Stereospecificity of myo-inositol hexakisphosphate hydrolysis by a ...

Appl Microbiol Biotechnol (2009) 82:95–103 DOI 10.1007/s00253-008-1734-5

BIOTECHNOLOGICALLY RELEVANT ENZYMES AND PROTEINS

Stereospecificity of myo-inositol hexakisphosphate hydrolysis by a protein tyrosine phosphatase-like inositol polyphosphatase from Megasphaera elsdenii Aaron A. Puhl & Ralf Greiner & L. Brent Selinger

Received: 22 July 2008 / Revised: 24 September 2008 / Accepted: 25 September 2008 / Published online: 14 October 2008 # Springer-Verlag 2008

Abstract Inositol polyphosphatases (IPPases), particularly those that can hydrolyze myo-inositol hexakisphosphate (Ins P6), are of biotechnological interest for their ability to reduce the metabolically unavailable organic phosphate content of feedstuffs and to produce lower inositol polyphosphates (IPPs) for research and pharmaceutical applications. Here, the gene coding for a new protein tyrosine phosphatase (PTP)-like IPPase was cloned from Megasphaera elsdenii (phyAme), and the biochemical properties of the recombinant protein were determined. The deduced amino acid sequence of PhyAme is similar to known PTPlike IPPases (29–44% identity), and the recombinant enzyme displayed strict specificity for IPP substrates. Optimal IPPase activity was displayed at an ionic strength of 250 mM, a pH of 5.0, and a temperature of 60°C. In order to elucidate its stereospecificity of Ins P6 dephosphorylation, a combination of high-performance ion-pair chromatography and kinetic studies was conducted. PhyAme displayed a stereospecificity that is unique among enzymes belonging to this class in that it preferentially cleaved Ins P6 at one of two phosphate positions, 1D-3 or 1D-4. PhyAme followed two distinct and specific routes of hydrolysis, predominantly degrading Ins P6 to Ins(2)P via: A. A. Puhl : L. B. Selinger (*) Department of Biological Sciences, University of Lethbridge, 4401 University Drive, Lethbridge, Alberta, Canada T1K 3M4 e-mail: [email protected] R. Greiner Department of Microbiology and Biotechnology, Max Rubner Institute, Federal Research Institute of Nutrition and Food, Haid-und-Neu-Strasse 9, 76131 Karlsruhe, Germany

(a) 1D-Ins(1,2,4,5,6)P5, 1D-Ins(1,2,5,6)P4, 1D-Ins(1,2,6)P3, and 1D-Ins(1,2)P2 (60%) and (b) 1D-Ins(1,2,3,5,6)P5, 1DIns(1,2,3,6)P4, Ins(1,2,3)P3, and D/L-Ins(1,2)P2 (35%). Keywords Phytase . Protein tyrosine phosphatase . myo-Inositol . Stereospecificity . Kinetics

Introduction myo-Inositol polyphosphates (IPPs) are ubiquitous products of inositol metabolism, and their biological importance in eukaryotic cells has been well established (Sasakawa et al. 1995; Shears 2001). IPPs have been implicated in myoinositol, phosphate, and cation storage (Batten and Lott 1986) and numerous regulatory functions involved with cell proliferation (Brailoiu et al. 2003; Hanakahi et al. 2000; Orchiston et al. 2004). Moreover, IPPs have been documented as having a number of novel metabolic effects (Ohkawa et al. 1984; Ruf et al. 1994; Zhang et al. 2005). The most abundant IPPs in most cells are the higher IPPs, myo-inositol hexakisphosphate (Ins P6) and myo-inositol pentakisphosphate (Ins P5; Sasakawa et al. 1995). The enzymes responsible for Ins P6 hydrolysis are a special class of phosphatases that are collectively known as phytases. Four distinct classes of phosphatases have been characterized in the literature as having phytase activity; i.e., histidine acid phosphatases, β-propeller phytases, purple acid phosphatases (Mullaney and Ullah 2003), and most recently, protein tyrosine phosphatase (PTP)-like myoinositol polyphosphatases (PTP-like IPPases; Chu et al. 2004; Puhl et al. 2007, 2008a, b). Phytases hydrolyze Ins P6 in a sequential and stepwise manner, yielding lower IPPs which may again become substrates for further hydrolysis (Konietzny and Greiner 2002). This occurs at different rates

