Báo cáo Y học: A new conceptual framework for enzyme catalysis Hydrogen tunneling coupled to enzyme dynamics in flavoprotein and quinoprotein enzymes docx

MINIREVIEW A new conceptual framework for enzyme catalysis Hydrogen tunneling coupled to enzyme dynamics in flavoprotein and quinoprotein enzymes Michael J. Sutcliffe 1,2 and Nigel S. Scrutton 1 Departments of 1 Biochemistry and 2 Chemistry, University of Leicester, UK Recent years have witnessed high levels of activity in iden- tifying enzyme systems that catalyse H-transfer by quantum tunneling. Rather than being restricted to a small number of specific enzymes as perceived initially, it has now become an accepted mechanism for H-transfer in a growing number of enzymes. Furthermore, H-tunneling is driven by the thermally induced dynamics of the enzyme. In some of those enzymes that break stable C–H bonds the reaction proceeds purely by quantum tunneling, without the need to partially ascend the barrier. Enzymes studied that fall into this cate- gory include the flavoprotein and quinoprotein amine dehydrogenases, which have proved to be excellent model systems. These enzymes have enabled us to study the rela- tionship between barrier shape and reaction kinetics. This has involved studies with ÔslowÕ and ÔfastÕ substrates and enzymes impaired by mutagenesis. A number of key ques- tions now remain, including the nature of the coupling between protein dynamics and quantum tunneling. The wide-ranging implications of quantum tunneling introduce a paradigm shift in the conceptual framework for enzyme catalysis, inhibition and design. Keywords: H-tunneling; transition state theory; protein dynamics; flavoprotein; quinoprotein; kinetic isotope effect; computational simulation; quantum mechanics; stopped-flow kinetics; molecular mechanics. INTRODUCTION The text-book description of catalysis states that enzymes reduce the energy required to surmount the barrier between reactants and products, which leads to enhanced rates. This classical over-the-barrier treatment, known as transition state theory (TST), has been used to depict enzyme- catalysed reactions over the last 50 years [3]. Indications that TST cannot be applied indiscriminately came in the late 1980s and the 1990s. In these instances [4–11], however, the experimental observations could be modelled satisfactorily by using a modified form of TST which incorporates an additional component, a quantum tunneling correction factor [1]; this permits tunneling below the saddle-point of the potential energy surface (i.e. in these instances, the saddle point of the potential energy surface is never reached). The first indication that TST (with tunneling correction) may not model faithfully all enzyme catalysed systems came in 1996 [12]; these data show large deviations from classical TST behaviour. The acid test of the generic applicability of TST to enzyme catalysed reactions came in 1999 from experimental studies on enzymes catalysing C–H bond breakage. Our own studies with methylamine dehy- drogenase [13], and almost simultaneously that of Klinman and coworkers independently with thermophilic alcohol dehydrogenase [14], identified that, rather than ascending the classical energy barrier prior to tunneling, the reaction proceeds solely by quantum tunneling. Furthermore, this work illustrated that quantum tunneling is driven by thermal vibrations of the enzyme-substrate complex, which serve to increase the tunneling probability (by reducing the width and/or height of the barrier) sufficiently for tunneling to occur (Fig. 1). Thus, at the dawn of the 21st century some enzymes were shown to gain their catalytic power from quantum mechanics—arguably the key scientific develop- ment of the 20th century; indeed the first suggestion that quantum mechanical tunneling may be a significant factor in chemical reactions involving the transfer of hydrogen was made by Hund some 70 years ago [15]. Although to some individuals biological systems and quantum mechanics seem poles apart, with hindsight the Correspondence to M. J. Sutcliffe or N. S. Scrutton, Department of Biochemistry, University of Leicester, University Road, Leicester LE1 7RH, UK. Fax: + 44 116252 3369, Tel.: + 44 116223 1337, E-mail: sjm@le.ac.uk or nss4@le.ac.uk Abbreviations:TST,transitionstatetheory;TTQ,tryptophantrypto- phylquinone; MADH, methylamine dehydrogenase; AADH, aroma- tic amine dehydrogenase; TMADH, trimethylamine dehydrogenase; TSOX, heterotetrameric sarcosine dehydrogenase; KIE, kinetic iso- tope effect; QM/MM, quantum mechanical/molecular mechanical. Definitions: Strictly, the term ÔsemiclassicalÕ [1] rather than ÔclassicalÕ is used to indicate the difference in zero point vibrational energies of C–H and C–D bonds in studies using the kinetic isotope effect as a probe of