Portal de Periódicos da CAPES

Crystal Structure of the Human Estrogen Sulfotransferase-PAPS Complex

2002; Elsevier BV; Volume: 277; Issue: 20 Linguagem: Inglês

10.1074/jbc.m111651200

1083-351X

Lars C. Pedersen, Evgeniy V. Petrotchenko, Sergei Shevtsov, Masahiko Negishi,

Drug Transport and Resistance Mechanisms

Estrogen sulfotransferase (EST) transfers the sulfate group from 3′-phosphoadenosine 5′-phosphosulfate (PAPS) to estrogenic steroids. Here we report the crystal structure of human EST (hEST) in the context of the V269E mutant-PAPS complex, which is the first structure containing the active sulfate donor for any sulfotransferase. Superimposing this structure with the crystal structure of hEST in complex with the donor product 3′-phosphoadenosine 5′-phosphate (PAP) and the acceptor substrate 17β-estradiol, the ternary structure with the PAPS and estradiol molecule, is modeled. These structures have now provided a more complete view of the SN2-like in-line displacement reaction catalyzed by sulfotransferases. In the PAPS-bound structure, the side chain nitrogen of the catalytic Lys47 interacts with the side chain hydroxyl of Ser137 and not with the bridging oxygen between the 5′-phosphate and sulfate groups of the PAPS molecule as is seen in the PAP-bound structures. This conformational change of the side chain nitrogen indicates that the interaction of Lys47 with Ser137 may regulate PAPS hydrolysis in the absences of an acceptor substrate. Supporting the structural data, the mutations of Ser137 to cysteine and alanine decrease gradually kcat for PAPS hydrolysis and transfer activity. Thus, Ser137 appears to play an important role in regulating the side chain interaction of Lys47 with the bridging oxygen between the 5′-phosphate and the sulfate of PAPS. Estrogen sulfotransferase (EST) transfers the sulfate group from 3′-phosphoadenosine 5′-phosphosulfate (PAPS) to estrogenic steroids. Here we report the crystal structure of human EST (hEST) in the context of the V269E mutant-PAPS complex, which is the first structure containing the active sulfate donor for any sulfotransferase. Superimposing this structure with the crystal structure of hEST in complex with the donor product 3′-phosphoadenosine 5′-phosphate (PAP) and the acceptor substrate 17β-estradiol, the ternary structure with the PAPS and estradiol molecule, is modeled. These structures have now provided a more complete view of the SN2-like in-line displacement reaction catalyzed by sulfotransferases. In the PAPS-bound structure, the side chain nitrogen of the catalytic Lys47 interacts with the side chain hydroxyl of Ser137 and not with the bridging oxygen between the 5′-phosphate and sulfate groups of the PAPS molecule as is seen in the PAP-bound structures. This conformational change of the side chain nitrogen indicates that the interaction of Lys47 with Ser137 may regulate PAPS hydrolysis in the absences of an acceptor substrate. Supporting the structural data, the mutations of Ser137 to cysteine and alanine decrease gradually kcat for PAPS hydrolysis and transfer activity. Thus, Ser137 appears to play an important role in regulating the side chain interaction of Lys47 with the bridging oxygen between the 5′-phosphate and the sulfate of PAPS. Sulfuryl transfer, also referred to as sulfation or sulfonation, is the transfer reaction of the sulfate group from the ubiquitous donor 3′-phosphoadenosine 5′-phosphosulfate (PAPS) 1The abbreviations used are: PAPS3′-phosphoadenosine 5′-phosphosulfateMES2-(N-morpholino)ethanesulfonic acidPAP3′-phosphoadenosine 5′-phosphateE217β-estradiolhESThuman estrogen sulfotransferasemESTmouse estrogen sulfotransferase to an acceptor substrate (1Leyh T.S. Crit. Rev. Biochem. Mol. Biol. 1993; 28: 515-542Crossref PubMed Scopus (91) Google Scholar, 2Duffel M.W. Guegenerich F.P. Comprehensive Toxicology. 