INTRODUCTION
Crotalus snakebites cause serious problems and can be fatal unless adequately treated. The high lethality occurs due to the frequency with which Crotalus envenomation causes acute renal failure, one of the major causes of death due to snake bites1. In addition to neurotoxicity, myotoxic action and clotting disorders may cause micro-bleeding, which characterizes the systemic effect of envenomation2–3. The local effects that can be observed are mild pain and local or regional paresthesia that may persist for different periods of time, slight edema, and erythema at the bite site4.
Protein isoforms present in C. durissus venom are also of great clinical importance, due to the fact that the venom pool used for antivenom production may be inadequate. This leads to reduced immunogenicity in animals and results in a product that cannot adequately neutralize the clinical manifestations of patients bitten by these snakes5–6.
This study aims to compare the toxic effects (local and systemic) induced by the venom of Crotalus subspecies, C. durissus terrificus, C. durissus cascavella, and C. durissus collilineatus, in individuals bitten by Crotalus snakes. Moreover, this study demonstrates that the variations found can be related to differences in neutralization rates by antivenom actions against the venoms from different subspecies, which is consistent with the results published by Boldrini-França et al.6 and Oliveira et al.7.
METHODS
Venoms and Animals
Cdt, Cdcolli, and Cdcasc venoms were acquired from Instituto Butantan, São Paulo-SP and were maintained at -20ºC in the Amazon Venom Bank at CEBio-UNIR-FIOCRUZ-RO (licenses: CGEN/CNPq 010627/2011-1 and IBAMA 27131-2). All Crotalus sp. venoms used were from a pool of venoms from adults male and female snakes from different parts of Brazil (carried by IBAMA, ONGs, firemen, or physical people).
Male Swiss mice (18-20 g) were housed in temperature-controlled rooms and received water and food ad libitum until used (approved protocol number 2012/09 from the FIOCRUZ-RO Animals’ Ethics Committee).
Venom biochemical characterization
Cdt, Cdcasc, and Cdcolli venoms’ chromatographic profile was produced by molecular exclusion chromatography (Sephadex G75) in an FPLC (Akta Purifier System/GE Healthcare®). The method was performed according to Bercovici et al.8 in which about 35 mg of venom was suspended in 1 mL of 0.05 M ammonium formate buffer, pH 3.5. The peaks obtained were analyzed by SDS-PAGE on 12.5% and/or 15% polyacrylamide gels. The proteins’ concentrations were measured using the Bio-Rad DC Protein Assay method (BIO-RAD).
Proteolytic activity
150 µL of 2% azocasein solution (substrate) was added to 7 µL of each venom (20 µg of Cdt, Cdcolli, or Cdcasc) in a 96-well plate (substrate) and incubated for 1 h at 37ºC. Subsequently, the reaction (Azocasein+venom) was stopped by adding 150 µL of 20% trichloroacetic acid (TCA). After 30 min, the samples were centrifuged at 180 xg for 10 min. Next, 100 µL of the supernatant was added to a 96-well plate, and 100 µL of 500 mM NaOH were added. The absorbance at 440 nm was monitored using spectrophotometer (Biotek)9. Bothrops’ venom (from Amazon Venom Bank at CEBio-UNIR-FIOCRUZ-RO) was used as a positive control.
Phospholipase activity
Phospholipase activity was determined using the method described by Holzer and Mackessy10 adapted to 96-well plates. Buffer containing the chromogenic substrate 4-nitrophenyl (3-octanoyloxy) benzoic acid (4N3OAB) was added to 10 µL of each venom (20 µg Cdt, Cdcolli, or Cdcasc) or water (negative control). The solution was analyzed in a spectrophotometer at 425 nm over 30 min for 30 s intervals10.
LAAO Activity
For LAAO activity, the method described by Torii11 adapted to a 96-well plate was followed according to Pontes et al.12. In brief, each venom (10 μg) was added separately to the reaction mixture containing horseradish peroxidase (50 μg/mL), 100 μM l-leucine, and 10 μM 3’3’diaminobenzidine in 100 mM Tris-HCl buffer (pH 7.8). The reaction was incubated at 37°C for 30 min. The reaction was stopped using 10% citric acid and the absorbance was measured on a spectrophotometer at 490 nm.
