• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of biochemjBJ Latest papers and much more!
Biochem J. Jan 1, 2007; 401(Pt 1): 39–47.
Published online Dec 11, 2006. Prepublished online Aug 30, 2006. doi:  10.1042/BJ20061066
PMCID: PMC1698673

Crystal structure of human phosphoribosylpyrophosphate synthetase 1 reveals a novel allosteric site

Abstract

PRPP (phosphoribosylpyrophosphate) is an important metabolite essential for nucleotide synthesis and PRS (PRPP synthetase) catalyses synthesis of PRPP from R5P (ribose 5-phosphate) and ATP. The enzymatic activity of PRS is regulated by phosphate ions, divalent metal cations and ADP. In the present study we report the crystal structures of recombinant human PRS1 in complexes with SO42− ions alone and with ATP, Cd2+ and SO42− ions respectively. The AMP moiety of ATP binds at the ATP-binding site, and a Cd2+ ion binds at the active site and in a position to interact with the β- and γ-phosphates of ATP. A SO42− ion, an analogue of the activator phosphate, was found to bind at both the R5P-binding site and the allosteric site defined previously. In addi-tion, an extra SO42− binds at a site at the dimer interface between the ATP-binding site and the allosteric site. Binding of this SO42− stabilizes the conformation of the flexible loop at the active site, leading to the formation of the active, open conformation which is essential for binding of ATP and initiation of the catalytic reaction. This is the first time that structural stabilization at the active site caused by binding of an activator has been observed. Structural and biochemical data show that mutations of some residues at this site influence the binding of SO42− and affect the enzymatic activity. The results in the present paper suggest that this new SO42−-binding site is a second allosteric site to regulate the enzymatic activity which might also exist in other eukaryotic PRSs (except plant PRSs of class II), but not in bacterial PRSs.

Keywords: allosteric site, nucleotide synthesis, oligomerization, PRPP synthetase, regulation mechanism
Abbreviations: PRPP, phosphoribosylpyrophosphate; PRS, PRPP synthetase; bsPRS, Bacillus subtilis PRPP synthetase; hPRS, human PRPP synthetase; mjPRS, Methanocaldococcus jannaschii PRPP synthetase; PPi, pyrophosphate; R5P, ribose 5-phosphate; rmsd, root mean square deviation; mATP, α,β-methylene ATP; mADP, α,β-methylene ADP

INTRODUCTION

PRPP (phosphoribosylpyrophosphate) synthetases (ATP:D-ribose-5-phosphate pyrophosphotransferase; EC 2.7.6.1) are a family of enzymes that catalyse the synthesis of PRPP from ATP and R5P (ribose 5-phosphate) by transferring the β,γ-diphos-phoryl moiety of ATP to the C1-hydroxy group of R5P [13]. PRPP is a key intermediate of metabolism that is required for syn-thesis of the purine and pyrimidine nucleotides, the pyridine nu-cleotide cofactors NAD and NADP, and the amino acids histidine and tryptophan [4,5]. Superactivity of PRS (PRPP synthetase) is an inherited X-chromosome-linked disorder [6] and the excessive enzymatic activity is associated with uric acid overproduction, gout and neurodevelopmental abnormalities [710]. Three classes of PRSs have been reported which are divided based on their dependence on phosphate ions for activity, their allosteric regul-ation mechanism and their diphosphoryl donor specificity [1115]. Most PRSs belong to class I, which require Mg2+ and phos-phate for enzymatic activity, but can be inhibited allosterically by ADP and possibly other nucleotides [1624]. Class II PRSs are found specifically in plants which are not dependent on phosphate for activity and lack an allosteric site for ADP [11,12]. While class I PRSs transfer the diphosphoryl group only from ATP or dATP to R5P [17,19], class II PRSs have a much broader specificity for a diphosphoryl donor, including ATP, dATP, UTP, CTP and GTP [12,13]. Recently, a novel class III PRS has been identified from Methanocaldococcus jannaschii which is activated by phosphate and uses ATP and dATP as a diphosphoryl donor, but also lacks an allosteric site for ADP [15].

Mg2+ forms a complex with ATP (Mg–ATP) to act as the actual substrate of the enzyme although other divalent cations, such as Mn2+, Ni2+, Co2+ or Cd2+ can serve as substitutes for Mg2+ with relatively lower activity [16,17,19,22,2426]. Phosphate has multiple effects on the activity and structure of the enzyme. It usually acts as an activator for the activity of bacterial and mammalian PRSs although SO42− can mimic the effect of phosphate at approx. 10-fold higher concentrations [14,16,19,22,24,27]. However, phosphate or SO42− has to compete with the inhibitor ADP at the same allosteric site for their function. On the other hand, at high concentrations both phosphate and SO42− can ex-hibit an inhibitory effect because of their competitive binding at the R5P binding site [14]. ADP functions as the most potent inhibitor in either a competitive or allosteric manner depending on the presence and concentration of substrates [18,21,22,24,28]. The competitive inhibition by ADP is owing to its competitive binding against ATP at the ATP binding site and the allosteric inhibition acts through its competitive binding with phosphate at the allosteric site [26,29].

