Synthetic antibodies block receptor binding and current‐inhibiting effects of α‐cobratoxin from Naja kaouthia

Abstract

Each year, thousands of people fall victim to envenomings caused by cobras. These incidents often result in death due to paralysis caused by α‐neurotoxins from the three‐finger toxin (3FTx) family, which are abundant in elapid venoms. Due to their small size, 3FTxs are among the snake toxins that are most poorly neutralized by current antivenoms, which are based on polyclonal antibodies of equine or ovine origin. While antivenoms have saved countless lives since their development in the late 18th century, an opportunity now exists to improve snakebite envenoming therapy via the application of new biotechnological methods, particularly by developing monoclonal antibodies against poorly neutralized α‐neurotoxins. Here, we describe the use of phage‐displayed synthetic antibody libraries and the development and characterization of six synthetic antibodies built on a human IgG framework and developed against α‐cobratoxin – the most abundant long‐chain α‐neurotoxin from Naja kaouthia venom. The synthetic antibodies exhibited sub‐nanomolar affinities to α‐cobratoxin and neutralized the curare‐mimetic effect of the toxin in vitro. These results demonstrate that phage display technology based on synthetic repertoires can be used to rapidly develop human antibodies with drug‐grade potencies as inhibitors of venom toxins.

1. INTRODUCTION

According to the World Health Organization, snakebite envenoming is a neglected tropical disease that affects millions of people worldwide. Annually, this disease causes 80,000 to 150,000 deaths and leaves around 400,000 individuals with permanent disabilities. ¹ Passive immunotherapy with animal‐derived antivenoms, containing polyclonal immunoglobulins or their derived fragments, isolated from animal plasma, remain the only specific treatment for snakebite envenoming more than 120 years after they were first demonstrated to be efficacious against envenomings. Animal‐derived therapies, however, can also induce adverse reactions triggered by contaminations or impurities, anti‐complementary activity, ² and immunogenicity of the heterologous proteins present in the antivenoms. ³ , ⁴ , ⁵ In turn, these effects can result in severe anaphylaxis and even death. ⁶ , ⁷ Moreover, the majority of the antibodies in current antivenoms target components with no or negligible contribution to venom toxicity. ⁴ , ⁸ Thus, only a fraction of the antibodies present in current antivenoms contribute to therapeutic efficacy, translating to low potency, additional cost, and potential safety issues. ⁹ , ¹⁰

The impact of snakebites is especially detrimental in developing countries in South and Southeast Asia, where incidence and mortality rates are among the highest in the world. Elapids of the Naja genus, for example, annually cause tens of thousands of snakebite‐related deaths. ¹¹ The venoms of these snakes are predominantly neurotoxic, ¹² containing post‐synaptically‐acting three‐finger toxins (3FTxs), such as Type I (short‐chain) and Type II (long‐chain) α‐neurotoxins. ¹³ , ¹⁴ Long‐chain α‐neurotoxins play an important role in the pathology of envenoming by binding tightly to nicotinic acetylcholine receptors (nAChRs) at neuromuscular junctions with nanomolar to sub‐nanomolar affinities. ¹⁵ This binding blocks the receptor and abolishes neurotransmission, leading to paralysis and death by respiratory arrest. ¹⁶ Since α‐neurotoxins have a high abundance in many elapid snake venoms and possess potent toxicity, ¹⁷ these toxins are considered key targets for antivenom development. ¹⁴ , ¹⁸ , ¹⁹

The monocled cobra (Naja kaouthia) is one of the most widely distributed venomous snakes in South and Southeast Asia. Recent proteomic studies of the venom of N. kaouthia, aimed at characterizing venom components and the binding and efficacy of a commonly used polyvalent antivenom, suggest that the α‐neurotoxins are the most medically important therapeutic targets in the venom, but that these are poorly recognized by antivenom antibodies. ²⁰ , ²¹ , ²² Therefore, improvement of therapy against envenomings by this snake species is warranted and could be achieved by improving the neutralization capacity against α‐neurotoxins. In this regard, recombinant antivenoms, based on defined oligoclonal mixtures of human antibodies, have been proposed as a cost‐competitive alternative to plasma‐derived antivenoms. ²³ These antivenoms may prove safer and more efficacious due to lower immunogenicity and the possibility of only including antibodies with proven therapeutic effects. ⁹ , ²⁴

To obtain neutralizing antibodies against α‐cobratoxin (α‐CTx) (the most potent and most abundant long‐chain α‐neurotoxin in the venom of N. kaouthia), efforts have focused on isolating antibodies or antibody fragments from immunized animals ²⁵ and phage‐displayed libraries based upon either naïve ²⁶ , ²⁷ , ²⁸ or immunized ²⁹ (human and/or animal) donors demonstrating neutralization of the toxin in vitro, ²⁵ and in some cases in vivo. ²⁸ , ²⁹ The 7.82 kDa (71 amino acid) α‐CTx is an ideal toxin to generate a proof of concept for the suitability of synthetic antibody repertoires for antivenom development, insofar as it has been structurally characterized, is commercially available in purified form, and existing antibodies offer the opportunity for comparison.