quantum tunneling. In this review, we have used the term classical to indicate over-the-barrier transfer to avoid confusion on the part of a reader less familiar with the concepts of quantised vibrational energy states. Quantum tunneling allows the hydrogen to travel through the barrier. This is made possible by wave–particle duality. A particle cannot pass through – it must pass over-the-barrier. However, wave– particle duality also gives the hydrogen wave-like properties, and this allows it to pass through a region (i.e. the barrier) from which a particle would be excluded. See reference [2] for a more detailed description of quantum tunneling. (Received 7 March 2002, revised 21 May 2002, accepted 6 June 2002) Eur. J. Biochem. 269, 3096–3102 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03020.x only major surprise is perhaps that the key role of quantum tunneling in enzyme catalysed H-transfer reactions is only now being realized. After all, quantum tunneling is an attractive means of transferring hydrogen from reactant to product for those enzyme-catalysed reactions with large activation barriers, where it is difficult to understand how the reaction can occur over-the-barrier. Additionally, all regions of the enzyme (not just the active site) likely contribute to the vibrations that drive quantum tunneling, thus providing a possible reason for why enzymes are much larger than the active site alone. Also, vibrationally assisted quantum tunneling is, for example, well established as a means of H-transfer in metals [16] and for enzyme-mediated electron transfer [17,18] (a proton is much heavier than an electron, reducing the probability for a proton to tunnel; this is one reason why enzymatic H-tunneling was not consid- ered a plausible mechanism until experimental results proved otherwise). H-transfer in nonbiological systems is also known to occur by quantum tunneling, but at low temperatures (cf. enzymatic H-tunneling, which occurs at room temperature); for example, along hydrogen bonds in benzoic acid dimers [19] and in the cyclic network of four hydrogen bonds in calix[4]arene [20]. Earlier reviews on enzyme catalysed tunneling have focussed on the inadequacies of TST for some enzyme reactions and the first descriptions of H-tunneling aided by protein motion along the reaction coordinate [2,21–23]. This review article summarizes our more recent kinetic work on H-transfer by tunneling in quinoprotein and flavoprotein enzymes that catalyse the oxidation of a number of amine substrates, and computational studies of enzymic H-tunneling in these enzymes. QUINOPROTEIN AND FLAVOPROTEIN AMINE DEHYDROGENASES The quinoprotein and flavoprotein amine dehydrogenases are ideally suited to studies of H-transfer. The reactions catalysed are conveniently divided into reductive and oxidative half-reactions. Enzyme reduction occurs by breakage of a substrate C–H bond, the kinetics of which are conveniently followed by absorbance spectrophotome- try owing to reduction of the redox centre (and concomitant change in absorbance spectrum) in the enzyme active site. The oxidative half-reaction usually involves long-range electron transfer to acceptor proteins (e.g. cytochromes, copper proteins or other flavoproteins). The ability to interrogate each half-reaction by stopped-flow methods simplifies substantially the kinetic analysis. Studies of steady-state reactions are often compromised by the inab- ility to focus on a single chemical step, owing to the existence of multiple barriers for binding, product release and a number of chemical steps, each of which may contribute to the overall catalytic rate. Using the stopped- flow method, the chemical step can often be isolated and the true kinetics of C–H bond breakage determined without complications arising from other events in the catalytic sequence. This feature of redox catalysis by the flavoprotein and quinoprotein enzymes makes them attractive targets for studies of H-transfer during substrate oxidation. For this reason, our work has focused on the tryptophan tryptophyl- quinone (TTQ)-dependent amine oxidases methylamine dehydrogenase (MADH) and aromatic amine dehydroge- nase (AADH), and also the flavoenzymes trimethylamine dehydrogenase (TMADH) and heterotetrameric sarcosine Fig. 1. Schematic representation of the three key steps involved in enzyme catalysed H-tun- neling. The TTQ-substrate iminoquinone adduct and the active site aspartate (Asp428) are represented as sticks. Protein dynamics (step 1) facilitate transfer of a proton from the TTQ-substrate iminoquinone adduct to Asp428 by quantum tunneling (step 2). In step 3, the proton is ÔtrappedÕ on the aspartate carboxyl group by subsequent protein vibrations. Ó FEBS 2002 Hydrogen tunneling in enzymes (Eur. J. Biochem. 