Structural studies of PRSs from Bacillus subtilis [14,30] and M. jannaschii [15] have been performed to characterize the binding sites for substrates, activator and inhibitor. Both PRSs consist of two domains related by a pseudo 2-fold symmetry, and each domain assumes the type I phosphoribosyltransferase fold. bsPRS (B. subtilis PRS) forms a hexamer in the crystal structure and is organized as a propeller with the N-terminal domains at the centre and the C-terminal domains at the outside. The catalytic active site, consisting of the ATP binding site and the R5P binding site, is located at the interface of two domains of one subunit, and the allosteric site for phosphate and ADP is located at the interface between three subunits of the hexamer [14,30]. mjPRS (M. jannaschii PRS) has a tetrameric quaternary structure with two homodimers stacking onto each other, and the active site is located at the domain interface of the subunit, but no allosteric site is found [15].

hPRSs (human PRPP synthetases) have three isoforms that share very high sequence identity (95.0% between hPRS1 and hPRS2; 94.3% between hPRS1 and hPRS3; and 91.2% between hPRS2 and hPRS3 respectively) [3134]. hPRS1 and hPRS2 genes are located on the X chromosome and are expressed in a wide range of tissues, but hPRS3 is an autosomal gene expressed specifically in testis. hPRS1 contains 318 amino acids and shares a moderate sequence identity with bsPRS (47% identity and 67% similarity respectively). It requires phosphate for activation and uses Mg2+ for activity [17,25,35]. The crystallization of hPRS1 has recently been reported [36]. In the present paper we describe the crystal structures of hPRS1 in complexes with SO42− ions and with ATP, Cd2+ and SO42− ions respectively. hPRS1 has an overall structure similar to that of bsPRS. Interestingly, in addition to binding at the R5P binding site and the allosteric site defined previously in bsPRS, an extra SO42− ion is found to bind at a new allosteric site at the dimer interface. Structural and biochemical data together reveal new insights into the allosteric regulatory mechanism of hPRS1 and possibly other eukaryotic PRSs (except for class II plant PRSs).

MATERIALS AND METHODS

Cloning, expression and purification of hPRS1

The hPRS1 gene encoding the full-length hPRS1 protein (318 amino acids) was obtained from the cDNA library of human CD34+ haematopoietic stem/progenitor cells [37]. The gene was cloned into the NdeI and XhoI restriction sites of the pET-22b(+) expression plasmid (Novagen) which inserts a His tag (LEHHHHHH) at the C-terminus. The plasmid was transformed into and expressed in Escherichia coli BL21(DE3) strain (Novagen). The bacterial cells were grown in Luria–Bertani medium supplemented with ampicillin (50 μg/ml) at 37 °C until D600 reached 0.8. After 8 h of expression induced with 0.5 mM IPTG (isopropyl-β-D-thiogalactopyranoside) at 20 °C, the cells were collected by centrifugation at 4000 g for 10 min and suspended in lysis buffer [50 mM NaH2PO4 (pH 8.0), 1 M NaCl, 15% (w/v) glycerol, 5 mM 2-mercaptoethanol and 1 mM PMSF]. The cells were further lysed on ice by sonication and the cell debris was precipitated by centrifugation at 13000 g for 40 min.

Protein purification was carried out by affinity chromatography using a 5 ml Ni–agarose column (Amersham Pharmacia) and an ion-exchange method using a 1 ml ANX weak ion-exchange column (Amersham Pharmacia). The lysis supernatant was loaded onto the Ni–agarose column equilibrated with the lysis buffer which was then washed with a washing buffer (lysis buffer supple-mented with 75 mM imidazole) to remove non-specific binding proteins. The target protein was eluted with an elution buffer [50 mM NaH2PO4 (pH 8.0), 1 M NaCl, 15% (w/v) glycerol and 75 mM EDTA]. This protein sample was dialysed against buffer A [20 mM NaH2PO4 (pH 8.0)] at 20 °C for 3 h and then loaded onto the ANX column. The column was washed with buffer A con-taining gradually increased concentrations of 1 M NaCl up to 50% (v/v) in 20 min and the target protein was washed down at 20–35% concentration of 1 M NaCl. The peak fractions were collected, concentrated and then stored at −70 °C in buffer B (buffer A supplemented with 400 mM NaCl) for biochemical and structural studies. All purification steps were carried out at 4 °C unless otherwise stated to minimize potential proteolysis of the protein. Reducing SDS/PAGE analyses of the protein sample showed a single band at approx. 35 kDa. DLS (dynamic light scattering) analysis was performed to characterize the aggregation state of the protein in solution (at a protein concentration of approx. 2 mg/ml). Expression plasmids of mutant hPRS1 were generated from the vector containing the wild-type enzyme using the QuikChange Site-Directed Mutagenesis Kit (Stratagene) and the sequences of these genes were confirmed by DNA sequencing. The mutant enzymes were expressed and purified following the same procedures as for the wild-type enzyme.

Crystallization and diffraction data collection

Crystallization was performed at 20 °C using the hanging-drop vapour diffusion method. Initial crystallization trials were performed using screening kits from Hampton Research. Protein solution (2 μl) was mixed with 2 μl of the reservoir solution. After several rounds of optimization, cluster-shaped crystals were obtained at the reservoir solution containing 100 mM citric acid (pH 4.15) and 1.3 M (NH4)2SO4. The hPRS1 mutants were crystallized in similar conditions following the same procedure, except that the pH was varied. Crystals of the hPRS1·ATP·SO42−·Cd2+ (quaternary complex) were obtained by adding a solution containing ATP and CdCl2 to a crystallization drop containing pre-grown crystals of the hPRS1·SO42− complex with the final concentration of about 10 mM for both ATP and CdCl2. This soaking process was allowed to equilibrate for more than 48 h before data collection. All of these crystals belong to space group R3. The diffraction data of wild-type and Y146M hPRS1 were collected from flash-cooled crystals at −176 °C using synchrotron radiation at beamline BL-6A at the Photon Factory, Japan. The diffraction data of S132A hPRS1 and the quaternary complex of hPRS1 were collected from flash-cooled crystals at −170 °C using an in-house Rigaku R-AXIS IV++. The statistics of diffraction data are summarized in Table 1.

