- Email: [email protected]
Heterologous expression and functional characterization of a plant alkaline phytase in Pichia pastoris
Protein Expression and Purification 74 (2010) 196–203
Contents lists available at ScienceDirect
Protein Expression and Purification journal homepage: www.elsevier.com/locate/yprep
Heterologous expression and functional characterization of a plant alkaline phytase in Pichia pastoris Steven C. Johnson, Mimi Yang, Pushpalatha P.N. Murthy * Department of Chemistry, Michigan Technological University, 1400 Townsend Drive, Houghton, MI 49931, USA
a r t i c l e
i n f o
Article history: Received 26 April 2010 and in revised form 16 July 2010 Available online 22 July 2010 Keywords: Alkaline phytase Phosphate contamination Animal feed Heterologous protein expression Pichia pastoris Pollen grains
a b s t r a c t Phytases catalyze the sequential hydrolysis of phytic acid (myo-insositol hexakisphosphate), the most abundant inositol phosphate in cells. Phytic acid constitutes 3–5% of the dry weight of cereal grains and legumes such as corn and soybean. The high concentration of phytates in animal feed and the inability of non-ruminant animals such as swine and poultry to digest phytates leads to phosphate contamination of soil and water bodies. The supplementation of animal feed with phytases results in increased bioavailability to animals and decreased environmental contamination. Therefore, phytases are of great commercial importance. Phytases with a range of properties are needed to address the specific digestive needs of different animals. Alkaline phytase (LlALP1 and LlALP2) which possess unique catalytic properties that have the potential to be useful as feed and food supplement has been identified in lily pollen. Substantial quantities of alkaline phytase are needed for animal feed studies. In this paper, we report the heterologous expression of LlALP2 from lily pollen in Pichia pastoris. The expression of recombinant LlALP2 (rLlALP2) was optimized by varying the cDNA coding for LlALP2, host strain and growth conditions. The catalytic properties of recombinant LlALP2 were investigated extensively (substrate specificity, pH- and temperature dependence, and the effect of Ca2+, EDTA and inhibitors) and found to be very similar to that of the native LlALP2 indicating that rLlALP2 from P. pastoris can serve as a potential source for structural and animal feed studies. Ó 2010 Elsevier Inc. All rights reserved.
Introduction Phytases are a class of phosphatases that catalyze the sequential hydrolysis of phytic acid [myo-inositol hexakisphosphate (Fig. 1)] to less phosphorylated inositol phosphates, and in some cases, to inositol [1–6]. Inositol phosphates play multiple roles in biological processes including signal transduction and calcium regulation [7– 11]. Phytic acid is the most abundant inositol phosphate in cells [12], thus phytases, the primary enzymes responsible for the hydrolysis of phytic acid, play important roles in the metabolism of inositol phosphates. A number of phytases with varying structural and catalytic properties have been found in plants, yeast, bacteria, and animals [2,3]. The enzymes differ in pH optima (acidic and alkaline), catalytic mechanisms (histidine acid phosphatase-like phytase, purple acid phosphatase-like phytase, cysteine phosphatase-like phytase, and b-propeller phytase), substrate specificity, specificity of hydrolysis, metal ion requirements, susceptibility to inhibitors, and thermostability [1–4]. A novel alkaline phytase was detected in the Abbreviations: IB, inclusion body; AOX1, alcohol oxidase; LiAc, lithium acetate. * Corresponding author. Fax: +1 906 487 2061. E-mail address: [email protected] (P.P.N. Murthy). 1046-5928/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.pep.2010.07.003
pollen grains of Lilium longiflorum (LlALP) [13]. Cloning and sequence analysis identified two isoforms of alkaline phytase cDNA, LlAlp1 and LlAlp2, which are 1463 and 1533 bp long and encode proteins of 487 and 511 amino acids, respectively [14]. The calculated molar mass of LlALP1 and LlALP2 monomers are 53.8 and 56.2 kDa, respectively; the active enzyme exists as a homodimer. The deduced amino acid sequences revealed that the enzyme is a histidine phosphatase, it contains the signature heptapeptide of histidine phosphatases, –RHGXRXP– near the N-terminus [14]. In cereal grains and legumes phytate constitutes 3–5% of the dry weight of seeds [12]. During seed and pollen development, phytic acid is deposited in membrane-bound bodies as a salt of essential mineral ions such as potassium, calcium, magnesium, iron, and zinc and this then serves as a source of phosphate, inositol, and metal ions during the germination of seeds and pollen grains [12]. Although high concentrations of phytate are present in corn and soybean, the major components of animal feed, the phosphate, inositol, and essential metals in phytate complexes are unavailable to humans and monogastric animals, such as poultry, swine and fish, because they lack phytase [1–4,15,16]. Therefore, for optimum animal growth, supplementation of animal feed with phosphate and essential minerals, at additional cost, is necessary [16]. The excretion of undigested phytate from animals results in elevated levels of orga-
S.C. Johnson et al. / Protein Expression and Purification 74 (2010) 196–203
197
Materials and methods Strains, plasmids, and media
Fig. 1. Dodecasodium salt of phytic acid.
nophosphate in soil and water bodies downstream of agriculturally intensive areas [16]. To address these nutritional and environmental contamination issues, governments have passed legislation to supplement animal feed with phytases [4,15]. The efficacy of supplementing animal feed with phytase to increase the bioavailability of inositol and phosphates and decrease phosphorous discharge from animal farms has been demonstrated in several experiments [4,15,17]. The commercial importance of phytases in animal feed has resulted in the need for substantial quantities of an array of phytases with a range of catalytic and stability properties for use in different end applications [15]. Investigation of the biochemical properties of alkaline phytase from lily pollen indicates that the enzyme has unique catalytic properties and suggests that the enzyme has the potential to be useful as a feed and food supplement [18]. Our initial efforts at heterologous expression of rLlALP2 from lily pollen in E. coli resulted in low levels of expression and the accumulation of LlALP2 in inclusion bodies [14]. Although recombinant proteins in inclusion bodies (IB) are frequently inactive, alkaline phytases in IBs did exhibit catalytic activity [14]. Numerous efforts to solubilize alkaline phytase from IBs and refold them into the more active form were unsuccessful. Once denatured, rLlALP2 did not refold into the biologically active form under any of the numerous conditions [19] we tried (unpublished data); the fact that the enzyme is a homodimer with a total of 20 cysteine residues may have contributed to the refolding challenges we encountered. The methylotropic yeast, Pichia pastoris, has been shown to be a good expression system for high-level production of intracellular and secreted foreign proteins [20–22]. Heterologous expression of proteins in P. pastoris has many advantages over bacterial expression hosts: P. pastoris is capable of posttranslational modifications such as folding, disulfide bond formation, and glycosylation; expression of foreign proteins is driven by the tightly controlled and highly inducible alcohol oxidase (AOX1) promoter; proteins can be targeted to the extracellular media serving as the first step in purification; it is relatively inexpensive compared to other eukaryotic expression systems; the ability to grow to high cell densities coupled with high expression rates results in high protein yields, expression levels can range from mg to many g per L of growth culture [21–24]. In this paper, we describe the heterologous expression of a plant alkaline phytase (lily) in P. pastoris. Our aim was to investigate the expression of the active enzyme in a soluble form in P. pastoris. As a first step, we decided to express the enzyme intracellularly so as to avoid the additional problems associated with the Pichia secretary pathway. In an effort to optimize the expression of LlALP2, the LlAlp2 insert, host strain and growth conditions were varied. Catalytic properties of the recombinant enzyme (rLlALP2) expressed in P. pastoris were investigated and compared to the wild-type enzyme. To our knowledge, this is the first report of the heterologous expression of a phytase from a plant source.
Escherichia coli TOP10F0 , P. pastoris strains X-33 and KM71H, and expression vector pPICZA were purchased from Invitrogen (Carlsbad, CA). Primers were purchased from Integrated DNA Technologies (Coralville, IA) and GoTaq Flexi DNA polymerase was purchased from Promega (Madison, WI). Restriction enzymes and T4 DNA ligase were purchased from New England Biolabs (Ipswich, MA). Peptone, tryptone, yeast extract, glycerol, biotin, and methanol were purchased from Fisher (Pittsburgh, PA). Yeast nitrogen base was purchased from BD Biosciences (Sparks, MD). HisPur cobalt was purchased from Thermo Fisher (Rockford, IL). Low salt LB medium (1% tryptone, 0.5% yeast extract, and 0.5% NaCl) was used for propagation of E. coli (TOP10F0 ). Low salt LB medium plates (1.5% agar) containing kanamycin (25 lg/mL) (Research Products International Corp., Mt. Prospect, IL) were used to select for transformants of TOP10F0 . P. pastoris strains were grown in either BMGY (2% peptone, 1% yeast extract, 100 mM potassium phosphate, pH 6.0, 1.34% yeast nitrogen base, 4 105% biotin, and 1% glycerol) or BMMY medium (2% peptone, 1% yeast extract, 100 mM potassium phosphate, pH 6.0, 1.34% yeast nitrogen base, 4 105% biotin, and 1% methanol). YPD agar plates [1% yeast extract, 2% peptone, 2% dextrose, 2% agar, and various concentrations (100, 500, 1000 or 2000 lg/mL) of zeocin, as indicated] were used for selection of P. pastoris transformants.
