Snakebite envenoming remains a devastating and neglected tropical disease, claiming over 100,000 lives annually and causing severe complications and long-lasting disabilities for many more^1,2. Three-finger toxins (3FTx) are highly toxic components of elapid snake venoms that can cause diverse pathologies, including severe tissue damage³ and inhibition of nicotinic acetylcholine receptors, resulting in life-threatening neurotoxicity⁴. At present, the only available treatments for snakebites consist of polyclonal antibodies derived from the plasma of immunized animals, which have high cost and limited efficacy against 3FTxs^5,6,7. Here we used deep learning methods to de novo design proteins to bind short-chain and long-chain α-neurotoxins and cytotoxins from the 3FTx family. With limited experimental screening, we obtained protein designs with remarkable thermal stability, high binding affinity and near-atomic-level agreement with the computational models. The designed proteins effectively neutralized all three 3FTx subfamilies in vitro and protected mice from a lethal neurotoxin challenge. Such potent, stable and readily manufacturable toxin-neutralizing proteins could provide the basis for safer, cost-effective and widely accessible next-generation antivenom therapeutics. Beyond snakebite, our results highlight how computational design could help democratize therapeutic discovery, particularly in resource-limited settings, by substantially reducing costs and resource requirements for the development of therapies for neglected tropical diseases.

The recognition of intracellular antigens by CD8 ⁺ T cells through T cell receptors (TCRs) is central for adaptive immunity against infections and cancer. However, the identification of TCRs from patient material remains complex. We present a rapid de novo minibinder (miBd) design platform leveraging state-of-the-art generative models to engineer miBds targeting the cancer-associated peptide-bound major histocompatibility complex (pMHC) SLLMWITQC/HLA-A*02:01 (NY-ESO-1). Incorporating in silico cross-panning enabled computational prescreening of specificity, and molecular dynamics simulations allowed for improved predictability of in vitro success. We identified a high-affinity NY-ESO-1 binder and confirmed its structure using cryo–electron microscopy, which, when incorporated in a chimeric antigen receptor, induced killing of NY-ESO-1 ⁺ melanoma cells. We further designed and validated binders to a neoantigen pMHC complex, RVTDESILSY/HLA-A*01:01, with unknown structure, demonstrating the potential for precision immunotherapy.

Animal venoms are complex mixtures containing peptides and proteins known as toxins, which are responsible for the deleterious effect of envenomations. Across the animal Kingdom, toxin diversity is enormous, and the ability to understand the biochemical mechanisms governing toxicity is not only relevant for the development of better envenomation therapies, but also for exploiting toxin bioactivities for therapeutic or biotechnological purposes. Most of toxinology research has relied on obtaining the toxins from crude venoms; however, some toxins are difficult to obtain because the venomous animal is endangered, does not thrive in captivity, produces only a small amount of venom, is difficult to milk, or only produces low amounts of the toxin of interest. Heterologous expression of toxins enables the production of sufficient amounts to unlock the biotechnological potential of these bioactive proteins. Moreover, heterologous expression ensures homogeneity, avoids cross-contamination with other venom components, and circumvents the use of crude venom. Heterologous expression is also not only restricted to natural toxins, but allows for the design of toxins with special properties or can take advantage of the increasing amount of transcriptomics and genomics data, enabling the expression of dormant toxin genes. The main challenge when producing toxins is obtaining properly folded proteins with a correct disulfide pattern that ensures the activity of the toxin of interest. This review presents the strategies that can be used to express toxins in bacteria, yeast, insect cells, or mammalian cells, as well as synthetic approaches that do not involve cells, such as cell-free biosynthesis and peptide synthesis. This is accompanied by an overview of the main advantages and drawbacks of these different systems for producing toxins, as well as a discussion of the biosafety considerations that need to be made when working with highly bioactive proteins.

Each year, snakebite envenoming claims thousands of lives and causes severe injury to victims across sub-Saharan Africa, many of whom depend on antivenoms derived from animal plasma as their sole treatment option ¹ . Traditional antivenoms are expensive, can cause adverse immunological reactions, offer limited efficacy against local tissue damage and are often ineffective against all medically relevant snake species ² . There is thus an urgent unmet medical need for innovation in snakebite envenoming therapy. However, developing broad-spectrum treatments is highly challenging owing to the vast diversity of venomous snakes and the complex and variable composition of their venoms ³ . Here we addressed this challenge by immunizing an alpaca and a llama with the venoms of 18 different snakes, including mambas, cobras and a rinkhals, constructing phage display libraries, and identifying high-affinity broadly neutralizing nanobodies. We combined eight of these nanobodies into a defined oligoclonal mixture, resulting in an experimental polyvalent recombinant antivenom that was capable of neutralizing seven toxin families or subfamilies. This antivenom effectively prevented venom-induced lethality in vivo across 17 African elapid snake species and markedly reduced venom-induced dermonecrosis for all tested cytotoxic venoms. The recombinant antivenom performed better than a currently used plasma-derived antivenom and therefore shows considerable promise for comprehensive, continent-wide protection against snakebites by all medically relevant African elapids.

Recombinantly produced biotherapeutics hold promise for improving the current standard of care for snakebite envenoming over conventional serotherapy. Nanobodies have performed well in the clinic, and in the context of antivenom, they have shown the ability to neutralize long α-neurotoxins in vivo. Here, we showcase a protein engineering approach to increase the valence and hydrodynamic size of neutralizing nanobodies raised against a long α-neurotoxin (α-cobratoxin) from the venom of the monocled cobraNaja kaouthia. Based on the p53 tetramerization domain, a panel of anti-α-cobratoxin nanobody-p53 fusion proteins, termed Quads, were produced with different valences, inclusion or exclusion of Fc regions for endosomal recycling purposes, hydrodynamic sizes, and spatial arrangements, comprising up to 16 binding sites. Measurements of binding affinity and stoichiometry showed that the nanobody binding affinity was retained when incorporated into the Quad scaffold, and all nanobody domains were accessible for toxin binding, subsequently displaying increased blocking potency in vitro compared to the monomeric format. Moreover, functional assessment using automated patch-clamp assays demonstrated that the nanobody and Quads displayed neutralizing effects against long α-neurotoxins from both N. kaouthia and the forest cobra N. melanoleuca. This engineering approach offers a means of altering the valence, endosomal recyclability, and hydrodynamic size of existing nanobody-based therapeutics in a simple plug-and-play fashion and can thus serve as a technology for researchers tailoring therapeutic properties for improved neutralization of soluble targets such as snake toxins.