Exchanging the active site between phytases for altering the functional properties of the enzyme

Improving phytase production in Pichia pastoris fermentations through de-repression and methanol induction optimization

Pichia pastoris (Komagataella phaffii) is a fast-growing methylotrophic yeast with the ability to assimilate several carbon sources such as methanol, glucose, or glycerol. It has been shown to have outstanding secretion capability with a variety of heterologous proteins. In previous studies, we engineered P. pastoris to co-express Escherichia coli AppA phytase and the HAC1 transcriptional activator using a bidirectional promoter. Phytase production was characterized in shake flasks and did not reflect industrial conditions. In the present study, phytase expression was explored and optimized using instrumented fermenters in continuous and fed-batch modes. First, the production of phytase was investigated under glucose de-repression in continuous culture at three dilution factors, 0.5 d-1 , 1 d-1 , and 1.5 d-1 . The fermenter parameters of these cultures were used to inform a kinetic model in batch and fed-batch modes for growth and phytase production. The kinetic model developed aided to design the glucose-feeding profile of a fed-batch culture. Kinetic model simulations under glucose de-repression and fed-batch conditions identified optimal phytase productivity at the specific growth rate of 0.041 h-1 . Validation of the model simulation with experimental data confirmed the feasibility of the model to predict phytase production in our newly engineered strain. Methanol was used only to induce the expression of phytase at high cell densities. Our results showed that high phytase production required two stages, the first stage used glucose under de-repression conditions to generate biomass while expressing phytase, and stage two used methanol to induce phytase expression. The production of phytase was improved 3.5-fold by methanol induction compared to the expression with glucose alone under de-repression conditions to a final phytase activity of 12.65 MU/L. This final volumetric phytase production represented an approximate 36-fold change compared to the flask fermentations. Finally, the phytase protein produced was assayed to confirm its molecular weight, and pH and temperature profiles. This study highlights the importance of optimizing protein production in P. pastoris when using novel promoters and presents a general approach to performing bioprocess optimization in this important production host.

Insights to the Structural Basis for the Stereospecificity of the Escherichia coli Phytase, AppA

AppA, the Escherichia coli periplasmic phytase of clade 2 of the histidine phosphatase (HP2) family, has been well-characterized and successfully engineered for use as an animal feed supplement. AppA is a 1D-6-phytase and highly stereospecific but transiently accumulates 1D-myo-Ins(2,3,4,5)P₄ and other lower phosphorylated intermediates. If this bottleneck in liberation of orthophosphate is to be obviated through protein engineering, an explanation of its rather rigid preference for the initial site and subsequent cleavage of phytic acid is required. To help explain this behaviour, the role of the catalytic proton donor residue in determining AppA stereospecificity was investigated. Four variants were generated by site-directed mutagenesis of the active site HDT amino acid sequence motif containing the catalytic proton donor, D304. The identity and position of the prospective proton donor residue was found to strongly influence stereospecificity. While the wild-type enzyme has a strong preference for 1D-6-phytase activity, a marked reduction in stereospecificity was observed for a D304E variant, while a proton donor-less mutant (D304A) displayed exclusive 1D-1/3-phytase activity. High-resolution X-ray crystal structures of complexes of the mutants with a non-hydrolysable substrate analogue inhibitor point to a crucial role played by D304 in stereospecificity by influencing the size and polarity of specificity pockets A and B. Taken together, these results provide the first evidence for the involvement of the proton donor residue in determining the stereospecificity of HP2 phytases and prepares the ground for structure-informed engineering studies targeting the production of animal feed enzymes capable of the efficient and complete dephosphorylation of dietary phytic acid.

A Brief History of De Novo Protein Design: Minimal, Rational, and Computational

Protein design has come of age, but how will it mature? In the 1980s and the 1990s, the primary motivation for de novo protein design was to test our understanding of the informational aspect of the protein-folding problem; i.e., how does protein sequence determine protein structure and function? This necessitated minimal and rational design approaches whereby the placement of each residue in a design was reasoned using chemical principles and/or biochemical knowledge. At that time, though with some notable exceptions, the use of computers to aid design was not widespread. Over the past two decades, the tables have turned and computational protein design is firmly established. Here, I illustrate this progress through a timeline of de novo protein structures that have been solved to atomic resolution and deposited in the Protein Data Bank. From this, it is clear that the impact of rational and computational design has been considerable: More-complex and more-sophisticated designs are being targeted with many being resolved to atomic resolution. Furthermore, our ability to generate and manipulate synthetic proteins has advanced to a point where they are providing realistic alternatives to natural protein functions for applications both in vitro and in cells. Also, and increasingly, computational protein design is becoming accessible to non-specialists. This all begs the questions: Is there still a place for minimal and rational design approaches? And, what challenges lie ahead for the burgeoning field of de novo protein design as a whole?