96

and in different orders among phytases, and may be evidence of the variety of biological roles played by these enzymes as well as their IPP substrates and products. IPPases, including phytases, have been the focus of many studies due to interest in their role in cellular regulation and proliferation, their ability to reduce the metabolically unavailable organic phosphate content of livestock feedstuffs, and their ability to produce lower IPPs (Greiner and Konietzny 1996; Konietzny and Greiner 2002; Sasakawa et al. 1995). The growing list of research and pharmaceutical applications for specific IPPs has greatly increased interest in the preparation of these compounds. The chemical synthesis of individual IPPs involves difficult steps and is performed at extreme conditions (Billington 1993), and the separation of individual isomers is problematic with most analytical approaches. Since phytases hydrolyze Ins P6 in an ordered and stepwise manner, the production of IPPs and free myoinositol using phytase is a promising alternative to chemical synthesis (Greiner and Konietzny 1996). A novel class of IPPase has recently been described that contain a PTP-like active site signature sequence (HCX5R) that facilitates a classical PTP mechanism of dephosphorylation but is highly specific for IPP substrates (Chu et al. 2004; Puhl et al. 2007). While the biological function of these enzymes remains unclear, the activities of many PTP superfamily enzymes have been found to be essential for regulating cellular signal transduction cascades (Cho et al. 2006). A number of putative PTP-like IPPase homologues have been partially cloned from a range of anaerobic bacteria (Nakashima et al. 2007). Here we report the cloning and sequencing of a new gene encoding PTP-like IPPase from Megasphaera elsdenii, a Gram-negative anaerobic coccus that is a normal inhabitant of the rumen as well as the gastrointestinal tract in humans (Elsden et al. 1956; Brancaccio and Legendre 1979; Haikara and Helander 2006). The biochemical properties of the recombinant protein were also determined. Isomer-specific high-performance ion-exchange chromatography (HPIC) combined with kinetic analysis was used to determine the stereospecificity of Ins P6 hydrolysis and the identity of the intermediates produced by the hydrolysis pathway.

Materials and methods Gene cloning M. elsdenii strain YR60 (Yanke et al. 1998) was cultured anaerobically (100% CO2) at 39°C in Hungate tubes with 5 ml of modified Scott and Dehority medium (Scott and Dehority 1965) containing 10% (v/v) rumen fluid, 0.2% (w/v) glucose, 0.2% (w/v) cellobiose, and 0.3% (v/v) starch. Genomic DNA was extracted using standard protocols

Appl Microbiol Biotechnol (2009) 82:95–103

(Priefer et al. 1984) and partially digested with HindIII. Identification and isolation of the subgenomic DNA corresponding to the approximate size of the putative IPPasecontaining fragment was performed as described previously (Puhl et al. 2008b). Gel-purified subgenomic DNA containing the gene of interest was ligated into the HindIII site of dephosphorylated pBluescript II SK (+) (Stratagene, La Jolla, CA, USA). Polymerase chain reaction (PCR) primers were generated from the known internal phyAme partial sequence (Nakashima et al. 2007) and were used in conjunction with M13 and T7 universal primers to generate PCR products from the ligation product corresponding to regions of phyAme adjacent to the known partial gene fragment. The PCR products were ligated into pGEM-T Easy (Promega, Madison, WI, USA) and sequenced by automated cycle sequencing at the University of Calgary Core DNA and Protein services facilities. Sequence data were analyzed with the aid of SequencherTM version 4.0 (Gene Codes, Ann Arbor, MI, USA) and MacDNAsis version 3.2 (Hitachi Software Engineering, San Bruno, CA, USA). Homology searches in GenBank (Fassler et al. 2000) were done using BLAST (Altschul et al. 1990), and preliminary sequence alignments were generated using Clustal W 1.82 (Chenna et al. 2003). Alignment optimization was carried out with GeneDoc (Nicholas et al. 1997) using methods for comparative structure-based sequence alignments (Greer 1981) and the experimentally determined structure of a PTP-like IPPase from Selenomonas ruminantium (Protein Data Bank (PDB) accession: 2B4P). Secondary structure predictions were generated with SSpro using recurrent neural networks (Pollastri et al. 2002) on the Scratch web server (Cheng et al. 2005). Recombinant phyAme expression construct The region coding for the mature M. elsdenii IPPase (phyAme; GeneBank accession number DQ257441; residues 26–360) was amplified from genomic DNA using PCR. The predicted signal peptide sequence was determined with SignalP 3.0 (Bendtsen et al. 2004). PhyAme expression construct primers were: GCCATATGGTTTTTTCGGCC ATGGGTAT and G CGAATTCTCAACGGTTATT GACTCTCA and included an NdeI and EcoRI site (underlined), respectively, for cloning and a 5′ GC cap. The PCR product was digested with NdeI and EcoRI and ligated into similarly digested pET28b vector (Novagen, San Diego, CA, USA). Constructs were verified with automated cycle sequencing. Protein production and purification Escherichia coli BL21 (DE3) cells (Novagen) were transformed with the phyAme expression construct. Over-


