A thermostable phytase from Bacillus sp. MD2: cloning, expression and high-level production in Escherichia coli
- First Online:
- Cite this article as:
- Tran, T.T., Mamo, G., Mattiasson, B. et al. J Ind Microbiol Biotechnol (2010) 37: 279. doi:10.1007/s10295-009-0671-3
- 527 Views
Phytase is used as a feed additive for degradation of antinutritional phytate, and the enzyme is desired to be highly thermostable for it to withstand feed formulation conditions. A Bacillus sp. MD2 showing phytase activity was isolated, and the phytase encoding gene was cloned and expressed in Escherichia coli. The recombinant phytase exhibited high stability at temperatures up to 100°C. A higher enzyme activity was obtained when the gene expression was done in the presence of calcium chloride. Production of the enzyme by batch- and fed-batch cultivation in a bioreactor was studied. In batch cultivation, maintaining dissolved oxygen at 20–30% saturation and depleting inorganic phosphate below 1 mM prior to induction by IPTG resulted in over 10 U/ml phytase activity. For fed–batch cultivation, glucose concentration was maintained at 2–3 g/l, and the phytase expression was increased to 327 U/ml. Induction using lactose during fed-batch cultivation showed a lag phase of 4 h prior to an increase in the phytase activity to 71 U/ml during the same period as IPTG-induced production. Up to 90% of the total amount of expressed phytase leaked out from the E. coli cells in both IPTG- and lactose-induced fed-batch cultivations.
KeywordsAlkaline phytaseBacillus sp.Fed-batch cultivationProtein secretion
Phytase is an important enzyme used as an additive in animal feed. The enzyme catalyses the sequential hydrolysis of phytate (myo-inositol 1,2,3,4,5,6-hexakisphosphate; IP6) present in plant material to less phosphorylated myo-inositol derivatives with concomitant release of inorganic phosphate . The phosphate is made available to monogastric animals, such as pigs, chickens and fish, that do not produce phytase and also their microflora cannot degrade phytate. This eliminates the need for external addition of phosphorous to the feed, which incurs costs and also contributes to environmental pollution [3, 23, 43]. Removal of phytate from the feed also has other advantages in terms of overcoming its negative nutritional effects such as decrease in the bioavailability of vital minerals [20, 21, 27, 30], impairing the function of digestive enzymes or decrease in the digestibility of feed protein [2, 3, 17]. For the phytase to be effective, it needs to be thermostable enough to withstand the high temperatures used during feed preparation.
A wide variety of phytases are known from different organisms; however, the focus has been on the enzymes of fungal origin, such as Aspergillus spp. [36, 38], which are significantly active at low pH as in stomach where phytate exists in a metal-free form but has low thermostability [19, 22, 28, 45]. A few phytases, known as alkaline phytases, which degrade phytate that is present as a metal-phytate complex in plants, have been reported from Bacillus spp. [3, 7, 12–14, 37] and pollen of some plants such as Typha latifolia  and Lilium longiflorum . Such phytases can potentially be used for treatment of the animal feed prior to feeding (i.e., during feed mixing, pelleting and storage).
Several phytase genes have been successfully cloned and expressed in different microbial hosts, corn kernels of transgenic maize and saliva of transgenic pigs [5, 12, 13, 17, 19, 39]. Escherichia coli is often the primary choice as host microorganism for production of recombinant proteins. Some of the earlier attempts with expression of Bacillus phytases in E. coli have resulted in production of inactive enzyme in the form of inclusion bodies [12, 33]. This entails additional steps for recovery of the active protein . Kim et al.  have reported production of Bacillus sp. DS11 phytase to 20% content of the total soluble protein in E. coli BL21(DE3) using the pET-22b(+) vector with the inducible T7 promoter.
Many strategies have been devoted to the establishment of high cell density processes to increase the level of production of recombinant target proteins by E. coli. Most of them are related to controlling the substrate concentration at certain levels in order to avoid overflow metabolism and the resultant acetate accumulation under glucose excess conditions [1, 10, 18, 31]. Åkesson and coworkers  have previously proposed a feeding strategy controlled automatically by the dissolved oxygen level such that aerobic conditions are maintained in spite of the limitations in oxygen transfer, i.e., the feeding rate is decreased when maximum oxygen transfer capacity is reached in the bioreactor.
