Interactions of phytate and myo-inositol phosphate esters

*Enzyme R & D, Genencor, Danisco A/S, DK 8220 Brabrand, Aarhus, Denmark; †Department of Biotechnology, Lund University, Lund, Sweden; ‡Poultry Research Foundation, Veterinary Science Faculty, University of Sydney, Camden,NSW 2570, Australia; and §Danisco Animal Nutrition, SN8 1AA Marlborough, United Kingdom 

Phytic acid (IP₆) and myo-inositol phosphate esters (IP_1-5), including IP₅ isomers prepared chemically and enzymatically with bacterial and fungal phytases, were examined for their effects on protein aggregation of soy protein and β-casein, interaction with Fe³⁺, and pepsin activity. The results indicated that the aggregating capabilities of IP esters (IP_1-6) on the 2 proteins decreased dramatically from IP₆ to IP₅ and became negligible with IP_1-4. Among the IP₅ isomers tested, InsP₅ (1,2,3,4,5) produced by 6-phytase was slightly less powerful in aggregating protein than InsP₅ (1,2,4,5,6) produced by 3-phytase (P = 0.001). For protein hydrolysis, IP esters of IP_3-4 still showed inhibition of pepsin though to a lesser extent than IP_5-6. The in vitro data with IP_1-5 generated with microbial 3- and 6-phytases indicate that, for complete alleviation of pepsin inhibition, IP₆ needs to be broken down to IP₁-2. In contrast to the aggregation with protein, the reactivity of IP_1-6 toward Fe³⁺ decreased proportionally from IP₆ to IP₃. Based on the radical decrease in turbidity of IP₆ -protein complex observed, as a result of IP₆ dephosphorylation to IP₅, a novel qualitative and semi-quantitative phytase plate assay was established using IP₆-protein complex incorporated into an agarose petri-dish as substrate. Phytase activity was shown as the development of clear halos on the agarose plate with time. This simple phytase plate assay method can be used at animal farms, control laboratories, and even for the screening of engineered phytase variants. The current study, thus, stresses the importance of the effi cient hydrolysis of IP₆ at lower pH range to alleviate the negative effect of phytic acid and its degradation products on protein and Fe³⁺ digestion.

Keywords: inositol phosphate, pepsin, phytase assay, phytic acid

It is well known that phytate (myo-inositol hexakisphosphate, IP₆) is the major form of inorganic phosphate storage in plant seed and is mobilized when plant seeds germinate. Phytic acid content in cereals is usually in the range of 0.8 to 4.0% of dry weight, and it can be a rich source of inorganic phosphate (Pi) for animals and humans (Reddy, 2002; Greiner and Konietzny, 2010). At acidic pH, IP₆ may interact with dietary protein, leading to the formation of IP₆-protein aggregates, which has been well documented in the literature (Reddy et al., 1989; Rajendran and Prakash, 1993). For example, Murthy and Rao (1984) reported the maximum interaction of IP₆ with mustard 12S protein occurred at pH 3.0 with the binding capacity of 170 mol Pi mol⁻¹ protein. These aggregates may have decreased accessibilities to protease, possibly resulting in ineffi cient protein digestion in the gastric phase, thereby leaving inappropriate protein substrate for the pancreatic protease (Knuckles et al., 1989; Kies et al., 2006). Phytic acid-protein complex occurs as a result of masking of positive charges of the protein by IP₆ (Reddy et al., 1989).

Figure 1. A sketch of the reactions c atalyzed by 3-phytase (EC 3.1.3.8) and 6-phytase (EC 3. 1.3.26). The products released beside inorganic P and myoinositol monophosphate can include metal divalent and trivalent ions, proteins, peptides, and AA (NH³⁺-CHR). The arrows indicate that the ester bond is initially attacked by the 3- and 6-phytase, respectively. The myo-inositol molecule was drawn in the so-called "turtle" confi guration (Agranoff, 2009).

Phytic acid also binds minerals, such as Fe³⁺, Zn²⁺, Ca²⁺, and Mn²⁺, which results in the formation of phytate (Weaver and Kannan, 2002; Tang et al., 2006) and other positively charged feed additives through electrostatic interactions (Figure 1). Such interactions can occur in cereals as protein bodies (globoids; Bohn et al., 2007), and also in the gastrointestinal tracts of animals and humans. Thus, the hydrolysis of IP₆ will not only release Pi, but also several positively charged nutrients (Figure 1). In this study, the effects of inositol phosphate ester (IP_1-6) and IP₅ positional isomers obtained chemically or prepared enzymatically using commercial microbial phytases were investigated for their effects on protein aggregation, pepsin activity, and interaction with Fe³⁺. 