expression was carried out according to the instructions in the pET Systems Manual (Novagen). Cells were induced with the addition of isopropylthiogalactoside (IPTG) to a final concentration of 1 mM and incubated for 4 h at 37°C. Induced cells were harvested by centrifugation and resuspended in lysis buffer: 25 mM KH2PO4 (pH 7), 0.6 M NaCl, 15 mM imidazole, 1 mM β-mercaptoethanol (BME), and one Complete Mini, ethylenediamine tetraacetic acid (EDTA)-free protease inhibitor tablet (Roche Applied Science; Laval, Quebec, Canada). Cells were disrupted by sonication, and debris was removed by centrifugation at 15,000×g for 15 min. The protein was purified to homogeneity using Ni2+-nitrilotriacetic acid (NTA) spin columns according to the supplied protocol (Qiagen). Protein was washed on the column with lysis buffer and then with wash buffer #2 (20 mM KH2PO4 (pH 7), 300 mM NaCl, 10% (v/v) glycerol, 15 mM imidazole, and 1 mM BME). Protein was eluted with lysis buffer containing 350 mM imidazole. Homogeneity was confirmed on a 12% (w/v) sodium dodecyl sulfatepolyacrylamide separating gel (SDS-PAGE) with a 4% (w/ v) stacking gel (Laemmli 1970) that was stained with Coomassie Brilliant Blue R-250. Purified protein was dialyzed into 20 mM Tris–HCl (pH 7), 300 mM NaCl, 0.1 mM EDTA, and 5 mM BME and stored at 4°C or dialyzed into 50 mM NH4(CO3)2 (pH 8), lyophilized, and stored at −20°C. The theoretical Mr and extinction coefficients of the proteins were determined using ProtParam tool (Gasteiger et al. 2005). The concentrations of protein solutions were determined with their absorbance at 280 nm. Assay and quantification of enzymatic activity Activity measurements were carried out at 37°C. Standard phosphatase assay mixtures consisted of a 600 μl buffered substrate solution and 150 μl of a 0.5 μM enzyme solution. The buffered substrate solution contained 50 mM sodium acetate (pH 4.5) and 2 mM sodium phytate. Ionic strength (I) was always held constant at 0.2 M with NaCl or, to examine the effect of I, varied from 0 to 0.8 M with NaCl. To determine the effect of pH, activity was measured at a range of pH values with overlapping buffer systems: [50 mM] glycine (pH 2–3), formate (3–4), sodium acetate (4–6), imidazole (6–7), and Tris–HCl (7–8). Activity was also measured at incremental temperatures from 10°C to 70°C. Substrate specificity was determined by replacing Ins P6 in the standard assay with various other phosphorester containing substrates; i.e., β-glycerophosphate, D/L-α-glycerophosphate, α-naphthyl phosphate, phospho (enol) pyruvate, phenolphthalein diphosphate, o-nitophenyl-β-Dgalactopyranoside-6-phosphate, phenyl phosphate, ρ-nitrophenyl phosphate (ρNPP), 5-bromo-4-chloro-3-indolyl

97

phosphate (BCIP), O-phospho-L-tyrosine, O-phospho-L-threonine, O-phospho-L-serine, adenosine triphosphate (ATP), adenosine diphosphate, pyrophosphate, D-fructose-1,6-diphosphate, D-fructose-6-phosphate, D-glucose-6-phosphate, and D-ribose-5-phosphate. Kinetic parameters were determined with the standard assay and a variable concentration (0.025–2 mM) of Ins P6 or another of the IPPs tested. Following a 3-min incubation period, the reactions were stopped, and the liberated phosphate was quantified using a modified ammonium molybdate method as described previously (Puhl et al. 2007, 2008b). Activity (U) was expressed as μmol phosphate liberated per minute. All assays were run in triplicate with at least three independent replicates performed for each investigation, and mean values have been reported. The steady-state kinetic constants (KM, kcat) were calculated from Michaelis–Menton plots. The data were analyzed with non-linear regression using Sigma-Plot 8.0 (Systat Software, Point Richmond, CA, USA). Preparation of individual myo-inositol phosphate isomers Phytases from Aspergillus niger, E. coli, and rye were used to generate 1D-Ins(1,2,4,5,6)P5, 1D-Ins(1,2,3,4,5)P4, and 1D-Ins(1,2,3,5,6)P3 from Ins P6. These isomers and the Ins P5 products generated by PhyAme were prepared as described previously (Greiner et al. 2002a, b). Identification of enzymatically formed hydrolysis products Standard phosphatase assays were run, and the periodically stopped reactions were resolved on a high-performance ion chromatography system using a Carbo Pac PA-100 (4× 250 mm) analytical column (Dionex, Sunnyvale, CA, USA) and a gradient of 5–98% HCl (0.5 M, 0.8 ml/min) as previously described (Skoglund et al. 1998). The eluants were mixed in a post-column reactor with 0.1% (w/v) Fe(NO3)3 in a 2% (w/v) HClO4 solution (0.4 ml/min; Phillippy and Bland 1988). The combined flow rate was 1.2 ml/min. myo-Inositol monophosphates were produced by incubation of 1.0 U of PhyAme with a limiting amount (0.1 μmol) of Ins P6 in a final volume of 500 μl of 50 mM NH4-acetate. The end products were identified using a gas chromatograph coupled with a mass spectrometer as previously described (Greiner et al. 2002a, b).

