Doxycycline-Mediated Inhibition of Snake Venom Phospholipase and Metalloproteinase

ABSTRACT

Introduction

Warfighters are exposed to life-threatening injuries daily and according to the Joint Trauma System Military Clinical Practice Guideline—Global Snake Envenomation Management snakebites are a concerning threat in all theaters of operation. Snake venom is a complex mixture of toxins including phospholipases A₂ (PLA₂) and snake venom metalloproteinases (SVMP) that produce myotoxic, hemotoxic, and cytotoxic injuries. Antibody-based antivenom is the standard of care but new approaches including small-molecule inhibitors have gained attention in recent years. Doxycycline is an effective inhibitor of human metalloproteinases and PLA₂. The enzymatic activities of 3 phylogenetically distinct snakes: Agkistrodon piscivorus, Naja kaouthia, and Daboia russelii were tested under inhibitory conditions using doxycycline.

Materials and Methods

Enzymatic activity of PLA₂ and SVMP was measured in N. kaouthia, D. russelii, and A. piscivorus venom alone and with doxycycline using EnzChek Phospholipase A₂ and Gelatinase Assay Kits. A 1-way ANOVA with Tukey’s post-hoc test was used to conduct comparative analysis. The median lethal dose of the venoms, the effective dose of doxycycline, and creatine kinase (CK) inhibition levels were measured in a murine model with adult Bagg Albino (BALB/c) mice using intramuscular injections. Median lethal and effective doses were determined using Spearman-Karber’s method and a 1-way ANOVA with Tukey’s post-hoc test was used to compare CK inhibition levels.

Results

Phospholipases A₂ activity was reduced to 1.5% to 44.0% in all 3 venoms in a dose-dependent manner using 0.32, 0.16, and 0.08 mg/mL doxycycline when compared to venom-only controls (P < .0001) (Fig. 1A). Snake venom metalloproteinases activity was reduced to 4% to 62% in all 3 venoms in a dose-dependent manner using 0.32, 0.16, and 0.08 mg/mL doxycycline (P < .0001) (Fig. 1B). The lethal dose (LD₅₀) values of the venoms in the murine model were calculated as follows: A. piscivorus = 20.29 mg/kg (Fig. 2A), N. kaouthia = 0.38 mg/kg (Fig. 2B), and D. russelii = 7.92 mg/kg (Fig. 2C). The effective dose (ED₅₀) of doxycycline in A. piscivorus was calculated to be 20.82 mg/kg and 72.07 mg/kg when treating D. russelii venom. No ED₅₀ could be calculated when treating N. kaouthia venom (Fig. 3). Creatine kinase activity was significantly decreased in all 3 venoms treated with doxycycline (P < .0001) (Fig. 4).

FIGURE 1.

Open in new tabDownload slide

In vitro doxycycline inhibition of PLA, and SVMP activity in whole venom. The AUC was calculated for PLA and SVMP activities for 20-min and 4-h time courses, respectively using EnzChek Phospholipase A₂, assay kits, and EnzChek Gelatinase (SVMP) assay kits. (A) Inhibition of PLA₂, with doxycycline (Dox) was tested at concentrations of 0.32 mg/mL, 0.16 mg/mL, and 0.08 mg/mL against A. piscivorus (4 μg/mL), N. kaouthia (4 μg/mL), and D. russelii (8 μg/mL) venom. A venom-only control was included. (B) Inhibition of SVMP with doxycycline was tested at concentrations of 0.32 mg/mL, 0.16 mg/mL, and 0.08 mg/mL against A. piscivorus (120 μg/ml.), N. kaouthia (125 μg/ml.), and D. russelii (250 μg/ml.). A venom-only control was included. An ordinary 1-way ANOVA with Tukey’s post-hoc test was performed for inhibition assays to compare treatments to their respective venom controls. All error bars show mean ± SD, n = 3, (****, P < .0001).

FIGURE 2.

Open in new tabDownload slide

Percent survival of mice receiving various dosages for each venom studied and LDs, calculation. Adult BALB/c mice received intramuscular injections of various venom dosages for each venom, (A) A. piscivorus, (B) N. kaouthia, and (C) D. russelii. The resulting percent survival was measured across 48 h and the lethal dose 50 (LD₅₀) was calculated using the Spearman-Karber method. In each graph, each treatment group (venom dosage) has n = 5.