Bovine β-casein, porcine pepsin (P7000, 370 UU≅  mg−1; P7012, 2,500 to 3,000 UU≅ mg−1), myo-inositol phosphate esters IP₁(2), IP₂(2,4), and IP₆ (sodium phytate) from rice, perchloric acid (60%, vol/vol), myoinositol hexasulfate hexapotassium salt (MIHS) and sodium hexametaphosphate (Sigma-Aldrich, St Louis, MO), and myo-inositol tris-, tetrakis-, and pentakisphosphates [IP₃(1,4,5); IP₄ (1,2,4,5), IP₅(1,2,4,5,6), IP₅(1,2,3,4,5), and IP₅(1,3,4,5,6); Sichem, Bremen, Germany] were obtained from commercial companies. One pepsin unit is defi ned as the increase of ΔA280 by 0.001 absorbance unit per min at pH2.0 at 37°C, measured as TCA-soluble products using hemoglobin as substrate. The amounts of these IP₃, IP₄, and IP₅ esters in 0.1 and 1 mg portions were used without weighing because, according to the manufacturer, the weight deviation was less than 5%. Dyed casein in the form of tablets [Protazyme AK containing azurine crosslinked (AZCL) casein; Megazyme, Wicklow, Ireland] was used for pepsin assay. Casein with fl uorescence labelling (BODIPY FL casein, E6638) was obtained from a commercial company (Invitrogen, Carlsbad, CA). The bacterial phytase [i.e., the E. coli K12 phytase (EC 3.1.3.26) AppA gene expressed in Schizosaccharomyces pombe (Phyzyme XP) and the subtilisin variant (EC 3.4.21.62; P3000, Protex 30L; Danisco A/S, Brabrand, Denmark)], the fungal phytase [i.e., the Aspergillus fi ccum phytase expressed in A. niger (EC 3.1.3.8; Natuphos; BASF, Ludwigshafen, Germany)], and soy [Glycine max (L.) Merr.] protein isolate (BodyLab Nutrition, Hadsund, Denmark) were obtained from commercial companies. Microplates of different formats (Nunc A/S, Roskilde, Denmark; Corning Inc., Corning, NY) and the corn (Zea mays L.) and soy based feed (Danish Technological Institute, Kolding, Denmark; Supplemental Table 1) were obtained accordingly. The feed was dosed with different amounts of E. coli phytase. The phytase found in the feed was 184, 223, 442, 1,129, and 2,301 phytase unit (FTU) U≅ kg−1. Phytase activity was determined according standard phytase assay method (Engelen et al., 2001). That is, commercial phytase or phytase feed extract was reacted with phytic acid at 37°C and pH 5.5 for 60 min. The reaction was stopped by adding acid vanadate-molybdate reagent, which produces colored complex with Pi produced. The optical density (OD) of the yellow colored complex was monitored at 415 nm. Commercial feed samples with unknown composition were obtained (Danisco A/S). 

Soluble soy protein fraction was prepared from soy protein isolate by the following procedure: soy protein isolate (0.5 g) was added to 25 mL 0.25 M GlyU≅ HCl with pH 2.5 (Buffer G), pH was adjusted to 2.5 or 3.0 with NaOH, centrifuged at 4,500 × g at 25°C for 20 min, and fi ltered through a glass fi ber filter (Whatman GF/B; Maidstone, Kent, UK). Protein concentration in the filtrate was determined by a dye binding method using BSA as standard according to the protocol from the manufacturer (Bio-Rad Laboratories, Hercules, CA). For assaying aggregation of myo-inositol phosphate ester (IP_x, where x denotes the number of phosphate ester bonds) with soy protein, the reaction mixture contained 5 μL of 1 of the IP_x (IP_1-6) at 3 mM in water, 100 μL soy protein, and Buffer G or 0.1 M sodium acetate to a fi nal volume of 120 μL. The final concentration of IP_x was 125 μM and the fi nal soy protein concentration was 1.5 mg mL−1. In certain experiments, this system also contained NaCl (0.3 M), and the pH of the system varied between 1.6 and 4.5, buffered by GlyU≅ HCl or sodium acetate. For testing IP₆ hydrolyzate formed by E. coli and A. niger phytase, 5 μL of IP_x were replaced with 10 μL hydrolyzate. After mixing the contents, aggregation in the form of turbidity was determined immediately at 25°C by following the increase in OD at 600 nm using a scanning microplate spectrophotometer (PowerWavex; Bio-TEK Instruments, Inc., Winooski, VT). For testing aggregation of β-casein with chemically pure IP_1-6, the experimental conditions were the same as for soy protein, and the fi nal concentration of β-casein was 1.5 mg mL−1. For testing β-casein aggregation with IP₆ hydrolyzate, the reaction mixture contained 2 μL hydrolyzate, 80 μL β-casein in Buffer G, and water to a fi nal volume of 96 μL. Final β-casein concentration was 1.5 mg mL−1 and fi nal IP₆ hydrolyzate concentration at time 0 min was 107 μM of IP₆. Optical density at 600 nm was read in a 384 well-microplate using a microplate reader (SpectraMax M5; Molecular Devices Corp., Sunnyvale, CA). 

For assaying phytase from feed extract in a petri-dish plate assay, water was added to 2.5 to 12.5 g feed to a fi nal volume of 50 mL and mixed until the feed pellet became disintegrated, which took usually 6 to 10 min. The supernatant obtained after standing still for 1 to 2 min was applied directly to the wells (20 μL) in an agarose plate containing IP₆-soy protein complex at pH 1.6 to 3.5, buffered by using 0.1 to 0.25 M GlyU≅ HCl, or at pH 3.5 to 4.5, buffered by using 0.1 M acetate. The agarose gel, which contained IP₆ (0.1 to 0.4 mM) or the phytase inhibitors of MIHS (0.1 mM) and NaF (10 mM), soy protein (1.5 mg mL−1), and agarose [1.5% (wt/vol)] was prepared by melting agarose containing IP₆ or the inhibitor or both in a water bath at 100°C or in a microwave oven. The melted agarose solution was mixed with soy protein solution and poured directly into a 9-cm petri dish and gelled at room temperature (23°C). After gelling, wells with a volume of approximately 20 μL were made in the gel manually using pipette tips, and numbered. The development of clear halos on the agarose gel with phytase extract was followed by incubating the plate at temperatures between 15 and 35°C. In control, phytase extract was replaced with extract made from the same feed but containing no added microbial phytase. For standards, feed extracts with the same feed samples to be tested and containing known amount of phytase were used. For negative control, 20 μL phytase extract was added to the agarose gel containing 0.1 mM of MIHS or 10 mM of NaF. The area of clear halos was measured in square centimeters for semi-quantifi cation of phytase activity. 