3. Oxford University Press, Oxford, United Kingdom1997: 365-386Google Scholar, 3Falany C.N. FASEB J. 1997; 11: 1-2Crossref PubMed Scopus (68) Google Scholar). A large family of enzymes known as sulfotransferases catalyzes these reactions. The active sites of sulfotransferase enzymes are conserved in all of the crystal structures solved for sulfotransferases (4Kakuta Y. Pedersen L.G. Carter C.W. Negishi M. Pedersen L.C. Nat. Struct. Biol. 1997; 4: 904-908Crossref PubMed Scopus (233) Google Scholar, 5Kakuta Y. Petrotchenko E. Pedersen L.C. Negishi M. J. Biol. Chem. 1998; 273: 27325-27330Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar, 6Kakuta Y. Pedersen L.G. Pedersen L.C. Negishi M. Trends Biochem. Sci. 1998; 23: 129-130Abstract Full Text Full Text PDF PubMed Scopus (143) Google Scholar, 7Kakuta Y. Sueyoshi T. Negishi M. Pedersen L.C. J. Biol. Chem. 1999; 274: 10673-10676Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar, 8Dajani R. Cleasby A. Neu M. Wonacott A.J. Jhoti H. Hood A.M. Modi S. Hersey A. Taskinen J. Cooke R.M. Manchee G.R. Coughtrie M.W.H. J. Biol. Chem. 1999; 274: 37862-37868Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar, 9Bidwell L.M. McManus M.E. Gaedigk A. Kakuta Y. Negishi M. Pedersen L. Martin J.F. J. Mol. Biol. 1999; 293: 521-530Crossref PubMed Scopus (99) Google Scholar, 10Pedersen L.C. Petrotchenko E.V. Negishi M. FEBS Lett. 2000; 475: 61-64Crossref PubMed Scopus (91) Google Scholar). In the mEST structure with both PAP and E2, conserved residues Ser138 and Lys48 directly interact with the 3′- and 5′-phosphate of the PAP molecule, respectively, whereas the conserved His108 directly coordinates to the acceptor group of the 17β-estradiol (E2) molecule, suggesting it to be the catalytic base. The crystal structure in the presence of PAP and orthovanadate mimics the transition state of the transfer reaction in which the lysine residue directly coordinates to the bridging oxygen of the 5′-phosphate of the PAP (5Kakuta Y. Petrotchenko E. Pedersen L.C. Negishi M. J. Biol. Chem. 1998; 273: 27325-27330Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar). This coordination suggested that the lysine could be a catalytic residue participating in the dissociation of the sulfate group from PAPS. Site-directed mutagenesis studies have confirmed the functional importance of this residue in various sulfotransferase enzymes (5Kakuta Y. Petrotchenko E. Pedersen L.C. Negishi M. J. Biol. Chem. 1998; 273: 27325-27330Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar,11Marsolais F. Varin L. J. Biol. Chem. 1995; 270: 30458-30463Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar, 12Sueyoshi T. Kakuta Y. Pedersen L.C. Wall F.E. Pedersen L.G. Negishi M. FEBS Lett. 1998; 433: 211-214Crossref PubMed Scopus (47) Google Scholar, 13Ong E. Yeh J.-C. Ding Y. Hindsgaul O. Pedersen L.C. Negishi M. Fukuda M. J. Biol. Chem. 1999; 274: 25608-25612Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar). Accordingly, the transfer reaction has been proposed to proceed through an SN2-like in-line displacement mechanism in which the conserved lysine and histidine play essential roles. 3′-phosphoadenosine 5′-phosphosulfate 2-(N-morpholino)ethanesulfonic acid 3′-phosphoadenosine 5′-phosphate 17β-estradiol human estrogen sulfotransferase mouse estrogen sulfotransferase The serine residue is conserved in all known sulfotransferases with no exception. Site-directed mutagenesis of the conserved serine has been shown to abrogate the activity of human natural killer cell-1 sulfotransferase (13Ong E. Yeh J.