FIGURE 3.

Open in new tabDownload slide

Percent survival of mice receiving various doxycycline dosages for each venom studied and ED calculation. Adult BALB/c mice received intramuscular injections of various doxycycline dosages combined with 3 times the LD₅₀ of cach venom [A. piscivorus (AP) (60.87 mg/kg), N. kaouthia (NK) (1.14 mg/kg). D. russelii (DR) (23.76 mg/kg)] and the resulting percent survival was measured across 48 hours and the effective dose 50 (ED₅₀) was calculated using the Spearman–Karber method. Each treatment group has n = 5.

FIGURE 4.

Open in new tabDownload slide

Venom-induced CK activity with doxycycline-mediated inhibition. The amount of CK produced by mice exposed to one-fourth of the LD₅₀, strength of A. piscivorus (AP) (5.07 mg/kg), N. kaouthia (NK) (0.095 mg/kg), or D. russelli (DR) (1.98 mg/kg) venom either with or without doxycycline (64 mg/kg). All error bars show mean, ± SD, n = 5 (NaCl—Dox, AP NaCl), n-4 (NK Dox), and n-3 (AP Dox. NK NaCl, DR Dox, DR + NaCl). Statistical differences were determined using an ordinary 1-way ANOVA with Tukey’s post-hoc test comparing each venom group within itself and to the doxycycline only control (**** P < .0001). (*** P < .001), (** P < .01). (* P < .05).

Conclusion

Doxycycline reduced PLA₂- and SVMP-related lethality, particularly in A. piscivorus envenomings and in a limited capacity with D. russelii revealing its promise as a treatment for snakebites. In addition, CK activity, a common indicator of muscle damage was inhibited in mice that received doxycycline-treated venom. The doxycycline concentrations identified in the ED₅₀ studies correspond to 1,456 to 5,061 mg dosages for a 70 kg human. Factors including venom yield and snake species would affect the actual dosage needed. Studies into high-dose doxycycline safety and its effectiveness against several snake species is needed to fully translate its use into humans. Based on this work, doxycycline could be used as a treatment en route to higher echelons of care, providing protection from muscle damage and reducing lethality in different snake species.

INTRODUCTION

The Joint Trauma System Military Clinical Practice Guideline on Global Snake Envenomation Management recognizes snakebites and other envenomings as a common occupational hazard for warfighters around the world.¹ This guideline recommends utilizing broad-spectrum antivenom rather than species-specific ones in order to not only to reduce the number of antivenoms on hand, but also to avoid the necessity of snake identification. Administration of antivenom is not to be delayed, regardless of if evacuation is needed or a dry bite (no venom injected) is suspected. Broad-spectrum and field-stable antivenoms are meant to be carried during operations while other regional antivenoms are to be kept at role 2 and 3 facilities. When antivenom is not available either in the field or at higher levels of care, the patient must be transferred to alternate facilities.¹ Antivenom production has room for improvement so that broad-spectrum antivenoms can be readily available.

Snake venom is a complex mixture of biomolecules with 90% of its composition made from more than 100 different proteins and peptides.² The major families of toxins include phospholipases A₂ (PLA₂), snake venom serine proteases, snake venom metalloproteinases (SVMP), and 3-finger toxins that primarily produce neurotoxic, hemotoxic, and cytotoxic injuries.^2,3 Recently, alternative methods of treatment have focused on small-molecule inhibitors of snake venom and its components.^4–6 The tetracycline class of antibiotics is effective in inhibiting PLA₂ and metalloproteinases.^7,8 Several modified tetracyclines have been shown to inhibit the PLA₂ activity of venom from Naja naja, the Indian Cobra, and Crotalus adamanteus, the Eastern Diamondback rattlesnake, including doxycycline—a drug commonly found in expeditionary packs. Doxycycline was tested against Bothrops asper, Fer-de-lance venom, and reduced coagulopathy and hemorrhage,⁷ and shown to be a particularly effective inhibitor of metalloproteinases and PLA_2, making it useful for anti-inflammatory and potential antivenom applications. In addition, doxycycline, apart from being cost-effective, is the preferred prophylactic medication within the U.S. Military for the prevention of malaria and other tropical diseases.^8,9