Ferric sulfosalicylic acid solution (also called Wade's solution) was made as described earlier (Rounds and Nielsen, 1993; Kwanyuen and Burton, 2005). It consisted of 0.015% (wt/vol) of FeCl3 · 6H2O and 0.15% (wt/vol) of sulfosalicylic acid. The myo-inositol ester (IP_1-6), 2.5 μL at 3 mM, was added to 75 μL of 10 mM methyl piperazine (pH 4.0) and 80 μL Fe³⁺-sulfosalicylic acid solution and mixed. The mixture had a fi nal pH of 2.4. The fi nal volume was 157.5 μL, and the fi nal IP_x concentration was 47.6 μM. Decrease in OD at 500 nm as a result of Fe³⁺ interaction with IP_1-6 was read spectrophotometrically 2 min after mixing at 23°C against a blank containing no IP_x. A standard curve was made by mixing IP₆ standards (1.5 to 21 nmol) with 75 μL Buffer G and 80 μL Fe³⁺- sulfosalicylic acid solution to ensure that OD data obtained were in the linear range.

Ferric perchloric acid solution was prepared as described earlier (Phillippy and Bland, 1988; Skoglund et al., 1998). The myo-inositol ester (IP_1-6), 1.5 μL at 3 mM, was added to 75 μL Buffer G and 40 μL Fe³⁺-perchloric acid solution containing 0.1% (wt/vol) Fe(NO₃)3 9H₂O and 2% (vol/vol) HClO₄. The fi nal volume was 116.5 μL, and the fi nal concentration of IP_x was 38.6 μM. Increase in OD at 290 nm, as a result of the interaction between Fe³⁺ and IP_x, was read spectrophotometrically 2 min after mixing at 23°C. A standard curve was made by mixing IP₆ standards (1.5 to 28 nmol) with 75 μL Buffer G and 40 μL Fe³⁺-perchloric acid solution to ensure that OD data obtained were in the linear range. 

The dry weight loss of sodium phytate [P3168, Lot 123k1413, purity of 95% (wt/wt); Sigma-Aldrich] was measured to be 14.74%, with an environmental humidity of 37%. It was determined by increasing the temperature from 23 to 160°C until a constant weight was reached using a HB43 halogen moisture analyzer (Mettler Toledo; Greifensee, Switzerland). The reaction mixture for measuring IP₆ hydrolysis consisted of 30 mL of 8.526 mM of sodium phytate, 4 FTU of the E. coli phytase, or 10 FTU of the A. niger phytase in Buffer G and Buffer G to a fi nal volume of 50 mL. One phytase unit was defi ned as the activity of phytase that generates 1 μmol of Pi min−1 from 5.1 mM of sodium phytate at pH 5.5 in 0.25 M sodium acetate at 37°C. Complete hydrolysis of IP₆ by the bacterial and fungal phytase would generate stoichiometrically 5.116 mM myo-inositol monophosphate (IP₁). The hydrolysis reactions were started by adding phytase to the rest of the reactants that had been incubated at 37°C for 10 min. The reaction was continued for 23 h and samples of 5 mL were taken at 0, 5, 10, 15, 20, 30, 40, and 50 min and 1, 1.5, 2, 3, and 23 h, heated in a boiling water bath for 5 min, cooled, and saved at −20°C before analyzing for Pi (Engelen et al., 2001) and IP_x by high performance ion exchange chromatography (HPIC; Skoglund et al., 1997). The hydrolysis samples (i.e., IP₆ hydrolyzates) collected were also tested for their effects on protein aggregation and inhibition of pepsin-catalyzed casein hydrolysis. 

The separation and detection of IP_1-6 were achieved by HPIC as described earlier (Skoglund et al., 1997, 1998) using the column (4 × 250 mm with a 4 × 50 mm pre-column; CarboPac Guard PA-100; Dionex, Sunnyvale, CA). MilliQ water and 1 N HCl prepared in water were used as Solvents A and B, respectively. The gradient program was: 2.5 to 50% (vol/vol) Solvent B in 30 min followed by isocratic at 50% (vol/vol) Solvent B for 3 min and isocratic at 2.5% (vol/vol) Solvent B for 2 min. The fl ow rate was 0.8 mL min−1. Phytic acid hydrolyzate sample of 0.5 mL in Buffer G was fi ltered and 20 μL or 40 μL with or without added internal IP₂-6 standards at 1.5 mM were injected into the column. The myo-inositol phosphate esters IP_1-6 standards (Sichem and Sigma-Aldrich) prepared in Buffer G were also analyzed under the same conditions with an injection volume of 40 μL. The eluant from the PA-100 column was reacted in-line in a T-tube with Fe³⁺ perchloric acid solution at a fl ow rate 0.4 mL min−1. Phytic acid IP₆ and IP_1-5 isomers were detected at 290 nm as positive peaks resulting from the formation of IP₆-Fe³⁺-ClO₄ − complex. Peaks were identifi ed by their retention time compared with standards and by their relative retention times compared with IP₆ for peaks that lack standards (Skoglund et al., 1998). Quantifi cation of IP_5-6 was performed by the comparison of the peak area with the respective IP_5-6 standards (Sichem). The linear range for IP₆ was 10 to 250 μg in an injection volume of 40 μL. 

