Stereospecific Proton Transfer by a Mobile Catalyst in Mammalian Fructose-1,6-bisphosphate Aldolase
2007; Elsevier BV; Volume: 282; Issue: 42 Linguagem: Inglês
10.1074/jbc.m704968200
ISSN1083-351X
Autores Tópico(s)Metabolism and Genetic Disorders
ResumoClass I fructose-1,6-bisphosphate aldolases catalyze the interconversion between the enamine and iminium covalent enzymatic intermediates by stereospecific exchange of the pro(S) proton of the dihydroxyacetone-phosphate C3 carbon, an obligatory reaction step during substrate cleavage. To investigate the mechanism of stereospecific proton exchange, high resolution crystal structures of native and a mutant Lys146 → Met aldolase were solved in complex with dihydroxyacetone phosphate. The structural analysis revealed trapping of the enamine intermediate at Lys229 in native aldolase. Mutation of conserved active site residue Lys146 to Met drastically decreased activity and enabled trapping of the putative iminium intermediate in the crystal structure showing active site attachment by C-terminal residues 360-363. Attachment positions the conserved C-terminal Tyr363 hydroxyl within 2.9Å of the C3 carbon in the iminium in an orientation consistent with incipient re face proton transfer. We propose a catalytic mechanism by which the mobile C-terminal Tyr363 is activated by the iminium phosphate via a structurally conserved water molecule to yield a transient phenate, whose developing negative charge is stabilized by a Lys146 positive charge, and which abstracts the C3 pro(S) proton forming the enamine. An identical C-terminal binding mode observed in the presence of phosphate in the native structure corroborates Tyr363 interaction with Lys146 and is consistent with transient C terminus binding in the enamine. The absence of charge stabilization and of a mobile C-terminal catalyst explains the extraordinary stability of enamine intermediates in transaldolases. Class I fructose-1,6-bisphosphate aldolases catalyze the interconversion between the enamine and iminium covalent enzymatic intermediates by stereospecific exchange of the pro(S) proton of the dihydroxyacetone-phosphate C3 carbon, an obligatory reaction step during substrate cleavage. To investigate the mechanism of stereospecific proton exchange, high resolution crystal structures of native and a mutant Lys146 → Met aldolase were solved in complex with dihydroxyacetone phosphate. The structural analysis revealed trapping of the enamine intermediate at Lys229 in native aldolase. Mutation of conserved active site residue Lys146 to Met drastically decreased activity and enabled trapping of the putative iminium intermediate in the crystal structure showing active site attachment by C-terminal residues 360-363. Attachment positions the conserved C-terminal Tyr363 hydroxyl within 2.9Å of the C3 carbon in the iminium in an orientation consistent with incipient re face proton transfer. We propose a catalytic mechanism by which the mobile C-terminal Tyr363 is activated by the iminium phosphate via a structurally conserved water molecule to yield a transient phenate, whose developing negative charge is stabilized by a Lys146 positive charge, and which abstracts the C3 pro(S) proton forming the enamine. An identical C-terminal binding mode observed in the presence of phosphate in the native structure corroborates Tyr363 interaction with Lys146 and is consistent with transient C terminus binding in the enamine. The absence of charge stabilization and of a mobile C-terminal catalyst explains the extraordinary stability of enamine intermediates in transaldolases. Stereospecificity is one of the hallmarks of enzyme catalysis. Aldolases, which are abundant and ubiquitous enzymes, catalyze stereospecific carbon-carbon bond formation, one of the most important transformations in living organisms. Their role is best known in glycolysis where fructose-1,6-bis(phosphate) (FBP) 3The abbreviations used are: FBP, fructose-1,6-bis(phosphate); G3P, d-glyceraldehyde 3-phosphate; DHAP, dihydroxyacetone phosphate; WT, recombinant wild type; DERA, d-2-deoxyribose-5-phosphate aldolase. aldolases (EC 4.1.2.13) promote the cleavage of FBP to triose phosphates, d-glyceraldehyde-3-phosphate (G3P), and dihydroxyacetone phosphate (DHAP). A common feature to class I enzymes is the use of a covalent mechanism for catalysis implicating iminium (protonated Schiff base) formation between a lysine residue on the enzyme and a ketose substrate (1Grazi E. Rowley P.T. Chang T. Tchola O. Horecker B.L. Biochem. Biophys. Res. Commun. 