Results Sequence analysis A 1.8-kbp DNA fragment (GenBank accession number, DQ257441) was isolated from the genome of M. elsdenii by cloning regions up- and downstream of a sequence fragment determined by Nakashima et al. (2007). BLAST

98


analysis of the sequenced product indicated the presence of one open reading frame (ORF; phyAme) and one partial ORF (orf2) with homologues in GenBank. orf2 is located 200 bp downstream of phyAme, and its predicted product is similar to the N terminus of a major envelope protein of S. ruminantium (GenBank accession number AB252707). phyAme encodes a 360-amino-acid polypeptide that contains a PTP-like active site signature sequence, HCEAGAGR. The predicted sequence of PhyAme is similar (29–44% amino acid sequence identity) to known PTP-like IPPases from S. ruminantium; PhyAsr (Chu et al. 2004; Puhl et al. 2007), S. ruminantium subsp. lactilytica (Puhl et al. 2008b), and Selenomonas lacticifex (Puhl et al. 2008a). A search of GenBank using the PhyAme sequence as a probe has identified a number of homologues putatively produced by Clostridial species (i.e., anaerobic Grampositive bacteria) as well as an assortment of Gramnegative bacteria, including plant pathogens (Acidovorax, Pseudomonas, Xanthomonas), a predacious bacterium (Bdellovibrio), a human pathogen (Legionella), and a member of the fruiting, gliding bacteria (Stigmatella). Consistent with other characterized bacterial PTP-like IPPases, PhyAme contains a predicted N-terminal signal peptide suggesting it is secreted. In addition to the

conserved PTP-like active-site signature sequence, a comparative structure-based sequence alignment (Fig. 1) suggests that PhyAme contains a conserved aspartic acid responsible for acid/base catalysis in PhyAsr (Puhl et al. 2007). Also notable is the sequence and predicted structural similarity between the proteins in regions corresponding to the small partial β-barrel domain of PhyAsr from S. ruminantium, a distinguishing feature from “typical” PTPs (Chu et al. 2004; Puhl et al. 2007).

Fig. 1 Amino acid sequence alignment of the M. elsdenii PTP-like IPPase and its characterized GeneBank homologues. Shading is according to alignment consensus as given by Gene Doc with similarity groups enabled (black 100%, dark grey 75%). The protein abbreviation, source and GenBank accession numbers are as follows: PhyAsr, S. ruminantium, AAQ13669; PhyAsl, S. lacticifex, ABC69367; PhyBsl, S. lacticifex, ABC69361; PhyAsrl, S. ruminantium subsp. lactylitica, ABC69359; PhyAme, M. elsdenii, ABC69358. Numbering is according to the sequence of PhyAme found in

GeneBank. The PTP-like signature sequence and the conserved upstream aspartic acid are identified by asterisks. Experimentally determined secondary structures are identified for PhyAsr (PDB accession: 1DKQ) above the sequences, and those predicted for PhyAme according to Recurrent Neural Networks (Baldi and Pollastri 2003) are identified below the sequences. Arrows represent β strands, and boxes indicate α helices. The secondary structures corresponding to the partial β-barrel domain of PhyAsr (Chu et al. 2004) are indicated by vertical stripes

Protein expression and purification Following induction with IPTG, overproduction of a polypeptide with an approximate Mr of 39 k was observed with SDSPAGE. This is consistent with the mass predicted from the sequence of the recombinant protein (predicted Mr =40 k). Ni2+-NTA purification was able to produce >99% homogeneity of PhyAme in a single step, as determined by SDS-PAGE and Coomassie Brilliant Blue-250 staining (data not shown). Enzymatic activity and substrate specificity The activity of PhyAme toward Ins P6 was examined because of its sequence similarity with known PTP-like


IPPases. PhyAme can hydrolyze Ins P6 with a maximum specific activity of 269.3 U mg−1. Thrombin cleavage of the 6xHis tag had no effect on activity towards Ins P6 (data not shown). The effects of ionic strength (I), pH, and temperature on the IPPase activity of PhyAme were examined with Ins P6 as a substrate (Fig. 2). Optimal activity was at ionic strengths between 0.2 and 0.3 M. PhyAme displayed activity over a narrow range of acidic pH values, and optimal activity was observed at pH 5. PhyAme displayed maximal activity at 60°C under the conditions of our standard assay.

99

We tested the ability of PhyAme to hydrolyze various other common phosphoester-containing substrates in order to characterize its specificity. The compounds that were hydrolyzed by PhyAme are listed in Table 1. PhyAme has an extremely narrow substrate specificity, showing significant activity only towards IPP substrates. PhyAme displayed very little activity towards the commonly used phosphatase substrates ρNPP and BCIP or towards any of phosphorylated amino acids tested. The rate of PhyAme-catalyzed phosphate release can be saturated by increasing the concentration of any of the IPP isomers tested and remains linear over the time period of the assay (data not shown). The apparent kcat and Km values for PhyAme with Ins P6 as a substrate were 122.1 s−1 and 64.2 μM, respectively. Kinetic parameters were also determined for the possible PhyAme Ins P5 hydrolysis products in order to determine the specific Ins P5 isomers generated and are displayed in Table 2. Products of Ins P6 dephosphorylation Isomer-specific HPIC analysis was used to identify the hydrolysis products generated by PhyAme. Purified PhyAme was incubated with excess Ins P6 for 60, 120, and 300 min, and the stopped reactions were resolved by HPIC (Fig. 3). Following 60 min of incubation, the quantity of Ins P6 had decreased, and D/L-Ins(1,2,4,5,6)P5 and D/LIns(1,2,3,4,5)P5 appeared as the major Ins P5 degradation products (60% and 35%, respectively), along with very small amounts of Ins(1,2,3,4,6)P5 (5% of Ins P5 products). Significant quantities of D/L-Ins(1,2,5,6)P4, D/L-Ins(1,2,3,4) P4, and Ins(1,2,3)P3 and/or D/L-Ins(1,2,6)P3 plus trace amounts of D/L-Ins(1,2,4,5)P4 and D/L-Ins(1,2,4,6)P4 were also found after 60 min incubation. Following 120 min of incubation, the chromatogram was similar to that after Table 1 Substrates that were dephosphorylated by PhyAme Substrate

