- Email: [email protected]
Alkaline phytase from lily pollen: Investigation of biochemical properties
Archives of Biochemistry and Biophysics 440 (2005) 133–140 www.elsevier.com/locate/yabbi
Alkaline phytase from lily pollen: Investigation of biochemical properties Sonali P. Jog, Barry G. Garchow, Bakul Dhagat Mehta, Pushpalatha P.N. Murthy ¤ Department of Chemistry, Michigan Technological University, Houghton, MI 49931, USA Received 28 March 2005, and in revised form 23 May 2005 Available online 11 July 2005
Abstract Phytases catalyze the hydrolysis of phytic acid (InsP6, myo-inositol hexakisphosphate), the most abundant inositol phosphate in cells. In cereal grains and legumes, it constitutes 3–5% of the dry weight of seeds. The inability of humans and monogastric animals such as swine and poultry to absorb complexed InsP6 has led to nutritional and environmental problems. The eYcacy of supplemental phytases to address these issues is well established; thus, there is a need for phytases with a range of biochemical and biophysical properties for numerous applications. An alkaline phytase that shows unique catalytic properties was isolated from plant tissues. In this paper, we report on the biochemical properties of an alkaline phytase from pollen grains of Lilium longiXorum. The enzyme exhibits narrow substrate speciWcity, it hydrolyzed InsP6 and para-nitrophenyl phosphate (pNPP). Alkaline phytase followed Michaelis–Menten kinetics with a Km of 81 M and Vmax of 217 nmol Pi/min/mg with InsP6 and a Km of 372 M and Vmax of 1272 nmol Pi/min/mg with pNPP. The pH optimum was 8.0 with InsP6 as the substrate and 7.0 with pNPP. Alkaline phytase was activated by calcium and inactivated by ethylenediaminetetraacetic acid; however, the enzyme retained a low level of activity even in Ca2+-free medium. Fluoride as well as myo-inositol hexasulfate did not have any inhibitory aVect, whereas vanadate inhibited the enzyme. The enzyme was activated by sodium chloride and potassium chloride and inactivated by magnesium chloride; the activation by salts followed the Hofmeister series. The temperature optimum for hydrolysis is 55 °C; the enzyme was stable at 55 °C for about 30 min. The enzyme has unique properties that suggest the potential to be useful as a feed supplement. 2005 Elsevier Inc. All rights reserved. Keywords: Phytate; Alkaline phytase; Pollen grains; Lilium longiXorum; Enzyme activation; Hofmeister series; Phosphate contamination; Animal feed
Phytases, a class of phosphatases, are the primary enzymes responsible for the hydrolysis of phytic acid [1– 6]. They catalyze the sequential hydrolysis of phytic acid to less phosphorylated inositol phosphates and, in some cases, to inositol [7]. A number of phytases with varying structural and catalytic properties have been found in plants, yeast, and bacteria [2–4]. Phytases have been classiWed on the basis of pH optima (acid and alkaline), catalytic mechanisms (histidine acid phosphatase-like phytase, purple acid phos-
*
Corresponding author. Fax: +1 906 487 2061. E-mail address: [email protected] (P.P.N. Murthy).
0003-9861/$ - see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2005.05.029
phatase-like phytase, and -propeller phytase), and speciWcity of hydrolysis (3-phytase, EC.3.1.3.8, 6-phytase, EC.3.1.3.26, and more recently 5-phytases, EC.3.1.3.72) [1–4]. Although acid phytases have been extensively studied, investigations of alkaline phytases have been relatively few [3,4]. Alkaline phytases were Wrst reported in pollen grains of cattail, Typha latifolia [8], and Lilium longiXorum [9]. Subsequently, the presence of alkaline phytase in a number of legume seeds was reported [10]. The membrane-associated alkaline phytase from lily pollen has unique catalytic properties; it has a pH optimum of 8.0, is activated by calcium ions, not inhibited by Xuoride, and yields Ins-1,2,3-P3 as the Wnal product (Fig. 1) [11–13]. In addition to pH optima, acid and alkaline
134
S.P. Jog et al. / Archives of Biochemistry and Biophysics 440 (2005) 133–140
Fig. 1. Hydrolysis of phytic acid by alkaline phytase from lily pollen.
phytases diVer in substrate speciWcity, speciWcity of hydrolysis, Wnal products produced and metal ion requirement [3,4]. Alkaline phytase from bacterial sources (Bacillus subtilis, Bacillus amyloliquefaciens, and Bacillus licheniformis) exhibit catalytic properties similar to plant alkaline phytases; bacterial enzymes also require calcium for activity, are inhibited by EDTA, and yield InsP3 (Ins(1,3,5)P3 and Ins(2,4,6)P3) as the Wnal product [14–21]. In addition to the diVerences in catalytic properties, the crystal structure of alkaline phytase from B. amyloliquefaciens is structurally distinct, it lacks the active site motif RHGxRxP found in many acid phytases, and has a six-bladed propeller structure rather than the / domain structure of acid phytases [2,4,21]. Phytate is the major storage form of phosphate and inositol in seeds [22]. It exists complexed with essential mineral ions such as Na+, K+, Zn2+, and Fe2+ as well as proteins [22]. Monogastric animals such as poultry and swine do not have phytase in their digestive tract and are not able to utilize the phosphates, inositol, or metal ions in phytate [1–4,19]. The unabsorbed phytate is excreted and contributes to phosphate pollution in water bodies downstream of agriculturally intensive areas [23]. In addition, animal feed has to be supplemented by phosphate and mineral ions. The detrimental aVect of high concentrations of phytic acid in corn and soybeans on animal nutrition, as well as soil and water contamination and eutrophication is well established [25]. To alleviate the detrimental aVects of high concentrations of phytate, pig and poultry feed is supplemented by phytases [24– 26]. Depending on the target application, there is a need for phytases with a range of diVering biochemical and biophysical properties. The commercial signiWcance of phytases has triggered great interest in phytases exhibiting a range of characteristics as well as the need to understand the underlying basis for structure–activity relationships. In this paper, we investigate the biochemical properties of an alkaline phytase from lily pollen.
Materials and methods Materials Pollen grains from L. longiXorum L. cv Nellie White (1988 and 1991 harvests) were kindly donated by Professor Frank A. Loewus, Washington State University, Pullman, WA. Column chromatography of proteins was
conducted on a Fast Protein Liquid Chromatography system (Pharmacia Biotech GradiFrac System). Dialysis membrane (Spectra/Por 6.4 mm diameter, molecular weight cutoV: 12–14 kDa) was from Spectrum Laboratories, CA. Ion-exchange columns, HiTrap Q-FF (5 and 20 ml), were purchased from Amersham Biosciences. Chromatofocusing column (Tricorn column, Mono P 5/ 200 GL, 5 £ 200 mm, 4 ml) was purchased from Amersham Biosciences. Centriplus and Centricon Wltering devices (YM-30, molecular weight cutoV: 30 kDa) were from Amicon. Sodium phytate and myo-inositol hexasulfate (MIHS)1 were purchased from Sigma Chemicals, St. Louis, MO. Calculations to determine the concentration of free calcium in assay media were conducted using the program, Maxchelator 2004 (WEBMAXC STANDARD at http://www.stanford. edu/~cpatton/maxc.html). Homogenization and protein puriWcation All procedures were conducted at 4 °C unless otherwise noted [12]. To 2 g of lily pollen (1988 or 1991 harvests) was added ice-cold BuVer A (16 ml of 10 mM Tris– HCl, pH 7.0, containing reduced glutathione, 0.5 mM). A comparison of the germination rates of pollen grains from the 1988, 1991, and 2004 harvests showed no signiWcant diVerence, nearly 90% of the pollen grains germinated in all cases indicating that the 1988 and 1991 pollen grains were viable [27,28]. The suspension was stirred with a glass rod until most of the pollen kit adhered to the glass rod and could be easily removed. To the suspension was added solid cetylpyridinium bromide (80 mg, Wnal concentration 0.5% w/v). The suspension was homogenized with an IKA Euroturrax T20 homogenizer (IKA Works, Wilmington, NC) at 27,000 rpm for 2 £ 1 min with a 1 min delay in-between to cool the homogenate. The resulting crude homogenate was centrifuged at 10,000g for 30 min. The supernatant containing alkaline phytase activity was collected and the pellet containing cellular debris was discarded. Heat labile proteins in the supernatant were precipitated by heating at 55 °C for 80 min in a constant temperature water bath with slow shaking and then allowed to stand on ice for 30 min. The precipitated proteins were removed by cen-
1 Abbreviations used: MIHS, myo-inositol hexasulfate; EDTA, ethylenediaminetetraacetic acid; pNPP, para-nitrophenylphosphate.