1962; 9: 38-43Crossref PubMed Scopus (68) Google Scholar) that entails stereospecific proton exchange in the covalent intermediate (2Rose I.A. Rieder S.V. J. Biol. Chem. 1958; 231: 315-329Abstract Full Text PDF PubMed Google Scholar). Of the three aldolase isozymes found in vertebrates (3Penhoet E.E. Rutter W.J. J. Biol. Chem. 1971; 246: 318-323Abstract Full Text PDF PubMed Google Scholar), the catalytic mechanism has been extensively studied using class I aldolase A from rabbit muscle and key intermediates are depicted in reaction Scheme 1. In the condensation direction, the reaction involves covalent intermediate formation with the keto triose phosphate DHAP followed by condensation with the aldehyde G3P to form the ketose of the acyclic FBP substrate (4Rose I.A. Warms J.V. Biochemistry. 1985; 24: 3952-3957Crossref PubMed Scopus (22) Google Scholar, 5Ray B.D. Harper E.T. Fife W.K. J. Am. Chem. Soc. 1983; 105: 3731-3732Crossref Scopus (15) Google Scholar). To form the C3-C4 bond of FBP, the enzyme stereospecifically abstracts the pro(S) C3 proton of the trigonal iminium 1 (6Jencks, W. P. (1969) Catalysis in Chemistry and Enzymology, pp. 120-121, McGraw-Hill Book Co., New YorkGoogle Scholar, 7Grazi E. Cheng T. Horecker B.L. Biochem. Biophys. Res. Commun. 1962; 7: 250-253Crossref PubMed Scopus (72) Google Scholar) that is formed from the Michaelis complex with DHAP thereby generating via the enamine 2 (2Rose I.A. Rieder S.V. J. Biol. Chem. 1958; 231: 315-329Abstract Full Text PDF PubMed Google Scholar) the carbanionic character at C3 of DHAP for the aldol reaction. The nascent carbon-carbon bond has the same orientation as the pro(S) α-hydrogen initially abstracted from the DHAP imine intermediate. The overall retention of configuration at C3 requires that proton abstraction from 1 to yield the enamine 2 and condensation with aldehyde in 3 must take place from the same direction on the enzyme (8Rose I.A. J. Am. Chem. Soc. 1958; 80: 5835-5836Crossref Scopus (64) Google Scholar). The iminium intermediate formed is then hydrolyzed and FBP is released by the inverse reaction sequence shown in Scheme 1. A distinguishing mechanistic feature of class I aldolases is the relative stability of the iminium 1 and enamine 2 forms, which is a consequence of the catalytic requirements. The enzyme must stabilize the enamine 2 and/or the preceding iminium 1 such that no decomposition occurs prior to reaction with the aldehyde as shown in 3. This stability is reflected in solution where the enzymatic populations 1 and 2 represent 20 and 60%, respectively, of bound DHAP on the muscle enzyme under equilibrium conditions (9Rose I.A. Warms J.V. Kuo D.J. J. Biol. Chem. 1987; 262: 692-701Abstract Full Text PDF PubMed Google Scholar). The interconversion between the two forms implicates the conserved C-terminal Tyr363 residue whose proteolysis inhibits the stereospecific proton exchange step, making it rate-limiting (10Rose I.A. O'Connell E.L. Mehler A.H. J. Biol. Chem. 1965; 240: 1758-1765Abstract Full Text PDF PubMed Google Scholar), whereas the penultimate residues (357-362) of the C-terminal region modulate the rate of exchange reaction (11Berthiaume L. Tolan D.R. Sygusch J. J. Biol. Chem. 1993; 268: 10826-10835Abstract Full Text PDF PubMed Google Scholar). The C-terminal region (residues 343-363) is conformationally mobile (12Adelman R.C. Morse D.E. Chan W. Horecker B.L. Arch. Biochem. Biophys. 1968; 126: 343-352Crossref PubMed Scopus (26) Google Scholar, 13Sygusch J. Beaudy D. Allaire M. Arch. Biochem. Biophys. 1990; 283: 227-233Crossref PubMed Scopus (8) Google Scholar), has an extended secondary structure (14Blom N. Sygusch J. Nat. Struct. Biol. 1997; 4: 36-39Crossref PubMed Scopus (111) Google Scholar, 15St-Jean M. Lafrance-Vanasse J. Liotard B. Sygusch J. J. Biol. Chem. 2005; 280: 27262-27270Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar), and distinguishes mammalian aldolases and their orthologues. Although a number of structural studies have been performed that have investigated intermediates in class I aldolases (15St-Jean M. Lafrance-Vanasse J. Liotard B. Sygusch J. J. Biol. Chem. 2005; 280: 27262-27270Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar, 16Choi K.H. Shi J. Hopkins C.E. Tolan D.R. Allen K.N. Biochemistry. 