Exploring the utility of ssDNA aptamers directed against snake venom toxins as new therapeutics for tropical snakebite envenoming

Snakebite is a neglected tropical disease that causes considerable death and disability in the tropical world. Although snakebite can cause a variety of pathologies in victims, haemotoxic effects are particularly common and are typically characterised by haemorrhage and/or venom-induced consumption coagulopathy. Antivenoms are the mainstay therapy for treating the toxic effects of snakebite, but despite saving thousands of lives annually, these therapies are associated with limited cross-snake species efficacy due to venom variation, which ultimately restricts their therapeutic utility to particular geographical regions. In this study, we sought to explore the potential of ssDNA aptamers as toxin-specific inhibitory alternatives to antibodies. As a proof of principle model, we selected snake venom serine protease toxins, which are responsible for contributing to venom-induced coagulopathy following snakebite envenoming, as our target. Using SELEX technology, we selected ssDNA aptamers against recombinantly expressed versions of the fibrinogenolytic SVSPs Ancrod from the venom of Calloselasma rhodostoma and Batroxobin from Bothrops atrox. From the resulting pool of specific ssDNA aptamers directed against each target, we identified candidates that exhibited low nanomolar binding affinities to their targets. Downstream ALISA, fibrinogenolysis, and coagulation profiling experiments demonstrated that the candidate aptamers were able to recognise native and recombinant SVSP toxins and inhibit toxin- and venom-induced prolongation of plasma clotting times and consumption of fibrinogen, with inhibitory potencies highly comparable to commercial polyvalent antivenoms. Our findings demonstrate that rationally selected toxin-specific aptamers can exhibit broad in vitro cross-reactivity against toxins found in different snake venoms and are capable of inhibiting toxins in pathologically relevant in vitro and ex vivo models of venom activity. These data highlight the potential utility of ssDNA aptamers as novel toxin-inhibiting therapeutics of value for tackling snakebite envenoming.