Construction of expression vector Alkaline phytase cDNA (LlAlp2) was amplified via PCR from a previously constructed plasmid [14]. Forward primers for LlAlp2Ser (EcoRI restriction site in bold and the codons corresponding to Ser and Ala underlined) (50 -TAGCATGAATTCAAAAATGTCTGC GTTCTCGCTTCACGC-30 ) and LlAlp2-Ala (50 -TAGCATGAATTC 0 AAAAATGGCGTTCTCGCTTCACGC-3 ) were used in conjunction with a reverse primer (NotI restriction site in bold, stop codon italicized) containing a stop codon (50 -AAAAGGAAAAGCGGCCGC TCACAGTTCTGTCTTCTGGTTACCGGTG-30 ) or without a stop codon (50 -AAAAGGAAA AGCGGCCGCCAGTTCTGTCTTCTGGTTACCGGTG-30 ). Absence of the stop codon allowed for the incorporation of a cmyc epitope tag followed by a (His)6 tag at the carboxyl terminus of the protein. PCR amplification (35 cycles of 94 °C for 30 s, 60 °C for 50 s and 72 °C for 2 min and a final elongation of 5 min at 72 °C) yielded a 1500 bp product coding for LlAlp2. The LlAlp2 cDNA was purified using a QIAquick Gel Extraction Kit (Qiagen, Valencia, CA). The purified LlAlp2 cDNA insert and pPICZA were separately digested, first with NotI for 18 h after which EcoRI was added and the mixture incubated for an additional 4 h, and then ligated with T4 DNA ligase (16 h at 14 °C) per manufacturer’s instructions. Competent TOP10F0 were transformed with recombinant plasmids according to Sambrook [25] and positive transformants were selected on low salt LB agar plates containing 25 lg/mL zeocin. The insertion of LlAlp2 cDNA into pPICZA was confirmed by PCR. Plasmids were isolated using the WizardÒ Plus DNA Purification System according to the manufacturer’s instructions (Promega, Madison, WI). In brief, 5 mL of LB medium was inoculated with a single colony and grown overnight at 37 °C. Cells were collected by centrifugation (10 min at 1400g) and the plasmid extracted as suggested by the manufacturer. PCR amplification using plasmid DNA and LlAlp2-specific, full-length primers, used above, yielded a 1500 bp product corresponding to LlAlp2 cDNA. A second PCR amplification using plasmid DNA and plasmid-specific primers that flank the insertion site (forward primer: 50 -GAC TGG TTC
198
S.C. Johnson et al. / Protein Expression and Purification 74 (2010) 196–203
CAA TTG ACA AGC-30 ; reverse primer: 50 -GCA AAT GGC ATT CTG ACA TCC-30 ) was conducted and the amplified product was sequenced to confirm that insertion had occurred in the correct reading frame (Nevada Genomics Center, Reno, NV). Transformation of P. pastoris Recombinant plasmid was isolated from TOP10F0 using a Qiagen Plasmid Maxi Kit according to the manufacturer’s instructions and linearized with PmeI (37 °C, 16 h). Competent cells were prepared according to the Pichia EasySelect manual [26]. In brief, 500 mL of YPD was inoculated with 0.1 mL of an overnight P. pastoris culture and grown at 29 °C to an Abs600 of 1.4. Cells were collected by centrifugation (15 min at 1500g, 4 °C; cell pellet C-1) and subsequently washed with successively smaller volumes of ice-cold sorbitol (1 M; 500, 250 and 20 mL). The cell pellet was finally resuspended in ice-cold sorbitol (1.5 mL, 1 M) and aliquots of this cell suspension used for transformation. The transformation efficiency under these conditions was low therefore several pre-treatment conditions were tried (Table 1). Cell pellet C-1 was pretreated by incubating (30 min) in either YPD HEPES (200 mM, pH 8.0) containing dithiothreitol (25 mM) or YPD HEPES (200 mM, pH 8.0) containing dithiothreitol (10 mM) and lithium acetate (100 mM) before successive washings in smaller volumes of ice-cold sorbitol (1 M). Linearized plasmid (10 lg in 5 lL of H2O) and competent cells (40 lL) were combined in an electroporation cuvette (0.2 cm gap), incubated for 5 min on ice, and pulsed at 2 kV with a MicroPulser (Bio-Rad, Hercules, CA). Two different post-treatment conditions were tried: ice-cold sorbitol (1 M) or ice-cold YPD containing sorbitol (1 M, YPDS) was added to the cuvette immediately and the cells transferred to a 15 mL Falcon tube and incubated at 30 °C from 2 to 4 h without shaking. Aliquots (10, 50 and 250 lL) were plated on YPDS agar plates containing 100 lg/mL zeocin [26]; colonies were visible in 2 d. To make glycerol stock, positive P. pastoris colonies were grown for 16 h (29 °C) in 3 mL of YPD. Aliquots of cell cultures (850 lL) were mixed with sterile glycerol (150 lL), flash frozen in liquid N2, and stored at 80 °C. The insertion of LlAlp2 cDNA into P. pastoris was confirmed by PCR. Glycerol stocks were streaked on YPD plates containing zeocin (25 lg/mL) and grown for 2 d (29 °C). Colonies were lifted from the plate and the cells were lysed by suspending in H2O (10 lL) and lyticase solution (20 lL, 8 U, Sigma, St. Louis, MO) and the mixture was incubated at 30 °C for 30 min. Samples were then frozen at 80 °C for 15 min and thawed. PCR amplification using genomic DNA and LlAlp2-specific, full-length primers used above yielded a 1500 bp product corresponding to LlAlp2 cDNA. Expression of alkaline phytase in P. pastoris and cell lysis In order to achieve high yields of LlALP2 the following growth parameters were optimized: (1) influence of culture volume was examined by performing expression using either 12.5 mL or 25 mL of media in a 125 mL baffle flask. (2) Optimal methanol
addition was investigated by adding varying amounts of methanol (0.5%, 1%, 2%, and 3% once daily, 1% twice daily) during expression. (3) The time point after methanol induction that yields the maximum activity (1, 2, 3, 4, and 5 d) was investigated. Changes to the conditions are indicated in the figures. In subsequent experiments, the following set of optimized conditions were employed: glycerol stock of clones were streaked on YPD agar plates containing zeocin (100 lg/mL) and grown at 29 °C for 2 d. Individual colonies were used to inoculate 25 mL of BMGY in a 125 mL baffled flask and grown at 30 °C for 20 h. The cells were harvested by centrifugation at 3000g for 5 min and resuspended in BMMY (25 mL). Every 24 h, sterile methanol (250 lL) and sterile H2O (2.25 mL) were added to a total of 10% of the culture volume. Cells were harvested at indicated times by centrifugation (10,000g) and resuspended in lysis buffer (Tris–HCl, 50 mM, pH 7.4; PMSF, 1 mM; 2 mL of lysis buffer per g wet weight of cells). The wet cell weight of P. pastoris grown in 25 mL of BMGY medium usually varied from 1.7 to 1.8 g, a narrow range. Greater variations were observed only when samples were infected, in these cases the samples were discarded. To ensure complete cell lysis and release of alkaline phytase activity, two methods of cell disruption were examined, a MiniBeadbeater-8 (BioSpec, Bartelsville, OK) or a Vortex Genie 2 (Fisher, Pittsburgh, PA). Resuspended cell pellet (1 mL) was added to 0.5 mL of 0.5 mm zirconia/silica disruption beads (Research Products International Corp., Mt. Prospect, IL). Samples were lysed for 30 s followed by 30 s incubation on ice for a total of 6 cycles, centrifuged at 10,000g, and the supernatant dialyzed for 18 h. Samples were assayed for LlALP2 activity according to the procedure below. In our hands, disruption using the Vortex mixer at the highest speed possible released 5-fold more total activity than the Beadbeater. Cell lysate was centrifuged (10,000g, 15 min, 4 °C) to separate the soluble proteins from the cell debris. The supernatant, which contains the soluble proteins, was dialyzed against Tris–HCl buffer (10 mM, pH 7.4) for 18 h to remove phosphates and salts before purification or assaying for enzyme activity.
Purification of rLlALP2 by immobilized metal affinity chromatography (IMAC) on Co2+-resin The enzyme LlALP2 -Ala with the c-myc and (His)6 tags at the Cterminus was expressed in P. pastoris strain X-33 and purified on HisPur cobalt resin. The dialyzed soluble lystate (3 mL) was incubated with urea (2 M) for 30 min on ice. HisPur Cobalt resin (350 lL) was added to a chromatography column and the storage buffer was allowed to drain by gravity. The resin was equilibrated with Binding buffer (1 mL, Tris–HCl, 50 mM, pH 7.4; NaCl, 300 mM) and the urea-treated soluble lystate was loaded onto the resin. The resin was washed with Binding buffer (3 1 mL). LlALP2-Ala-c-myc-(His)6 was then eluted with Binding buffer (500 lL) containing increasing concentrations of imidazole (7 mM, 30 mM, 60 mM, and 200 mM). The fractions were assayed for alkaline phytase activity and separated using SDS–PAGE.
Table 1 Effect of various pre-treatment and post-transformation recovery conditions on transformation efficiency. Method
A [26] B [32] C [33] D [33,34]
Pre-treatment conditions
Transformants/lg DNA
Recovery conditions
Media
DTT (mM)
LiAc (mM)
Media
Time (h)
No pre-treatment YPD + HEPES (pH 8.0) YPD + HEPES (pH 8.0) YPD + HEPES (pH 8.0)
0 25 10 10
0 0 100 100
1 M sorbitol 1 M sorbitol 1 M sorbitol YPD + 1 M sorbitol
2 4 4 4
Cells were prepared as described in the ‘‘Materials and Methods” section. Pre-treatment and recovery conditions are indicated above.
10 50 200 500
S.C. Johnson et al. / Protein Expression and Purification 74 (2010) 196–203
Alkaline phytase activity assay Alkaline phytase activity was assayed by measuring the phosphate released from sodium phytate by the enzyme [27]. The reaction mixture contained Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM), KCl (0.5 M), CaCl2 (1 mM), NaF (10 mM), and aliquots of dialyzed cell lysate in a total volume of 500 lL. Samples were incubated at 37 °C for 1 h and the reaction was stopped with the addition of TCA (50% w/v, 100 lL). The denatured enzyme was pelleted at 10,000g for 2 min and 300 lL of the supernatant was added to 700 lL of the phosphate detection solution (6:1 ratio of 0.42% ammonium molybdate w/v in 0.5 M H2SO4 and 10% w/v ascorbic acid in H2O). The mixture was incubated for 1 h at 37 °C and the absorbance measured at 820 nm. Inorganic phosphate concentration was determined from a calibration curve using KH2PO4 as the standard. One unit of activity is defined as the release of 1 nmol of Pi from sodium phytate in 1 h. Standard assay conditions served as controls in the experiments; deviations are indicated in the figure legends. Vanadate was dissolved in Tris–HCl (100 mM, pH 8.0) and used immediately.
Results Construction and transformation of expression vectors The design of expression constructs for P. pastoris was guided by the following considerations: (1) the translational efficiency in P. pastoris is thought to be primarily dependent on the rate of initiation, which in turn, is influenced by the sequence in the 50 untranslated region of mRNA [28]. In P. pastoris, the consensus sequence for high initiation rates includes 4As immediately upstream of the start codon, AUG. Therefore, AAAAAUG was the sequence around the start codon in all constructs (Fig. 2). (2) The half-life of a protein is determined by the nature of the penultimate amino acid at the N-terminus (N-2 position), the N-end rule [29,30]. Two amino acids that stabilize proteins in yeast include Ala and Ser [29]; the wild-type cDNA from lily pollen (LlAlp2) codes for Ala in the N-2 position. Two isoforms of cDNAs that code for alkaline phytase, one coding for Ala in the N-2 position (LlAlp2-Ala) and another coding for Ser in the N-2 position (LlAlp2-Ser), were synthesized by PCR amplification of the full-length LlAlp2 from a LlAlp2containing plasmid using appropriate primers (forward primers
199
containing -AAAAATGTCT- and -AAAAATGGCG- near the start codon) (Fig. 2) [14]. The alkaline phytase-coding sequences, LlAlp2Ala and LlAlp2-Ser, were then inserted into the expression vector pPICZA between the EcoRI and NotI sites to yield two expression constructs, pPICZA-LlAlp2-Ala and pPICZA-LlAlp2-Ser. After transformation, PCR amplification of both constructs using LlAlp2-specific full-length primers yielded a 1500 kb fragment, the expected length of full length alkaline phytase cDNA. Subsequently, the constructs were amplified with plasmid-specific primers and sequenced from both directions. The data indicated that the alkaline phytase cDNAs were inserted in frame in pPICZA; additionally, no PCR-induced mutations were detected in the expression constructs. Initial efforts to transform P. pastoris strains by sorbitol pretreatment and post-transformation recovery in the presence of sorbitol (2 h) were not very successful in our hands; the transformation efficiency was low (10 transformants per lg of DNA) (Table 1A) [26]. Transformational efficiency of P. pastoris is highly influenced by chemical pre-treatment methods that render the cells competent for DNA uptake during electroporation, electroporation conditions, and post-transformation recovery conditions [26,31–36]. Therefore, different pre-treatment and recovery conditions were examined (described in the ‘‘Materials and Methods” section). Pre-treatment with a reducing agent such as dithiothreitol (DTT) increased transformational efficiency 5-fold (Table 1B) [31]. Transformational efficiency increased 20-fold when both DTT and lithium acetate (LiAc) were included in the pre-treatment medium (Table 1C); 200 transformants per lg of DNA was observed [33]. Although the exact role of DTT and LiAc in facilitating entry of foreign DNA into yeast cells is not known, it has been suggested that reducing agents and metal ions may change the integrity of cells by modifying the mannan-protein complex that impart rigidity to yeast cell walls [35,36]. The inclusion of YPD medium during the recovery phase increased the transformational efficiency by an additional 2.5-fold; nearly 500 transformants were produced (Table 1D). All subsequent transformations were conducted under the following conditions: pre-treatment in medium containing DTT (10 mM) and LiAc (100 mM) and recovery in YPD medium containing sorbitol for a 4-h period. A number of host strains of P. pastoris that differ in their ability to utilize methanol and express foreign proteins are available for foreign gene expression [21,22]. Therefore, two strains of P. pastoris, X-33 (Mut+) and KM71H (Muts), were transformed with PmeI-
Fig. 2. Cloning of LlAlp2 into P. pastoris. (I and II): LlAlp2 cDNA coding for either Ala (I) or Ser (II) in the N-2 position was inserted between the EcoRI and NotI restriction sites into pPICZA. (III) LlAlp2 cDNA coding for Ala in the N-2 position and lacking a stop codon at the carboxyl end was inserted between the EcoRI and NotI restriction sites into pPICZA. The lack of a stop codon in LlAlp2 cDNA allowed the c-myc and (His)6 tag from the pPICZA plasmid to be fused at the C-terminus. The resulting recombinant constructs were then electroporated into X-33 or KM71H strains to produce proteins LlALP2-Ala (A) and (B), LlALP2-Ser (C) and (D), and LlALP2-Ala-c-myc-(His)6 (E).