Paw edema assay
Groups of 5 mice were inoculated with a subplantar injection of 30 µL of 100 μg/kg of Cdt, Cdcolli, or Cdcasc diluted in 150 mM sterile physiological saline in the right hind paw. As a control, sterile physiological saline (30 μL) was injected into the contralateral paw. Paw volume was measured immediately before sample injection (basal time 0 h) and at different time intervals thereafter (0.5, 1, 3, 6, 9, 12, and 24 h). Paw volume was measured using a hydroplethysmometer (Ugo Basile). Results were expressed as percentage paw volume increase in relation to the control paw13.
Myotoxic, Hepatotoxic, and Nephrotoxic activities
Groups of 5 mice received an intramuscular (i.m.) injection of 30 µL of Cdt, Cdcolli, or Cdcasc venom (100 µg/kg of venom diluted in 150 mM sterile physiological saline) in the gastrocnemius muscle. Control group animals received 30 µL (150 mM sterile physiological saline) in the same conditions. After 3, 6, 9, 12, and 24 h, mouse blood was withdrawn from their orbital plexus, dispensed into heparinized tubes, and centrifuged at 2205 xg for 5 min according to Teixeira et al. (2018)13.
Myotoxicity activity was evaluated by measuring creatine kinase (CK), creatine kinase MB isoenzyme (CK-MB), and lactate dehydrogenase (LDH). Hepatotoxic activity was evaluated by measuring alanine transaminase (ALT), aspartate transaminase (AST), gamma-glutamyl transferase (GGT), and alkaline phosphatase (AP) activity. Kidney function was evaluated by measuring plasma creatinine and urea biochemical parameters and total urine proteins and calcium. All determinations were conducted using commercial diagnostic kits purchased from Labtest Diagnostica SA (Brazil).
Coagulation tests in vivo
Groups of 5 mice received in the gastrocnemius muscle an i.m. injection of 30 µL of Cdt, Cdcolli, and Cdcasc (100 µg/kg of venom diluted in 150 mM sterile physiological saline). Control group animals received 30 µL (150 mM sterile physiological saline) in the same conditions. After 3 h, mouse blood was drawn from the inferior vena cava, dispensed into citrated tubes, and centrifuged at 2205 xg for 5 min in order to obtain platelet-poor plasma. Prothrombin time (TAP) and activated partial thromboplastin time (TAPP) were determined using Hemostasis (Labtest, Brazil).
Myographic Study
Cdt, Cdcolli, and Cdcasc venoms neuromuscular activities were assessed in mice phrenic-diaphragm preparations in accordance with the method previously described by Gallacci and Cavalcante14. The amplitudes of indirect and direct twitches were measured at 150 and 420 min.
RESULTS
Biochemical Characterization
Venoms were fractioned by molecular exclusion chromatography using the Sephadex G-75 (GE healthcare®, 10 x 60 cm). The C. durissus venoms chromatographic profile analysis showed three peaks, Cdt-I, Cdt-II, and Cdt-III, for Cdt (Figure 1A); three peaks, Cdcolli-I, Cdcolli-II, and Cdcolli-III, for Cdcolli (Figure 1B); and two peaks, Cdcasc-I and Cdcasc-II, for Cdcasc (Figure 1C).
After electrophoresis, convulxin (Cvx) and gyroxin (Grx) were visualized in the lanes associated with Cdt-I, Cdcolli-I, and Cdcasc-I with approximate molecular weights (MW) of ~72 kDa for Cvx and ~30 kDa for Grx. Moreover, crotoxin (Ctx) was found in the Cdt-II, Cdcolli-II, and Cdcasc-II lanes at ~24 kDa. However, crotamine (Ctm) alone was visualized in the Cdt-III and Cdcolli-III lanes at ~4.8 kDa, but was not observed in Cdcasc venom (Ctm-negative) (Figures A2, B1, and C1). It was observed that the estimated protein content for Cdt-I, Cdcolli-I, and Cdcasc-I peaks were 2%, 6%, and 18%, respectively. For Cdt-II, Cdcolli-II, and Cdcasc-II the protein contents were 81%, 67%, and 82%, respectively. Additionally, the estimated protein contents for Cdt-III and Cdcolli-III were 16% and 27%, respectively. The results above apply only to Cdt and Cdcolli because Cdcasc was found to be devoid of crotamine.
