HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 30, 23146-23153, July 28, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/30/23146    most recent M002544200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Jedrzejas, M. J. Articles by Krishnasamy, G. Search for Related Content PubMed PubMed Citation Articles by Jedrzejas, M. J. Articles by Krishnasamy, G. Social Bookmarking                      What's this?

Mechanism of Catalysis of the Cofactor-independent Phosphoglycerate Mutase from Bacillus stearothermophilus

CRYSTAL STRUCTURE OF THE COMPLEX WITH 2-PHOSPHOGLYCERATE^*

Mark J. Jedrzejas^§, Monica Chander^¶, Peter Setlow^¶, and Gunasekaran Krishnasamy

From the  Department of Microbiology, University of Alabama at Birmingham, Birmingham, Alabama 35294 and the ^¶ Department of Biochemistry, University of Connecticut Health Center, Farmington, Connecticut 06030

Received for publication, March 26, 2000, and in revised form, April 7, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

The structure of the complex between the 2,3-diphosphoglycerate-independent phosphoglycerate mutase (iPGM) from Bacillus stearothermophilus and its 3-phosphoglycerate substrate has recently been solved, and analysis of this structure allowed formulation of a mechanism for iPGM catalysis. In order to obtain further evidence for this mechanism, we have solved the structure of this iPGM complexed with 2-phosphoglycerate and two Mn²⁺ ions at 1.7-Å resolution. The structure consists of two different domains connected by two loops and interacting through a network of hydrogen bonds. This structure is consistent with the proposed mechanism for iPGM catalysis, with the two main steps in catalysis being a phosphatase reaction removing the phosphate from 2- or 3-phosphoglycerate, generating an enzyme-bound phosphoserine intermediate, followed by a phosphotransferase reaction as the phosphate is transferred from the enzyme back to the glycerate moiety. The structure also allowed the assignment of the function of the two domains of the enzyme, one of which participates in the phosphatase reaction and formation of the phosphoserine enzyme intermediate, with the other involved in the phosphotransferase reaction regenerating phosphoglycerate. Significant structural similarity has also been found between the active site of the iPGM domain catalyzing the phosphatase reaction and Escherichia coli alkaline phosphatase.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Phosphoglycerate mutases (PGMs)¹ catalyze three types of reactions including interconversion of 1,3-phosphoglycerate and 2,3-phosphoglycerate (23PGA) and of 3-phosphoglycerate (3PGA) and 2-phosphoglycerate (2PGA) as well as synthesis of 3PGA from 23PGA (1). There are two distinct types of PGMs, bisphosphoglycerate mutase and monophosphoglycerate mutase, and only the first type has the ability to perform all of the reactions listed above, while monophosphoglycerate mutases catalyze primarily the interconversion of 3PGA and 2PGA in both glycolysis and gluconeogenesis. There are also two classes of monophosphoglycerate mutases that are distinguished by their requirement for 23PGA for catalysis (1, 2). The PGMs requiring 23PGA for catalysis are termed 23PGA-dependent and are the predominant PGM in mammals, yeast, and some bacteria. They also have the ability to perform all reactions noted above but at significantly different rates. The monophosphoglycerate mutases that do not require DPG for catalysis are termed 23PGA-independent (iPGMs) and are the predominant PGM in plants and some other bacteria, including endospore-forming Gram-positive bacteria and their close relatives; iPGMs, unlike cofactor-dependent phosphoglycerate mutases and bisphosphoglycerate mutases, can only carry out the interconversion of 2PGA and 3PGA (1, 3, 4). The two classes of monophosphoglycerate mutases are extremely different in amino acid sequence, catalytic mechanism, and structure, both tertiary and quaternary. Despite the differences between these two types of PGMs, the iPGMs all have very conserved amino acid sequences, as do the cofactor-dependent phosphoglycerate mutases; the cofactor-dependent phosphoglycerate mutases are also very similar in sequence to bisphosphoglycerate mutases, reflecting their likely similarity in structure and in function (1, 5). Another difference between the cofactor-dependent phosphoglycerate mutases and iPGMs is that there is no metal ion requirement for catalysis by cofactor-dependent phosphoglycerate mutases (or bisphosphoglycerate mutases), while at least some iPGMs, including the enzymes of endospore-forming bacteria, have an absolute and specific requirement for Mn²⁺ ions for catalysis, which makes the activity of these enzymes exquisitely sensitive to pH (1, 4, 6-8). This pH sensitivity is physiologically relevant, since it allows for the regulation of iPGM activity during different parts of the developmental cycle of these organisms. During sporulation, the pH in the developing spore falls to ~6.5, which decreases the activity of this cell's iPGM, allowing for accumulation of 3PGA, which is stored in the dormant spore; this 3PGA is then utilized for ATP synthesis during spore germination, when the pH in the germinating spore rapidly rises to 7.5-8 (9-12). The iPGMs of non-spore-forming organisms that are closely related to spore formers also require Mn²⁺ for catalysis, suggesting that all iPGMs are related evolutionarily to an ancestral metal ion-requiring iPGM (4). However, the iPGMs of some organisms (e.g. Escherichia coli, whose iPGM contains Mn²⁺ (2), and plants for which the Mn²⁺ requirement is not yet clear) may have lost the extreme pH sensitivity of their enzyme activity during evolution (13), since this pH dependence may have no physiological importance.