269) 3097 dehydrogenase (TSOX). High resolution crystallographic structures are available for MADH and TMADH, which has also opened up complementary computational chemistry studies of H-transfer in these enzymes. TTQ-DEPENDENT METHYLAMINE DEHYDROGENASE AND HETEROTERA- MERIC SARCOSINE OXIDASE Our initial studies were focused on TTQ-dependent MADH. TTQ reduction is concerted with C–H bond cleavage from an iminoquinone intermediate that forms rapidly in the reductive half-reaction (Fig. 2). The rate of reduction of the TTQ cofactor has a large kinetic isotope effect (KIE ¼ 16.8 ± 0.5 at 298 K), larger than the upper value expected for reactions described by transition state theory, and is suggestive of tunneling. Tunneling reactions are associated with KIE values greater than unity, owing to the higher probability of proton over deuterium tunneling. The inflated KIE for MADH prompted us to study the temperature dependence of this reaction. Reactions that proceed purely by quantum tunneling are independent of temperature, and thus the KIE should likewise be inde- pendent of temperature. Our studies of TTQ reduction in MADH indicated that the value of the KIE was tempera- ture independent, but significantly the reaction rate was strongly dependent on temperature! Our explanation of this anomalous finding was to couple protein dynamics to the reaction coordinate (Fig. 1); in other words, temperature dependent fluctuations of the enzyme-substrate complex are required to ÔdistortÕ the active site into a geometry that is compatible with a pure tunneling reaction. These (isotope independent) fluctuations required to drive the tunneling reaction give rise to the temperature dependence of the reaction. The inferences drawn from our experimental data are congruent with theoretical models of H-tunneling in enzymes that invoke motion in the protein and/or substrate as part of the tunneling reaction [24–26]. An earlier study [27] had observed temperature-independent KIE values ( 2–3) in steady-state reactions catalysed by serine proteases performed in deuterated solvent, and these were suggested to indicate tunneling. Note, however, that the effect of D 2 O on the reaction dynamics is potentially complicated owing to the exchange of protons throughout the protein scaffold. The data were modelled on earlier theoretical treatments of H-tunneling propounded by Dogonadzhe and coworkers [27] in which thermal vibrations bring the solvent into a configuration favourable to tunneling. The observation of pure tunneling coupled to protein dynamics represents a major departure from the more traditional Ôstatic barrierÕ, Ôquantum correctionÕ depictions of TST that have been used to rationalize H-tunneling effects in enzymes. Pure tunneling is an attractive means of promoting a reaction that has a high potential energy barrier. However, H-tunneling occurs over relatively short distances (e.g.  0.5 A ˚ ). A key feature of the Ôdynamic barrierÕ model is the role of protein motion in transiently compressing the width of the potential energy barrier, which promotes the tunneling reaction. Dynamic fluctuations in protein structure also prevent transfer from the product to reactant side of the potential energy surface. Following tunneling from donor to acceptor atoms, distortion of the active site geometry away from the optimal configuration effectively traps the H nucleus on the product side of the barrier. Pure tunneling (i.e. tunneling without first ascending the barrier) facilitated by protein dynamics is a radically different view of enzyme catalysis compared with the alternative over-the-barrier depictions, but how general is this phenomenon? Soon after our own findings with MADH, Klinman and colleagues demonstrated extreme tunneling coupled to protein motion in a thermophilic alcohol dehydrogenase [14]. They also made the interesting finding that the tunneling contribution was less at mesophi- lic temperatures where the low frequency vibrational modes of the protein are less excited. Our own work has been extended in the direction of H-tunneling with other amine oxidases to demonstrate the general importance of pure tunneling coupled to enzyme dynamics. We have demon- strated that the C–H/C–D bond breakage catalysed by TSOX gives rise to a temperature independent KIE and that reaction rate is strongly dependent on temperature, consis- tent with a pure tunneling reaction driven by motion of the enzyme-substrate complex [28]. TSOX is a flavoprotein, and our work with this enzyme, together with that of Klinman’s work with thermophilic alcohol dehydrogenase, was an early indication that pure tunneling reactions may occur in different enzyme families. More recent reports have also made the connection between enzyme dynamics and tunneling [29] and in at least one case tunneling has been