Enzymatic Characterization
Results suggested that Cdt, Cdcolli, and Cdcasc venoms do not exert a proteolytic effect on azocasein compared to the negative control (Figure 2A).
Cdcolli and Cdcasc venoms (20 µg) had significant LAAO activity compared to that of the control. This result was determined by the peroxidase reaction, wherein H2O2 was produced. However, Cdt venom showed no LAAO activity (Figure 2B).
Cdcolli venom (20 µg) demonstrated low phospholipase activity on the substrate 4N3OAB at the concentration used. However, Cdcasc and Cdt venoms, using the same substrate and the same concentration, displayed significant catalytic activity compared to control (Figure 2C).
Local and Systemic Effect Characterization
Cdcolli and Cdcasc venoms induced paw edema increases of 76.7% and 40%, respectively, after 1 h. Additionally, Cdt venom induced a paw edema increase of 33.3% after 3 h (Figure 3).
Mice blood CK levels are depicted in Figure 4A. Cdcolli and Cdcasc induced a significant increase in total-CK liberation compared to control animals 3 h after inoculation. After 6, 9, and 12 h, all venoms induced a significant total-CK increase compared to that of the control (182.3 U/L). Total-CK liberation induced by Cdcolli was statistically different from Cdt from 3 to 12 h after inoculation. However, total-CK liberation induced by Cdcasc was statistically different from Cdt just after 12 h of inoculation. In Figure 4B, animals inoculated with venom had significantly higher MB-CK levels compared to control animals after 3 h of inoculation. At this time, animals inoculated with Cdcolli differed statistically from those inoculated with Cdcasc. This was not observed in mice envenomated with Cdt venom after 6 h of inoculation. After 9 h, all envenomated animals had basal MB-CK levels. However, after 12 h, Cdcolli and Cdt venoms injected in mice induced a significant increase in MB-CK levels compared to that in control animals. Additionally, 24 h after inoculation, all venoms induced a significant MB-CK increase compared to the controls.
Figure 4C shows that all venoms studied increased LDH levels 3 and 6 h after inoculation compared to control animals. However, the animals inoculated with Cdt alone differed statistically from those inoculated with Cdcasc. Moreover, Cdcolli and Cdt venoms increased LDH levels after 12 h compared to controls.
As shown in Figure 4D, none of the venoms changed ALT levels after 3 h compared to control animals. However, Cdcasc and Cdcolli venoms increased ALT levels compared to the controls 6 and 9 h after inoculation, respectively. After 9 h, animals envenomated with Cdcolli differed statistically from those with Cdt and Cdcasc. After 12 h past Cdcolli and Cdt inoculation, ALT levels significantly increased compared to Cdcasc and the control group. Additionally, 24 h after Cdcolli, Cdcasc, or Cdt injection, ALT levels significantly increased compared to controls.
In Figure 4E, significantly higher levels of AST can be seen in animals injected with Cdcasc, Cdcolli, or Cdt compared to those in control animals after 3, 6, 9, and 24 h. However, only animals injected with Cdcolli venom did not display an increase in AST levels 12 h after inoculation. Animals inoculated with Cdcasc differed statistically from those inoculated with Cdcolli and Cdt 3 and 12 h after inoculation. At 9 h after Cdcasc inoculation, Cdcasc action differed from animals inoculated with Cdcasc and Cdt. However, animals inoculated with Cdt and Cdcasc differed from those inoculated with Cdcolli 24 h after inoculation.
The animals injected with Cdt venom showed an increase in ALP levels after 6 h (Figure 4F). Animals injected with Cdcolli venom induced higher ALP levels compared to control animals 9, 12, and 24 h after inoculation. ALP levels in animals inoculated with Cdcolli differed statistically from those inoculated with Cdt and Cdcasc 12 and 24 h after injections.