Table 1
Statistics of diffraction data and structure refinement

Structure determination and refinement

The structure of wild-type hPRS1 was solved with the MR (mole-cular replacement) method as implemented in the program CNS [38] using the dimeric bsPRS structure as the search model [14]. The structures of the two mutants and the quaternary complex were determined with MR using the wild-type hPRS1 structure as the search model. Structure refinement was carried out using CNS and REFMAC5 [39] and model building was facilitated using the program O [40]. There are two monomers in the asymmetric unit; therefore strict 2-fold NCS (non-crystallographic symmetry) constraints were applied in the early stage of refinement, but released in the later stage of refinement. In the initial difference Fourier maps there was strong electron density at both the R5P binding site and the allosteric site defined in the structure of the bsPRS·SO42− complex [14]. The position and shape of the electron density match the SO42− ion very well and therefore they were modelled as SO42− ions. In addition, one extra SO42− ion was found to bind at the homodimer interface in each subunit with well-defined electron density and co-ordination geometry which appeared to be a second allosteric site (see Results and discussion section). In the quaternary complex, there was well-defined elec-tron density for the AMP moiety of an ATP molecule at the ATP binding site of each subunit, but very poor density for the β- and γ-phosphate moieties of ATP and therefore we modelled only the AMP moiety of ATP in the final model. In addition, there was electron density for a metal ion at the ATP binding site which occupies a similar position to that in the bsPRS structure [30]. Since only the Cd2+ ion existed in the crystallization solution, we modelled the metal ion as a Cd2+ ion with its occupancy being adjusted to 0.6 according to the difference density. The statistics of structure refinement are summarized in Table 1.

Enzymatic activity assay

The enzymatic activity of wild-type and mutant hPRS1 was deter-mined by the 32P transfer assay performed at 37 °C in 50 mM tri-ethanolamine (pH 8.0) using a modified method based on that described previously [41]. First the protein was dialysed against a dialysis buffer [50 mM triethanolamine (pH 8.0) and 1 mM ATP] at 20 °C for 2 h, and then diluted to 0.1 mg/ml with the same buffer. The reaction solution containing 2 μl of the enzyme and 18 μl of the reaction buffer [50 mM triethanolamine, 2 mM ATP, 5 mM R5P, 5 mM MgSO4, 0 or 100 μM ADP and phosphate of varying concentrations (0, 5 and 50 mM)] was incubated at 37 °C and then 10 μl aliquots of the reaction solution were withdrawn after 5 min and mixed with 5 μl of 0.33 M formic acid. The mixture (1.5 μl) was applied to poly(ethyleneimine) sheets which were developed in 0.85 M KH2PO4 (pH 3.4) at 4 °C and then dried and exposed using a PhosphorImager (Molecular Dynamics). In the control experiments, phosphate was omitted from the reaction solution. All experiments were repeated at least three times under the same conditions.

RESULTS AND DISCUSSION

Overall structure of hPRS1

The crystal structure of wild-type hPRS1 was determined using the MR method. The final structure model contains two hPRS1 subunits forming a homodimer, six SO42− ions and 183 water molecules in the asymmetric unit (Figure 1). Subunit A contains 305 amino acid residues from Asn3 to Phe313 except for Lys197–Val202, and subunit B contains 308 amino acid residues from Asn3 to Pro317 except for Arg196–Val202. The two NCS-related subunits in the asymmetric unit are essentially identical as revealed by superposition of all corresponding Cα atoms in each subunit [an rmsd (root mean square deviation) of 0.54 Å (1 Å=0.1 nm)]. The statistics of structure refinement and the protein model are summarized in Table 1.

Figure 1
Structure of human PRS1

Structural comparison indicates that the overall structure of hPRS1 is very similar to that of bsPRS [14,30] [an rmsd of 1.02 Å compared with the bsPRS·SO42− complex, 1.18 Å compared with the bsPRS·mADP (α, β-methylene ADP) complex and 1.06 Å compared with the bsPRS·mATP·SO42−·Cd2+ (mATP, α,β-methylene ATP) complex respectively] (Figure 1C and Supplementary Figure S1 at http://www.BiochemJ.org/bj/401/bj4010039add.htm). Like bsPRS, hPRS1 consists of two domains with a similar sandwich-like α/β structure (Figure 1A). At the N-terminal domain, the central five-stranded parallel β-sheet is surrounded by four α-helices and one 310-helix; at the C-terminal domain, it is flanked by two α-helices on one side and three α-helices on the other. Both domains also contain a short anti-parallel β-sheet that protrudes from the central core. The strong conservation of the overall structure of two different PRSs with a moderate sequence homology underscores the functional importance of the enzymes in the metabolism. Despite the overall structural similarity, there are a few notable conformational differences between the two PRSs (Figure 1C). Region Gln278–His283 in hPRS1 forms an α-helix (α8), whereas the equivalent region (Leu280–Lys285) in bsPRS adopts a loop conformation. The region Ser160–Asn164 in hPRS1 forms a short 310-helix (η2), whereas the corresponding region in bsPRS forms an extended loop. In addition, conformational differences are observed in sev-eral loops, including Ser10–Asp14, Ser58–Gly61 and Tyr97–Thr113 (the flexible loop).