Insights into the unfolding pathway and identification of thermally sensitive regions of phytase from Aspergillus niger by molecular dynamics simulations

Thermal stability is of great importance in the application of commercial phytases. Phytase A (PhyA) is a monomeric protein comprising twelve α-helices and ten β-sheets. Comparative molecular dynamics (MD) simulations (at 310, 350, 400, and 500 K) revealed that the thermal stability of PhyA from Aspergillus niger (A. niger) is associated with its conformational rigidity. The most thermally sensitive regions were identified as loops 8 (residues 83-106), 10 (161-174), 14 (224-230), 17 (306-331), and 24 (442-444), which are present on the surface of the protein. It was observed that solvent-exposed loops denature before or show higher flexibility than buried residues. We observed that PhyA begins to unfold at loops 8 and 14, which further extends to loop 24 at the C-terminus. The intense movement of loop 8 causes the helix H2 and beta-sheet B3 to fluctuate at high temperature. The high flexibility of the H2, H10, and H12 helices at high temperature resulted in complete denaturation. The high mobility of loop 14 easily transfers to the adjacent helices H7, H8, and H9, which fluctuate and partially unfold at high temperature (500 K). It was also observed that the salt bridges Asp110-Lys149, Asp205-Lys277, Asp335-Arg136, Asp416-Arg420, and Glu387-Arg400 are important influences on the structural stability but not the thermostability, as the lengths of these salt bridges did not increase with rising temperature. The salt bridges Glu125-Arg163, Asp299-Arg136, Asp266-Arg219, Asp339-Lys278, Asp335-Arg136, and Asp424-Arg428 are all important for thermostability, as the lengths of these bridges increased dramatically with increasing temperature. Here, for the first time, we have computationally identified the thermolabile regions of PhyA, and this information could be used to engineer novel thermostable phytases. Numerous homologous phytases of fungal as well as bacterial origin are known, and these homologs show high sequence similarity. Our findings could prove useful in attempts to increase the thermostability of homologous phytases via protein engineering.

Catalytic efficiency of HAP phytases is determined by a key residue in close proximity to the active site

The maximum activity of Yersinia enterocolitica phytase (YeAPPA) occurs at pH 5.0 and 45 °C, and notably, its specific activity (3.28 ± 0.24 U mg(-1)) is 800-fold less than that of its Yersinia kristeensenii homolog (YkAPPA; 88% amino acid sequence identity). Sequence alignment and molecular modeling show that the arginine at position 79 (Arg79) in YeAPPA corresponding to Gly in YkAPPA as well as other histidine acid phosphatase (HAP) phytases is the only non-conserved residue near the catalytic site. To characterize the effects of the corresponding residue on the specific activities of HAP phytases, Escherichia coli EcAPPA, a well-characterized phytase with a known crystal structure, was selected for mutagenesis-its Gly73 was replaced with Arg, Asp, Glu, Ser, Thr, Leu, or Tyr. The results show that the specific activities of all of the corresponding EcAPPA mutants (17-2,400 U mg(-1)) were less than that of the wild-type phytase (3,524 U mg(-1)), and the activity levels were approximately proportional to the molecular volumes of the substituted residues' side chains. Site-directed replacement of Arg79 in YeAPPA (corresponding to Gly73 of EcAPPA) with Ser, Leu, and Gly largely increased the specific activity, which further verified the key role of the residue at position 79 for determining phytase activity. Thus, a new determinant that influences the catalytic efficiency of HAP phytases has been identified.

Assembly of mutations for improving thermostability of Escherichia coli AppA2 phytase

We previously identified a number of mutations in Escherichia coli AppA2 phytase for enhancing its thermostability. The objective of the present study was to determine if these mutations (K46E, K65E, G103S, D112N, D144N, S209G, V227A, and G344D) could be sequentially added to further improve the thermostability of AppA2. Compared with the wild-type enzyme, two variants (D144N/V227A and D144N/V227A/G344D) out of the eight resulting mutants showed 15% enhancement in thermostability (as measured by residual activity after being heated at 80 degrees C for 10 min) and 4 to 5 degrees C increases in the melting temperatures (T (m)). Based on the structural predictions with a highly homologous AppA phytase, the substitution D144N introduces a side-chain-side-chain hydrogen bond, thereby stabilizing the loop region (Gln137-Asn144), and the V227A substitution might eliminate structural hindrance between Val222 and Val227 that face each other in the beta-hairpin structure. In addition, overall catalytic efficiency (k (cat)/K (m)) of the two mutants was also improved (P < 0.05) compared to the wild type. However, no further improvement in thermostability was observed by adding other mutations to D144N/V227A/G344D, which might result from unfavorable electrostatic interactions or structural perturbation. In conclusion, our results underscore the potential as well as difficulty of predicting synergistic effects of multiple mutations on thermostability within phytase.

