Biochimie xxx (2014) 1e3

Contents lists available at ScienceDirect

Biochimie journal homepage: www.elsevier.com/locate/biochi

Short communication

Exploring the aryl esterase catalysis of paraoxonase-1 through solvent kinetic isotope effects and phosphonate-based isosteric analogues of the tetrahedral reaction intermediate Aljosa Bavec a, 1, Damijan Knez b, 1, Toma z Makovec a, Jure Stojan a, Stanislav Gobec b, a , * Marko Goli cnik a b

Institute of Biochemistry, Faculty of Medicine, University of Ljubljana, Vrazov trg 2, 1000 Ljubljana, Slovenia Chair of Pharmaceutical Chemistry, Faculty of Pharmacy, University of Ljubljana, Askerceva 7, 1000 Ljubljana, Slovenia

a r t i c l e i n f o

a b s t r a c t

Article history: Received 8 July 2014 Accepted 21 August 2014 Available online xxx

Although a recent study of Debord et al. in Biochimie (2014; 97:72e77) described the thermodynamics of the catalysed hydrolysis of phenyl acetate by human paraoxonase-1, the mechanistic details along the reaction route of this enzyme remain unclear. Therefore, we briefly present the solvent kinetic isotope effects on the phenyl acetate esterase activity of paraoxonase-1 and its inhibition with the phenyl methylphosphonate anion, which is a stable isosteric analogue that mimics the high-energy tetrahedral intermediate on the hydroxide-promoted hydrolysis pathway. The data show normal isotope effects, while proton inventory analysis indicates that two protons contribute to the kinetic isotope effect. Coherently, moderate competitive inhibition with the phenyl methylphosphonate anion reveals that the rate-limiting transition state suboptimally resembles the tetrahedral intermediate. The implications of these findings can be attributed to two possible reaction mechanisms that might occur during the paraoxonase-1ecatalysed hydrolysis of phenyl acetate. © 2014 Elsevier Masson SAS. All rights reserved.

Keywords: Paraoxonase-1 Solvent kinetic isotope effect Inventory proton analysis Enzyme reaction mechanism

1. Introduction Human paraoxonase-1 (huPON1) is a calcium-dependent enzyme that can catalyse the hydrolysis of a diverse group of substrates, including aryl esters, lactones and organophosphate compounds. As huPON1 is known to be difficult to express and purify in large quantities, recombinant PON1 (rePON1) enzymes have been created using directed evolution via gene shuffling of the human, rabbit, mouse and rat PON1 genes [1]. These rePON1 proteins can be successfully produced in the E. coli bacterial system as they evolve to higher solubility, whereas the catalytic properties of the rePON1eG2E6 variant with aryl esters do not deviate significantly from huPON1 [1]. Although structural and site-directed mutagenesis studies have provided us much information about the role of individual amino-acid residues in the active site of

* Corresponding author. Tel.: þ386 1 5437669; fax: þ386 1 5437641. E-mail address: [email protected] (M. Goli cnik). 1 These authors contributed equally to this work.

rePON1 [2e4], our knowledge about the nature of very interactions that promote catalysis with this enzyme is still far from complete. Its lactonase and aryl esterase activities are proposed to be hydroxide-ion generated via general base catalysis by the H115eH134 dyad [2,4], and quantitative structureeactivity relationships [5] and recent thermodynamic studies [6] have suggested that this mechanism might be correct. However, the proton transfer and the transition state (TS) geometry, and charge stabilisation through Ca2þ-promoted oxyanion hole during PON1 catalysis have not been characterised to date. Hence, the purpose in this report is to briefly present the solvent kinetic isotope effects (SKIEs) on the phenyl acetate esterase activity of rePON1 and its inhibition with the phenyl methylphosphonate anion, which is stable isosteric analogue that mimics the high-energy tetrahedral intermediate (INT) in the substrate hydrolysis pathway (see Fig. 1). Our data are consistent with previously reported studies [2e6], and they suggest that proton bridging has the main role in PON1 basecatalysed aryl ester hydrolysis, whereas only a supporting role can be attributed to the influence of Ca2þ ions on the catalytic power contribution of PON1.

http://dx.doi.org/10.1016/j.biochi.2014.08.011 0300-9084/© 2014 Elsevier Masson SAS. All rights reserved.

Please cite this article in press as: A. Bavec, et al., Exploring the aryl esterase catalysis of paraoxonase-1 through solvent kinetic isotope effects and phosphonate-based isosteric analogues of the tetrahedral reaction intermediate, Biochimie (2014), http://dx.doi.org/10.1016/ j.biochi.2014.08.011

2

A. Bavec et al. / Biochimie xxx (2014) 1e3

equation was fitted to progress curves using non-linear regression with the GraphPad Prism 5 software [9]. Further details on the data analysis are provided in the Supplementary Information.