The structure and general mechanism of catalysis of cofactor-dependent phosphoglycerate mutases were established a number of years ago (1, 14), but until recently there was no corresponding information on iPGMs, although over 20 years ago crystals were obtained of at least one iPGM that might have allowed structural studies using x-ray crystallography (see appendix of Ref. 15). However, the structure of Bacillus stearothermophilus iPGM complexed with its 3PGA substrate as well as with two Mn²⁺ ions was recently solved by x-ray crystallography at 1.9-Å resolution (5). Analysis of this structure located the two Mn²⁺ ions and the 3PGA in the enzyme's active site and allowed formulation of a likely catalytic mechanism that utilizes a phosphoserine-enzyme intermediate. In addition to information obtained by analysis of the structure of the 3PGA-iPGM complex, the proposed mechanism was also consistent with the effects on enzyme activity of a number of site-directed mutations altering residues in or near the active site. However, another important source of information on which the proposed catalytic mechanism was based were studies in which the 2PGA substrate/product was modeled in the active site of the enzyme. Since the latter information was obtained only from modeling studies, we decided to obtain further information relevant to our understanding of the mechanism for this iPGM by determining the crystal structure of B. stearothermophilus iPGM complexed with 2PGA. With this latter information, there will be two structures available, which have both substrates/products bound in the active site of this iPGM, thus closing the catalytic cycle originating with 3PGA and finishing with 2PGA or vice versa. The availability of the structural details of 2PGA binding in the active site of the enzyme, in contrast to the modeled information previously available, further allowed definitive assignment of specific mechanistic functions to different parts of the enzyme.

Since previous work had indicated that there is some limited sequence similarity between the metal binding regions of E. coli alkaline phosphatase (AP) and iPGMs (16),² we also manually compared the structure of the active site of B. stearothermophilus iPGM with that of the active site of E. coli AP and have found striking structural similarity, but limited only to these regions of the two proteins. Since the catalytic mechanism for E. coli AP has been studied in great detail (18-19), the similarity in the part of the active site structure between AP and iPGM is discussed in view of the mechanism proposed for iPGMs.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Crystallization and X-ray Data Collection-- B. stearothermophilus iPGM was obtained as described previously (20). The enzyme was dialyzed against 10 mM Hepes-KOH buffer, pH 7.4, 100 mM KCl, 5 mM MnCl₂, 1 mM dithiothreitol, 50 µM EDTA and concentrated to 25 mg/ml. The 2PGA complex with iPGM was crystallized at room temperature using the hanging drop vapor diffusion method (21) and 24-well Linbro culture plates. The crystallization drops contained equal volumes (1-2 µl) of the enzyme, 150 mM 2PGA in water, and the reservoir solution. The drops were equilibrated against 0.3 ml of reservoir solution, which contained 2.1 M ammonium sulfate and 100 mM sodium citrate, pH 6.7, 25 mM zinc acetate, 20 mM CsCl₂, 3% polyethylene glycol 200, and 1 mM -mercaptoethanol. Diffraction quality crystals were obtained in less than a week, and they diffracted x-rays to 1.7-Å resolution; the presence of 2PGA was essential for crystallization. For the diffraction experiments, the crystals were cryoprotected using 23% ethylene glycol (v/v), 2.5 M ammonium sulfate, and 100 mM sodium citrate, pH 6.7, and flash frozen at 170 °C in a nitrogen stream using a Cryostream Cooler (Oxford Cryosystems, Oxford, UK). The x-ray diffraction data were collected at the Argonne National Laboratory, Advanced Photon Source, beamline 19-ID of the Structural Biology Center at an x-ray wavelength of 1.1 Å, and a 1.7-Å resolution diffraction data set was obtained. The data were processed and scaled using the HKL2000 package (22). The unit cell of the crystals was determined to be as follows: a = 58.41 Å, b = 205.69 Å, c = 123.71 Å with the orthorhombic space group C222₁. The crystal volume per unit of molecular weight was consistent with one molecule of the enzyme in the asymmetric unit and a solvent content of 63% (23).

