English

• The α-ketoglutarate (α-KG) dependent dioxygenases are the major class of non-heme Fe(Ⅱ)-dependent enzymes catalyzing a great variety of reactions ranging from the simple hydroxylation to the complicated skeletal rearrangement. This enzyme superfamily, consisting of more than 60 enzymes, are widely distributed throughout the living organisms including human beings [1]. They play essential roles in a lot of biological processes including DNA repairing, fatty acid metabolism, oxygen sensing, and biosynthesis secondary metabolites [2-5]. Generally, the α-KG dependent dioxygenases need α-ketoglutarate and dioxygen to oxidize various substrates (Scheme S1 in Supporting information). These reactions share two common features including the decarboxylation of α-KG in the form of carbon dioxide and the formation of a reactive Fe(Ⅳ)-oxo intermediate [6]. 4-Hydroxyphenylpyruvate dioxygenase (HPPD) is a conventional α-KG dioxygenase utilizing the dioxygen to converts 4-hydroxyphenylpyruvate acid (HPP) to homogentisic acid (HG) (Scheme S2 in Supporting information). This reaction is the key step for the cascade reaction of tyrosine catabolism and biosynthesis of plastoquinones and tocopherols [7] happened in plant. In addition, the HPPD is also able to catalyze the hydroxylation of other α-keto compound including phenylpyruvate and isocaproate [8, 9]. Due to the unique roles of HPPD in life, HPPD has been recognized as an important target for herbicides [10-15] and for diseases such as the treatment of type I tyrosinemia and alkaptonuria [16-18].

  Interestingly, the uncatalyzed conversion reaction of HPP to HG is spin forbidden because of inconsistent spin states for the reactants and the products (Scheme 1). Thus, the HPPD should play an important role in inverting the spins of reaction system during the catalyzed reaction. It is well known that the catalytic center of HPPD (as well as other α-dependent dioxygenases) is in the quintet state [19], in which the ferrous ion is in high spin with S_Fe = 2. Considering the addition of a triplet O₂ molecule and singlet HPP to a quintet ferrous center, there are three possible spin states that can be resulted: triplet, quintet, and septet. Previous study suggested that the septet dioxygen adducts, where a triplet dioxygen weakly interact with the ferrous center, was the most stable pre-reactive complex [6, 20, 21]. On the other hand, the Fe(Ⅳ)-oxo intermediate was proposed to be in quintet spin based on an Mössbauer and electron paramagnetic resonance spectroscopies experiment on Taurine/α-ketoglutarate dioxygenase [22]. Therefore, the dioxygen activation catalyzed by HPPD and other α-KG dependent dioxygenase should cross different spin potential surfaces.

  Scheme 1

  

  Scheme 1.  Spin states for each reactant and product for the un-catalyzed conversion reaction of HPP to HG.

  DownLoad: Full-Size Img PowerPoint

  Three reaction routes for the dioxygen activation reaction catalyzed by α-KG dependent dioxygenase activation reaction have been advanced so far [6, 23-25]. In the first route, this reaction mainly runs on a quintet PES consisting of steps including the formation of an Fe(Ⅲ)-O₂ adduct, and/or the generation of bicyclic superoxo intermediate [26], decarboxylation, and O—O bond heterolytic cleavage [6, 20, 25]. In the second route, the initial two steps including the formation of the O₂ adduct and the decarboxylation and O—O bond cleavage run on a septet surface. Then the generated Fe-oxo intermediate switches to the quintet spin by intersystem crossing [6, 27]. In the last route, the binding of triplet dioxygen to the quintet catalytic metal center affords a triplet Fe(Ⅳ)-alkyl peroxo bridged intermediate. Then this structure decay to a quintet Fe(Ⅱ)-peracid intermediate via a triplet-quintet spin crossing, followed by a heterolytic O—O cleavage step (on quintet surface) affording the Fe(Ⅳ)-oxo intermediate [28]. Wójcik et al. carefully studied these reaction routes with dispersion-corrected DFT methods employing B3LYP functionals [6], the robust of which was supported by CCSD(T) benchmark calculations. Their calculation results support the first reaction route.