Fig. 2 Effects of pH (a), ionic strength (b), and temperature (c) on PhyAme activity. Values are normalized to pH 5 (a), 0.25 M ionic strength (b), and 60°C (c). The data presented are mean values with error bars representing the standard deviation between three independent experiments

Ins P6 ATP D-Fructose-1,6-diphosphate α-Naphthyl acid phosphate ρNPP Phospho (enol) pyruvate BCIP O-Phospho-L-tyrosine D-Ribose-5-phosphate O-Phospho-L-threonine Phenolphthalein diphosphate

Specific activity (U mg−1)

Relative activity (%)

269.30 1.37 0.80 0.79 0.78 0.76 0.73 0.69 0.38 0.35 0.28

100.00 0.51 0.30 0.29 0.29 0.28 0.27 0.26 0.14 0.13 0.10

For determination of relative activity, rate of Ins P6 hydrolysis was taken as 100%. A full list of substrates tested is presented in “Materials and methods”

100


Table 2 Kinetic parameters for enzymatic myo-inositol polyphosphate dephosphorylation by PhyAme, where values given represent the average ± standard deviation of at least three separate experimental runs Substrate

Km (μM)

kcat (s−1)

Ins(1,2,3,4,5,6)P6

64.2±0.61 61.3±0.57 61.8±0.52 102.5±0.67 61.1±0.49 61.5±0.53

122.1±1.6 134.5±1.8 135.3±1.9 78.4±1.1 133.9±1.6 135.8±1.7

D-Ins(1,2,4,5,6)P5 D-Ins(1,2,3,5,6)P5 D-Ins(1,2,3,4,5)P5 a a

a

InsP5—D/l-Ins(1,2,4,5,6)P5 InsP5—D/l-Ins(1,2,3,4,5)P5

Generated by the PTP-like phytase from M. elsdenii

Fig. 3 High-perfomance ion chromatography analysis of hydrolysis products of myo-inositol polyphosphates by PhyAme. a Reference sample. The source of the reference myo-inositol phosphates is as indicated in (Skoglund et al. 1998); Peaks: 1 Ins(1,2,3,4,5,6)P6; 2 D/L-Ins (1,2,4,5,6)P5; 3 D/L-Ins (1,2,3,4,5)P5; 4 Ins(1,2,3,4,6)P5; 5 Ins(2,4,5,6)P4; 6 D/L-Ins (1,2,5,6)P4; 7 D/L-Ins(1,2,4,5)P4; 8 D/L-Ins(1,2,3,4)P4; 9 D/L-Ins (1,2,4,6)P4; 10 D/L-Ins(1,4,5)P3, D/L-Ins(2,4,5)P3; 11 Ins(1,2,3) P3, D/L-Ins(1,2,6)P3; 12 D/L-Ins (1,2,4)P3; 13 D/L-Ins(1,2)P2, D/ L-Ins(4,5)P2, Ins(2,5)P2. b PhyAme incubated with Ins P6 for 60 min. c PhyAme incubated with Ins P6 for 120 min. d PhyAme incubated with Ins P6 for 300 min

60 min except the overall major product had become Ins (1,2,3)P3 and/or D/L-Ins(1,2,6)P3, and trace amounts of D/LIns(1,2)P2 and/or D/L-Ins(4,5)P2 and/or Ins(2,5)P2 had been produced. After 300 min of incubation, PhyAme had degraded all of the Ins P6 and Ins P5s. Ins(1,2,3)P3 and/or D/L-Ins(1,2,6)P3 were found as the major products in addition to D/L-Ins(1,2,3,4)P4 and D/L-Ins(1,2)P2 and/or D/ L-Ins(4,5)P2 and/or Ins(2,5)P2. The end products of Ins P6 degradation were determined by incubating excess protein with a limiting substrate concentration. The results of a gas chromatography-mass spectrometry analysis revealed that the end product is Ins(2)P.