S.P. Jog et al. / Archives of Biochemistry and Biophysics 440 (2005) 133–140
trifugation at 10,000g for 20 min and the supernatant was Wltered through a syringe Wlter containing 5 m cellulose ester Wlter paper (Advantec MFS, Pleasanton, CA). Solid ammonium sulfate was slowly added to the extract to 41.1% saturation (1.6 M) with gentle stirring. The solution was stirred for 1 h and then allowed to stand for 1 h. The precipitated proteins were removed by centrifugation at 15,000g for 30 min and the supernatant fractionated further with the slow addition of solid ammonium sulfate to 56.5% saturation (2.2 M) with gentle stirring. The solution was stirred for 1 h and then allowed to stand overnight. The precipitated proteins (41–56% pellet) were collected by centrifugation at 15,000g for 30 min and the supernatant was discarded. The 41–56% pellet was suspended in a minimum volume of BuVer A, dialyzed against Tris buVer (10 mM Tris– HCl, pH 8.0, containing reduced glutathione, 0.5 mM) to remove residual salts and assayed for alkaline phytase activity. Alternatively, the pellets may be frozen at ¡20 °C for up to 3 months without measurable loss of alkaline phytase activity. To the dialyzed enzyme extract was added an appropriate volume of NaCl (2 M) solution to a Wnal concentration of 0.13 M and then loaded onto a HiTrap Q-FF (Amersham Biosciences), a strong anion-exchange column equilibrated in BuVer B (20 mM Tris–HCl, pH 8.0, containing 0.13 M NaCl). Optimum puriWcation was obtained when 20–50 mg of protein was loaded on a 20 ml column or 3–15 mg was loaded on a 5 ml column. The loaded column was thoroughly washed with BuVer B. The bound proteins were eluted in Tris–HCl (20 mM, pH 8.0) with a linear gradient of 0.13–0.4 M NaCl over 80 min at a Xow rate of 5 ml/min. UV absorbance at 280 nm was monitored continuously. Fractions (10 ml) were collected and assayed for alkaline phytase activity. The active fractions were pooled, concentrated to a volume of about 1 ml by centrifugal Wltration in a Centriplus Wlter at 1,500g and desalted in the same Wlter by washing the protein solution with 3 £ 1 ml of BuVer A to a Wnal volume of 1 ml. In some cases, when the protein was further puriWed on a chromatofocusing column, BuVer C (25 mM imidazole–HCl, pH 7.4) was used in place of BuVer A. Active fractions from anion-exchange column were pooled and concentrated to 1 ml by centrifuging the protein solution in a Centriplus Wlter at 1500g. The Tris buVer was replaced with BuVer C by centrifuging in a Centricon Wlter at 3000g. Chromatofocusing column (Tricorn Mono P 5/200 GL, 5 £ 200 mm, 4 ml) was equilibrated with BuVer C [29]. PolybuVer 74 from Amersham Biosciences was diluted 1:10 (v/v), adjusted to pH 4.0 with HCl (Eluent buVer D), per manufacturer’s instructions, and one column volume (4 ml) of Eluent buVer D was passed through the column prior to the loading of sample. The sample was loaded on the column and
135
eluted with 15 column volumes (60 ml) of Eluent buVer D over 2 h at a Xow rate of 0.5 ml/min. Absorption at 280 nm was monitored continuously. Thirty fractions of 2 ml each were collected and assayed for alkaline phytase activity. The presence of Eluent buVer D did not aVect the alkaline phytase assay, however, buVer had to be removed before SDS– or native–PAGE could be conducted with the sample. PolybuVer from the fraction was replaced with BuVer A using a Centricon Wlter and the sample was concentrated to 0.5 ml in the same Wlter. Enzyme assays Alkaline phytase assay Assay with sodium phytate as substrate. Alkaline phytase activity was assayed by measuring the inorganic phosphate (Pi) released by the enzyme [12]. The assay mixture contained Tris–HCl buVer (100 mM, pH 8.0), NaCl (0.5 M), CaCl2 (1 mM), sodium phytate (1 mM), NaF (10 mM), and an aliquot of enzyme solution (0.35–7 g of enzyme) in a total volume of 250 l. The enzyme was puriWed 30- to 35-fold after anion-exchange chromatography, it was greater than 80% pure as judged by SDS gel electrophoresis; no acid phytase activity was detected after anion-exchange chromatography. The assay mixture was incubated at 37 °C for 1 h and the reaction was stopped by the addition of 50 l of 50% TCA. The release of inorganic phosphate was linear for 1 h; about 10–20% of the substrate was consumed in 1 h, depending on the amount of enzyme in the assay. In kinetic experiments to determine Km and Vmax, the reactions were carried out for 20 min. Inorganic phosphate was analyzed as previously described [12]. In brief, ammonium molybdate solution (700 l of a 1:6 solution of 10% w/v ascorbic acid and 0.42% ammonium molybdate (w/v) in 0.5 M H2SO4) was added and the solution was incubated at 37 °C for 1 h. Absorbance at 820 nm was measured and the inorganic phosphate concentration was determined from a calibration curve using KH2PO4 as the standard. One unit of enzyme is deWned as the amount of enzyme that releases 1 mol of Pi from sodium phytate per minute under these conditions. The standard assay conditions described here serve as the control in many of the experiments. Changes to standard assay conditions such as addition of salts, CaCl2, ethylenediaminetetraacetic acid (EDTA), or inhibitors are noted in Wgure legends. A solution of sodium vanadate was prepared by the method of Gallagher and Leonard [30]. To determine substrate speciWcity, other phosphorylated substrates (1 mM) were added to the assay media in place of sodium phytate. Activity of alkaline phytase towards para-nitrophenylphosphate (pNPP) was assayed by two methods: by measuring the Pi released (as described above) and by measuring the release of p-nitrophenolate anion spectrophotometrically at 410 nm [32]. Both methods gave the same results. To assay activity against pNPP by spectrophotometric method, the following procedure
136
S.P. Jog et al. / Archives of Biochemistry and Biophysics 440 (2005) 133–140
was employed: the assay reaction mixture contained Tris– HCl (100 mM, pH 7.0), CaCl2 (1 mM), NaF (10 mM), pNPP (0.05–1.0 mM), and an aliquot of protein homogenate in a total volume of 100 l. The assay mixture was incubated at 37 °C for 20 min and the reaction was stopped by the addition of 100 l of 10 M NaOH. The reaction mixture was incubated at room temperature for 10 min and the solution turned bright yellow due to the formation of p-nitrophenolate. Absorbance at 410 nm was measured and the concentration of p-nitrophenol was calculated using a molar extinction coeYcient of 18,525 M¡1 cm¡1. The molar extinction coeYcient was determined from a standard curve constructed by measuring absorption of varying concentrations of p-nitrophenolate under the same conditions as the assay. Acid phytase assay Acid phytase activity was assayed in a solution containing Na-acetate buVer (100 mM, pH 5.0), sodium phytate (1 mM), and CaCl2 (1 mM). NaF was not added to this assay mixture. The assay mixture was incubated at 37 °C for 1 h and the reaction was stopped by the addition of 50 l of 50% TCA. Pi released in the reaction was quantitated as described above. Total protein assay Protein concentrations were estimated by the Bradford dye binding method using the Bio-Rad assay reagent (Bio-Rad Laboratories, Hercules, CA) according to manufacturer’s instructions. Bovine serum albumin was used as the standard.
Results Lily pollen contains both acid and alkaline phytase activities [11]. Acid phytase activity can be separated from alkaline phytase activity by selective precipitation (heat and ammonium sulfate) followed by anionexchange chromatography [10,11]; no acid phytase activity was detected after anion-exchange chromatography. Therefore, anion-exchange-puriWed enzyme was used in all experiments unless otherwise noted. Additionally, Loewus et al. [11] had previously demonstrated that acid phytase and non-speciWc alkaline phosphatase activities in lily pollen extract are inhibited by Xuoride whereas alkaline phytase activity is unaVected, therefore, Xuoride was included in the buVer solutions when investigating alkaline phytase. InsP5 was the only product detected in the duration (20 min to 1 h) of enzyme assays. Substrate speciWcity and kinetic parameters Substrate speciWcity of phytases towards organophosphates varies widely. Generally, alkaline phytases exhibit
narrow substrate speciWcity and hydrolyze only phytate, whereas acid phytases hydrolyze a variety of substrates including pNPP, glucose-6-phosphate, fructose-6-phosphate, and ATP [2,4]. Alkaline phytase from lily pollen did not exhibit broad substrate speciWcity, it hydrolyzed InsP6 and pNPP at pH 8.0 (Fig. 2). The highest activity was against pNPP, 2.5-fold higher than against phytate. This observation diVers from that reported by Loewus et al. [11] who observed signiWcantly reduced activity against pNPP (5%) compared to phytate (100%). The reason for this discrepancy remains unexplained at this time. The ability of alkaline phytase from lily pollen to hydrolyze pNPP distinguishes it from alkaline phytases from B. subtilis, B. amyloliquefaciens, and T. latifolia that do not exhibit hydrolytic activity against pNPP [3,4]. Alkaline phytase exhibited very low activity (<10%) against other physiological substrates tested including glucose-6-phosphate, -glycerophosphate, fructose-6-phosphate, AMP, ADP, and ATP. Km values of phytate-degrading enzymes vary from <10 to 650 M [3,4,31]. Initial reaction rates of alkaline phytase at pH 8.0 were determined at various phytate and pNPP concentrations in the presence of calcium. Alkaline phytase exhibited Michaelis–Menten kinetics. The Km value for phytate was 81 M and Vmax was 217 nmol Pi/min/mg. InsP5 was the only product produced in the 20 min time period of the assay; thus, the kinetic parameters are related to the hydrolysis of the Wrst phosphate on the inositol ring [12]. The rates of hydrolysis of the second and third phosphates are signiWcantly lower [12]. When pNPP was used as the substrate, both the Km (327 M) and Vmax (1272 nmol Pi/ min/mg) values were higher. The Km value of alkaline phytase with phytate as substrate was comparable to values of other phytases from plant tissues (wheat bran, 77 M; soybean, 48–61 M; and lupine, 80 M), Km of bacterial enzymes is generally lower, 10–40 M [3]. Thus, alkaline phytase from lily pollen was similar to enzymes
Fig. 2. Hydrolytic activity of alkaline phytase with various phosphorylated substrates. Enzyme assay was conducted in buVer containing Tris–HCl (100 mM, pH 8.0), CaCl2 (1 mM), NaF (10 mM), NaCl (0.5 M), and diVerent substrates (1 mM). Error bars represent §SEM, n D 2.