  Despite the significance of the theoretical studies mentioned above, these works were based on quantum mechanical (QM) calculations on various active site models. The credibility of the related conclusions should be checked by fully accounting for the effects of the protein environment. Herein, we report a systematic quantum mechanics/molecular mechanics (QM/MM) reaction coordinate study on the HPPD-catalyzed O₂ activation reaction. QM/MM takes advantage of the accuracy of QM method and the efficiency of MM method. This methodology has been widely used in studying a great variety systems ranging from biological system [29] to organometallic catalysts [30]. In these studies, we investigated the reaction of O₂ with HPP-HPPD complex running on different potential energy surfaces with triplet/quintet/septet spin. In addition, the intersection of potential surface with different spin states were also investigated. These studies suggested that the most favorable reaction pathway successively runs on a septet and a quintet PES, which are joined by a minimum energy crossing point (MECP), namely ^5-7M2. For the sake of clarity, in this study intermediates, transition states and MECP structures are labeled as ^mXn, where X stands for either I (intermediate) or T (transition state) or M (MECP structure) and m stands for multiplicity. Computational details are described in Supporting information. The definition of QM/MM region are defined in Fig. S1 (Supporting information). Below is a detailed description of this work.

  There are five possible spin states for the HPP-HPPD-O₂ complex, namely singlet, open singlet, triplet, quintet and septet. In our previous studies [21], the singlet and open singlet spin were found to be ~12.3–28.4 kcal/mol higher in total energy than the other three spin states. Hence, only triplet, quintet and septet spin are considered in this study. The QM/MM geometrical optimizations on the ternary HPP-HPPD-O₂ complex at these spin states leading to four pre-reactive structures, namely, ³I1′, ³I1, ⁵I1 and ⁷I1′ (Fig. S2 in Supporting information). The relative energies of these structures are collected in Table S1 (Supporting information). Previous study proved that the HPP conformation presented in the HPPD catalytic site was an active conformation [29]. In triplet state, two pre-reactive complexes, i.e., ³I1′ and ³I1 are formed (Figs. S2A and B). In ³I1′ the dioxygen loosely interacts with the ferrous ion with O-Fe distance of 2.7 Å, whereas in ³I1 the dioxygen tightly binds with the ferrous ion with O-Fe distance of 1.9 Å. In addition, the dioxygen in ³I1 also approaches the keto carbon of HPP with O—C distance of 2.3 Å. In quintet state, only one local minima structure ⁵I1 is located in which the dioxygen binds with the metal ion with Fe-O distance of 2.1 Å (Fig. S2C). Similarly, at septet state only one minima ⁷I1′ (Fig. S2D) is located. In this structure the dioxygen is detached from the ferrous ion with Fe-O distance of 3.1 Å. The geometrical parameters for ³I1′, ³I1, ⁵I1 and ⁷I1′ complexes are in consistent with the results obtained from previous studies on QM models of α-KG dependent dioxygenase by Wójcik et al. [6] and Ye et al. [31]. Spin and charge analysis shows that in ³I1′ and ⁷I1′ the high-spin (S_O2 = 1) dioxygen respectively coupled antiferromagnetically and ferromagnetically with the high-spin (S_Fe = 2) ferrous ion without charge transfer. Compared with the electronic structure of ³I1′, in ³I1 the dioxygen changes into doublet (S_O2 = 1/2) O₂⁻ upon exchanging one beta electron with two alpha electrons from the ferrous atom, and the latter becomes doublet (S_Fe = 1/2) Fe(Ⅲ). In ⁵I1 the dioxygen is roughly in singlet state and the ferrous ion is in high spin (S_Fe = 2), suggesting that one alpha or beta electron from the dioxygen has been flipped to the opposite spin during the binding of triplet O₂ to the quintet ferrous ion. According to Table S1, the ³I1′ and ⁷I1′ are the most two stable structures with relative total energy of 0.0 and 1.0 kcal/mol, respectively. These values are slightly different to the ones obtained from Salter's work which predicted that the septet structure was 0.5 kcal/mol more stable than the triplet one. This difference may be caused by the protein environment. The other two complexes are 6.4–21.7 kcal/mol less stable than ³I1′ or ⁷I1′, indicating that the binding of O₂ to the ferrous ion will generally increase the internal energy of the HPP-HPPD-O₂ complex.

