Engineering toxins for 21st century therapies

Poisons, toxungens, and venoms: redefining and classifying toxic biological secretions and the organisms that employ them

Despite extensive study of poisonous and venomous organisms and the toxins they produce, a review of the literature reveals inconsistency and ambiguity in the definitions of 'poison' and 'venom'. These two terms are frequently conflated with one another, and with the more general term, 'toxin.' We therefore clarify distinctions among three major classes of toxins (biological, environmental, and anthropogenic or man-made), evaluate prior definitions of venom which differentiate it from poison, and propose more rigorous definitions for poison and venom based on differences in mechanism of delivery. We also introduce a new term, 'toxungen', thereby partitioning toxic biological secretions into three categories: poisons lacking a delivery mechanism, i.e. ingested, inhaled, or absorbed across the body surface; toxungens delivered to the body surface without an accompanying wound; and venoms, delivered to internal tissues via creation of a wound. We further propose a system to classify toxic organisms with respect to delivery mechanism (absent versus present), source (autogenous versus heterogenous), and storage of toxins (aglandular versus glandular). As examples, a frog that acquires toxins from its diet, stores the secretion within cutaneous glands, and transfers the secretion upon contact or ingestion would be heteroglandular-poisonous; an ant that produces its own toxins, stores the secretion in a gland, and sprays it for defence would be autoglandular-toxungenous; and an anemone that produces its own toxins within specialized cells that deliver the secretion via a penetrating wound would be autoaglandular-venomous. Adoption of our scheme should benefit our understanding of both proximate and ultimate causes in the evolution of these toxins.

Model for studying Clostridium botulinum neurotoxin using differentiated motor neuron-like NG108-15 cells

Cancerous cell lines have traditionally shown low sensitivity to laboratory or pharmaceutical preparations of botulinum neurotoxin. The work presented here demonstrates that the mouse neuroblastoma/rat glioma hybrid cell line NG108-15 is capable of more sensitively detecting BoNT/A1 than any cell line previously described. This cell line has previously been described to have motor neuron like characteristics, therefore making it a good model to study BoNTs. Differentiation of NG108-15 cells in serum-free medium containing retinoic acid and purmorphamine dramatically increased sensitivity of the neurons to BoNT/A (EC(50) = ~16 LD(50) U). Additional pre-treatment with triasialoganglioside GT1B prior to toxin exposure reduced the EC(50) further to ~11 LD(50) U. Co-culture of the neurons with C2C12 myotubes also significantly increased BoNT/A sensitivity of NG108-15 cells (EC(50) = 26 U) in the absence of differentiation factors.

Novel application of human neurons derived from induced pluripotent stem cells for highly sensitive botulinum neurotoxin detection

Human induced pluripotent stem cells (hiPSC) hold great promise for providing various differentiated cell models for in vitro toxigenicity testing. For Clostridium botulinum neurotoxin (BoNT) detection and mechanistic studies, several cell models currently exist, but none examine toxin function with species-specific relevance while exhibiting high sensitivity. The most sensitive cell models to date are mouse or rat primary cells and neurons derived from mouse embryonic stem cells, both of which require significant technical expertise for culture preparation. This study describes for the first time the use of hiPSC-derived neurons for BoNT detection. The neurons used in this study were differentiated and cryopreserved by Cellular Dynamics International (Madison, WI) and consist of an almost pure pan-neuronal population of predominantly gamma aminoisobutyric acidergic and glutamatergic neurons. Western blot and quantitative PCR data show that these neurons express all the necessary receptors and substrates for BoNT intoxication. BoNT/A intoxication studies demonstrate that the hiPSC-derived neurons reproducibly and quantitatively detect biologically active BoNT/A with high sensitivity (EC(50) ∼0.3 U). Additionally, the quantitative detection of BoNT serotypes B, C, E, and BoNT/A complex was demonstrated, and BoNT/A specificity was confirmed through antibody protection studies. A direct comparison of BoNT detection using primary rat spinal cord cells and hiPSC-derived neurons showed equal or increased sensitivity, a steeper dose-response curve and a more complete SNARE protein target cleavage for hiPSC-derived neurons. In summary, these data suggest that neurons derived from hiPSCs provide an ideal and highly sensitive platform for BoNT potency determination, neutralizing antibody detection and for mechanistic studies.