Quaternary structure of hPRS1

PRSs normally form oligomers in solution and aggregation ap-pears to be necessary for the enzymatic activity [17,19,42]. In the bsPRS structures, the enzyme forms a hexamer and both the catalytic active site and the allosteric regulatory site are formed by conserved residues from more than one subunit, leading to the suggestion that the hexamer is the physiological functional unit for PRSs [14,30]. The quaternary structure of hPRS1 is also similar to that of bsPRS: three homodimers form a hexamer in a propeller shape with a 32 point group symmetry, the N-terminal do-mains form the inner circle close to the 3-fold axis and the C-terminal domains form the blades towards the outside, further supporting the notion that the hexameric form of the enzyme is the minimal physiological functional unit (see Supplementary Figure S2 at http://www.BiochemJ.org/bj/401/bj4010039add.htm). This arrangement, however, is different from that of mjPRS which forms a tetramer with two homodimers stacking onto each other [15]. There are two types of inter-subunit interface in the hexamer: the interface between the two subunits of the homodimer in the asymmetric unit (designated as the dimer interface) and the interface between the two subunits of the neighbouring homo-dimers near the 3-fold axis (designated as the trimer interface). Contacts between adjacent subunits are very tight, burying 3878 Å2 or 27.7% of the accessible surface area of each subunit (among them the dimer interface buries 1570 Å2 or 11.2% and the trimer interface buries 1845 Å2 or 13.2% respectively). Both dimer and trimer interfaces involve extensive hydrophobic and hydrophilic interactions between highly conserved residues. Genetic and biochemical studies have shown that hPRS superactivity can be caused by a number of mutations in hPRS1 resulting in alteration of the allosteric regulatory mechanisms of activation by phosphate and inhibition by ADP [9,10]. Structural analysis of hPRS1 indicates that the majority of those mutations (including N114S, D183H, A190V and H193Q) are located at the dimer interface and their changes would probably affect the formation of the homodimer and thus the allosteric regulatory site and the enzymatic activity.

The active site

The active site of PRS comprises the binding sites for ATP and R5P. Like bsPRS, the ATP binding site of hPRS1 is located at the interface of two domains of one subunit and is composed of conserved residues of primarily three structural elements, namely the flexible loop (residues Phe92–Ser108, corresponding to residues Tyr97–Thr113 of bsPRS), the PPi (pyrophosphate) binding loop (residues Asp171–Gly174, corresponding to residues Asp174–Gly177 of bsPRS), and the flag region (residues Val30–Ile44 of an adjacent subunit, corresponding to residues Cys35–Ile49 of bsPRS) (Fig-ure 2A). In the hPRS1·ATP· SO42−·Cd2+ complex structure, only the AMP moiety of ATP and a Cd2+ ion were found to bind at the ATP binding site while the PPi moiety of ATP was disordered, which is similar to the bsPRS·mATP·SO42−·Cd2+ complex [30]. The bound AMP has interactions with residues Arg96, Gln97, Asp101 and His130 of one subunit and Phe35, Asn37 and Glu39 of the other subunit, which are highly conserved in all PRSs (Figure 2B). The adenine N1 atom of AMP forms a hydrogen bond with the side chain of Arg96 and the N6 atom forms two hydrogen bonds with the side chains of Asn37 and Glu39 respectively. The ribose 2′- and 3′-hydroxy groups of AMP form two hydrogen bonds with the side chains of Gln97 and Asp101 respectively. The α-phosphate group of AMP has hydrogen-bonding interactions with the side chains of Lys99 and His130. In addition, the side chains of Lys99 and Lys176 are in positions to interact with the β- and γ-phosphates of ATP. In the bsPRS structure, Arg104 (equivalent to Lys99 of hPRS1) appears to have a similar function; however, Ser179 (equivalent to Lys176 of hPRS1) has a much shorter side chain and may have no interaction with the PPi group of ATP. Biochemical data have shown that Lys197 and Arg199 of the flexible catalytic loop in bsPRS play important roles in stabilization of the transition state and their mutations can drastically reduce the Vmax of the enzyme [43]. In the hPRS1 structures, the majority of the flexible catalytic loop remains disordered and the resolved parts adopt a conformation similar to that observed in the bsPRS·mATP·SO42−·Cd2+ complex [30]. The side chain of Arg196 (corresponding to Arg199 of bsPRS) is pointed towards the bound ATP and the distance between the guanidinium group of Arg196 and the α-phosphate of ATP is about 7.4 Å. Therefore it is very probable that the side chain of Arg196 is involved in the interaction with the PPi and stabilization of the transition state. Though the side chain of Lys194 (corresponding to Lys197 of bsPRS) is pointed away from the substrate, it is possible that a conformational change of the flexible catalytic loop as observed in the bsPRS structure may take place, so it could also be involved in the stabilization of the transition state of the substrate.

Figure 2
Structure of the catalytic active site in the hPRS1·ATP·SO42−·Cd2+ complex

Biochemical data show that a Mg2+ ion can steadily bind to both β- and γ-phosphates of ATP, forming an Mg–ATP complex to act as the actual substrate of the enzyme [16,19,25,44]. In the hPRS1·ATP·SO42−·Cd2+ complex structure, the bound Cd2+ ion has four ligands from the bidentate carboxylates of both Asp171 and Asp220 (equivalent to Asp174 and Asp223 of bsPRS respectively), and is in a position to interact with the β- and γ-phos-phates of ATP, but is about 5 Å away from the α-phosphate (Fig-ure 2B). It is possible that this Cd2+ ion might mimic the Mg2+ ion in the Mg–ATP complex in the catalysis. In the bsPRS·mATP·SO42−·Cd2+ complex structure, a second Cd2+ ion binds near His135 (corresponding to His130 of hPRS1) which is suggested to be the binding site of a free cation in the catalytic reaction [30]. However, no Cd2+ ion was found to bind at the equivalent site in the hPRS1 complex.

The flexible loop of bsPRS exhibits the largest conformational change in different structures: it is partially disordered when either a SO42− ion is bound at the allosteric site and/or mATP is bound at the active site, but is entirely ordered when only mADP is bound at the allosteric site [14]. However, in the structures of the hPRS1 complexes in the absence and presence of ATP, the flexible loop assumes a well-defined, open conformation similar to that ob-served in the bsPRS·mATP·SO42−·Cd2+ structure (Figure 1C). This conformational difference appears to be correlated with the binding of the SO42− ion at the novel allosteric site and plays an important role in the regulation of the enzymatic activity (see discussion later).