Introduction
Snakebite envenoming is a significant public health issue as more than 5.4 million people are bitten annually by venomous snakes, and it mainly affects resource-poor communities in the rural tropics and subtropics across Africa, the Americas, Asia, the Middle East, and Australasia (1)(2)(3). Since 2017, snakebite has been classified by the World Health Organization as a neglected tropical disease (4).
Snake venom is a potentially lethal and complex mixture of hundreds of functional toxins that vary extensively among snake species (5, 6). Although snake venom exhibits major variations, the various toxins present can be broadly classified into causing three major categories of pathology in snakebite victims: haemotoxicity, neurotoxicity and cytotoxicity (7).
The venom toxins primarily thought to be responsible for causing VICC are members of the snake venom metalloproteinase (SVMP) and snake venom serine protease (SVSP) toxin families (14,15).
Each of these toxin families is multi-locus in nature, in that related genes produce multiple related protein isoforms in the venom glands of snake species and, due to isoform diversity as the result of protein neofunctionalisation, these isoforms typically exhibit distinct functional activities (7,13,16).
Snakebite coagulopathy disorders result from the consumption of clotting factors (e.g., Factor X, V and prothrombin) (15) via initial toxin-induced cleavage stimulating activation of the coagulation cascade, resulting in the abnormal and continual activation of downstream clotting factors, and a loss of clotting capability via depletion of fibrinogen (13,16). However, some toxins also act directly on fibrinogen in a fibrinogenolytic manner (15), and many venoms contain both upstream clotting factor activators and fibrinogenolytic toxins simultaneously. Thus, venom toxins can result in hypofibrinogenaemia, with either no formation of fibrin due to the cleavage of the fibrinogen α-chain and/or β-chain by thrombin-like enzymes or fibrinogenases (typically SVSP toxins), or afibrinogenaemia (the absence of circulating fibrinogen) due to depletion via upstream coagulation cascade activation (17). While this combination of procoagulant and anticoagulant toxins present in snake venom (18) can rapidly lead to a loss of clotting capability in envenomed victims (9,15,17), the 4 severity of envenoming can be further exacerbated by other toxins (e.g., SVMPs) simultaneously stimulating widespread haemorrhage by disrupting the integrity of the microvasculature via cleavage of extracellular matrix proteins (19,20).
Antivenoms, which consist of animal-derived polyclonal antibodies derived from hyperimmunised serum/plasma, remain the only specific therapeutic to combat the toxicity of snakebite envenoming (21). Although antivenoms are life-saving therapeutics, they are also associated with several limitations that affect their efficacy, safety, and utility in the tropical world. For example, the intravenous delivery of animal-derived antibodies comes with an associated risk of adverse reactions, which may include vomiting, urticaria, generalised rash or, in rarer cases, more severe effects (e.g., anaphylactic shock) (22)(23)(24)(25). In addition, only 10-20% of antivenom antibodies are typically specific to the venom immunogens (26), as venom-immunised animals are exposed to other environmental stimuli and toxin-specificity is typically not selected for from the resulting antibody pool. Perhaps most importantly, antivenoms exhibit limited cross-snake species efficacy as the direct result of variation in venom composition among medically important snake species (6,27). Thus, antivenom efficacy is largely restricted to the snake species, or those closely-related to or with similar venom compositions to, those used during the immunisation process (28). The result of limited paraspecific antivenom efficacy is a fragmented drug market, with numerous antivenoms manufactured around the world with specificity against certain snake species and thus restricted for use in specific geographical regions. Nonetheless, antivenoms are often unaffordable to patients, as treatment courses frequently necessitate the administration of multiple vials (ranging from 3-30 vials depending on region and product), resulting in costs exceeding USD 1,000 in parts of sub-Saharan Africa, for example (29). This cost pushes many tropical snakebite victims further into poverty (30,31).
In response to the above, much recent research has focused on applying alternate therapeutic strategies to circumvent the limitations currently associated with conventional antivenoms. Promising approaches explored to date include the use of toxin-specific monoclonal antibodies (32), smallmolecule drugs (33), decoy receptor binding proteins (34) and aptamers (35)(36)(37). Although arguably the least studied of these approaches in the context of snakebite, single-stranded DNA (ssDNA) aptamers have emerged as promising alternatives to antibodies in various biosensing platforms and as the biological recognition element for therapeutic applications and diagnostic tools for different diseases (38)(39)(40). Aptamers are short ssDNA (or ssRNA) molecules that can bind to their target and fold into complex and stable three-dimensional shapes (41,42) with high specificity and affinity through hydrogen bonding, van der Waals forces, hydrophobic, salt bridges and other electrostatic interactions (43)(44)(45)(46). Consequently, aptamers act like antibodies by binding and inhibiting target antigens, and thus have been referred to as chemical antibodies due to their synthetic production (47, 5 48). ssDNA aptamers possess a number of advantages over antibodies as, in addition to seemingly exhibiting comparable high-affinity recognition properties (i.e., dissociation constants in the low nanomolar (nM)/picomolar (pM) range) (43,44,49), they are highly specific, sensitive, chemically stable with long shelf lives (50)(51)(52), and require facile artificial synthesis (53) resulting in high purity, low inter-batch variability and cost-effective production (54)(55)(56). Further, ssDNA aptamers are poorly immunogenic which aids their tolerance, and due to their small size they diffuse readily into tissues (57). However, this latter characteristic also causes challenges, as the in vivo half-life of ssDNA aptamers is much shorter (~20 minutes) than antibodies (e.g., potentially a few days to weeks) (58), though prolonged circulation can be improved by chemical modification (57,(59)(60)(61). Despite the therapeutic promise of aptamers, to date the only FDA approved aptamer remains the macular degeneration treatment Macugen® (Pegaptanib) which was first licenced in 2005 (62). Since then, no other aptamers have been entered the market, although many candidate aptamers have entered clinical trials over the past decade, including as promising new treatments for heart disease, cancer and type II diabetes (62).
ssDNA aptamers are typically selected by Systematic Evolution of Ligands by Exponential enrichment (SELEX) technology, which was introduced by Ellington and Tuerk three decades ago (43,44). This process is based on the isolation of high-affinity ligands from a combinatorial ssDNA library (approximately 10 17 random sequence oligonucleotides) through repeated cycles (usually requiring seven to fifteen rounds) of binding, elution, and amplification to select for oligonucleotides with desired specificity and/or sensitivity (48,63,64). Following cyclical SELEX enrichment, the final ssDNA aptamer pool is subjected to sequencing to identify the optimal resulting binding sequences, which are then manufactured by chemical synthesis and downstream optimised by introduction of chemical modifications to further improve their pharmacokinetic and/or pharmacodynamic profiles (65). To date, several ssDNA aptamers have been successfully selected for and applied against a wide range of targets that include, proteins, peptides, small organic compounds, carbohydrates, metal ions and biological cofactors (66)(67)(68)(69)(70)(71)(72). In addition, ssDNA aptamers have been shown to bind to highly toxic antigens, which, due to toxicity, cannot be achieved in animal-based methods to generate specific antibodies (73). In the context of animal toxins, little work has been done, though aptamers have been selected against the three-finger toxin α-bungarotoxin from Bungarus multicinctus (74) and shown to exhibit cross-recognition and inhibitory activity against cytotoxins from Naja atra venom that share similar tertiary structures (37). Another recent study selected ssDNA aptamers against Indian Bungarus caeruleus (common krait) venom and used them in a diagnostic context to discriminate envenomings by this species from those of other medically important snakes found in the region (75).