S.P. Jog et al. / Archives of Biochemistry and Biophysics 440 (2005) 133–140
137
from bacterial sources (B. subtilis and B. amyloliquefaciens) in that it expresses narrow substrate speciWcity. InXuence of pH on enzyme activity As reported by Scott and Loewus [9], the maximum activity with phytate as the substrate was observed at pH 8.0; no activity was detected at pH 5.0. The inXuence of pH on enzyme activity towards pNPP was monitored from pH 4 to 9 (sodium-acetate buVer, pH 4.0–5.6; Trismaleate buVer, pH 5.5–9.0; and Tris–HCl buVer, pH 7.0– 9.0) (Fig. 3). With pNPP as the substrate, maximum activity was observed at pH 7.0, one pH unit less than that observed with phytate. The lowering of pH optimum with pNPP compared to physiologically occurring phosphate monoesters is consistent with observations with other phosphatases [33,34] and phytases [35,36]. The unusually low pKa2 of pNPP (5.5) compared to other phosphate monoesters (6.5) results in higher concentrations of dianionic pNPP, the species that binds at the enzyme active site, at lower pH [33,34]. EVect of metal ions Enhancement of alkaline phytase activity by CaCl2 was Wrst observed by Scott and Loewus [9]. Subsequently, activation of alkaline phytases by CaCl2 has been observed in enzymes from bacteria as well [4,15,18]. To determine the concentration of Ca2+ required for maximum activity, Ca2+–EDTA buVers were used; the concentration of CaCl2 was varied at Wxed EDTA concentration (0.5 mM) (Fig. 4). The concentration of free Ca2+, [Ca2+]f, was calculated using the software, Maxchelator. Fig. 4 indicates that maximum catalytic activity was observed when [Ca2+]f was 0.5–1.5 mM [9]. The addition of CaCl2 (1 mM) increased the activity of alkaline phytase 3.5- to 4-fold, however, some enzyme activity (17%), albeit low, was observed even in Ca2+-free environment. Higher concentrations of CaCl2 (ratio of CaCl2 to phytate above 5) led to precipitation of calcium phytate [9]. When higher con-
Fig. 3. The inXuence of pH on alkaline phytase activity. Enzyme assays were performed in 100 mM buVer solutions [sodium acetate (–䊉–), pH 4.0–5.6; Tris-maleate (–䉱–), pH 5.5–9.0; and Tris–HCl (–䊏–), pH 7.0– 9.0] containing substrate (1 mM), CaCl2 (1 mM), NaF (10 mM), and NaCl (0.5 M) at 37 °C for 1 h. Error bars represent §SEM, n D 2.
Fig. 4. The inXuence of Ca2+ on alkaline phytase activity. Concentration of free Ca2+ in the presence of EDTA was calculated using the program, Maxchelator 2004 (WEBMAXC STANDARD at http:// www.stanford.edu/~cpatton/maxc.html). Kd of Ca2+–EDTA complex was 1.116 £ 10¡8. The concentrations of total and free Ca2+ in the diVerent solutions were as follows: Total CaCl2 D 0.5 mM, [Ca2+]f D 0.0023 mM; total CaCl2 D 1.0 mM, [Ca2+]f D 0.5 mM; total [Ca2+]f D 1.5 mM; total CaCl2 D 5.0 mM, CaCl2 D 2.0 mM, [Ca2+]f D 4.5 mM; total CaCl2 D 10.0 mM, [Ca2+]f D 9.5 mM; and total CaCl2 D 20 mM, [Ca2+]f D 19.5 mM. Error bars represent §SEM, n D 3.
centrations of EDTA (0.5–5 mM) were added to remove endogenous Ca2+, catalytic activity was not completely eliminated; alkaline phytase activity (»17%) was observed even at relatively high EDTA concentration (5 mM). In this respect, alkaline phytase from lily pollen diVers markedly from those from bacterial sources that do not show hydrolytic activity in Ca2+-free medium [15,19]. That catalytic activity was observed in the presence of EDTA could be because Ca2+ in the enzyme active site is very tightly bound and not extracted by EDTA or because alkaline phytase is able to hydrolyze InsP6 at a low level even in the absence of CaCl2. EVect of salts on enzyme activity The ability of salts to increase the activity of alkaline phytase from lily pollen was discovered by accident. When the concentration of NaCl in the assay mixture was varied from 0 to 1 M, maximum activity (»100% enhancement) was observed in the presence of 0.5 M NaCl (data not shown). When the eVect of diVerent kinds of salts was investigated, it was discovered that the enhancement of catalytic activity was salt-speciWc and inXuenced by ionic strength rather than salt concentration; enzyme activity was enhanced by (NH4)2SO4, K2SO4, NH4Cl, KCl, and Na2SO4, all salts that are known to “salt-out” proteins (cosmotropic salts) (Fig. 5A) [37–41]. However, MgSO4 and MgCl2, both chaotropic salts that “salt-in” proteins, completely inactivated alkaline phytase at the same ionic strength. When all salts were added at the same concentration (0.5 M), the ionic strength in solutions containing (NH4)2SO4 and Na2SO4 is 1.5 compared to 0.5 in NaCl and KCl; catalytic enhancement due to (NH4)2SO4 and Na2SO4 was not observed indicating that ionic strength, not concentration, was important for rate enhancements (Fig. 5B).
138
S.P. Jog et al. / Archives of Biochemistry and Biophysics 440 (2005) 133–140
The eVect of salts on alkaline phytase activity followed the Hofmeister series, KCl produced the maximum enhancement, about 2.5-fold [37–41]. Cosmotropic salts that are known to stabilize protein structure brought about rate enhancement, whereas chaotropic salts that destabilize protein conformation and increase solubility inactivated alkaline phytase [37–41]. These data suggest that high concentrations (0.5 M) of salts such as NaCl and KCl increased alkaline phytase activity by inducing structural change and stabilizing the active conformation. When both NaCl and CaCl2 were added, enhancement of catalytic activity was additive. NaCl increased activity 3-fold and CaCl2 by 4-fold, the two together increased activity 6-fold. These data suggest that the enhancement by Ca2+ and NaCl may be due to interactions at diVerent domains, NaCl due to global changes in enzyme conformation and Ca2+ due to binding at the enzyme active site [4,18,20].
AVect of inhibitors Fluoride, vanadate, and myo-inositol hexasulfate (MIHS) are strong inhibitors of phosphatases and acid phytases [3,42,43]. A major diVerence between acid and alkaline phytases is that Xuoride ion (0.1–0.5 mM) inhibits acid phytase activity, whereas alkaline phytases are not inhibited [9,15]. Alkaline phytase from lily pollen was not inhibited even at relatively high concentrations, up to 15 mM of NaF (Fig. 6) [9]. However, when vanadate, a transition state analog, was added to the reaction mixture (0.25–3 mM), the reaction was inhibited by 78– 83%. Vanadate also inhibited alkaline phytase from B. subtilis, but to a lesser extent, about 25–50%. MIHS, a structural analog of phytic acid in which all the phosphate groups are substituted by sulfates, is a potent inhibitor of histidine acid phytase from A. Wcuum (phyA and phyB); phyA (pH 2.5) was completely inhibited at 10 M, phyA (pH 5.0) at 200 M and phyB at 1.0 M [43]. However, alkaline phytase from B. subtilis was not inhibited by MIHS [15]. The aVect of MIHS on alkaline phytase from lily pollen was investigated; no inhibition was observed up to 100 M and slight inhibition (»10%) was observed at high concentrations (500 M) (Fig. 6). Thermal stability Thermal stability of phytases and the ability to refold to an active conformation are important properties from a commercial perspective. With alkaline phytase, maximum activity was observed at 55 °C [9]. The thermal stability of alkaline phytase and the ability of the heat-denatured enzyme to refold were investigated by pre-incubating the enzyme at various temperatures (30– 80 °C) in the presence or absence of CaCl2, allowing the samples to refold at room temperature in the presence of
Fig. 5. The eVect of salts on catalytic activity of alkaline phytase. (A) The assay mixtures contained alkaline phytase (41–56% ammonium sulfate pellet), Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM) NaF (10 mM), CaCl2 (1 mM), and diVerent salts as indicated in the Wgure. The ionic strength of salt solutions was maintained at 0.5. The concentrations of KCl, NaCl, NH4Cl, and MgSO4 were 0.5 M. The concentrations of all other salts were 0.166 M. (B) The assay mixtures were the same as above except that the concentration of all salts was maintained at 0.5 M. Error bars represent §SEM, n D 2.
Fig. 6. The aVect of inhibitors, Xuoride, vanadate, and MIHS, on alkaline phytase activity. The assay mixtures contained Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM), NaCl (0.5 M), CaCl2 (1 mM), and diVerent concentrations of NaF, vanadate, and MIHS as indicated in the Wgure. Error bars represent §SEM, n D 2.
S.P. Jog et al. / Archives of Biochemistry and Biophysics 440 (2005) 133–140
CaCl2, and subsequently measuring residual activity (Fig. 7). No signiWcant reduction in activity was observed up to 50 °C, however, when the protein was heated at 55 °C for 10 min, the enzyme lost »40% of its original activity. It has been observed that in the case of some bacterial phytases, the presence of CaCl2 during the pre-incubation period helps retain the active conformation [16,19,44], however, when alkaline phytase was incubated with CaCl2 (1 mM), the reduction in activity
139
was greater—at 55 °C, only »40% of the activity remained in assays containing CaCl2 compared to 60% in the absence of CaCl2 (Fig. 7). To investigate if the partially denatured protein could refold to the active conformation, the duration and temperature of refolding were varied. Samples were heated at 50 and 55 °C and allowed to refold for 3 min at room temperature (Fig. 8A) or for 1 h at room temperature (Fig. 8B) or 4 °C (Fig. 8C). Fig. 8 indicates that the partially denatured enzyme did not regain activity under the conditions used. After further puriWcation by chromatofocusing chromatography, the enzyme was stable for up to 30 min at 55 °C; however, the denatured enzyme did not regain activity.
Discussion
Fig. 7. The aVect of pre-heating on alkaline phytase activity. Reaction mixtures containing alkaline phytase, Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM), NaF (10 mM), and NaCl (0.5 M) were preheated for 10 min at various temperatures in the presence (white bars) or absence (gray bars) of CaCl2 (1 mM). After pre-incubation, the enzyme was allowed to refold for 1 h at room temperature, »24 °C. CaCl2 (1 mM) was added to the reaction mixtures that were pre-heated in the absence of CaCl2. Residual activity was measured at 37 °C. C, control experiment without pre-heating assay solution. Error bars represent §SEM, n D 3.
Fig. 8. Refolding of heat-denatured enzyme. Assay mixtures containing alkaline phytase, Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM), NaF (10 mM), and NaCl (0.5 M) were pre-heated for 10 min at 50 or 55 °C in the presence (white bars) or absence (gray bars) of CaCl2 (1 mM). After pre-incubation, the enzyme mixtures were allowed to refold at (A) 3 min at room temperature, (B) 1 h at room temperature, or (C) 1 h at 4 °C. CaCl2 (1 mM) was added to the reaction mixtures that were pre-heated in the absence of CaCl2. Residual activity was measured at 37 °C. C, control experiment without preheating assay solution. Error bars represent §SEM, n D 3.