200
S.C. Johnson et al. / Protein Expression and Purification 74 (2010) 196–203
linearized DNA of expression vectors pPICZA-LlAlp2-Ala and pPICZA-LlAlp2-Ser to produce four transformants, X-33-LlAlp2-Ala (Fig. 2;A), X-33-LlAlp2-Ser (Fig. 2;C), KM71H-LlAlp2-Ala (Fig. 2;B), and KM71H-LlAlp2-Ser (Fig. 2;D).
Screening for rLlALP2 phytase expression In order to identify the clone that produced the highest levels of alkaline phytase, the four transformants, X-33-LlAlp2-Ala, X-33LlAlp2-Ser, KM71H-LlAlp2-Ala, and KM71H-LlAlp2-Ser, were screened in parallel. Ten clones of each transformant were randomly selected and grown in baffle shake flasks. To ensure that the clones contained copies of alkaline phytase cDNA, PCR was conducted using LlAlp2-specific, full-length primers and genomic DNA as the template. All clones tested amplified a 1500 kb fragment, the expected length for full length alkaline phytase cDNA. rLlALP2 expression was induced by transferring the cells from glycerol-containing growth medium (BMGY) to medium containing methanol (BMMY). After 3 d of induction, the intracellular content of rLlALP2 was determined by lysing the cells and assaying for phytase activity. rLlALP2 activity was detected in the intracellular fraction; no activity was detected in un-induced cells. Fig. 3 indicates that clones with the same expression vector showed significant variability in total rLlALP2 activity (undetectable levels to 1200 U/25 mL of medium). Clones of KM71H and X-33 that con-
Fig. 3. Expression levels of rLlALP2 in different clones. Ten clones of each of the four transformants were selected at random (KM71H-LlALP2-Ser, KM71H-LlALP2-Ala, X-33-LlALP2-Ser, X-33-LlALP2-Ala). Alkaline phytase expression was determined by assaying for enzymatic activity in cell lysates. Reaction mixtures contained Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM), KCl (0.5 M), CaCl2 (1 mM), NaF (10 mM), and dialyzed cell lysate. Error bars represent ±SEM, n = 2.
tained LlAlp2 cDNA with Ala in the N-2 position showed a high proportion (17 of 19) of clones with total enzymatic activity greater than 200 U/25 mL compared to X-33 and KM71H transformed with the LlAlp2-Ser cDNA (2 of 20). There appeared to be little difference in total enzymatic activity between X-33 and KM71H strains when the same expression construct was used for transformation; both host strains produced a high proportion of clones with high activity (8 of 10 and 9 of 9) when LlAlp2-Ala cDNA was the construct and significantly less activity when LlAlp2-Ser cDNA was the construct. Although both Ser and Ala are considered stabilizing residues, these data are consistent with the idea that the presence of Ala in the N-2 position increases the half-life of LlALP2 in P. pastoris significantly. Optimization of expression conditions and purification of rLlALP2 To optimize expression conditions for the production of rLlALP2, the clone which produced the highest level of enzyme activity in KM71H-LlAlp2-Ala transformants was grown under a variety of conditions. Aeration is of great importance during methanol-induced expression so it is generally recommended that baffle flasks be employed for fermentation and culture volumes remain below 30% of the container volume to allow for good oxygen availability [26]. To investigate the effect of culture volume, rLlALP2 was expressed in either 25 mL (20% flask volume) or 12.5 mL (10% flask volume) of growth medium (Fig. 4A). Total rLlALP2 activity was greater (35–100%) when clones were grown in 25 mL of BMMY in the two clones investigated. The effect of methanol addition was examined by measuring total rLlALP2 activity in intracellular fraction after 3 d of induction (Fig. 4B). Varying amounts of methanol were added (0.5, 1, 2, or 3% v/v) once a day or 1% (v/v) methanol was added twice a day as indicated. Total rLlALP2 activity varied very little (±10%) under the methanol conditions examined. To determine the time point after methanol induction when intracellular accumulation of rLlALP2 activity level was the highest, cells were harvested 1–5 d post induction and rLlALP2 activity was measured (Fig. 4C). After 1 d of induction, cells had already accumulated 60% of the total activity. Cells had maximum activity after 2 d, after which total activity decreased from days 3 to 5. Optimal expression conditions employed for the rest of the experiments were as follows: baffle flasks (125 mL) contained 20% of medium (25 mL) and 1% methanol was added once a day for a period of 2 d. To facilitate the purification of the rLlALP2, cDNA coding for LlAlp2 with Ala in the N-2 position was inserted into pPICZA without
Fig. 4. Optimization of expression of rLlALP2 in P. pastoris. rLlALP2 expression was determined by assaying for enzymatic activity in cell lysate. Reaction mixtures contained Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM), KCl (0.5 M), CaCl2 (1 mM), NaF (10 mM) and dialyzed cell lysate. (A) The effect of culture volume on alkaline phytase expression in two clones (A and B) was determined using either 12.5 mL (10% flask volume) or 25 mL of media (20% flask volume) in a 125 mL baffle flask. (B) The effect of methanol addition on alkaline phytase expression was determined by adding varying amounts of methanol, 0.5%, 1%, 2%, or 3% once per day or 1% twice per day. (C) Time course of alkaline phytase expression after methanol induction: samples were harvested 1, 2, 3, 4, and 5 d after transferring to BMMY, the cells lysed and assayed for activity. Error bars represent ± SEM, n = 2.
S.C. Johnson et al. / Protein Expression and Purification 74 (2010) 196–203
201
did bind to the Co2+-resin in the presence of urea (2 M) but we observed no binding to the Ni2+-resin. The enzyme was eluted from the Co2+-resin with buffer containing increasing concentrations of imidazole and the fractions assayed for enzyme activity and analyzed by SDS–PAGE (Fig. 5). Fractions containing rLlALP2 activity indicated a dominant protein band of Mr. close to 60 kDa (Fig. 5, lanes 6 and 7), the calculated molar mass of rLlALP2 is 58 kDa; fraction corresponding to lane 7 demonstrated nearly twice the enzymatic activity as the fraction corresponding to lane 6 (data not shown). The yield obtained was 8 mg/L of broth. Fig. 5. Separation by SDS–PAGE of column fractions during rLlALP2 purification on Co2+-resin. The fusion protein rLlALP2-Ala-c-myc-(His)6 was expressed in X-33. Cells were lysed and the soluble fraction was loaded on a Co2+-resin column, washed and then eluted with buffer containing increasing concentrations of imidazole. Lane 1: molar mass markers; lane 2: crude lysate incubated in urea (2 M); lane 3: flow through; lane 4: wash 1 (Tris–HCl, pH 7.4, 300 mM NaCl); lane 5: elution 1 (Wash Buffer with imidazole, 10 mM); lane 6: elution 2 (Wash Buffer with imidazole, 30 mM); lane 7: elution 3 (Wash Buffer with imidazole 60 mM).
the stop codon so that the c-myc epitope and (His)6 tag encoded by the vector was fused at the C-terminus (Fig. 2III). Attempts to purify the rLlALP2 with immobilized metal affinity chromatography (IMAC) using Ni2+-nitrilotriacetic acid resin or Co2+-resin indicated that rLlALP2 did not bind to either resin. All enzyme activity was recovered in the flow-through fraction. The lack of binding may be due to the fact that the C-terminal (His)6 tag was not exposed for binding in the recombinant dimeric protein. Low concentrations of urea (2 M) were included in the binding buffer in an effort to expose the (His)6 tag without denaturing the enzyme. rLlALP2
Functional characterization of alkaline phytase In an effort to functionally characterize the rLlALP2 produced by P. pastoris and compare it to the native enzyme from lily pollen, the influence of a number of factors (pH, substrate specificity, temperature dependence, calcium ions, EDTA and inhibitors) on catalytic activity was probed. The influence of pH on the activity of rLlALP2 was monitored in Tris–maleate buffer from pH 5.5 to 9.0 (Fig. 6A). As observed with alkaline phytase from lily pollen [18], maximum activity was detected at pH 8.0. Lowering the pH to 6 caused a 90% drop in activity, whereas raising pH to 9 caused a 40% decrease. The pH activity profile is similar to that of wild-type alkaline phytase isolated from lily pollen [18]. LlALP from lily pollen exhibits a relatively high temperature optimum, 55 °C. The influence of temperature on the enzymatic activity of rLlALP2 was monitored at temperatures ranging from 30 °C to 70 °C (Fig. 6B). rLlALP2 had the highest total activity at
Fig. 6. Functional characterization of rLlALP2. rLlALP2 activity was assayed by measuring the phosphate released from sodium phytate at 37 °C in 60 min. The enzyme assay mixture contained Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM), KCl (0.5 M), CaCl2 (1 mM), NaF (10 mM) and dialyzed cell lysate, unless otherwise noted. (A) The effect of pH on activity was determined in Tris–maleate buffer (100 mM) at various pH. (B) The effect of temperature was determined by pre-equilibrating the assay mixtures at appropriate temperatures and the activity assayed at corresponding temperatures. The amount of phosphate released in 20 min was measured. (C) rLlALP2 activity in an assay mixture containing different phosphorylated substrates (1 mM). (D) rLlALp2 activity in the standard mixture with calcium and/or EDTA. (E) rLlALP2 activity in the absence or presence of inhibitors (N). Error bars represent ± SEM, n = 2.
202
S.C. Johnson et al. / Protein Expression and Purification 74 (2010) 196–203
55 °C, a sharp decrease in activity was observed at temperatures above 55 °C, consistent with the observations with the native LlALP [18]. The substrate specificity of LlALP from lily pollen is unique among phytases; it differs from that of acid phytases, which show broad substrate specificity, and from other alkaline phytases, which exhibit narrow substrate specificity including lack of activity against para-nitrophenyl phosphate (pNPP). LlALP from lily pollen hydrolyzes phytate and pNPP. The rLlALP2 expressed in P. pastoris hydrolyzed phytate and pNPP and exhibited no activity towards ATP (Fig. 6C). Activity against pNPP was nearly 2.5-fold higher than against phytate, similar to what is observed with the native enzyme [18]. The native enzyme from lily pollen exhibits a 3.5–4.0-fold enhancement of activity in the presence of added CaCl2 (1 mM), however, some hydrolytic activity is observed even in the presence of relatively high concentrations of EDTA (5 mM) (Fig. 6D). The effect of Ca2+ ions on enzymatic activity of rLlALP2 was investigated, with and without added CaCl2 (1 mM) and EDTA (5 mM). Consistent with observations from the native enzyme, the recombinant enzyme showed a 4-fold enhancement when CaCl2 was added; the enzyme retained some enzyme activity (20%), albeit low, even in the presence of high concentration of EDTA (5 mM) (Fig. 6D). This could be because of the presence of tightly bound Ca2+ at the enzyme active site that cannot be extracted with EDTA or because LlALP is able to catalyze phytate at a low level even in the absence of CaCl2. A major difference between acid and alkaline phytases is the effect of inhibitors. LlALP from lily pollen is inhibited by vanadate, a transition state analog, but not inhibited by fluoride, a strong inhibitor of many histidine acid phytases [18]. Addition of relatively high concentrations of NaF (10 mM) had only a slight effect on total activity (10% decrease) whereas rLlALP2 was strongly inhibited by vanadate (85% inhibition at 1 mM) (Fig. 6E), consistent with what was observed with the native enzyme [18].