The repurposing of FDA-approved drugs is an effective strategy to develop new therapeutic options while saving the cost and time. Repurposing doxycycline has been investigated in many other studies including treatments for COVID-19, Rocky Mountain spotted fever, certain cancers, and some neurodegenerative diseases.^10–13

This study explores the ability of doxycycline to inhibit the envenomation effects of 3 phylogenetically distinct snakes: Agkistrodon piscivorus (Cottonmouth, taxa Crotalinae), Naja kaouthia (Monocled Cobra, taxa Elapidae), and Daboia russelii (Russell’s Viper, taxa Viperinae). An in vitro model was used to assess doxycycline-mediated inhibition of PLA₂ and SVMP activity and an in vivo murine model was used to assess improvements in survival and inhibition of creatine kinase (CK) activity a marker of muscle damage. We hypothesized that PLA₂ and SVMP activities would be inhibited in in vitro experiments. For subsequent in vivo experiments, we hypothesized that the therapeutic effects of doxycycline on A. piscivorus, N. kaouthia, and D. russelii venoms would correlate with levels of PLA₂ and SVMP in the venom. The goal of this research was to validate the potential of doxycycline as an emergency and supportive treatment used en route to a hospital, where proper antivenom can be administered.

MATERIALS AND METHODS

Experimental Animals

All procedures involving animals were approved before implementation by the Texas A&M University Animal Care and Use Committee (IACUC approval # November 30, 2018/1429). All animals were housed in an Association for Assessment and Accreditation of Laboratory Animal Care accredited facility and all procedures were conducted in accordance with the “Guide for the Care and Use of Laboratory Animals.” Additionally, animal work was approved by the Veterinary Affairs of the Bureau of Medicine and Surgery (Navy animal use research database number NRD—1186). This study utilized a population of adult females (50%) and males (50%) BALB/c mice weighing 18 to 20 g. The mice were purchased from Jackson Laboratory, JAX® Mice & Services (Bar Harbor, ME) and bred at the National Natural Toxins Research Center at Texas A&M Kingsville. Mice were maintained under 20 °C environment with alternating 12-h light/dark cycles, receiving ad libitum access to feed and water. Cluster-randomized controlled experiments were performed, where individuals were grouped according to the treatment to be received in each assay. In addition, blinding during the outcome assessment and data analysis was followed. Following data collection, all mice were euthanized through CO₂ inhalation, in accordance with the IACUC-approved protocol.

Venom and Antibiotic Preparation

Lyophilized snake venoms from A. piscivorus, N. kaouthia, and D. russelii were purchased (Sigma-Aldrich, St. Louis, MO) and stored at −80 °C. Each venom was reconstituted with either 0.85% or 0.9% sodium chloride (w/v) and stored at −20 °C. Doxycycline hyclate was purchased from Sigma-Aldrich and suspended in phosphate-buffered saline (pH 7.2) at a concentration of 3 mg/mL. Doxycycline hyclate liquid stocks were stored at 4 °C.

Enzymatic Assays

Phospholipase A₂ activity levels were measured for each venom using the EnzChek Phospholipase A₂ Assay Kit (Invitrogen, Waltham, MA) following the manufacturer’s instructions. Measurements were taken every minute for 20 min. The venom concentrations—16, 8, 4, 2, and 1 µg/mL—were kept consistent across all types to ensure that resulting fluorescence fell within the assay’s detectable range. Fluorescence rates guided the selection of venom concentration for subsequent testing. Each concentration was tested in triplicate alongside a saline (diluent) control.

Metalloproteinase activity levels were measured using the EnzChek Gelatinase (SVMP) Assay Kit (Invitrogen, Altham, MA) in accordance with the manufacturer’s instructions. Measurements were taken every 30 min for 4 h. Venom concentrations were optimized using the following ranges: A. piscivorus = 240, 120, 60, and 30 µg/mL, N. kaouthia = 250, 125, 62.5, and 31.25 µg/mL, and D. russelli = 500, 250, 125, and 62.5 µg/mL. The fluorescence rates were used to determine the venom concentration used for subsequent testing.