3. Results and discussion

Fig. 1. The generally accepted additioneelimination mechanism of PON1-catalysed phenyl acetate hydrolysis on the reaction pathway between the enzymeesubstrate (E$S) and enzymeeproduct (E$P) complexes where bases Bs refer to histidines [2,4]. Phosphonate-based ligand should mimic the transition states near the tetrahedral intermediate.

2. Materials and methods 2.1. Recombinant PON1 The plasmid containing the G2E6 variant of the rePON1 gene was kindly provided by Prof. D.S. Tawfik of the Weizmann Institute (Rehovot, Israel). Expression and purification of the bacterially expressed rePON1 enzyme were performed as previously reported [1], and the procedures are described in detail in the Supplementary Information. The purity of the rePON1 protein was monitored by SDS-PAGE, and its concentration was determined using the Bradford assay (BioRad). 2.2. Phosphonate-based inhibitor synthesis The synthesis of phenyl methylphosphonic acid was carried out as reported previously [7]. Diphenyl methyphosphonate was prepared from commercially available methylphosphonic dichloride and phenol in the presence of triethylamine. Selective hydrolysis of interim diphenyl methyphosphonate in 0.3 M NaOH, followed by the subsequent treatment with concentrated hydrochloric acid and extraction, yielded phenyl methylphosphonic acid (for structure see Fig. 1). Further details on the synthesis, purification and NMR chemical shifts are provided in the Supplementary Information.

The mechanisms of enzyme-catalysed reactions are described through the structure and energy of any INTs and TSs in the enzyme active site that are formed between these INTs, reactants and products on the reaction pathway. Although snapshots of enzymes in high-energy states are employed extensively in computational studies [10], it is unrealistic to explore them directly through experimental measurements given how elusive these states inevitably are. Hence, two indirect measures to explore the nature of the PON1 catalytic mechanism were sought for practical use in the present study: i.e., the SKIEs and inhibition with a phosphonatebased isosteric analogue that mimics the tetrahedral INT. While the values for kcat ¼ 784(±8) s1 and KM ¼ 1.24(±0.02) mM are in agreement with previously reported values for rePON1eG2E6 [4,5] and purified human enzyme [6], the introduction of deuterium in the place of protium in the hydrogenic exchangeable sites of the water and the enzyme produces SKIEs with the kinetic parameters Dkcat ¼ kcat,HOH/kcat,DOD ¼ 3.3 and D KM ¼ KM,HOH/KM,DOD ¼ 1.7. These data signify a loosening at the hydrogenic sites upon the formation of the rate-determining TS (Dkcat > 1) and an almost two-fold tighter enzymeesubstrate binding in D2O than in H2O. Proton inventory analysis provides further details regarding the nature of the hydrogens that are involved in the enzyme-catalysed mechanism. The partial SKIEs on D kcat acquired for eight H2OeD2O mixtures with the mole fraction of deuterium (nD) are shown in Fig. 2, where the graph of Dkcat versus nD is bowl shaped [8]. The two-site model equation D kcat¼(1  nD þ F$nD)2 with an isotopic fractionation factor F of 0.55(±0.01) fit the data well. Poorer fits are obtained for the linear one-site Dkcat ¼ (1  nD þ F$nD) and the exponential infinite-sites D kcat ¼ Fn models (see Table T2 in the Supplementary Information for goodness of fits). This indicates that two protons can undergo changes of state upon activation of the Michaelis E$S complex to a rate-controlling TS. Plain inference in the rePON1-catalysed reaction mechanism can be concluded from our SKIE data. The first step in the enzymecatalysed reaction with about two-fold weaker substrate binding in H2O than in D2O, is partly derived from the lower solubility of phenyl acetate in D2O. Consequently, the non-polar substrate is favourably transferred from the ‘heavy’ aqueous solution into the hydrophobic cleft of the enzyme. Deuterium bonds are shorter and stronger than hydrogen bonds, and as such, there is an additional enthalpic penalty for creating a cavity with a non-polar substrate in

2.3. Kinetic assays The hydrolysis of phenyl acetate was followed spectrophotometrically at l ¼ 270 nm (ε ¼ 1310 M1 cm1) in the reaction buffer (50 mM bis-TRIS propane, 1 mM CaCl2, 1% ethanol, pH 8, 0.15 M ionic strength). Mixtures of H2O-based and D2O-based buffers for SKIEs on the enzyme activity and proton inventory analysis were prepared as described in the literature [8]. The kinetic constants kcat and KM for the substrate and the inhibition constant Ki for the phosphonatebased competitive inhibitor were determined by recording progress curves at 25  C, with initial substrate concentrations of one-to three-fold the Michaelis constant [6]. The integrated Michaelis

Fig. 2. Dkcat versus the atom fraction nD of deuterium in the solvent. The solid line is the theoretical curve for the two-proton model equation Dkcat ¼ (1  nD þ F$nD)2.

