Codon optimization of xylA gene for recombinant glucose isomerase production in Pichia pastoris and fed-batch feeding strategies to fine-tune bioreactor performance

Avoiding overflow metabolite formation in Komagataella phaffii fermentations to enhance recombinant protein production

BACKGROUND

Komagataella phaffii (K. phaffii), formerly known as Pichia pastoris, is a widely utilized yeast for recombinant protein production. However, due to the formation of overflow metabolites, carbon yields may be reduced and product recovery becomes challenging. This study investigates the impact of oxygen availability, different glucose concentrations and feeding strategies on overflow metabolite formation and recombinant protein production in K. phaffii.

RESULTS

High glucose concentrations in batch fermentation, as applied in literature, lead to substantial ethanol accumulation, adversely affecting biomass yield and product formation. Increasing dissolved oxygen setpoints does not significantly reduce ethanol formation, indicating that glucose surplus, rather than oxygen availability, drives overflow metabolism. Decreasing the initial glucose concentration to 5 g/L and adapting the feeding strategy of the fed-batch phase, effectively mitigates overflow metabolite formation, improving biomass yield by up to 9% and product concentration by 40% without increasing process time.

CONCLUSIONS

These findings underscore the importance of a suitable glucose-feeding strategy in K. phaffii fermentation processes and highlight the detrimental effects of overflow metabolites on productivity. By optimizing carbon source utilization, it is possible to enhance fermentation efficiency and recombinant protein production with K. phaffii.

A Systematic Review of the Potential of Pichia pastoris (Komagataella phaffii) as an Alternative Host for Biologics Production

The methylotrophic yeast Pichia pastoris is garnering interest as a chassis cell factory for the manufacture of recombinant proteins because it effectively satisfies the requirements of both laboratory and industrial set up. The optimisation of P. pastoris cultivation is still necessary due to strain- and product-specific problems such as promoter strength, methanol utilisation type, and culturing conditions to realize the high yields of heterologous protein(s) of interest. Techniques integrating genetic and process engineering have been used to overcome these problems. Insight into the Pichia as an expression system utilizing MUT pathway and the development of methanol free systems are highlighted in this systematic review. Recent developments in the improved production of proteins in P. pastoris by (i) diverse genetic engineering such as codon optimization and gene dosage; (ii) cultivating tactics including co-expression of chaperones; (iii) advances in the use of the 2A peptide system, and (iv) CRISPR/Cas technologies are widely discussed. We believe that by combining these strategies, P. pastoris will become a formidable platform for the production of high value therapeutic proteins.

Alter codon bias of the P. pastoris genome to overcome a bottleneck in codon optimization strategy development and improve protein expression

Apart from its role in translation, codon bias is also an important mechanism to regulate mRNA levels. The traditional frequency-based codon optimization strategy is rather efficient in organisms such as N. crassa, but much less in yeast P. pastoris which is a popular host for heterologous protein expression. This is because that unlike N. crassa, the preferred codons of P. pastoris are actually AU-rich and hence codon optimization for extremely low GC content comes with issues of pre-mature transcriptional termination or low RNA stability in spite of translational advantages. To overcome this bottleneck, we focused on three reporter genes in P. pastoris first and confirmed the great advantage of GC-prone codon optimization on mRNA levels. Then we altered the codon bias profile of P. pastoris by introducing additional rare tRNA gene copies. Prior to that we constructed IPTG-regulated tRNA species to enable chassis cells to switch between different codon bias status. As demonstrated again with reporter genes, protein yield of luc and 0788 was successfully increased by 4-5 folds in chassis cells. In summary, here we provide an alternative codon optimization strategy for genes with unsatisfactory performance under traditional codon frequency-based optimization.