The inhibitory effects of doxycycline on PLA₂ and SVMP activities within each venom were quantified using previously detailed methodology. Results from previous assays were used to determine the venom concentrations for this round of tests. For PLA₂ activity, the following final venom concentrations were used: A. piscivorus = 4 µg/mL; N. kaouthia = 4 µg/mL; and D. russelii = 8 µg/mL. For SVMP activity, the following final venom concentrations were used: A. piscivorus = 120 µg/mL, N. kaouthia = 125 µg/mL, and D. russelii = 250 µg/mL. Each venom was diluted to the above concentrations in a 1:1 ratio with doxycycline and allowed to sit at room temperature for 5 min before the completion of each assay. Mixtures were diluted into reaction buffer resulting in the above venom concentrations and to produce final doxycycline concentrations of 0.32 mg/mL, 0.16 mg/mL, and 0.08 mg/mL. Each condition was tested in triplicate with saline only, venom only, and doxycycline only controls. The results of these tests were used to determine the concentration of doxycycline for subsequent in vivo analyses.

In Vivo Venom Median Lethal Dose

The median lethal dose (LD₅₀) for each venom was based on survival rates at specific venom doses per kilogram of mouse weight: A. piscivorus = 105.26, 52.63, 26.32, 13.16, and 6.58 mg/kg, N. kaouthia = 1.71, 0.86, 0.43, 0.21, and 0.11 mg/kg, D. russelii = 12.21, 6.11, 3.05, 1.53, and 0.76 mg/kg. Additionally, a 0.85% saline solution was used as a control. Venom and control solutions were stored at 4 °C and prewarmed to body temperature before injection. A volume of 0.2 mL of each solution was injected into the right, hind quadriceps femoris muscle of adult BALB/c mice weighing 18 to 20 g. The injections were administered using a 1 mL syringe fitted with a 30-gauge, 1.27-cm needle. For each venom, 6 treatment groups were set up (5 venom concentrations and 1 saline control), each comprising 5 mice, resulting in 30 mice per venom and 90 for LD₅₀ experiments. Mice were monitored for 48 h post-injection and percent survival was measured.

In Vivo Doxycycline Median Effective Dose

The previously determined LD₅₀ concentrations of each venom (A. piscivorus = 20.29 mg/kg, N. kaouthia = 0.38 mg/kg, and D. russelii = 7.92 mg/kg) were tested in vivo to determine the median effective dose (ED₅₀) of doxycycline. Doxycycline at concentrations of 12.20, 6.10, 3.05, 1.53, and 0.76 mg/mL were diluted by mixing in a 1:1 ratio with 30 × LD₅₀ equivalents of each venom to produce doxycycline concentrations of 6.10, 3.05, 1.53, 0.76, and 0.38 mg/mL, in addition to 15 × LD₅₀. The venom/doxycycline mixtures were vortexed and incubated at 37 °C for 30 min before administration in 0.2 mL volumes, corresponding to final doxycycline dosages of 64, 32, 16, 8, and 4 mg/kg and 3 × LD₅₀ equivalents of venom (A. piscivorus = 60.87 mg/kg, N. kaouthia = 1.14 mg/kg, and D. russelii = 23.76 mg/kg). Each mixture was injected into the right, hind quadriceps femoris muscle of adult BALB/c mice weighing 18 to 20 g using a 1 mL syringe fitted with a 30-gauge, 1.27-cm needle. The venom-only and doxycycline-only solutions were applied in the same concentrations and under the same administration conditions as described above. The study comprised 7 treatment groups (5 venom/doxycycline mixtures, venom-only, and doxycycline-only), each group containing 5 mice, resulting in 35 mice per venom and 105 mice for ED₅₀ studies. All mice were monitored over a 48-h period post-treatment and percent survival was observed to determine ED₅₀.