invoked in the reappraisal of the catalytic mechanism of the aspartate proteinase family of enzymes [30]. Our own work, and that of others, on the link between dynamics and tunneling is inferred from the results of kinetic studies. The findings are of potential fundamental importance, thus an independent method of assessing the role of tunneling in enzymes was sought. Our approach here has been to use Fig. 2. Reproductive half-reaction of MADH. (A) A reaction mechanism for the oxidation of methylamine by MADH. The boxed reaction step is the step studied computationally. The base in this reaction corresponds to an aspar- tate residue (Asp428) in MADH. (B) The active site of MADH; the QM region is shown unshaded with link atoms circled. 3098 M. J. Sutcliffe and N. S. Scrutton (Eur. J. Biochem. 269) Ó FEBS 2002 computational chemistry methods, which are described in the following section. COMPUTATIONAL STUDIES OF H-TUNNELING IN METHYLAMINE DEHYDROGENASE How can we gain a detailed picture of enzymic H-tunneling reactions at the atomic level? Computational modelling methods provide an answer in the form of combined quantum mechanical/molecular mechanical (QM/MM) methods. These can simulate the contribution of quantum tunneling to enzyme-catalysed reactions. In the QM/MM approach, a small region at the active site is treated quan- tum mechanically, and is coupled to a simpler molecular mechanics description of the surrounding protein and solvent (Fig. 2). This allows the reaction catalysed by the enzyme to be modelled whilst including the effects of the protein environment. H-Tunneling in the oxidative demethylation of methyl- amine by MADH is a system well suited to study using the QM/MM approach; a crystal structure of MADH has been determined, and we have a large body of experimental data, and the H-tunneling step is rate-limiting. The H-tunneling step, the step we have studied computationally, involves the abstraction by the active site base (Asp428) of a proton (C–H bond breakage) from the iminoquinone (Fig. 2). Our approach [31] was to determine the potential energy surface over which the reaction proceeds, and then to calculate the extent of tunneling by following the reaction over this surface. Details of the approach are as follows. First, the structure of the iminoquinone was produced by adding methylamine to the TTQ in the crystal structure of Methylophilus methylotrophus MADH (Fig. 2B). The par- tial charges of the iminoquinone were calculated using SPARTAN (Wavefunction Inc., Irvine, CA, USA); the entire protein (including iminoquinone) was then protonated and solvated, and energy minimized. Next the QM region was defined as comprising the sidechain of the active site base and the sidechain of the catalytically active modified tryptophan of the TTQ that forms an adduct with methylamine (Fig. 2B). QM/MM calculations were then performed to determine the reactant, product and transition state structures, keeping the link atoms and MM atoms fixed [32] and using GAUSSIAN 94 [33] and AMBER 4.1 [34]. A reaction path profile was then generated, using POLYRATE [35] and the transmission coefficient (extent of tunneling) calculated. This computational study suggested that approximately 96% of the reaction proceeds by tunneling through the barrier, whereas only  4% of the reaction occurs via the classical over-the-barrier route. Similar results were found in an independent study [36], where 99% of the reaction was calculated to proceed by tunneling through the barrier and 1% over-the-barrier. This degree of tunneling with MADH is significantly larger than that observed in other protein systems; the next largest is approximately 60% of the reaction proceeding via tunneling in liver alcohol dehydrog- enase [37]. Also, a significant tunneling correction is needed to get closer to the experimental KIE value at 298 K; no tunneling correction yields a KIE of 6.1, the largest tunneling correction yields 11.1 and the experimental value is 16.8 ± 0.5 [13]. Interestingly, in the independent study mentioned [36], the calculated KIE with tunneling correc- tion was 18.3, falling to 5.9 when tunneling was omitted. AROMATIC AMINE DEHYDROGENASE: SLOWER SUBSTRATES COMPROMISE REACTION RATES BY DIFFERENT MEANS Aromatic amine dehydrogenase (AADH), like MADH, is a TTQ-dependent amine oxidase. AADH transfers electrons, derived from the deamination of primary amines (aromatic amines are generally preferred over simple aliphatic amines), to azurin [38]. As with MADH, the rate-limiting step in the reductive half-reaction is abstraction by an active site base of a proton (C–H bond breakage) from an iminoquinone intermediate (Fig. 2). We used stopped-flow kinetics to study C–H bond breakage in three different substrates by Alcaligenes faecalis AADH [39], the fast