CHEM. RES. CHINESE UNIVERSITIES 2010, 26(5), 833 837 Insight into Reaction Mechanism of Sirtuins via Molecular Simulation and Density Functional Theory Study ZHAO Yong-shan 1,3, HOU Rui-zhe 2, ZHANG Hong-xing 1, ZHENG Qing-chuan 1* and SUN Chia-chung 1 1. State Key Laboratory of Theoretical and Computational Chemistry, Institute of Theoretical Chemistry, 2. Norman Bethune College of Medicine, Jilin University, Changchun 130021, P. R. China; 3. School of Life Sciences & Biopharmaceutical Sciences, Shenyang Pharmaceutical University, Shenyang 110016, P. R. China Abstract With density functional theory(dft) and molecular mechanics method, the catalytic mechanism of silent information regulator(sirtuins) has been investigated. The calculations support the S N 2-like reaction of the initial step of the catalysis, and are consistent with experiment results. We further explored the second step of the catalysis and proposed that this step took place in a concerted reaction. In addition, the side chain of Phenylalanine33 may help to shield the glycosidic bond from water and be in a position to protect the developing oxacabenium transition state from hydrolysis. Our results of the calculations support this hypothesis that the phenylalanine33 plays a critical role in the sirtuins biology function. Keywords Sirtuin; NAD + ; p53; Model; Density functional theory Article ID 1005-9040(2010)-05-833-05 1 Introduction Sirtuins(or Sir2 protein deacetylases) comprise a broadly conserved family of NAD + -dependent protein deacetylases and mono ADP-ribosyltransferases whose function is crucial to the apoptosis and cell survival, transcriptional silencing, neurodegeneration and calorie restriction [1,2]. As a key player in a broad variety of biological processes, sirtuins may be attractive targets for the treatment of cardiovascular disease, diabetes, cancers and aging. A thorough understanding of the catalytic mechanism of these enzymes will facilitate the development of specific, mechanism-based regulators of sirtuins activity in order to treat the diseases related with these enzymes [3]. However, there are many questions about the reaction mechanism of sirtuins that are difficult to answer by experimental means alone. A key question regarding the reaction mechanism of sirtuins is the initial catalytic step. We are particularly interested in the role of Phenylalanine33(Phe33), which is located in the invariant flexible loop, in the initial reaction step and the inhibitory nicotinamide exchange reaction. By far, no theoretical studies have been found about active-site of sirtuins. In the present study, we use the classical molecular mechanics(mm) and the hybrid density functional theory(dft) method B3LYP [4] to explore the reaction mechanism of sirtuins, which has been extensively used to study other enzyme reaction mechanisms in recent years. 2 Chemical Models In the present study, the models of the active site of sirtuins were constructed on the basis of the crystal structure of Sir2Tm in the complex with NAD + and acetylated p53 peptide(pdb code 2H4F) [5]. Coordinates were extracted from the PDB files and hydrogen atoms were added via InsightII/Biopolymer module [6]. In the first model(denoted as model A), the following groups are included(fig.1): (1) NAD +, which included nicotinamide group and N-ribose group, truncated in one side after the ribose ring. The adenine group and the part of the phosphoric acid remain their position during the reaction process as the NAD + binds in the substrate cleft *Corresponding author. E-mail: zhengqc@jlu.edu.cn Received January 12, 2010; accepted March 31, 2010. Supported by the National Natural Science Foundation of China(No.20573042, 20903045), Specialized Research Fund for the Doctoral Program of Higher Education of China(No.20070183046) and Specialized Fund for the Basic Research of Jilin University, China(No.200810018).