Better library design: data-driven protein engineering

Data-driven protein engineering is increasingly used as an alternative to rational design and combinatorial engineering because it uses available knowledge to limit library size, while still allowing for the identification of unpredictable substitutions that lead to large effects. Recent advances in computational modeling and bioinformatics, as well as an increasing databank of experiments on functional variants, have led to new strategies to choose particular amino acid residues to vary in order to increase the chances of obtaining a variant protein with the desired property. Strategies for limiting diversity at each position, design of small sub-libraries, and the performance of scouting experiments, have also been developed or even automated, further reducing the library size.

A modified consensus approach to mutagenesis inverts the cofactor specificity of Bacillus stearothermophilus lactate dehydrogenase

Lactate dehydrogenase from Bacillus stearothermophilus is specific for NAD+. There have been several attempts to alter the cofactor specificity of this enzyme, but these have yielded enzymes with relatively low activities that still largely prefer NAD+. A modified consensus approach was used to create a library of phylogenetically preferred amino acids situated near the cofactor binding site, and variants were screened for their ability to utilize NMN+. A triple mutant (Mut31) was discovered that proved to be more catalytically efficient than wild-type. Mut31 was also better at utilizing NAD+ than the wild-type enzyme and was weakly active with NADP+ and NMN+. An analysis of single amino acid substitutions suggested that all three mutations worked in a concerted fashion to yield robust cofactor utilization. When two previously identified amino acid substitutions were introduced into the Mut31 background, the resultant quintuply substituted enzyme not only utilized NADP+ far better than the wild-type enzyme, it actually inverted its preference for NAD+ and NADP+.

Isolation, Characterization, and Molecular Cloning of the cDNA Encoding a Novel Phytase from Aspergillus niger 113 and High Expression in Pichia pastoris

Phytases catalyze the release of phosphate from phytic acid. Phytase-producing microorganisms were selected by culturing the soil extracts on agar plates containing phytic acid. Two hundred colonies that exhibited potential phytase activity were selected for further study. The colony showing the highest phytase activity was identified as Aspergillus niger and designated strain 113. The phytase gene from A. niger 113 (phyI1) was isolated, cloned, and characterized. The nucleotide and deduced amino acid sequence identity between phyI1 and phyA from NRRL3135 were 90% and 98%, respectively. The identity between phyI1 and phyA from SK-57 was 89% and 96%. A synthetic phytase gene, phyI1s, was synthesized by successive PCR and transformed into the yeast expression vector carrying a signal peptide that was designed and synthesized using P. pastoris biased codon. For the phytase expression and secretion, the construct was integrated into the genome of P. pastoris by homologous recombination. Over-expressing strains were selected and fermented. It was discovered that ~4.2 g phytase could be purified from one liter of culture fluid. The activity of the resulting phytase was 9.5 U/mg. Due to the heavy glycosylation, the expressed phytase varied in size (120, 95, 85, and 64 kDa), but could be deglycosylated to a homogeneous 64 kDa species. An enzymatic kinetics analysis showed that the phytase had two pH optima (pH 2.0 and pH 5.0) and an optimum temperature of 60 degrees C.

Putting engineering back into protein engineering: bioinformatic approaches to catalyst design

Complex multivariate engineering problems are commonplace and not unique to protein engineering. Mathematical and data-mining tools developed in other fields of engineering have now been applied to analyze sequence-activity relationships of peptides and proteins and to assist in the design of proteins and peptides with specified properties. Decreasing costs of DNA sequencing in conjunction with methods to quickly synthesize statistically representative sets of proteins allow modern heuristic statistics to be applied to protein engineering. This provides an alternative approach to expensive assays or unreliable high-throughput surrogate screens.

Enhancement of transglycosylation activity by construction of chimeras between mesophilic and thermophilic beta-glucosidase