substrates dopam- ine and tryptamine, and the slow substrate benzylamine. Again,aswithMADHandTSOX,anindicationasto whether H-transfer occurs classically or by quantum tunneling was gained by investigating the temperature dependence of the rates of C–H and C–D bond breakage, and analysing this using an Eyring plot. This indicated that, whilst the rates of both C–H and C–D bond breakage are temperature dependent for all three substrates, the KIEs are temperature independent. Also, for dopamine and benzylamine (a) there was no significant difference between the apparent Ôactiva- tionÕ energy (or to be more precise the enthalpy of activation) for C–H and C–D bond breakage (C–H bond breakage in tryptamine was too rapid to observe above 277 K), and (b) the ratio of the Arrhenius-like pre- exponential factors [13] was comparable with the KIE. This illustrates that protium and deuterium do not ascend the potential energy barrier and that vibrationally assisted quantum tunneling is the mechanism for H- and D-transfer for all three substrates. Additionally, the enthalpy of activation for benzylamine (67.1 ± 0.9 kJÆmol )1 )is  15 kJÆmol )1 higher than both that for trypta- mine (53.5 ± 1.2 kJÆmol )1 ) and that for dopamine (51.9±1.1kJÆmol )1 ), suggesting that more energy is required to deform the enzyme-substrate iminoquinone intermediate with tryptamine. How does barrier shape change with substrate? The relative rates of C–H and C–D bond breakage in dopamine, tryptamine and benzylamine, and the relative KIEs, give important insight into the shape of the potential energy barrier separating reactants from products. Although a fluctuating potential energy barrier is consistent with our experimental observations, the tunneling event can be visualized as a two-step process (see, for example, [2,13,21]). (Fig. 1). The first step is dynamic, and is required to activate the enzyme–substrate complex by thermal vibration. In essence, this leads to a crossing over of the potential energy surfaces of the enzyme-substrate and enzyme–product complexes. Once this crossing point is populated, the second step (H-transfer by quantum tunneling) can occur. Thus, although the enzyme is dynamic, the barrier can be considered rigid for the lifetime of the tunneling event. Our conceptual framework therefore uses a rigid barrier depic- tion of H-tunneling. To understand the effect of barrier shape on tunneling rates, the factors that enhance tunneling need to be Ó FEBS 2002 Hydrogen tunneling in enzymes (Eur. J. Biochem. 269) 3099 considered. These are (a) a small particle mass and (b) a small area under the potential energy barrier. The barrier also needs to be sufficiently high to favour tunneling rather than classical over-the-barrier reactions; high, narrow barriers are particularly favoured for efficient tunneling. Based on these criteria, we investigated the compatibility of different barrier shapes with our experimental rates and KIEs [39]. These data are inconsistent with both a rectangular energy barrier and a truncated parabolic energy barrier (Fig. 3), two commonly used, idealized barrier shapes. However, the data are consistent with more complex barrier shapes (Fig. 3), the true nature of which remain to be established. COMPROMISING MUTATIONS AND ENZYMATIC H-TUNNELING Over the years, the transition state theory has been the foundation for our quantitative understanding of the effects of compromising mutations on enzyme catalysis. Altered enzymic rates have often been modelled as changes in the stabilities of transition states and ground states. But how do we rationalize the effects of compromising mutations in those enzymes known to catalyse C–H bond breakage by pure tunneling? Is reduced catalytic rate solely attributable to changes in barrier width and height? Clearly, increases in width and height will lead to lower tunneling probability. Mutations within the active site will also likely affect the thermal fluctuations coupled to the reaction coordinate, and may lead to differences in (a) the energetics of barrier compression (as seen for benzylamine and AADH; [39]) and/or (b) extent of barrier compression (i.e. is the active site more ÔrelaxedÕ in a mutant enzyme compared with the native enzyme?). In the latter scenario, the equilibrium distance between donor and acceptor atoms is larger and thus a greater degree of thermal motion is required to form a geometry consistent with quantum tunneling. This has been discussed recently in reactions catalysed by lipoxygenase and mutants thereof [40]. We have attempted to gain an early insight into the effects of compromising mutations on catalysis by studying C–H and C–D bond breakage in TMADH. In the wild-type enzyme, C–H bond cleavage is fast (> 1200 s )1 [41]), and a number of transient kinetic, computational