Discussion Enzymatic activity and substrate specificity Under the conditions of our standard assay, the specific activity of PhyAme towards Ins P6 (269.3 U mg−1) falls within the range of other characterized PTP-like IPPases that have displayed a range of activities; from 12 U mg−1 (PhyB from S. lacticifex; Puhl et al. 2008a) to 668.11 U mg−1 (PhyAsr from S. ruminantium; Puhl et al. 2007). Similar to other characterized PTP-like IPPases, PhyAme activity was very sensitive to slight changes in I. The substantial charge carried by IPP substrates and the corresponding highly charged substrate-binding pocket of the enzyme likely exacerbates the salt dependence that is expected of all enzymes over some range (Puhl et al. 2007). For this reason, ionic strength was strictly controlled in all components of this study. Similar responses to ionic strength were reported for PTP-like IPPases from S. ruminantium; PhyAsr (Chu et al. 2004; Puhl et al. 2007), S. ruminantium subsp. lactilytica (Puhl et al. 2008b), and S. lacticifex (Puhl et al. 2008a). The ionic strength of rumen fluid depends on diet and has been reported in the range of 0.085 to 0.15 M (Wohlt et al. 1973; Koppolu and Clements, 2004). The observed effect of ionic strength on these enzymes may be an adaptation to the rumen environment. PhyAme exhibits a very narrow substrate specificity; this is similar to other known PTP-like IPPases (Puhl et al. 2007, 2008a, b) but unlike most non-PTP-like acid IPPases that exhibit a broad specificity for substrates with phosphate esters (Konietzny and Greiner, 2002). Among the IPPs tested, PhyAme has a slight catalytic preference for the pentakisphosphate substrates, consistent with most characterized PTP-like IPPases (Puhl et al. 2007, 2008a, b).

101

for the PhyAme-catalyzed hydrolysis of 1D-Ins(1,2,4,5,6) P5 (134.5 s−1 and 61.3 μM, respectively) and suggests that they are the same isomer. Similarly, the kcat and Km for the hydrolysis of the D/L-Ins(1,2,3,4,5)P5 intermediate produced by PhyAme were 135.8 s−1 and 61.5 μM, respectively. These values are most similar to the kcat and Km for the hydrolysis of 1D-Ins(1,2,3,5,6)P5 (135.3 s−1 and 61.8 μM, respectively). Thus, PhyAme initiates hydrolysis of Ins P6 primarily at the 1D-3 (60%) and 1D-4 (35%) phosphate positions. Other known PTP-like IPPases have been shown to exclusively hydrolyze (>90%) either the 1D3 or 1D-5-phosphate positions of Ins P6 (Puhl et al. 2007, 2008a, b). A similar mixed-position specificity was previously identified for non-PTP-like phytases cloned from the basidiomycete fungi Agrocybe pediades, Ceriporia sp., and Trametes pubescens (Lassen et al. 2001), but no distinction was made between the enantiomers of their Ins P6 hydrolysis products. Many acid IPPases have been found to liberate all five equatorial phosphate groups of Ins P6 (Konietzny and Greiner 2002), including the known PTP-like IPPases (Puhl et al. 2007, 2008a, b). PhyAme also displays the ability to cleave all five equatorial phosphates, resulting in a final product of Ins(2)P. HPIC and kinetic analysis indicate that,

Ins P6 hydrolysis pathway Based on the position of the first phosphate hydrolyzed, three types of phytases are recognized by the Enzyme Nomenclature Committee of the International Union of Biochemistry; i.e., 3-phytase (EC 3.1.3.8), 4-phytase (EC 3.1.3.26), and 5-phytase (EC 3.1.3.72). To date, most of the known phytases are 1D-3-, 1D-4-, or 1D-6-phytases (Konietzny and Greiner 2002). Here, HPIC analysis indicated that D/L-Ins(1,2,4,5,6)P5 and D/L-Ins(1,2,3,4,5)P5 were the major Ins P5 degradation products of PhyAmecatalyzed Ins P6 hydrolysis. To determine the specific Ins P5 isomers generated, kinetic parameters were determined for the possible Ins P5 hydrolysis products and compared to those for the actual intermediates generated by PhyAme. The kcat and Km for the hydrolysis of the D/L-Ins(1,2,4,5,6) P5 intermediate generated by PhyAme were 133.9 s−1 and 61.1 μM. These values are very similar to the kcat and Km

Fig. 4 Degradation pathways of Ins P6 by PhyAme. Larger arrows indicate major pathways, smaller arrows indicate minor pathways. Solid arrows represent routes verified by HPIC and kinetic data; open arrows designate possible routes of hydrolysis as predicted from HPIC data. Values (%) above respective major pathways indicate proportion of hydrolysis products generated by that route