In a typical chromophore labeled casein assay, 1 AZCLcasein tablet, 1 mini-magnetic bar, 1 mL Buffer G, and 0.5 mL IP₆ hydrolyzate or 0.5 mL Buffer G (positive control) were added to a test tube. After pre-incubation at 40°C for 5 min for the hydration of the AZCL-casein, 0.5 mL pepsin solution (12.5 μg mL−1; Sigma P7000; Sigma-Aldrich) in Buffer G, or 0.5 mL Buffer G (blank) were added to start the reaction. The reaction was allowed to proceed at 40°C for 60 min and was stopped by adding 10 mL 2% (wt/vol) of trisodium phosphate. After centrifugation at 3,000 × g for 10 min at 23°C, the supernatant was fi ltered and OD of the fi ltrate was measured at 590 nm against the blank.

In a typical fl uorophore labeled casein assay, the reaction mixture contained casein labeled with the fl uorescence dye BODIPY FL (Invitrogen) at 5 μg mL−1 in 10 mM HCl, 0.125 units of porcine pepsin (Sigma P7012; Sigma-Aldrich), and increasing amounts of IP₆ (0 to 1.5 mg mL−1) in a fi nal volume of 160 μL. The reaction was performed at 30°C. Fluorescence was monitored by using a microplate reader (SpectraMax M5; Molecular Devices Corp.) with excitation set at 485 nm and emission set at 524 nm. Reading sensitivity was set at 6, and 6 readings were recorded on average at 1-min intervals. The microplate was shaken before each reading. A half-area 96-well ultraviolet microplate was used for top reading.

A detailed experiment design fl ow chart is given in Figure 2 to illustrate the plans and goals set for the current study. A computer program (Excel 2007; Microsoft Danmark ApS, Hellerup, Denmark) was used for data processing and preparation of the fi gures. Mean ± SD are presented in all fi gures. Statistical comparisons for differences between IP₅(1,2,3,4,5) and IP₅(1,2,4,5,6) isomers in their aggregation of soy protein and β-casein were performed by Student t-test.

Figure 2. Experimental design fl ow chart.

Soy protein and bovine β-casein were studied for their aggregate formation with IP_1-5 esters compared with IP₆. Figure 3 shows the aggregation of soy protein as a function of IP₆ concentration. Under the assay conditions, the aggregation reached its maximum, as observed from the turbidity (OD at 600 nm) at an IP₆ concentration of about 0.08 mg mL−1, and further increasing the IP₆ concentration did not lead to increased turbidity. The amount of protein remaining in solution after centrifugation of the reaction mixture decreased rapidly at increased IP₆ concentration up to 0.1 mg mL−1. It is noted, however, that even at 0.33 mg mL−1 of IP₆, there were still 29.4 ± 2.4% soy protein in solution. Interaction and formation of aggregate were also observed between IP₆ and β-casein. In contrast to soy protein, β-casein was more liable in forming aggregate at the same IP₆ concentration. For instance, to reach a turbidity of 1.0, the IP₆ concentration needed for β-casein was 1.80 times less than that for soy protein under the same conditions. The IP₆ concentration needed to aggregate these 2 proteins is relevant in vivo for food and feed, which can have 0.5 mg IP₆ mL−1 slurry, assuming that food and feed have a reduced concentration of IP₆ of 0.2% (wt/wt; 2 mg mL−1), and, in the digestive tract, they are diluted by a factor of 4. Several factors, including pH and salts, affected the interaction between IP₆ and protein. For example, in the case of soy protein, aggregates were formed in a pH range of 2.5 to 3.8 in GlyU≅  HCl buffer, whereas at pH lower than 2.0, the solution was clear (OD at 600 nm ≤ 0.05). This is due to the fact that IP₆ has 6 dissociable protons with a pKa of approximately 2 (Reddy et al., 1989). Salts like CaCl2 (Murthy and Rao, 1984) and NaCl affected IP₆-protein interaction negatively. For instance, the turbidity of IP₆-soy complex was abolished in the presence of 0.3 M NaCl (OD ≤ 0.05). Beside IP₆ that forms aggregates with positive charged proteins, other poly-anions such as MIHS, the analogue of IP₆, and hexametaphosphate also formed aggregates with soy protein at 0.1 mM (data not shown).