2001; 40: 13868-13875Crossref PubMed Scopus (76) Google Scholar, 17Lorentzen E. Pohl E. Zwart P. Stark A. Russell R.B. Knura T. Hensel R. Siebers B. J. Biol. Chem. 2003; 278: 47253-47260Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 18Heine A. DeSantis G. Luz J.G. Mitchell M. Wong C.H. Wilson I.A. Science. 2001; 294: 369-374Crossref PubMed Scopus (259) Google Scholar), characterization of the interconversion between the iminium and the enamine has so far proven elusive at the structural level. To investigate the mechanism of stereoselective proton transfer in class I aldolases and how the enzyme is able to make efficient use of intermediates that are intrinsically unstable molecules, high resolution crystallographic studies were under-taken of rabbit muscle class I FBP aldolase in complex with DHAP using aldolase crystals incubated in the presence of DHAP in non-acidic buffer (pH 7.5), similar to kinetic studies (9Rose I.A. Warms J.V. Kuo D.J. J. Biol. Chem. 1987; 262: 692-701Abstract Full Text PDF PubMed Google Scholar). Flash freezing of a native rabbit muscle aldolase crystal soaked in a saturating DHAP solution trapped the enamine intermediate. On the other hand, a crystal of the active site mutant Lys146 → Met revealed DHAP bound as the iminium intermediate and interacting with the C-terminal tyrosine in the active site. Attachment by the C-terminal region yielded reaction geometry conducive for incipient proton transfer at the DHAP C3 carbon and explained proton exchange chirality. Purification and Crystallization—Expression and purification of recombinant native (WT) and Lys146 → Met mutant (K146M) rabbit muscle aldolases were performed as described previously (11Berthiaume L. Tolan D.R. Sygusch J. J. Biol. Chem. 1993; 268: 10826-10835Abstract Full Text PDF PubMed Google Scholar, 15St-Jean M. Lafrance-Vanasse J. Liotard B. Sygusch J. J. Biol. Chem. 2005; 280: 27262-27270Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar, 19Morris A.J. Tolan D.R. J. Biol. Chem. 1993; 268: 1095-1100Abstract Full Text PDF PubMed Google Scholar) and using Escherichia coli strain BL21 SI for overexpression of recombinant proteins (Invitrogen). Aldolase concentration was determined using an extinction coefficient of 0.91 (mg/ml)-1 at 280 nm (20Baranowski T. Niederland T.R. J. Biol. Chem. 1949; 180: 543-551Abstract Full Text PDF PubMed Google Scholar). WT and K146M aldolases were crystallized using the conditions reported previously (15St-Jean M. Lafrance-Vanasse J. Liotard B. Sygusch J. J. Biol. Chem. 2005; 280: 27262-27270Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar). Proton Exchange—Stereospecific pro(S) proton exchange at the C3 carbon of DHAP catalyzed by WT and K146M aldolases and WT aldolase crystals was followed by appearance of the tritium isotope label as (S)-[3-3H1]DHAP from tritiated water. Incorporation of the tritiated label at (S)-C3 of DHAP in the presence of [3H]H2O was determined using an ion exchange protocol (Dowex Cl- resin) described in the literature (8Rose I.A. J. Am. Chem. Soc. 1958; 80: 5835-5836Crossref Scopus (64) Google Scholar). Carboxypeptidase A-treated aldolase was prepared according to conditions described previously (10Rose I.A. O'Connell E.L. Mehler A.H. J. Biol. Chem. 1965; 240: 1758-1765Abstract Full Text PDF PubMed Google Scholar). Digestion of aldolase by carboxypeptidase A was monitored by loss of enzymatic activity using a coupled assay and following NADH oxidation at 340 nm (21Racker E. J. Biol. Chem. 1947; 167: 843-854Abstract Full Text PDF PubMed Google Scholar). The proteolyzed enzyme was separated from carboxypeptidase A by size exclusion chromatography when aldolase activity was decreased to 5% of the original activity. Data Collection and Processing—WT aldolase crystals were soaked either for 2 min in ligand buffer containing DHAP (mother liquor plus 1 mm DHAP) or in phosphate