-C. Ding Y. Hindsgaul O. Pedersen L.C. Negishi M. Fukuda M. J. Biol. Chem. 1999; 274: 25608-25612Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar). Despite the apparent functional importance, the structural implication of the conserved serine is not clear at this time. In all known crystal structures, the conserved serine forms a hydrogen bond with the 3′-phosphate of the PAP molecule and is away from the 5′-phosphate group where the transfer occurs. Previous structures provided no evidence for the involvement of the serine residue in catalysis, although the interaction with the 3′-phosphate group suggested that it might be conserved for the specific binding of PAPS. The lack of a sulfotransferase structure with the active donor PAPS has limited our understanding of the reaction mechanism and the roles conserved residues like the serine might play in catalysis. We have now crystallized the V269E mutant (a surface mutation) of hEST in the presence of PAPS. This hEST mutant converts the enzyme from a dimer to a monomer in solution but has no observable effect on the activity (14Petrotchenko E.V. Pedersen L.C. Brochers C.H. Tomer K.B. Negishi M. FEBS Lett. 2001; 490: 39-43Crossref PubMed Scopus (90) Google Scholar). All residues involved in interacting directly with the donor and acceptor substrates are found to be conserved in the hEST as observed in the crystal structures of mEST (4Kakuta Y. Pedersen L.G. Carter C.W. Negishi M. Pedersen L.C. Nat. Struct. Biol. 1997; 4: 904-908Crossref PubMed Scopus (233) Google Scholar). However in the hEST-PAPS structure, a conformational change in the position of Lys47is observed. In this structure, Lys47 forms a hydrogen bond with Ser137 instead of with the bridging oxygen between the 5′-phosphate and the sulfate group of PAPS. Complemented with site-directed mutagenesis studies on Ser137 and His107, we herein present experimental evidence that leads us to propose the reaction mechanism in which the 3′-phosphate-Ser interaction helps regulates the action of the lysine in controlling the dissociation of the 5′-sulfate group from PAPS in the absence of substrate. Upon binding of substrate and initiation of catalysis by His107, Lys47 undergoes a conformational change and interacts with the bridging oxygen of the phosphate-sulfate bond of PAPS to promote the dissociation of the PAP leaving group. A hEST cDNA, kindly provided by Charles Falany, was cloned into pGEX-4T3 plasmid. Site-directed mutagenesis was performed using QuikChange kit (Stratagene) and verified by DNA sequencing. Each recombinant plasmid was transformed into Escherichia coli BL21DE3 cell, and the expressed recombinant enzyme was bound on glutathione-Sepharose from which the pure enzyme was eluted by thrombin digestion. For crystallization, the purified enzyme was then washed over a bensoyl-Sepharose microcolumn followed by extensive dialysis against 0.5 mmNaH2PO4 and 100 mm NaCl at pH 7.2. The crystals of hEST were obtained using the sitting drop vapor diffusion technique. Protein concentrated at 15 mg/ml in 0.5 mmNaH2PO4, 100 mm NaCl, and 4 mm PAP (Sigma), pH 7.2, was mixed in equal volume with 0.1m MES, pH 6.0, and 18% polyethylene glycol 8000 and then placed at 20 °C. Typical crystals appeared after 10 days and grew to 0.5 × 0.5 × 0.02 mm after one month. For data collection, crystals were transferred into the cold room at 4 °C. After the temperature in the drop equilibrated, the crystals were transferred three times into 60 μl of 0.1 m MES, pH 6.0, 22% polyethylene glycol 8000, and 10 mm PAPS and allowed to sit for 4 h. Crystals were then transferred in four steps of increasing ethylene glycol concentration until a final concentration of 15% ethylene glycol in the soaking solution was obtained. Crystals were frozen in the nitrogen stream at −180 °C. Data were collected on a RaxisIV image plate detector with a RU3H-rotating anode generator. All data were processed using Denzo and Scalepack (15Otwinowski Z. Minor W. Methods Enzymol. 