By using the IP₆-protein aggregate system as control, the aggregation capabilities of IP_1-4 and IP₅ isomers with defi ned structures with the 2 proteins were examined. It was observed that the turbidity of the system decreased radically with a decrease in phosphorylation from IP₆ to IP₃ (Figure 4). The 2 IP₅ isomers, IP₅(1,3,4,5,6) and IP₅ (1,2,4,5,6), showed a 4.6 ± 0.1 fold decrease in aggregating soy protein compared with IP₆. myo-Inositol pentakis-phosphate IP₅(1,2,3,4,5), an intermediate product of E. coli 6-phytase, showed the least aggregation degree (6.6 ± 0.5-fold turbidity reduction) compared with IP₅(1,3,4,5,6) and IP₅ (1,2,4,5,6) generated by A. niger 3-phytase (P = 0.001). It was noted that, under the same conditions, the ability of IP₄(1,2,4,5), IP₃(1,4,5), IP₂ (2,4), and IP₁(2) to aggregate soy protein was negligible, as the OD of these mixtures was ≤0.007 ± 0.002. As with soy protein, a similar interaction pattern with IP_1-4 and IP₅ isomers, compared with IP₆, was also observed with β-casein (data not shown). These observations were in contrast to our assumption that phosphorylation degrees of myo-inositol (i.e., IP_1-6) were directly proportional to their protein aggregation capabilities.

Figure 3. Aggregation of soy protein as a function of phytic acid (IP₆) concentrations. The assay mixture contained 0 to 20 μL IP₆ (1 mg mL−1), 0 to 20 μL 0.25 M GlyU≅ HCl (pH 2.5, Buffer G), and 100 μL soy protein made in Buffer G with a fi nal protein concentration of 0.6 mg mL−1. Turbidity [optical density (OD) 600 nm] was measured at 23°C, and each data point is the average of 4 measurements with SD

Figure 4. The aggregation of soy protein by myo-inositol phosphate esters (IP_1-6) and IP₅ positional isomers. The assay contained 5 μL IP_1-6 (3 mM), Buffer G, and 100 μL soy protein (1.8 mg mL−1). Turbidity [optical density (OD) 600 nm] was measured at 23°C after a 1 min mixing, and each data point is the average of 4 measurements with SD.

 

Under in vivo conditions, IP_1-6 occurs as a mixture of various esters and isomers because of the action of phytases of plant, microbial, or mucosal origin (Greiner and Konietzny, 2010). It was, therefore, essential to examine various IP_1-6 mixtures generated by commercial microbial phytases for their aggregation of dietary proteins and inhibition of pepsin-catalyzed protein hydrolysis. The presence of IP₆ at 0.911 mg mL−1 resulted in 86.9 ± 0.5% loss in pepsin activity (Figure 5). The IP₆ concentration used (0.9 mg mL−1) is relevant in vivo in feed slurries as discussed before. Phytic acid inhibited protein hydrolysis by pepsin was also observed with fl uorophore-labeled casein as substrate (BODIPY FL casein), pepsin activity being lost by 24.7 ± 9.4% at IP₆ concentration of 0.027 mg mL−1. The inhibition of pepsin catalyzed AZCL-casein hydrolysis by IP₆ was reversed by the addition of either E. coli phytase (Figure 6) or A. niger phytase (data not shown).

The hydrolysis of IP₆ by phytases results in the formation of inositol pentakis-, tetrakis-, tris-, bis-, and monophosphate in a stepwise manner (Wyss et al., 1999; Greiner et al., 2000). Commercial phytases are bacterial and fungal origin, which produce myo-inositol monophosphate IP₁ and Pi as end products after thorough hydrolysis (Greiner, 2007; Selle and Ravindran, 2007). Figure 6 shows the time course for the degradation of IP₆ in the presence of 0.08 FTU mL−1 of the E. coli phytase at pH 2.5. At 3 h after incubation, IP₁ and IP₂ could be detected, while at 23 h (data not shown), only IP₁ was detected.

From Figure 6, it can be observed that a mixture of Pi, IP₁, and IP₂ produced at 3 h gave no longer inhibition on pepsin activity (106.6 ± 1.1 vs. 100.0% in the absence of IP₆ hydrolyzate) and the sample collected at 23 h containing solely Pi and IP₁ (pepsin activity being 108.2 ± 2.9%). The reaction mixtures containing other IP esters including IP₃, IP₄, IP₅, and IP₆ all inhibited pepsin catalyzed protein hydrolysis to various degrees. For instance, at 2 h with a mixture containing IP₃ along with Pi, IP₁, and IP₂, the pepsin activity was reduced to 64.8 ± 9.6% (Figure 6). Decreased pepsin inhibition with the progress of IP₆ hydrolysis was also observed with A. niger phytase (data not shown). These results, showing that IP₃-6 (and not IP₁-2) still inhibited pepsin activity (Figure 6), are coincidently in agreement with those observed earlier using IP_1-5 preparations prepared chemically with unknown structures (Knuckles et al., 1989).