Figure 4G shows that 3 h after inoculation, only animals injected with Cdcolli venom induced a GGT increase compared to the control and Cdt animals. Moreover, after 6 h, all animals injected with the Crotalus subspecies venoms presented a significant increase in GGT levels. At this time, the GGT levels of animals inoculated with Cdt differs from those animals inoculated with Cdcolli and Cdcasc. After 9 h, mice injected with Cdcolli and Cdcasc venoms had a reduced GGT levels compared to control and Cdt mice.
Figure 4H shows that 3 h after Cdcasc and Cdcolli mice envenomation there was a significant increase in urea levels compared to control animals. Only Cdcolli venom induced a significant increase in urea levels 6 h after envenomation. Additionally, 9 h after Cdt and Cdcolli injection there was a significant increase in urea levels compared to control animals.
For urine urea levels at 3, 6, 12, and 24 h, all Crotalus venoms did not induce urea liberation in mouse urine. However, Cdcolli induced a significant increase in urine urea levels at 9 h after envenomation (Figure 4I). Cdcolli and Cdt induced a significant increase in serum creatinine compared to Cdcasc animals 9 h after envenomation (Figure 4J).
Results showed that 3 h after Crotalus venom injection there was no increase in urine creatinine levels (Figure 4K). In contrast, after 6 h, mice injected with Cdcasc venom had significant increases in urine creatinine levels. At 9 h post injection, increased urine creatinine levels were significant in mice injected with Cdcolli or Cdt venom. Additionally, a significant increase in urine protein levels was observed 3, 6, 9, and 12 h after venom injection (Figure 4L).
Cdcolli venom induced an increase in prothrombin time of 120 sec, which is significantly different from the control group’s prothrombin time of 28 sec. However, Cdt and Cdcasc venom did not increase coagulation time. With respect to activated partial thromboplastin time, Cdcolli and Cdcasc venom induced an increase of 240 sec, statistically different from the control group (92.5 sec, Figure 4N).
Cdt, Cdcolli, and Cdcasc venoms induced a time- and concentration-dependent blockade of indirectly evoked twitches (Figures 5A and 5B), and at the higher concentration studied (5 μg/mL), they also paralyzed the directly evoked twitches (Figure 5C). As shown in Table 1, the times required for 50% blockade (t50) of indirect twitches were significantly lower than those necessary to blockade the direct contractions. While the blockade of indirect contraction is an unequivocal indicator of a neurotoxic activity, the blockade of direct contractions frequently denotes a myotoxic activity affecting muscle contractility.
Experimental Group | Indirect | Direct | |
---|---|---|---|
1 μg/mL | 5 μg/mL | 5 μg/mL | |
C.d. terrificus Venom | 84.36 ± 8.48 (n=4) | 42.70 ± 2.3 (n=5) # | 56.80 ± 6.43 (n=4) * |
C.d. collilineatus Venom | 71.20 ± 6.35 (n=6) | 42.24 ± 0.88 (n=4) # | 64.26 ± 4.01 (n=4) * |
C.d. cascavella Venom | 100 ± 7.79 (n=4) | 57.96 ± 4.43 (n=6) # | 249.6 ± 11.7 (n=4) * |
#Indicates differences in t50 of indirectly evoked preparations exposed to the same venom at different concentrations (1 μg/mL vs 5 μg/mL). *Indicates differences in t50 between directly and indirectly evoked preparations exposed to the same venom and concentration (5 μg/mL).
DISCUSSION
Molecular exclusion chromatography of Crotalus durissus venom identified four major toxins: Cvx15, Gvx16, Ctx17, and Ctm18. Therefore, C. durissus venom has a variable composition, and Ctm may or may not be present in it19. Venom composition is directly associated to age, sex, captivity, and the individual glands of a snake20–22. Ontogenetic and seasonal variations also contribute to molecular diversity and venom complexity7,23.
Oliveira et al.7used proteomic and functional analyses of 22 Cdcolli individuals’ venoms to explore their qualitative and quantitative variations. Moreover, these authors found that different Cdcolli venoms caused envenomings with different changes in biochemical and immunological parameters.
Lourenço et al.24 found Ctm venom heterogeneity but did not observe a statistical difference in C. durissus venom proteolytic activity. The findings of this study are in accordance with our data.