buffer for 5 min (mother liquor plus 20 mm NaH2PO4). A K146M mutant aldolase crystal was soaked for 10 min in a ligand buffer containing a higher concentration of DHAP (mother liquor plus 20 mm DHAP). Prior to data collection, single crystals were cryoprotected in a soaking buffer containing 15% glycerol and immediately flash frozen in a stream of gaseous N2 cooled to 100 K. Data collections were performed at beamline X8-C of the National Synchrotron Light Source (Brookhaven National Laboratory, Upton, NY) and diffracted intensities were measured using a Quantum4 charge-coupled device detector (Area Detector Systems, Poway, CA). All data sets were processed with HKL2000 (22Otwinowski Z. Minor W. Methods Enzymol. 1997; 276: 307-326Crossref PubMed Scopus (38573) Google Scholar) and the results are summarized in Table 1.TABLE 1Data collection and refinement statisticsWT-DHAPWT-PiK146M-DHAPData collectionResolution (Å)50-1.88 (1.98-1.88)aAll values in parentheses are given for the highest resolution shell.50-2.22 (2.34-2.22)aAll values in parentheses are given for the highest resolution shell.50-1.98 (2.07-1.98)aAll values in parentheses are given for the highest resolution shell.Wavelength (Å)0.97950.97951.100Unique reflections/redundancy111,679/3.7 (14,586/3.1)67,988/3.6 (9,692/3.4)96,378/3.5 (9,641/2.5)Completeness (%)98.5 (90.4)99.6 (99.8)96.5 (77.3)Average I/∑(I)15.2 (2.4)7.9 (2.4)21.4 (3.8)RsymbRsym = ∑hkl∑i|Ii(hkl) - Īi(hkl)| /∑hkl∑iIi(hkl), with i running over the number of independent observations of reflection hkl.0.087 (0.466)0.146 (0.607)0.052 (0.258)Space groupP21P21P21Unit cell parametersa (Å), b (Å), c (Å), β (°)82.8, 102.9, 84.2, 98.583.0, 102.7, 84.6, 98.283.9, 103.7, 84.9, 98.8RefinementNumber of atomsProtein10,76710,97310,820Water2,8602,5922,230Hetero363536∑ cutoff; I/∑(I) >111Rcryst (%)cRcryst = ∑hkl||Io(hkl)| - |Ic(hkl)|| /∑hkl|Io(hkl)|.15.914.415.7Rfree (%)dRfree = ∑hkl∈T||Io(hkl) - |Ic(hkl)||/∑hkl∈T|Io(hkl)|, where T is a test data set randomly selected from the observed reflections prior to refinement. Test data set was not used throughout refinement and contains 5, 5, and 10% of the total unique reflections for WT-DHAP, WT-Pi, and K146M-DHAP, respectively.19.119.119.5Root mean square deviationBond length (Å)0.0050.0060.005Bond angle (°)1.2471.3301.211Average B-factor (Å2)28.828.625.1Ramachandran analysiseAnalyzed by PROCHECK (26).Most favorable91.690.291.4Allowed8.49.88.6Luzzati positional error (Å)0.170.180.18a All values in parentheses are given for the highest resolution shell.b Rsym = ∑hkl∑i|Ii(hkl) - Īi(hkl)| /∑hkl∑iIi(hkl), with i running over the number of independent observations of reflection hkl.c Rcryst = ∑hkl||Io(hkl)| - |Ic(hkl)|| /∑hkl|Io(hkl)|.d Rfree = ∑hkl∈T||Io(hkl) - |Ic(hkl)||/∑hkl∈T|Io(hkl)|, where T is a test data set randomly selected from the observed reflections prior to refinement. Test data set was not used throughout refinement and contains 5, 5, and 10% of the total unique reflections for WT-DHAP, WT-Pi, and K146M-DHAP, respectively.e Analyzed by PROCHECK (26Laskowski R.A. MacArthur M.W. Moss D.S. Thornton J.M. J. Appl. Crystallogr. 1993; 26: 283-290Crossref Google Scholar). Open table in a new tab Structure Determination and Refinement—Liganded crystal structures were isomorphous with the crystal structure of native aldolase and belong to the monoclinic space group P21. Structures were solved by difference electron density maps using a native aldolase homotetramer structure as reference model (Protein Data Bank code 1ZAH) (15St-Jean M. Lafrance-Vanasse J. Liotard B. Sygusch J. J. Biol. Chem. 2005; 280: 27262-27270Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar). Each asymmetric unit contains one homotetramer, consistent with the biologically active form of the enzyme. Refinement was performed using all reflections having an I/σ(I) > 1, however, electron density maps were calculated to the resolution indicated in the Table 1 to ensure at least ∼70% completeness in the highest resolution shell with an I/σ(I) > 2. Each refinement cycle was performed as reported previously (15St-Jean M. Lafrance-Vanasse J. Liotard B. Sygusch J. J. Biol. Chem. 2005; 280: 27262-27270Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar) using the Crystallography and NMR System (23Brunger A.T. Adams P.D. Clore G.M. DeLano W.L. Gros P. Grosse-Kunstleve R.W. Jiang J.