In Vivo CK Inhibition Assay

To evaluate the in vivo CK activity regarding myotoxic response to the 3 venoms both individually and in combination with doxycycline, a colorimetric assay kit was used (Biovision, Milpitas, California). A 1:1 mixture of venom/doxycycline was prepared depending upon the respective lethal venom dosage previously determined. The mixtures were prepared to correspond to a sub-lethal dose or one-fourth of the LD₅₀ strength (A. piscivorus = 5.07 mg/kg, N. kaouthia = 0.095 mg/kg, and D. russelii = 1.98 mg/kg) and a final doxycycline concentration of 6.1 mg/mL (a 64 mg/kg dose). There were 7 groups total including doxycycline + 0.85% saline (control) (n = 5), venom (A. piscivorus (n = 5), N. kaouthia (n = 3), or D. russelii (n = 3)) + 0.85% saline (control), and venom (A. piscivorus (n = 3), N. kaouthia (n = 4), or D. russelii (n = 3)) + doxycycline (treatment). A total of 26 mice were used in CK activity assays. All solutions were administered in volumes of 0.2 mL into the right, hind quadriceps femoris muscle of adult BALB/c mice (18-20 g) using 1 mL syringe with a 30-gauge, 1.27-cm needle. Blood samples were collected via cardiac puncture 3 h after treatment, treated with 1% EDTA, and plasma was obtained through centrifugation. The mice were excluded if they died prematurely, preventing the collection of blood sample. Enzymatic activity (expressed in mU/mL), where 1 unit equals the quantity of enzyme producing 1.0 µmol of nicotinamide adenine dinucleotide and hydrogen per minute at pH 9.0 and 37 °C, was determined using the colorimetric assay kit following manufacturer’s instructions.

Statistical Analysis

An ordinary 1-way ANOVA with Tukey’s post-hoc test was performed on GraphPad Prism 6 using the area under the curve (AUC) of fluorescence signal from PLA₂ and metalloproteinase inhibition assays. An ordinary 1-way ANOVA with Tukey’s post-hoc test was used as well for CK activity assays. In vitro PLA₂ and metalloproteinase assays were performed in triplicate with a saline (diluent) control. The LD₅₀ of each venom and the corresponding doxycycline ED₅₀ were calculated using the Spearman-Karber method. Group sizes for animal studies were followed according to the standard operating procedure at the National Natural Toxins Research Center Texas A&M Kingsville for changes in CK activity and animal survival. Sample size calculations for in vivo work was performed using an alpha = .05 (false positive) and beta = .2 (false negative). An n = 5 per treatment group was used for survival experiments and n = 3 to n =5 was used for CK activity experiments. Data are expressed as mean ± SD and statistical significance is set at P ≤ .05.

RESULTS

Enzymatic Inhibition With Doxycycline

Phospholipase A₂ activity of A. piscivorus, N. kaouthia, and D. russelii venoms were measured and optimal signals were observed at 4, 8, and 4 µg/mL for A. piscivorus, N. kaouthia, and D. russelii venom, respectively. The inhibitory capacity of doxycycline against PLA₂ was compared using the AUC of the timecourse fluorescence signal for comparison. Phospholipase A₂ activity was reduced in A. piscivorus to 1.5%, 5%, and 22%, in N. kaouthia to 5%, 19%, and 30%, and in D. russelii to 8%, 21%, and 44% using 0.32, 0.16, and 0.08 mg/mL doxycycline, respectively, when compared to venom only controls (P ≤ .0001) (Fig. 1A).

Metalloproteinase activities of the venoms were evaluated and the following concentrations were used for doxycycline inhibition testing: A. piscivorus = 120 µg/mL, N. kaouthia = 125 µg/mL, and D. russelii = 250 µg/mL. Comparison between the AUC values for doxycycline-treated venom and venom only controls revealed that SVMP activity was reduced in A. piscivorus to 4%, 28%, and 62%, in N. kaouthia to 4%, 19%, and 39%, and in D. russelii to 5%, 22%, and 58% using 0.32, 0.16, and 0.08 mg/mL doxycycline, respectively (P ≤ .0001) (Fig. 1B).