and mutagenesis studies [42–45] have indi- cated that the mechanism of flavin reduction in TMADH involves nucleophilic attack of the substrate lone pair on the C4a atom of the flavin followed by C–H bond breakage (Fig. 4). Evidence now favours the transfer of hydrogen from the substrate methyl to the flavin N5. The rate of C–H bond breakage is lowered > 100-fold in a H172Q mutant TMADH [44], and by an additional factor of 4 in a Y169F mutant TMADH [43]. Both His172 and Tyr169 are located near the substrate-binding site and the flavin isoalloxazine ring as part of a His-Tyr-Asp triad [42], and their mutation will likely lead to altered dynamics within the active site. A KIE accompanies flavin reduction in the wild-type and mutant enzymes [44,46]. The very fast flavin reduction rates of wild-type TMADH with protiated substrate has preven- ted us from performing detailed temperature-dependence studies of this reaction. However, we have shown with H172Q TMADH that the KIE is independent of tempera- ture over the experimental range (277–297 K) and that reaction rates are strongly dependent on temperature [46]. This suggests, as with MADH [13] and TSOX [28], that the reaction proceeds by pure tunneling driven by protein motion. Comparable studies with Y169F TMADH revealed a small temperature dependence on the KIE. The data cannot be understood in terms of the TST. We have previously suggested this might be due to partial thermal excitation of substrate (i.e. partial ascent of the barrier) Fig. 3. Schematic illustrating how changing barrier width affects tunneling through a variety of potential energy barriers. Arrows indicate the paths of H and D nuclei; solid lines denote the classically allowed regions and dashed lines the classically disallowed (quantum tun- neling) regions. The top two barriers (the rectangular barrier and truncated parabolic barrier) are commonly used idealized barrier shapes; these do not agree with the experi- mentally observed trends in rates and KIEs [39]. The bottom barrier is a possible barrier shape that is consistent with the experiment- ally observed trends in both rates and KIEs [39]. In this barrier, it is the narrowest part of the barrier, rather than the whole barrier, that becomes progressively wider, with the concave shoulder becoming progressively less pro- nounced. For a full discussion of the effects of barrier shape on tunneling rate and KIE val- ues see reference [39]. 3100 M. J. Sutcliffe and N. S. Scrutton (Eur. J. Biochem. 269) Ó FEBS 2002 prior to H-tunneling by a vibrationally assisted mechan- ism, although a Boltzman analysis suggests a very small population in anything other than the vibrational ground state. An alternative explanation might be found in invoking a more ÔrelaxedÕ active site with larger degree of vibrational motion required to reach a geometry consis- tent with tunneling. Our observations with Y169F TMADH also find parallels with our recent data for the reaction of MADH with the ÔslowÕ substrate ethanolamine [39]. FUTURE PERSPECTIVES Recent studies have established the importance of H- tunneling in enzyme catalysis. In the last three years, pure tunneling (i.e. without partial barrier ascent) driven by protein motion has become established as a mechanism for the enzymic breakage of C–H bonds; this may be a general strategy for these energetically difficult reactions. Our current understanding of how protein motion is coupled to the reaction coordinate is lacking, and unravelling this represents a major challenge for the future. As this understanding is gained, H-tunneling can then be used as a tool for (a) increasing the catalytic efficiency of enzymes in the biotechnology industry (by enhancing the coupling of dynamics to the reaction coordinate), and (b) producing more effective enzyme inhibitors in the pharmaceutical industry (by dampening those vibrations coupled to the reaction coordinate). Such advances have broader implica- tions, as they will also give insight into the role of dynamics in driving classical over-the-barrier reactions—these issues also impact on the tuning of enzyme performance under extreme conditions (e.g. high/low temperature). Thus, the key challenge for the future is elucidating the inseparable relationship between protein dynamics and classical/quan- tum enzyme mechanisms. ACKNOWLEDGEMENTS The work described in this article was funded by the UK Biotechnology and Biological Sciences Research Council, the Wellcome Trust and the Lister Institute of Preventive Medicine. N. S. S. is a Lister Institute Research Professor. REFERENCES 1. Bell, R.P. (1980) The application of tunnel corrections in chemical kinetics. In The Tunnel Effect in Chemistry. pp. 51–140. Chapman &Hall,London. 2. Sutcliffe, M.J. & Scrutton, N.S. (2000) Enzymology takes a quantum leap forward. Phil.Trans.Roy.Soc.SeriesA.358, 367–386. 