834 CHEM. RES. CHINESE UNIVERSITIES Vol.26 which forms extensive hydrogen bonds and van der Waals interactions with the Rossmann fold domain [7]. (2) Acetyl p53 peptide substrate, which was modeled on the basis of the structure of the acetylated peptide, in which the acetyl lysine analog was remained. (3) The phenyl group of Phe33, to test the proposal that this group plays an important role in the catalytic reaction process. (4) The imidazolyl group of His116, which abstracts a proton in the second step of the catalytic mechanism [8]. At last the quantum chemical model employed comprises 71 atoms and consists of the parts mentioned above. This approach is used to keep the various groups in place to as much as possible resemble the crystal structure. These atoms are indicated by arrows in Fig.1(A) and 1(B). 3 Computational Details 3.1 Molecular Mechanics To obtain the reasonable structure for the complex of Sir2Tm and modified O-alkylamidate intermediate, a conjugate gradient energy minimization was performed until the root mean-square gradient was lower than 4.18 10 4 kj/(mol nm). The calculation mentioned above was performed on the SGI O3900 workstations via InsightII software package developed by Accelrys [9]. The Extensible and Systematic Force Field(ESFF) [10] was used for energy minimization. 3.2 Density Functional Theory Calculations Fig.1 Model A of 2H4F active site in the present calculation(a) and model B of optimized 2H4F complexes(b) Arrows indicate centers frozen to their X-ray coordinates. To explore the second step of the catalytic mechanism and the role of Phe33, the second model(denoted as model B) was also generated from the X-ray crystal structure of 2H4F. O-Alkylamidate intermediate was formed by removing the nicotiamide part of NAD + and linking the oxygen atom of the acetyl lysine to the C1 of the N-ribose, and then the energy of the entire modified system was minimized via MM method. Finally, model B was truncated and included the O-alkylamidate intermediate analog, the side chain of His116 and Phe33(Fig.1). With all these parts, the net charges of these two models are +1. In the geometry optimization, some certain atoms, typically those at which the truncation was done, were kept frozen to their X-ray positions. All of the Quantum Mechanics(QM) calculations were performed with Gaussian 03 program package [11] and all the geometries and energies of the models which were truncated from the crystal structure in the present study were computed with the three-parameter hybrid exchange functional of Becke and Lee, Yang, and Parr correlation functional(b3lyp) method [4]. Geometry optimizations were performed with the double-zeta plus polarization basis set 6-31G(D,P). On the basis of these geometries, the larger basis set 6-311+G(2D,2P) was employed for single-point calculations to obtain more accurate energies with the solvation energies added via the conductor-like solvation model COSMO at the B3LYP/6-31G(D,P) level [12]. In these models, a cavity around the system was surrounded by a polarizable dielectric continuum. The dielectric constant chosen is ε=4, as is known that the standard value is used to model the protein surroundings [13,14]. To obtain the transition states, the potential energy curve scans were performed. Hessians were calculated to confirm the nature of the stationary points, with no negative eigenvalues for minima and only one negative eigenvalue for the transition state. In the present study, some heavy atoms were fixed in their crystal structure position that may give rise to a few small negative eigenvalues for the optimized structures. Fortunately, these are very small in the order from 5i to 15i cm 1, and do not affect the obtained results [15,16].
No.5 ZHAO Yong-shan et al. 835 4 Results and Discussion 4.1 Nucleophilicity of Acetyl-oxygen and Disassociation of Nicotinamide Previous studies [17,18] suggest two different mechanisms for the first step. Except for the S N 2-like mechanism which was proposed by Avalos et al. [17,18], there is another possible mechanism called the dissociative mechanism based on the results of experiments done by Slama et al. [19] and Zhao et al. [20], which suggests that the acetyl group does not chemically take part in nicotinamide cleavage, but helps to position NAD + in a destabilizing conformation that makes nicotinamide cleavage, forming a distinct oxocarbenium intermediate instead. Recently, the experimental results have shed light on the first step in the proposed mechanism [21,22]. Together with the previously experimental data, the first step was explored by QM method in detail and calculational results allow us to propose a S N 2-like mechanism. The