Pentose metabolism and conversion to biofuels and high-value chemicals in yeasts

Pentose sugars are widespread in nature and two of them, D-xylose and L-arabinose belong to the most abundant sugars being the second and third by abundance sugars in dry plant biomass (lignocellulose) and in general on planet. Therefore, it is not surprising that metabolism and bioconversion of these pentoses attract much attention. Several different pathways of D-xylose and L-arabinose catabolism in bacteria and yeasts are known. There are even more common and really ubiquitous though not so abundant pentoses, D-ribose and 2-deoxy-D-ribose, the constituents of all living cells. Thus, ribose metabolism is example of endogenous metabolism whereas metabolism of other pentoses, including xylose and L-arabinose, represents examples of the metabolism of foreign exogenous compounds which normally are not constituents of yeast cells. As a rule, pentose degradation by the wild-type strains of microorganisms does not lead to accumulation of high amounts of valuable substances; however, productive strains have been obtained by random selection and metabolic engineering. There are numerous reviews on xylose and (less) L-arabinose metabolism and conversion to high value substances; however, they mostly are devoted to bacteria or the yeast Saccharomyces cerevisiae. This review is devoted to reviewing pentose metabolism and bioconversion mostly in non-conventional yeasts, which naturally metabolize xylose. Pentose metabolism in the recombinant strains of S. cerevisiae is also considered for comparison. The available data on ribose, xylose, L-arabinose transport, metabolism, regulation of these processes, interaction with glucose catabolism and construction of the productive strains of high-value chemicals or pentose (ribose) itself are described. In addition, genome studies of the natural xylose metabolizing yeasts and available tools for their molecular research are reviewed. Metabolism of other pentoses (2-deoxyribose, D-arabinose, lyxose) is briefly reviewed.

High-level expression of Thermobifida fusca glucose isomerase for high fructose corn syrup biosynthesis

Glucose isomerase (GIase), an efficient enzyme in the isomerization of d-glucose to d-fructose, has been widely used in food processing. In this study, an efficient expression system for a Thermobifida fusca GIase (GIase_Tfus) in Escherichia coli was firstly designed via a two-stage feeding strategy for improving expression level. The cultivation strategy was performed at an exponential feeding rate during the pre-induction phase, followed by a gradient-decreasing feeding rate at the induction phase in a 3-L fermenter. During this process, the effect of induction conditions and the complex nitrogen supplementation in feeding solutions on GIase_Tfus production were investigated and optimized. Under the optimal conditions, the yield of GIase_Tfus reached 124.1 U/mL, which is the highest expression level of GIase by recombinant E. coli reported to date. Additionally, the obtained GIase_Tfus was performed to produce high fructose corn syrup (HFCS) with conversion approacing 55 % from glucose (45 %, w/v) to fructose. According to the molecular dynamic simulation, a number of hydrogen bonds existed in the enzyme-substrate complex could stablilize the transient states, and a appreciate reaction distance of M1 catalytic site and oxygen atom of glucose make the reaction proceed easily, thus resulting in the efficient biosynthesis of HFCS. The function of GIase_Tfus renders it a valuable catalyst for HFCS-55 (containing 55 % d-fructose) manufacturing, the most favorable industrial product of HFCS. The efficient expression of GIase_Tfus and its efficient HFCS production lays the foundation for its proming industrial application.

Efficient Expression of Xylanase by Codon Optimization and Its Effects on the Growth Performance and Carcass Characteristics of Broiler

Simple Summary

The aim of this work was to combine xylanase expression and broiler production. The xylanase (XynB) gene from Trichoderma reesei was optimized to increase its expression level in Pichia pastoris. The maximum activity of xylanase (optiXynB) secreted by P. pastoris pPICZaA-optiXynB was 1299 U/mL after 96 h induction. The recombinase was highly specific towards birchwood xylan, beechwood xylan, and oat-spelt xylan. Dietary 1000 and 1500 IU/kg optiXynB significantly increased (p < 0.05) final weight and body weight gain; dietary 500, 1000, and 1500 IU/kg optiXynB significantly increased (p < 0.05) pre-evisceration weight, dressed percentage, and eviscerated weight compared with the control group. Results suggested that the optiXynB from P. pastoris pPICZaA-optiXynB has great application in broiler production.

Abstract

The aim of the present study was to improve the expression level of Trichoderma reesei xylanase (XynB) in Pichia pastoris through a codon optimization strategy and evaluate its effects on the growth performance and carcass characteristics of broiler. According to the codon bias of Pichia genome, the XynB gene from T. reesei was optimized and synthesized by whole gene assembly to improve its expression level in P. pastoris. Approximately 180 target mutations were successfully introduced into natural XynB. The maximum activity of xylanase (optiXynB) secreted by P. pastoris pPICZaA-optiXynB was 1299 U/mL after 96 h induction. Purified recombinant optiXynB had the molecular weight of 24 kDa. The optiXynB presented highest activity in pH 5.0 and 50 °C. The recombinase was highly specific towards birchwood xylan, beechwood xylan, and oat-spelt xylan. In the broiler experiment, a total of 200 Arbor Acre broilers (one day old) were randomly allocated into four groups fed with basal diets containing 0 (control group), 500, 1000, and 1500 IU/kg optiXynB. Dietary 1000 and 1500 IU/kg optiXynB significantly increased (p < 0.05) final weight and body weight gain; dietary 500, 1000, and 1500 IU/kg optiXynB significantly increased (p < 0.05) pre-evisceration weight, dressed percentage, and eviscerated weight compared with the control group. Inclusion of optiXynB in broiler diets linearly increased final weight, body weight gain, breast muscle weight and leg muscle weight, but linearly decreased feed conversion rate (p < 0.05). Furthermore, inclusion of optiXynB in broiler diets linearly and quadratically increased pre-evisceration weight, dressed percentage, and eviscerated weight (p < 0.05). The recombinant optiXynB from P. pastoris pPICZaA-optiXynB was beneficial in improving growth performance and carcass characteristics of broilers.