3. Kraut, J. (1988) How do enzymes work? Science 242, 533–540. 4. Cha, Y., Murray, C.J. & Klinman, J.P. (1989) Hydrogen tun- neling in enzyme reactions. Science 243, 1325–1330. 5. Grant, K.L. & Klinman, J.P. (1989) Evidence that protium and deuterium undergo significant tunneling in the reaction catalyzed by bovine serum amine oxidase. Biochemistry 28, 6597–6605. 6. Jonsson, T., Edmondson, D.E. & Klinman, J.P. (1994) Hydrogen tunneling in the flavoenzyme monoamine oxidase B. Biochemistry 33, 14871–14878. 7. Kohen, A., Jonsson, T. & Klinman, J.P. (1997) Effects of protein glycosylation on catalysis: changes in hydrogen tunneling and enthalpy of activation in the glucose oxidase reaction. Biochem- istry 36, 2603–2611. 8. Alston, W.C., 2nd, Kanska, M. & Murray, C.J. (1996) Secondary H/T and D/T isotope effects in enzymatic enolization reactions. Coupled motion and tunneling in the triosephosphate isomerase reaction. Biochemistry 35, 12873–12881. 9. Karsten, W.E., Hwang, C.C. & Cook, P.F. (1999) Alpha-sec- ondary tritium kinetic isotope effects indicate hydrogen tunneling and coupled motion occur in the oxidation of 1-malate by NAD- malic enzyme. Biochemistry 38, 4398–4402. 10. Whittaker, M.M., Ballou, D.P. & Whittaker, J.W. (1998) Kinetic isotope effects as probes of the mechanism of galactose oxidase. Biochemistry 37, 8426–8436. 11. Nesheim, J.C. & Lipscomb, J.D. (1996) Large kinetic isotope effects in methane oxidation catalyzed by methane mono- oxygenase: evidence for C-H bond cleavage in a reaction cycle intermediate. Biochemistry 35, 10240–10247. 12. Jonsson, T., Glickman, M.H., Sun, S. & Klinman, J.P. (1996) Experimental evidence for extensive tunneling of hydrogen in the lipoxygenase reaction: implications for enzyme catalysis. J.Am. Chem. Soc. 118, 10319–10320. 13. Basran, J., Sutcliffe, M.J. & Scrutton, N.S. (1999) Enzymatic H-transfer requires vibration-driven extreme tunneling. Bio- chemistry 38, 3218–3222. 14. Kohen, A., Cannio, R., Bartolucci, S. & Klinman, J.P. (1999) Enzyme dynamics and hydrogen tunnelling in a thermophilic alcohol dehydrogenase. Nature 399, 496–499. 15. Hund, F. (1927) Interpretation of molecular spectra. III. Remarks on the oscillation and rotation spectra of molecules with more than two nuclei. Z. Phys. 43, 805–826. 16. Stoneham, A. (1990) Quantum diffusion in solids. J. Chem. Soc. Faraday Trans. 86, 1215–1220. 17. DeVault, D. (1980) Quantum mechanical tunnelling in biological systems. Q. Rev. Biophys. 13, 387–564. 18. Marcus, R.A., & Sutin, N. (1985) Electron transfers in chemistry and biology. Biochim. Biophys. Acta. 811, 265–316. 19. Neumann, M., Brogham, D.F., McGloin, C.J., Johnson, M.R., Horsewill, A.J. & Tromsdorff, H.P. (1998) Proton tunneling in benzoic acid crystals at intermediate temperatures: nuclear mag- netic resonance and neutron scattering studies. J. Chem. Phys. 109, 7300–7311. Fig. 4. Our proposed mechanism for the oxidation of trimethylamine by TMADH, which is based on mechanistic studies [45,46] and by analogy with the work of Edmondson on monoamine oxidase [47]. It is important to note that steps 1 and 2 of this scheme are likely to be concerted; the second species shown is therefore not intended to represent a true enzyme intermediate. Enz, enzyme. Ó FEBS 2002 Hydrogen tunneling in enzymes (Eur. J. Biochem. 269) 3101 20. Horsewill, A.J., Jones, N.H. & Caciuffo, R. (2001) Evidence for coherent proton tunneling in a hydrogen bond network. Science 291, 100–103. 21. Scrutton, N.S., Basran, J. & Sutcliffe, M.J. (1999) New insights into enzyme catalysis: ground state tunnelling driven by protein dynamics. Eur. J. Biochem. 264, 666–671. 22. Kohen, A. & Klinman, J.P. (1999) Hydrogen tunneling in biol- ogy. Chem. Biol. 6, R191–R198. 23. Sutcliffe, M.J. & Scrutton, N.S. (2000) Enzyme catalysis: over-the- barrier or through-the barrier. Trends Biochem. Sci. 25, 405–408. 24. Bruno, W.J. & Bialek, W. (1992) Vibrationally enhanced tun- neling as a mechanism for enzymatic hydrogen transfer. Biophys. J. 63, 689–699. 25. Antoniou, D. & Schwartz, S.D. (1997) Large kinetic isotope effects in enzymatic proton transfer and the role of substrate oscillations. Proc. Natl Acad. Sci. USA 94, 12360–12365. 26. Kuznetsov, A.M. & Ulstrup, J. (1999) Proton and hydrogen atom tunnelling in hydrolytic and redox enzyme catalysis. Can. J. Chem. 77, 1085–1096. 27. Krishtalik, L.I. (1986) Charge transfer reactions in electrochemical and chemical processes. pp. 244–298. Consultants Bureau, New York. 28. Harris, R.J., Meskys, R., Sutcliffe, M.J. & Scrutton, N.S. (2000) Kinetic studies of the mechanism of C-H bond breakge by the heterotetrameric sarcosine oxidase of Arthrobacter sp 1-IN. Bio- chemistry 39, 1189–1198. 