Figure 5. Phytic acid inhibition of porcine pepsin-catalyzed hydrolysis of azurine cross-linked casein. The assay was performed at 40°C. Each data point is the average of duplicate

 

Based on the observations that when IP₆ was converted to IP₅ and IP_1-4, the capability to interact with proteins decreased (Figure 4), we set out to examine if such a radical change would also be observed in their chelating with Fe³⁺ using the same IP₁−IP₆ and IP₅ isomer preparations. The Fe³⁺ is known to form a complex with IP₆, which is indigestible by phytase (Tang et al., 2006). From Figure 7, one can see that there was no dramatic changes in chelating power from IP₆ to IP₅, expressed either as the Fe³⁺ bound by IP₆ or the OD decrease caused by the formation of Fe³⁺- IP₆ complex, which had a molar ratio of 2 to 1 under the assay conditions. The 3 IP₅ isomers, IP₅(1,3,4,5,6), IP₅(1,2,4,5,6), and IP₅(1,2,3,4,5), had 76.0 ± 8.2, 70.4 ± 9.6, and 73.0 ± 5.7% Fe³⁺ reactivity relative to IP₆, respectively. The IP₄(1,2,4,5) and IP₃(1,4,5) had 35.0 ± 6.7 and 29.5 ± 5.7% reactivity, respectively, while IP₂ (1,4) and IP₁(2) had no detectable reactivity (Figure 7). A standard curve between the OD decrease at 500 nm and IP₆ concentration (1.5 to 21 nmol) gave good linearity (R² = 0.997; data not shown), indicating the direct correlation between the OD and the Fe³⁺ bound. The OD value obtained with 7.5 nmol of IP₆ used in Figure 7, with which all IP_1-5 were compared, was in the linear range. These data indicate that, in sulfosalicylic acid solution at pH 2.4, the binding of Fe³⁺ with inositol phosphate esters decreased with a decrease in the degree of inositol phosphorylation, and IP₃ was still capable of binding Fe³⁺ at 29.5 ± 5.7% strength of IP₆. The observation shown in Figure 7 is in line with the results obtained with IP₃-6 using human intestinal cell lines (Han et al., 1994) and in human studies (Sandberg et al., 1999), which showed that inositol phosphate esters with a phosphorylation degree as low as IP₃ still contributed to the poor digestion of iron.

Figure 6. Time course of phytic acid hydrolysis by E. coli phytase (Phyzyme XP, Danisco A/S, Brabrand, Denmark; 0.08 phytase unitU≅ mL−1) and inhibition of porcine pepsin catalyzed azurine cross-linked casein hydrolysis by the hydrolyzates. Phytic acid hydrolysis was performed at 37°C; pepsin activity assay was carried out at 40°C. Each data point is an average of 2 separate experiments.

The interaction of IP_1-6 with Fe³⁺ complexed with perchloric acid was further examined at pH 2.0 to confi rm the aforementioned results when Fe³⁺ was complexed with sulfosalicylic acid at pH 2.4. Just as the Fe³⁺-sulfosalicylic acid reagent, the reagent of Fe³⁺ complexed with perchloric acid is commonly used inline in the detection of IP₂-IP₆ after chromatographic separations (Skoglund et al., 1998). When reacted with Fe³⁺-perchloric acid solution at pH 2.0, the 3 IP₅ isomers IP₅(1,3,4,5,6), IP₅(1,2,4,5,6), and IP₅(1,2,3,4,5) had 66.4 ± 5.2, 65.7 ± 6.7, and 58.4 ± 3.8%, reactivity relative to IP₆, respectively. The IP₄, IP₃, IP₂, and IP₁ had 24.1 ± 4.2, 16.2 ± 3.5, 9.2 ± 2.1, and 4.6 ± 2.4% reactivity of IP₆, respectively. A standard curve of OD and IP₆ concentration (1.5 to 28 nmol) was made, which showed a linear response (R2 = 0.997; data not shown).

Ferric ion was chosen in the current study for its interaction with IP₁−IP₆ instead of Ca2+ and Zn2+ to address the dual issue of iron nutrition and IP₂-6 analysis. The Fe³⁺ reagents have been traditionally used for the quantifi cation of IP₆ by weighing the ferric phytate precipitate formed (Kikunaga et al., 1985). Two Fe³⁺ reagents are currently used for the quantifi cation of IP₂-6 in HPIC (Rounds and Nielsen, 1993; Skoglund et al., 1998, Oberleas and Harland, 2007). The results from our 2 studies described indicate that errors in quantifi cation of IP_1-5 could occur if one assumes that the relative reactivity of IP₁−IP₅ to IP₆ to be 83.3 (5/6), 66.7 (4/6), 50 (3/6), 33.3 (2/6), and 16.7% (1/6) of IP₆, respectively, which was assumed by investigators in quantifying IP₁−IP₅ (Skoglund et al., 1997).

Figure 7. Interaction of myo-inositol phosphate esters (IP₁−IP₆) and IP₅ positional isomers with Fe³⁺. The ferric reagent used was Fe³⁺ sulfosalicylic acid solution. The reaction mixture had a fi nal pH of 2.4. Each data point measured at 23°C is the average of 4 measurements with SD.

 

Based on the observation that the turbidity of IP₆- protein complex decreased radically when IP₆ was converted to IP₅ (Figure 4), and the correlation between the phytase dose and the time needed to clarify the IP₆-soy protein solution (R2 = 0.984; data not shown), a phytase plate assay was designed. The plate assay was based on the incorporation of 1.5% (wt/vol) agarose to the IP₆- protein system such that an agarose gel plate was formed with a whitish background because of the complex formation between IP₆ and soy protein. Addition of phytase to the wells in the gel created a clear halo around the wells after incubation (Figure 8; Supplemental Figure 1). The transparent halo was usually observed in 10 to 20 min at 23°C for a phytase dose of 1,000 to 2,000 FTU kg−1 feed. For a dose of 250 to 800 FTU kg−1, clear halos were observed at 23°C in 3 to 7 h or overnight. The incubation temperature was chosen between 20 and 35°C. Feed sample with no added phytase was used as negative control. Agarose gels containing phytase inhibitors (10 mM NaF or 0.1 mM MIHS) were also found to be good controls as addition of phytase did not result in the development of halos in gels (data not shown).