The family 3 beta-glucosidase from Thermotoga maritima is a highly thermostable enzyme (85 degrees C) that displays transglycosylation activity. In contrast, the beta-glucosidase from Cellvibrio gilvus is mesophilic (35 degrees C) and displays no such transglycosylation activity. Both enzymes consist of two domains, an N-terminal and a C-terminal domain, and the amino acid identities between the two enzymes in these domains are 32.4 and 36.4%, respectively. In an attempt to identify the molecular basis underpinning the display of transglycosylation activity and the requirements for thermal stability, eight chimeric genes were constructed by shuffling the two parental beta-glucosidase genes at four selected borders, two in the N-terminal domain and two in the C-terminal domain. Of the eight chimeric genes constructed, only two chimeric enzymes (Tm578/606Cg and Tm638/666Cg) gave catalytically active forms and these were the ones shuffled in the C-terminal domain. For these active chimeric enzymes, 80% (Tm578/606Cg) and 88% (Tm638/666Cg) of their amino acid sequences originated from T. maritima. With regard to their thermal profiles, the two active chimeric enzymes, Tm578/606Cg and Tm638/666Cg, displayed profiles intermediate to those of the two parental enzymes as they were optimally active at 65 and 70 degrees C, respectively. These two chimeric enzymes were optimally active at pH 4.1 and 3.9, which is closer to that observed for the T. maritima enzyme (pH 3.2-3.5) than that for the C. gilvus enzyme (pH 6.2-6.5). Kinetic parameters for the chimeric enzymes were investigated with five different substrates including pNP-beta-D-glucopyranoside. The kinetic parameters obtained for the chimeric enzymes were closer to those of the T. maritima enzyme than to those of the C. gilvus enzyme. Transglycosylation activity was observed for both chimeric enzymes and the activity of the Tm578/606Cg chimera was at a level twice that observed with the T. maritima enzyme. This study is an effective demonstration of the usefulness of chimeric enzymes in altering the characteristics of an enzyme.

Site-directed mutagenesis of Aspergillus niger NRRL 3135 phytase at residue 300 to enhance catalysis at pH 4.0

Increased phytase activity for Aspergillus niger NRRL 3135 phytaseA (phyA) at intermediate pH levels (3.0-5.0) was achieved by site-directed mutagenesis of its gene at amino acid residue 300. A single mutation, K300E, resulted in an increase of the hydrolysis of phytic acid of 56% and 19% at pH 4.0 and 5.0, respectively, at 37 degrees C. This amino acid residue has previously been identified as part of the substrate specificity site for phyA and a comparison of the amino acid sequences of other cloned fungal phytases indicated a correlation between a charged residue at this position and high specific activity for phytic acid hydrolysis. The substitution at this residue by either another basic (R), uncharged (T), or acidic amino acid (D) did not yield a recombinant enzyme with the same favorable properties. Therefore, we conclude that this residue is not only important for the catalytic function of phyA, but also essential for imparting a favorable pH environment for catalysis.

Engineering of phytase for improved activity at low pH

For industrial applications in animal feed, a phytase of interest must be optimally active in the pH range prevalent in the digestive tract. Therefore, the present investigation describes approaches to rationally engineer the pH activity profiles of Aspergillus fumigatus and consensus phytases. Decreasing the negative surface charge of the A. fumigatus Q27L phytase mutant by glycinamidylation of the surface carboxy groups (of Asp and Glu residues) lowered the pH optimum by ca. 0.5 unit but also resulted in 70 to 75% inactivation of the enzyme. Alternatively, detailed inspection of amino acid sequence alignments and of experimentally determined or homology modeled three-dimensional structures led to the identification of active-site amino acids that were considered to correlate with the activity maxima at low pH of A. niger NRRL 3135 phytase, A. niger pH 2.5 acid phosphatase, and Peniophora lycii phytase. Site-directed mutagenesis confirmed that, in A. fumigatus wild-type phytase, replacement of Gly-277 and Tyr-282 with the corresponding residues of A. niger phytase (Lys and His, respectively) gives rise to a second pH optimum at 2.8 to 3.4. In addition, the K68A single mutation (in both A. fumigatus and consensus phytase backbones), as well as the S140Y D141G double mutation (in A. fumigatus phytase backbones), decreased the pH optima with phytic acid as substrate by 0.5 to 1.0 unit, with either no change or even a slight increase in maximum specific activity. These findings significantly extend our tools for rationally designing an optimal phytase for a given purpose.

Cloned and expressed fungal phyA gene in alfalfa produces a stable phytase

The phyA gene from Aspergillus ficuum that codes for a 441-amino-acid full-length phosphomonoesterase (phytase) was cloned and expressed in Medicago sativa (alfalfa) leaves. The expressed enzyme from alfalfa leaves was purified to homogeneity and biochemically characterized, and its catalytic properties were elucidated. The expressed phytase in alfalfa leaves retained all the biochemical properties of the benchmark A. ficuum phytase. Although the characteristic bi-hump pH optima were retained in the cloned phytase, the optimal pH shifted downward from 5.5 to 5.0. Also, the recombinant phytase was inhibited by the pseudo-substrate myo-inositol hexasulfate and also by antibody raised against a 20-mer peptide belonging to fungal phytase. The expressed phytase in alfalfa could also be modified by phenylglyoxal. Taken together, the results indicate that fungal phytase when cloned and expressed in alfalfa leaves produces stable and catalytically active phytase while retaining all the properties of the benchmark phytase. This affirms our view that "molecular biofarming" could be an alternative means of producing stable hydrolytic enzymes such as phytase.