Discussion The studies reported here demonstrate that LlALP2 from lily pollen can be expressed in P. pastoris in a soluble active form. Our attempts to express rLlALP2 in E. coli resulted in accumulation of the enzyme in inclusion bodies and numerous efforts to solubilize and refold alkaline phytase were unsuccessful. Related to this is our observation from previous studies that when soluble native alkaline phytase from lily pollen was denatured under mild conditions (0.5% sodium dodecyl sulfate and b-mercaptoethanol), we were not successful in renaturing it to the active form despite extensive efforts [37]. The difficulty in refolding denatured alkaline phytase into the active form may be due to the fact that the native active alkaline phytase is a homodimer with 20 Cys residues (each monomer has 10 Cys) and disulfide linkages may play an important role in protein folding. The fact that P. pastoris can express an active dimeric protein with many Cys residues suggests that the intracellular environment of P. pastoris promotes correct disulfide bond formation, a significant advantage over bacterial expression systems. Intracellular protein degradation will affect the levels of foreign proteins that can be harnessed from P. pastoris. Although both Ala and Ser at the N-2 position are stabilizing amino acids in S. cerevisiae, we isolated significantly more protein when Ala was in the N2 position compared to when Ser was at N-2 (Fig. 3); on an average N-2-Ala clones produced 7-fold more LlALP2 than N-2-Ser clones. This difference in yield could be due to differences in protein degradation mechanisms between P. pastoris and S. cerevisiae or due to unique characteristics of the protein.
The rLlALP2 expressed in P. pastoris showed Mr. and catalytic properties very similar to the wild-type enzyme. The substrate specificity and temperature dependence of catalysis as well as the effect of pH, inhibitors, calcium ions, and EDTA were very similar to that observed with the wild-type enzyme from lily pollen, suggesting that the folding of the recombinant enzyme is very similar to that of the wild-type enzyme. (His)6 and c-myc peptide tags at the N- or the C-terminal are widely used to detect proteins with specific antibodies and purify proteins by affinity chromatography [38]. However, the success of these methods depends on the presence of unmasked tags. We observed no binding of rLlALP2 to Ni2+-NTA or Co2+-resin. When low concentration of urea (2 M) were added to the binding buffer, rLlALP2 bound to the Co2+-resin, possibly due to partial unfolding and unmasking of the (His)6 tag, no binding to the Ni2+-NTA resin was observed even in the presence of urea. This observation is similar to that noted by Li et al. [39]. Active rLlALP2 was eluted from the column with buffer containing imidazole. An observation related to this issue is that, although phytate hydrolyzing activity was readily detected in enzyme assays of cell lysate, our efforts to detect the presence of the fusion protein by Western blot analysis with anti-c-myc antibodies or anti-(His)6 antibodies were unsuccessful. Taken together, these data suggest that both the cmyc and (His)6 tags are masked in the folded protein. We have demonstrated in this study that LlALP2 from lily pollen can be heterologously expressed in P. pastoris in a soluble active form. This is the first report of the heterologous expression of a plant phytase in an active soluble form. Transformed P. pastoris can serve as a potential source of substantial amounts of active rLlALP2 for structural, biochemical, and animal feed studies. Acknowledgments We gratefully acknowledge MTU for Graduate Student Fellowship to S.C.J. We thank Dr. Bakul Dhagat Mehta for constructing the LlAlp2 cDNA-containing pPET vector and Dr. Sonali Jog for technical discussions. We thank Dr. W.H. Campbell (Nitrate Elimination Company) for many helpful discussions on protein expression in P. pastoris. References [1] B.C. Oh, W.C. Choi, S. Park, Y.O. Kim, T.K. Oh, Biochemical properties and substrate specificities of alkaline and histidine acid phytases, Appl. Microbiol. Biotechnol. 63 (2004) 362–372. [2] E.J. Mullaney, A.H.J. Ullah, The term phytase comprises several different classes of enzymes, Biochem. Biophys. Res. Commun. 312 (2003) 179–184. [3] U. Konietzny, R. Greiner, Molecular and catalytic properties of phytatedegrading enzymes (phytases), Int. J. Food Sci. Technol. 37 (2002) 791–812. [4] R.J. Wodzinski, A.H.J. Ullah, Phytase, Adv. Appl. Microbiol. 42 (1996) 263–302. [5] J.E. Coleman, Structure and mechanism of alkaline-phytase, Annu. Rev. Biophys. Biomol. Struct. 21 (1992) 441–483. [6] J.B. Vincent, M.W. Crowder, B.A. Averill, Hydrolysis of phosphate monoesters: a biological problem with multiple chemical solutions, Trends Biochem.Sci. 17 (1992) 105–110. [7] A.R. Alcazar-Roman, S.R. Wente, Inositol polyphosphates: a new frontier for regulating gene expression, Chromosoma 117 (2008) 1–13. [8] R.H. Michell, Inositol derivatives: evolution and functions, Nat. Rev. Mol. Cell Biol. 9 (2008) 151–161. [9] A.T. Miller, P.P. Chamberlain, M.P. Cooke, Beyond IP3 – roles for higher order inositol phosphates in immune cell signaling, Cell Cycle 7 (2008) 463–467. [10] F.A. Loewus, P.P.N. Murthy, Myo-inositol metabolism in plants, Plant Sci. 150 (2000) 1–19. [11] M.J. Berridge, Inositol trisphosphate and calcium signaling, Nature 361 (1993) 315–325. [12] N.R. Reddy, S.K. Sathe, D.K. Salunkhe, Phytates in legumes and cereals, Adv. Food Res. 28 (1982) 1–92. [13] B.G. Baldi, J.J. Scott, J.D. Everard, F.A. Loewus, Localization of constitutive phytases in lily pollen and properties of the pH 8 form, Plant Sci. 56 (1988) 137–147. [14] B.D. Mehta, S.P. Jog, S.C. Johnson, P.P.N. Murthy, Lily pollen alkaline phytase is a histidine phosphatase similar to mammalian multiple inositol polyphosphate phosphatase (MINPP), Phytochemistry 67 (2006) 1874–1886.
S.C. Johnson et al. / Protein Expression and Purification 74 (2010) 196–203 [15] X.G. Lei, C.H. Stahl, Biotechnological development of effective phytases for mineral nutrition and environmental protection, Appl. Microbiol. Biotechnol. 57 (2001) 474–481. [16] A.N. Sharpley, S.C. Chapra, R. Wedepohl, J.T. Sims, T.C. Daniel, K.R. Reddy, Managing agricultural phosphorus for protection of surface waters – issues and options, J. Environ. Qual. 23 (1994) 437–451. [17] T.S. Nelson, T.R. Shieh, R.J. Wodzinski, J.H. Ware, Effect of supplemental phytase on the utilization of phytate phosphorus by chicks, J. Nutr. 101 (1971) 1289–1293. [18] S.P. Jog, B.G. Garchow, B.D. Mehta, P.P.N. Murthy, Alkaline phytase from lily pollen: investigation of biochemical properties, Arch. Biochem. Biophys. 440 (2005) 133–140. [19] N. Armstrong, A. De Lencastre, E. Gouaux, A new protein folding screen: application to the ligand binding domains of a glutamate and kainate receptor and to lysozyme and carbonic anhydrase, Protein Sci. 8 (1999) 1475–1483. [20] P.Z. Li, A. Anumanthan, X.G. Gao, K. Ilangovan, V.V. Suzara, N. Duzgunes, V. Renugopalakrishnan, Expression of recombinant proteins in Pichia pastoris, Appl. Biochem. Biotechnol. 142 (2007) 105–124. [21] J.M. Cregg, J.L. Cereghino, J.Y. Shi, D.R. Higgins, Recombinant protein expression in Pichia pastoris, Mol. Biotechnol. 16 (2000) 23–52. [22] J.L. Cereghino, J.M. Cregg, Heterologous protein expression in the methylotrophic yeast Pichia pastoris, FEMS Microbiol. Rev. 24 (2000) 45–66. [23] R. Daly, M.T.W. Hearn, Expression of heterologous proteins in Pichia pastoris: a useful experimental tool in protein engineering and production, J. Mol. Recognit. 18 (2005) 119–138. [24] S. Macauley-Patrick, M.L. Fazenda, B. McNeil, L.M. Harvey, Heterologous protein production using the Pichia pastoris expression system, Yeast 22 (2005) 249–270. [25] J. Sambrook, D. Russel (Eds.), Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, New York, 2001. [26] Invitrogen, EasySelectTM Pichia Expression Kit, Carlsbad, CA, 2009. [27] L. Barrientos, J.J. Scott, P.P.N. Murthy, Specificity of hydrolysis of phytic acid by alkaline phytase from lily pollen, Plant Physiol. 106 (1994) 1489–1495.
203
[28] R. Hamilton, C.K. Watanabe, H.A. de Boer, Compilation and comparison of the sequence context around the AUG startcodons in Saccharomyces cerevisiae mRNAs, Nucleic Acids Res. 15 (1987) 3581–3593. [29] A. Varshavsky, The N-end rule: functions, mysteries, uses, Proc. Natl. Acad. Sci. USA 93 (1996) 12142–12149. [30] A. Mogk, R. Schmidt, B. Bukau, The N-end rule pathway for regulated proteolysis: prokaryotic and eukaryotic strategies, Trends Cell Biol. 17 (2007) 165–172. [31] J. Lin-Cereghino, W.W. Wong, S. Xiong, W. Giang, L.T. Luong, J. Vu, S.D. Johnson, G.P. Lin-Cereghino, Condensed protocol for competent cell preparation and transformation of the methylotrophic yeast Pichia pastoris, BioTechniques 38 (2005) 44–48. [32] J.R. Thompson, E. Register, J. Curotto, M. Kurtz, R. Kelly, An improved protocol for the preparation of yeast cells for transformation by electroporation, Yeast 14 (1998) 565–571. [33] S. Wu, G. Letchworth, High efficiency transformation by electroporation of Pichia pastoris pretreated with lithium acetate and dithiothreitol, BioTechniques 36 (2004) 152–154. [34] J.M. Cregg, K.A. Russell, Transformation, in: D.R. Higgins, J.M. Cregg (Eds.), Pichia Protocols, Humana Press, Totowa, 1998, pp. 27–39. [35] B. Brzobohaty, L. Kovac, Factors enhancing genetic-transformation of intact yeast-cells modify cell-wall porosity, J. Gen. Microbiol. 132 (1986) 3089–3093. [36] H. Ito, K. Murata, A. Kimura, Transformation of intact yeast-cells treated with alkali cations or thiol compounds, Agric. Biol. Chem. 48 (1984) 341–347. [37] S. Jog, Biochemical Characterization and Recombinant Expression of Alkaline Phytase from Lilium longiflorum, Ph.D. Dissertation, Michigan Technological University, Houghton, MI 49931, USA, 2010. [38] H. Block, B. Maertens, A. Spriestersbach, N. Brinker, J. Kubicek, R. Fabis, J. Labahn, F. Schafer, Immobilized-metal Affinity Chromatography (IMAC): A Review, Guide to Protein Purification, second ed., Elsevier Academic Press Inc., San Diego, 2009. pp. 439–473. [39] Z. Li, A. Moy, K. Sohal, C. Dam, P. Kuo, J. Whittaker, M. Whittaker, N. duzgunes, K. Konopka, A.H. Franz, J. Lin-Cereghino, G.P. Lin-cereghino, Expression and characterization of recombinant human secretory leukocyte protease inhibitor (SLPI) protein from Pichia pastoris, Protein Expr. Purif. 67 (2009) 175–181.