Structure Determination and Refinement-- The space group as well as the cell parameters for the crystals of the iPGM·2PGA·Mn²⁺ complex were isomorphous to those of the iPGM·3PGA·Mn²⁺ complex (5, 20), and thus molecular replacement was not needed for the initial phase determination. The rigid body refinements (24) of the protein component of the iPGM·3PGA·Mn²⁺ complex structure against the diffraction data set from the iPGM·2PGA·Mn²⁺ crystals was sufficient. The R_free procedure, using a data set containing 5% of the total unique reflections, was used to judge the refinement progress (25). The R_cryst after such refinements decreased to 0.35, whereas the R_free was 0.39. After manual inspection and evaluation of the model using the graphics program QUANTA (26), adjustments to the protein model were made followed by refinements using the X-plor least-squares procedures (24), after which the R_cryst decreased to 0.30. Using the 2F_o  F_c and F_o  F_c electron density maps, the geometry of the structure was reexamined using a Ramachandran plot (27), and the model was corrected manually as above. The bulk solvent correction and the overall B-factor corrections were then applied to the data during the subsequent structure refinement using the molecular dynamics procedures of the X-plor protocol (24). Two Mn²⁺ ions were located between the two domains of the structure using [vert]F_o  F_c[vert] electron density maps with a 6 cut-off. The simulated annealing omit maps (25) calculated in the region where the 2PGA was expected to bind (5) were used to examine as well as to locate and place the 2PGA in the enzyme's active site. The topology and parameter files for 2PGA were generated using standard files based on the energetically expected values for the distances and angles between atoms of 2PGA provided in the Quanta program (26). Subsequently, another round of least-squares and molecular dynamics refinements was performed followed by individual B-factor refinements (25). Solvent molecules were added to the model based on the inspection of the 3 level [vert]F_o  F_c[vert] map and the analysis of the contacts with either the protein or with the 2PGA. The stereochemistry of the model examined using the Ramachandran plot (27) does not contain any residues outside of the energetically allowed areas. Figures were drawn using Ribbons software (28).

Sequence Alignment-- The conserved regions of the B. stearothermophilus iPGM and E. coli AP sequences were aligned based on the structural alignment of the active site residues of both enzymes. All manipulations as well as a least-squares refinement of the structure-based alignment were performed using the graphics program O (29). All 14 segments encompassing 149 amino acid residues were superimposed with a root mean square deviation of 1.67 Å. However, all sequence-based alignment methods that were tried failed to recognize most of these similarities between these two enzymes.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Structure of the 2PGA·iPGM Complex-- The crystal structure of the B. stearothermophilus iPGM·2PGA complex was solved at 1.7-Å resolution using the structure of the iPGM·3PGA complex as a model (5, 20). The iPGM·2PGA structure consists of residues from Lys³ to Val⁵¹⁰, two Mn²⁺ ions, 2PGA, and 172 water molecules (Fig. 1). Ser² and the last residue, Lys⁵¹¹, were not included in the final model due to the lack of their corresponding electron density. The Ramachandran plot analysis (27) showed no residues, even in loop areas, in the energetically disallowed regions. The final refinement statistics are provided in Table I.

View larger version (65K): [in this window] [in a new window]   Fig. 1.   Ribbon drawing of the B. stearothermophilus iPGM·2PGA·Mn2+ complex. The Mn2+ ions and the 2PGA are shown in a ball and stick fashion. Domains A and B as well as their connecting loops are marked.

