Expression, purification and characterization of human interferon-γ in Pichia pastoris

Abstract

Human interferon-γ (hIFN-γ) is a multifunctional protein known to possess immunoregulatory, antiviral and anticancer functions. In the present study, in order to explore the biological roles of hIFN-γ and its mechanisms of action, IFN-γ was cloned and expressed in Pichia pastoris (P. pastoris) under the control of alcohol oxidase promoter 1 (AOX1). The protein was secreted by two different signal peptides, the native secretion signal peptide of hIFN-γ and the Saccharomyces cerevisiae α signal peptide. Following 96 h of methanol induction, Tricine-SDS-PAGE Coomassie staining, western blot analysis and N-terminal protein sequencing revealed that the level of recombinant hIFN-γ (rhIFN-γ) secreted by the native secretion signal was barely detectable, while the α signal peptide secreted ~300 mg/l. rhIFN-γ was purified by Vivaflow 200, SP Sepharose Fast Flow and Vivaspin 2 ml, yielding >96% of a highly purified rhIFN-γ preparation, with a specific activity of 1 x 10(7)-1.4 x 10(7) IU/mg protein as determined by an antiviral assay. The results demonstrated that the experimental procedures developed are capable of producing a large quantity of active rhIFN-γ from P. pastoris.

Is Pichia pastoris a realistic platform for industrial production of recombinant human interferon gamma?

“…This study was based on the study by Wang et al (2014) using native and P. pastoris codon-optimised sequences of hIFNg and expanded the study using eight combinations of P. pastoris strains, vectors and sequences. Surprisingly, expression were orders of magnitudes lower than previously reported [10]. Based on our low expression/secretion results for all constructs and in agreement with very recently published results [13], we conclude that the P. pastoris expression/secretion system is at present not economically viable for commercial production of eukaryotic recombinant hIFNg (Animation 1).…”

“…The P. pastoris expression system has only recently been investigated for the commercial production of hIFNg, reporting secretion of around 300 mg L À1 with a specific activity of 1 Â 10 7 e1.4 Â 10 7 IU mg À1 , using the pPICZa vector and the alcohol oxidase (AOX1) promoter [10]. Our yields of hIFNg were much lower and these outcomes were not improved using additional strains, vectors, codon-optimised sequences.…”

“…The Western blot result, however, shown in Ref. [10] identified a 15 kDa band as hIFNg which is theoretically impossible unless the target protein was truncated and unglycosylated, which highlights the possibility of misidentification in the small-scale experiments. In contrast, the authors verified the nature of the secreted protein by N-terminal Table 4 hIFNg cDNA (¼ mRNA) copy number of GS115-pPIC9-COS1 P. pastoris transformants (Mean ± SD, n ¼ 3).…”

“…Our yields of hIFNg were much lower and these outcomes were not improved using additional strains, vectors, codon-optimised sequences. In contrast to Wang, Ren [10], protein quantification was carried out in this study via ELISA instead of using the Bradford assay and purity-determination by HPLC. As we did not use the same restriction sites, it could be possible that KEX2 and/or STE13 cleavage was not efficient, leading to ineffective secretion, because the processing of the signal sequence of human interferon gamma occurs in two steps, firstly cleavage by KEX2 then STE13.…”

“…To overcome these limitations, expression of recombinant hIFNg was attempted in various hosts like Saccharomyces cerevisiae 20B-12 [4], insect cells lines Spodoptera frugiperda, Spodoptera exigua, and Spodoptera litura. [5], Chinese hamster ovary (CHO) [3,6], wild-type mice strain C57BL/6 [7], rat cell line 3Y1-B [8], monkey and human cells [9]; however; high costs of cultivation and purification, contamination, low yields, low biological activity and short half-life of the product also adversely impacted on the use of these expression systems [10].…”