It can be seen from Figure 8, the detection limit for phytase can be down to 193 FTUU≅ kg−1 feed for an incubation time of 7 and 20 h at 23°C, a dosage usually used for laying hen diets. The relationship between phytase dosage used (223 to 2,301 FTUU≅ kg−1 feed) and the halo area developed can also be expressed mathematically (R2 = 1.0, Figure 8). Factors that may affect the phytase plate assay include buffers, buffer pH, IP₆ concentrations, ratio of feed to extraction solution, and extraction time. The optimal conditions for the phytase plate assay established in this study were: 0.25 M of Gly or acetate buffer at pH 3.0 to 3.5, IP₆ at 0.10 to 0.15 mM, water to feed ratio of 4:1, and an extraction time of 10 to 30 min at 23°C in the case of E. coli phytase (Phyzyme XP; Danisco A/S).

The assay method for feed phytase is designed to be both qualitative and semi-quantitative. The CV for intraassay variation was up to 16% when tested for the phytase dose range of 233 to 2,301 FTUU≅ kg−1 (Figure 8). The CV for inter-assay variation was up to 13% when tested for the dose of 442 FTUU≅ kg−1 at different days.

From Figure 8, it can be seen that there was a good relationship between the phytase plate assay method and the standard method based on Pi release (Engelen et al., 2001). However, for certain feed samples, such relationship was not always clear. For example, among 9 feed samples collected from different feed mills, 2 of them, which contained 276 and 433 FTUU≅ kg−1 by the standard method showed no halo development. This could be due to the interference to the standard method caused by endogenous Pi (Kim and Lei, 2005). It should be noted the lack of halo development for these 2 samples was not due to possible phytase inhibitors in the feed, as all 9 feed samples gave good response to the externally added phytases at 500 and 1,000 FTU kg−1 (data not shown). 

There are numerous reports describing the interactions of IP₆ with proteins and iron in the last several decades, but few of them have examined the interactions between proteins and defi ned IP₆ degradation products. The present in vitro study, using pure IP_1-4, IP₅ isomers, and IP₆ hydrolyzate containing IP_1-5 generated by commercial bacterial and fungal 6- and 3-phytases, indicates for the fi rst time that the solubility of soy protein and β-casein increased radically when IP₆ was converted to IP₅ and further to IP₄. Based on these observations, a novel, simple, yet phytasespecifi c, qualitative and semi-quantitative assay using agarose- petri dishes suitable for microbial phytase screening in phytase protein engineering, as well as for use in feed laboratories and at animal farms for controlling phytase activity, was developed. This method is complementary to the tedious, lengthy, yet quantitative phytase standard method based on the assay of Pi released. The myo-inositol phosphate esters IP₁-3, which had no infl uence on protein solubility, still interacted with Fe³⁺, and IP₃ inhibited pepsin catalyzed protein hydrolysis, though to a much lesser extent than IP₆. These studies indicate that an early and thorough hydrolysis of phytate by phytase in the upper digestive tract is essential for an improved phosphorus, minerals (e.g., Fe³⁺), and protein digestion.

Figure 8. The plate assay of phytase and its correlation to the standard phytase assay method based on inorganic P (Pi) release. The x-axis, the amount of E. coli phytase assayed by the standard method based on Pi release; y-axis, halo area, which was measured after incubation for 7 (?) and 20 h (?) at 23°C with the phytase feed extract. Each data point is the average of 4 assays with SD. Standard deviation bars are shown when they are larger than the symbols. FTU = phytase unit.

 

Supplemental Figure 1. Clear halo development as a function of phytase dosage applied and incubation time in the agarose-petri-dishes. Feed water extracts were made by mixing 5 g corn soy based feed (Supplemental Table 1) with water to a final volume of 50 mL. The supernatant of the extract 20μl obtained 10 min after mixing was applied to the wells and incubated at 23EC. Halo diameters were measured and photos were taken at 7 h (1a) and 20 h (1b, n = 4). The wells marked with 0, 193, 233, 442, 1,129 and 2,301 indicate the amount of E. coli phytase (Phyzyme XP, Danisco A/S, Brabrand, Denmark) in phytase unit (FTU)Ξkg^-1 feed assayed by standard phytase assaying method using vanadate-molybdate reagent for Pi determination (Engelen et al., 2001).

Supplemental Figure 1a, Ms ID E-2011-3866

Supplemental Figure 1b, Ms ID E-2011-3866

 

Agranoff, B. W. 2009. Turtles all the way: Refl ections on myo-inositol. J. Biol. Chem. 284:21121−21126.

Bohn, L., L. Josefsen, A. S. Meyer, and S. K. Rasmussen. 2007. Quantitative analysis of phytate globoids isolated from wheat bran and characterization of their sequential dephosphorylation by wheat phytase. J. Agric. Food Chem. 55:7547−7552.

Engelen, A. J., F. C. van der Heeft, P. H. Randsdorp, and W. A. Somers. 2001. Determination of phytase activity in feed by a colorimentric enztmatic method: Collaboratory srudy. J. AOAC Int. 84:629−633.