As found previously for the iPGM·3PGA·Mn²⁺ complex, the B. stearothermophilus iPGM·2PGA·Mn²⁺ complex assumes a compact, globular shape with two domains, A (Lys³-Ala⁷¹ and Thr³¹⁶-Val⁵¹⁰) and B (Leu⁷⁹-Tyr³⁰⁵). Both domains have similar folds consisting of a central eight-stranded -sheet structure flanked by -helices, eight in domain A and nine in domain B. Overall, this iPGM assumes an / type structure (5, 30). Both domains have complementary surfaces, and they interact through a network of hydrogen bond interactions and are connected by two loops, Gly⁷²-Ser⁷⁸ and Val³⁰⁶-Asn³¹⁵. There is a well defined solvent-, metal ion-, and substrate accessible cleft between the two domains. The sides of this cleft are lined by residues primarily from the loops connecting -helices and -strands of both domains. Both Mn²⁺ ions, Mn1 and Mn2, and 2PGA are located in this cleft (Fig. 1). The phosphate of 2PGA and both Mn1 and Mn2 interact with residues of domain A, and the phosphate interacts only with Arg²⁶¹ from domain A, while the glycerate end of the substrate interacts with residues only from domain B. Both Mn²⁺ binding sites are located in a crevice between strands of domain A. The ordered water molecules are located either on the enzyme's surface or inside the interdomain cleft.

Active Site Environment-- The active site is well defined and is composed of 15 amino acid residues interacting with 2PGA, Mn1, or Mn2 (Fig. 2A). The amino acid residues originate from both domains: Asp¹², Ser⁶², Lys³³⁶, Asp⁴⁰³, His⁴⁰⁷, Asp⁴⁴⁴, His⁴⁴⁵, His⁴⁶² in domain A and His¹²³, Arg¹⁵³, Asp¹⁵⁴, Arg¹⁸⁵, Arg¹⁹¹, Arg²⁶¹, and Arg²⁶⁴ in domain B (Fig. 2A and Table II). The residues interacting directly with Mn1 are Asp⁴⁰³, His⁴⁰⁷, and His⁴⁶² (average Mn1 to ligand distance of 2.17 Å), and residues interacting directly with Mn2 are Asp¹² (both NE1 and NE2 atoms), Ser⁶², His⁴⁴⁴, and His⁴⁴⁵ (average Mn2 to ligand distance of 2.37 Å). Mn1 also interacts with two phosphate oxygen atoms, O1P and O3P, of 2PGA (average distance of 2.46 Å) (Fig. 2A and Table III). The common coordinating hard ligands of Mn²⁺ ions are carboxylates of aspartate residues and nitrogens of histidines, although other ligands are sometimes encountered; these include oxygens of serine residues and phosphate groups on substrates (31, 32). The geometry of ligand coordination to both Mn²⁺ ions is distorted square pyramidal, which is typical for Mn²⁺ ions (Fig. 2A and Table III) (32, 33). The apex coordination is occupied by the NE2 nitrogen atom of His⁴⁶² for Mn1 and by the OG oxygen atom of Ser⁶² for Mn2. Additional interactions that are important for catalysis (5) are the bidentate interaction of Arg²⁶¹ with the phosphate group oxygens O2P and O4P (average distance of 2.81 Å) as well as the bidentate interaction of Arg²⁶⁴ with the substrate carboxyl group (average distance of 2.87 Å) (Table II). The O3P phosphate group's oxygen atom semibridges the two Mn²⁺ ions with a Mn2 to O3P distance of 3.33 Å and an O3P to Mn1 distance of 2.32 Å; the Mn1-O3P-Mn2 angle is 134.0° (Fig. 2A, Table III).

View larger version (20K): [in this window] [in a new window]   Fig. 2.   A, a stereo view of the active site residues plus the 2PGA and Mn2+ ions. A total of 15 residues constitute the active site of iPGM. B, the superposition of 2PGA and 3PGA structures based on the x-ray structures of their respective complexes with iPGM. One-letter amino acid codes are shown.

