Botulinum neurotoxins (BoNT) are established biopharmaceuticals for neuromuscular and secretory conditions based on their ability to block neurotransmitter release from neurons by proteolyzing specific soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) proteins. Recently, a mutant catalytic domain of serotype E (LC/E) exhibiting 16 mutations was reported to cleave the phosphatase and tensin homolog (PTEN). This molecule represents an attractive new target in neurons as several reports support PTEN knockdown as a strategy to stimulate axonal regeneration after injury. Though this LC/E mutant was shown to cleave PTEN in primary neurons through lentivirus-based expression, its expression and functionality as mutated full-length BoNT/E have not been studied. Hence, we assembled the 16 mutations stepwise in a bacterial expression plasmid for LC/E and purified several multiple mutants of LC/E. Biochemical characterization showed that the 16-fold mutant did not exhibit a detectable activity toward SNAP-25 up to 10 µM final concentration while it displayed an EC50 of approximately 200 nM for PTEN, exceeding 1000-fold that for LC/E-wt on the native substrate SNAP-25. Unexpectedly, expression of the full length 16-fold mutated BoNT/E did not provide soluble protein, possibly due to an interference of the interaction between LC and the translocation domain. Reversion of individual mutations revealed the E159L and S162Q substitutions, critical for redirecting LC/E activity toward PTEN, as main culprits for the solubility issue. To overcome this problem, we applied a methodology proved successful years ago, harnessing a proteolytically inactive variant of BoNT type D (BoNT/Di) as neurospecific delivery system for cargo proteins. The fusion protein LCE-16x-BoNT/Di could be produced in sufficient yields. Activity tests using rat cerebellar granule neurons showed BoNT/E-like activity for LC/E-wt-BoNT/Di, but no PTEN-directed activity for LC/E-16x-BoNT/Di.
Read full abstract