Integration event induced changes in recombinant protein productivity in Pichia pastoris discovered by whole genome sequencing and derived vector optimization

Abstract

BackgroundThe classic AOX1 replacement approach is still one of the most often used techniques for expression of recombinant proteins in the methylotrophic yeast Pichia pastoris. Although this approach is largely successful, it frequently delivers clones with unpredicted production characteristics and a work-intense screening process is required to find the strain with desired productivity.ResultsIn this project 845 P. pastoris clones, transformed with a GFP expression cassette, were analyzed for their methanol-utilization (Mut)-phenotypes, GFP gene expression levels and gene copy numbers. Several groups of strains with irregular features were identified. Such features include GFP expression that is markedly higher or lower than expected based on gene copy number as well as strains that grew under selective conditions but where the GFP gene cassette and its expression could not be detected. From these classes of strains 31 characteristic clones were selected and their genomes sequenced. By correlating the assembled genome data with the experimental phenotypes novel insights were obtained. These comprise a clear connection between productivity and cassette-to-cassette orientation in the genome, the occurrence of false-positive clones due to a secondary recombination event, and lower total productivity due to the presence of untransformed cells within the isolates were discovered. To cope with some of these problems, the original vector was optimized by replacing the AOX1 terminator, preventing the occurrence of false-positive clones due to the secondary recombination event.ConclusionsStandard methods for transformation of P. pastoris led to a multitude of unintended and sometimes detrimental integration events, lowering total productivity. By documenting the connections between productivity and integration event we obtained a deeper understanding of the genetics of mutation in P. pastoris. These findings and the derived improved mutagenesis and transformation procedures and tools will help other scientists working on recombinant protein production in P. pastoris and similar non-conventional yeasts.Electronic supplementary materialThe online version of this article (doi:10.1186/s12934-016-0486-7) contains supplementary material, which is available to authorized users.

Effect of Plasmid Design and Type of Integration Event on Recombinant Protein Expression in Pichia pastoris

“…Copy number estimations pinpointed CN variations as the major influence on reporter protein fluorescence in our setting (Fig. 4), as supported by more detailed studies on multicopy integration orientation (26).…”

“…While these studies provide insights into P. pastoris clonal variation, our alternative study design focused on different aspects and overcame some experimental limitations. For instance, Schwarzhans et al noticed contamination in their sequencing efforts, possibly arising from mixed cultures due to the use of an auxotrophic HIS marker and the lack of dilution plating (26). Our integration plasmids (Fig.…”

“…A considerable decrease in the cost of next-generation sequencing and WGS over the last decade (54) has permitted comparisons of interclonal variation rather than sequencing only a single strain. Recently, Karl Friehs' group thoroughly investigated integration event-induced changes in recombinant protein productivity in P. pastoris, focusing on the nature of multicopy integration events and clones with abnormal colony morphology in two studies by Schwarzhans et al (26,27). While these studies provide insights into P. pastoris clonal variation, our alternative study design focused on different aspects and overcame some experimental limitations.…”

“…1) were based on phleomycin D1 (zeocin) antibiotic selection, as auxotrophic markers had resulted in mixed cultures in previous studies (26). We linearized the expression plasmids with the respective restriction enzymes and gel purified the bands (to avoid carryover of uncut vector or foreign DNA present [26,27]). P. pastoris cells were transformed in triplicate with each plasmid to test variability and potential influence of the transformation events.…”

“…With further decreases in the cost of these technologies, systematic studies of the underlying mechanisms for clonal variation in P. pastoris transformants have become feasible. Recently, milestone papers by the group of Karl Friehs investigated the influence of integration events on protein expression (26,27), providing insight into various mechanisms of multicopy integration (e.g., orientation of the cassettes [26]) and noncanonical integration events influencing growth phenotypes (27). While these integration events can have a profound influence on strain productivity and physiology, it remains partly unclear to what extent they occur, how they are influenced by expression cassette properties, and how they bias typical standard expression approaches in P. pastoris.…”