-S. Kuszewski J. Nilges N. Pannu N.S. Read R.J. Rice L.M. Simonson T. Warren G.L. Acta Crystallogr. Sect. D Biol. Crystallogr. 1998; 54: 905-921Crossref PubMed Scopus (16967) Google Scholar) and O (24Jones T.A. Zou J.Y. Cowan S.W. Kjeldgaard M. Acta Crystallogr. Sect. A Found. Crystallogr. 1991; 47: 110-119Crossref PubMed Scopus (13011) Google Scholar). The PRODRG server was used to generate ligand topology and parameter files (25Schuttelkopf A.W. van Aalten D.M. Acta Crystallogr. Sect. D Biol. Crystallogr. 2004; 60: 1355-1363Crossref PubMed Scopus (4291) Google Scholar). The presence of ligands in the final models was confirmed by observation of simulated annealing Fo - Fc omit maps. Final model statistics, calculated with CNS and PROCHECK (26Laskowski R.A. MacArthur M.W. Moss D.S. Thornton J.M. J. Appl. Crystallogr. 1993; 26: 283-290Crossref Google Scholar), are shown in Table 1. The coordinates and structure factors of WT aldolase soaked with DHAP, WT aldolase soaked with inorganic phosphate (Pi), and K146M aldolase soaked with DHAP have been deposited with the Protein Data Bank (codes 2QUT, 2QUV, and 2QUU, respectively). The final structure models have Rcryst (Rfree) values of 0.159 (0.191), 0.144 (0.191), and 0.157 (0.195), respectively. The corresponding positional errors in atomic coordinates using Luzzati plots were estimated at 0.17, 0.18, and 0.18 Å, respectively. Errors in hydrogen bond distances and positional differences are reported as standard deviations and were estimated based on their value in each subunit of the aldolase homotetramers unless specified otherwise. All figures were prepared using the program PyMOL (27DeLano W.L. The PyMOL Molecular Graphics System. DeLano Scientific LLC, San Carlos, CA2004Google Scholar). Superpositions were performed also with the program PyMOL overlaying Cα atom coordinates of aldolase residues 158-259 that are invariant to binding events as noted previously (15St-Jean M. Lafrance-Vanasse J. Liotard B. Sygusch J. J. Biol. Chem. 2005; 280: 27262-27270Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar). Chemical identity of covalent intermediates trapped in the WT-DHAP and K146M-DHAP structures was selected from simulated annealing Fo - Fc omit maps using the real space statistic, Rfact, calculated in O (RS_FIT command). The statistic assesses the fit of Lys229 Cϵ and Nζ atoms and DHAP C1, C2, C3, and O3 atoms in enamine and iminium forms to the electron density. In the enamine, Lys229 Nζ would be sp3 hybridized, whereas in the iminium, the hybridization would be sp2. The discrimination of this difference in hybridization between the iminium and the enamine was quantified using a paired Student's t test comparing pairwise Rfact statistics for identical subunits having the bound DHAP refined as either iminium or enamine form. Enamine Intermediate—Flash freezing of a WT aldolase crystal in the presence of DHAP trapped a covalent intermediate in each aldolase subunit. Continuous electron density, extending beyond Lys229 Nζ in each subunit, shown in Fig. 1, indicates formation of a stable covalent adduct with DHAP. The planar shape of the electron density observed about the DHAP C2 carbon indicates trigonal hybridization, whereas non-planar shape of the electron density about the Lys229 Nζ atom suggests tetrahedral hybridization and is consistent with trapping of a cis-enamine intermediate in each aldolase subunit (Fig. 2A). Comparison of average B-factors between bound DHAP and interacting side chains, 24 ± 3 and 21 ± 3 Å2, respectively, indicates full active site occupancy by DHAP. To validate the electron density interpretation, real space Rfact was evaluated (24Jones T.A. Zou J.Y. Cowan S.W. Kjeldgaard M. Acta Crystallogr. Sect. A Found. Crystallogr. 