bioRxiv Science⌕ Search

Biology subjects

Miller, P. S.

Publications and source records attributed to Miller, P. S..

2 recordsLinked to original sources

Molecular mechanism of action of a blood brain barrier shuttle antibody

The transferrin receptor has emerged as a prime target for transcytosis of antibody shuttles from the bloodstream into the brain parenchyma to deliver therapeutic payloads that treat neurological disorders. However, how the transferrin receptor-antibody binding mode impacts avidity, degradation, pH sensitivity and delivery remains underexplored. To address this, we determined the cryo-EM structure of mouse transferrin receptor 1 bound to the model brain shuttle antibody 8D3. In combination with cell binding and localisation studies we show that 8D3 can structurally support small-scale inter-receptor cross-linking, such as in self-contained pairs, that cause avidity but do not induce receptor redistribution or degradation. The structure can also explain pH-dependent binding modes, and we show how some of these antibody variants with graded sensitivities regulate brain penetration in vivo. Overall, our study illuminates how distinct molecular features of antibody binding impact transferrin receptor behaviour and brain delivery to inform on future shuttle design.

neuroscience↗

A robust platform for recombinant production of animal venom toxin modulators of ion channels

Background and PurposePeptide toxins isolated from animal venom are potent and selective modulators of ion channels, and promising therapeutic leads. Due to intricate disulphide bridge patterns, they are often challenging to produce in standard laboratory settings, which limits engineering approaches to manipulate their structure-function properties. Given the low cost, wide accessibility, and versatility of recombinant expression systems for protein production, we set out to establish a straightforward high-yield strategy across a broad panel of peptide toxins from snakes, spiders and scorpions. Experimental Approach13 toxin DNA sequences were genetically fused to the C-terminus of either bivalent or monovalent human IgG1 antibody fragment crystallisable (Fc) domain sequences and expressed recombinantly from mammalian Expi293F cells. Affinity-purified proteins were evaluated by SDS-PAGE and size-exclusion chromatography (SEC). Function was assessed by Ca2+ flux assays on CN21 cells, or whole-cell electrophysiology on human embryonic kidney (HEK293T) cells, Chinese hamster ovary (CHO) cells, or dorsal root ganglion (DRG) neurons. Immunocytochemistry using HEK293T cells and mouse DRG neurons assessed Fc-toxin fusion binding. Key ResultsMonovalent Fc-toxin fusions consistently yielded 1-6 mg of pure, non-proteolytically cleaved protein from 20-70 ml cultures for several toxin types, including three-finger toxins from snakes, inhibitory cystine knot (ICK) toxins from spiders, and -toxins from scorpions, substantially surpassing the performance of unfused toxins or bivalent Fc-toxin fusions which gave low or no yield. Snake toxins targeting nicotinic acetylcholine receptors retained high single digit nanomolar inhibitory potency. Spider and scorpion toxins targeting the voltage-gated Na+ channel Nav1.7 retained pharmacological function and selectivity across a panel of five Nav subtypes, albeit with reduced potencies that did not exceed [~]70 nM. Conclusions and ImplicationsWe present a strategy for straightforward robust production of pure, monodisperse, and functional animal venom-derived toxins. This lowers the barrier to toxin production in a standard laboratory setting for follow-on engineering purposes.

pharmacology and toxicology↗