structures of the optimized reactant, transition state and intermediate are shown in Fig.2. As seen from Fig.2(A), this structure shows a quite high resemblance to the crystal structure, where the distance between the oxygen of acetyl lysine and C1 of the ribose is 0.316 nm. This critical distance is 0.217 and 0.151 nm at the transition state and intermediate, respectively(fig.2b, C), and the α face of Fig.2 Optimized structures of model A(A), transition state for the acetyl-oxygen attack on C1 of N-ribose(B) and resulting O-alkylamidate intermediate and the nicotiamide molecule(c) N-ribose is routed in order to prepare for attack of acetyl-lysine. The C1 of the ribose flats about 0.165 nm to complete the reaction according to the results mentioned above, which is consistent with the experiment [5]. Frequency calculations confirmed the nature of the transition state with only imaginary frequency of 1624i cm 1. The reactant has overcome the barrier energy 111.94 kj/mol, and arrives to the transition state. This high energy barrier is due in great part to the small model that is comprised of only two residues of the enzyme and a hydrogen bond formed between the oxygen of the acetyl lysine and the 3 hydroxyl of the ribose at the reactant, which was broken during the O-alkylamidate intermediate forming. We made several attempts to locate an oxocarbenium intermediate, i.e., corresponding to a dissociative mechanism, but no success was obtained. By this, we can of course not completely rule out this possibility, but the results obviously support that the initial step of the catalysis is an S N 2-like reaction. 4.2 The Role of Phe33 The Phe33 which is conserved in all the family of sirtuins appears to play a critical role in the initial reaction step and affects the functions of the enzyme. An experimental study has proved that the mutation of the corresponding residue in yeast sirtuin results in a decrease of deacetylation activity of 90% and a loss of silencing in vivo [23]. A large amount of reports suggest that the phenyl ring of the Phe33 moves to stack above ribose oxygen in order to protect the glycosidic bond from water and maybe allow for pi-electron stabilization when the transient carbocation is formed in the ternary complex. This position has been supposed to prevent the dissociated nicotinamide from reacting with the O-alkylamidate intermediate [24,25]. As shown in Fig.2(C), the phenyl ring of the Phe33 changes in position, with the movement of the ribose when the O-alkylamidate intermediate is formed. The distance between the center of phenyl and C1 of the ribose is 0.638 nm, which indicates that the phenylalanine has no effect on the pi-electron stabilization. To investigate the position of Phe33 when the O-alkylamidate was formed and the nicotinamide was released, the structure of model B was optimized(fig.3). As seen in Fig.3, the phenyl ring of Phe33 shifts to a position above the β face of the intermediate and helps to shield it from the solvent.
836 CHEM. RES. CHINESE UNIVERSITIES Vol.26 Based on the results mentioned above, we can conjecture that once the nicotinamide is cleaved and released, Phe33 is allowed to rotate and stack against the β face of the O-alkylamidate intermediate in order to shield the intermediate from the solvent and inhibit the base-exchange reaction. In contrast, the long distance between the center of phenyl and C1 of the ribose suggests Phe33 may not play the role in stabilizing the transient carbocation through pi-electron stabilization, which is in good agreement with the experiment by Sanders et al. [25]. Fig.3 Optimized structure of model B used in our calculations 4.3 Protons Transfer Between His116 and N-Ribose Model B without Phe33 was employed to study the second step of the reaction mechanism for Phe33 contribution to the reaction energy. In the optimized reactant structure(fig.4a), His116 forms a hydrogen bond with 3 hydroxyl of ribose and the length of this H-bond is 0.175 nm. To test whether the second step is a stepwise reaction or concerted reaction, a linear transit scan was performed in which the H N distance was fixed in the steps between 0.175 nm(the distance at the reactant) and 0.100 nm, while all other degrees of freedom were optimized. As is seen in Fig.5, no energy minimum could be found corresponding to an intermediate where the proton is transferred to the His116. Based upon the computational results, we can suggest that the His116, acting as a general base, abstracts a proton from 3 hydroxyl and the concerted proton transfers from 2 hydroxyl to 3 hydroxyl, activating the 2 oxygen to attack the acetyl group, forming the cyclic amino-acetyl intermediate. We have managed to locate