Contents lists available at ScienceDirect
Protein Expression and Purification journal homepage: www.elsevier.com/locate/yprep
Heterologous expression and functional characterization of a plant alkaline phytase in Pichia pastoris Steven C. Johnson, Mimi Yang, Pushpalatha P.N. Murthy * Department of Chemistry, Michigan Technological University, 1400 Townsend Drive, Houghton, MI 49931, USA
a r t i c l e
i n f o
Article history: Received 26 April 2010 and in revised form 16 July 2010 Available online 22 July 2010 Keywords: Alkaline phytase Phosphate contamination Animal feed Heterologous protein expression Pichia pastoris Pollen grains
a b s t r a c t Phytases catalyze the sequential hydrolysis of phytic acid (myo-insositol hexakisphosphate), the most abundant inositol phosphate in cells. Phytic acid constitutes 3–5% of the dry weight of cereal grains and legumes such as corn and soybean. The high concentration of phytates in animal feed and the inability of non-ruminant animals such as swine and poultry to digest phytates leads to phosphate contamination of soil and water bodies. The supplementation of animal feed with phytases results in increased bioavailability to animals and decreased environmental contamination. Therefore, phytases are of great commercial importance. Phytases with a range of properties are needed to address the specific digestive needs of different animals. Alkaline phytase (LlALP1 and LlALP2) which possess unique catalytic properties that have the potential to be useful as feed and food supplement has been identified in lily pollen. Substantial quantities of alkaline phytase are needed for animal feed studies. In this paper, we report the heterologous expression of LlALP2 from lily pollen in Pichia pastoris. The expression of recombinant LlALP2 (rLlALP2) was optimized by varying the cDNA coding for LlALP2, host strain and growth conditions. The catalytic properties of recombinant LlALP2 were investigated extensively (substrate specificity, pH- and temperature dependence, and the effect of Ca2+, EDTA and inhibitors) and found to be very similar to that of the native LlALP2 indicating that rLlALP2 from P. pastoris can serve as a potential source for structural and animal feed studies. Ó 2010 Elsevier Inc. All rights reserved.
Introduction Phytases are a class of phosphatases that catalyze the sequential hydrolysis of phytic acid [myo-inositol hexakisphosphate (Fig. 1)] to less phosphorylated inositol phosphates, and in some cases, to inositol [1–6]. Inositol phosphates play multiple roles in biological processes including signal transduction and calcium regulation [7– 11]. Phytic acid is the most abundant inositol phosphate in cells [12], thus phytases, the primary enzymes responsible for the hydrolysis of phytic acid, play important roles in the metabolism of inositol phosphates. A number of phytases with varying structural and catalytic properties have been found in plants, yeast, bacteria, and animals [2,3]. The enzymes differ in pH optima (acidic and alkaline), catalytic mechanisms (histidine acid phosphatase-like phytase, purple acid phosphatase-like phytase, cysteine phosphatase-like phytase, and b-propeller phytase), substrate specificity, specificity of hydrolysis, metal ion requirements, susceptibility to inhibitors, and thermostability [1–4]. A novel alkaline phytase was detected in the Abbreviations: IB, inclusion body; AOX1, alcohol oxidase; LiAc, lithium acetate. * Corresponding author. Fax: +1 906 487 2061. E-mail address: [email protected] (P.P.N. Murthy). 1046-5928/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.pep.2010.07.003
pollen grains of Lilium longiflorum (LlALP) [13]. Cloning and sequence analysis identified two isoforms of alkaline phytase cDNA, LlAlp1 and LlAlp2, which are 1463 and 1533 bp long and encode proteins of 487 and 511 amino acids, respectively [14]. The calculated molar mass of LlALP1 and LlALP2 monomers are 53.8 and 56.2 kDa, respectively; the active enzyme exists as a homodimer. The deduced amino acid sequences revealed that the enzyme is a histidine phosphatase, it contains the signature heptapeptide of histidine phosphatases, –RHGXRXP– near the N-terminus [14]. In cereal grains and legumes phytate constitutes 3–5% of the dry weight of seeds [12]. During seed and pollen development, phytic acid is deposited in membrane-bound bodies as a salt of essential mineral ions such as potassium, calcium, magnesium, iron, and zinc and this then serves as a source of phosphate, inositol, and metal ions during the germination of seeds and pollen grains [12]. Although high concentrations of phytate are present in corn and soybean, the major components of animal feed, the phosphate, inositol, and essential metals in phytate complexes are unavailable to humans and monogastric animals, such as poultry, swine and fish, because they lack phytase [1–4,15,16]. Therefore, for optimum animal growth, supplementation of animal feed with phosphate and essential minerals, at additional cost, is necessary [16]. The excretion of undigested phytate from animals results in elevated levels of orga-
S.C. Johnson et al. / Protein Expression and Purification 74 (2010) 196–203
197
Materials and methods Strains, plasmids, and media
Fig. 1. Dodecasodium salt of phytic acid.
nophosphate in soil and water bodies downstream of agriculturally intensive areas [16]. To address these nutritional and environmental contamination issues, governments have passed legislation to supplement animal feed with phytases [4,15]. The efficacy of supplementing animal feed with phytase to increase the bioavailability of inositol and phosphates and decrease phosphorous discharge from animal farms has been demonstrated in several experiments [4,15,17]. The commercial importance of phytases in animal feed has resulted in the need for substantial quantities of an array of phytases with a range of catalytic and stability properties for use in different end applications [15]. Investigation of the biochemical properties of alkaline phytase from lily pollen indicates that the enzyme has unique catalytic properties and suggests that the enzyme has the potential to be useful as a feed and food supplement [18]. Our initial efforts at heterologous expression of rLlALP2 from lily pollen in E. coli resulted in low levels of expression and the accumulation of LlALP2 in inclusion bodies [14]. Although recombinant proteins in inclusion bodies (IB) are frequently inactive, alkaline phytases in IBs did exhibit catalytic activity [14]. Numerous efforts to solubilize alkaline phytase from IBs and refold them into the more active form were unsuccessful. Once denatured, rLlALP2 did not refold into the biologically active form under any of the numerous conditions [19] we tried (unpublished data); the fact that the enzyme is a homodimer with a total of 20 cysteine residues may have contributed to the refolding challenges we encountered. The methylotropic yeast, Pichia pastoris, has been shown to be a good expression system for high-level production of intracellular and secreted foreign proteins [20–22]. Heterologous expression of proteins in P. pastoris has many advantages over bacterial expression hosts: P. pastoris is capable of posttranslational modifications such as folding, disulfide bond formation, and glycosylation; expression of foreign proteins is driven by the tightly controlled and highly inducible alcohol oxidase (AOX1) promoter; proteins can be targeted to the extracellular media serving as the first step in purification; it is relatively inexpensive compared to other eukaryotic expression systems; the ability to grow to high cell densities coupled with high expression rates results in high protein yields, expression levels can range from mg to many g per L of growth culture [21–24]. In this paper, we describe the heterologous expression of a plant alkaline phytase (lily) in P. pastoris. Our aim was to investigate the expression of the active enzyme in a soluble form in P. pastoris. As a first step, we decided to express the enzyme intracellularly so as to avoid the additional problems associated with the Pichia secretary pathway. In an effort to optimize the expression of LlALP2, the LlAlp2 insert, host strain and growth conditions were varied. Catalytic properties of the recombinant enzyme (rLlALP2) expressed in P. pastoris were investigated and compared to the wild-type enzyme. To our knowledge, this is the first report of the heterologous expression of a phytase from a plant source.
Escherichia coli TOP10F0 , P. pastoris strains X-33 and KM71H, and expression vector pPICZA were purchased from Invitrogen (Carlsbad, CA). Primers were purchased from Integrated DNA Technologies (Coralville, IA) and GoTaq Flexi DNA polymerase was purchased from Promega (Madison, WI). Restriction enzymes and T4 DNA ligase were purchased from New England Biolabs (Ipswich, MA). Peptone, tryptone, yeast extract, glycerol, biotin, and methanol were purchased from Fisher (Pittsburgh, PA). Yeast nitrogen base was purchased from BD Biosciences (Sparks, MD). HisPur cobalt was purchased from Thermo Fisher (Rockford, IL). Low salt LB medium (1% tryptone, 0.5% yeast extract, and 0.5% NaCl) was used for propagation of E. coli (TOP10F0 ). Low salt LB medium plates (1.5% agar) containing kanamycin (25 lg/mL) (Research Products International Corp., Mt. Prospect, IL) were used to select for transformants of TOP10F0 . P. pastoris strains were grown in either BMGY (2% peptone, 1% yeast extract, 100 mM potassium phosphate, pH 6.0, 1.34% yeast nitrogen base, 4 105% biotin, and 1% glycerol) or BMMY medium (2% peptone, 1% yeast extract, 100 mM potassium phosphate, pH 6.0, 1.34% yeast nitrogen base, 4 105% biotin, and 1% methanol). YPD agar plates [1% yeast extract, 2% peptone, 2% dextrose, 2% agar, and various concentrations (100, 500, 1000 or 2000 lg/mL) of zeocin, as indicated] were used for selection of P. pastoris transformants.
