From channel proteins to industrial biomimetic membrane technology.

Biomimetic membrane technology, based on the use of nano-scale functional additives in the form of channel proteins or artificially made channel structures, represents an attractive way of optimizing membrane separation technology. However, the nano-scale nature of the additives inherently points to the challenge in up-scaling the membranes to square meter areas. Thus, the ability to up-scale the processes involved in manufacturing will be crucial for translating the protein/nano-science into technology. Here we discuss how highly selective aquaporin proteins can be used to enhance the performance of the classical thin film composite membrane, and how this can be used in relevant membrane elements and module form factors. A particular up-scaling challenge lies in securing large scale membrane protein production. We demonstrate our framework for making batch amounts which are compatible with the large scale production of biomimetic membranes for water purification based on the use of the E. coli expression system.

Purification and functional comparison of nine human Aquaporins produced in Saccharomyces cerevisiae for the purpose of biophysical characterization.

The sparse number of high-resolution human membrane protein structures severely restricts our comprehension of molecular physiology and ability to exploit rational drug design. In the search for a standardized, cheap and easily handled human membrane protein production platform, we thoroughly investigated the capacity of S. cerevisiae to deliver high yields of prime quality human AQPs, focusing on poorly characterized members including some previously shown to be difficult to isolate. Exploiting GFP labeled forms we comprehensively optimized production and purification procedures resulting in satisfactory yields of all nine AQP targets. We applied the obtained knowledge to successfully upscale purification of histidine tagged human AQP10 produced in large bioreactors. Glycosylation analysis revealed that AQP7 and 12 were O-glycosylated, AQP10 was N-glycosylated while the other AQPs were not glycosylated. We furthermore performed functional characterization and found that AQP 2, 6 and 8 allowed flux of water whereas AQP3, 7, 9, 10, 11 and 12 also facilitated a glycerol flux. In conclusion, our S. cerevisiae platform emerges as a powerful tool for isolation of functional, difficult-to-express human membrane proteins suitable for biophysical characterization.