  Started from the ³I1 structure, the dioxygen activating reaction was investigated by reaction coordinate calculations. The results show that the reaction run on triplet surface consists only one-step in which electrophilic attacking of dioxygen and decarboxylation happens simultaneously. Depicted in Fig. S3 (Supporting information) are the key geometrical parameters for the transition state ³T2 and subsequent intermediate ³I3 along the reaction pathway. In ³T2 the attacking atom, namely Oa, has already attached with the keto carbon C2′ with Oa-C2′ distance of 1.4 Å, and the carboxylate group is leaving the keto carbon with C2′-C3′ distance 1.8 Å. Compared to ³I1, in this structure the net spin for the -CO₂ group raised to −0.45 and 0.39, whereas the spin for the dioxygen decreased to 0.37, suggesting that partial β electron density has been transferred from the carboxylic group to the dioxygen. In ³I3 the carboxylic group leaves the keto carbon with C2′-C3′ distance 2.9 Å. However, this group is still in a V-shaped conformation, indicating that it does not changes into the expected linear-shaped carbon dioxide. Electronic structure analysis on ³I3 also shows that the carboxylic group is still bonded with the nearby iron atom, as evidenced by the net spin value of 0.85 and net charge value of −0.46 for this group.

  The reaction route runs on the septet surface consists of two steps, dioxygen addition and decarboxylation (Fig. S4 in Supporting information). Geometrical parameters for the first transition state ⁷T1 are quite resembles the ones in ³T2 with Oa-Fe distance of 1.4 Å and C2′-C3′ distance 1.8 Å. However, this reaction step leads to a peroxyl adduct ⁷I2 in which the carboxyl group is still weakly bonded with the keto carbon with C2′-C3′ distance of 2.1 Å. In the subsequent decarboxylation step, the carboxyl group leaves the keto group affording the intermediate ⁷I3. Similar to the case of ³I3, in ⁷I3 the V-shaped carboxyl group is still covalently bonded with the metal ion with Fe-O distance 2.1 Å. Electronic structure analysis shows that during these two steps two β electrons, one from the carboxyl group and the other from the metal ion, are transferred to the dioxygen.

  Reaction coordinate calculations show that the reaction run on quintet surface consists of three steps, namely, dioxygen addition, decarboxylation, and O—O heterolysis (Fig. S5 in Supporting information). In the dioxygen addition step (Figs. S5A and B), the dioxygen attaches with the HPP keto carbon affording intermediate ⁵I2. During the time, one α electron is transferred from the metal ion to the dioxygen. As results, the spin for the metal ion decreases from 3.82 (in ⁵I1) to 2.84 (in ⁵I2), whereas the spin for the dioxygen increases from −0.15 (in ⁵I1) to 1.07 (in ⁵I2). In the subsequent decarboxylation step (Figs. S5C and D), a carbon dioxide molecule is released from the C2′ atom, leading to intermediate ⁵I3. In this step, one α electron has been transferred from the carboxylic group to the metal ion and one β electron is transferred from the same group to the dioxygen. In ⁵I3 (Fig. S5D) the spin for the metal ion and dioxygen are respectively 3.72 and 0.10, as compared with the counterpart values of 2.84 and 1.07 in ⁵I2 (Fig. S5B). After losing a pair of electrons, the carboxylic group is able to leave the C2′ as carbon dioxide. In the following O—O heterolysis step (Fig. S6 in Supporting information), an HPA is generated along with a ferryl oxygen. During this step, the high-spin oxo atom has taken one α electron from the metal ion. Accordingly, the spin for the metal ion decreases from 3.72 (⁵I3) to 3.01 (⁵I4). The reaction route and electronic structures described here for the quintet spin is generally consistent with the results obtained by Wójcik et al. [6] and by Ye et al. [31].