the exact transition state for this reaction,that is to say, the mechanism for the second step reaction is a concerted reaction. The structures of the optimized reactant, the transition state and the product are displayed in Fig.4(A) (C), respectively. The reactant has overcome the barrier energy of 71.18 kj/mol to arrive the transition state. Both the bond lengths of O H in N-ribose increase to 0.127 nm, leading to the break of the bonds to form the transition state. Frequency calculations confirm the nature of the transition state with only imaginary frequency of 1068i cm 1. Fig.5 Potential energy curves for moving the proton from the 2 oxygen of the intermediate to NE of the His116 5 Conclusions Fig.4 Optimized structures of model B without Phe33 residue(a), transition state for hydrogen atom transfer(b), resulting cyclic amino-acetyl intermediate(c) In the present study, the reaction mechanism of sirtuins has been investigated with MM and QM methods. The calculations indicate that the first step of the reaction of sirtuins enzyme takes place in a concerted S N 2-like step. The second step of the catalytic mechanism has been explored and the results show that subsequent proton is abstracted from 2 hydroxyl of N-ribose to His116 coupled with proton transferred
No.5 ZHAO Yong-shan et al. 837 from 2 hydroxyl to 3 hydroxyl, which is a concerted reaction. In addition, we have also examined the role of Phe33, and the results of the calculations show that the Phe33 plays an important role in the first step of catalytic mechanism, which positions above the ribose oxygen adjacent to the ribose C1 to prevent nicotinamide exchange reaction. Our results may be helpful for further kinetic and structural investigations of the importance of sirtuins. References [1] Sauve A. A., Wolberger C., Schramm V. L., Boeke J. D., Annu. Rev. Biochem., 2006, 75, 435 [2] Tanny J. C., Dowd G. J., Huang J., Hilz H., Moazed D., Cell, 1999, 99, 735 [3] Yamamoto H., Schoonjans K., Auwerx J., Mol. Endocrinol., 2007, 21, 1745 [4] Becke A. D., J. Chem. Phys., 1993, 98, 5648 [5] Hoff K. G., Avalos J. L., Sens K., Wolberger C., Structure, 2006, 14, 1231 [6] Biopolymer User Guide, Accelrys Inc., San Diego, 1999 [7] Avalos J. L., Bever K. M., Wolberger C., Mol. Cell, 2005, 17, 855 [8] Smith B. C., Denu J. M., Biochemistry, 2006, 45, 272 [9] InsightII, Version 98.0., Accelrys Inc., San Diego, 1998 [10] Discover3 User Guide, Accelrys Inc., San Diego, 1999 [11] Frisch M. J., Trucks G. W., Schlegel H. B., Scuseria G. E., Robb M. A., Cheeseman J. R., Montgomery J. A. Jr., Vreven T., Kudin K. N., Burant J. C., Millam J. M., Iyengar S. S., Tomasi J., Barone V., Mennucci B., Cossi M., Scalmani G., Rega N., Petersson G. A., Nakatsuji H., Hada M., Ehara M., Toyota K., Fukuda R., Hasegawa J., Ishida M., Nakajima T., Honda Y., Kitao O., Nakai H., Klene M., Li X., Knox J. E., Hratchian H. P., Cross J. B., Bakken V., Adamo C., Jaramillo J., Gomperts R., Stratmann R. E., Yazyev O., Austin A. J., Cammi R., Pomelli C., Ochterski J. W., Ayala P. Y., Morokuma K., Voth G. A., Salvador P., Dannenberg J. J., Zakrzewski V. G., Dapprich S., Daniels A. D., Strain M. C., Farkas O., Malick D. K., Rabuck A. D., Raghavachari K., Foresman J. B., Ortiz J. V., Cui Q., Baboul A. G., Clifford S., Komaromi I., Martin R. L., Fox D. J., Keith T., Al-Laham M. A., Peng C. Y., Nanayakkara A., Challacombe M., Gill P. M. W., Johnson B., Chen W., Wong M. W., Gonzalez C., Pople J. A., Gaussian 03, Revision C.02, Gaussian, Inc., Wallingford, 2004 [12] Cammi R., Mennucci B., Tomasi J., J. Phys. Chem. A, 1999, 103, 9100 [13] Hopmann K. H., Hallberg B. M., Himo F., J. Am. Chem. Soc., 2005, 127, 14339 [14] Himo F., Siegbahn P. E. M., J. Am. Chem. Soc., 2001, 123, 10280 [15] Himo F., Theor. Chim. Acta., 2006, 116, 232 [16] Curtiss L. A., Krishnan R., Redfern P. C., Pople J., J. Chem. Phys., 1997, 106, 1063 [17] Avalos J. L., Boeke J. D., Wolberger C., Mol. Cell, 2004, 13, 639 [18] Chang J. H., Kim H. C., Hwang K. Y., Lee J. W., Jackson S. P., Bell S. D., Cho Y., J. Biol. Chem., 2002, 277, 34489 [19] Slama J. T., Simmons A. M., Biochemistry, 1988, 27, 183 [20] Zhao K., Harshaw R., Chai X., Marmorstein R., Proc. Natl. Acad. Sci. USA, 2004, 101, 8563 [21] Smith B. C., Denu J. M., J. Am. Chem. Soc., 2007b, 129, 5802 [22] Hawse W. F., Hoff K. G., Fatkins D. G., Daines A., Zubkova O. V., Schramm V. L., Zheng W., Wolberger C., Structure, 2008, 16, 1368 [23] Armstrong C. M., Kaeberlein M., Imai S. I., Guarente L., Mol. Biol. Cell, 2002, 13, 1427 [24] Zhu X., Heine A., Monnat F., Houk K. N., Janda K. D., Wilson I. A., J. Mol. Biol., 2003, 329, 69 [25] Sanders B. D., Zhao K. H., Slama J. T., Marmorstein R., Mol. Cell, 2007, 25, 463