not induce a significant inflammatory response as evaluated by oedema formation, increased vascular permeability and cell migration. The subcutaneous injection of the venom, or its addition to cell cultures, had an inhibitory effect on the spreading and phagocytosis of resident macrophages, without affecting the viability of the cells. This effect was not observed when the venom was added to cultures of thioglycollate elicited macrophages, but it was able to inhibit these macrophage functions when the cells were obtained from animals injected simultaneously with the venom and thioglycollate. These observations suggest that the venom interferes with the mechanisms of macrophage activation. Leukocyte migration induced by intraperitoneal injection of thioglycollate was also inhibited by previous venom injection. This down-regulatory activity of the venom on macrophage functions could account for the mild inflammatory response observed in the site of the snake bite in Crotalus durissus terrificus envenomation in man.


Introduction
For many years, clinical observations have demonstrated that the venom of Viperid snakes are highly phlogistic. 1'2 A classical exception, found among the genus Crotalus, is the South American rattlesnake, Crotalus durissus terrificus (Cdt). The venom of this snake does not induce significant inflammatory reaction at the site of the bite in human victims. 1 '3 Moreover, recent studies have demonstrated that Cdt venom inhibits inflammatory signals like pain and oedema in some experimental models. <5 Based on this information, the present study was undertaken in order to investigate some parameters of the inflammatory response experimentally induced in mice by Cdt venom injection, as well as the effect of this venom on some activities of resident peritoneal and inflammatory macrophages, an important component of the inflammatory process. Material

and Methods
Animals.. Male mice from the Swiss NIH strain and Nu/Nu mice from the Balb/c strain, weighing [18][19][20][21][22] g, were used throughout this study. The animals were kept with free access to food and water. 18 Mediators of Inflammation Vol 5 1996 Venom: A pool of lyophilized venom was obtained from various adult specimens of Crotalus durissus terrificus snakes at the Laboratory of Herpetology of the Instituto Butantan. The venom solutions were prepared (w/v) in sterile saline, immediately before use. Histological analysis.. Histological sections of tissues of the abdominal wall of mice injected s.c. with 1.5tg (501al) of Cdt venom were analysed by using light microscopy. The tissues were collected 1, 2, 4 and 6 h after venom injection, fixed in Bouin fixative and processed for embedding in paraffin. Histological sections were stained with Haematoxilin-eosin. Vascular permeability evaluation: Vascular permeability was evaluated by extravasation of Evans blue in animals injected in the foot pad with 1.5 l.tg of Cdt venom. Animals were injected i.v. with a 2.5% solution of Evans blue (100 mg/kg of body weight), immediately after venom injection or at different times.
Twenty min after the injection of the dye the animals were anaesthetized and sacrificed. The paw was cut off in the tibio-tarsic junction and the fragments were placed in tubes containing 2.0ml of formamide (Merck) maintained for 48 h at 37C for complete extraction of the dye. ether and sacrificed by exsanguination by sec-(d) athymic nude mice, 4 days after s.c. injection tioning the cervical vessels. The peritoneal cavity of 1.51.tg Cdt venom; and (e) control groups was washed with 3 ml of cold phosphate-bufinjected with saline in the same volume and by fered saline (PBS), pH 7.4. After a gentle massage the same routes.
of the abdominal wall, the peritoneal fluid was collected in a plastic syringe. In some experi-Effect of venom on the inflammatory response ments 4% thioglycollate was injected into the induced by thioglycollate: The migration of cells peritoneal cavity and inflammatory peritoneal into the peritoneal cavity 4 h after the injection of cells were harvested 4 days later. Cell viability 4% thioglycollate (1.0ml i.p.)was investigated in was assessed by the Trypan blue exclusion test. groups injected s.c. I h earlier with 1.5 I.tg Cdt Total and differential counts of peritoneal cell venom, and compared to control groups injected suspensions were made as above, with saline.
Spreading of peritoneal macrophages: The Statistical analysis.. Data were expressed as the spreading capacity of peritoneal macrophages mean _ _ _ S.E.M. Student's t-test for impaired was estimated according to the method descri-samples and F-test 1 were used to compare bed by Rabinovitch and De Stefano. s Briefly, experimental and the respective control groups. 50 t.tl of cell suspension (1 x 10/ml) was placed Difference of results were considered statistically onto glass coverslips and left to adhere for 10 significant when p < 0.05. min at room temperature. The coverslips were washed with PBS and incubated in medium 199 Rul, (with Earle's salts and t-glutamine) at 37C for 1 h. Cells were fixed in a 2.5% glutaraldehyde Inflammatory response induced by the venom: solution and the index of spreaded cells was determined by examination under phase contrast Peritoneal cavity. The cell migration to the perimicroscopy. The spreading activity index was toneal cavity induced by Cdt venom was similar defined as the ratio between spreaded cells and to the control group in all times analysed 200 cells counted.
( Table 1).  Subcutaneous tissue. The histological analysis of the sims of inoculation of the Cdt venom showed a mild inflammatory response characterized by the presence of small cellular infiltrate and oedema. The cellular infiltrate was predominantly composed of polymorphonuclear cells and was a maximum 6 h after the venom injection (Fig. 1). vascular permeabilRy: The s.c. injection of 1.5 I.tg of venom in the foot pad of mice caused a slight increase in vascular permeability that started immediately after the injection and decreased thereafter (Fig. 2).