Toward this aim, phage display selections were conducted on α‐CTx using a highly validated synthetic antibody library based upon an optimized human framework. ³⁰ Toxin‐specific monoclonal antibodies were isolated, expressed, and characterized for their biophysical properties and functional neutralization of α‐CTx. The resultant synthetic antibodies could be produced at high yields, were highly specific, bound with sub‐nanomolar affinities, and neutralized the toxin in two distinct assays. The demonstration of potent neutralization mediated by human monoclonal antibodies from a synthetic library with no additional optimization suggests that such antibodies may find broad utility for recombinant antivenom development.

2. RESULTS

2.1. Phage display selections and antibody characterization

Using a synthetic antigen‐binding fragment (Fab) library, ³⁰ four rounds of selections against biotinylated α‐CTx (Latoxan, lot #L8114) were conducted. Following selection and isolation of individual clones, DNA fragments encoding Fab variable regions were amplified by polymerase chain reaction (PCR) and subjected to DNA sequence analysis. Subsequent analysis of the complementary determining regions (CDRs) identified 19 sequence‐unique Fab clones. The binding specificity of these unique clones was assessed using phage enzyme‐linked immunosorbent assays (ELISAs) (Figure 1a). To estimate affinities and prioritize clones for protein production in the IgG format, competitive phage ELISAs were conducted in the presence of 50 nM non‐biotinylated α‐CTx and compared to antibody binding in the absence of toxin. This resulted in the identification of six Fabs, whose binding was inhibited >50% (Figure 1b). These clones were selected for conversion to the human IgG1 format.

FIGURE 1.

Characterization of Fab‐phage clones and antibodies by ELISA. (a) Binding of unique Fab‐phage clones to immobilized α‐CTx and control proteins. (b) Binding inhibition of unique Fab‐phage clones to immobilized α‐CTx by 50 nM solution‐phase α‐CTx. Dashed line indicates 50% inhibition. (c) CDR sequences of the six IgGs for which the binding to immobilized α‐CTx was blocked by >50%. Amino acid positions are indicated at the top of each column and numbered according to IMGT standards. ³¹ (d) Visualization of expressed IgG by Coomassie‐stained SDS‐PAGE. (e) Plots of the binding signals of serially diluted IgGs to immobilized α‐CTx obtained by ELISA. (f) A summary table of key characteristics for the IgG proteins

2.2. IgG expression and characterization

Genes encoding the variable regions of the heavy and light chains of the six clones were sub‐cloned into pSCSTa hIg1 and hk IgG expression vectors designed to express heavy and light chains, respectively, ²⁹ and these were used to produce IgG proteins in Expi293 cells. The sequences encoding the CDRs of the six IgGs are shown in Figure 1c, and the purity of the antibodies obtained from a single‐step Protein A purification was verified to be >95% by SDS‐PAGE (Figure 1d). To evaluate antibody binding, ELISAs were performed by exposing five‐fold serial dilutions of IgG to immobilized biotinylated α‐CTx. Observed binding signals versus concentration of IgG were plotted, confirming dose‐dependent and saturable binding (Figure 1e), and EC₅₀ values were determined (Figure 1f). Five of the six IgGs (16034, 16036, 16037, 16038, and 16039), bound with EC₅₀ values in the sub‐nanomolar range, while exhibiting no binding to streptavidin or BSA control proteins.

Antibody binding kinetics were determined by biolayer interferometry (BLI) versus biotinylated α‐CTx comparing to an unrelated biotinylated protein of similar size as control. In accordance with EC₅₀ estimates of apparent affinity obtained by ELISA, BLI measurements confirmed single digit or sub‐nanomolar apparent K_D values for all antibodies, as shown in Figure 1f (BLI sensorgrams are provided in Figure S1). Notably, IgG 16036 possessed the lowest apparent K_D of 0.1 nM, arising from both the fastest k_on and slowest k_off, while the weakest binding IgG (16035) possessed a k_on value similar to other antibodies, but exhibited a k_off that was >60 faster than the tightest binding antibodies. In summary, estimates of affinity by two separate assays confirmed the tight binding and specificity of antibodies selected from a synthetic Fab phage library against α‐CTx, which suggests their suitability as potential neutralizing agents.