Phospholipase A2 (PLA2) can be responsible for edema induction, since it is directly related to envenoming pathophysiology that accounts for various local and systemic disorders25–27. Furthermore, Cdt, Cdcolli, and Cdcasc venoms were uniform in relation to phospholipase activity with the substrate 4N3OBA. The capacity of some PLA2s to recognize and act on specific targets can explain these differences. The same results were obtained by Santoro et al.28 with Cdcasc and Cdcolli.
Unlike other viperid venoms, Crotalus venoms do not induce significant inflammatory reactions at the bite site in animals or humans29–30. However, there was a report of an edematogenic response induced by Cdt venom that was not dose-dependent and had a fast and transient course41. Some studies have reported that PLA2 induced edema, an effect that in some cases are dependent on PLA2s binding to specific membrane proteins31.
Other than the association with edema induction, other studies have emphasized the participation of PLA2s in myotoxicity caused by crotalic envenoming. A large part of this action at least is due to two components: Ctm and Ctx32–34. The toxicity induced by Ctx is generated by the CB subunit, which is a PLA235.
Several biological activities including cytotoxicity; mild myonecrosis; apoptosis induction; platelet aggregation, induction, and/or inhibition; as well as hemorrhagic, hemolytic, edematogenic, antibacterial, antiproliferative, antiparasitic, and anti-HIV activities36have been attributed to LAAO. In 201537, a very small amount (1.8% of the total venom) of the first LAAO from Cdt yellow venom, Bordenein-L, was isolated. Few studies have been conducted to determine the mechanisms of action of LAAOs in the induction of edema compared to other classes of snake toxins. Studies investigating different toxins revealed that the action of these proteins is related to the release of inflammatory mediators such as histamine, prostaglandin, kinins, and serotonin38. However, the edematogenic activity of LAAOs does not seem to be mediated by the same mechanisms described for other toxins, since these enzymes do not lose their edematogenic activity in the presence of antihistamines.
There are essentially two clinical models for myotoxicity: local and systemic. For instance, rhabdomyolysis constitutes a generalized muscle breakdown and causes myoglobin and CK increases in circulation that may lead to renal dysfunction, which is one of the envenoming characteristics of crotalics2,28,39.
Results showed that all Crotalus subspecies can induce alterations in CK (Total and MB) levels in different ways. The variance between the extravasation of Total-CK induced by crotalic venoms can be explained by the venom’s PLA240 and LAAO activity. Cdcolli had the highest activity in the muscular system, which differs from results found by Santoro et al.28that showed that Cdt had its highest activity in muscles. This could be the result of the two toxins involved in myotoxicity, Ctm and Ctx, working synergistically. Both are found in high concentration in Cdcolli venom.
The significant increase in MB-CK circulation caused by Cdt and Cdcolli venoms can be explained by PLA2 action on the myocardium muscle, the occurrence of early lesions, and the release of mediators that contribute to delayed injury. However, Cupo et al.,41observed an absence of clinical signs of myocardiotoxicity and the presence of normal serial ECG and echocardiography. Siqueira et al.42 reported myocardium damage in a victim bitten by a C.d. terrific snake. C. durissus venom’s toxicity in heart muscle is controversial and poorly understood. It was observed by electrocardiogram changes43 and acute myocardial infarction41 in Cdt bitten patients. TOTAL-CK and MB-CK results in the present study are in agreement with the works mentioned above in which CK increases were verified in the first 2 to 3 h after venom inoculation. Additionally, the myotoxic activity results of our work are in line with the data of Saraiva et al.44. They measured mice plasma CK levels 3 h after Cdt venom intramuscular injection in the right gastrocnemius muscle. According to Barraviera et al.45 in their retrospective study of Crotalus bitten victims, 100% of the analyzed patients presented increased CK levels.
Cupo et al.41,43 showed that the LDH (heart fraction) concentrations in Cdt bitten patients were higher than those of other LDH isoforms. Rowlands et al.46 performed autopsies on Pseudechis australis envenoming victims and observed rhabdomyolysis with necrosis foci in the myocardium. De Siqueira et al.42 reported damage to the myocardium after Cdt envenoming. Autopsies showed diffuse edema myocytolysis and rare micro-infarcts foci.