Construction of expression vector Alkaline phytase cDNA (LlAlp2) was amplified via PCR from a previously constructed plasmid [14]. Forward primers for LlAlp2Ser (EcoRI restriction site in bold and the codons corresponding to Ser and Ala underlined) (50 -TAGCATGAATTCAAAAATGTCTGC GTTCTCGCTTCACGC-30 ) and LlAlp2-Ala (50 -TAGCATGAATTC 0 AAAAATGGCGTTCTCGCTTCACGC-3 ) were used in conjunction with a reverse primer (NotI restriction site in bold, stop codon italicized) containing a stop codon (50 -AAAAGGAAAAGCGGCCGC TCACAGTTCTGTCTTCTGGTTACCGGTG-30 ) or without a stop codon (50 -AAAAGGAAA AGCGGCCGCCAGTTCTGTCTTCTGGTTACCGGTG-30 ). Absence of the stop codon allowed for the incorporation of a cmyc epitope tag followed by a (His)6 tag at the carboxyl terminus of the protein. PCR amplification (35 cycles of 94 °C for 30 s, 60 °C for 50 s and 72 °C for 2 min and a final elongation of 5 min at 72 °C) yielded a 1500 bp product coding for LlAlp2. The LlAlp2 cDNA was purified using a QIAquick Gel Extraction Kit (Qiagen, Valencia, CA). The purified LlAlp2 cDNA insert and pPICZA were separately digested, first with NotI for 18 h after which EcoRI was added and the mixture incubated for an additional 4 h, and then ligated with T4 DNA ligase (16 h at 14 °C) per manufacturer’s instructions. Competent TOP10F0 were transformed with recombinant plasmids according to Sambrook [25] and positive transformants were selected on low salt LB agar plates containing 25 lg/mL zeocin. The insertion of LlAlp2 cDNA into pPICZA was confirmed by PCR. Plasmids were isolated using the WizardÒ Plus DNA Purification System according to the manufacturer’s instructions (Promega, Madison, WI). In brief, 5 mL of LB medium was inoculated with a single colony and grown overnight at 37 °C. Cells were collected by centrifugation (10 min at 1400g) and the plasmid extracted as suggested by the manufacturer. PCR amplification using plasmid DNA and LlAlp2-specific, full-length primers, used above, yielded a 1500 bp product corresponding to LlAlp2 cDNA. A second PCR amplification using plasmid DNA and plasmid-specific primers that flank the insertion site (forward primer: 50 -GAC TGG TTC
198
S.C. Johnson et al. / Protein Expression and Purification 74 (2010) 196–203
CAA TTG ACA AGC-30 ; reverse primer: 50 -GCA AAT GGC ATT CTG ACA TCC-30 ) was conducted and the amplified product was sequenced to confirm that insertion had occurred in the correct reading frame (Nevada Genomics Center, Reno, NV). Transformation of P. pastoris Recombinant plasmid was isolated from TOP10F0 using a Qiagen Plasmid Maxi Kit according to the manufacturer’s instructions and linearized with PmeI (37 °C, 16 h). Competent cells were prepared according to the Pichia EasySelect manual [26]. In brief, 500 mL of YPD was inoculated with 0.1 mL of an overnight P. pastoris culture and grown at 29 °C to an Abs600 of 1.4. Cells were collected by centrifugation (15 min at 1500g, 4 °C; cell pellet C-1) and subsequently washed with successively smaller volumes of ice-cold sorbitol (1 M; 500, 250 and 20 mL). The cell pellet was finally resuspended in ice-cold sorbitol (1.5 mL, 1 M) and aliquots of this cell suspension used for transformation. The transformation efficiency under these conditions was low therefore several pre-treatment conditions were tried (Table 1). Cell pellet C-1 was pretreated by incubating (30 min) in either YPD HEPES (200 mM, pH 8.0) containing dithiothreitol (25 mM) or YPD HEPES (200 mM, pH 8.0) containing dithiothreitol (10 mM) and lithium acetate (100 mM) before successive washings in smaller volumes of ice-cold sorbitol (1 M). Linearized plasmid (10 lg in 5 lL of H2O) and competent cells (40 lL) were combined in an electroporation cuvette (0.2 cm gap), incubated for 5 min on ice, and pulsed at 2 kV with a MicroPulser (Bio-Rad, Hercules, CA). Two different post-treatment conditions were tried: ice-cold sorbitol (1 M) or ice-cold YPD containing sorbitol (1 M, YPDS) was added to the cuvette immediately and the cells transferred to a 15 mL Falcon tube and incubated at 30 °C from 2 to 4 h without shaking. Aliquots (10, 50 and 250 lL) were plated on YPDS agar plates containing 100 lg/mL zeocin [26]; colonies were visible in 2 d. To make glycerol stock, positive P. pastoris colonies were grown for 16 h (29 °C) in 3 mL of YPD. Aliquots of cell cultures (850 lL) were mixed with sterile glycerol (150 lL), flash frozen in liquid N2, and stored at 80 °C. The insertion of LlAlp2 cDNA into P. pastoris was confirmed by PCR. Glycerol stocks were streaked on YPD plates containing zeocin (25 lg/mL) and grown for 2 d (29 °C). Colonies were lifted from the plate and the cells were lysed by suspending in H2O (10 lL) and lyticase solution (20 lL, 8 U, Sigma, St. Louis, MO) and the mixture was incubated at 30 °C for 30 min. Samples were then frozen at 80 °C for 15 min and thawed. PCR amplification using genomic DNA and LlAlp2-specific, full-length primers used above yielded a 1500 bp product corresponding to LlAlp2 cDNA. Expression of alkaline phytase in P. pastoris and cell lysis In order to achieve high yields of LlALP2 the following growth parameters were optimized: (1) influence of culture volume was examined by performing expression using either 12.5 mL or 25 mL of media in a 125 mL baffle flask. (2) Optimal methanol
addition was investigated by adding varying amounts of methanol (0.5%, 1%, 2%, and 3% once daily, 1% twice daily) during expression. (3) The time point after methanol induction that yields the maximum activity (1, 2, 3, 4, and 5 d) was investigated. Changes to the conditions are indicated in the figures. In subsequent experiments, the following set of optimized conditions were employed: glycerol stock of clones were streaked on YPD agar plates containing zeocin (100 lg/mL) and grown at 29 °C for 2 d. Individual colonies were used to inoculate 25 mL of BMGY in a 125 mL baffled flask and grown at 30 °C for 20 h. The cells were harvested by centrifugation at 3000g for 5 min and resuspended in BMMY (25 mL). Every 24 h, sterile methanol (250 lL) and sterile H2O (2.25 mL) were added to a total of 10% of the culture volume. Cells were harvested at indicated times by centrifugation (10,000g) and resuspended in lysis buffer (Tris–HCl, 50 mM, pH 7.4; PMSF, 1 mM; 2 mL of lysis buffer per g wet weight of cells). The wet cell weight of P. pastoris grown in 25 mL of BMGY medium usually varied from 1.7 to 1.8 g, a narrow range. Greater variations were observed only when samples were infected, in these cases the samples were discarded. To ensure complete cell lysis and release of alkaline phytase activity, two methods of cell disruption were examined, a MiniBeadbeater-8 (BioSpec, Bartelsville, OK) or a Vortex Genie 2 (Fisher, Pittsburgh, PA). Resuspended cell pellet (1 mL) was added to 0.5 mL of 0.5 mm zirconia/silica disruption beads (Research Products International Corp., Mt. Prospect, IL). Samples were lysed for 30 s followed by 30 s incubation on ice for a total of 6 cycles, centrifuged at 10,000g, and the supernatant dialyzed for 18 h. Samples were assayed for LlALP2 activity according to the procedure below. In our hands, disruption using the Vortex mixer at the highest speed possible released 5-fold more total activity than the Beadbeater. Cell lysate was centrifuged (10,000g, 15 min, 4 °C) to separate the soluble proteins from the cell debris. The supernatant, which contains the soluble proteins, was dialyzed against Tris–HCl buffer (10 mM, pH 7.4) for 18 h to remove phosphates and salts before purification or assaying for enzyme activity.
Purification of rLlALP2 by immobilized metal affinity chromatography (IMAC) on Co2+-resin The enzyme LlALP2 -Ala with the c-myc and (His)6 tags at the Cterminus was expressed in P. pastoris strain X-33 and purified on HisPur cobalt resin. The dialyzed soluble lystate (3 mL) was incubated with urea (2 M) for 30 min on ice. HisPur Cobalt resin (350 lL) was added to a chromatography column and the storage buffer was allowed to drain by gravity. The resin was equilibrated with Binding buffer (1 mL, Tris–HCl, 50 mM, pH 7.4; NaCl, 300 mM) and the urea-treated soluble lystate was loaded onto the resin. The resin was washed with Binding buffer (3 1 mL). LlALP2-Ala-c-myc-(His)6 was then eluted with Binding buffer (500 lL) containing increasing concentrations of imidazole (7 mM, 30 mM, 60 mM, and 200 mM). The fractions were assayed for alkaline phytase activity and separated using SDS–PAGE.
Table 1 Effect of various pre-treatment and post-transformation recovery conditions on transformation efficiency. Method
A [26] B [32] C [33] D [33,34]
Pre-treatment conditions
Transformants/lg DNA
Recovery conditions
Media
DTT (mM)
LiAc (mM)
Media
Time (h)
No pre-treatment YPD + HEPES (pH 8.0) YPD + HEPES (pH 8.0) YPD + HEPES (pH 8.0)
0 25 10 10
0 0 100 100
1 M sorbitol 1 M sorbitol 1 M sorbitol YPD + 1 M sorbitol
2 4 4 4
Cells were prepared as described in the ‘‘Materials and Methods” section. Pre-treatment and recovery conditions are indicated above.
10 50 200 500
S.C. Johnson et al. / Protein Expression and Purification 74 (2010) 196–203
Alkaline phytase activity assay Alkaline phytase activity was assayed by measuring the phosphate released from sodium phytate by the enzyme [27]. The reaction mixture contained Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM), KCl (0.5 M), CaCl2 (1 mM), NaF (10 mM), and aliquots of dialyzed cell lysate in a total volume of 500 lL. Samples were incubated at 37 °C for 1 h and the reaction was stopped with the addition of TCA (50% w/v, 100 lL). The denatured enzyme was pelleted at 10,000g for 2 min and 300 lL of the supernatant was added to 700 lL of the phosphate detection solution (6:1 ratio of 0.42% ammonium molybdate w/v in 0.5 M H2SO4 and 10% w/v ascorbic acid in H2O). The mixture was incubated for 1 h at 37 °C and the absorbance measured at 820 nm. Inorganic phosphate concentration was determined from a calibration curve using KH2PO4 as the standard. One unit of activity is defined as the release of 1 nmol of Pi from sodium phytate in 1 h. Standard assay conditions served as controls in the experiments; deviations are indicated in the figure legends. Vanadate was dissolved in Tris–HCl (100 mM, pH 8.0) and used immediately.
Results Construction and transformation of expression vectors The design of expression constructs for P. pastoris was guided by the following considerations: (1) the translational efficiency in P. pastoris is thought to be primarily dependent on the rate of initiation, which in turn, is influenced by the sequence in the 50 untranslated region of mRNA [28]. In P. pastoris, the consensus sequence for high initiation rates includes 4As immediately upstream of the start codon, AUG. Therefore, AAAAAUG was the sequence around the start codon in all constructs (Fig. 2). (2) The half-life of a protein is determined by the nature of the penultimate amino acid at the N-terminus (N-2 position), the N-end rule [29,30]. Two amino acids that stabilize proteins in yeast include Ala and Ser [29]; the wild-type cDNA from lily pollen (LlAlp2) codes for Ala in the N-2 position. Two isoforms of cDNAs that code for alkaline phytase, one coding for Ala in the N-2 position (LlAlp2-Ala) and another coding for Ser in the N-2 position (LlAlp2-Ser), were synthesized by PCR amplification of the full-length LlAlp2 from a LlAlp2containing plasmid using appropriate primers (forward primers
199
containing -AAAAATGTCT- and -AAAAATGGCG- near the start codon) (Fig. 2) [14]. The alkaline phytase-coding sequences, LlAlp2Ala and LlAlp2-Ser, were then inserted into the expression vector pPICZA between the EcoRI and NotI sites to yield two expression constructs, pPICZA-LlAlp2-Ala and pPICZA-LlAlp2-Ser. After transformation, PCR amplification of both constructs using LlAlp2-specific full-length primers yielded a 1500 kb fragment, the expected length of full length alkaline phytase cDNA. Subsequently, the constructs were amplified with plasmid-specific primers and sequenced from both directions. The data indicated that the alkaline phytase cDNAs were inserted in frame in pPICZA; additionally, no PCR-induced mutations were detected in the expression constructs. Initial efforts to transform P. pastoris strains by sorbitol pretreatment and post-transformation recovery in the presence of sorbitol (2 h) were not very successful in our hands; the transformation efficiency was low (10 transformants per lg of DNA) (Table 1A) [26]. Transformational efficiency of P. pastoris is highly influenced by chemical pre-treatment methods that render the cells competent for DNA uptake during electroporation, electroporation conditions, and post-transformation recovery conditions [26,31–36]. Therefore, different pre-treatment and recovery conditions were examined (described in the ‘‘Materials and Methods” section). Pre-treatment with a reducing agent such as dithiothreitol (DTT) increased transformational efficiency 5-fold (Table 1B) [31]. Transformational efficiency increased 20-fold when both DTT and lithium acetate (LiAc) were included in the pre-treatment medium (Table 1C); 200 transformants per lg of DNA was observed [33]. Although the exact role of DTT and LiAc in facilitating entry of foreign DNA into yeast cells is not known, it has been suggested that reducing agents and metal ions may change the integrity of cells by modifying the mannan-protein complex that impart rigidity to yeast cell walls [35,36]. The inclusion of YPD medium during the recovery phase increased the transformational efficiency by an additional 2.5-fold; nearly 500 transformants were produced (Table 1D). All subsequent transformations were conducted under the following conditions: pre-treatment in medium containing DTT (10 mM) and LiAc (100 mM) and recovery in YPD medium containing sorbitol for a 4-h period. A number of host strains of P. pastoris that differ in their ability to utilize methanol and express foreign proteins are available for foreign gene expression [21,22]. Therefore, two strains of P. pastoris, X-33 (Mut+) and KM71H (Muts), were transformed with PmeI-
Fig. 2. Cloning of LlAlp2 into P. pastoris. (I and II): LlAlp2 cDNA coding for either Ala (I) or Ser (II) in the N-2 position was inserted between the EcoRI and NotI restriction sites into pPICZA. (III) LlAlp2 cDNA coding for Ala in the N-2 position and lacking a stop codon at the carboxyl end was inserted between the EcoRI and NotI restriction sites into pPICZA. The lack of a stop codon in LlAlp2 cDNA allowed the c-myc and (His)6 tag from the pPICZA plasmid to be fused at the C-terminus. The resulting recombinant constructs were then electroporated into X-33 or KM71H strains to produce proteins LlALP2-Ala (A) and (B), LlALP2-Ser (C) and (D), and LlALP2-Ala-c-myc-(His)6 (E).