102

following initial hydrolysis at the 1D-3- or 1D-4-phosphate positions, PhyAme follows distinct and specific routes of hydrolysis with each subsequent product. PhyAme can produce Ins(2)P via the routes indicated in Fig. 4. PhyAme predominantly degrades Ins P6 to Ins(2)P via: (A) 1D-Ins (1,2,4,5,6)P5, 1D-Ins(1,2,5,6)P4, 1D-Ins(1,2,6)P3, and 1DIns(1,2)P2 and (B) 1D-Ins(1,2,3,5,6)P5, 1D-Ins(1,2,3,6)P4, Ins(1,2,3)P3, and D/L-Ins(1,2)P2 (60% and 35% respectively). The two major pathways are nearly identical except in the order of removal of the 1D-3 phosphate, i.e., the first phosphate removed or the fourth removed, respectively. All characterized PTP-like IPPases to date use very ordered and stereospecific routes of Ins P6 hydrolysis. Moreover, with the characterization of PhyAme, this class includes representatives that can dephosphorylate the 1D-3, 4, or 5 phosphate positions of Ins P6, and could thus offer a convenient means of producing any number of specific lower IPPs. Further analysis of the relationship between specific residues or other structural features and the stereospecificity of members of this novel class of IPPase are currently underway by our group and ideally would lead to the ability to modify an enzyme from this class that could dephosphorylate Ins P6 to any desired IPP. Acknowledgements L. Brent Selinger receives funding from the Natural Sciences and Engineering Research Council (NSERC) and the Advanced Foods and Materials Network (AFMNet). Thanks to L. J. Yanke, Agriculture and Agri-Food Canada (Lethbridge, Alberta), for supplying the M. elsdenii cultures. Analysis of the isomers of the individual myo-inositol phosphate derivatives by N.-G. Carlsson, Chalmers University of Technology (Göteborg, Sweden), is also gratefully acknowledged.

References Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment search tool. J Mol Biol 215:403–410 Baldi P, Pollastri G (2003) The principled design of large-scale recursive neural network architectures-dag-rnns and the protein structure prediction problem. J Mach Learn Res 4:575–603 Batten GD, Lott JNA (1986) The influence of phosphorus nutrition on the appearance and composition of globoid crystals in wheat aleurone cells. Cereal Chem 63:14–18 Bendtsen JD, Nielson H, Von Heijne G, Brunak S (2004) Improved prediction of signal peptides: SignalP 3.0. J Mol Biol 340:783–795 Billington DC (1993) The inositol phosphates. Chemical synthesis and biological significance. VCH, New York, NY Brailoiu E, Miyamoto MD, Dun NJ (2003) Inositol derivatives modulate spontaneous transmitter release at the frog neuromuscular junction. Neuropharmacology 45:691–701 Brancaccio M, Legendre GG (1979) Megasphaera elsdenii Endocarditis. J Clin Microbiol 10:72–74 Cheng J, Randall AZ, Sweredoski MJ, Baldi P (2005) Scratch: a protein structure and structural feature prediction server. Nucl Acids Res 33:W72–W76 Chenna R, Sugawara H, Koike T, Lopez R, Gibson TJ, Higgins DG, Thompson JD (2003) Multiple sequence alignment with the clustal series of programs. Nucl Acids Res 31:3497–3500

Appl Microbiol Biotechnol (2009) 82:95–103 Cho CY, Koo SH, Wang Y, Callaway S, Hedrick S, Mak PA, Orth AP, Peters EC, Saez E, Montminy M, Schultz PG, Chanda SK (2006) Identification of the tyrosine phosphatase ptp-meg2 as an antagonist of hepatic insulin signaling. Cell Metab 3:367–378 Chu HM, Guo RT, Lin TW, Chou CC, Shr HL, Lai HL, Tang TY, Cheng KJ, Selinger BL, Wang AHJ (2004) Structures of Selenomonas ruminantium phytase in complex with persulfated phytate: DSP-phytase fold and mechanism for sequential substrate hydrolysis. Structure 12:2015–2024 Elsden SR, Volcani BE et al (1956) Properties of a fatty acid forming organism isolated from the rumen of sheep. J Bacteriol 72:681–689 Fassler J, Nadel C, Richardson N, McEntyre J, Schuler G, McGinnis S, Pongor S (2000) NCBI website. from http://www.ncbi.nlm.nih.gov/ Gasteiger E, Hoogland C, Gattiker A, Duvaud S, Wilkins MR, Appel RD, Bairoch A (2005) Protein identification and analysis tools on the expasy server. In: Walker JM (ed) The proteomics protocols handbook. Humana Press, New York, pp 571–607 Greer J (1981) Comparative model-building of the mammalian serine proteases. J Mol Biol 153:1027–1042 Greiner R, Konietzny U (1996) Construction of a bioreactor to produce special breakdown products of phytate. J Biotechnol 48:153–159 Greiner R, Alminger ML, Carlsson NG, Muzquiz M, Burbano C, Cuadrado C, Pedrosa MM, Goyoaga C (2002a) Pathway of dephosphorylation of myo-inositol hexakisphosphate by phytases of legume seeds. J Agric Food Chem 50:6865–6870 Greiner R, Farouk A, Alminger ML, Carlsson NG (2002b) The pathway of dephosphorylation of myo-inositol hexakisphosphate by phytate-degrading enzymes of different Bacillus spp. Can J Microbiol 48:986–994 Haikara A, Helander I (2006) Pectinatus, Megasphaera and Zymophilus. In: Dworkin M, Falkow S, Rosenberg E, Schleifer K-H, Stackebrandt E (eds) The prokaryotes: a handbook on the biology of bacteria, volume 4. Springer Science & Business Media, New York, pp 965–981 Hanakahi LA, Bartlet-Jones M, Chappell C, Pappin D, West SC (2000) Binding of inositol phosphate to DNA-pk and stimulation of double-strand break repair. Cell 102:721–729 Konietzny U, Greiner R (2002) Molecular and catalytic properties of phytate-degrading enzymes (phytases). Int J Food Sci Technol 37:791–812 Koppolu A, Clements LD (2004) Ruminal waste stream as a source of industrial chemicals. Resour Conserv Recycl 41:215–226 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685 Lassen SF, Breinholt J, Ostergaard PR, Brugger R, Bischoff A, Wyss M, Fuglsang CC (2001) Expression, gene cloning, and characterization of five novel phytases from four basidiomycete fungi: Peniophora lycii, Agrocybe Pediades, ceriporia sp., and Trametes pubescens. Appl Environ Microbiol 67:4701–4707 Mullaney EJ, Ullah AHJ (2003) The term phytase comprises several different classes of enzymes. Biochem Biophys Res Commun 312:179–184 Nakashima BA, McAllister TA, Sharma R, Selinger LB (2007) Diversity of phytases in the rumen. Microb Ecol 53:82–88 Nicholas KB, Nicholas HB Jr, Deerfield DW II (1997) GeneDoc: Analysis and visualization of genetic variation. from http://www. psc.edu/biomed/genedoc Ohkawa T, Ebisuno S, Kitagawa M, Morimoto S, Miyazaki Y, Yasukawa S (1984) Rice bran treatment for patients with hypercalciuric stones: experimental and clinical studies. J Urol 132:1140–1145 Orchiston EA, Bennett D, Leslie NR, Clarke RG, Winward L, Downes CP, Safrany ST (2004) Pten m-cbr3, a versatile and selective regulator of inositol 1,3,4,5,6-pentakisphosphate (ins (1,3,4,5,6)p-5)—evidence for Ins(1,3,4,5,6)p-5 as a proliferative signal. J Biol Chem 279:1116–1122