6
In this study, we sought to explore the potential utility of ssDNA aptamers against snake venom SVSP toxins responsible for contributing to venom induced coagulopathy following snakebite envenoming.
Using SELEX technology, we selected ssDNA aptamers against recombinantly expressed versions of the fibrinogenolytic SVSPs Ancrod from the venom of Calloselasma rhodostoma and Batroxobin from Bothrops atrox. Downstream ssDNA Aptamer Linked Immobilised Sorbent Assay (ALISA), fibrinogenolysis, and coagulation experiments demonstrated that the rationally selected ssDNA aptamers were able to recognise native and recombinant SVSP toxins and inhibit toxin and venominduced prolongation of plasma clotting times and consumption of fibrinogen. These data highlight the potential utility of ssDNA aptamers as novel toxin-inhibiting therapeutics and alternative binding scaffolds for future treatment strategies targeting tropical snakebites.

Materials and methods ssDNA library and primer design
The ssDNA library and PCR primers were chemically synthesised and purified by Integrated DNA Technologies, Inc. (90). The random ssDNA library used in the first SELEX cycle consisted of (3 nM of 2 × 10 17 sequences) nucleotides. The ssDNA library consisted of a central random region of 72 nucleotides flanked by two primers consisting of 16 nucleotides at the 5ʹ and 3ʹ end (5ʹ-TCCCTACGGCGCTAAC-N72-GTTGTAGCACGGTGGC-3ʹ) for amplification of the library sequence. To facilitate the separation of ssDNA from amplified double-stranded (dsDNA) PCR products and to quantify DNA during selection, the forward and reverse primers were designed and modified with fluorescein and a HEG linker (the spacer is designed to block the PCR extension step during amplification). The modified PCR forward primer used during the selection cycles was (5ʹ-6-FAM-TCCCTACGGCGCTAAC-3ʹ), and the reverse primer was (5ʹ-poly dA20-HEG-spacer-GTTGTAGCACGGTGGC-3ʹ). An unmodified primer set was used for PCR amplification and cloning when SELEX rounds were completed and enriched.

Target conjugation
The recombinant toxins used in this study, Ancrod from C. rhodostoma and Batroxobin from B. atrox, were previously expressed in HEK293 mammalian cell lines (76). In this study, we used these toxins to conjugate to NHS-activated Sepharose® 4 Fast Flow beads (Sigma-Aldrich, UK) to generate ssDNA aptamers as novel toxin inhibiting therapeutics. To do so, the manufacturer protocol was followed, with first a 300 μg/ml solution of each toxin prepared in coupling buffer (0.1 M NaHCO3, 0.5 M NaCl, pH 8.3). Next, 2 mL of NHS-activated Sepharose beads were rinsed multiple times with 1 mM HCl for 7 15 minutes to remove the additives and preserve the activity of the reactive groups. Each venom toxin was then added immediately to the washed Sepharose beads independently (1:1 v:v ratio) and mixed by end-over-end rotation overnight at 4 ⁰C. After incubation, the beads were centrifuged at 14 x g for 5 minutes, the supernatant discarded, and the samples washed with 1 ml of 1 M NaCl. This centrifugation step was repeated before 2 ml of 1 M NaCl was added, and the samples were incubated with rotation for 2 hours at 4 ⁰C to block the unreacted amine groups on the beads. After blocking, the toxin-conjugated Sepharose was washed three times with 4 ml in an alternating manner, first with 0.1 M acetic acid/sodium acetate, pH 4 containing 0.5 M NaCl 0.1 M, followed by Tris-HCl buffer, pH 8 containing 0.5 M NaCl. Thereafter, the toxin-conjugated Sepharose was stored in 50 mM Tris-HCl buffer (pH 7.5), 0.05% sodium azide at 4 ⁰C until use.