Alkaline phytase from lily pollen shares many biochemical similarities with alkaline phytases from bacterial sources [4,16] including optimum pH of 8.0, high optimum temperature of catalysis (55 °C), narrow substrate speciWcity, activation by calcium, inhibition by EDTA, and the production of InsP3 as the Wnal product. X-ray analysis has conWrmed the presence of Ca2+ at the enzyme active site of bacterial alkaline phytase, six Ca2+ atoms facilitate the complexation and orientation of the phosphate anions at the active site [46]. This is especially important for the phytate anion which carries nine negative charges at pH 8.0; six protons with pKa values between 1.0 and 2.2 and three with pKa values between 5.5 and 7.5 are ionized, the three least acidic protons with pKa between 9.0 and 9.5 remain attached to the oxygen [45]. In addition, the increased acidity of water held between two Ca2+ ions facilitates the formation of the nucleophilic hydroxide species which hydrolyzes phytate by the direct displacement of phosphate [46]. Therefore, Ca2+ may play a similar role in lily pollen alkaline phytase, namely facilitate the complexation and orientation of phytate as well as provide the hydroxide for nucleophilic attack at the phosphorous atom. However, alkaline phytase from lily pollen exhibited signiWcant diVerences in catalytic details compared to bacterial alkaline phytase including speciWcity of hydrolysis on the inositol ring and lack of inhibition by Xuoride. Alkaline phytase from lily pollen Wrst hydrolyzes the phosphate at the D-5 position followed by hydrolysis at the adjacent D-6 position and Wnally the D-4 phosphate [13]. Attack of phosphates adjacent to a free hydroxyl group has been observed in many phytases; it has been proposed that this may be due to the greater nucleophilicity of the hydroxyl oxygen compared to the ester bond oxygen and thus stronger binding of the hydroxyl to an electrophile adjacent to the cleavage site [47]. On the other hand, the -propeller alkaline phytase from Bacillus Wrst hydrolyzes the phosphate at the D-1
140
S.P. Jog et al. / Archives of Biochemistry and Biophysics 440 (2005) 133–140
(or enantiotopic D-3) or D-4 (or enantiotopic D-6) positions and subsequent hydrolysis occurs at every second phosphate to yield Ins(1,3,5)P3 or Ins(2,4,6)P3 [15,46]. Xray crystallographic analysis of the Bacillus alkaline phytase enzyme revealed that this is due to the preferential binding of adjacent phosphate groups at two nonequivalent phosphate binding sites with hydrolysis occurring at one of the sites [46]. Hydrolysis at the second phosphate rather than the one adjacent to the free hydroxyl suggests that the active site structure may be quite diVerent. In addition, X-ray crystallography of bacterial alkaline phytase suggests that Xuoride ion, an uncompetitive inhibitor, inhibits enzyme activity by replacing the hydroxide ion at the enzyme active site [46]. However, Xuoride was not an inhibitor of alkaline phytase from lily pollen. These biochemical diVerences suggest that the active sites of the lily pollen and bacterial alkaline phytases may diVer substantially. EVorts to determine the amino acid sequence and gather structural information that will shed more light on structure–activity relationships are ongoing.
Acknowledgments The authors thank Professor Frank A. Loewus (Institute of Biological Chemistry, Washington State University, Pullman, WA) for his generous gift of pollen grains. This work was supported by a grant from the Research Corporation (Grant RA0299). We gratefully acknowledge MTU for Graduate Student Fellowships to S.P.J., B.G.G., and B.D.M. We acknowledge Ms. Katie Gilles for preliminary work on salt eVect.
References [1] R.J. Wodzinski, A.H.J. Ullah, Adv. Appl. Microbiol. 42 (1996) 263–302. [2] E.J. Mullaney, A.H.J. Ullah, Biochem. Biophys. Res. Commun. 312 (2003) 179–184. [3] U. Konietzny, R. Greiner, Int. J. Food Sci. Technol. 37 (2002) 791– 812. [4] B.-C. Oh, W.-C. Choi, S. Park, Y.-O. Kim, T.-K. Oh, Appl. Microbiol. Biotechnol. 63 (2004) 362–372. [5] J.B. Vincent, M.W. Crowder, B.A. Averill, Trends Biochem. Sci. 17 (1992) 105–110. [6] J.E. Coleman, Annu. Rev. Biophys. Biomol. Struct. 21 (1992) 441– 483. [7] D.J. Cosgrove, Inositol Phosphates: Their Chemistry, Biochemistry, and Physiology, Elsevier, Amsterdam, 1980 , pp. 85–98. [8] A. Hara, S. Ebina, A. Kondo, T. Funaguma, Agric. Biol. Chem. 49 (1985) 3539–3544. [9] J.J. Scott, F.A. Loewus, Plant Physiol. 82 (1986) 333–335. [10] J.J. Scott, Plant Physiol. 95 (1991) 1298–1301. [11] B.G. Baldi, J.J. Scott, J.D. Everard, F.A. Loewus, Plant Sci. 56 (1988) 137–147. [12] L. Barrientos, J.J. Scott, P.P.N. Murthy, Plant Physiol. 106 (1994) 1489–1495.
[13] L.G. Barrientos, NMR investigation of the speciWcity of phytases, Ph.D. dissertation, Michigan Technological University, Houghton, MI 49931, USA, 1995. [14] J. Kerovuo, M. Lauraeus, P. Nurminen, N. Kalkkinen, J. Apajalahti, Appl. Environ. Microbiol. 64 (1998) 2079–2085. [15] J. Kerovuo, J. Rouvinen, F. Hatzack, Biochem. J. 352 (2000) 623– 628. [16] J. Kerovuo, I. Lappalainen, T. Reinikainen, Biochem. Biophys. Res. Commun. 268 (2000) 365–369. [17] Y.-O. Kim, J.-K. Lee, H.-K. Kim, J.-H. Yu, T.-K. Oh, FEMS Microbiol. Lett. 162 (1998) 185–191. [18] B.-C. Oh, B.S. Chang, K.-H. Park, N.-C. Ha, H.-K. Kim, B.-H. Oh, T.-K. Oh, Biochemistry 40 (2001) 9669–9676. [19] Y.-O. Kim, H.-K. Kim, K.-S. Bae, J.-H. Yu, T.-K. Oh, Enzyme Microb. Technol. 22 (1998) 2–7. [20] N.-C. Ha, B.-C. Oh, S. Shin, H.-J. Kim, T.-K. Oh, Y.-O. Kim, K.Y. Choi, B.-H. Oh, Nat. Struct. Biol. 7 (2000) 147–153. [21] A.J. Tye, F.K.Y. Siu, T.Y.C. Leung, B.L. Lim, Appl. Microbiol. Biotechnol. 59 (2002) 190–197. [22] N.R. Reddy, S.K. Sathe, D.R. Salunkhe, Adv. Food Res. 28 (1982) 1–92. [23] A.N. Sharpley, S.C. Chapra, R. Wedepohl, J.T. Sims, T.C. Daniel, K.R. Reddy, J. Environ. Qual. 23 (1994) 437–451. [24] T.S. Nelson, T.R. Shieh, R.J. Wodzinski, J.H. Ware, J. Nutr. 101 (1971) 1289–1294. [25] X.G. Lei, C.H. Stahl, J. Appl. Anim. Res. 17 (2000) 97–112. [26] X.G. Lei, C.H. Stahl, Appl. Microbiol. Biotechnol. 57 (2001) 474– 481. [27] D.B. Dickinson, Physiol. Plant. 20 (1967) 118–127. [28] F.A. Loewus, M.W. Loewus, Plant Cell Incompatibility Newsl. 22 (1990) 32–36. [29] L. Giri, Methods Enzymol. 182 (1990) 380–392. [30] S.R. Gallagher, R.T. Leonard, Plant Physiol. 70 (1982) 1335– 1340. [31] M. Wyss, R. Brugger, A. Kronenberger, R. Remy, R. Fimbel, G. Oesterhelt, M. Lehmann, A.P.G.M. van Loon, Appl. Environ. Microbiol. 65 (1999) 367–373. [32] K. Ostanin, E.H. Harms, P.E. Stevis, R. Kuciel, M.M. Zhou, R.L. Van Etten, J. Biol. Chem. 267 (1992) 22830–22836. [33] R.L. Van Etten, Ann. N. Y. Acad. Sci. 390 (1982) 27–51. [34] R.L. Van Etten, Phosphorous, Sulfur Silicon 76 (1993) 107–110. [35] D.B. Mitchell, K. Vogel, B.J. Weimann, L. Pasamontes, A.P.G.M. van Loon, Microbiology 143 (1997) 25–252. [36] R. Greiner, U. Konietzny, Kl.-D. Jany, Arch. Biochem. Biophys. 303 (1993) 107–113. [37] S.N. TimasheV, T. Arakawa, in: B.D. Hames, D. Rickwood (Eds.), second ed., Gel Electrophoresis of Proteins: A Practical Approach, Oxford University Press, New York, NY, 1990, pp. 331–345. [38] V.A. Parsegian, Nature 378 (1995) 335–336. [39] R. Leberman, A.K. Soper, Nature 378 (1995) 364–366. [40] P.T. WingWeld, in: J.E. Coligan, B.M. Dunn, D.W. Speicher, P.T. WingWeld (Eds.), Current Protocols in Protein Science, Wiley, New York, 1998, pp. A.3F.1–A.3F.8. [41] E.M. Wondrak, J.M. Louis, S. Oroszlan, FEBS Lett. 280 (1991) 344–346. [42] D.C. Crans, J.J. Smee, E. Gaidamauskas, L. Yang, Chem. Rev. 104 (2004) 849–902. [43] A.H.J. Ullah, K. Setumadhavan, Biochem. Biophys. Res. Commun. 251 (1998) 260–263. [44] Y.M. Choi, H.J. Suh, J.M. Kim, J. Protein Chem. 20 (2001) 2878– 2892. [45] A.J.R. Costello, T. Glonek, T.C. Myers, Carbohydr. Res. 46 (1976) 159–171. [46] S. Shin, N.-C. Ha, B.-C. Oh, T.-K. Oh, B.-H. Oh, Structure 9 (2001) 851–858. [47] D.J. Cosgrove, Inositol Phosphates: Their Chemistry, Biochemistry, and Physiology, Elsevier, Amsterdam, 1980 , pp. 99–105.
