Variation in the distribution and venom composition of Bothrops species.

A) Geographical distribution of the Bothrops species relevant to this manuscript. The distribution is based on the ICUN Red List accessed in April 2025 presented using QGIS 3.4. The images of B. asper, B. atrox, B. jararaca, B. moojeni and B. neuwiedi are held under all rights reserved copyright and have been published with permission by Wolfgang Wüster, while the image of B. lanceolatus is published with permission by Jonathan Florentin. The image of B. alternatus was accessed via Wikimedia commons and is held respectively under Creative Commons Attribution-Share Alike 2.0 Generic licence by Cláudio Timm. B) A summary of the published percentage venom compositions for the three key toxins of relevance to this manuscript; snake venom metalloproteinases (SVMP) in beige, snake venom serine proteases (SVSP) in orange, and snake venom phospholipase A2 (PLA2) in green. The extracted data (Supplementary Table 1) vary from an individual to large pools of specimens including captive and wild caught. The data is summarised from the following number of publications per species; B. alternatus (n=2), B. asper (n=1), B. atrox (n=6), B. jararaca (n=3), B. lanceolatus (n=2), B. moojeni (n=2) and B. neuwiedi (n=1). The data is displayed as a box and violin plot (interleaved low-high with line at the mean) using Prism software version 10 (GraphPad).

In vitro profiling of venom samples from seven different Bothrops species.

A) Protein profiles of the venom protein components per species. Whole venom (5 µg per lane) was prepared under denaturing and reducing conditions, then separated by SDS-PAGE using a 4-20% gel. The gel was then stained with Coomassie blue and destained to show all proteinaceous components. M = molecular weight marker, and approximate masses of the markers are given in kDa. B-D) enzymatic activity of the seven Bothrops species in toxin specific assays. The results were analysed using area under the curve (AUC) in seconds and plotted using Prism version 10 software (GraphPad). A) Snake venom metalloproteinase (SVMP) activity using fluorogenic substrate ES010 read at an excitation wavelength of 320 nm and an emission wavelength of 405 nm over 108 minutes (n=3, ±SE), with 1 µg of venom per reaction. C) phospholipase A2 (PLA2) activity using commercially available secretory PLA2 kit (Abcam) read at an absorbance wavelength of 405 nm over 12 minutes (n= ≥2, ±SE), with 100 ng of venom per reaction. D) snake venom serine protease (SVSP) activity using commercially available chromogenic substrate S-2288 read at an absorbance of 405 nm over 35 minutes (n=4 ±SE) with 1 µg of venom per reaction.

In vitro coagulation profile of seven different Bothrops species and inhibition of procoagulant activity by small molecule drugs.

A) The coagulopathic profile of the seven Bothrops species at a comparable 100 ng dose in bovine plasma over 25 minutes at an absorbance of 595 nm (n=2, ±SD). B) Coagulation profile in the absence of venom for SVMP inhibitors (DMPS, a metal chelator, and the MMP inhibitors marimastat and prinomastat), the PLA2 inhibitor varespladib and the serine protease inhibitor nafamostat. The profiles indicate no direct coagulopathy for any of the inhibitors except for nafamostat which has a strong anticoagulant profile at this dose (n=2, ±SD). C-I) The percentage inhibition in the coagulation assay of the small molecules (excluding nafamostat due to the inherent anticoagulant activity) in each of the seven Bothrops venoms, tested at 5 µM (n=2, ±SE). The adjusted dose for each venom indicated in the graph title was selected to provide comparable profiles for all seven venoms. J) Dose response testing was performed for the four inhibitors in the coagulation assays and EC50s (nM) calculated using Prism v10 software (GraphPad) as presented this table (n=2, ±95% CI).

In vitro inhibition of seven different Bothrops species by small molecule drugs.

The enzymatic activity of the seven Bothrops species with relevant inhibitors in the three toxin specific assays are presented, with the results analysed using area under the curve (AUC) and plotted using Prism version 10 software (GraphPad). A) Dose response testing was performed in the SVMP and PLA2 assays using varespladib for the latter, but a wider panel of MMPis and metal chelators with prior evidence of SVMP inhibition in the SVMP assay. This allowed for the calculation of EC50s (with 95% CI shown) using Prism v10 software (GraphPad) which are displayed in this table (n=2). B) Snake venom metalloproteinase (SVMP) activity and inhibition by the MMP inhibitor, marimastat (AUC over 178 minutes, n=4, ±SE). C) phospholipase A2 (PLA2) activity and inhibition by the PLA2 inhibitor, varespladib (AUC over 13 minutes, n= ≥3, ±SE). D) snake venom serine protease (SVSP) activity and inhibition by the SP inhibitor, nafamostat (AUC over 35 minutes, n=4, ±SE).

In vitro coagulation profile of two coagulopathic Bothrops species and inhibition by combination therapy of two small molecule drugs.

The effect of Bothrops jararaca (A-B, purple) and Bothrops atrox (C-D, green) with and without small molecule inhibitors is presented alongside the direct effect of these inhibitors (E and F) in the coagulation assay. (A and C) The kinetic curves demonstrate the procoagulant effect of both venoms, compared to normal clotting control (black circles). The SVSP inhibitor nafamostat (orange triangles) fails to rescue the procoagulant effect, whilst marimastat (5 µM, white triangles) fully inhibits the procoagulant effect of B. atrox and partial rescue of B. jararaca (n=6, ± SD). (B and D) Percentage inhibition of coagulopathy by Bothrops species in single treatment by small molecule inhibitors (n=6, ± SE) matched that of the kinetic curves for the top 5 µM dose of marimastat with subsequent lower doses slightly reducing the percentage inhibition, nafamostat resulted in minor inhibition at any doses. Combination testing of marimastat and nafamostat, for both venoms appears to reflect an additive effect, however in the no-venom control (E, kinetic curve; F, % inhibition), nafamostat shows a dose-dependent increase beyond 100% in the absence of venom, demonstrating that it is directly anticoagulant at higher doses. Marimastat has no effect on plasma clotting, indicating that the effects seen in graphs (A to D) are through inhibition of venom.

Thromboelastography (TEG) profiles of two representative Bothrops species and inhibition by the SVMP inhibitor marimastat.

A and B) Thromboelastography profile of respectively B. atrox and B. jararaca using an XY plot of clotting time vs clotting strength. Both venoms (0.6 µg per 300 µL reaction) have strong procoagulant activity (circles) compared to the no venom controls (black diamond) or drug only control (grey diamond). Marimastat inhibits the procoagulant activity of both venoms (squares) (n >3). C) The parameters reported by the ROTEM instrument are displayed including the clotting time (CT; seconds), maximum clot firmness (MCF; mm), clot formation time (CFT; seconds), alpha angle (°) and amplitude at 5 minutes (A5; mm), each is the mean of the triplicate results ± the related range.

Survival curves of chicken embryos dosed with B. atrox venom with and without representative small molecule drugs.

Groups of 6-day post-fertilisation chicken egg embryos were dosed on to the vitelline vein with either PBS (negative control) or B. atrox venom (20 µg) with or without a subsequent dose of 0.5 µg, 1 µg and 5 µg of A) marimastat, B) DMPS and C) nafamostat. All treatment group sizes were n=5, the PBS control contained an n=12, while the venom only control contained an n=20. D) Representative pathology images of chicken embryos 1 hour post-dosing with B. atrox venom with and without 5 µg of the inhibitory drugs, showing varying degrees of protection against the vascular damage shown in the venom only control (see also Supplementary Figure 1).