In this study, we report cloning and expression of a thermostable alkaline phytase from a newly isolated Bacillus sp. MD2 in recombinant Escherichia coli, and its production by high cell density cultivation using a strategy based on dissolved oxygen level to control stirrer speed and to regulate feeding to avoid anaerobic conditions and acetate accumulation.
Materials and methods
Bacterial strains, plasmids and media
Isolate MD2 was obtained from an excrement sample in Hanoi (Vietnam) on meat peptone agar (MPA) medium containing 1% sodium phytate. Escherichia coli strains DH5α and BL21(DE3) were purchased from Invitrogen (Carlsbad, CA) and Novagen (Madison, WI), respectively. The cloning vector pJET1.2 and expression vector pET-22b(+) were purchased from Fermentas (MBI Fermentas, Germany) and Novagen, respectively. pE10C2 plasmid was constructed in this study from pET-22b(+) and the phytase gene from Bacillus sp. MD2. The recombinant E. coli strain was cultivated either in LB medium supplemented with 100 μg/ml of ampicillin or synthetic glucose-minimal medium composed of (per liter): 10 g glucose, 2 g (NH4)2SO4, 4.807 g K2HPO4 and 1.067 g NaH2PO4.H2O, 0.5 g (NH4)2HC6H5O7, 2 mM MgSO4, 2 ml trace element solution (composed of 0.5 g CaCl2, 16.7 g FeCl3, 0.18 g ZnSO4.7H2O, 0.16 g CuSO4.5H2O, 0.15 g MnSO4.H2O, 0.18 g CoCl2.6H2O, and 20.1 g sodium EDTA per liter) , and 100 mg ampicillin.
Identification of Bacillus sp. MD2
Genomic DNA was extracted from the isolate MD2 according to Sambrook et al. . The 16S rRNA gene of isolate was amplified and sequenced using a forward primer 8–27f (AGAGTTTGATCCTGGCTCAG) and a reverse primer 1492R (GGTTACCTTGTTACGACTT). The 16S rDNA sequence was compared with sequences available in public databases. Sequencing was done using BigDye® Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA) following the manufacturer’s protocols. Sequence reactions were electrophoresed using the ABI PRISM® 3100 DNA sequencer.
Cloning and expression of a phytase gene from the isolate MD2
Genomic DNA was extracted from the isolate MD2 following the method of Sambrook et al. . The phytase encoding gene was amplified from the genomic DNA using a forward primer Phyf (TATAAAGCTTCTGTCTGATCCTTATCATTTTACCG) and a reverse primer Phyr (TCTCTCGAGTTTTCCGCTTCTGTCGGTCAG), which were designed based on the phytase gene sequence of closest organisms identified by 16S rDNA sequence similarity search. The gene amplification was performed by polymerase chain reaction (PCR) using 5 μl of Pfu DNA polymerase buffer (Fermentas), 1.25 units of Pfu DNA polymerase (Fermentas), 1.25 units of Vent DNA polymerase (New England BioLabs), 1 μg genomic DNA and 0.25 μM primers in a total volume of 50 μl. The reaction was run for 15 cycles of 30 s at 94°C, 30 s at 55°C and 1 min at 72°C. Finally, an additional post-cycle extension was carried out for 7 min at 72°C. The amplified fragments were purified and cloned into blunt cloning vector pJET1.2 and transformed to E. coli DH5α competent cells. After colony PCR screening, plasmids were purified from the insert positive clones by alkaline lysis method and then digested with HindIII/XhoI. The released inserts were re-cloned into expression vector pET-22b(+) to construct pE10C2 that was transformed into E. coli BL21(DE3) cells. All the cloning steps were done according to Sambrook et al. .