1996; 276: 307-326Crossref Scopus (38617) Google Scholar). The structure factor phases were determined by molecular replacement using the program AmoRe (16Navaza J. Acta Crystallogr. Sec. A. 1994; 50: 157-168Crossref Scopus (5030) Google Scholar). The hEST crystals obtained were from the mutant V269E. A monomer of mEST coordinates with a PDB identification of 1AQU was used as the initial search model against data from the V269E mutant in complex with PAP crystal. The V269E structure was refined with multiple cycles of model building using the program "O" (17Jones T.A. Zou J.Y. Cowan S.W. Kjeledgaard M. Acta Crystallogr. 1991; 47: 110-119Crossref PubMed Scopus (13014) Google Scholar) for model building and CNS (18Brunger A.T. Adams P.D. Clore G.M. DeLano W.L. Gros P. Acta Crystallogr. Sec. D. 1998; 54: 905-921Crossref PubMed Scopus (16979) Google Scholar) for refinement. These coordinates were then used as a starting model for the structures reported here, which were refined and built in a similar fashion (Table I). The quality of the models was checked using Procheck (19Collaborative Computational Project Number 4Acta Crystallogr. Sec. D. 1994; 50: 760-763Crossref PubMed Scopus (19797) Google Scholar). The position of Lys47 was confirmed using Fo− Fc, 2 Fo −Fc, and Fo −Fc difference maps as well as omit maps. In addition, the Lys residue was modeled and refined in both orientations to determine the correct side chain orientation. The PDB identifier code is 1HY3 for these coordinates.Table ICrystallographic data statisticsData setPAP + estradiolPAPSUnit cell dimensionsa = 61.64, b = 96.65, c = 62.79a = 61.93, b = 97.26, c = 62.73α = 90 °, β = 92.70 °, γ = 90 °α = 90 °, β = 91.13 °, γ = 90 °Space groupP21P21No. of observations81,110182,966Unique reflections25,47263,064Rsym (%) (last shell)1-aRsym = ΣhΣi‖Ii(h) − ‖/ΣhΣiIi(h).7.8 (33.0)4.9 (22.9)I/ςI (last shell)12.5 (2.7)12.87 (2.2)Mosaicity0.680.62Completeness (%) (last shell)98.3 (86.8)91.8 (52.1)Refinement statisticsResolution (Å)50–2.550–1.8Rcryst(%)1-aRsym = ΣhΣi‖Ii(h) − ‖/ΣhΣiIi(h).18.7019.7Rfree22.521.7No. of waters2684501. Root mean square deviation from ideal values Bond length (Å)0.0080.006 Bond angle (°)1.21.3 Dihedral angle (°)21.421.4 Improper angle (°)0.820.72 Mean B value (Å2)27.927.32. Ramachandran statisticsResidues in: most favored regions (%)91.192.2 additionally allowed regions8.77.4 generously allowed regions0.20.0 disallowed regions0.00.2b Rcryst = Σ‖‖Fo‖ − ‖Fc‖/Σ‖Fo‖ calculated from working data set. Rfree is calculated from 5% of data randomly chosen not to be included in refinement.1-a Rsym = ΣhΣi‖Ii(h) − ‖/ΣhΣiIi(h). Open table in a new tab b Rcryst = Σ‖‖Fo‖ − ‖Fc‖/Σ‖Fo‖ calculated from working data set. Rfree is calculated from 5% of data randomly chosen not to be included in refinement. PAPS (Sigma) was purified through MonoQ HR 5/5 column using 20 mm Tris-HCl buffer, pH 7.5, and linear gradient of NaCl (0–1.0 m). A previously published method (20Kakuta Y. Pedersen L.C. Chae K. Song W.C. Leblanc D. London R. Carter C.W. Negishi M. Biochem. Pharmacol. 1998; 55: 313-317Crossref PubMed Scopus (28) Google Scholar) was used to measure estrogen sulfotransferase activity and to calculate Km of PAPS and kcatof sulfation. To assay PAPS hydrolysis, the mixture containing enzyme and excess pure PAPS in 20 mm Tris-HCl buffer, pH 7.5, was incubated for 15 min at 37 °C. 25 μl of aliquot of the reaction mixtures was separated on MonoQ HR 5/5 column using the chromatographic conditions mentioned above. Molar extinction coefficients of 15,400 (at 259 nm) and 53,340 (at 280 nm) were used for the quantitative determination of PAPS (also PAP) and hEST, respectively. The crystal structures of hEST have now been solved in the presence of PAPS and of PAP and E2. The protein structure of hEST is a classical sulfotransferase-fold. The main core of the molecule is composed of an α/β motif. This motif contains a β sheet comprised of five parallel β-strands surrounded by α-helices on both sides and a conserved helix running across the top of the fold. This portion of the molecule contains the conserved 