Murine Venom Median Lethal Dosage

With doxycycline demonstrating venom inhibition through the in vitro enzymatic assays, the next step was to evaluate its inhibitory effects for these 2 compounds in vivo using a murine animal model. To understand the potential inhibitory effects of doxycycline in vivo, the median lethal dose (LD₅₀) of each venom must first be established. A range of dosages were selected for each venom and were administered intramuscularly into adult BALB/c mice weighing 18 to 20 g. Across 48 h, the survival rates were determined and used to calculate the LD₅₀ for each venom (Fig. 2). Mortalities were observed after dosages higher than 13.16 mg/kg for A. piscivorus venom, 0.21 mg/kg for N. kaouthia venom, and higher than 3.05 mg/kg for D. russelii venom were administered to the mice. Based on the cumulative results from this study, the obtained LD₅₀ values in these mice are as follows: A. piscivorus = 20.29 mg/kg (Fig. 2A), N. kaouthia = 0.38 mg/kg (Fig. 2B), and D. russelii = 7.92 mg/kg (Fig. 2C).

Median Effective Dosage of Doxycycline Venom Inhibition

Once the LD₅₀ for each venom was determined, the median effective dose (ED₅₀) for inhibition of venom lethality could be assessed. To test this, a range of doxycycline dosages were combined with 3 times the LD₅₀ for each venom (A. piscivorus = 60.87 mg/kg, N. kaouthia = 1.14 mg/kg, and D. russelii = 23.76 mg/kg) and administered intramuscularly into adult BALB/c mice weighing 18 to 20 g. The mice were monitored during the 48 h following treatment and the survival rate was determined (Fig. 3). This information was then used to calculate the ED₅₀ of doxycycline for each venom. The doxycycline ED₅₀ for A. piscivorus survival was calculated to be 20.82 mg/kg, with 20% of the mice surviving with a 16 mg/kg dose and 100% surviving with a 32 mg/kg dose. For N. kaouthia, 0% survival was seen for all doxycycline dosages tested in this study. The ED₅₀ of doxycycline with the D. russelii venom was calculated to be 72.07 mg/kg, with 40% of the mice surviving with a 64 mg/kg dose.

Inhibition of Venom-induced Myotoxicity by Doxycycline

A CK activity assay was performed to determine the potential myotoxicity inhibition properties of doxycycline interacting with each venom. For this, each venom was tested alone or at one-fourth of the LD₅₀ and in combination with doxycycline. Additionally, doxycycline alone was tested for this study. Both the venom and doxycycline were combined with saline as a comparison to their venom/doxycycline mixture counterparts. Under exposure to all venoms, doxycycline treatment displayed decreased CK activity, indicative of myotoxicity inhibition. Mice injected with one-fourth of the LD₅₀ strength of A. piscivorus venom 5.07 mg/kg and 64 mg/kg of doxycycline saw CK activity decrease from 708.43 ± 134.43 mU/mL to 186.52 ± 156.47 mU/mL (P ≤ .001). Naja kaouthia venom at 0.095 mg/kg saw a decrease of CK activity from an average of 407 ± 150 mU/mL to 66.38 ± 23.55 mU/mL with the same doxycycline treatment (P ≤ .01). A D. russelii venom dose of 1.98 mg/kg and doxycycline saw CK activity decrease from 813.35 ± 55.63 mU/mL to 136.72 ± 35.52 mU/mL (P ≤ .0001). All venom-only measurements were higher than the doxycycline-only control (P ≤ .0001-.05) and no doxycycline-treated venom sample was different than the doxycycline control (Fig. 4).

DISCUSSION

Doxycycline has the potential to protect against snakebites because of its ability to reduce PLA₂ and SVMP activity. When administered intramuscularly, doxycycline demonstrated systemic biochemical relevance for all 3 venoms through lowered CK activity in mouse blood samples. High CK activity is an indicator of myotoxicity commonly measured in snake venom studies and is mediated in part by PLA₂ and SVMP.^14–16 As is seen in Figure. 4, CK activity was lowered through doxycycline mediated inhibition of venom suggesting a protective effect on muscle tissue overall that can be examined more closely in future studies through histological studies to show local muscle protection. Administered as an immediate response to snakebite victims, doxycycline has the potential to conserve wound site tissue, thereby reducing morbidity during post-recovery. This suggests that doxycycline could be a useful intervention in snakebite envenomings.