Comparison with the Structure of the iPGM·3PGA·Mn²⁺ Complex-- There are no global differences between the structures of the iPGM·2PGA complex and the iPGM·3PGA complex; the fold as well as the arrangement of the domains is, as expected, the same (5). The active site geometry is similar, although there are differences around the carboxyl end of the 2PGA including the precise positions of residues Arg¹⁵³, Arg¹⁵⁴, and Arg¹⁸⁵, all from domain B. Mn1 and Mn2 and their ligands as well as the residues interacting with the phosphate of 2PGA are also in similar positions and possess similar side chain orientation as in the 3PGA complex. As proposed earlier (5), 2PGA is located in the same area of the enzyme's active site as is 3PGA in the 3PGA·iPGM complex; the differences are in the positioning and orientation of the glycerate moiety (Fig. 2B). In general, the position of the phosphate and carboxyl moieties of 2PGA and 3PGA are the same. The O3 oxygen of 2PGA is placed in proximal position to the O2 atom of 3PGA, and both atoms point in the direction of Asp¹⁵⁴. In the iPGM·2PGA complex the O3 to OD1 Asp¹⁵⁴ distance is 2.62 Å, while in the 3PGA complex the distance between the corresponding oxygen atom, O2, and OD1 of Asp¹⁵⁴ is 2.75 Å. The largest differences between the 3PGA and 2PGA are in the positions for the C2 (from 3PGA) and C3 (from 2PGA) carbon atoms, which differ by 1.21 Å and 1.31 Å, respectively. Carbon atoms C2 for 2PGA and C3 for 3PGA are bridging atoms with the phosphate group, and their positions are displaced only by 0.63 Å, suggesting that the placement of the phosphate group is very important and is a stringent requirement for catalysis (Fig. 2B).

Comparison of the iPGM Structure with That of Alkaline Phosphatase-- Comparison of the amino acid sequences of iPGMs from different sources, including bacteria (both spore formers and non-spore formers) and plants, has shown that all active site residues are conserved in these enzymes (5). Comparison of iPGM sequences with the sequences of several members of the alkaline phosphatase family has also shown that the residues directly involved in metal binding are also conserved between these two types of enzymes (16).² Alkaline phosphatases contain two Zn²⁺ ions, while those iPGMs in which metal ions have been identified contain Mn²⁺ ions, with two Mn²⁺ ions identified in the iPGM of B. stearothermophilus (2, 4, 5, 18).² It appears that all iPGMs are likely to be metalloenzymes (34, 35), although the presence of Zn²⁺ instead of Mn²⁺ in some of these enzymes has not yet been excluded.²

The conservation of metal-binding residues in APs and iPGMs suggested that there might also be some structural similarities between these two classes of proteins. However, structural similarity comparisons using the DALI algorithm (36) of our iPGM structure with structures in the Protein Data Bank gave relatively low scores, indicating that iPGM has no significant overall structural similarity to known proteins including E. coli AP (37, 38) (Fig. 3a). Despite this initial lack of success, the AP and iPGM structures were aligned manually, focusing on the metal-binding residues using the graphics software QUANTA (26). To our surprise, the metal-binding residues aligned very well (Fig. 3, b and c). In addition, the position of the phosphate group in the active site of E. coli AP (37) nearly overlaps with that of the phosphate group of the 2PGA bound to B. stearothermophilus iPGM, and the residue in AP that is phosphorylated during the reaction (Ser¹⁰²) is also present in a similar location in iPGM (Ser⁶²). In AP, this Ser residue is part of a triad, Asp¹⁰¹-Ser¹⁰²-Glu¹⁰³, which is conserved in all APs; a similar triad, Asn⁶¹-Ser⁶²-Glu⁶³, is also conserved among all iPGMs (5, 16).

View larger version (46K): [in this window] [in a new window]   Fig. 3.   Comparison of the structures of all of (a) or the active sites of (b) B. stearothermophilus iPGM (green) and E. coli AP (red) (Protein Data Bank code 1ALK) or the amino acid sequences of B. stearothermophilus iPGM and E. coli AP (c). In a, the domains were aligned based only on the C backbone atoms of the central -sheet regions of both enzymes using the least-squares procedures implemented in the program O (29). In b, the active sites were aligned based on the coordinates of corresponding resides from both active sites. In c, the sequences were aligned based on the three-dimensional structural alignment of both enzymes, with iPGM sequences on top and AP sequences on the bottom. Only sequence segments with good structural overlap are shown. The metal-binding active site residues, the catalytic triad, Asn61-Ser62-Glu63, as well as the Arg261 residue maintaining the bidentate interaction with the phosphate group are colored in red. Other identical aligned residues are highlighted in yellow.