Shake flask cultivation of the recombinant E. coli
The E. coli BL21 (DE3) cells harboring the recombinant vector pE10C2 were grown on a LB-agar plate containing 100 μg/ml of ampicillin at 30°C. A single colony was picked from the plate and cultured for 5 h in 5 ml LB broth containing 100 μg/ml of ampicillin at 30°C and 200 rpm. This culture was used to inoculate 100 ml of the same medium, which was incubated under similar conditions in a gyratory shaker incubator. When the optical density (OD) of the culture at 620 nm was about 0.7, isopropyl β-d-thiogalactoside (IPTG) was added to a final concentration of 1 mM. Culture samples (3 ml) were withdrawn at different time intervals for analysis of enzyme activity in extracellular, cytoplasmic and periplasmic fractions according to Novagen pET System manual (http://www.novagen.com). The extracellular activity was determined in the cell-free supernatant obtained after centrifugation of the sample at 8,000 g for 10 min at 4°C. The cell pellet was washed with physiological saline and resuspended in 30 mM Tris–HCl buffer pH 8 containing 20% (w/v) sucrose and 5 mM CaCl2, and then EDTA was added to a final concentration of 1 mM. After gentle shaking at room temperature, the cells were harvested by centrifugation and subjected to osmotic shock by re-suspending in ice-cold 5 mM MgSO4, and shaken for 10 min on ice. After centrifugation, periplasmic activity was determined in the supernatant, while the remaining cell pellet was re-suspended in 0.1 M Tris–HCl buffer, pH 7, containing 5 mM CaCl2, and the cytoplasmic content was released by four rounds of sonication. All the assays were performed in triplicates.
Batch- and fed-batch cultivations of recombinant E. coli in a bioreactor
A 3 liter bioreactor (Chemoferm FCL-B-3, Hägersten, Sweden) containing 2.5 l of minimal medium was autoclaved for 40 min at 121°C. After cooling, the sterilized medium was supplemented with sterilized solutions of glucose, trace elements, MgSO4 and ampicillin, respectively, to the desired concentration. The medium was inoculated with 100 ml culture of E. coli BL21(DE3) harboring pE10C2 grown in the same medium for 16 h at 30°C with shaking at 200 rpm. For batch cultivations, phosphate buffer in the medium was replaced by 50 mM Tris-maleate buffer pH 7.0.
Cultivations in the bioreactor were performed at 30°C with stirring at an initial speed of 200 rpm. The dissolved oxygen (DO) level was monitored using a DO probe, and when reduced to 30% saturation, the stirrer speed was increased automatically, and the DO was thereafter controlled at 20–30% saturation. The pH of the culture was maintained by addition of 3.57 M ammonia solution, which was controlled by a signal from the pH probe in the bioreactor. The volume of the ammonia solution consumed during the cultivation was recorded. The concentration of glucose was monitored off line in duplicates every 30 min by Accucheck glucose test strips. Inorganic phosphate (Pi) in the fermentation broth was monitored in triplicates at regular time intervals according to the method of Shimizu  (see ahead under phytase activity assay). For induction of the phytase production, the IPTG or lactose was added to the bioreactor to a desired concentration when the concentration of inorganic phosphate had decreased to less than 1 mM.
Fed-batch cultivation was started using a continuous feed of a solution containing 50% (w/v) of glucose, 50 ml/l of 1 M MgSO4 and 10 ml/l of trace element solution in order to maintain the glucose concentration at 2–3 g/l in the bioreactor.
Phytase activity assay
The phytase activity was assayed according to the method described by Shimizu . Unless otherwise mentioned, a mixture of appropriately diluted enzyme sample and sodium phytate (1.5 mM) dissolved in 100 mM Tris–HCl buffer, pH 7, containing 5 mM CaCl2 was incubated at 70°C for 20 min. The reaction was stopped by addition of an equal volume of 15% trichloroacetic acid (TCA) solution. The precipitate obtained was removed by centrifugation, and 500 μl of the clear supernatant was transferred to an Eppendorf tube to which was then added 500 μl ammonium molybdate reagent (containing 1:4 mixture of 2.7% FeSO4 and 1.5% ammonium molybdate in 5.5% H2SO4). After 5 min incubation at room temperature, absorbance at 700 nm was read, which was used to calculate the concentration of Pi from a standard curve made using sodium dihydrogen phosphate solution in the concentration range of 0–600 μmol/l Pi. All the samples were assayed in triplicates. One unit of phytase activity was defined as the amount of enzyme that releases 1 μmole of Pi per minute under the assay conditions.