  The relative free energies and Gibbs free energies for the intermediate and transition state structures located at potential surfaces with different spins are collected in Fig. 1. Among the three profiles described therein, the septet surface has the lowest free energy barrier (19.9 kcal/mol), whereas the quintet surface has highest free energy barrier (37.4 kcal/mol). The free energy barriers for septet and triplet are consistent with the values of 19.1 kcal/mol for septet and 28.8 kcal/mol for triplet obtained by Wójcik et al. [6]. However, the free energy barrier for quintet surface present in this study differs significantly to the corresponding value of 18.8 kcal/mol from that work. This difference may be caused by the protein environment of HPPD. Despite the relatively low free energy barrier on the septet surface, the reaction is not likely to run totally on this surface because the routine product ⁷I3 are thermodynamically unfavorable with Gibbs free energy lies 14.6 kcal/mol above that of the pre-reactive complex ⁷I1′. Instead, ⁵I3 is the most favorable decarboxylated intermediate with relative Gibbs free energy of −35.7 kcal/mol. This implies that at least part of quintet surface involves in the lowest energy reaction pathway (LERP). On the other hand, the most stable pre-reactive complexes are ³I1′ and ⁷I1′ complex, suggesting that the initial part of the LERP should be run on triplet or septet surface. Put all the discussions together, it reaches that the electron spin inversion should happen in the dioxygen attacking process.

  Figure 1

  

  Figure 1.  Relative energies and Gibbs free energies of stationary points and MECPs located in HPPD-catalyzed dioxygen activation reaction obtained at the QM/MM B3LYP-D3/6-311++G(d, p): AMBER level. The reaction coordinate was set as R_C2′-C3′ - R_Oa-C2′ (the difference of the interatomic distance between atom pairs C2′-C3′ and Oa-C2′) for the steps covering I1 to I3 and was set as R_Oa-Ob for the step covering I3 to I4. Profile color codes: yellow for triplet, red for quintet, green for septet and blue for MECP.

  DownLoad: Full-Size Img PowerPoint

  To investigate the possible electron spin inversion in the reaction pathway, a series of geometrical optimization from different initial points on the potential surfaces were performed in order to find the MECP points connecting the triplet/quintet and quintet/septet potential energy surface. To do so, a revised Harvey's algorithm [32] was employed along with the Gaussian-Amber QM/MM program. Four MECP points, namely ^5-7M2 (Fig. 2), ^3-5M1, ^3-5M2 and ^5-7M1 (Fig. S7 in Supporting information), were located in these calculations. We characterized MECP structures by performing geometrical optimization using the respective spins. These calculations showed that ^3-5M1 and ^5-7M1 connected the pre-reactive complexes ³I1/⁵I1 and ⁵I1/⁷I1′, respectively. Thus, these MECPs will not affect the overall energetic of the reaction. ^3-5M2 connects the pre-reactive ³I1 and intermediate ⁵I3. This MECP is thermodynamically unfavorable with relative energy of 29.0 kcal/mol (Fig. 1). ^5-7M2 is located near the transition states ⁵T2 and ⁷T1 (Fig. S8 in Supporting information). Collected in Table S2 (Supporting information) are the key interatomic distances as well as reaction coordinates for these three structures. As can be seem from this table, the parameters for ^5-7M2 are rightly between the corresponding ones for ⁵T2 and ⁷T1. Geometrical optimizations on ^5-7M2 with quintet and septet spin proved that this MECP connects the ⁷I1′ at the one end and ⁵I3 at the other. From the spin values list in Fig. 2, it could be found that in ^5-7M2 the spin value for the ferrous ion is nearly the same in both spin state. This implies that one α electron from the orbital(s) formed by dioxygen and the citrate group flips to the opposite spin in this point. Interestingly, the ^5-7M2 lies 3.0 kcal/mol below ⁷T1, 20.5 kcal/mol below ⁵T2 and 15.9 kcal/mol below ³T2 (Fig. 1), suggesting that the ⁷I1′-^5-7M2-⁵I3 routine is the most favorable pathway for the dioxygen activation reaction catalyzed by HPPD. The MECP ^5-7M2 structure presented in this study was not reported in previous works, which were based on QM calculations. The finding of ^5-7M2 structure implies that the protein environment will largely re-shape the potential surface of the catalytic core of HPPD and may as well as other α-KG dependent dioxygenases.