Accordingly, it was suggested that the Cdcasc venom induced liver injury 6 h after inoculation, whereas Cdcolli and Cdt venoms induced the same reaction after 9 h and 12 h, respectively. AST levels also changed after individual inoculation of these three venoms. These differences in activities triggered by the venoms can be attributed to Ctx actions and its PLA2 subunit and/or LAAO action on specific sites in the hepatocytes membranes47.
Elevation of ALT and AST may be related to side effects caused by biological factors released by tissue injury. ALP and GGT alterations are shown here reinforcing the hepatotoxicity induced by Crotalus envenoming. Barraviera et al.48 showed a positive correlation between bromsulphalein retention and ALT serum levels as well as AST and ALT serum increase in Cdt bitten patients. The authors proposed that these alterations were associated with liver dysfunction. In other research, Barraviera et al.45 performed an anatomopathological exam on a patient who died and were diagnosed with extensive hepatic necrosis. However, França et al.47 measured rat serum levels of ALT, AST, ALP, and GGT after Cdt venom inoculation and showed acute hepatotoxicity. Snake venom from other families, such as Elapidae Nana naja venom, also induces liver changes49. However, the mechanisms involved in hepatic injury are still not understood, which makes them an important question in the field.
Given the creatinine excretion reduction caused by envenoming, it can be inferred that it is related to bloodstream retention. However, the presence of albumin in urine indicates kidney injury because this protein has a high molecular weight and is not able to cross the glomerular membrane in physiological conditions. Reduction or loss of kidney function, known as acute renal failure, is the main complication of crotalic envenoming. This condition can be attributed to systemic manifestations, dehydration, and especially myotoxicity. Furthermore, it can develop into a severe renal ischemia, leading to kidney function loss2,39,50. However, Amora et al.51 used isolated rat kidney to find that Ctx and PLA2 were involved in this process, since the renal effects observed would be due to the venom components’ synergistic action.
Another systemic manifestation of crotalic envenoming is the alteration of the blood coagulation system52. These characteristics may be attributed to the enzymes present (serineproteases and metalloproteinases) that primarily affect the hemostatic system53. The difference in these enzymes expression patterns in these three subspecies might explain the different results shown in coagulation.
De Oliveira et al.54 reported on the activity of fibrinolytic enzymes, mainly in the fibrinogen Bβ and Aα subunits in the Cdt and Cdcolli venoms. This is considered a characteristic of the serineproteases present in these venoms. Accordingly, the blood incoagulability observed in severe cases of C. durissus bites are derived from fibrinogen consumption52.
The studied venoms were able to induce a contraction blockade. The blockade of indirect contraction is an unequivocal indicator of a neurotoxic activity, the blockade of direct contractions usually denotes myotoxic activity affecting muscle contractility55.
C. durissus subspecies venoms contain large amounts of Ctx6,28,56–58, a potent β-neurotoxin that induces neuromuscular transmission blockade and progressive muscle paralysis59. In addition, Ctx also induces in vivo and in vitro myotoxic activity, which may be underlying the blockade of direct muscle contractions induced by crotalic venoms60–62. Prior to the establishment of the neuromuscular blockade, Cdt and Cdcolli venoms induce initial facilitation of muscle contractions. This effect can be attributed to the presence of Ctm in both venoms28,58. This myonecrotic toxin binds to voltage-sensitive Na+ channels on the skeletal muscle sarcolemma, leading to a large influx of Na+ ions, which causes depolarization, strong contraction, injury, and myonecrosis63–65. In contrast, the absence of this molecule in Cdcasc venom may explain the t50 for direct evoked twitches, which was six times higher than those observed for the other venoms (Table 1).
Taken together, the data obtained in this study created a new view of the intraspecific variation of local and systemic effects caused by Cdt, Cdcolli, and Cdcasc venoms. Based on this evidence and the changes in the liver, kidney, muscle systems, and coagulation induced by these envenoming processes, besides variations in protein and enzymatic composition, we can evaluate the differences between local and systemic effects caused by subspecies of C. durissus. This highlights the clinical and biochemical effects produced by their respective venoms. The differences in some pharmacological activities observed in our study are in accordance with published data6,66,44.