Both iPGM and AP are / proteins, although AP is a homodimer. The monomers of each protein consist of a central -sheet surrounded by -helices. When the AP backbone is superimposed on domain A of our iPGM structure (where the metal binding residues reside), the central -sheet and the surrounding -helices superimpose with a significant degree of overlap (Fig. 3a). However, the areas far from the central -sheet of both AP and iPGM overlap poorly if at all, and the structural similarity between these two proteins was significant only in the active sites and the metal binding regions (Fig. 3b).

Catalytic Mechanisms of E. coli AP and B. stearothermophilus iPGM-- The structural similarity between the metal binding regions of the active sites of E. coli AP and B. stearothermophilus iPGM suggested that the reaction mechanisms of these two enzymes would also be similar. E. coli AP is a phospomonoesterase, although it can catalyze a phosphotransferase reaction under some conditions, and the catalytic mechanism of this enzyme is well established (18). The active site of the enzyme contains two Zn²⁺ ions and one Mg²⁺ ion, although the latter ion does not seem to have any role in catalysis (Fig. 3b) (18, 37). Asp³²⁷ (bidentate ligand), His³³¹, His⁴¹², and one of the phosphate oxygen atoms of AP coordinate with Zn1; Asp⁵¹, Asp³⁶⁹, Asp³⁷⁰, and another oxygen atom of the phosphate coordinate with Zn2, forming a phosphate bridge between Zn1 and Zn2. An arginine residue, Arg¹⁶⁶, forms a bidentate hydrogen bond with the remaining two phosphate oxygen atoms. The catalytic Ser¹⁰² initially coordinates with Zn2, and catalysis has been proposed to involve an intermediate five-coordinate phosphate group with trigonal bipyramidal geometry. The fifth position is occupied initially by Ser¹⁰² coordinating with the phosphate, and this is replaced with a water molecule later in the reaction. Throughout catalysis, the bidentate interaction between the phosphate oxygen atoms and Arg¹⁶⁶ is maintained and contributes to the known retention of phosphate configuration (39). Upon phosphorylation of the active site Ser¹⁰² residue, the tetrahedral environment of the phosphorus atom becomes trigonal bipyramidal. In the enzyme-substrate complex, one of the phosphate oxygen atoms coordinates with both metal ions by forming a µ-bridge.

A catalytic mechanism for iPGMs has been proposed based on the structure of the B. stearothermophilus iPGM-3PGA complex and studies modeling 2PGA in the iPGM active site; results from site-directed mutagenesis studies were consistent with this mechanism (5). The analysis of the structure of the iPGM·2PGA complex supports this proposed mechanism, which had initially relied on structural information for the iPGM·2PGA complex obtained by modeling and energetic refinement methods. One important piece of new information provided in the structure of the iPGM·2PGA complex is that, as predicted, the glycerate moiety does not have to move significantly to allow for catalysis (Fig. 2B). The glycerate parts of both 2- and 3PGA occupy exactly the same space in the active site; therefore, it is likely that positional isomerization (reorientation of glycerate without affecting its interaction with Mn1 and the bidentate interaction of its carboxyl group with the enzyme) of the glycerate, during which the bidentate interaction of its carboxyl group with Arg²⁶⁴ as well as the interaction of either O2 (3PGA) or O3 (2PGA) with Mn1 are maintained, can take place. In the conversion of 3PGA to 2PGA, this positional isomerization brings the O2 oxygen ion into close proximity with the phosphate group of the phosphoserine intermediate after the hydrogen of the O2H group is abstracted by Asp¹⁵⁴.