5′-phosphosulfate-binding (PSB)-loop and helix 6 that constitute the binding site for the donor substrate PAPS molecule. A noncrystallographic 2-fold rotation axis relating the two molecules in the asymmetric unit places the loops from the two molecules containing residues 265–275 in close contact. These loops form the identical interaction as seen in human hydroxysteroid sulfotransferase and human aryl sulfotransferase 3 crystals that has been implicated to be the physiological dimerization interface (14Petrotchenko E.V. Pedersen L.C. Brochers C.H. Tomer K.B. Negishi M. FEBS Lett. 2001; 490: 39-43Crossref PubMed Scopus (90) Google Scholar). Apparently, the V269E mutation did not prevent the hEST from forming the proper dimer in the crystal lattice. The C-α trace of the acceptor substrate-binding pocket of hEST is very similar to that of mEST (4Kakuta Y. Pedersen L.G. Carter C.W. Negishi M. Pedersen L.C. Nat. Struct. Biol. 1997; 4: 904-908Crossref PubMed Scopus (233) Google Scholar). These substrate-binding pockets accommodate the E2 molecule in the same orientation. The side chains from residues Lys105 and His107 form hydrogen bonds with the acceptor hydroxyl of the E2 molecule as is also seen in the mEST structures (4Kakuta Y. Pedersen L.G. Carter C.W. Negishi M. Pedersen L.C. Nat. Struct. Biol. 1997; 4: 904-908Crossref PubMed Scopus (233) Google Scholar). The major differences in the substrate-binding site occur near the 17β-hydroxyl group end of the E2 molecule. This region is represented by the amino acid substitutions from mEST to hEST: Tyr20 to Asp21, His148 to Try149, Ile246 to Met247, Asp22 to Arg23, Met89 to Ile90, Leu242 to Met243, and Lys85 to Asn86. Within this region of the hEST structure, a stretch of protein from Val145 to Ser153, appears to have a slightly different conformation from that of mEST and appears to have collapsed slightly on the substrate. The PAP portion of PAPS molecule in the hEST-PAPS structure superimposes well with the PAP molecules in the hEST-PAP-E2 structure. The hEST-PAPS crystal structure reveals the key catalytic residues that coordinate with the sulfate moiety (Fig. 1). The atom O2S of the sulfate group is in position to form a hydrogen bond with NE2 of His107 (3.2 Å) and the backbone amide from Lys47 (3.0 Å). The sulfate coordination to these residues is similar to that of the vanadate molecule as observed in the structure of the mEST-PAP-orthovanadate (5Kakuta Y. Petrotchenko E. Pedersen L.C. Negishi M. J. Biol. Chem. 1998; 273: 27325-27330Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar). The sulfur atom is located 2.8 Å from the acceptor 3-phenolic group of the E2 molecule when the hEST-PAPS structure is superimposed with the hEST-PAP-E2 structure (Fig. 2). The position of the sulfate group with respect to the acceptor substrate molecule in this superposition is consistent with the proposed in-line transfer reaction mechanism based on the geometry.Figure 2Superposition of the hEST-PAPS structure with the hEST-PAP-E2 structure. The hydrogen bonding interactions in the PAPS bound structure are represented by black dotted lines, whereas those in the PAP-E2-bound structure are indicated with red dotted lines. Molscript (23Kraulis P.J. J. Appl. Crystallogr. 1991; 24: 946-950Crossref Google Scholar) and Raster3D (24Merritt E.A. Bacon D.J. Methods Enzymol. 1997; 277: 505-524Crossref PubMed Scopus (3878) Google Scholar) were used to create this figure.