Please cite this article in press as: A. Bavec, et al., Exploring the aryl esterase catalysis of paraoxonase-1 through solvent kinetic isotope effects and phosphonate-based isosteric analogues of the tetrahedral reaction intermediate, Biochimie (2014), http://dx.doi.org/10.1016/ j.biochi.2014.08.011

A. Bavec et al. / Biochimie xxx (2014) 1e3

D2O instead of H2O. Although this effect apparently increases the affinity for the enzymeesubstrate complex in D2O, the relative solubilities of small neutral molecules such as phenyl acetate are only up to 1.3-fold lower in D2O than in H2O [8]. Hence, the rest of the SKIEs for the Michaelis constant KM can be attributed to the possibility that rePON1 conformational factors [3] are also involved in the SKIEs upon substrate binding. The second step in the phenyl acetate hydrolysis is rate determining, and the proton inventory data describe a TS in which each of two concerted protons makes a normal contribution of 1/0.55 ¼ 1.8 to an overall SKIE of Dkcat ¼ 3.3. The magnitudes of the isotopic fractionation factors F for the twoproton model are consistent with a postulated concerted protonrelay mechanism that involves general base catalysis by the imidazole rings of two histidine residues (H115&134 schematically denoted with Bs in Fig. 1) [2,4], which are within the range F ¼ 0.50e0.60 that is typically obtained for enzymes where histidines act as general catalysts [11]. The deprotonation of a water molecule via a two-proton shuttle mechanism appears to be orchestrated particularly through the hydroxide ion (of effective concentration kcat/knon ¼ 784 s1/0.53 M1 s1 z 1500 M [12]) attack at the carbonyl centre of the substrate (Fig. 1, TS1). The reaction model is also in agreement with the recent thermodynamic studies of Debord et al. in this journal [6], as the negative activation entropy DS# ¼ 63.1 J/(mol K) [6] hints at the close synchronisation between multiple sites that participate in the rate-limiting TS1 structure formation. However, we are aware that SKIEs can be best interpreted only when the mechanism of the studied reaction is reasonably well understood. Hence, we were also interested in the geometry and the charge of the bonds undergoing major changes during this enzyme-catalysed reaction. The PON1 affinity for the altered phenyl acetate in the rate-controlling TS matches or exceeds the enzyme catalytic proficiency (kcat/KM)/kuncat ¼ 1.2  1012 M1 (kuncat ¼ knon  [OH] ¼ 5.3  107 s1 at pH ¼ 8 [12]), and consequently, a tightly bound enzymeeligand complex is predicted with a stable molecule that resembles the geometry and charge of such a TS. Although the methyl phenylphosphonate anion appears to be reasonably isosteric, with both TSs near the tetrahedral INT (see Fig. 1), it is geometrically more similar to the elimination TS2 in terms of the molecular volumes, bond lengths, and bond angles, than it is to the addition TS1 [13]. In terms of electrostatic properties, the phosphonate groups show a region of negative potential around their oxygen atoms, which is also considerably more intense than that around the corresponding atoms of TS1 [13]. Equally tight binding of the phosphonate and the obviously ratecontrolling TS1 in the enzyme active site is therefore not likely, and results in competitive inhibition of rePON1 with only a moderate inhibition constant Ki ¼ 0.21(±0.01) mM for the methyl phenylphosphonate anion. Although the active site of rePON1 is hydrophobic [2], and as such, it might be unsuitable a-priori for binding charged ligands, only an approximately six-fold tighter enzyme binding to the phosphonate-based inhibitor than to the substrate reveals that the Ca2þ-promoted oxyanion hole stabilises oxyanions reasonably, but suboptimally, to avoid overstabilisation of the substrate ground state [14]. However, these findings are only partially consistent with previous quantitative structureeactivity relationship studies [5]. Aryl esters with small substituents at the meta and ortho positions show almost no dependence of kcat on the leaving group pKa [5], and consequently bond breaking involving the leaving group (Fig. 1, TS2) does not appear to influence the ratelimiting step. Coherently, the variation to a less negative bLg value €nsted plot of log(kcat) versus the pKa of the (i.e. slope of the Bro leaving group) for the enzyme-catalysed reaction in relation to