200
S.C. Johnson et al. / Protein Expression and Purification 74 (2010) 196–203
linearized DNA of expression vectors pPICZA-LlAlp2-Ala and pPICZA-LlAlp2-Ser to produce four transformants, X-33-LlAlp2-Ala (Fig. 2;A), X-33-LlAlp2-Ser (Fig. 2;C), KM71H-LlAlp2-Ala (Fig. 2;B), and KM71H-LlAlp2-Ser (Fig. 2;D).
Screening for rLlALP2 phytase expression In order to identify the clone that produced the highest levels of alkaline phytase, the four transformants, X-33-LlAlp2-Ala, X-33LlAlp2-Ser, KM71H-LlAlp2-Ala, and KM71H-LlAlp2-Ser, were screened in parallel. Ten clones of each transformant were randomly selected and grown in baffle shake flasks. To ensure that the clones contained copies of alkaline phytase cDNA, PCR was conducted using LlAlp2-specific, full-length primers and genomic DNA as the template. All clones tested amplified a 1500 kb fragment, the expected length for full length alkaline phytase cDNA. rLlALP2 expression was induced by transferring the cells from glycerol-containing growth medium (BMGY) to medium containing methanol (BMMY). After 3 d of induction, the intracellular content of rLlALP2 was determined by lysing the cells and assaying for phytase activity. rLlALP2 activity was detected in the intracellular fraction; no activity was detected in un-induced cells. Fig. 3 indicates that clones with the same expression vector showed significant variability in total rLlALP2 activity (undetectable levels to 1200 U/25 mL of medium). Clones of KM71H and X-33 that con-
Fig. 3. Expression levels of rLlALP2 in different clones. Ten clones of each of the four transformants were selected at random (KM71H-LlALP2-Ser, KM71H-LlALP2-Ala, X-33-LlALP2-Ser, X-33-LlALP2-Ala). Alkaline phytase expression was determined by assaying for enzymatic activity in cell lysates. Reaction mixtures contained Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM), KCl (0.5 M), CaCl2 (1 mM), NaF (10 mM), and dialyzed cell lysate. Error bars represent ±SEM, n = 2.
tained LlAlp2 cDNA with Ala in the N-2 position showed a high proportion (17 of 19) of clones with total enzymatic activity greater than 200 U/25 mL compared to X-33 and KM71H transformed with the LlAlp2-Ser cDNA (2 of 20). There appeared to be little difference in total enzymatic activity between X-33 and KM71H strains when the same expression construct was used for transformation; both host strains produced a high proportion of clones with high activity (8 of 10 and 9 of 9) when LlAlp2-Ala cDNA was the construct and significantly less activity when LlAlp2-Ser cDNA was the construct. Although both Ser and Ala are considered stabilizing residues, these data are consistent with the idea that the presence of Ala in the N-2 position increases the half-life of LlALP2 in P. pastoris significantly. Optimization of expression conditions and purification of rLlALP2 To optimize expression conditions for the production of rLlALP2, the clone which produced the highest level of enzyme activity in KM71H-LlAlp2-Ala transformants was grown under a variety of conditions. Aeration is of great importance during methanol-induced expression so it is generally recommended that baffle flasks be employed for fermentation and culture volumes remain below 30% of the container volume to allow for good oxygen availability [26]. To investigate the effect of culture volume, rLlALP2 was expressed in either 25 mL (20% flask volume) or 12.5 mL (10% flask volume) of growth medium (Fig. 4A). Total rLlALP2 activity was greater (35–100%) when clones were grown in 25 mL of BMMY in the two clones investigated. The effect of methanol addition was examined by measuring total rLlALP2 activity in intracellular fraction after 3 d of induction (Fig. 4B). Varying amounts of methanol were added (0.5, 1, 2, or 3% v/v) once a day or 1% (v/v) methanol was added twice a day as indicated. Total rLlALP2 activity varied very little (±10%) under the methanol conditions examined. To determine the time point after methanol induction when intracellular accumulation of rLlALP2 activity level was the highest, cells were harvested 1–5 d post induction and rLlALP2 activity was measured (Fig. 4C). After 1 d of induction, cells had already accumulated 60% of the total activity. Cells had maximum activity after 2 d, after which total activity decreased from days 3 to 5. Optimal expression conditions employed for the rest of the experiments were as follows: baffle flasks (125 mL) contained 20% of medium (25 mL) and 1% methanol was added once a day for a period of 2 d. To facilitate the purification of the rLlALP2, cDNA coding for LlAlp2 with Ala in the N-2 position was inserted into pPICZA without
Fig. 4. Optimization of expression of rLlALP2 in P. pastoris. rLlALP2 expression was determined by assaying for enzymatic activity in cell lysate. Reaction mixtures contained Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM), KCl (0.5 M), CaCl2 (1 mM), NaF (10 mM) and dialyzed cell lysate. (A) The effect of culture volume on alkaline phytase expression in two clones (A and B) was determined using either 12.5 mL (10% flask volume) or 25 mL of media (20% flask volume) in a 125 mL baffle flask. (B) The effect of methanol addition on alkaline phytase expression was determined by adding varying amounts of methanol, 0.5%, 1%, 2%, or 3% once per day or 1% twice per day. (C) Time course of alkaline phytase expression after methanol induction: samples were harvested 1, 2, 3, 4, and 5 d after transferring to BMMY, the cells lysed and assayed for activity. Error bars represent ± SEM, n = 2.
S.C. Johnson et al. / Protein Expression and Purification 74 (2010) 196–203
201
did bind to the Co2+-resin in the presence of urea (2 M) but we observed no binding to the Ni2+-resin. The enzyme was eluted from the Co2+-resin with buffer containing increasing concentrations of imidazole and the fractions assayed for enzyme activity and analyzed by SDS–PAGE (Fig. 5). Fractions containing rLlALP2 activity indicated a dominant protein band of Mr. close to 60 kDa (Fig. 5, lanes 6 and 7), the calculated molar mass of rLlALP2 is 58 kDa; fraction corresponding to lane 7 demonstrated nearly twice the enzymatic activity as the fraction corresponding to lane 6 (data not shown). The yield obtained was 8 mg/L of broth. Fig. 5. Separation by SDS–PAGE of column fractions during rLlALP2 purification on Co2+-resin. The fusion protein rLlALP2-Ala-c-myc-(His)6 was expressed in X-33. Cells were lysed and the soluble fraction was loaded on a Co2+-resin column, washed and then eluted with buffer containing increasing concentrations of imidazole. Lane 1: molar mass markers; lane 2: crude lysate incubated in urea (2 M); lane 3: flow through; lane 4: wash 1 (Tris–HCl, pH 7.4, 300 mM NaCl); lane 5: elution 1 (Wash Buffer with imidazole, 10 mM); lane 6: elution 2 (Wash Buffer with imidazole, 30 mM); lane 7: elution 3 (Wash Buffer with imidazole 60 mM).
the stop codon so that the c-myc epitope and (His)6 tag encoded by the vector was fused at the C-terminus (Fig. 2III). Attempts to purify the rLlALP2 with immobilized metal affinity chromatography (IMAC) using Ni2+-nitrilotriacetic acid resin or Co2+-resin indicated that rLlALP2 did not bind to either resin. All enzyme activity was recovered in the flow-through fraction. The lack of binding may be due to the fact that the C-terminal (His)6 tag was not exposed for binding in the recombinant dimeric protein. Low concentrations of urea (2 M) were included in the binding buffer in an effort to expose the (His)6 tag without denaturing the enzyme. rLlALP2
Functional characterization of alkaline phytase In an effort to functionally characterize the rLlALP2 produced by P. pastoris and compare it to the native enzyme from lily pollen, the influence of a number of factors (pH, substrate specificity, temperature dependence, calcium ions, EDTA and inhibitors) on catalytic activity was probed. The influence of pH on the activity of rLlALP2 was monitored in Tris–maleate buffer from pH 5.5 to 9.0 (Fig. 6A). As observed with alkaline phytase from lily pollen [18], maximum activity was detected at pH 8.0. Lowering the pH to 6 caused a 90% drop in activity, whereas raising pH to 9 caused a 40% decrease. The pH activity profile is similar to that of wild-type alkaline phytase isolated from lily pollen [18]. LlALP from lily pollen exhibits a relatively high temperature optimum, 55 °C. The influence of temperature on the enzymatic activity of rLlALP2 was monitored at temperatures ranging from 30 °C to 70 °C (Fig. 6B). rLlALP2 had the highest total activity at
Fig. 6. Functional characterization of rLlALP2. rLlALP2 activity was assayed by measuring the phosphate released from sodium phytate at 37 °C in 60 min. The enzyme assay mixture contained Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM), KCl (0.5 M), CaCl2 (1 mM), NaF (10 mM) and dialyzed cell lysate, unless otherwise noted. (A) The effect of pH on activity was determined in Tris–maleate buffer (100 mM) at various pH. (B) The effect of temperature was determined by pre-equilibrating the assay mixtures at appropriate temperatures and the activity assayed at corresponding temperatures. The amount of phosphate released in 20 min was measured. (C) rLlALP2 activity in an assay mixture containing different phosphorylated substrates (1 mM). (D) rLlALp2 activity in the standard mixture with calcium and/or EDTA. (E) rLlALP2 activity in the absence or presence of inhibitors (N). Error bars represent ± SEM, n = 2.
202
S.C. Johnson et al. / Protein Expression and Purification 74 (2010) 196–203
55 °C, a sharp decrease in activity was observed at temperatures above 55 °C, consistent with the observations with the native LlALP [18]. The substrate specificity of LlALP from lily pollen is unique among phytases; it differs from that of acid phytases, which show broad substrate specificity, and from other alkaline phytases, which exhibit narrow substrate specificity including lack of activity against para-nitrophenyl phosphate (pNPP). LlALP from lily pollen hydrolyzes phytate and pNPP. The rLlALP2 expressed in P. pastoris hydrolyzed phytate and pNPP and exhibited no activity towards ATP (Fig. 6C). Activity against pNPP was nearly 2.5-fold higher than against phytate, similar to what is observed with the native enzyme [18]. The native enzyme from lily pollen exhibits a 3.5–4.0-fold enhancement of activity in the presence of added CaCl2 (1 mM), however, some hydrolytic activity is observed even in the presence of relatively high concentrations of EDTA (5 mM) (Fig. 6D). The effect of Ca2+ ions on enzymatic activity of rLlALP2 was investigated, with and without added CaCl2 (1 mM) and EDTA (5 mM). Consistent with observations from the native enzyme, the recombinant enzyme showed a 4-fold enhancement when CaCl2 was added; the enzyme retained some enzyme activity (20%), albeit low, even in the presence of high concentration of EDTA (5 mM) (Fig. 6D). This could be because of the presence of tightly bound Ca2+ at the enzyme active site that cannot be extracted with EDTA or because LlALP is able to catalyze phytate at a low level even in the absence of CaCl2. A major difference between acid and alkaline phytases is the effect of inhibitors. LlALP from lily pollen is inhibited by vanadate, a transition state analog, but not inhibited by fluoride, a strong inhibitor of many histidine acid phytases [18]. Addition of relatively high concentrations of NaF (10 mM) had only a slight effect on total activity (10% decrease) whereas rLlALP2 was strongly inhibited by vanadate (85% inhibition at 1 mM) (Fig. 6E), consistent with what was observed with the native enzyme [18].