Despite inhibiting CK activity in vivo and demonstrating in vitro inhibition of PLA₂ and SVMP, doxycycline’s effectiveness varied across venoms. It failed to protect mice from N. kaouthia venom, moderately countered D. russelii, and successfully inhibited A. piscivorus. This variance is likely because of venom composition. Agkistrodon piscivorus venom, composed mainly of PLA₂ and MP (54-79% of total mass), enabled doxycycline to neutralize most toxins. Conversely, D. russelii venom’s diverse toxin range, including a combined PLA₂/SVMP mass of 40 to 60% and variable concentrations of potent hemotoxins and cytotoxins (7-51% C-type lectins and Kunitz peptides),³ reduced doxycycline’s efficacy. This is even more apparent in N. kaouthia where 13% to 30% of the venom can be composed of PLA₂/SVMP but with potent neurotoxins, 3 finger toxins at 57% to 78%.³ No ED₅₀ could be calculated for N. kaouthia as none of the mice survived.

In addition to composition, LD₅₀s and ED₅₀s play a role in the context of venom yield during an envenoming. The amount of venom or injected during an attack is highly variable and depends on several factors including snake size, age, diet, multiple bites, and depleted venom reserves.^17,18 Bites can occur as dry bites (especially in defensive bites) where no venom is injected but for illustration A. piscivorus has been reported to inject 0 to 170 mg with one report stating typical adults inject 90 to 170 mg.^18,19 The actual lethal dose in humans is not known although some predictions estimate 100 to 150 mg.²⁰ Based on our LD₅₀ (20.29 mg/kg), it would require 1,457 mg of venom intramuscularly while another group predicted a lethal dose of 1,806 mg subcutaneously. The wide range in lethal doses likely stems from route of administration; intravenous LD₅₀s have been reported as 3.42 to 5.13 mg/kg,^19,20 which do not correlate with 100 to 150 mg dosages, and subcutaneously as 25.8 mg/kg in mice.²⁰ When administered intramuscularly, the ED₅₀ for doxycycline concentration against A. piscivorus was 20.8 mg/kg which equates to 1,456 mg for a 70 kg human patient. Most snakebites are naturally subcutaneous or intramuscular and if the route of injection for A. piscivorus venom affects its lethality so greatly that is promising for doxycycline’s use.²¹ In addition, ED₅₀ was calculated with 3 × LD₅₀s and given the putative venom yields doxycycline may be more potent in reality.

When considering D. russelii, one study reported venom yields from 6 to 147 mg during a single defensive bite.²² Intravenous LD₅₀s in mice have ranged from 0.11 to 0.42 mg/kg,^23–25 which corresponds to 7.7 to 29.4 mg for a 70 kg human, although one report predicted 40 to 70 mg as a lethal human dose.²⁶ In either case, the lethal dose is well within the capabilities of D. russelii, although we reported an intramuscular LD₅₀ of 7.92 mg/kg or a total dose of 554 mg in a human. Even considering the different route of administration the LD₅₀ appears high compared to previously reported numbers. The doxycycline ED₅₀ of 72.3 mg/kg would require a 5,061 mg dose to counteract 3 × LD₅₀s from our study, which would be much higher than experienced in nature.

Naja kaouthia venom yields ranged from 0 to 125 mg in one study while another found an average of 263 mg.^18,27 LD₅₀s of 0.148 to 0.9 mg/kg intravenously in mice have been reported and our LD₅₀ falls within the range at 0.38 mg/kg.^28,29 At only 26.6 mg needed for a lethal dose according to our numbers a bite from N. kaouthia of this severity is likely. Doxycycline was unable to save at any concentration and even at low levels of venom it is likely to be ineffective at saving because of potent neurotoxins but has promise at inhibiting local tissue myotoxic effects.

The previously mentioned doxycycline dosages are high especially compared to when taken orally the standard dose is 100 mg/day for mild bacterial infections or 300 mg/day for severe infections. For malarial prevention, 100 mg/day is the standard in health care and has been adopted by the U.S. Military for deployed warfighters in endemic areas.^9,30 Intramuscular injections are not commonly practiced in human patients but are used for animals including pigs, sheep, goats, and at least one study testing 100 mg/kg in a bird model.^31–34 These concentrations provide a foundation that will allow for further testing in humans at necessary high dosages to treat several snake species envenomings.