The R5P binding site of bsPRS was identified to be localized in the region of Asp223–Thr231 of the C-terminal domain of each subunit based on structural similarity with type I phosphoribosyl-transferases [45,46]. Similarly, the R5P binding loop of hPRS1 can be defined as residues Asp220–Thr228. In the structures of both hPRS1·SO42− and hPRS1·ATP·SO42−·Cd2+ complexes, a SO42− ion is bound at the R5P binding site and occupies the position of the 5′-phosphate of R5P, similar to that in the bsPRS·SO42− complex [14]. The SO42− ion has direct hydrogen-bonding interactions with the main chain amides of the strictly conserved residues Asp224, Thr225, Cys226, Gly227 and Thr228, and the side chains of Thr225 and Thr228, as well as two water molecules (Figure 2C).

Compared with the hPRS1·SO42− structure, the binding of ATP induces movement of the tip of the flag region about 2–3 Å closer to the ATP binding site to interact with the adenine moiety of ATP. In addition, the side chain of Gln97 of the flexible loop rotates about 180° to form a hydrogen bond with the ribose 2′-hydroxy group of ATP. It is also noteworthy that the side chain of the strictly conserved Arg96 of the flexible loop at the ATP binding site has two different conformations in the two subunits of the hPRS·SO42− complex: one assumes a conformation pointing towards the side chain of Asp224 of the R5P binding loop as observed in the structure of the bsPRS·mATP·SO42−·Cd2+ complex, while the other takes a different conformation pointing towards the side chain of Asp171 of the PPi binding loop. However, in the presence of ATP, the side chain of Arg96 assumes the same conformation in both subunits and forms hydrogen bonds with the adenine N1 of ATP and the side chain of Asp224 of the R5P binding loop. It is possible that Arg96 may be involved not only in the binding of ATP but also in the binding and release of the product PRPP.

The allosteric site

The enzymatic activity of PRS is regulated by both phosphate and ADP, and the activator phosphate and the inhibitor ADP compete for binding at a common allosteric regulatory site [29]. The allosteric site in bsPRS is located at the interface among three subunits of the hexamer [14]. The allosteric site in hPRS1 is composed of conserved residues Gln135, Asp143, Asn144 and Ser308–Phe313 of one subunit, residues Lys100–Arg104 of the flexible loop of the second subunit, and residues Ser47, Arg49, Ala80 and Ser81 of the third subunit (Figure 2A). In the structures of both hPRS1·SO42− and hPRS1·ATP·SO42−·Cd2+ complexes, this site is occupied by a SO42− ion, similar to that in the bsPRS·SO42− structure. The bound SO42− ion occupies the position of the β-phosphate of ADP and has hydrogen-bonding interactions with the side chains of Ser47, Arg49, Arg104, Ser308 and Ser310, the main chains of Val109 and Ser310, and two water molecules (Figure 3A).

Figure 3
Structures of the allosteric sites

A novel allosteric site

In the structures of both hPRS1·SO42− and hPRS1·ATP·SO42−·Cd2+ complexes, one extra SO42− ion is identified in each hPRS1 subunit compared with the bsPRS·SO42− structure [14]. This SO42− binding site is located at the homodimer interface and positioned between the ATP binding site (about 9 Å away from the ribose moiety of ATP) and the previously defined allosteric site (about 13 Å away from the β-phosphate group of ADP and 5 Å away from the adenine moiety of ADP; Figure 2A). This SO42− ion has hydrogen-bonding interactions with the side chains of Ser132, Gln135, Asn144 and Tyr146 of one subunit, and the main chain amides of Lys100, Asp101 and Lys102 of the flexible loop of the other subunit as well as a water molecule (Figure 3B). All of the residues involved in binding of the SO42− ion (Lys100, Asp101, Lys102, Ser132, Gln135, Asp144 and Tyr146) are strictly conserved in other eukaryotic PRSs (except plant PRSs of class II which are not dependent on phosphate for activity and lack of an allosteric site for ADP), but can be replaced by other residues in bacterial PRSs (corresponding to Lys105, Ala106, Arg107, Pro137, Gln140, His149 and Met151 in bsPRS respectively; see Supplementary Figure S1 at http://www.BiochemJ.org/bj/401/bj4010039add.htm). Binding of this SO42− is accompanied by substantial conformational change of the flexible loop at the active site. As discussed above, in the bsPRS structure, the flexible loop is partially disordered and adopts different conformations in the absence or presence of ATP and is entirely ordered only when mADP is bound at the allosteric site [14,30]. However, in the hPRS1 structures, the flexible loop is well-defined in the presence and absence of ATP and adopts an open conformation similar to that in the bsPRS·mATP·SO42−·Cd2+ structure. It was suggested that binding of the metal ion at the active site might induce conformational change and stabilize the open conformation of the flexible loop [30]. However, from analysis of the hPRS1 structures and its comparison with the bsPRS structures, it is apparent that the binding of the extra SO42− induces the conformation change at the active site and stabilizes the open conformation of the flexible loop. Since this SO42− binding site is very close to the active site and the flexible loop plays an important role in ATP binding and catalysis, we were tempted to speculate that it might be a novel allosteric site for hPRS1.