Expression yields for the six antibodies evaluated in triplicate ranged from 170 to 310 mg/L of culture (Figure 1f), and compared favorably to a yield ~200 mg/L for the benchmark IgG trastuzumab. Furthermore, by size exclusion chromatography, all IgGs were verified to elute predominantly as monodisperse peaks with monomeric fractions ranging from 88% to >98%, similar again to trastuzumab (Figure 1f) except for IgG 16038, for which there was no detectable elution peak, suggesting that it may interact with the column resin. Thus, in addition to possessing high affinity and specificity, high yields and favorable biophysical properties support the development of these antibodies for the treatment of snakebite envenoming.

2.3. α‐CTx: receptor blocking assay

To rank the IgGs according to their ability to block the binding between the nAChR and α‐CTx, a receptor blocking assay was conducted using the α7 subunit of the nAChR (α7‐AChR), excluding IgG 16035 due to its lower apparent affinity. As evident from Figure 2, the remaining five IgGs were able to prevent the binding between α7‐AChR and α‐CTx, albeit with different potencies. A dendrotoxin‐specific control antibody had no effect on the interaction between α7‐AChR and α‐CTx. Based on the blocking assay, IgGs 16036 and 16038 exhibited the highest potencies, with IC₅₀ values of 3.6 (95% CI: 2.5–5.1) and 6.2 (95% CI: 5.1–7.5) nM, respectively. The IC₅₀ values of the other antibodies ranged from 40 to 240 nM. Relating the IC₅₀ values to the concentration of α‐CTx used in the experiment (12.7 nM), 16036 inhibits one toxin molecule from binding to α7AChR per 0.57 IgG molecules, which is very close to the theoretical maximum being one toxin molecule per 0.5 IgG molecule, each IgG having two binding sites.

FIGURE 2.

Quantification of IgG‐mediated inhibition of the α‐CTx:α7‐AChR interaction. The binding of a constant amount of α‐CTx to immobilized α7‐AChR was evaluated following pre‐incubation with a range of serially diluted anti‐α‐CTx IgG to estimate the IC₅₀ value for the inhibition of the toxin:receptor interaction by each IgG. Fitted plots of binding versus IgG concentration enabled ranking of the top six IgGs according to how effectively the interaction between α‐CTx and α7‐AChR was inhibited

2.4. Electrophysiological neutralization assay

To evaluate whether the ability of the top two IgGs to block the binding between α‐CTx and α7‐AChR translates into a protection of nAChR function, an electrophysiological study was conducted using whole‐cell patch clamp assays. The effect of α‐CTx, preincubated with varing IgG concentrations, on nAChR‐dependent current was determined, employing an immortal cell line endogenously expressing the nAChR. A dendrotoxin‐specific IgG negative control had no current‐protective effect, but as shown in Figure 3, both anti‐α‐CTx IgGs potently abrogated the current‐inhibiting effect of α‐CTx. IC₅₀ values for the two IgGs were, just as for the receptor blocking assay, not significantly different (2.7 (95% CI: 1.7–4.6) and 2.8 (95% CI: 2.0–3.8) nM for 16036 and 16038, respectively, p = .99). Relating to the concentration of α‐CTx used in the study (4 nM), 16036 neutralized one toxin molecule per 0.6 IgG molecules. Once again, this value is very close to the theoretical maximum of one toxin molecule per 0.5 IgG molecules.

FIGURE 3.

In vitro neutralization of nAChR inhibition by α‐CTx. Assessment of the in vitro neutralization potency of the top two IgGs was performed via electrophysiological measurements using whole cell patch‐clamp. The blockade of ACh‐dependent current by purified α‐CTx was reversed by pre‐incubation of the toxin with serial dilutions of blocking IgG. Signals were normalized to full response (in the absence of α‐CTx and IgG). Concentration response plots were fitted, and IC₅₀ values were obtained from the fitted curves of normalized peak current versus IgG concentrations