Optimizing assembly and production of native bispecific antibodies by codon de-optimization

When production of bispecific antibodies requires the co-expression and assembly of three or four polypeptide chains, low expression of one chain can significantly limit assembly and yield. κλ bodies, fully human bispecific antibodies with native IgG structure, are composed of a common heavy chain and two different light chains, one kappa and one lambda. No engineering is applied to force pairing of the chains, thus both monospecific and bispecific antibodies are secreted in the supernatant. In this context, stoichiometric expression of the two light chains allows for maximal assembly of the bispecific antibody. In this study, we selected a κλ body with suboptimal characteristics due to low kappa chain expression. Codon optimization to increase expression of the kappa chain did not improve bispecific yield. Surprisingly, progressive introduction of non-optimal codons into the sequence of the lambda chain resulted in lowering its expression for an optimal tuning of the relative distribution of monospecific and bispecific antibodies. This codon de-optimization led to doubling of the κλ body yield. These results indicate that assembly of different proteins into a recombinant complex is an interconnected process and that reducing the expression of one polypeptide can actually increase the overall yield.

Engineering strategies for enhanced production of protein and bio-products in Pichia pastoris: A review

Pichia pastoris has been recognized as one of the most industrially important hosts for heterologous protein production. Despite its high protein productivity, the optimization of P. pastoris cultivation is still imperative due to strain- and product-specific challenges such as promoter strength, methanol utilization type and oxygen demand. To address the issues, strategies involving genetic and process engineering have been employed. Optimization of codon usage and gene dosage, as well as engineering of promoters, protein secretion pathways and methanol metabolic pathways have proved beneficial to innate protein expression levels. Large-scale production of proteins via high cell density fermentation additionally relies on the optimization of process parameters including methanol feed rate, induction temperature and specific growth rate. Recent progress related to the enhanced production of proteins in P. pastoris via various genetic engineering and cultivation strategies are reviewed. Insight into the regulation of the P. pastoris alcohol oxidase 1 (AOX1) promoter and the development of methanol-free systems are highlighted. Novel cultivation strategies such as mixed substrate feeding are discussed. Recent advances regarding substrate and product monitoring techniques are also summarized. Application of P. pastoris to the production of biodiesel and other value-added products via metabolic engineering are also reviewed. P. pastoris is becoming an indispensable platform through the use of these combined engineering strategies.

Multi-omics data driven analysis establishes reference codon biases for synthetic gene design in microbial and mammalian cells

In this study, we analyzed multi-omics data and subsets thereof to establish reference codon usage biases for codon optimization in synthetic gene design. Specifically, publicly available genomic, transcriptomic, proteomic and translatomic data for microbial and mammalian expression hosts, Escherichia coli, Saccharomyces cerevisiae, Pichia pastoris and Chinese hamster ovary (CHO) cells, were compiled to derive their individual codon and codon pair frequencies. Then, host dependent and -omics specific codon biases were generated and compared by principal component analysis and hierarchical clustering. Interestingly, our results indicated the similar codon bias patterns of the highly expressed transcripts, highly abundant proteins, and efficiently translated mRNA in microbial cells, despite the general lack of correlation between mRNA and protein expression levels. However, for CHO cells, the codon bias patterns among various -omics subsets are not distinguishable, forming one cluster. Thus, we further investigated the effect of different input codon biases on codon optimized sequences using the codon context (CC) and individual codon usage (ICU) design parameters, via in silico case study on the expression of human IFNγ sequence in CHO cells. The results supported that CC is more robust design parameter than ICU for improved heterologous gene design.