Alkaline phytase from lily pollen: Investigation of biochemical properties Sonali P. Jog, Barry G. Garchow, Bakul Dhagat Mehta, Pushpalatha P.N. Murthy ¤ Department of Chemistry, Michigan Technological University, Houghton, MI 49931, USA Received 28 March 2005, and in revised form 23 May 2005 Available online 11 July 2005
Abstract Phytases catalyze the hydrolysis of phytic acid (InsP6, myo-inositol hexakisphosphate), the most abundant inositol phosphate in cells. In cereal grains and legumes, it constitutes 3–5% of the dry weight of seeds. The inability of humans and monogastric animals such as swine and poultry to absorb complexed InsP6 has led to nutritional and environmental problems. The eYcacy of supplemental phytases to address these issues is well established; thus, there is a need for phytases with a range of biochemical and biophysical properties for numerous applications. An alkaline phytase that shows unique catalytic properties was isolated from plant tissues. In this paper, we report on the biochemical properties of an alkaline phytase from pollen grains of Lilium longiXorum. The enzyme exhibits narrow substrate speciWcity, it hydrolyzed InsP6 and para-nitrophenyl phosphate (pNPP). Alkaline phytase followed Michaelis–Menten kinetics with a Km of 81 M and Vmax of 217 nmol Pi/min/mg with InsP6 and a Km of 372 M and Vmax of 1272 nmol Pi/min/mg with pNPP. The pH optimum was 8.0 with InsP6 as the substrate and 7.0 with pNPP. Alkaline phytase was activated by calcium and inactivated by ethylenediaminetetraacetic acid; however, the enzyme retained a low level of activity even in Ca2+-free medium. Fluoride as well as myo-inositol hexasulfate did not have any inhibitory aVect, whereas vanadate inhibited the enzyme. The enzyme was activated by sodium chloride and potassium chloride and inactivated by magnesium chloride; the activation by salts followed the Hofmeister series. The temperature optimum for hydrolysis is 55 °C; the enzyme was stable at 55 °C for about 30 min. The enzyme has unique properties that suggest the potential to be useful as a feed supplement. 2005 Elsevier Inc. All rights reserved. Keywords: Phytate; Alkaline phytase; Pollen grains; Lilium longiXorum; Enzyme activation; Hofmeister series; Phosphate contamination; Animal feed
Phytases, a class of phosphatases, are the primary enzymes responsible for the hydrolysis of phytic acid [1– 6]. They catalyze the sequential hydrolysis of phytic acid to less phosphorylated inositol phosphates and, in some cases, to inositol [7]. A number of phytases with varying structural and catalytic properties have been found in plants, yeast, and bacteria [2–4]. Phytases have been classiWed on the basis of pH optima (acid and alkaline), catalytic mechanisms (histidine acid phosphatase-like phytase, purple acid phos-
*
Corresponding author. Fax: +1 906 487 2061. E-mail address: [email protected] (P.P.N. Murthy).
0003-9861/$ - see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2005.05.029
phatase-like phytase, and -propeller phytase), and speciWcity of hydrolysis (3-phytase, EC.3.1.3.8, 6-phytase, EC.3.1.3.26, and more recently 5-phytases, EC.3.1.3.72) [1–4]. Although acid phytases have been extensively studied, investigations of alkaline phytases have been relatively few [3,4]. Alkaline phytases were Wrst reported in pollen grains of cattail, Typha latifolia [8], and Lilium longiXorum [9]. Subsequently, the presence of alkaline phytase in a number of legume seeds was reported [10]. The membrane-associated alkaline phytase from lily pollen has unique catalytic properties; it has a pH optimum of 8.0, is activated by calcium ions, not inhibited by Xuoride, and yields Ins-1,2,3-P3 as the Wnal product (Fig. 1) [11–13]. In addition to pH optima, acid and alkaline
134
S.P. Jog et al. / Archives of Biochemistry and Biophysics 440 (2005) 133–140
Fig. 1. Hydrolysis of phytic acid by alkaline phytase from lily pollen.
phytases diVer in substrate speciWcity, speciWcity of hydrolysis, Wnal products produced and metal ion requirement [3,4]. Alkaline phytase from bacterial sources (Bacillus subtilis, Bacillus amyloliquefaciens, and Bacillus licheniformis) exhibit catalytic properties similar to plant alkaline phytases; bacterial enzymes also require calcium for activity, are inhibited by EDTA, and yield InsP3 (Ins(1,3,5)P3 and Ins(2,4,6)P3) as the Wnal product [14–21]. In addition to the diVerences in catalytic properties, the crystal structure of alkaline phytase from B. amyloliquefaciens is structurally distinct, it lacks the active site motif RHGxRxP found in many acid phytases, and has a six-bladed propeller structure rather than the / domain structure of acid phytases [2,4,21]. Phytate is the major storage form of phosphate and inositol in seeds [22]. It exists complexed with essential mineral ions such as Na+, K+, Zn2+, and Fe2+ as well as proteins [22]. Monogastric animals such as poultry and swine do not have phytase in their digestive tract and are not able to utilize the phosphates, inositol, or metal ions in phytate [1–4,19]. The unabsorbed phytate is excreted and contributes to phosphate pollution in water bodies downstream of agriculturally intensive areas [23]. In addition, animal feed has to be supplemented by phosphate and mineral ions. The detrimental aVect of high concentrations of phytic acid in corn and soybeans on animal nutrition, as well as soil and water contamination and eutrophication is well established [25]. To alleviate the detrimental aVects of high concentrations of phytate, pig and poultry feed is supplemented by phytases [24– 26]. Depending on the target application, there is a need for phytases with a range of diVering biochemical and biophysical properties. The commercial signiWcance of phytases has triggered great interest in phytases exhibiting a range of characteristics as well as the need to understand the underlying basis for structure–activity relationships. In this paper, we investigate the biochemical properties of an alkaline phytase from lily pollen.
Materials and methods Materials Pollen grains from L. longiXorum L. cv Nellie White (1988 and 1991 harvests) were kindly donated by Professor Frank A. Loewus, Washington State University, Pullman, WA. Column chromatography of proteins was
conducted on a Fast Protein Liquid Chromatography system (Pharmacia Biotech GradiFrac System). Dialysis membrane (Spectra/Por 6.4 mm diameter, molecular weight cutoV: 12–14 kDa) was from Spectrum Laboratories, CA. Ion-exchange columns, HiTrap Q-FF (5 and 20 ml), were purchased from Amersham Biosciences. Chromatofocusing column (Tricorn column, Mono P 5/ 200 GL, 5 £ 200 mm, 4 ml) was purchased from Amersham Biosciences. Centriplus and Centricon Wltering devices (YM-30, molecular weight cutoV: 30 kDa) were from Amicon. Sodium phytate and myo-inositol hexasulfate (MIHS)1 were purchased from Sigma Chemicals, St. Louis, MO. Calculations to determine the concentration of free calcium in assay media were conducted using the program, Maxchelator 2004 (WEBMAXC STANDARD at http://www.stanford. edu/~cpatton/maxc.html). Homogenization and protein puriWcation All procedures were conducted at 4 °C unless otherwise noted [12]. To 2 g of lily pollen (1988 or 1991 harvests) was added ice-cold BuVer A (16 ml of 10 mM Tris– HCl, pH 7.0, containing reduced glutathione, 0.5 mM). A comparison of the germination rates of pollen grains from the 1988, 1991, and 2004 harvests showed no signiWcant diVerence, nearly 90% of the pollen grains germinated in all cases indicating that the 1988 and 1991 pollen grains were viable [27,28]. The suspension was stirred with a glass rod until most of the pollen kit adhered to the glass rod and could be easily removed. To the suspension was added solid cetylpyridinium bromide (80 mg, Wnal concentration 0.5% w/v). The suspension was homogenized with an IKA Euroturrax T20 homogenizer (IKA Works, Wilmington, NC) at 27,000 rpm for 2 £ 1 min with a 1 min delay in-between to cool the homogenate. The resulting crude homogenate was centrifuged at 10,000g for 30 min. The supernatant containing alkaline phytase activity was collected and the pellet containing cellular debris was discarded. Heat labile proteins in the supernatant were precipitated by heating at 55 °C for 80 min in a constant temperature water bath with slow shaking and then allowed to stand on ice for 30 min. The precipitated proteins were removed by cen-
1 Abbreviations used: MIHS, myo-inositol hexasulfate; EDTA, ethylenediaminetetraacetic acid; pNPP, para-nitrophenylphosphate.