PCR amplification
Each of the recombinant toxin-conjugated Sepharose mixtures were washed five times by adding 500 μl binding buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 2 mM MgCl2) to 100 μl of Sepharose, using centrifugal tube filters (0.45 µm) (Costar, USA). In the first SELEX cycle, 50 μl (3 nM) of the ssDNA library solution in 450 μl binding buffer was used, whereas 50 pM of the DNA elution from each round was used for subsequent cycles. The DNA and binding buffer mixture were heated to 90 ⁰C on a heat block for five minutes, cooled at 4 ⁰C for ten minutes and then held at room temperature (RT) for five minutes. Next, 300 μl of DNA and binding buffer mixture was added to the washed beads and incubated end-over-end for 1 hour at RT. After incubation, the mixtures were washed and centrifuged five times with 500 μl of binding buffer at 82 x g for 2 minutes, to ensure the elimination of unbound DNA. The flow-through from the first wash was retained for use as a positive control in downstream agarose and denaturing polyacrylamide gel electrophoresis (PAGE) experiments. DNA bound to the toxin-conjugated beads was collected from each sample using elution buffer (7 M urea in binding buffer) and incubated at 90 ⁰C for 10 minutes on a heat block. The elution step has performed a total of five times via the addition of 300 μl elution buffer until no fluorescence was detected in the eluted aliquots, indicating the complete elution of bound DNA from the beads. The eluted aliquots were then desalted with filtered water and concentrated using an ultrafiltration device with a 3 kDa cut-off membrane (Amicon Ultra-0.5 Centrifugal Filter Unit; Merck, UK) to remove any urea and salt residues. A ssDNA product with different lengths for each strand was also generated via PCR amplification by using 20 parallel 50 μL reactions, using the same conditions as outlined above, except with the use of 0.2 μM of the labelled forward primer. Following amplification, the 20 reactions were pooled and aliquoted into three Eppendorf tubes, each containing 250 μl of pooled ssDNA product, 50 μl 3M sodium acetate, pH 5.2 and 750 μl absolute ethanol. The tubes were then mixed and stored at -80 ⁰C for 2 hours, before centrifugation at 13,000 x g for 30 minutes at 4 ⁰C. The supernatant was discarded, and the remaining pellet was dried at 90 ⁰C for a couple of minutes on a heat block before resuspension in 400 μl of water and formamide (1:1 v/v) and incubation at 90 ⁰C for 5 minutes.

Separation of ssDNA by denaturing PAGE
We employed a 10% denaturing PAGE to separate the resulting ssDNA from the double-stranded

Counter-selection
Following enrichment, counter (negative) selection was employed by using blank Sepharose beads instead of those conjugated to toxins. The goal here was to increase enrichment to the toxin target via elimination of aptamers directed towards the beads. Sepharose beads were subjected to the aforementioned conjugation process with 0.1 M Tris-HCl buffer, pH 8.3 containing 0.5 M NaCl only (i.e., in the absence of toxins), and counter selection followed the same experimental strategy as described above and was used to filter out negative targets.