3

non-enzymatic hydrolysis indicates that general acid-catalysed protonation or metal-catalysed stabilisation of the accumulating negative charge that forms the leaving group might facilitate bond breakage. Hence, our study indicates that PON1-catalysed phenyl acetate hydrolysis potentially proceeds via two different reaction mechanisms that cannot be distinguishable from the experimental data: (i) a stepwise additioneelimination mechanism where apriori stabilisation of additional TS1 is achieved by the generation of two-proton general base catalysis through the coherent His-dyad machinery (see Fig. 1 and refs. [2,4]); and (ii) a concerted mechanism with single TS (not shown in Fig. 1) that features simultaneous addition and elimination steps that are also accompanied by twoproton transfers [10], yet the two-proton His-dyad shuttle system might be replaced by general base activation of the water nucleophile with autonomous H115 in conjunction with synchronous general acid protonation of the phenyl oxide leaving group presumably by D269 [3]. However, it is proposed that the catalytically rich environment of various amino-acid residues around the Ca2þ can assist in the alignment and activation of the water nucleophile, or in the protonation of the leaving group [3]. In conclusion, despite the presence of the Ca2þ-promoted oxyanion hole, we conclude from proton inventory and phosphonatebased inhibition studies of the hydrolysis of phenyl acetate by rePON1 that the catalytic power of rePON1 can be mostly rationalised by concerted two-proton exchange reorganisation at the active site. However, computational studies [10] will be able to further help us to identify the genuine mechanism of the enzymecatalysed reaction pathway. Conflict of interest The authors declare that they have no conflict of interest. Acknowledgements The authors are grateful to Prof. Daniel S. Tawfik for providing the plasmid for the rePON1eG2E6 enzyme. This study was supported by the Slovenian Research Agency (grants P1e170 and P1e208). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.biochi.2014.08.011. References [1] A. Aharoni, L. Gaidukov, S. Yagur, L. Toker, I. Silman, D.S. Tawfik, Proc. Natl. Acad. Sci. U S A 101 (2004) 482e487. [2] M. Harel, A. Aharoni, L. Gaidukov, B. Brumshtein, O. Khersonsky, R. Meged, H. Dvir, R.B.G. Ravelli, A. McCarthy, L. Toker, I. Silman, J.L. Sussman, D.S. Tawfik, Nat. Struct. Mol. Biol. 11 (2004) 412e419. [3] M. BeneDavid, M. Elias, J.J. Filippi, E. Dunach, I. Silman, J.L. Sussman, D.S. Tawfik, J. Mol. Biol. 418 (2012) 181e196. [4] O. Khersonsky, D.S. Tawfik, J. Biol. Chem. 281 (2006) 7649e7656. [5] O. Khersonsky, D.S. Tawfik, Biochemistry 44 (2005) 6371e6382. [6] J. Debord, J.C. Bollinger, M. Harel, T. Dantoine, Biochimie 97 (2014) 72e77. [7] C. McWhirter, E.A. Lund, G. Feng, Q.I. Sheikh, A.C. Hengge, N.H. Williams, J. Am. Chem. Soc. 130 (2008) 13673e13682. [8] K.B. Schowen, R.L. Schowen, Methods Enzymol. 87 (1982) 551e606. [9] M. Goli cnik, Biochem. Eng. J. 63 (2012) 116e123. [10] D. Xie, D. Xu, L. Zhang, H. Guo, J. Phys. Chem. B 109 (2005) 5259e5266. [11] M.S. Matta, D.T. Vo, J. Am. Chem. Soc. 108 (1986) 5316e5318. [12] T.C. Bruice, G.L. Schmir, J. Am. Chem. Soc. 79 (1957) 1663e1667. [13] D.J. Tantillo, K.N. Houk, J. Org. Chem. 64 (1999) 3066e3076. [14] L. Simon, J.M. Goodman, J. Org. Chem. 75 (2010) 1831e1840.

Please cite this article in press as: A. Bavec, et al., Exploring the aryl esterase catalysis of paraoxonase-1 through solvent kinetic isotope effects and phosphonate-based isosteric analogues of the tetrahedral reaction intermediate, Biochimie (2014), http://dx.doi.org/10.1016/ j.biochi.2014.08.011

Exploring the aryl esterase catalysis of paraoxonase-1 through solvent kinetic isotope effects and phosphonate-based isosteric analogues of the tetrahedral reaction intermediate.

Although a recent study of Debord et al. in Biochimie (2014; 97:72-77) described the thermodynamics of the catalysed hydrolysis of phenyl acetate by h...
315KB Sizes 0 Downloads 6 Views