Greiner, R. 2007. Phytate-degrading enzymes: regulation of synthesis in microorganisms and plants. Page 78−97 in Inositol Phosphates: Linking Agriculture and the Environment. Turner, B. L., A. E. Richardson, and E. Mullaney, ed. CAB International, Wallingford, Oxfordshire, UK.

Greiner, R., N.-G. Carlsson, and M. L. Alminger. 2000. Stereospecifi city of myo-inositol hexakisphosphate dephosphorylation by a phytatedegrading enzyme of Escherichia coli. J. Biotechnol. 84:53−62.

Greiner, R., and U. Konietzny. 2010. Phytases: Biochemistry, enzymology, and characteristics relevant to animal feed use. Page 96−128 in Enzymes in Farm Animal Nutrition. 2nd ed. M. Bedford and G. Patridge, ed. CAB International, Wallingford, UK.

Han, O., M. L. Failla, A. D. Hill, E. R. Morris, and J. C. Smith, Jr. 1994. Inositol phosphates inhibit uptake and transport of iron and zinc by a human intestinal cell line. J. Nutr. 124:580−587.

Kies, A. K., L. H. De Jonge, P. A. Kemme, and A. W. Jongbloed. 2006. Interaction between protein, phytate, and microbial phytase. In vitro studies. J. Agric. Food Chem. 54:1753−1758.

Kikunaga, S., M. Takahashi, and H. Huzisige. 1985. Accurate and simple measurement of phytic acid contents in cereal grains. Plant Cell Physiol. 26:1323−1330.

Kim, T. W., and X. G. Lei. 2005. An improved method for a rapid determination of phytase activity in animal feed. J. Anim. Sci. 83:1062–10671.

Knuckles, B. E., D. D. Kuzmicky, M. R. Gumbmann, and A. A. Betschart. 1989. Effect of myo-inositol phosphate esters on in vitro and in vivo digestion of protein. J. Food Sci. 54:1348−1350.

Kwanyuen, P., and J. W. Burton. 2005. A simple and rapid procedure for phytate determination in soys and soy products. J. AOCS. 80:81− 85.

Murthy, N. V. K. K., and M. S. N. Rao. 1984. Interaction of phytate with mustard 12S protein. J. Agric. Food Chem. 32:493−498.

Oberleas, D. and B. Harland. 2007. Validation of a column liquid chromatographic method for phytate. J. AOAC Int. 90:1635−1638.

Phillippy, B. Q., and J. M. Bland. 1988. Gradient ion chromatography of inositol phosphates. Anal. Biochem. 175:162−166.

Rajendran, S. and V. Prakash. 1993. Interaction of myo-inositol hexaphosphate (MIHP) with beta-globulin from Sesamum indicum L. Kinetics and thermodynamics. Int. J. Pept. Protein Res. 42:78− 83.

Reddy, N. R. 2002. Occurrence, distribution, content, and dietary intake of phytate. Page 25−51 in Food Phytates. N. R. Reddy and S. K. Sathe, ed. CRC Press, Boca Raton, FL. Reddy, N. R., M. D. Pierson, S. K. Sathe, and D. K. Salunkhe. 1989. Interactions of phytate with proteins and minerals. Page 57− 70 in Phytates in Cereals and Legumes. N. R. Reddy et al., ed. CRC Press. Boca Raton, FL.

Rounds, M. A., and S. S. Nielsen. 1993. Anion-exchange high performance liquid chromatography with post-column detection for the analysis of phytic acid and other inositol phosphates. J. Chromatogr. A 653:148−152.

Sandberg, A.-S., M. Brune, N.-G. Carlsson, L. Hallberg, E. Skoglund, and L. Rossander-Hulthén. 1999. Inositol phosphates with different numbers of phosphate groups infl uence iron absorption in humans. Am. J. Clin. Nutr. 70:240−246.

Selle, P. H., and V. Ravindran. 2007. Microbial phytase in poultry nutrition. An. feed Sci. Technol. 135:1−41.

Skoglund, E., N.-G. Carlsson, and A.-S. Sandberg. 1997. Determination of isomers of inositol mono- to hexaphosphates in selected foods and intestinal contents using high-performance ion chromatography. J. Agric. Food Chem. 45:431–436.

Skoglund, E., N.-G. Carlsson, and A.-S. Sandberg. 1998. Highperformance chromatographic separation of inositol phosphate isomers on strong anion exchange columns. J. Agric. Food Chem. 46:1877−1882.

Tang, J., Leung, A., Leung, C., and B. L. Lim. 2006. Hydrolysis of precipitated phytate by three distinct families of phytases. Soil Biol. Biochem. 38:1316−1324.

Weaver, C. M., and S. Kannan. 2002. Phytate and mineral availability. Page 211−223 in Food Phytates. N. R. Reddy et al., ed. CRC Press. Boca Raton, FL.

Wyss, M., R. Brugger, A. Kronenberger, R. Rémy, R. Fimbel, G. Oesterhelt, M. Lehmann, and A. P. G. M. van Loon. 1999. Biochemical characterization of fungal phytases (myo-inositol hexakisphosphate phosphohydrolase): Catalytic properties. Appl. Environ. Microbiol. 65:367−373.