Cloning and sequencing ssDNA aptamers
After ssDNA recovery, the eluted ssDNA from the 14 th round of selection was amplified using the unlabelled PCR primer (5ʹ-TCCCTACGGCGCTAAC-3ʹ) following the same conditions as PCR amplification: 94 ⁰C for 10 minutes, followed by 17 cycles of 94 ⁰C for 1 minute, 54 ⁰C for 1 minute and 72 ⁰C for 1 minute, and a final extension step of 10 minutes at 72 ⁰C. Next, ligation was performed by using the TOPO TA Cloning Kit for Sequencing and One Shot TOP10 Chemically Competent E. coli (Sigma-Aldrich, UK), and incubated at 37 ⁰C overnight. Following incubation, ten random positive colonies (white colonies) were picked for colony PCR analysis: 5 µL 10x PCR buffer, 0.5 µL deoxyribonucleoside triphosphates (dNTPs) mix (50 mM), 0.5 µL of forward and reverse M13 primers (0.1 µg/µL each), 41.5 µL nuclease-free water and 1 µL Taq polymerase (1 unit/µL). Colony PCR was performed using the following cycling parameters: initial denaturation at 94 ⁰C for 2 minutes, followed by 25 cycles of denaturation at 94 ⁰C for 1 minute, annealing at 55 ⁰C for 1 minute and extension at 72 ⁰C for 1 minute, with a final extension step at 72 ⁰C for 7 minutes. To confirm insert amplification, 2% agarose gel electrophoresis was performed.
One hundred positive bacterial colonies were sterilely selected and inoculated with 5 ml of LB medium containing 50 µL Ampicillin (20 mg/ml) and incubated overnight at 37 ⁰C with shaking at 200 rpm.
Next, plasmid DNA was purified using a QIAprep Spin Miniprep Kit (Qiagen, Germany). The overnight culture was centrifuged at 6000 x g for 10 minutes at RT. The pellet was then resuspended in 250 µl Buffer P1 (50 mM Tris-HCl pH 8.0, 10 mM EDTA, 100 μg/ml RnaseA) and transferred to a microcentrifuge tube. Then, 250 µl Buffer P2 (lysis buffer: 200 mM NaOH, 1%) was added and mixed thoroughly by inverting the tube gently 4-6 times. After that, 350 μl Buffer N3 was added and mixed immediately and thoroughly by inverting the tube 4-6 times. Next, samples were centrifuged for 10 minutes at 17,900 x g, and the supernatant was transferred to QIAprep 2.0 Spin Columns before brief centrifugation at 2,817 x g for one minute with the eluate discarded. Columns were then washed with 750 µl Buffer PE via centrifugation at 2,817 x g for one minute, the eluate discarded, and columns centrifuged for an additional minute to remove any residual buffer. Finally, 50 μl Buffer EB (10 mM Tris-HCl, pH 8.5) was added to each column and incubated at RT for two minutes prior to elution via centrifugation at 2,817 x g for 1 minute. The resulting PCR products were then sequenced commercially via Sanger sequencing (SourceBioScience, UK), and the resulting ssDNA aptamer sequences were analysed and aligned using the PRALINE website: (https://www.ibi.vu.nl/programs/pralinewww/).

Dissociation constants (KdS)
The binding affinities of the identified ssDNA aptamers against each of the corresponding toxins used for selection were evaluated using a fluorescence-based assay. Both forward and reverse primer-

Fibrinogenolysis via degradation SDS-PAGE
To assess the capability of the selected ssDNA aptamers to inhibit the fibrinogenolytic activity of the recombinant toxins and corresponding native snake venoms, we used a degradation SDS-PAGE

Fibrinogen consumption via the Clauss method
The Clauss method is a quantitative, clot-based assay. It measures the ability of thrombin to convert fibrinogen to fibrin clot followed by manual time measurements of clotting (79). Here we applied this Tubes were gently tilted at regular intervals (returning to the water bath between tilting), and the time for the formation of a clot recorded. All experiments were performed in duplicate.