Although there is a lack of safety data concerning intramuscular doxycycline injections in humans, especially at the necessary doses identified in this study, it is important to note that there are no known instances of doxycycline overdose in humans resulting from a single dose. However, one study in rats showed skeletal muscle damage and cardiomyopathy when receiving 50 mg/kg twice daily for 10 days.³⁵ Although safety data should be gathered for high dose intramuscular and intravenous injections, the benefits likely outweigh the risks given the immediate threat of the venom toxins, short administration timeframe, affordability compared to traditional antivenom, and general safety of doxycycline.

As a standard, doxycycline is carried by military medics which makes it a convenient candidate for administration immediately after an envenoming and through the first several hours of treatment, although antibiotic administration is generally not advised after snakebites because infections are rare.¹ The potential broad-spectrum capabilities of doxycycline for snakebites directly address the needs of warfighters who need antivenom in the field and will aid in reducing evacuations. Studies considering Africa to Germany medevacs range from $17,000 to $265,000 depending on the location.^36,37 This also highlights the need for improved treatments for deployed forces as snakebite cases are famously underreported. In addition to the potential monetary loss a decrease in mission success is seen by lost workdays by the patient, their replacement, and the transport teams.

This research aimed to establish the basic venom inhibition characteristics of doxycycline as an en route to hospital treatment for venomous snakebites. Further investigation into time to death after envenoming and treatment, the types of venom, high-concentration doxycycline safety, effectiveness of oral administration vs. intravenous, and doxycycline-antivenom interactions will help add this drug to emergency antivenom support.

CONCLUSION

We have gathered data that the inhibition of PLA₂ and SVMP by doxycycline is an effective way to increase survival rate and provides biochemical protection of muscle tissue following intramuscular administration of snake venom in a pre-clinical setting. We expect that in the time between injury and treatment at a medical facility, a single high dosage of this common antibiotic could be administered to the warfighter alone or in tandem with antivenom that is on hand. This would increase its effectiveness and therefore improve the survivability and protect muscle tissue that once injured cannot be saved by future antivenom. Venom composition will be critical to consider when administering doxycycline, noting that venoms with high neurotoxin content and other toxins will retain a higher amount of their activity. Doxycycline’s ubiquity and low price within civilian and military spheres will allow physicians to leverage its advantages to increase warfighter effectiveness with reduced strain on medical resources and minimal increases to money and effort.

ACKNOWLEDGMENTS

None declared.

CLINICAL TRIAL REGISTRATION

Not applicable.

INSTITUTIONAL REVIEW BOARD (HUMAN SUBJECTS)

Not applicable.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC)

IACUC approval # November 30, 2018/1429; Veterinary Affairs of the Bureau of Medicine and Surgery, Navy animal use research database number NRD—1186. “The study protocol was reviewed and approved by the Texas A&M Kingsville Institutional Animal Care and Use Committee in compliance with all applicable federal regulations governing the protection of animals and research.”

INDIVIDUAL AUTHOR CONTRIBUTION STATEMENT

D.K.A. and M.A.R. analyzed the data, wrote the original manuscript, and edited the manuscript. E.M.S., M.A.H., E.Y.H., and A.A.F. collected the data and editedthe manuscript drafts. D.W.T. edited the manuscript drafts. E.E.S. and Y.Y.H. provided experimental design, coordination, and editing of manuscripts. E.E.S. provided subject-matter expertise and Y.Y.H. procured the funding.

COPYRIGHT STATEMENT

I am a military service member or federal/contracted employee of the U.S. Government. This work was prepared as part of my official duties. Title 17 U.S.C. 105 provides that “copyright protection under this title is not available for any work of the United States Government.” Title 17 U.S.C. 101 defines a U.S. Government work as work prepared by a military service member or employee of the U.S. Government as part of that person’s official duties.

INSTITUTIONAL CLEARANCE

Institutional clearance obtained.

FUNDING

This work was supported by the Defense Health Agency, work unit number G1716.

CONFLICT OF INTEREST STATEMENT

No competing interests to declare.

DATA AVAILABILITY

The data that support the findings of this study are available on request from the corresponding author. All data are freely accessible.

REFERENCES

Author notes