3. DISCUSSION

The global health and economic burden that snakebite envenoming imposes is substantial, yet largely neglected. Existing therapies continue to employ century‐old technology, and while efforts to improve efficacy, breadth, and potency are underway, the field of envenoming therapy is in its infancy. To improve existing antivenoms, a broad‐based effort has been undertaken with the following aims: (1) defining venom compositions ¹³ , ²² ; (2) determining the toxic properties of individual venom components or fractions ²¹ ; (3) characterizing the toxicokinetics of toxins ³² ; (4) determining the pharmacokinetics of antivenoms ³³ ; (5) evaluating the efficacy of antivenoms ³⁴ ; and (6) identifying neutralizing recombinant antibodies to key toxins that effectively protect against envenomation. ²⁹ These combined studies have yielded key insights, revealing that 3FTxs of elapids (including both long‐ and short‐chain α‐neurotoxins, as well as cytotoxins), such as the ones from the monocled cobra, are amongst the most abundant ²² , ³⁵ , ³⁶ and lethal ³⁵ components of cobra venoms, but in general, are poorly immunogenic. In fact, combined venomics and lethality studies on venom fractions from N. kaouthia led investigators to conclude that “overall venom toxicity towards mice and lizards appears to result from α‐cobratoxin,” ³⁵ offering strong justification for the development of antibodies that neutralize this toxin.

Though previous studies using naïve, natural repertoires failed to generate antibodies that effectively neutralize venom by in vivo lethality challenge, ²⁶ , ²⁷ subsequent efforts have shown that high affinity antibodies to α‐CTx, selected from either a phage library derived from an immunized repertoire or obtained after a round of light‐chain shuffling, can provide complete protection during in vivo mouse challenges with lethal doses of α‐CTx or whole venom, similar to a polyclonal antivenom, when pre‐incubated prior to envenoming. ²⁸ , ²⁹ This suggests that an effective recombinant antivenom could, in some cases, theoretically be achieved with a single monoclonal antibody that can neutralize the central venom component(s), thus facilitating development efforts.

Of the six antibodies converted to the IgG format, five exhibited sub‐nanomolar EC₅₀ values. In close concordance, three of the six antibodies were determined to possess sub‐nanomolar apparent K_D values against immobilized α‐CTx. Two of the tightest binding antibodies were subsequently determined to both potently inhibit binding of α‐CTx to the nAChR and to protect ACh‐elicited currents in a functional neutralization assay performed on whole cells, both at molar ratios close to the theoretical minimum of two toxins per IgG molecule. Notably, their apparent affinities are on par with antibodies reported to completely protect against lethal doses of toxin in an in vivo mouse model. ²⁸ , ²⁹

Though the in vivo activity of these antibodies has not yet been established in a clinically relevant model, a single framework, recombinant antibody system that is amenable to re‐formatting and engineering for optimization of pharmacologically important features (i.e., high solubility, monodispersity, low poly‐reactivity, etc.) will facilitate future in vivo studies. Moreover, IC₅₀ values obtained for inhibition of both the binding and electrophysiological assays in this article are similar to those recently reported for antibodies shown to protect mice from lethal whole venom challenge in vivo, ²⁸ suggesting that these antibodies could possess similar in vivo activity or at least represent validated leads for evaluation in more relevant models.

The selection of recombinant antibodies from a phage‐displayed library based upon synthetic immune repertoires offers a proof of concept that high affinity antibodies that block toxin:receptor interactions and the resultant current‐inhibiting effects of α‐CTx can be obtained from naïve synthetic libraries; something that has only been possible from naïve libraries derived from natural sources when light‐chain shuffling has been applied or from libraries derived from immunized animals. ²⁶ , ²⁷ , ²⁸ Using a library that incorporates considerations regarding antibody developability into the design, while employing a human framework optimized for clinical use (long half‐life, high thermostability, and low immunogenicity), ³⁰ we aimed to show that highly potent neutralizing antibodies with biophysical properties that support in vivo use could be obtained without additional optimization. By focusing on potent neutralization of one of the most abundant and toxic components of venoms, we aim to eliminate poorly‐ or non‐neutralizing antivenom components, while providing a consistent and renewable supply of antibodies of defined sequence and optimizable characteristics. Together, this moves us closer to the goal of highly efficacious, cost‐effective, and safe biosynthetic oligoclonal antivenoms to address the unmet medical need that snakebites represent.