Prothrombin Time (PT)
To measure the inhibitory capability of selected aptamers against toxins acting on the extrinsic coagulation pathway, we quantified differences in the PT between toxin and venom samples in the presence and absence of aptamers. Measurements of PT were undertaken by first adding 100 µl of Calcium Rabbit Brain Thromboplastin (Diagnostic Reagents Ltd, UK) to a glass test tube (10 x 75 mm) and incubating at 37 ⁰C for 60-120 seconds in a water bath. Next, 50 µl of PPP or cryo-AHF was spiked with 50 µl containing 0.6 ng of Ancrod, Batroxobin, C. rhodostoma venom or B. atrox venom, or saline solution as the negative control. All samples were also repeated in the presence of 1 pM of the selected ssDNA aptamers or 0.5 µg of the commercial antivenoms, following preincubation at 37 ⁰C for 5 minutes. Time measurements were commenced, tubes were gently tilted at regular intervals (returning to the water bath between tilting), and the time for the formation of a clot was recorded.
All experiments were performed in duplicate. We did not use FFP for quantification of the PT as abundant platelets have been shown to interact with test reagents and increase the concentration of phospholipids, resulting in false shortening of clotting times (80).

Activated Partial Thromboplastin Time (aPTT)
To measure the inhibitory capability of selected aptamers against toxins acting on the intrinsic and tube, and the tube was gently tilted until the resulting clot time was recorded. All experiments were performed in duplicate. As with PT, we did not use FFP for quantification of the aPTT due to the potential for interference by platelets.

ssDNA aptamer selection
Here, we report the identification of ssDNA aptamers that target the recombinantly expressed SVSPs toxins Ancrod from C. rhodostoma and Batroxobin from B. atrox. To do so, we employed a parallel process consisting of 14 rounds of SELEX selections against each toxin target using a ssDNA library pool of 10 17 random sequences. The selection process was monitored throughout by PCR amplification of the eluted DNA and agarose gel electrophoresis ( Figure 1A and 1B), and by measuring the fluorescence intensity of the eluted ssDNA solution ( Figure 1C and 1D).
Our findings revealed that initial DNA recovery against each target was low, though this increased with progressive rounds of selection cycles, until a marked increase was observed at round 11, indicating enrichment of the DNA pool with sequences specific to the toxins coupled to the beads ( Figure 1).
Notably, the patterns of fluorescent intensities obtained from the sequential selection rounds were highly comparable between the two toxin targets (Figure 1). Counter selection (CS) was performed at round 12 to eliminate non-specific ssDNA interacting with the sepharose beads instead of the toxins, and resulted in an anticipated drop in the fluorescence intensity of the recovered DNA ( Figure 1C and   1D). Thereafter, two additional rounds of SELEX selection were performed to secondarily enrich toxinspecific aptamers, which resulted in an increase in fluorescence intensity at round 13, but no further increase at round 14, suggesting that target binding sites might be saturated ( Figure 1C and 1D).
Consequently, SELEX selection was stopped and the resulting ssDNA from the last round (round 14) was cloned for downstream aptamer identification.
Cloning was performed by ligation into a cloning vector and transformation into E. coli competent cells, with validation performed on ten random colonies to demonstrate that the inserted DNA (300 bp) was successfully ligated into the pCR2.1-TOPO cloning vector (3956 bp) ( Figure 1E and 1F).
Thereafter, 100 colonies were picked for each toxin target, and subjected to PCR followed by Sanger sequencing. Bioinformatic analyses of the resulting DNA identified various identical sequences, indicating enrichment of the DNA pool, and resulted in thirteen and nine unique aptamer sequences for Ancrod and Batroxobin, respectively (Table 1).

Dissociation constants (Kds) of identified aptamers
The dissociation constants (Kds) for each of the selected ssDNA aptamers, thirteen against Ancrod and

Quantifying aptamer binding to recombinant toxins and venom by ALISA
An ALISA was next used to quantify the binding levels of the two highest affinity aptamers (Ancrod-55 and Batroxobin-26) against each of the recombinant toxins, and the native venoms from which the recombinant toxins were derived (i.e., C. rhodostoma and B. atrox venom) in a reciprocal and comparative manner. The results of these binding assays, which were performed at two aptamer concentrations (0.4 nM and 2 nM), revealed that the two aptamers tested exhibited considerable cross-recognition and binding to the two SVSP toxins (i.e., irrespective of which toxin the aptamer was selected against) and the two homologous venoms (Figure 3). Indeed, the aptamer Batroxobin-26 exhibited highly comparable binding levels to Batroxobin, Ancrod and the two native snake venoms ( Figure 3B). These binding levels were, however, lower than those obtained with the Ancrod-55 aptamer for all comparisons ( Figure 3A).
Next, the ALISA was used to explore the capability of each aptamer to bind to a panel of 11 distinct venoms that exhibit considerable variation in toxin composition and were sourced from taxonomically diverse snake species (81). The binding patterns obtained from these experiments suggest that the two aptamers, each selected against a single SVSP toxin, recognise distinct snake venoms to a similar extent as the toxins and homologous snake venoms described above. Thus, at the 2 nM concentration