To investigate this possibility, we performed enzymatic activity assays of both wild-type and mutant hPRS1s containing mutations of some residues at this putative allosteric site (Figure 3C). As described above, the SO42− ion at the tentative allosteric site of hPRS1 has hydrogen-bonding interactions with the side chains of Ser132, Gln135, Asn144 and Tyr146. All four residues are strictly conserved in eukaryotic PRSs (see Supplementary Figure S1 at http://www.BiochemJ.org/bj/401/bj4010039add.htm). However, Ser132 of hPRS1 has a conserved glutamic acid residue in most of the bacterial PRSs but a proline residue (Pro137) in bsPRS; Gln135 remains unchanged in all bacterial PRSs; Asn144 is conserved in most bacterial PRSs except bsPRS which has a histidine residue (His149) at the equivalent position; and Tyr146 has a conserved phenylalanine residue at the equivalent position in most bacterial PRSs but has a methionine residue (Met151) in bsPRS. The muta-genesis results showed that the mutation of Ser132 to alanine substantially reduced the enzymatic activity at a low concentration of phosphate (5 mM), but the impairment could be restored to a comparable level with that of wild-type hPRS1 at a high concen-tration of phosphate (50 mM). However, mutation of Ser132 to phenylalanine appeared to have a less significant effect on the enzyme activity at both low and high concentrations of phosphate. Similarly, mutation of Tyr146 to methionine (the corresponding residue in bsPRS) caused a marked reduction of the activity at both concentrations of phosphate, but a change of Tyr146 to phenyl-alanine had no substantial effect on the enzyme activity at both concentrations of phosphate. On the other hand, mutation of Asn144 to histidine (the corresponding residue in bsPRS) appeared to have no obvious effect on the enzyme activity at both concen-trations of phosphate. The enzyme activity of both wild-type and mutant hPRS1s can be inhibited by ADP.

At the same time, we determined the crystal structures of S132A and Y146M mutant hPRS1s. The overall structures of the two mutants are almost identical with that of wild-type hPRS1 except for very fine conformational differences at the SO42− binding site (Figure 3D). In the S132A mutant structure, the SO42− lost a hydrogen bond with the side chain of Ser132, but retained a hydrogen bond with the main chain carbonyl of Ala132 via a water molecule compared with the wild-type hPRS1 structure. This may explain the biochemical data that the binding of phosphate (or SO42−) with the S132A mutant is weakened at low concentrations of the activator due to the loss of the side chain interaction, but can be restored at high concentrations of the activator because of the preservation of the main chain interaction. In the Y146M mutant structure, the new SO42− binding site is fully occupied by a SO42− ion in one subunit, but partially occupied with a low occupancy (0.5) in the other. Moreover, in the latter subunit, the surrounding residues interacting with the SO42− ion have relatively higher B factors, an indication of higher flexibility. These results indicate that the binding of the SO42− ion is weaker in at least one subunit of the homodimer due to Y146M mutation, consistent with the biochemical data that mutation Y146M resulted in a decrease in the enzyme activity. A modelling study also shows that mutation of Asn144 to histidine could maintain a hydrogen bond between the side chain and the SO42− ion and therefore, has no significant effect on the activity. Taking the structural and biochemical data together, we suggest that this new SO42− binding site is a second allosteric site which can regulate the enzymatic activity.

Implications for the allosteric regulatory mechanism

The crystal structures of hPRS1 complexes with SO42− and with ATP, together with the biochemical data, have revealed a second allosteric site in addition to the allosteric site defined previously in bsPRS. Binding of the activator SO42− to the new allosteric site stabilizes the flexible loop, an important structural element involved in Mg–ATP binding, resulting in the formation of the ac-tive, open conformation similar to that in the structure of the bsPRS·mADP complex [30]. This stabilization is essential to the binding of the substrate Mg–ATP and the initiation of the catalytic reaction. Moreover, binding of the activator SO42− to the second allosteric site brings the flexible loop closer to its co-ordination sphere which makes the tip of the loop in partial overlap with the binding site of the adenine moiety of ADP and prevents entering and/or binding of the inhibitor ADP, leading to the activation of the enzyme. Conversely, binding of the inhibitor ADP to the first allosteric site would block the second allosteric site and prevent conformational rearrangement of structural elements at the active site, leading to the inhibition of the enzyme. This is the first time that structural stabilization at the active site induced by binding of the activator SO42− to the allosteric sites has been observed. Since the residues forming the second allosteric site are strictly conserved in eukaryotic PRSs (except class II plant PRSs), but can be replaced by other residues in bacterial PRSs, it seems plausible that this new allosteric site might also exist in other eukaryotic PRSs, but not in bacterial PRSs. This second allosteric site might be developed during evolution to provide a very fine regulatory mechanism of the enzymatic activity for eukaryotic PRSs because of the complexity of cellular activity in eukaryotes.

Online data

Supplementary Materials:

Acknowledgments

We thank the staff members at Photon Factory, Japan, for technical support in diffraction data collection; Qiuhua Huang of Shanghai Institute of Hematology, Rui-jin Hospital, for providing human PRS1 plasmid; Dr Jinqiu Zhou for assistance in enzymatic activity assay; and other members of our group for helpful discussion. This work was supported by NSFC grants (30125011 and 30570379), MOST 973 and 863 grants (2002BA711A13, 2004CB520801, and 2004CB720102), and CAS grant (KSCX1-SW-17).