S.P. Jog et al. / Archives of Biochemistry and Biophysics 440 (2005) 133–140
trifugation at 10,000g for 20 min and the supernatant was Wltered through a syringe Wlter containing 5 m cellulose ester Wlter paper (Advantec MFS, Pleasanton, CA). Solid ammonium sulfate was slowly added to the extract to 41.1% saturation (1.6 M) with gentle stirring. The solution was stirred for 1 h and then allowed to stand for 1 h. The precipitated proteins were removed by centrifugation at 15,000g for 30 min and the supernatant fractionated further with the slow addition of solid ammonium sulfate to 56.5% saturation (2.2 M) with gentle stirring. The solution was stirred for 1 h and then allowed to stand overnight. The precipitated proteins (41–56% pellet) were collected by centrifugation at 15,000g for 30 min and the supernatant was discarded. The 41–56% pellet was suspended in a minimum volume of BuVer A, dialyzed against Tris buVer (10 mM Tris– HCl, pH 8.0, containing reduced glutathione, 0.5 mM) to remove residual salts and assayed for alkaline phytase activity. Alternatively, the pellets may be frozen at ¡20 °C for up to 3 months without measurable loss of alkaline phytase activity. To the dialyzed enzyme extract was added an appropriate volume of NaCl (2 M) solution to a Wnal concentration of 0.13 M and then loaded onto a HiTrap Q-FF (Amersham Biosciences), a strong anion-exchange column equilibrated in BuVer B (20 mM Tris–HCl, pH 8.0, containing 0.13 M NaCl). Optimum puriWcation was obtained when 20–50 mg of protein was loaded on a 20 ml column or 3–15 mg was loaded on a 5 ml column. The loaded column was thoroughly washed with BuVer B. The bound proteins were eluted in Tris–HCl (20 mM, pH 8.0) with a linear gradient of 0.13–0.4 M NaCl over 80 min at a Xow rate of 5 ml/min. UV absorbance at 280 nm was monitored continuously. Fractions (10 ml) were collected and assayed for alkaline phytase activity. The active fractions were pooled, concentrated to a volume of about 1 ml by centrifugal Wltration in a Centriplus Wlter at 1,500g and desalted in the same Wlter by washing the protein solution with 3 £ 1 ml of BuVer A to a Wnal volume of 1 ml. In some cases, when the protein was further puriWed on a chromatofocusing column, BuVer C (25 mM imidazole–HCl, pH 7.4) was used in place of BuVer A. Active fractions from anion-exchange column were pooled and concentrated to 1 ml by centrifuging the protein solution in a Centriplus Wlter at 1500g. The Tris buVer was replaced with BuVer C by centrifuging in a Centricon Wlter at 3000g. Chromatofocusing column (Tricorn Mono P 5/200 GL, 5 £ 200 mm, 4 ml) was equilibrated with BuVer C [29]. PolybuVer 74 from Amersham Biosciences was diluted 1:10 (v/v), adjusted to pH 4.0 with HCl (Eluent buVer D), per manufacturer’s instructions, and one column volume (4 ml) of Eluent buVer D was passed through the column prior to the loading of sample. The sample was loaded on the column and
135
eluted with 15 column volumes (60 ml) of Eluent buVer D over 2 h at a Xow rate of 0.5 ml/min. Absorption at 280 nm was monitored continuously. Thirty fractions of 2 ml each were collected and assayed for alkaline phytase activity. The presence of Eluent buVer D did not aVect the alkaline phytase assay, however, buVer had to be removed before SDS– or native–PAGE could be conducted with the sample. PolybuVer from the fraction was replaced with BuVer A using a Centricon Wlter and the sample was concentrated to 0.5 ml in the same Wlter. Enzyme assays Alkaline phytase assay Assay with sodium phytate as substrate. Alkaline phytase activity was assayed by measuring the inorganic phosphate (Pi) released by the enzyme [12]. The assay mixture contained Tris–HCl buVer (100 mM, pH 8.0), NaCl (0.5 M), CaCl2 (1 mM), sodium phytate (1 mM), NaF (10 mM), and an aliquot of enzyme solution (0.35–7 g of enzyme) in a total volume of 250 l. The enzyme was puriWed 30- to 35-fold after anion-exchange chromatography, it was greater than 80% pure as judged by SDS gel electrophoresis; no acid phytase activity was detected after anion-exchange chromatography. The assay mixture was incubated at 37 °C for 1 h and the reaction was stopped by the addition of 50 l of 50% TCA. The release of inorganic phosphate was linear for 1 h; about 10–20% of the substrate was consumed in 1 h, depending on the amount of enzyme in the assay. In kinetic experiments to determine Km and Vmax, the reactions were carried out for 20 min. Inorganic phosphate was analyzed as previously described [12]. In brief, ammonium molybdate solution (700 l of a 1:6 solution of 10% w/v ascorbic acid and 0.42% ammonium molybdate (w/v) in 0.5 M H2SO4) was added and the solution was incubated at 37 °C for 1 h. Absorbance at 820 nm was measured and the inorganic phosphate concentration was determined from a calibration curve using KH2PO4 as the standard. One unit of enzyme is deWned as the amount of enzyme that releases 1 mol of Pi from sodium phytate per minute under these conditions. The standard assay conditions described here serve as the control in many of the experiments. Changes to standard assay conditions such as addition of salts, CaCl2, ethylenediaminetetraacetic acid (EDTA), or inhibitors are noted in Wgure legends. A solution of sodium vanadate was prepared by the method of Gallagher and Leonard [30]. To determine substrate speciWcity, other phosphorylated substrates (1 mM) were added to the assay media in place of sodium phytate. Activity of alkaline phytase towards para-nitrophenylphosphate (pNPP) was assayed by two methods: by measuring the Pi released (as described above) and by measuring the release of p-nitrophenolate anion spectrophotometrically at 410 nm [32]. Both methods gave the same results. To assay activity against pNPP by spectrophotometric method, the following procedure
136
S.P. Jog et al. / Archives of Biochemistry and Biophysics 440 (2005) 133–140
was employed: the assay reaction mixture contained Tris– HCl (100 mM, pH 7.0), CaCl2 (1 mM), NaF (10 mM), pNPP (0.05–1.0 mM), and an aliquot of protein homogenate in a total volume of 100 l. The assay mixture was incubated at 37 °C for 20 min and the reaction was stopped by the addition of 100 l of 10 M NaOH. The reaction mixture was incubated at room temperature for 10 min and the solution turned bright yellow due to the formation of p-nitrophenolate. Absorbance at 410 nm was measured and the concentration of p-nitrophenol was calculated using a molar extinction coeYcient of 18,525 M¡1 cm¡1. The molar extinction coeYcient was determined from a standard curve constructed by measuring absorption of varying concentrations of p-nitrophenolate under the same conditions as the assay. Acid phytase assay Acid phytase activity was assayed in a solution containing Na-acetate buVer (100 mM, pH 5.0), sodium phytate (1 mM), and CaCl2 (1 mM). NaF was not added to this assay mixture. The assay mixture was incubated at 37 °C for 1 h and the reaction was stopped by the addition of 50 l of 50% TCA. Pi released in the reaction was quantitated as described above. Total protein assay Protein concentrations were estimated by the Bradford dye binding method using the Bio-Rad assay reagent (Bio-Rad Laboratories, Hercules, CA) according to manufacturer’s instructions. Bovine serum albumin was used as the standard.
Results Lily pollen contains both acid and alkaline phytase activities [11]. Acid phytase activity can be separated from alkaline phytase activity by selective precipitation (heat and ammonium sulfate) followed by anionexchange chromatography [10,11]; no acid phytase activity was detected after anion-exchange chromatography. Therefore, anion-exchange-puriWed enzyme was used in all experiments unless otherwise noted. Additionally, Loewus et al. [11] had previously demonstrated that acid phytase and non-speciWc alkaline phosphatase activities in lily pollen extract are inhibited by Xuoride whereas alkaline phytase activity is unaVected, therefore, Xuoride was included in the buVer solutions when investigating alkaline phytase. InsP5 was the only product detected in the duration (20 min to 1 h) of enzyme assays. Substrate speciWcity and kinetic parameters Substrate speciWcity of phytases towards organophosphates varies widely. Generally, alkaline phytases exhibit
narrow substrate speciWcity and hydrolyze only phytate, whereas acid phytases hydrolyze a variety of substrates including pNPP, glucose-6-phosphate, fructose-6-phosphate, and ATP [2,4]. Alkaline phytase from lily pollen did not exhibit broad substrate speciWcity, it hydrolyzed InsP6 and pNPP at pH 8.0 (Fig. 2). The highest activity was against pNPP, 2.5-fold higher than against phytate. This observation diVers from that reported by Loewus et al. [11] who observed signiWcantly reduced activity against pNPP (5%) compared to phytate (100%). The reason for this discrepancy remains unexplained at this time. The ability of alkaline phytase from lily pollen to hydrolyze pNPP distinguishes it from alkaline phytases from B. subtilis, B. amyloliquefaciens, and T. latifolia that do not exhibit hydrolytic activity against pNPP [3,4]. Alkaline phytase exhibited very low activity (<10%) against other physiological substrates tested including glucose-6-phosphate, -glycerophosphate, fructose-6-phosphate, AMP, ADP, and ATP. Km values of phytate-degrading enzymes vary from <10 to 650 M [3,4,31]. Initial reaction rates of alkaline phytase at pH 8.0 were determined at various phytate and pNPP concentrations in the presence of calcium. Alkaline phytase exhibited Michaelis–Menten kinetics. The Km value for phytate was 81 M and Vmax was 217 nmol Pi/min/mg. InsP5 was the only product produced in the 20 min time period of the assay; thus, the kinetic parameters are related to the hydrolysis of the Wrst phosphate on the inositol ring [12]. The rates of hydrolysis of the second and third phosphates are signiWcantly lower [12]. When pNPP was used as the substrate, both the Km (327 M) and Vmax (1272 nmol Pi/ min/mg) values were higher. The Km value of alkaline phytase with phytate as substrate was comparable to values of other phytases from plant tissues (wheat bran, 77 M; soybean, 48–61 M; and lupine, 80 M), Km of bacterial enzymes is generally lower, 10–40 M [3]. Thus, alkaline phytase from lily pollen was similar to enzymes
Fig. 2. Hydrolytic activity of alkaline phytase with various phosphorylated substrates. Enzyme assay was conducted in buVer containing Tris–HCl (100 mM, pH 8.0), CaCl2 (1 mM), NaF (10 mM), NaCl (0.5 M), and diVerent substrates (1 mM). Error bars represent §SEM, n D 2.