29. Northrop, D.B. & Cho, Y.K. (2000) Effect of pressure on deu- terium isotope effects of yeast alcohol dehydrogenase: evidence for mechanical models of catalysis. Biochemistry 39, 2406–2412. 30. Northrop, D.B. (2001) Follow the protons: a low barrier hydro- gen bond unifies the mechanisms of the aspartic proteases. Acc Chem Res. 34, 790–797. 31. Faulder, P.F., Tresadern, G., Chohan, K.K., Scrutton, N.S., Sutcliffe, M.J., Hillier, I.H. & Burton, N.A. (2001) QM/MM studies show substantial tunneling for the hydrogen-transfer reaction in methylamine dehydrogenase. J.Am.Chem.Soc.123, 8604–8605. 32. Hall, R.J., Hindle, S.A., Burton, N.A. & Hillier, I.H.J. (2000) Aspects of hybrid QM/MM calculations: The treatment of the QM/MM interface region and geometry optimization with an application to chorismate mutase. J. Comput. Chem. 21, 1433–1441. 33. Frisch, M.J., Trucks, G.W., Schlegel, H.B., Gill, P.M.W., Johnson, B.G., Robb, M.A., Cheeseman, J.R.A.K.T., Petersson, G.A., Montgomery, J.A., Raghavachari, K. et al. (1995) Gaussian94. Gaussian Inc, Pittsburgh PA. 34. Pearlman, D.A., Case, D.A., Caldwell, J.W., Ross, W.S., Cheatham, T.E., Ferguson, D.M., Seibel, G.L., Singh, U.C., Weiner, P.K. & Kollman, P.A. (1995) AMBER. University of California, San Francisco. 35. Corchado, J.C., Chuang, Y.Y., Fast, P.L., Villa, W.P., Hu, W.P., Liu, Y.P., Lynch, G.C., Nguyen, K.A., Jackels, C.F., Gu, M.Z. et al. (2000) POLYRATE. University of Minnesota, Minneapolis. 36.Alhambra,C.,Sanchez,M.L.,Corchado,J.C.,Gao,J.& Truhlar, D.G. (2001) Quantum mechanical tunneling in methyl- amine dehydrogenase. Chem. Phys. Lett. 347, 512–518. 37. Alhambra, C., Corchado, J., Sanchez, M., Gao, J. & Truhlar, D. (2000) Quantum dynamics of hydride transfer in enzyme cata- lysis. J.Am.Chem.Soc.122, 8197–8203. 38. Hyun, Y L. & Davidson, V.L. (1995) Electron-transfer reactions between aromatic amine dehydrogenase and azurin. Biochemistry 34, 12249–12254. 39. Basran, J., Patel, S., Sutcliffe, M.J. & Scrutton, N.S. (2001) Importance of barrier shape in enzyme-catalyzed reactions. Vibrationally assisted hydrogen tunneling in tryptophan trypto- phylquinone-dependent amine dehydrogenases. J. Biol. Chem. 276, 6234–6242. 40. Knapp, M.J., Rickert, K. & Klinman, J.P. (2002) Temperature- dependent isotope effects in soybean lipoxygenase-1: correlating hydrogen tunneling with protein dynamics. J.Am.Chem.Soc. 124, 3865–3874. 41. Jang, M H., Basran, J., Scrutton, N.S. & Hille, R. (1999) The reaction of trimethylamine dehydrogenase with trimethylamine. J. Biol. Chem. 274, 13147–13154. 42. Trickey, P., Basran, J., Lian, L Y., Chen, Z W., Barton, J.D., Sutcliffe, M.J., Scrutton, N.S. & Mathews, F.S. (2000) Struc- tural and biochemical characterization of recombinant wild-type and a C30A mutant of trimethylamine dehydrogenase from Methylophilus methylotrophus (sp. W3A1). Biochemistry 39, 7678–7688. 43. Basran, J., Sutcliffe, M.J., Hille, R. & Scrutton, N.S. (1999) The role of Tyr 169 of trimethylamine dehydrogenase in substrate oxidation and magnetic interaction between FMN cofactor and the 4Fe/4S center. J. Biol. Chem. 274, 13155–13161. 44. Basran, J., Sutcliffe, M.J., Hille, R. & Scrutton, N.S. (1999) Reductive half-reaction of the H172Q mutant of trimethylamine dehydrogenase: evidence against a carbanion mechanism and assignment of kinetically influential ionizations in the enzyme– substrate complex. Biochem. J. 341, 307–314. 45. Basran, J., Sutcliffe, M.J. & Scrutton, N.S. (2001) Optimizing the Michaelis complex of trimethylamine dehydrogenase: identifica- tion of interactions that perturb the ionization of substrate and facilitate catalysis with trimethylamine base. J. Biol. Chem. 276, 42887–42892. 46. Basran, J., Sutcliffe, M.J. & Scrutton, N.S. (2001) Deuterium isotope effects during C-H bond cleavage by trimethylamine dehydrogenase: implications for mechanism and vibrationally assisted H-tunneling in wild-type and mutant enzymes. J.Biol. Chem. 276, 24581–24587. 47. Miller, J.R. & Edmondson, D.E. (1999) Structure–activity relationships in the oxidation of para-substituted benzylamine analogues by recombinant human liver monoamine oxidase A. Biochemistry 38, 13670–13683. 3102 M. J. Sutcliffe and N. S. Scrutton (Eur. J. Biochem. 269) Ó FEBS 2002 . MINIREVIEW A new conceptual framework for enzyme catalysis Hydrogen tunneling coupled to enzyme dynamics in flavoprotein and quinoprotein enzymes Michael J. Sutcliffe 1,2 and Nigel S. Scrutton 1 Departments. these enzymes. QUINOPROTEIN AND FLAVOPROTEIN AMINE DEHYDROGENASES The quinoprotein and flavoprotein amine dehydrogenases are ideally suited to studies of H-transfer. The reactions catalysed are. this understanding is gained, H -tunneling can then be used as a tool for (a) increasing the catalytic efficiency of enzymes in the biotechnology industry (by enhancing the coupling of dynamics to the reaction