Discussion The studies reported here demonstrate that LlALP2 from lily pollen can be expressed in P. pastoris in a soluble active form. Our attempts to express rLlALP2 in E. coli resulted in accumulation of the enzyme in inclusion bodies and numerous efforts to solubilize and refold alkaline phytase were unsuccessful. Related to this is our observation from previous studies that when soluble native alkaline phytase from lily pollen was denatured under mild conditions (0.5% sodium dodecyl sulfate and b-mercaptoethanol), we were not successful in renaturing it to the active form despite extensive efforts [37]. The difficulty in refolding denatured alkaline phytase into the active form may be due to the fact that the native active alkaline phytase is a homodimer with 20 Cys residues (each monomer has 10 Cys) and disulfide linkages may play an important role in protein folding. The fact that P. pastoris can express an active dimeric protein with many Cys residues suggests that the intracellular environment of P. pastoris promotes correct disulfide bond formation, a significant advantage over bacterial expression systems. Intracellular protein degradation will affect the levels of foreign proteins that can be harnessed from P. pastoris. Although both Ala and Ser at the N-2 position are stabilizing amino acids in S. cerevisiae, we isolated significantly more protein when Ala was in the N2 position compared to when Ser was at N-2 (Fig. 3); on an average N-2-Ala clones produced 7-fold more LlALP2 than N-2-Ser clones. This difference in yield could be due to differences in protein degradation mechanisms between P. pastoris and S. cerevisiae or due to unique characteristics of the protein.
The rLlALP2 expressed in P. pastoris showed Mr. and catalytic properties very similar to the wild-type enzyme. The substrate specificity and temperature dependence of catalysis as well as the effect of pH, inhibitors, calcium ions, and EDTA were very similar to that observed with the wild-type enzyme from lily pollen, suggesting that the folding of the recombinant enzyme is very similar to that of the wild-type enzyme. (His)6 and c-myc peptide tags at the N- or the C-terminal are widely used to detect proteins with specific antibodies and purify proteins by affinity chromatography [38]. However, the success of these methods depends on the presence of unmasked tags. We observed no binding of rLlALP2 to Ni2+-NTA or Co2+-resin. When low concentration of urea (2 M) were added to the binding buffer, rLlALP2 bound to the Co2+-resin, possibly due to partial unfolding and unmasking of the (His)6 tag, no binding to the Ni2+-NTA resin was observed even in the presence of urea. This observation is similar to that noted by Li et al. [39]. Active rLlALP2 was eluted from the column with buffer containing imidazole. An observation related to this issue is that, although phytate hydrolyzing activity was readily detected in enzyme assays of cell lysate, our efforts to detect the presence of the fusion protein by Western blot analysis with anti-c-myc antibodies or anti-(His)6 antibodies were unsuccessful. Taken together, these data suggest that both the cmyc and (His)6 tags are masked in the folded protein. We have demonstrated in this study that LlALP2 from lily pollen can be heterologously expressed in P. pastoris in a soluble active form. This is the first report of the heterologous expression of a plant phytase in an active soluble form. Transformed P. pastoris can serve as a potential source of substantial amounts of active rLlALP2 for structural, biochemical, and animal feed studies. Acknowledgments We gratefully acknowledge MTU for Graduate Student Fellowship to S.C.J. We thank Dr. Bakul Dhagat Mehta for constructing the LlAlp2 cDNA-containing pPET vector and Dr. Sonali Jog for technical discussions. We thank Dr. W.H. Campbell (Nitrate Elimination Company) for many helpful discussions on protein expression in P. pastoris. References [1] B.C. Oh, W.C. Choi, S. Park, Y.O. Kim, T.K. Oh, Biochemical properties and substrate specificities of alkaline and histidine acid phytases, Appl. Microbiol. Biotechnol. 63 (2004) 362–372. [2] E.J. Mullaney, A.H.J. Ullah, The term phytase comprises several different classes of enzymes, Biochem. Biophys. Res. Commun. 312 (2003) 179–184. [3] U. Konietzny, R. Greiner, Molecular and catalytic properties of phytatedegrading enzymes (phytases), Int. J. Food Sci. Technol. 37 (2002) 791–812. [4] R.J. Wodzinski, A.H.J. Ullah, Phytase, Adv. Appl. Microbiol. 42 (1996) 263–302. [5] J.E. Coleman, Structure and mechanism of alkaline-phytase, Annu. Rev. Biophys. Biomol. Struct. 21 (1992) 441–483. [6] J.B. Vincent, M.W. Crowder, B.A. Averill, Hydrolysis of phosphate monoesters: a biological problem with multiple chemical solutions, Trends Biochem.Sci. 17 (1992) 105–110. [7] A.R. Alcazar-Roman, S.R. Wente, Inositol polyphosphates: a new frontier for regulating gene expression, Chromosoma 117 (2008) 1–13. [8] R.H. Michell, Inositol derivatives: evolution and functions, Nat. Rev. Mol. Cell Biol. 9 (2008) 151–161. [9] A.T. Miller, P.P. Chamberlain, M.P. Cooke, Beyond IP3 – roles for higher order inositol phosphates in immune cell signaling, Cell Cycle 7 (2008) 463–467. [10] F.A. Loewus, P.P.N. Murthy, Myo-inositol metabolism in plants, Plant Sci. 150 (2000) 1–19. [11] M.J. Berridge, Inositol trisphosphate and calcium signaling, Nature 361 (1993) 315–325. [12] N.R. Reddy, S.K. Sathe, D.K. Salunkhe, Phytates in legumes and cereals, Adv. Food Res. 28 (1982) 1–92. [13] B.G. Baldi, J.J. Scott, J.D. Everard, F.A. Loewus, Localization of constitutive phytases in lily pollen and properties of the pH 8 form, Plant Sci. 56 (1988) 137–147. [14] B.D. Mehta, S.P. Jog, S.C. Johnson, P.P.N. Murthy, Lily pollen alkaline phytase is a histidine phosphatase similar to mammalian multiple inositol polyphosphate phosphatase (MINPP), Phytochemistry 67 (2006) 1874–1886.
S.C. Johnson et al. / Protein Expression and Purification 74 (2010) 196–203 [15] X.G. Lei, C.H. Stahl, Biotechnological development of effective phytases for mineral nutrition and environmental protection, Appl. Microbiol. Biotechnol. 57 (2001) 474–481. [16] A.N. Sharpley, S.C. Chapra, R. Wedepohl, J.T. Sims, T.C. Daniel, K.R. Reddy, Managing agricultural phosphorus for protection of surface waters – issues and options, J. Environ. Qual. 23 (1994) 437–451. [17] T.S. Nelson, T.R. Shieh, R.J. Wodzinski, J.H. Ware, Effect of supplemental phytase on the utilization of phytate phosphorus by chicks, J. Nutr. 101 (1971) 1289–1293. [18] S.P. Jog, B.G. Garchow, B.D. Mehta, P.P.N. Murthy, Alkaline phytase from lily pollen: investigation of biochemical properties, Arch. Biochem. Biophys. 440 (2005) 133–140. [19] N. Armstrong, A. De Lencastre, E. Gouaux, A new protein folding screen: application to the ligand binding domains of a glutamate and kainate receptor and to lysozyme and carbonic anhydrase, Protein Sci. 8 (1999) 1475–1483. [20] P.Z. Li, A. Anumanthan, X.G. Gao, K. Ilangovan, V.V. Suzara, N. Duzgunes, V. Renugopalakrishnan, Expression of recombinant proteins in Pichia pastoris, Appl. Biochem. Biotechnol. 142 (2007) 105–124. [21] J.M. Cregg, J.L. Cereghino, J.Y. Shi, D.R. Higgins, Recombinant protein expression in Pichia pastoris, Mol. Biotechnol. 16 (2000) 23–52. [22] J.L. Cereghino, J.M. Cregg, Heterologous protein expression in the methylotrophic yeast Pichia pastoris, FEMS Microbiol. Rev. 24 (2000) 45–66. [23] R. Daly, M.T.W. Hearn, Expression of heterologous proteins in Pichia pastoris: a useful experimental tool in protein engineering and production, J. Mol. Recognit. 18 (2005) 119–138. [24] S. Macauley-Patrick, M.L. Fazenda, B. McNeil, L.M. Harvey, Heterologous protein production using the Pichia pastoris expression system, Yeast 22 (2005) 249–270. [25] J. Sambrook, D. Russel (Eds.), Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, New York, 2001. [26] Invitrogen, EasySelectTM Pichia Expression Kit, Carlsbad, CA, 2009. [27] L. Barrientos, J.J. Scott, P.P.N. Murthy, Specificity of hydrolysis of phytic acid by alkaline phytase from lily pollen, Plant Physiol. 106 (1994) 1489–1495.
203
[28] R. Hamilton, C.K. Watanabe, H.A. de Boer, Compilation and comparison of the sequence context around the AUG startcodons in Saccharomyces cerevisiae mRNAs, Nucleic Acids Res. 15 (1987) 3581–3593. [29] A. Varshavsky, The N-end rule: functions, mysteries, uses, Proc. Natl. Acad. Sci. USA 93 (1996) 12142–12149. [30] A. Mogk, R. Schmidt, B. Bukau, The N-end rule pathway for regulated proteolysis: prokaryotic and eukaryotic strategies, Trends Cell Biol. 17 (2007) 165–172. [31] J. Lin-Cereghino, W.W. Wong, S. Xiong, W. Giang, L.T. Luong, J. Vu, S.D. Johnson, G.P. Lin-Cereghino, Condensed protocol for competent cell preparation and transformation of the methylotrophic yeast Pichia pastoris, BioTechniques 38 (2005) 44–48. [32] J.R. Thompson, E. Register, J. Curotto, M. Kurtz, R. Kelly, An improved protocol for the preparation of yeast cells for transformation by electroporation, Yeast 14 (1998) 565–571. [33] S. Wu, G. Letchworth, High efficiency transformation by electroporation of Pichia pastoris pretreated with lithium acetate and dithiothreitol, BioTechniques 36 (2004) 152–154. [34] J.M. Cregg, K.A. Russell, Transformation, in: D.R. Higgins, J.M. Cregg (Eds.), Pichia Protocols, Humana Press, Totowa, 1998, pp. 27–39. [35] B. Brzobohaty, L. Kovac, Factors enhancing genetic-transformation of intact yeast-cells modify cell-wall porosity, J. Gen. Microbiol. 132 (1986) 3089–3093. [36] H. Ito, K. Murata, A. Kimura, Transformation of intact yeast-cells treated with alkali cations or thiol compounds, Agric. Biol. Chem. 48 (1984) 341–347. [37] S. Jog, Biochemical Characterization and Recombinant Expression of Alkaline Phytase from Lilium longiflorum, Ph.D. Dissertation, Michigan Technological University, Houghton, MI 49931, USA, 2010. [38] H. Block, B. Maertens, A. Spriestersbach, N. Brinker, J. Kubicek, R. Fabis, J. Labahn, F. Schafer, Immobilized-metal Affinity Chromatography (IMAC): A Review, Guide to Protein Purification, second ed., Elsevier Academic Press Inc., San Diego, 2009. pp. 439–473. [39] Z. Li, A. Moy, K. Sohal, C. Dam, P. Kuo, J. Whittaker, M. Whittaker, N. duzgunes, K. Konopka, A.H. Franz, J. Lin-Cereghino, G.P. Lin-cereghino, Expression and characterization of recombinant human secretory leukocyte protease inhibitor (SLPI) protein from Pichia pastoris, Protein Expr. Purif. 67 (2009) 175–181.