The purity of the phytase samples prepared from the cultivation broth was checked by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), using 12.5% (w/v) polyacrylamide.
Protein concentration was measured by bicinchoninic acid method using bovine serum albumin as standard. The recorded values were an average of measurements of triplicate samples.
Results and discussion
Isolation and identification of phytase positive strain
The isolate MD2 showing positive phytase activity (0.2 units/ml) was obtained from a sample of infant excreta after 3 days of incubation at 37°C in MPA medium containing 1% sodium phytate. The 1,407-nucleotide-long 16S rRNA gene sequence (GenBank accession number GU143091) of the isolate showed the highest similarity (99%) to Bacillus subtilis, B. amyloliquefaciens and B. velezensis sequences.
Sequence analysis of Bacillus sp. MD2 phytase
The phytase encoding gene (GenBank accession number GU143090) of Bacillus sp. MD2 was amplified and sequenced so that the amino acid sequence could be deduced. The sequence comprised a mature peptide of 354 amino acids, and like other phytases of Bacillus origin, it lacked RHGXRXP, a conserved motif at catalytic site of acidic phytases . The sequence is highly similar to that of ts-phy from Bacillus amyloliquefaciens DS11 , the only difference being at positions 81 and 148 where Ala and Asn in ts-phy are replaced by Val and Asp, respectively. Close similarity is also seen with phyC from Bacillus subtilis VTTE68013 , the sequences differing by eight residues.
Expression and localization of recombinant Bacillus sp. MD2 phytase
Supplementation of the medium with CaCl2 during induction resulted in expression of higher phytase activity, with maximum activity of 3.6 U/ml being obtained at 10 mM CaCl2. The activity was detected in all three fractions, although the extracellular activity was the most predominant. The periplasmic phytase activity started to decrease after 10 h of induction while the activity in the extracellular fraction increased to about 50% of the total activity (Fig. 1b). The recombinant phytase was estimated to constitute about 19% of the total soluble protein present in the cells. The higher activity obtained on addition of CaCl2 seems to be due to stimulation of the activity and increased stability of the phytase in the presence of the salt. Increasing the cultivation temperature to 37°C decreased the level of expression, but increased the extent of secretion of the recombinant enzyme into the culture medium (data not shown).
The possibility to secrete the target protein into the extracellular medium provides several advantages for enzyme production including a better folding environment free of cell-associated proteolytic degradation and easier recovery without the need for cell disruption and separation of cell debris . There are several reports on the recombinant proteins directed to the periplasm being exported out of the cells [26, 29, 41]. This has been attributed to the stress on the outer membrane due to periplasmic accumulation of proteins resulting in loss of membrane integrity or triggering of autolytic response caused by change in cultivation conditions. Leakage of the periplasmic enzymes by increasing the growth temperature is likely to be caused by induction of heat-shock response and activation of the phospholipase activity in the outer membrane . There have also been many efforts to engineer E. coli to excrete proteins extracellularly . Recently, secretion of Bacillus hydrolytic enzymes using their native signal peptides or that of E. coli outer membrane protein, OmpA has been achieved . Miksch et al.  have used the coexpression of Kil protein to secrete a recombinant E. coli phytase.
Properties of the recombinant Bacillus sp. MD2 phytase
In order to characterize the enzyme, the recombinant Bacillus sp. MD2 phytase was purified to homogeneity from the crude cell homogenate by heat treatment, immobilized metal affinity chromatography and gel filtration to yield an enzyme with a specific activity of 39 U/mg. The pure enzyme had a molecular weight of 47.5 kDa and was optimally active at pH 6–7 and 67–73°C. The presence of 2–5 mM Ca2+ was essential for the enzyme to be active at 70°C. It was highly stable in the pH range of 3.6–9.6 at room temperature, and exhibited remarkable thermostability in the presence of 5 mM CaCl2, retaining about 40% of the activity at pH 7 after 10 min at 100°C (data not shown). This is a desirable feature for the enzymes for animal feed applications in order to tolerate pelleting temperature of 80–85°C. Indeed, evaluation of the ts-phytase from B. amyloliquefaciens, to which the MD2 phytase is related, has shown promise for application in feed treatment .