S.P. Jog et al. / Archives of Biochemistry and Biophysics 440 (2005) 133–140
137
from bacterial sources (B. subtilis and B. amyloliquefaciens) in that it expresses narrow substrate speciWcity. InXuence of pH on enzyme activity As reported by Scott and Loewus [9], the maximum activity with phytate as the substrate was observed at pH 8.0; no activity was detected at pH 5.0. The inXuence of pH on enzyme activity towards pNPP was monitored from pH 4 to 9 (sodium-acetate buVer, pH 4.0–5.6; Trismaleate buVer, pH 5.5–9.0; and Tris–HCl buVer, pH 7.0– 9.0) (Fig. 3). With pNPP as the substrate, maximum activity was observed at pH 7.0, one pH unit less than that observed with phytate. The lowering of pH optimum with pNPP compared to physiologically occurring phosphate monoesters is consistent with observations with other phosphatases [33,34] and phytases [35,36]. The unusually low pKa2 of pNPP (5.5) compared to other phosphate monoesters (6.5) results in higher concentrations of dianionic pNPP, the species that binds at the enzyme active site, at lower pH [33,34]. EVect of metal ions Enhancement of alkaline phytase activity by CaCl2 was Wrst observed by Scott and Loewus [9]. Subsequently, activation of alkaline phytases by CaCl2 has been observed in enzymes from bacteria as well [4,15,18]. To determine the concentration of Ca2+ required for maximum activity, Ca2+–EDTA buVers were used; the concentration of CaCl2 was varied at Wxed EDTA concentration (0.5 mM) (Fig. 4). The concentration of free Ca2+, [Ca2+]f, was calculated using the software, Maxchelator. Fig. 4 indicates that maximum catalytic activity was observed when [Ca2+]f was 0.5–1.5 mM [9]. The addition of CaCl2 (1 mM) increased the activity of alkaline phytase 3.5- to 4-fold, however, some enzyme activity (17%), albeit low, was observed even in Ca2+-free environment. Higher concentrations of CaCl2 (ratio of CaCl2 to phytate above 5) led to precipitation of calcium phytate [9]. When higher con-
Fig. 3. The inXuence of pH on alkaline phytase activity. Enzyme assays were performed in 100 mM buVer solutions [sodium acetate (–䊉–), pH 4.0–5.6; Tris-maleate (–䉱–), pH 5.5–9.0; and Tris–HCl (–䊏–), pH 7.0– 9.0] containing substrate (1 mM), CaCl2 (1 mM), NaF (10 mM), and NaCl (0.5 M) at 37 °C for 1 h. Error bars represent §SEM, n D 2.
Fig. 4. The inXuence of Ca2+ on alkaline phytase activity. Concentration of free Ca2+ in the presence of EDTA was calculated using the program, Maxchelator 2004 (WEBMAXC STANDARD at http:// www.stanford.edu/~cpatton/maxc.html). Kd of Ca2+–EDTA complex was 1.116 £ 10¡8. The concentrations of total and free Ca2+ in the diVerent solutions were as follows: Total CaCl2 D 0.5 mM, [Ca2+]f D 0.0023 mM; total CaCl2 D 1.0 mM, [Ca2+]f D 0.5 mM; total [Ca2+]f D 1.5 mM; total CaCl2 D 5.0 mM, CaCl2 D 2.0 mM, [Ca2+]f D 4.5 mM; total CaCl2 D 10.0 mM, [Ca2+]f D 9.5 mM; and total CaCl2 D 20 mM, [Ca2+]f D 19.5 mM. Error bars represent §SEM, n D 3.
centrations of EDTA (0.5–5 mM) were added to remove endogenous Ca2+, catalytic activity was not completely eliminated; alkaline phytase activity (»17%) was observed even at relatively high EDTA concentration (5 mM). In this respect, alkaline phytase from lily pollen diVers markedly from those from bacterial sources that do not show hydrolytic activity in Ca2+-free medium [15,19]. That catalytic activity was observed in the presence of EDTA could be because Ca2+ in the enzyme active site is very tightly bound and not extracted by EDTA or because alkaline phytase is able to hydrolyze InsP6 at a low level even in the absence of CaCl2. EVect of salts on enzyme activity The ability of salts to increase the activity of alkaline phytase from lily pollen was discovered by accident. When the concentration of NaCl in the assay mixture was varied from 0 to 1 M, maximum activity (»100% enhancement) was observed in the presence of 0.5 M NaCl (data not shown). When the eVect of diVerent kinds of salts was investigated, it was discovered that the enhancement of catalytic activity was salt-speciWc and inXuenced by ionic strength rather than salt concentration; enzyme activity was enhanced by (NH4)2SO4, K2SO4, NH4Cl, KCl, and Na2SO4, all salts that are known to “salt-out” proteins (cosmotropic salts) (Fig. 5A) [37–41]. However, MgSO4 and MgCl2, both chaotropic salts that “salt-in” proteins, completely inactivated alkaline phytase at the same ionic strength. When all salts were added at the same concentration (0.5 M), the ionic strength in solutions containing (NH4)2SO4 and Na2SO4 is 1.5 compared to 0.5 in NaCl and KCl; catalytic enhancement due to (NH4)2SO4 and Na2SO4 was not observed indicating that ionic strength, not concentration, was important for rate enhancements (Fig. 5B).
138
S.P. Jog et al. / Archives of Biochemistry and Biophysics 440 (2005) 133–140
The eVect of salts on alkaline phytase activity followed the Hofmeister series, KCl produced the maximum enhancement, about 2.5-fold [37–41]. Cosmotropic salts that are known to stabilize protein structure brought about rate enhancement, whereas chaotropic salts that destabilize protein conformation and increase solubility inactivated alkaline phytase [37–41]. These data suggest that high concentrations (0.5 M) of salts such as NaCl and KCl increased alkaline phytase activity by inducing structural change and stabilizing the active conformation. When both NaCl and CaCl2 were added, enhancement of catalytic activity was additive. NaCl increased activity 3-fold and CaCl2 by 4-fold, the two together increased activity 6-fold. These data suggest that the enhancement by Ca2+ and NaCl may be due to interactions at diVerent domains, NaCl due to global changes in enzyme conformation and Ca2+ due to binding at the enzyme active site [4,18,20].
AVect of inhibitors Fluoride, vanadate, and myo-inositol hexasulfate (MIHS) are strong inhibitors of phosphatases and acid phytases [3,42,43]. A major diVerence between acid and alkaline phytases is that Xuoride ion (0.1–0.5 mM) inhibits acid phytase activity, whereas alkaline phytases are not inhibited [9,15]. Alkaline phytase from lily pollen was not inhibited even at relatively high concentrations, up to 15 mM of NaF (Fig. 6) [9]. However, when vanadate, a transition state analog, was added to the reaction mixture (0.25–3 mM), the reaction was inhibited by 78– 83%. Vanadate also inhibited alkaline phytase from B. subtilis, but to a lesser extent, about 25–50%. MIHS, a structural analog of phytic acid in which all the phosphate groups are substituted by sulfates, is a potent inhibitor of histidine acid phytase from A. Wcuum (phyA and phyB); phyA (pH 2.5) was completely inhibited at 10 M, phyA (pH 5.0) at 200 M and phyB at 1.0 M [43]. However, alkaline phytase from B. subtilis was not inhibited by MIHS [15]. The aVect of MIHS on alkaline phytase from lily pollen was investigated; no inhibition was observed up to 100 M and slight inhibition (»10%) was observed at high concentrations (500 M) (Fig. 6). Thermal stability Thermal stability of phytases and the ability to refold to an active conformation are important properties from a commercial perspective. With alkaline phytase, maximum activity was observed at 55 °C [9]. The thermal stability of alkaline phytase and the ability of the heat-denatured enzyme to refold were investigated by pre-incubating the enzyme at various temperatures (30– 80 °C) in the presence or absence of CaCl2, allowing the samples to refold at room temperature in the presence of
Fig. 5. The eVect of salts on catalytic activity of alkaline phytase. (A) The assay mixtures contained alkaline phytase (41–56% ammonium sulfate pellet), Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM) NaF (10 mM), CaCl2 (1 mM), and diVerent salts as indicated in the Wgure. The ionic strength of salt solutions was maintained at 0.5. The concentrations of KCl, NaCl, NH4Cl, and MgSO4 were 0.5 M. The concentrations of all other salts were 0.166 M. (B) The assay mixtures were the same as above except that the concentration of all salts was maintained at 0.5 M. Error bars represent §SEM, n D 2.
Fig. 6. The aVect of inhibitors, Xuoride, vanadate, and MIHS, on alkaline phytase activity. The assay mixtures contained Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM), NaCl (0.5 M), CaCl2 (1 mM), and diVerent concentrations of NaF, vanadate, and MIHS as indicated in the Wgure. Error bars represent §SEM, n D 2.
S.P. Jog et al. / Archives of Biochemistry and Biophysics 440 (2005) 133–140
CaCl2, and subsequently measuring residual activity (Fig. 7). No signiWcant reduction in activity was observed up to 50 °C, however, when the protein was heated at 55 °C for 10 min, the enzyme lost »40% of its original activity. It has been observed that in the case of some bacterial phytases, the presence of CaCl2 during the pre-incubation period helps retain the active conformation [16,19,44], however, when alkaline phytase was incubated with CaCl2 (1 mM), the reduction in activity
139
was greater—at 55 °C, only »40% of the activity remained in assays containing CaCl2 compared to 60% in the absence of CaCl2 (Fig. 7). To investigate if the partially denatured protein could refold to the active conformation, the duration and temperature of refolding were varied. Samples were heated at 50 and 55 °C and allowed to refold for 3 min at room temperature (Fig. 8A) or for 1 h at room temperature (Fig. 8B) or 4 °C (Fig. 8C). Fig. 8 indicates that the partially denatured enzyme did not regain activity under the conditions used. After further puriWcation by chromatofocusing chromatography, the enzyme was stable for up to 30 min at 55 °C; however, the denatured enzyme did not regain activity.
Discussion
Fig. 7. The aVect of pre-heating on alkaline phytase activity. Reaction mixtures containing alkaline phytase, Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM), NaF (10 mM), and NaCl (0.5 M) were preheated for 10 min at various temperatures in the presence (white bars) or absence (gray bars) of CaCl2 (1 mM). After pre-incubation, the enzyme was allowed to refold for 1 h at room temperature, »24 °C. CaCl2 (1 mM) was added to the reaction mixtures that were pre-heated in the absence of CaCl2. Residual activity was measured at 37 °C. C, control experiment without pre-heating assay solution. Error bars represent §SEM, n D 3.
Fig. 8. Refolding of heat-denatured enzyme. Assay mixtures containing alkaline phytase, Tris–HCl (100 mM, pH 8.0), sodium phytate (1 mM), NaF (10 mM), and NaCl (0.5 M) were pre-heated for 10 min at 50 or 55 °C in the presence (white bars) or absence (gray bars) of CaCl2 (1 mM). After pre-incubation, the enzyme mixtures were allowed to refold at (A) 3 min at room temperature, (B) 1 h at room temperature, or (C) 1 h at 4 °C. CaCl2 (1 mM) was added to the reaction mixtures that were pre-heated in the absence of CaCl2. Residual activity was measured at 37 °C. C, control experiment without preheating assay solution. Error bars represent §SEM, n D 3.