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The most significant structural difference in the protein-PAPS interaction, which differs from the PAP-protein interaction, is the side chain conformation of Lys47. The NZ atom of Lys47 interacts with the leaving oxygen of 5′-phosphate group in the PAP bound structures as previously observed in the other structures (4Kakuta Y. Pedersen L.G. Carter C.W. Negishi M. Pedersen L.C. Nat. Struct. Biol. 1997; 4: 904-908Crossref PubMed Scopus (233) Google Scholar, 7Kakuta Y. Sueyoshi T. Negishi M. Pedersen L.C. J. Biol. Chem. 1999; 274: 10673-10676Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar, 8Dajani R. Cleasby A. Neu M. Wonacott A.J. Jhoti H. Hood A.M. Modi S. Hersey A. Taskinen J. Cooke R.M. Manchee G.R. Coughtrie M.W.H. J. Biol. Chem. 1999; 274: 37862-37868Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar, 10Pedersen L.C. Petrotchenko E.V. Negishi M. FEBS Lett. 2000; 475: 61-64Crossref PubMed Scopus (91) Google Scholar). The last torsion angle of the lysine has changed considerably in the hEST-PAPS structure compared with those in the PAP-bound structure. As a result, the NZ of Lys47 is found to coordinate to the side chain oxygen of Ser137 in the PAPS bound structure (2.8 Å) and not to the bridging oxygen between the 5′-phosphate and sulfate groups of the PAPS molecule. In addition, the position of two water molecules has also been affected by the position of Lys47. A Fo −Fc omit map clearly shows the position of the lysine side chain in the PAPS structure (Fig. 3). A Fo −Fc Fourier difference map was also calculated for the PAPS structure with Lys47 modeled and refined in the PAP-E2 bound conformation. The negative density at the position of the NZ atom as well as the position of the positive density peak provides additional evidence supporting that the position of the lysine side chain in the PAPS-bound structure is modeled correctly (Fig. 3). The difference in the conformation of Lys47 may be because of a charge difference on the bridging oxygen between the 5′-phosphate and sulfate moiety. The effective charge on the 5′-phosphate of PAP is −2. However, the charge on the same phosphate in PAPS is −1. In the PAP-bound structure, there is a significant negative charge at the position of the bridging oxygen because the oxygen is not binding a sulfuryl group. Because there is no negative charge accumulated on the bridging oxygen in the PAPS molecule, the NZ of Lys47interacts with the OG atom of Ser137 that also forms a hydrogen bond with the negatively charged 3′-phosphate of the PAPS molecule. The direct interaction of Ser137 with the catalytic residue Lys47 supports a possible catalytic role for Ser137 as well. The conserved lysine (e.g. Lys47 in hEST) is implicated as a catalytic residue that assists in the dissociation of the sulfate group from PAPS during the sulfotransferase reaction (4Kakuta Y. Pedersen L.G. Carter C.W. Negishi M. Pedersen L.C. Nat. Struct. Biol. 1997; 4: 904-908Crossref PubMed Scopus (233) Google Scholar, 5Kakuta Y. Petrotchenko E. Pedersen L.C. Negishi M. J. Biol. Chem. 1998; 273: 27325-27330Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar,11Marsolais F. Varin L. J. Biol. Chem. 1995; 270: 30458-30463Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar, 21Marsolais F. Varin L. Eur. J. Biochem. 1997; 247: 1056-1062Crossref PubMed Scopus (21) Google Scholar). For example, we previously showed that the mutation of the corresponding Lys48 to methionine completely abolished estrogen sulfotransferase activity of the mEST, whereas the mESTK48R mutant retained a significant degree of the residual activity (5Kakuta Y. Petrotchenko E. Pedersen L.C. Negishi M. J. Biol. Chem. 1998; 273: 27325-27330Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar). Complementing the crystal structure of the mEST-PAP-vanadate complex that mimicked the transient state of the sulfuryl transfer reaction, these site-directed mutagenesis studies indicated that the role of the conserved lysine in catalysis is to act as a proton donor to the leaving PAP group (5Kakuta Y. Petrotchenko E. Pedersen L.C. Negishi M. J. Biol. Chem. 1998; 273: 27325-27330Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar). In light of the fact that the side chain nitrogen of Lys47 coordinates to Ser137 