A distinguishing feature of the MD2 phytase with respect to the other alkaline phytases that possess strict specificity for phytate substrate [4, 28] was that the enzyme, although being most active with phytate, displayed even some degree of activity with other phosphate substrates, such as ADP, ATP, α-fructose-1,6-diphosphate, p-nitrophenyl phosphate and sodium pyrophosphate.
Production of the recombinant phytase by batch cultivation in a bioreactor
Production of the recombinant phytase by fed-batch cultivation and induction by IPTG
Comparison of phytase production in batch and fed-batch cultivations in this study with those reported in literature
Cultivation type, inducer
Total phytase production (U/ml)
Lactose-induced production of the recombinant phytase during fed-batch cultivation
Expression of phytase activity by recombinant E. coli cells induced by different concentrations of lactose
Inducer concentration (mM)
Total activity (U/ml) at different hours after induction
IPTG, 1 mM
Lactose, 5 mM
Lactose, 10 mM
Lactose, 15 mM
Lactose, 20 mM
Lactose, 25 mM
Production of the recombinant phytase was then studied by fed-batch cultivation with 20 mM lactose as the inducer. The higher concentration of lactose was used assuming that some of it could be consumed by the E. coli cells at a low glucose concentration (2–3 g/l) . All the parameters were controlled in the same way as described above for IPTG-induced cultivations. After induction, cells continued growing for 4 h at a maximal specific growth rate of 0.49; however, phytase production was very low in contrast to similar cultivation conditions using 1 mM IPTG as the inducer. Phytase production increased dramatically after 4 h of induction. Delay in enzyme production on lactose induction during fed-batch cultivation of E. coli BL21(DE3) cells has been reported earlier [32, 42] and could be explained by the time needed for lactose to be processed to allolactose, an inducer for lac UV5 in E. coli chromosome and lac o in pET-22b(+) [11, 25].
Higher cell density (57 g cell dry weight/l) was reached at the end of cultivation due to longer exponential growth phase. However, the total phytase activity after 5 h induction was only 71 U/ml (1.25 U/g cell dry weight), which is approximately 22% of that obtained on induction by IPTG (Fig. 5). This indicated that the conditions used for induction by lactose during high cell density cultivation were not optimal. It has previously been suggested that the residual lactose concentration plays an important role in the induction process; too high concentrations may lead to inhibition, while limitation occurs at concentrations below a critical value . It is thus likely that optimization of the amount of lactose and mode of addition could further increase the phytase activity levels. It is possible that prolonged incubation (beyond 5 h) could lead to higher enzyme activity in case of both IPTG and lactose-induced fed-batch cultivations but bioreactor capacity posed a limitation.
Distribution of the recombinant phytase during the fed-batch cultivation induced by lactose is shown in Fig. 5b. As in the IPTG-induced cultivation, up to 90% of total phytase activity had leaked out of the cells into the medium. The possibility to use lactose as an inducer for recombinant enzyme production would substantially lower the production cost, especially when added in the form of whey or whey permeate .
Bacillus sp. MD2 phytase is a highly thermostable enzyme and has good potential as a supplement to feed for monogastric animals. In this study, selection of a suitable expression host and fermentation conditions have resulted in a drastic increase in the production levels of the enzyme. Although large-scale production is normally done using fungal or other microbial hosts that are able to secrete high enzyme amounts, E. coli serves as a convenient host for laboratory-scale studies allowing structure-function studies and also evaluation of the product in the desired application. Moreover, extracellular expression is beneficial for avoiding the formation of inclusion bodies or degradation by host proteases, and also lowers the risk of bacterial endotoxins entering the enzyme product.
Currently, the enzyme is being evaluated with regard to feed utilization and animal growth rates. Studies are also on-going at the molecular level to investigate the possibility of increasing the activity of the enzyme in the acidic range.
The financial support of the Swedish Agency for Development Cooperation (Sida/SAREC) is gratefully acknowledged.