Effect of Cdt venom on peritoneal macrophage
,reading and phagocytosis in vitro: As shown in Fig.4, the addition of Cdt venom to culture medium at the concentration of 0.5 or 1.01.tg caused significant reduction in the spreading index of resident peritoneal macrophages when compared to a control group. The inhibitory effect on spreading and phagocytosis of the venom was not observed when it was added to cultures of inflammatory peritoneal macrophages elicited by thioglycollate (Fig. 4). The addition of serum from animals injected with Cdt venom to the culture medium did not alter the spreading index of the cells.

Effect of in vivo administration of Cdt venom on
peritoneal macrophage spreading and phagocytosis: The .i.p. or s.c. injection of the venom into normal mice induced a significant inhibition of peritoneal macrophage spreading after different time intervals (Fig. 5). A maximal inhibition was observed after 4 days of venom injection and to a lesser extent even after 16 days. The same inhibitory effect was obseeeed on the phagocytic activity of these cells. The injection of Cdt venom in athymic nude mice also determined a significant inhibition of peritoneal macrophage spreading when compared to a control group. Effect of the venom on inflammatory cell migration: The injection of Cdt venom (1.5 l.tg, s.c.) I h before the i.p. injection of thioglycollate, induced a significant reduction in the total number of cells in the peritoneal cavity 4h after thioglycollate injection when compared to control animals not injected with the venom ( Table 2).
The differential count of these cells showed a significant reduction in the number of polymorphonuclear cells.
Influence of venom administration on thioglycollate elicited activities of peritoneal cells: A significan reduction of both phagocytic and spreading index of peritoneal macrophages, was observed 4 days after the concomitant injection of venom (1.5 g s.c.) and of thioglycollate solution (i.p.) in mice (Fig. 6).
Results demonstrate that the Cdt venom injection into the subcutaneous tissue of mice did not induce a significant inflammatory response, confirming earlier clinical and experimental observa-111 tions.
The mild oedemmous response induced by the venom was not dosedependent, transient and comparable to the oedematogenic response induced by serotonin or histamine into the hind paw of rats. 12 The oedema induced by Cdt venom was completely abolished by pretreatment with an antihistamine drug. These findings are corroborated by the observation of Vargaftig et al. 13 that antihistamine drugs inhibit the increased vascular permeability induced by this venom. Taken together, these results suggest that histamine may be the main mediator of the acute phase of the mild inflammatory response induced by Cdt venom.
The main toxin found in Cdt venom is crotoxin which is composed of two subunits, crotapotin and phospholipase A2.14 Phospholipase A2 (PEA2) present in several snakes venoms is endowed with potent oedema forming activity. [15][16][17] However, the mild oedematous response of crude Cdt venom might be the result of a low enzymatic activity of PEA 2 when it is associated with crotapotin, is Furthermore, it has been demonstrated   that crotapotin has a marked inhibitory effect on the carrageenin-induced paw oedema, a multimediated inflammatory response. 5 The investigation of the Cdt venom on some of the macrophage functions was suggested by the observation that this venom has inhibitory effects in some inflammatory models 4'5 and by the fact that macrop_hages play a pivotal role in 19 2O these phenomena. The Cdt venom showed an inhibitory effect on the spreading and the phagocytic activities of peritoneal macrophages. At concentrations used in this experiment, this venom did not induce alterations in cell viability in vitro, as opposed to the effect of some cytotoxins from other snake species which, by affecting the plasmatic membrane permeability, may cause cell lysis. 2 These observations suggest a down-regulatory property of the venom on the investigated functions of macrophages, which was also shown in viva Although, in general, snake venoms are rapidly removed from circulation, 22'23 the inhibitory effects of Cdt venom observed on macrophage functions lasted for several days. The low turnover of Pieritoneal macrophages in mice (around 15 days) 4 could explain the long-lasting downregulatory effect of the venom on these cells. This low turnover of these cells might also account for the progressive reduction of the inhibitory effect induced by the venom.
On the other hand, the inhibitory effect of Cdt venom on macrophage functions might be a consequence of the generation by the venom of serum factors, since the envenomation by Crotalus durissus terrificus can cause renal failure, and the serum from rats with renal failure inhibits, in vitro, the phagocytic activity of peritoneal macrophages. 25 However, sera obtained from mice injected with Cdt venom did not alter, in vitro, the spreading ability of macrophages. The inhibitory effect of the venom on the spreading index of peritoneal macrophage from athymic Nu/Nu mice suggests that this phenomenon is not mediated by lymphocyte-released products.
The fact that, when added in vitro, Cdt venom did not inhibit the spreading and phagocytic capacities of macrophages elicited with thioglycollate but it was able to inhibit these actMties in macrophages obtained from animals simultaneously injected with Cdt venom (s.c.) and thioglycollate (i.p.) suggests that the venom interferes with the activation of macrophages but not with the activity of those already elicited. In fact, Cdt venom also inhibited the acute inflammatory response induced by thioglycollate, demonstrated by the reduction of inflammatory cell migration to the peritoneal cavity of animals injected with this drug plus venom. Preliminary data also indicate that Cdt venom significantly reduces the peritoneal cell migration induced by zymozan (Sousa e Silva and Cury, unpublished observations).
As inflammatory leukocyte migration is mediated by cytokines, and this includes those liberated by macrophages  there is reason to speculate that Cdt venom affects the inflammatory process by inhibiting the liberation of cytokines. This hypothesis is under investigation. snakes. In: Lee CY, ed. Handbook of Experimental Pharmacology, vol.