Background

The production of active recombinant proteases at concentrations that allow for their use in downstream applications can be particularly challenging (Colige, 2020). The yeast Pichia pastoris has been successfully used as a eukaryotic expression system for foreign proteins (Macauley-Patrick et al., 2005). Unlike mammalian cells, Pichia does not require complex media or growth conditions, which makes its use cost-effective and scalable since high cell densities can be achieved in minimal media (Macauley-Patrick et al., 2005). Pichia strains are also genetically tractable, and the variety of available promoters and selectable markers facilitate gene expression (Daly and Hearn, 2005). Moreover, the presence of a eukaryotic protein synthesis pathway in Pichia allows high levels of posttranslational modification including glycosylation, proteolytic processing, and disulfide bond formation—processes that are limited in prokaryotic expression systems (Macauley-Patrick et al., 2005; Ahmad et al., 2014; Macek et al., 2019).

Generating large quantities of recombinant proteins is achievable with bioreactors; however, this is not feasible for most and, in general, recombinant proteins are produced in shaking flasks, similar to our approach (Schwettmann and Tschesche, 2001; Fernandez et al., 2013). The relatively small volumes of culture and limited cell densities that are ultimately achieved with this approach have made the isolation of recombinant proteases in large quantities difficult. Our goal was to develop a standardized protocol that would allow for the routine expression and isolation of a proteolytically active secreted metalloprotease, with high purity and in large enough quantities to facilitate downstream applications. We made use of the AOX1 methanol-induced promoter in Pichia that ensures that transcription is tightly regulated by a repression/derepression mechanism (Ahmad et al., 2014). Carbon sources such as glucose or glycerol, which are nutritional requirements for Pichia’s cell growth, cannot be used during the induction phase since both would repress transcription. The challenge arises during induction with methanol, since growth is severely inhibited preventing high levels of biomass production, which can ultimately negatively impact protein production. We found that the addition of sorbitol, a non-repressible and non-fermentable carbon source in yeast, during the methanol-induction phase, significantly improved cell viability and biomass production. The inclusion of sorbitol boosted protein production to milligram quantities. Although implementing non-repressing carbon sources such as alanine, mannitol, trehalose, and sorbitol had previously been explored, it has not been readily adopted (Inan and Meagher, 2001; Celik et al., 2009).

Here, we provide a detailed protocol for the expression and isolation of a proteolytically active metalloprotease from a neurotropic fungal pathogen, which has been applied to our work involving protein-nanoparticle conjugates as drug-delivery platforms for the central nervous system (Aaron and Gelli, 2020).