Aptamers inhibit toxin-induced fibrinogenolysis and fibrinogen depletion
Many SVSP toxins exert a functional activity akin to thrombin by acting to cleave the α-or/and βchains of fibrinogen, typically resulting in the release of fibrinopeptides rather than fibrin and thus ultimately contributing to dysregulation of coagulation via depletion of fibrinogen (82,83). To assess whether the binding exhibited by the highest affinity ssDNA aptamers were capable of inhibiting the functional activities known from SVSPs, we first used degradation SDS-PAGE to explore protection against fibrinogenolysis. Our findings revealed that, in line with the known functional activities of native ancrod and batroxobin and recent work (76) both recombinantly expressed toxins cleaved the α-chain of fibrinogen ( Figure 4A and 4B). We then assessed the inhibitory capability of the two candidate ssDNA aptamers at preventing this functional activity. Both Ancrod-55 and Batroxobin-26 aptamers protected against fibrinogenolysis caused by the corresponding recombinant toxin, resulting in visualisation of the α-chain of fibrinogen on the gel ( Figure 4A and 4B). We next sought to explore whether these two aptamers could also protect against native venom, which might contain additional related (or unrelated) proteins that could contribute to global fibrinogenolysis. Despite this possibility, our data showed that the aptamer Ancrod-55 inhibited the fibinogenolytic effect of C. rhodostoma venom on both the α-and β-chains, while Batroxobin-26 aptamer prevented B. atrox venom-induced cleavage of the α-chain of fibrinogen ( Figure 4C and 4D). These findings suggest that the binding specificities obtained during aptamer selection translate into inhibitory capabilities.
The clotting time of diluted plasma with a standard thrombin concentration is inversely related to the fibrinogen concentration (79). To further explore the inhibitory effect of the selected high-affinity aptamers against both SVSP toxins and native snake venom, we measured fibrinogen depletion using However, the aptamers were more effective in reducing the consumption of fibrinogen in both PPP and Cryo-AHF, which may at least be partly due to the depletion of platelets in these samples, and despite the toxins seemingly exhibiting increased potency in the Cryo-AHF sample ( Figure 5C). In both cases, however, the two aptamers inhibited the depletion of fibrinogen stimulated by both toxins and native venoms to near control levels (PPP, 2.7-2.9 g/L vs 3.45 g/L g/L; Cryo-AHF, 1.8-2.2 g/L vs 2.25 g/L, respectively) ( Figure 5B and 5C). Crucially, irrespective of the plasma sample used, the aptamers showed highly comparable inhibitory potencies with those obtained with gold-standard commercial antivenoms, as evidenced by the resulting equivalent fibrinogen concentrations observed ( Figure 5).