4. MATERIALS AND METHODS

4.1. Long‐chained alpha neurotoxins

Pure N. kaouthia α‐CTx (Uniprot ID: P01391) purchased from Latoxan (lot #L8114, Valence, France) was dissolved in 1X PBS (pH 7.4) to a final concentration of 1 mg/ml. The toxin was biotinylated using NHS‐PEG₄‐Biotin (Thermo Fisher Scientific, A39259), reacting at room temperature (RT) for 30 min at molar ratios of 1:1 and 1:2 (toxin:biotin) following the manufacturer's recommendations. Unreacted NHS‐PEG₄ ‐biotin was removed by washing with 3 ml PBS using a centrifugal concentrator (Amicon Ultra‐4, 3000 Da Molecular Weight Cut‐Off, Merck Millipore, UFC800324). Biotinylation of the toxin in the retentate was confirmed by an SDS‐PAGE migration shift assay, comparing the migration of toxin incubated in the presence or absence of an excess of streptavidin.

4.2. Phage display selection

Phage selections were conducted using a validated phage‐displayed, synthetic antigen‐binding fragment (Fab) library with a diversity of 3 × 10¹⁰ unique clones, as previously described, ³⁷ with the following modification: the phage library was mixed and incubated with 100 nM biotinylated α‐CTx in solution for 2 hr with gentle nutation. A 96‐well capture plate (Thermo Fisher Scientific) was coated overnight with a 2 μg/ml streptavidin solution, blocked with 0.2% BSA solution for 1 hr, and washed four times with PBS 0.05% Tween (PT) buffer, before capture of phage‐bound biotinylated toxin during a 15 min incubation. Captured phages were eluted with 0.1 M HCl and neutralized with 1 M Tris (pH 10), before infecting Omnimax cells at log phase growth for isolation and amplification of single Fab‐phage clones. Clones that bound to captured α‐CTx, but not to streptavidin, BSA, or a biotinylated control protein of similar size in monoclonal phage ELISAs, were subjected to DNA sequencing to decode antibody variable region genes.

4.3. Enzyme‐linked immunosorbent assay

To assess the specificity and estimate the affinity of Fab‐phage clones for α‐CTx, we used monoclonal phage ELISAs as previously described. ³⁸ Direct‐binding ELISAs were conducted by preparing, blocking, then coating plates with 100 nM biotinylated α‐CTx (as above), then exposing serial dilutions of IgG to toxin‐coated plates for 30 min, before developing with an anti‐kappa‐HRP antibody (1:10,000 dilution in PBT). Single‐point competitive ELISAs were similarly conducted using streptavidin‐coated plates, in which 100 nM biotinylated α‐CTx was captured as above. For competition, a sub‐saturating Fab‐phage concentration was pre‐incubated with 50 nM non‐biotinylated α‐CTx for 60 min before transferring the mixture to toxin‐coated plates, incubating for 15 min, then washing and developing with anti‐M13 antibodies (GE Healthcare) as previously described. ³⁸ Plots of IgG concentration versus binding signal were fitted using Prism 4 (GraphPad, v. 4.0) to estimate the EC₅₀ and IC₅₀ by nonlinear regression.

4.4. IgG production and purification

The DNA fragments encoding the V_L and V_H of selected clones were subcloned into pSCSTa‐hk and pSCST‐hIg1 vectors for transient expression of light chain and heavy chain, respectively. Following plasmid purification, 2.6 × 10⁶ cells/ml of Expi293F cells (Thermo Fisher Scientific) were transfected using FectoPRO reagent (Polyplus‐transfection), according to the manufacturer's instructions, and expression atmosphere of 5% CO₂ in air at 37°C for 4 days. Purification of antibodies from cell supernatants was performed via affinity chromatography using protein A Sepharose (Pierce, Thermo Fisher Scientific). IgGs were eluted from the column using IgG elution buffer (Pierce, pH 2.8) and neutralized with 1 M Tris (Invitrogen, pH 8). Samples were buffer exchanged with PBS (pH 7.5).

4.5. Biolayer interferometry

The kinetics of the antibodies binding to α‐CTx were determined using an Octet HTX instrument (ForteBio) at 1,000 rpm and 25°C. Biotinylated α‐CTx was first coated on streptavidin biosensors at 9 ng/ml in PBT assay buffer followed by a quench step of 180 s with 100 μg/ml biotin. To determine binding affinity, the α‐CTx‐coated sensors were equilibrated with assay buffer and then dipped for 600 s into anti‐α‐CTx IgGs at various concentrations in a three‐fold series from 1.8 to 50 nM prior to dissociation for 600 s. A biotinylated peptide with a similar size was used as a negative control. Binding response data were reference subtracted and were globally fitted with a 1:1 binding model using the ForteBio Data Analysis software 9.0.

