Differential Antivenom and Small-Molecule Inhibition of Novel Coagulotoxic Variations in Atropoides, Cerrophidion, Metlapilcoatlus, and Porthidium American Viperid Snake Venoms

Abstract

Within Neotropical pit-vipers, the Mexican/Central-American clade consisting of Atropoides, Cerrophidion, Metlapilcoatlus, and Porthidium is a wide-ranging, morphologically and ecologically diverse group of snakes. Despite their prevalence, little is known of the functional aspects of their venoms. This study aimed to fill the knowledge gap regarding coagulotoxic effects and to examine the potential of different therapeutic approaches. As a general trait, the venoms were shown to be anticoagulant but were underpinned by diverse biochemical actions. Pseudo-procoagulant activity (i.e., thrombin-like), characterized by the direct cleavage of fibrinogen to form weak fibrin clots, was evident for Atropoides picadoi, Cerrophidiontzotzilorum, Metlapilcoatlus mexicanus, M. nummifer, M. occiduus, M. olmec, and Porthidium porrasi. In contrast, other venoms cleaved fibrinogen in a destructive (non-clotting) manner, with C. godmani and C. wilsoni being the most potent. In addition to actions on fibrinogen, clotting enzymes were also inhibited. FXa was only weakly inhibited by most species, but Cerrophidion godmani and C. wilsoni were extremely strong in their inhibitory action. Other clotting enzymes were more widely inhibited by diverse species spanning the full taxonomical range, but in each case, there were species that had these traits notably amplified relatively to the others. C. godmani and C. wilsoni were the most potent amongst those that inhibited the formation of the prothrombinase complex and were also amongst the most potent inhibitors of Factor XIa. While most species displayed only low levels of thrombin inhibition, Porthidium dunni potently inhibited this clotting factor. The regional polyvalent antivenom produced by Instituto Picado Clodomiro was tested and was shown to be effective against the diverse anticoagulant pathophysiological effects. In contrast to the anticoagulant activities of the other species, Porthidium volcanicum was uniquely procoagulant through the activation of Factor VII and Factor XII. This viperid species is the first snake outside of the Oxyuranus/Pseudonaja elapid snake clade to be shown to activate FVII and the first snake venom of any kind to activate FXII. Interestingly, while small-molecule metalloprotease inhibitors prinomastat and marimastat demonstrated the ability to prevent the procoagulant toxicity of P. volcanicum, neither ICP antivenom nor inhibitor DMPS showed this effect. The extreme variation among the snakes here studied underscores how venom is a dynamic trait and how this can shape clinical outcomes and influence evolving treatment strategies.

Keywords: anticoagulant; antivenom; procoagulant; pseudo-procoagulant.

Figure 1

Ancestral state reconstruction of coagulotoxic venom effects on human fibrinogen (left) and plasma (right). The spontaneous clotting time of human plasma was 358.9 s. Cooler colors represent the inhibition of clotting in plasma (anticoagulant effect) and no clotting of fibrinogen. Warmer colors represent clotting in plasma and fibrinogen. The maximum machine reading time was 999 s. Bars represent 95% confidence intervals for the estimate at each node. The phylogeny used is based upon Alencar et al. [24] and timetree.org. The correlations between fibrinogen clotting and plasma clotting were congruent with direct action upon fibrinogen rather than the activation of clotting factors. However, for P. volcanicum, only plasma was clotted, with no direct action upon fibrinogen, which was suggestive of the activation of one or more clotting factors.

Figure 2

Thromboelastography traces showing Atropoides/Cerrophidion/Metlapilcoatlus venoms ability to clot human plasma in comparison to the spontaneous control. Samples showed to possess pseudo-procoagulant (Atropoides/Cerrophidion/Metlapilcoatlus) and procoagulant (P. volcanicum) venoms. Blue traces represent spontaneous controls, green traces represent thrombin control, and red traces represent samples incubated with venoms. SP = split point, which is the time taken until the clot began to form (min). R = time to initial clot formation, where formation is 2 mm + (min). A (amplitude) = clot strength (mm). Assays were performed in triplicate (n = 3) with data representing the mean ± SD.

Figure 3

Factor activation assay showing P. volcanicum’s procoagulant ability to activate FVII, FX, FXI, and FXII. Activation is shown as the relative percentage of zymogen converted to its active form compared with the active-zymogen positive control. Assays were performed in triplicate (n = 3) with data representing the mean ± SD.

Figure 4

Coagulation-cascade-factor-inhibition assays showing the inhibitory effects of Cerrophidion godmani, C. wilsoni, P. dunni, P. nasutum, P. ophryomegas, and P. yucatanicum on thrombin, FXa, prothrombinase, FIXa, and FXIa. Venoms that induced clotting could not be used for the prothrombinase inhibition assay as plasma was clotted before machine recording. Assays with venoms missing were due to a lack of stock, and such venoms were unable to be tested. Assays were performed in triplicate (n = 3) with data representing the mean ± SD. Values were statistically analyzed using one-way ANOVAs with a multiple-comparison tests compared with the negative control. Statistical significance from the negative control is indicated by * p < 0.1, ** p < 0.01, *** p < 0.001, or **** p < 0.0001.

Figure 5

Thromboelastography traces showing effects of two Cerrophidion and seven Porthidium species venoms that did not directly clot fibrinogen. A solution of fibrinogen was incubated for 30 min either alone (for the thrombin control) or with venoms, followed by the addition of thrombin. The green trace represents the 30 min fibrinogen incubation for all samples. Blue traces represent samples in which thrombin was added to fibrinogen in the absence of venom. Red traces represent samples in which venoms were incubated with a fibrinogen solution for 30 min, followed by the addition of thrombin. SP = Split point, i.e., the time taken until a clot began to form (min). R = time to initial clot formation, where formation is 2 mm + (min). A (amplitude) = clot strength (mm). Assays were performed in triplicate (n = 3) with data representing the mean ± SD.

Figure 6

Concentration–response curves showing the pseudo-procoagulant effects and the relative efficacy of the PoliVal-ICP antivenom. Curves represent the clotting time of both the venom-only assay (red) and the incubation-with-antivenom assay (blue). The bar graph represents the X-fold shift value of each venom when incubated with antivenom. Calculated values represent antivenom neutralization, where 0 is no neutralization and >0 indicates neutralization. Assays were performed in triplicate (n = 3), excluding P. porrasi, which was in duplicate (n = 2) due to a shortage of venom supply. Data represent the mean ± SD. Note: some data points have error bars that are smaller than the size of the symbol.

Figure 7

Concentration–response curves showing the anticoagulant effects of two Cerrophidion venoms and the relative efficacy of the PoliVal-ICP antivenom in neutralizing FXa-inhibition and the inhibition of the prothrombinase complex. Curves represent the venom-only assay (red) and the incubation-with-antivenom assay (blue). Bar graphs indicate the percentage drop-in venom activity of venom incubated with antivenom relative to the assay of venom incubated without antivenom. Assays were performed in triplicate (n = 3) with data representing the mean ± SD. Note: some data points have error bars that are smaller than the size of the symbol.

Figure 8

Concentration–response curves showing the procoagulant effect of P. volcanicum venom on plasma and the relative efficacy of three SVMP inhibitors and the PoliVal-ICP antivenom in neutralizing the venom effects. Curves represent the venom-only assay (red) and the incubation-with-inhibitor/antivenom assay (blue). Assays were performed in triplicate (n = 3) with data representing the mean ± SD. Note: some data points have error bars that are smaller than the size of the symbol.