Appl Microbiol Biotechnol (2009) 82:95–103 Phillippy BQ, Bland JM (1988) Gradient ion chromatography of inositol phosphates. Anal Biochem 175:162–166 Pollastri G, Przybylski D, Rost B, Baldi P (2002) Improving the prediction of protein secondary structure in three and eight classes using recurrent neural networks and profiles. Proteins 47:228–235 Priefer U, Simon R, Pühler A (1984) Cloning with cosmids. In: Pühler A, Timmis KN (eds) Advanced molecular genetics. Springer, New York, pp 190–201 Puhl AA, Gruninger RJ, Greiner R, Janzen TW, Mosimann SC, Selinger LB (2007) Kinetic and structural analysis of a bacterial protein tyrosine phosphatase-like myo-inositol polyphosphatase. Protein Sci 16:1368–1378 Puhl AA, Greiner R, Selinger LB (2008a) Kinetics, substrate specificity, and stereospecificity of two new protein tyrosine phosphatase-like inositol polyphosphatases from Selenomonas lacticifex. Biochem Cell Biol 86:322–330 Puhl AA, Greiner R, Selinger LB (2008b) A protein tyrosine phosphatase-like inositol polyphosphatase from Selenomonas ruminantium subsp. lactilytica has specificity for the 5-phosphate of myo-inositol hexakisphosphate. Int J Biochem Cell Biol 40:2053–2064

103 Ruf JC, Ciavatti M, Gustafsson T, Renaud S (1994) In-vitro effect of D-myo-inositol 1,2,6-trisphosphate (PP-56) on aggregation of platelets from normal and diabetic rats—relationship to malondialdehyde release and phosphoinositide pathway. Can J Physiol Pharmacol 72:644–649 Sasakawa N, Sharif M, Hanley MR (1995) Metabolism and biological-activities of inositol pentakisphosphate and inositol hexakisphosphate. Biochem Pharmacol 50:137–146 Scott HW, Dehority BA (1965) Vitamin requirements of several cellulolytic bacteria. J Bacteriol 89:1169–1175 Shears SB (2001) Assessing the omnipotence of inositol hexakisphosphate. Cell Signal 13:151–158 Skoglund E, Carlsson NG, Sandberg AS (1998) High-performance chromatographic separation of inositol phosphate isomers on strong anion exchange columns. J Agric Food Chem 46:1877–1882 Wohlt JF-, Sniffen CJ, Hoover WH (1973) Measurement of protein solubility in common feedstuffs. J Dairy Sci 56:1052–1057 Yanke LJ, Bae HD, Selinger LB, Cheng K-J (1998) Survey of phytase activity in anaerobic rumen bacteria. Microbiol 144:1565–1573 Zhang Z, Song Y, Wang XL (2005) Inositol hexaphosphate-induced enhancement of natural killer cell activity correlates with suppression of colon carcinogenesis in rats. World J Gastroenterol 11:5044–5046