1991; 47: 110-119Crossref PubMed Scopus (13011) Google Scholar) to objectively assess model fit to the electron density. The resulting paired Student's t test statistic calculated based on Rfact values for DHAP modeled as either an enamine or an iminium was discriminatory with p = 0.042, the enamine having consistently lower Rfact values in all subunits. This distinction is statistically significant and confirms trapping of a genuine enamine intermediate. Proton exchange in WT aldolase crystals as measured by production of (S)-[3-3H1]DHAP (Table 2) indicates 10 turnovers within the 2-min soaking period of DHAP into the aldolase crystal and corroborates equilibrium trapping of an enamine intermediate in all subunits. Coplanarity of DHAP C1, C2, C3, and O3 atoms, a requisite structural feature in the enamine intermediate, further supports identification of the intermediate as the enamine.FIGURE 2Discrimination of enzymatic intermediates formed with DHAP in aldolase active site. Geometry of enzymatic intermediates was used to identify the chemical entity of the trapped DHAP-aldolase complexes. A, an enamine intermediate was identified in all subunits in the structure of native aldolase on the basis of sp3 hybridization of Lys229 Nζ and enables hydrogen bond formation between the Ser300 hydroxyl and Lys229 Nζ. Difference electron density shown was calculated as described in the legend to Fig. 1 and contoured at 4.0σ. B, an iminium intermediate was identified in all subunits in the structure of the K146M mutant aldolase complexed with DHAP. sp2 hybridization of Lys229 Nζ precludes hydrogen bond formation with Ser300. Difference electron density was calculated from a 1.98-Å simulated annealing Fo - Fc omit map encompassing Lys229 and DHAP and contoured at 4.0σ.View Large Image Figure ViewerDownload Hi-res image Download (PPT)TABLE 2Stereospecific pro(S) proton exchange at C3 of DHAPAldolase1H/3H exchange rateaError calculated using individual errors derived at each step of the measurement protocol.min−1Native9.7 × 102 ± 1.3 × 102Native crystalsbSize of crystals used was comparable to crystals used for data collection.5.14 ± 0.69K146M0.0733 ± 0.0086Carboxypeptidase-treated aldolase0.457 ± 0.061BackgroundcExchange rate measured in the absence of aldolase.0.0118 ± 0.0011a Error calculated using individual errors derived at each step of the measurement protocol.b Size of crystals used was comparable to crystals used for data collection.c Exchange rate measured in the absence of aldolase. Open table in a new tab Numerous interactions with active site residues stabilize the covalent intermediate, as shown in Fig. 1. These interactions are identical to those reported in the structure of the iminium intermediate formed with FBP substrate by the same enzyme (15St-Jean M. Lafrance-Vanasse J. Liotard B. Sygusch J. J. Biol. Chem. 2005; 280: 27262-27270Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar) including an unusual short hydrogen bond (2.4 ± 0.1 Å) with Ser271 side chain. Asp33 and Lys146 further stabilize the intermediate by hydrogen bonding with DHAP C3 hydroxyl and on the basis of consistency with the hydrogen bonding pattern, Lys146 is protonated. An additional hydrogen bond between Ser300 hydroxyl and Lys229 Nζ specifically stabilizes the enamine intermediate (Fig. 2A), which is not possible in the carbanion or the iminium requiring sp2 hybridization at Lys229 Nζ. Stabilization of the enamine by the hydrogen bond is reflected in the reduced pro(S) proton exchange rate at C3 of DHAP when measured for mutations Ser300 to Ala and Cys and compared with native enzyme (28Munger C. Enzymatic mechanism of the fructose 1,6-bisphosphate aldolase from rabbit muscle: Role study of serines 271 and 300 from the active site. Master's thesis, Université de Montréal, Qc, Canada2002Google Scholar). Elimination of hydrogen bonding potential in mutation S300A yielded a 5-fold reduction in exchange rate, whereas weakening the same hydrogen bond in mutation S300C reduced the rate 3-fold. Active site binding of DHAP induces identical conformational changes in each aldolase subunit with respect to native enzyme that results in the asymmetric narrowing of the active site cleft, similar to the conformational change observed in the structure of the iminium intermediate formed with FBP (15St-Jean M. Lafrance-Vanasse J. Liotard B. Sygusch J. J. Biol. Chem. 2005; 280: 27262-27270Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar). From Table 3, conformational displacements of these helical regions, although smaller than for the FBP bound structure, are