The major intermediates in the proposed catalytic mechanism have been previously described in detail (5) and are as follows (Figs. 2a and 4). (a) 2PGA binds in the active site of iPGM, during which the phosphate oxygen atoms O1P and O3P coordinate to Mn1; the two remaining phosphate oxygen atoms, O2P and O4P, coordinate with Arg²⁶¹ in a bidentate fashion. Ser⁶² coordinates with Mn2 and the O2P oxygen atom of the phosphate group, and the phosphorus atom is 3.18 Å from the OG oxygen atom of Ser⁶² (Fig. 4B, Table II). (b) Ser⁶² moves to occupy the fifth coordination site of the phosphorus atom, which temporarily has a trigonal bipyramidal geometry. This results in the formation of a phosphoserine intermediate with breakage of the bond between the phosphate and the glycerate part of the substrate. During this event, the phosphate group maintains its bidentate interaction with Arg²⁶¹, and the carboxyl group of glycerate maintains its bidentate interaction with Arg²⁶⁴ (Fig. 4B). (c) The newly formed negatively charged oxygen atom, O2, of the glycerate coordinates with Mn1, and Asp¹⁵⁴ abstracts a proton from the O3 oxygen atom of the glycerate. The glycerate undergoes repositioning (positional isomerization), which brings the O3 oxygen atom close to the phosphorus atom (Fig. 4C). (d) The repositioned O3 oxygen occupies the position of the previous phosphomonoester oxygen of the substrate and interacts with the phosphoserine intermediate, which again generates a trigonal bipyramidal phosphate group; this leads to breaking of the phosphoserine ester bond and transfer of the phosphate group to the O3 oxygen of the glycerate. The newly formed 3PGA then moves away from Ser⁶², which remains coordinated with Mn2 (as it likely was during the whole catalytic process) (Fig. 4D). (e) Finally, the 3PGA·iPGM complex dissociates into enzyme and 3PGA with the help of the ordered active site water molecule, Wat63, which is polarized by Mn1 into OH and H⁺. The Wat63 hydroxyl group, OH, probably replaces the 3PGA ester oxygen in coordination with the Mn1 ion, and the remaining H⁺ neutralizes the charge on the O2 oxygen of the leaving 3PGA. The OD2 oxygen atom of Asp⁴⁰³ probably completes the square-pyramidal coordination geometry of Mn1 (Fig. 4E). The exchange of protons with the microenvironment probably completes the cycle; one proton originating with the structured Wat63 molecule leaves with the 3PGA product while the proton abstracted by Asp¹⁵³ is exchanged with the water microenvironment and ultimately returned to the OH coordinating with Mn1. The reaction in the reverse direction from 3PGA to 2PGA is equally probable.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   The proposed catalytic mechanism for B. stearothermophilus iPGM. Catalysis consists of both phosphatase and a phosphotransferase component. The major intermediates in the proposed mechanism for conversion of 2PGA to 3PGA are binding of 2PGA in the active site (rendition based on the x-ray coordinates) (A); formation of a phosphoserine intermediate (phosphatase component) (B); repositioning of the glycerate moiety (C); transfer of the phosphate group from the serine residue to the repositioned glycerate to form 3PGA (phosphotransferase component) (3PGA complex rendition is based on the x-ray coordinates) (D); and dissociation of 3PGA from the enzyme (E).

Different Roles of Domains A and B in Catalysis-- As described above, the iPGM reaction has two main steps, the first being a phosphatase activity generating a phosphoserine intermediate and the second being a phosphotransferase reaction including positional isomerization of the glycerate moiety and transfer of the phosphate from the phosphoserine back to the glycerate. Based on the iPGM structures with either 3PGA (5) or 2PGA (this work), it can be seen that domain B is primarily involved in the phosphotransferase step of the reaction (note that domain B residues are not involved in Mn²⁺ binding) (Fig. 2A). In contrast, the first step of the reaction, the phosphatase activity, which removes the phosphate group from the phosphoglycerate and transfers it to Ser⁶², utilizes primarily amino acid residues interacting with both Mn²⁺ ions, all of which reside in domain A, as is also the case for AP. Indeed, only the structure of domain A's part of iPGM's active site is structurally similar to the AP active site.

During the course of iPGM catalysis, Mn1 coordinates with the O2 atom (previously O1P) of the glycerate part of 2PGA. This interaction has a dual function; it assists with movement of O1P (O2) away from the phosphate group of the phosphoserine intermediate, and it also assists in the proper repositioning of the glycerate part of the substrate by maintaining the Mn1-O2 interaction. This interaction, together with the bidentate interaction with the glycerate's carboxyl group, properly positions the glycerate and its O3 oxygen for nucleophilic attack on the phosphorus of the phosphoserine intermediate.

The primary residues in domain B interacting with the substrate are Asp¹⁵⁴ (hydrogen removal from the O3-H group of the glycerate), Arg²⁶⁴ (maintenance of the bidentate interaction with glycerate's carboxyl group), and Arg²⁶¹ (bidentate positioning and maintenance of the configuration of the phosphate group). There is no structural similarity between this domain and any part of AP. However, Arg²⁶¹ of iPGM has a corresponding residue in AP, Arg¹⁶⁶, which also maintains a bidentate interaction with the phosphate in the AP active site (37) (Fig. 3, b and c).