and not to the bridging oxygen in the PAPS-bound structure, we hypothesized that the side chain interaction of Lys47 with Ser137 may regulate the catalytic activity of Lys47. To test this hypothesis, Ser137 of hEST was mutated to alanine or cysteine, and the mutated enzymes were analyzed for the PAPS hydrolysis (Fig. 4). First, the Km of PAPS for estrogen sulfotransferase activity was measured for the mutants to select the PAPS concentration used for the determination of kcat.hydrolysis. Subsequently, PAPS hydrolysis of the wild-type and mutated enzymes was measured in the presence of 2–10-fold higher PAPS concentration than their Km of PAPS values. High PAPS concentration would help minimize the effect different mutations have on the rates of catalysis upon change in the binding affinity for PAPS. The hEST mutants profoundly increased their Km of PAPS values (Table II). This finding suggested that that Ser137 was critical for the binding of PAPS to the enzyme, which was consistent with its direct binding to the 3′-phosphate group of the PAPS molecule. The possible role of Ser137 in regulating the catalytic function of Lys47 was better correlated with Kcat of PAPS. Based on the kcat of hydrolysis, the hESTS137Amutant increased PAPS hydrolysis ∼6-fold compared with the wild-type enzyme, whereas the hESTT137C mutant displayed ∼2-fold greater hydrolysis (Table II). Thus, the conservation of Ser137 may be to discourage hydrolysis in the absence of the substrate. Concomitant with the kcat of hydrolysis increase, the mutations also increased kcat for estrogen sulfotransferase activity of the enzymes. These results indicate that Ser137 is capable of regulating the hydrolysis as well as the sulfation of the substrate. Because Ser137 does not directly interact with the bridging oxygen between the 5′-phosphate and sulfate groups of the PAPS molecule, Ser137 may regulate the dissociation through its side chain interaction with Lys47. Removing or weakening the interaction with Ser137 could possibly cause the side chain of Lys47 to adopt a more favorable position coordinating to the bridging oxygen in the order of hESTS137A → hESTS137C → wild-type hEST. These mutations suggest that this Ser137, which is not found near the 5′-phosphate or sulfate moieties, can regulate the hydrolysis and sulfotransferase activity of the enzyme. Taken these mutation studies into consideration, the crystal structures are consistent with a reaction scheme whereby this regulation is carried out though the interactions of Ser137 with Lys47.Table IICatalytic properties of the wild-type and mutated hEST enzymesEnzymeKm of PAPSkcatof sulfationkcat of hydrolysisμms−1min−1Wild-type0.66 ± 0.20.020 ± 0.0040.019 ± 0.096S137C336 ± 290.045 ± 0.0050.030 ± 0.042S137A26.5 ± 6.50.123 ± 0.0120.121 ± 0.039H107NndndndThe sulfotransferase activity was measured as described previously, and a previously reported method was used to calculate the Km of PAPS and kcat of sulfation values (20Kakuta Y. Pedersen L.C. Chae K. Song W.C. Leblanc D. London R. Carter C.W. Negishi M. Biochem. Pharmacol. 1998; 55: 313-317Crossref PubMed Scopus (28) Google Scholar). For kcat of hydrolysis, the activity of PAPS hydrolysis was measured using 1 μm of each enzyme with 84 and 840 μm PAPS. The kcat of hydrolysis values are the average of at least two separate measurements for each enzyme. nd, not detectable. Open table in a new tab The sulfotransferase activity was measured as described previously, and a previously reported method was used to calculate the Km of PAPS and kcat of sulfation values (20Kakuta Y. Pedersen L.C. Chae K. Song W.C. Leblanc D. London R. Carter C.W. Negishi M. Biochem. Pharmacol. 