References

1. Kornberg A., Lieberman I., Simms E. S. Enzymatic synthesis and properties of 5-phosphoribosylpyrophosphate. J. Biol. Chem. 1955;215:389–402. [PubMed]
2. Khorana H. G., Fernandes J. F., Kornberg A. Pyrophosphorylation of ribose 5-phosphate in the enzymatic synthesis of 5-phosphorylribose 1-pyrophosphate. J. Biol. Chem. 1958;230:941–948. [PubMed]
3. Miller G. A., Jr, Rosenzweig S., Switzer R. L. Oxygen-18 studies of the mechanism of pyrophosphoryl group transfer catalyzed by phosphoribosylpyrophosphate synthetase. Arch. Biochem. Biophys. 1975;171:732–736. [PubMed]
4. Hove-Jensen B. Mutation in the phosphoribosylpyrophosphate synthetase gene (prs) that results in simultaneous requirements for purine and pyrimidine nucleosides, nicotinamide nucleotide, histidine and tryptophan in Escherichia coli. J. Bacteriol. 1988;170:1148–1152. [PMC free article] [PubMed]
5. Hove-Jensen B. Phosphoribosylpyrophosphate (PRPP)-less mutants of Escherichia coli. Mol. Microbiol. 1989;3:1487–1492. [PubMed]
6. Yen R. C., Adams W. B., Lazar C., Becker M. A. Evidence for X-linkage of human phosphoribosylpyrophosphate synthetase. Proc. Natl. Acad. Sci. U.S.A. 1978;75:482–485. [PMC free article] [PubMed]
7. Zoref E., De Vries A., Sperling O. Mutant feedback-resistant phosphoribosylpyrophosphate synthetase associated with purine overproduction and gout. Phosphoribosylpyrophosphate and purine metabolism in cultured fibroblasts. J. Clin. Invest. 1975;56:1093–1099. [PMC free article] [PubMed]
8. Becker M. A., Losman M. J., Kim M. Mechanisms of accelerated purine nucleotide synthesis in human fibroblasts with superactive phosphoribosylpyrophosphate synthetases. J. Biol. Chem. 1987;262:5596–5602. [PubMed]
9. Roessler B. J., Nosal J. M., Smith P. R., Heidler S. A., Palella T. D., Switzer R. L., Becker M. A. Human X-linked phosphoribosylpyrophosphate synthetase superactivity is associated with distinct point mutations in the PRPS1 gene. J. Biol. Chem. 1993;268:26476–26481. [PubMed]
10. Becker M. A., Smith P. R., Taylor W., Mustafi R., Switzer R. L. The genetic and functional basis of purine nucleotide feedback-resistant phosphoribosyl-pyrophosphate synthetase superactivity. J. Clin. Invest. 1995;96:2133–2141. [PMC free article] [PubMed]
11. Krath B. N., Eriksen T. A., Poulsen T. S., Hove-Jensen B. Cloning and sequencing of cDNAs specifying a novel class of phosphoribosyl diphosphate synthase in Arabidopsis thaliana. Biochim. Biophys. Acta. 1999;1430:403–408. [PubMed]
12. Krath B. N., Hove-Jensen B. Class II recombinant phosphoribosyl diphosphate synthase from spinach. Phosphate independence and diphosphoryl donor specificity. J. Biol. Chem. 2001;276:17851–17856. [PubMed]
13. Krath B. N., Hove-Jensen B. Implications of secondary structure prediction and amino acid sequence comparison of class I and class II phosphoribosyl diphosphate synthases on catalysis, regulation, and quaternary structure. Protein Sci. 2001;10:2317–2324. [PMC free article] [PubMed]
14. Eriksen T. A., Kadziola A., Bentsen A. K., Harlow K. W., Larsen S. Structural basis for the function of Bacillus subtilis phosphoribosyl-pyrophosphate synthetase. Nat. Struct. Biol. 2000;7:303–308. [PubMed]
15. Kadziola A., Jepsen C. H., Johansson E., McGuire J., Larsen S., Hove-Jensen B. Novel class III phosphoribosyl diphosphate synthase structure and properties of the tetrameric, phosphate-activated, non-allosterically inhibited enzyme from Methanocaldococcus jannaschii. J. Mol. Biol. 2005;354:815–828. [PubMed]
16. Switzer R. L. Regulation and mechanism of phosphoribosylpyrophosphate synthetase. I. Purification and properties of the enzyme from Salmonella typhimurium. J. Biol. Chem. 1969;244:2854–2863. [PubMed]
17. Fox I. H., Kelley W. N. Human phosphoribosylpyrophosphate synthetase. Distribution, purification, and properties. J. Biol. Chem. 1971;246:5739–5748. [PubMed]
18. Switzer R. L., Sogin D. C. Regulation and mechanism of phosphoribosylpyrophosphate synthetase. V. Inhibition by end products and regulation by adenosine diphosphate. J. Biol. Chem. 1973;248:1063–1073. [PubMed]
19. Roth D. G., Shelton E., Deuel T. F. Purification and properties of phosphoribosyl pyrophosphate synthetase from rat liver. J. Biol. Chem. 1974;249:291–296. [PubMed]
20. Roth D. G., Deuel T. F. Stability and regulation of phosphoribosyl pyrophosphate synthetase from rat liver. J. Biol. Chem. 1974;249:297–301. [PubMed]
21. Gibson K. J., Schubert K. R., Switzer R. L. Binding of the substrates and the allosteric inhibitor adenosine 5′-diphosphate to phosphoribosylpyrophosphate synthetase from Salmonella typhimurium. J. Biol. Chem. 1982;257:2391–2396. [PubMed]
22. Hove-Jensen B., Harlow K. W., King C. J., Switzer R. L. Phosphoribosylpyrophosphate synthetase of Escherichia coli. Properties of the purified enzyme and primary structure of the prs gene. J. Biol. Chem. 1986;261:6765–6771. [PubMed]
23. Kita K., Otsuki T., Ishizuka T., Ishijima S., Tatibana M. Rat liver phosphoribosylpyrophosphate synthetase: existence as heterogeneous aggregates and identification of the catalytic subunit. Adv. Exp. Med. Biol. 1989;253:1–6. [PubMed]
24. Arnvig K., Hove-Jensen B., Switzer R. L. Purification and properties of phosphoribosyldiphosphate synthetase from Bacillus subtilis. Eur. J. Biochem. 1990;192:195–200. [PubMed]
25. Fox I. H., Kelley W. N. Human phosphoribosylpyrophosphate synthetase. Kinetic mechanism and end product inhibition. J. Biol. Chem. 1972;247:2126–2131. [PubMed]