4.6. Size‐exclusion chromatography

To assess monodispersity, 50 μg of IgG was injected onto a TSKgel BioAssist G3Wxl (Tosoh) pre‐equilibrated in a PBS mobile phase using an NGC chromatography system via a C69 autosampler (Biorad). Protein elution was monitored by absorbance at 215 nm during a 1.5 CV isocratic elution in PBS.

4.7. α‐CTx:receptor blocking assay

The α‐CTx:receptor blocking assay was adapted from Ratanabanangkoon et al. ³⁹ The α7 subunit of nAChR (α7‐AChR) was coated in a Maxisorp well, before adding different dilutions of the six IgGs, preincubated for 30 min at room temperature with 12.7 nM biotinylated α‐CTx in PBS, 0.1% BSA. Binding between α7‐AChR and α‐CTx was detected using streptavidin conjugated to europium (Perkin Elmer, 1244‐360), using a VICTOR Nivo plate reader. All concentrations were repeated in technical triplicates. The obtained signals were plotted relative to the full signal achieved for the toxin only (without addition of IgG), curves were fitted, and IC₅₀ values were determined using Prism 9 (GraphPad v. 9.2.0). A dendrotoxin‐specific IgG was used as a negative control.

4.8. Electrophysiological determination of neutralization of α‐CTx

Whole‐cell patch‐clamp experiments were carried out on an automated planar patch‐clamp system (Qpatch II, Sophion Bioscience), using 48‐channel patch chips with 10 parallel patch holes per channel (patch hole diameter ∼1 μm, resistance 2.00 ± 0.02 MΩ). A human‐derived rhabdomyosarcoma RD cell line (CCL‐136, from ATCC), endogenously expressing the muscle type nAChR, composed of the the α1, β1, δ, γ, and ε subunits, was used. For patching, an extracellular solution (145 mM NaCl, 10 mM HEPES, 4 mM KCl, 1 mM MgCl₂, 2 mM CaCl₂, and 10 mM glucose, pH adjusted to 7.4, and osmolality adjusted to 296 mOsm) and an intracellular solution (140 mM CsF, 10 mM HEPES, 10 mM NaCl, 10 mM EGTA, pH adjusted to 7.3, and osmolality adjusted to 290 mOsm) were employed.

In the experimental setup, nAChR‐mediated current was elicted by addition of 70 μM acetylcholine, corresponding approximately to the EC₈₀ value. After compound wash‐out, 2 U acetylcholinesterase was added to fully remove acetylcholine before the succeeding application. Acetylcholine was applied three times to allow for stabilization of the response, before the fourth application was used to evaluate the current‐inhibiting effect of α‐CTx preincubated with different concentrations of IgG protein. Four nM of unbiotinylated α‐CTx was used, yielding approximately 80% inhibition of the response obtained by application of acetylcholine. α‐CTx was preincubated with the IgG for at least 30 min at room temperature and added 5 min before acetylcholine application to allow for preincubation with the patched cells. The forth response (preincubated mixture) was normalized to the third response (full response). The relative responses were plotted as a function of the IgG concentration, and the data were fitted by nonlinear regression, from which ED₅₀/IC₅₀‐values for the IgGs were extracted. Data analysis was performed using Sophion Analyzer (Sophion Bioscience v 6.6) and Prism 9 (GraphPad v. 9.2.0). A dendrotoxin‐specific IgG was used as a negative control.

AUTHOR CONTRIBUTIONS

Shane Miersch: Investigation (equal); project administration (lead); supervision (equal); writing – original draft (equal). L. Guillermo de la Rosa: Conceptualization (equal); investigation (equal). Rasmus Friis: Investigation (equal). Line Ledsgaard: Investigation (equal); writing – review and editing (supporting). Kim Boddum: Investigation (supporting). Andreas Laustsen: Conceptualization (equal); funding acquisition (equal); investigation (equal). Sachdev Sidhu: Conceptualization (lead); funding acquisition (equal); project administration (equal).

CONFLICT OF INTERESTS

The authors declare no potential conflict of interest.

Supporting information

Figure S1 BLI sensorgrams which depict the interaction between recombinant IgGs and immobilized α‐CTx for the indicated antibodies.

Miersch S, de la Rosa G, Friis R, Ledsgaard L, Boddum K, Laustsen AH, et al. Synthetic antibodies block receptor binding and current‐inhibiting effects of α‐cobratoxin from Naja kaouthia . Protein Science. 2022;31(5):e4296. 10.1002/pro.4296

DATA AVAILABILITY STATEMENT

Data are available on request from the authors.