Alkaline phytase from lily pollen shares many biochemical similarities with alkaline phytases from bacterial sources [4,16] including optimum pH of 8.0, high optimum temperature of catalysis (55 °C), narrow substrate speciWcity, activation by calcium, inhibition by EDTA, and the production of InsP3 as the Wnal product. X-ray analysis has conWrmed the presence of Ca2+ at the enzyme active site of bacterial alkaline phytase, six Ca2+ atoms facilitate the complexation and orientation of the phosphate anions at the active site [46]. This is especially important for the phytate anion which carries nine negative charges at pH 8.0; six protons with pKa values between 1.0 and 2.2 and three with pKa values between 5.5 and 7.5 are ionized, the three least acidic protons with pKa between 9.0 and 9.5 remain attached to the oxygen [45]. In addition, the increased acidity of water held between two Ca2+ ions facilitates the formation of the nucleophilic hydroxide species which hydrolyzes phytate by the direct displacement of phosphate [46]. Therefore, Ca2+ may play a similar role in lily pollen alkaline phytase, namely facilitate the complexation and orientation of phytate as well as provide the hydroxide for nucleophilic attack at the phosphorous atom. However, alkaline phytase from lily pollen exhibited signiWcant diVerences in catalytic details compared to bacterial alkaline phytase including speciWcity of hydrolysis on the inositol ring and lack of inhibition by Xuoride. Alkaline phytase from lily pollen Wrst hydrolyzes the phosphate at the D-5 position followed by hydrolysis at the adjacent D-6 position and Wnally the D-4 phosphate [13]. Attack of phosphates adjacent to a free hydroxyl group has been observed in many phytases; it has been proposed that this may be due to the greater nucleophilicity of the hydroxyl oxygen compared to the ester bond oxygen and thus stronger binding of the hydroxyl to an electrophile adjacent to the cleavage site [47]. On the other hand, the -propeller alkaline phytase from Bacillus Wrst hydrolyzes the phosphate at the D-1
140
S.P. Jog et al. / Archives of Biochemistry and Biophysics 440 (2005) 133–140
(or enantiotopic D-3) or D-4 (or enantiotopic D-6) positions and subsequent hydrolysis occurs at every second phosphate to yield Ins(1,3,5)P3 or Ins(2,4,6)P3 [15,46]. Xray crystallographic analysis of the Bacillus alkaline phytase enzyme revealed that this is due to the preferential binding of adjacent phosphate groups at two nonequivalent phosphate binding sites with hydrolysis occurring at one of the sites [46]. Hydrolysis at the second phosphate rather than the one adjacent to the free hydroxyl suggests that the active site structure may be quite diVerent. In addition, X-ray crystallography of bacterial alkaline phytase suggests that Xuoride ion, an uncompetitive inhibitor, inhibits enzyme activity by replacing the hydroxide ion at the enzyme active site [46]. However, Xuoride was not an inhibitor of alkaline phytase from lily pollen. These biochemical diVerences suggest that the active sites of the lily pollen and bacterial alkaline phytases may diVer substantially. EVorts to determine the amino acid sequence and gather structural information that will shed more light on structure–activity relationships are ongoing.
Acknowledgments The authors thank Professor Frank A. Loewus (Institute of Biological Chemistry, Washington State University, Pullman, WA) for his generous gift of pollen grains. This work was supported by a grant from the Research Corporation (Grant RA0299). We gratefully acknowledge MTU for Graduate Student Fellowships to S.P.J., B.G.G., and B.D.M. We acknowledge Ms. Katie Gilles for preliminary work on salt eVect.
References [1] R.J. Wodzinski, A.H.J. Ullah, Adv. Appl. Microbiol. 42 (1996) 263–302. [2] E.J. Mullaney, A.H.J. Ullah, Biochem. Biophys. Res. Commun. 312 (2003) 179–184. [3] U. Konietzny, R. Greiner, Int. J. Food Sci. Technol. 37 (2002) 791– 812. [4] B.-C. Oh, W.-C. Choi, S. Park, Y.-O. Kim, T.-K. Oh, Appl. Microbiol. Biotechnol. 63 (2004) 362–372. [5] J.B. Vincent, M.W. Crowder, B.A. Averill, Trends Biochem. Sci. 17 (1992) 105–110. [6] J.E. Coleman, Annu. Rev. Biophys. Biomol. Struct. 21 (1992) 441– 483. [7] D.J. Cosgrove, Inositol Phosphates: Their Chemistry, Biochemistry, and Physiology, Elsevier, Amsterdam, 1980 , pp. 85–98. [8] A. Hara, S. Ebina, A. Kondo, T. Funaguma, Agric. Biol. Chem. 49 (1985) 3539–3544. [9] J.J. Scott, F.A. Loewus, Plant Physiol. 82 (1986) 333–335. [10] J.J. Scott, Plant Physiol. 95 (1991) 1298–1301. [11] B.G. Baldi, J.J. Scott, J.D. Everard, F.A. Loewus, Plant Sci. 56 (1988) 137–147. [12] L. Barrientos, J.J. Scott, P.P.N. Murthy, Plant Physiol. 106 (1994) 1489–1495.
[13] L.G. Barrientos, NMR investigation of the speciWcity of phytases, Ph.D. dissertation, Michigan Technological University, Houghton, MI 49931, USA, 1995. [14] J. Kerovuo, M. Lauraeus, P. Nurminen, N. Kalkkinen, J. Apajalahti, Appl. Environ. Microbiol. 64 (1998) 2079–2085. [15] J. Kerovuo, J. Rouvinen, F. Hatzack, Biochem. J. 352 (2000) 623– 628. [16] J. Kerovuo, I. Lappalainen, T. Reinikainen, Biochem. Biophys. Res. Commun. 268 (2000) 365–369. [17] Y.-O. Kim, J.-K. Lee, H.-K. Kim, J.-H. Yu, T.-K. Oh, FEMS Microbiol. Lett. 162 (1998) 185–191. [18] B.-C. Oh, B.S. Chang, K.-H. Park, N.-C. Ha, H.-K. Kim, B.-H. Oh, T.-K. Oh, Biochemistry 40 (2001) 9669–9676. [19] Y.-O. Kim, H.-K. Kim, K.-S. Bae, J.-H. Yu, T.-K. Oh, Enzyme Microb. Technol. 22 (1998) 2–7. [20] N.-C. Ha, B.-C. Oh, S. Shin, H.-J. Kim, T.-K. Oh, Y.-O. Kim, K.Y. Choi, B.-H. Oh, Nat. Struct. Biol. 7 (2000) 147–153. [21] A.J. Tye, F.K.Y. Siu, T.Y.C. Leung, B.L. Lim, Appl. Microbiol. Biotechnol. 59 (2002) 190–197. [22] N.R. Reddy, S.K. Sathe, D.R. Salunkhe, Adv. Food Res. 28 (1982) 1–92. [23] A.N. Sharpley, S.C. Chapra, R. Wedepohl, J.T. Sims, T.C. Daniel, K.R. Reddy, J. Environ. Qual. 23 (1994) 437–451. [24] T.S. Nelson, T.R. Shieh, R.J. Wodzinski, J.H. Ware, J. Nutr. 101 (1971) 1289–1294. [25] X.G. Lei, C.H. Stahl, J. Appl. Anim. Res. 17 (2000) 97–112. [26] X.G. Lei, C.H. Stahl, Appl. Microbiol. Biotechnol. 57 (2001) 474– 481. [27] D.B. Dickinson, Physiol. Plant. 20 (1967) 118–127. [28] F.A. Loewus, M.W. Loewus, Plant Cell Incompatibility Newsl. 22 (1990) 32–36. [29] L. Giri, Methods Enzymol. 182 (1990) 380–392. [30] S.R. Gallagher, R.T. Leonard, Plant Physiol. 70 (1982) 1335– 1340. [31] M. Wyss, R. Brugger, A. Kronenberger, R. Remy, R. Fimbel, G. Oesterhelt, M. Lehmann, A.P.G.M. van Loon, Appl. Environ. Microbiol. 65 (1999) 367–373. [32] K. Ostanin, E.H. Harms, P.E. Stevis, R. Kuciel, M.M. Zhou, R.L. Van Etten, J. Biol. Chem. 267 (1992) 22830–22836. [33] R.L. Van Etten, Ann. N. Y. Acad. Sci. 390 (1982) 27–51. [34] R.L. Van Etten, Phosphorous, Sulfur Silicon 76 (1993) 107–110. [35] D.B. Mitchell, K. Vogel, B.J. Weimann, L. Pasamontes, A.P.G.M. van Loon, Microbiology 143 (1997) 25–252. [36] R. Greiner, U. Konietzny, Kl.-D. Jany, Arch. Biochem. Biophys. 303 (1993) 107–113. [37] S.N. TimasheV, T. Arakawa, in: B.D. Hames, D. Rickwood (Eds.), second ed., Gel Electrophoresis of Proteins: A Practical Approach, Oxford University Press, New York, NY, 1990, pp. 331–345. [38] V.A. Parsegian, Nature 378 (1995) 335–336. [39] R. Leberman, A.K. Soper, Nature 378 (1995) 364–366. [40] P.T. WingWeld, in: J.E. Coligan, B.M. Dunn, D.W. Speicher, P.T. WingWeld (Eds.), Current Protocols in Protein Science, Wiley, New York, 1998, pp. A.3F.1–A.3F.8. [41] E.M. Wondrak, J.M. Louis, S. Oroszlan, FEBS Lett. 280 (1991) 344–346. [42] D.C. Crans, J.J. Smee, E. Gaidamauskas, L. Yang, Chem. Rev. 104 (2004) 849–902. [43] A.H.J. Ullah, K. Setumadhavan, Biochem. Biophys. Res. Commun. 251 (1998) 260–263. [44] Y.M. Choi, H.J. Suh, J.M. Kim, J. Protein Chem. 20 (2001) 2878– 2892. [45] A.J.R. Costello, T. Glonek, T.C. Myers, Carbohydr. Res. 46 (1976) 159–171. [46] S. Shin, N.-C. Ha, B.-C. Oh, T.-K. Oh, B.-H. Oh, Structure 9 (2001) 851–858. [47] D.J. Cosgrove, Inositol Phosphates: Their Chemistry, Biochemistry, and Physiology, Elsevier, Amsterdam, 1980 , pp. 99–105.