26. Willemoes M., Hove-Jensen B. Binding of divalent magnesium by Escherichia coli phosphoribosyl diphosphate synthetase. Biochemistry. 1997;36:5078–5083. [PubMed]
27. Meyer L. J., Becker M. A. Human erythrocyte phosphoribosylpyrophosphate synthetase. Dependence of activity on state of subunit association. J. Biol. Chem. 1977;252:3919–3925. [PubMed]
28. Becker M. A., Kostel P. J., Meyer L. J. Human phosphoribosylpyrophosphate synthetase. Comparison of purified normal and mutant enzymes. J. Biol. Chem. 1975;250:6822–6830. [PubMed]
29. Willemoes M., Hove-Jensen B., Larsen S. Steady state kinetic model for the binding of substrates and allosteric effectors to Escherichia coli phosphoribosyldiphosphate synthase. J. Biol. Chem. 2000;275:35408–35412. [PubMed]
30. Eriksen T. A., Kadziola A., Larsen S. Binding of cations in Bacillus subtilis phosphoribosyldiphosphate synthetase and their role in catalysis. Protein Sci. 2002;11:271–279. [PMC free article] [PubMed]
31. Iizasa T., Taira M., Shimada H., Ishijima S., Tatibana M. Molecular cloning and sequencing of human cDNA for phosphoribosylpyrophosphate synthetase subunit II. FEBS Lett. 1989;244:47–50. [PubMed]
32. Sonoda T., Taira M., Ishijima S., Ishizuka T., Iizasa T., Tatibana M. Complete nucleotide sequence of human phosphoribosylpyrophosphate synthetase subunit I (PRSI) cDNA and a comparison with human and rat PRPS gene families. J. Biochem. 1991;109:361–364. [PubMed]
33. Taira M., Iizasa T., Shimada H., Kudoh J., Shimizu N., Tatibana M. A human testis-specific mRNA for phosphoribosylpyrophosphate synthetase that initiates from a non-AUG codon. J. Biol. Chem. 1990;265:16491–16497. [PubMed]
34. Nosal J. M., Switzer R. L., Becker M. A. Overexpression, purification, and characterization of recombinant human 5-phosphoribosyl-pyrophosphate synthetase isozymes I and II. J. Biol. Chem. 1993;268:10168–10175. [PubMed]
35. Fox I. H., Kelley W. N. Human phosphoribosylpyrophosphate (PP-ribose-P) synthetase: properties and regulation. Adv. Exp. Med. Biol. 1973;41:79–86. [PubMed]
36. Tang W., Li X., Zhu Z., Tong S., Li X., Zhang X., Teng M., Niu L. Expression, purification, crystallization and preliminary X-ray diffraction analysis of human phosphoribosyl pyrophosphate synthetase 1 (PRS1) Acta Cryst. 2006;F62:432–434. [PMC free article] [PubMed]
37. Zhang Q. H., Ye M., Wu X. Y., Ren S. X., Zhao M., Zhao C. J., Fu G., Shen Y., Fan H. Y., Lu G., et al. Cloning and functional analysis of cDNAs with open reading frames for 300 previously undefined genes expressed in CD34+ hematopoietic stem/progenitor cells. Genome Res. 2000;10:1546–1560. [PMC free article] [PubMed]
38. Brunger A. T., Adams P. D., Clore G. M., DeLano W. L., Gros P., Grosse-Kunstleve R. W., Jiang J. S., Kuszewski J., Nilges M., Pannu N. S., et al. Crystallography and NMR system: a new software suite for macromolecular structure determination. Acta Cryst. 1998;D54:905–921. [PubMed]
39. Murshudov G. N., Vagin A. A., Dodson E. J. Refinement of macromolecular structures by the maximum-likelihood method. Acta Cryst. 1997;D53:240–255. [PubMed]
40. Jones T. A., Zou J. Y., Cowan S. W., Kjeldgaard M. Improved methods for building protein models in electron density maps and the location of errors in these models. Acta Cryst. 1991;A47:110–119. [PubMed]
41. Jensen K. F., Houlberg U., Nygaard P. Thin-layer chromatographic methods to isolate 32P-labeled 5-phosphoribosyl-α-1-pyrophosphate (PRPP): determination of cellular PRPP pools and assay of PRPP synthetase activity. Anal. Biochem. 1979;98:254–263. [PubMed]
42. Schubert K. R., Switzer R. L., Shelton E. Studies of the quaternary structure and the chemical properties of phosphoribosylpyrophosphate synthetase from Salmonella typhimurium. J. Biol. Chem. 1975;250:7492–7500. [PubMed]
43. Hove-Jensen B., Bentsen A.-K. K., Harlow K. W. Catalytic residues Lys197 and Arg199 of Bacillus subtilis phosphoribosyl diphosphate synthetase: alanine-scanning mutagenesis of the flexible catalytic loop. FEBS J. 2005;272:3631–3639. [PubMed]
44. Liao J. C., Sun S., Chandler D., Oster G. The conformational states of Mg. ATP in water. Eur. Biophys. J. 2004;33:29–37. [PubMed]
45. Focia P. J., Craig S. P., 3rd, Nieves-Alicea R., Fletterick R. J., Eakin A. E. A 1.4 Å crystal structure for the hypoxanthine phosphoribosyltransferase of Trypanosoma cruzi. Biochemistry. 1998;37:15066–15075. [PubMed]
46. Shi W., Munagala N. R., Wang C. C., Li C. M., Tyler P. C., Furneaux R. H., Grubmeyer C., Schramm V. L., Almo S. C. Crystal structures of Giardia lamblia guanine phosphoribosyltransferase at 1.75 Å Biochemistry. 2000;39:6781–6790. [PubMed]

Articles from Biochemical Journal are provided here courtesy of The Biochemical Society
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • Gene
    Gene
    Gene links
  • GEO Profiles
    GEO Profiles
    Related GEO records
  • HomoloGene
    HomoloGene
    HomoloGene links
  • MedGen
    MedGen
    Related information in MedGen
  • Pathways + GO
    Pathways + GO
    Pathways, annotations and biological systems (BioSystems) that cite the current article.
  • Protein
    Protein
    Published protein sequences
  • PubMed
    PubMed
    PubMed citations for these articles
  • Structure
    Structure
    Published 3D structures
  • Substance
    Substance
    PubChem Substance links
  • Taxonomy
    Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...