significant when compared with residues 158-259 of the β-barrel, which make up part of the active site and inter-subunit contacts.TABLE 3Conformational changes induced by DHAP bindingResidues selectedWT-DHAPWT-FBPaProtein Data Bank code 1ZAI (15).158-2590.090.0933-65bCalculation of root mean square differences for these residues was based on superposition of liganded complexes onto native enzyme using Cα atom coordinates of residues 158-259.0.721.02302-329bCalculation of root mean square differences for these residues was based on superposition of liganded complexes onto native enzyme using Cα atom coordinates of residues 158-259.0.660.8435bCalculation of root mean square differences for these residues was based on superposition of liganded complexes onto native enzyme using Cα atom coordinates of residues 158-259.0.661.2538bCalculation of root mean square differences for these residues was based on superposition of liganded complexes onto native enzyme using Cα atom coordinates of residues 158-259.0.300.62303bCalculation of root mean square differences for these residues was based on superposition of liganded complexes onto native enzyme using Cα atom coordinates of residues 158-259.0.700.74a Protein Data Bank code 1ZAI (15St-Jean M. Lafrance-Vanasse J. Liotard B. Sygusch J. J. Biol. Chem. 2005; 280: 27262-27270Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar).b Calculation of root mean square differences for these residues was based on superposition of liganded complexes onto native enzyme using Cα atom coordinates of residues 158-259. Open table in a new tab Superposition of the native and enamine-bound structures identified two conserved water molecules, W1 and W2, whose positions remain invariant upon DHAP attachment (Fig. 1). W1 that has a low B-factor (23 ± 4Å2), similar to that of Cα atoms of residues 10-343 (20 ± 8Å2), is positioned by hydrogen bonds with Glu187 and W2, at 3.1 ± 0.1 Å from both DHAP C2 and C3 carbons. The orientation of W1 is perpendicular to the enamine plane and more in line with the DHAP C2 carbon. A similar in line geometry with the C3 carbon position would require only a slight positional displacement by the water molecule. Phosphate Binding—Phosphate anion binding to aldolase is inhibited by DHAP (29Ginsburg A. Mehler A.H. Biochemistry. 1966; 5: 2623-2634Crossref PubMed Scopus (83) Google Scholar) indicating competition for the P1 phosphate binding site. To address the specificity of the induced conformational changes in stabilization of the enamine intermediate, a WT aldolase crystal was soaked in a phosphate buffer. Structural analysis revealed subunit heterogeneity in response to phosphate binding with respect to the extent of asymmetric narrowing of the active site cleft and binding specificity was controlled by active site attachment of C-terminal residues 360-363. In one subunit (D), phosphate anion binding coincides with the DHAP P1 phosphate site in the enamine structure and concomitant with the C terminus occluding the active site (Fig. 3). The conformational changes induced by phosphate bound at the P1 phosphate binding site were greatest in this subunit and comparable with those induced by DHAP with respect to the native enzyme. In addition, Ser38 from the helical region 33-65 was displaced by 0.6 Å along the helical axis to avoid close contact with Tyr363 in the active site. An additional partially occupied phosphate binding site was found at the subunit interface and interacting with residues Gln202, Arg258, and Lys12, the latter from an adjacent subunit. In two other subunits (A and C), the phosphate anion binding site is shifted ∼2 Å from the previous DHAP P1 phosphate binding locus toward the surface of the active site (not shown). Binding at this site does not involve active site attachment by the C terminus region nor significant conformational changes and is of lower affinity as phosphate positions refined to partial occupancy. The C-terminal regions are fully bound at their proper subunit interface as in the enamine structure and C terminus Tyr363 interacts with the same residues involved in phosphate binding as those implicated at the subunit interface of D. In the remaining subunit (B), electron density
Referência(s)