Activity of the Enzyme in the Crystallization Conditions-- The free energy change, G⁰, for the reaction catalyzed by PGM is ~+4.7 kJ/mol (40), although this value can vary somewhat depending on the reaction conditions. Using this value of G⁰, the equilibrium constant for 3PGA conversion to 2PGA is ~0.15, indicating that 2PGA mixed with PGM should be largely converted to 3PGA. However, in our crystallization conditions for the iPGM·2PGA complex, we saw only 2PGA in the enzyme's active site; we also saw only 3PGA in the active site of the crystals of the iPGM·3PGA complex (5). The reason for these latter findings appears to be the inactivity of the enzyme under the crystallization conditions, since this was less than 0.01% of maximum activity (data not shown). The reasons for enzyme inactivity are not clear, but probably include the crystallization temperature (B. stearothermophilus iPGM has maximum activity at 65 °C), the low pH (4), and the presence of a variety of salts including Zn²⁺ ions, which inhibit iPGM.³ However, the presence of Mn²⁺ but not Zn²⁺ ions in the iPGM active site was clearly established based on the crystallographic analyses and refinements. The square pyramidal coordination geometry of the active site metal ions is also indicative of the presence of Mn²⁺, since Zn²⁺ would have tetrahedral geometry (17).

	ACKNOWLEDGEMENT

Use of the Argonne National Laboratory Structural Biology Center beamline 19-ID at the Advanced Photon Source is acknowledged for the diffraction data collection.

	FOOTNOTES

^* This work was supported in part by a supplement to National Institutes of Health Grant GM19698 (to P. S.) for the determination of high resolution structures (to P. S. and M. J. J.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The atomic coordinates and the structure factors (code 2ejj) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).

^§ To whom correspondence should be addressed: Dept. of Microbiology, University of Alabama at Birmingham, 933 19th St. S., CHSB-19 Rm. 545, Birmingham, AL 35294-2041. Tel.: 205-975-7627; Fax: 205-975-5424; E-mail: jedrzejas@uab.edu.

Published, JBC Papers in Press, April 10, 2000, DOI 10.1074/jbc.M002544200

² M. J. Jedrzejas and P. Setlow, submitted for publication.

³ M. Chander and P. Setlow, unpublished results.

	ABBREVIATIONS

The abbreviations used are: PGM, phosphoglycerate mutase; 23PGA, 2,3-phosphoglycerate; AP, alkaline phosphatase; iPGM, cofactor-independent phosphoglycerate mutase; 2PGA, 2-phosphoglycerate, 3PGA, 3-phosphoglycerate.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				B. Macek, I. Mijakovic, J. V. Olsen, F. Gnad, C. Kumar, P. R. Jensen, and M. Mann The Serine/Threonine/Tyrosine Phosphoproteome of the Model Bacterium Bacillus subtilis Mol. Cell. Proteomics, April 1, 2007; 6(4): 697 - 707. [Abstract] [Full Text] [PDF]

				M. Nukui, L. V. Mello, J. E. Littlejohn, B. Setlow, P. Setlow, K. Kim, T. Leighton, and M. J. Jedrzejas Structure and Molecular Mechanism of Bacillus anthracis Cofactor-Independent Phosphoglycerate Mutase: A Crucial Enzyme for Spores and Growing Cells of Bacillus Species Biophys. J., February 1, 2007; 92(3): 977 - 988. [Abstract] [Full Text] [PDF]

				Y. Zhang, J. M. Foster, S. Kumar, M. Fougere, and C. K. S. Carlow Cofactor-independent Phosphoglycerate Mutase Has an Essential Role in Caenorhabditis elegans and Is Conserved in Parasitic Nematodes J. Biol. Chem., August 27, 2004; 279(35): 37185 - 37190. [Abstract] [Full Text] [PDF]

				C. S. Bond, M. F. White, and W. N. Hunter High Resolution Structure of the Phosphohistidine-activated Form of Escherichia coli Cofactor-dependent Phosphoglycerate Mutase J. Biol. Chem., January 26, 2001; 276(5): 3247 - 3253. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/30/23146    most recent M002544200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Jedrzejas, M. J. Articles by Krishnasamy, G. Search for Related Content PubMed PubMed Citation Articles by Jedrzejas, M. J. Articles by Krishnasamy, G. Social Bookmarking                      What's this?