Aptamers reduce toxin-and venom-induced prolongations of the PT and aPTT
The PT measures the time for citrated plasma to clot and specifically assesses the clotting capability of the extrinsic and common coagulation cascades. A prolonged PT can result from an absence or deficiency of one of the clotting factors X, VII, V, II or fibrinogen. Our recombinant toxin/native venom spiking experiments demonstrated that both Ancrod and Batroxobin, and the corresponding venoms from C. rhodostoma and B. atrox, result in a prolongation of the PT in PPP compared to the saline control (i.e., >18 seconds vs ~14 seconds, respectively) ( Figure 6A). Noticeably, when co-incubated with the corresponding toxins/native venoms, the aptamers Ancrod-55 and Batroxobin-26 caused a substantial reduction in the PT compared with the toxin/native venom-only samples, resulting in the PT approaching the level of control readings (14.5-15.5 seconds vs ~14.0 seconds, respectively), and thus demonstrating inhibition of toxin activity ( Figure 6A). These reductions were highly comparable, and thus equipotent, to those obtained with commercially available antivenoms (14-16 seconds).
Contrastingly, when repeated in the Cryo-AHF samples, which contain the clotting factors XIII, XI, VIII, V, Fibrinogen and vWF, neither of the toxins or snake venoms consistently affected the PT ( Figure 6B).
We antivenoms consisting of monoclonal antibodies (mAbs) selected for their desirable specificities show much promise in this regard, particularly since different antibody formats can be adapted to specific pharmacodynamic and pharmacokinetic needs (84). However, to date, little effort has been devoted to exploring antibody alternatives despite outstanding concerns relating to the potential cost of mAbbased snakebite treatments. In other fields, ssDNA aptamers have emerged as promising alternatives to antibodies for both therapeutic application and their use as diagnostic tools (38)(39)(40), and several ssDNA aptamers have been successfully selected and applied against a wide range of targets including small molecules (85) and toxins (37,(86)(87)(88)(89). Despite this prior research, ssDNA aptamers remain largely unexplored for their potential utility as snakebite therapeutics. Thus, in this proof-of-concept study we sought to select ssDNA aptamers against fibrinogenolytic SVSP toxins found in medically important viper venoms, and to evaluate their potential utility as novel toxin-inhibiting molecules via in vitro cross-reactivity and neutralisation studies. Nonetheless, the proof-of-concept findings described here reveal that rationally selected ssDNA aptamers hold great value for exploration as novel inhibitory molecules capable of broadly inhibiting snake venom toxins, such as SVSPs. However, the potential therapeutic utility of aptamers does not come without a number of challenges. One major limitation is that ssDNA aptamers are susceptible to nucleases and this, combined with their small size, results in short half-lives in vivo, which can be as low as two minutes due to DNA degradation and rapid clearance (49). However, aptamers also possess great flexibility and, as such, can be modified to increase their stability and their resulting halflife, with conjugation to a higher molecular weight protein carrier, for example, liposomes, cholesterol or PEG, previously been demonstrated to successfully decrease the clearance rate from plasma to acceptable levels (57,59,91). Other limitations include the initial selection process often being labour intensive and time-consuming, though downstream production of the resulting identified ssDNA aptamers offers a number of advantages over mAb production, for example, due to their size and chemical synthesis. Finally, to date, it is worth re-emphasising that only a single aptamer has passed through an FDA-approval process (Macugen® [Pegaptanib]) (62), and as such regulatory challenges are likely to remain for exploiting ssDNA aptamers as new therapeutics.
Despite these limitations, aptamers possess several clear advantages in terms of therapeutic characteristics over the polyclonal antibodies currently used for treating snakebite. For example, aptamers are chemically synthesised and thus do not require the use of ethically and financially costly experimental animals, and the resulting product exhibits a relative lack of batch to batch variation, unlike biologic antivenom (92). Further, aptamers exhibit desirable stability and are likely more resistant to harsh conditions than antibodies (93), which may be of particular benefit in cold-chain unstable parts of the tropics that suffer a high burden of snakebite. Finally, the large-scale cost of manufacturing aptamers seems likely to be relatively inexpensive compared with existing polyclonal antivenoms. Currently, manufacturing costs for aptamers range from USD 140-280 per dose (91), and while the dose efficacy of aptamers in the context of snakebite remains to be established, it seems likely that aptamer-based treatments could reduce the high costs currently borne by tropical snakebite victims, which often exceed USD 1,000 in sub-Saharan Africa, for example (31). Although much work needs to be undertaken before such benefits can be realised, the proof-of-concept data described herein add further weight to the potential utility of aptamers as future snakebite treatments and strongly advocate for their continued exploration as a novel therapeutic modality for inhibiting pathogenic snake venom toxins.   The data on the left shows the binding levels obtained against the two recombinant toxins that the aptamers were selected against and the corresponding snake venoms that these toxins are derived from. The data on the right shows binding levels obtained against a broad panel of 11 distinct snake venoms. All data represent the mean of duplicates measured at 450 nm, and error bars represent the SD of the duplicate measurements.