1998; 55: 313-317Crossref PubMed Scopus (28) Google Scholar). For kcat of hydrolysis, the activity of PAPS hydrolysis was measured using 1 μm of each enzyme with 84 and 840 μm PAPS. The kcat of hydrolysis values are the average of at least two separate measurements for each enzyme. nd, not detectable. For the transfer reaction to proceed, the side chain of Lys47 should switch from interacting with Ser137 to interacting with the bridging oxygen during catalysis. The question remains as to what promotes this conformational switch? One possibility is that the binding of E2 could cause a structural alteration that would influence the side chain position of Lys47. However, no other significant structural change is observed upon binding E2. Another possibility is that an electrostatic change in the local environment of the lysine could dictate the switch. In the transfer reaction, the catalytic base histidine deprotonates the acceptor 3-hydroxyl of the E2 molecule and increases the nucleophilic character of this hydroxyl group. Subsequently, the nucleophilic acceptor group attacks the sulfur atom, which builds up a partial negative charge on the bridging oxygen of the PAPS molecule. This charge accumulation may be enough to alter the electrostatic equilibrium of the lysine from interacting with the serine to interacting with the 5′-phosphate. To examine this possibility, His107 was mutated to asparagine to resemble histidine without the negative charge, and the enzymatic characteristics of the mutant were analyzed. Unlike previous alanine mutants that were unstable (5Kakuta Y. Petrotchenko E. Pedersen L.C. Negishi M. J. Biol. Chem. 1998; 273: 27325-27330Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar, 21Marsolais F. Varin L. Eur. J. Biochem. 1997; 247: 1056-1062Crossref PubMed Scopus (21) Google Scholar, 22Homma K. Ogawa K. Hirono Y. Morioka Y. Hirota M. Tanahashi I. Matsui M. Biochim. Biophys. Acta. 1996; 1296: 159-166Crossref PubMed Scopus (16) Google Scholar), the hESTH107N mutant was expressed as a soluble protein to as high a level as the wild-type enzyme in E. coli cells. The mutant was subjected to enzyme assays for estrogen sulfotransferase and PAPS hydrolysis activities. No detectable activity was observed with the hESTH107N mutant for both sulfotransferase and hydrolysis, confirming the absolute requirement of His107 for catalysis (Table II). These results suggest that the action of the catalytic histidine is essential for the side chain nitrogen of Lys47 to switch from Ser137 to the bridging oxygen and to advance the catalysis. Taken all structural features and mutational analyses together, the proposed in-line displacement reaction mechanism is depicted in Fig. 5. The conserved serine plays an important role in controlling PAPS hydrolysis activity of sulfotransferase. The binding of PAPS to the enzyme elicits Ser137 to form an interaction with Lys47. This Ser-Lys interaction removes the side chain nitrogen of Lys47 from the bridging oxygen, preventing PAPS hydrolysis. Upon binding of E2 substrate, His107 makes the 3-phenol group a better nucleophile that attacks the sulfur atom, thus building up a partial negative charge on the bridging oxygen. The negative charge then drives the positively charged side chain of Lys47 to switch to the bridging oxygen, thereby aiding in sulfate dissociation and transfer by stabilizing the transition state and possibly donating the proton to the bridging oxygen. Thus, a principle of this reaction can be viewed as charge redistribution on a coordination chain starting from the 3′-phosphate of PAPS molecule to the bridging oxygen between the 5′-phosphate and sulfate group. The key residues, serine, lysine, and histidine, appear to act in concert to regulate the charge redistribution, thus advancing the transfer reaction. We thank Dr. Charles Falany (University of Alabama at Birmingham) for providing a hEST cDNA. We sincerely appreciate Drs. Lee Pedersen (University of North Carolina at Chapel Hill) and Lee Bartolotti (North Carolina Supercomputer Center) for valuable discussion.