  Figure 2

  

  Figure 2.  Key geometrical parameters and spins for atoms in the catalytic site of ^5-7M2.

  DownLoad: Full-Size Img PowerPoint

  The above discussions suggest that the release of carbon dioxide can only occur in the ⁵I3-formating reaction step. It is beneficial to address why quintet spin is necessary for the CO₂ releasing. Collected in Fig. 3 are the qualitative molecular orbital (MO) schemes of ^3/5I1, ⁷I1′, and ^3/5/7I3. More MO schemes are presented in Figs. S9 and S10 (Supporting information). In the pre-reactive complex ³I1 (Fig. 3A), five electrons are distributed in the Fe 3d-based t_2g MO in a low-spin manner (S_Fe = 1/2), leaving the high-energy e_g MO empty. The dioxygen has withdrawn an electron from the metal ion affording a peroxyl radical anion (S_O2 = 1/2). In the corresponding intermediate ³I3 (Fig. 3B), this dioxygen further withdraws a β electron from the σ MO of the carboxyl group, leaving an unpaired α electron in that σ MO. The carboxyl group must give out this α electrons in order to leave the substrate as carbon dioxide. The only possible destination for this transfer is the e_g MOs of the ferric ion, since the dioxygen π^* MOs are fully occupied after accepting two electrons. However, this transfer is thermodynamically unfavorable due to the high energy of the e_g MOs. Thus, this α electron will stay within the carboxyl group leading to the failure of carbon dioxide release. Similar α electron transferring problem happens in the septet spin. In the pre-reactive complex ⁷I1′ (Fig. 3E), four unpaired α electrons are distributed in ferrous t_2g and e_g MO in a high-spin manner (S_Fe = 2). In addition, two α electrons align parallelly in the π^* MO of dioxygen. In the intermediate ⁷I3, the dioxygen has taken one β electron from the carboxyl group. However, neither the dioxygen nor the ferrous ion is able to taken the remaining α electron from the carboxyl group since all destination MOs of dioxygen and ferrous ion are occupied by unpaired α electrons (Fig. 3F). Different to ³I1 or ⁷I1′, in ⁵I1 structure (Fig. 3C) the dioxygen is in a singlet state (S_O2 = 0). This unique structure enables the dioxygen capable of accommodating a pair of electrons with opposite spins delivered from the carboxyl group (Fig. 3D) and thus facilitate the departure of this group as carbon dioxide. The present analysis shows that the electronic structure of dioxygen in the pre-reaction complex dominate the decarboxylation reaction.

  Figure 3

  

  Figure 3.  The transmission of electrons for the decarboxylation reaction run in different spin state.

  DownLoad: Full-Size Img PowerPoint

  According to the calculation results presented in this study, the dioxygen activation reaction in HPPD is rate-determined by decarboxylation step with predicted free energy barrier of 16.9 kcal/mol. This value is in good agreement with the experimentally determined value of 16–17 kcal/mol (rate constants of 1.1–7.8 s⁻¹ [33-36], suggesting that the present theoretical calculations are reliable.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  The work was supported by the National Key R & D Program (No. 2021YFD1700100) and National Natural Science Foundation of China (Nos. 21837001, 21273089), the Open Project Fund of the Key Laboratory of the Pesticides and Chemical Biology of Central China Normal University (No. 2018-A01), the Fundamental Research Funds for the Central Universities, the Fundamental Research Funds for the South-Central University for Nationalities (No. CZW20020). We thank the high-performance computing center (HPC) platform of Huazhong University of Science and Technology and HPC of University of Kentucky for providing computing resources.

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2022.107803.

  
  
• 1. [1]

     Y. Nakashima, T. Mori, H. Nakamura, et al., Nat. Commun. 9 (2018) 104. doi: 10.1038/s41467-017-02371-w

  2. [2]

     G.R. Moran, Arch. Biochem. Biophys. 433 (2005) 117–128. doi: 10.1016/j.abb.2004.08.015

  3. [3]

     B.I. Fedeles, V. Singh, J.C. Delaney, et al., J. Biol. Chem. 290 (2015) 20734–20742. doi: 10.1074/jbc.R115.656462

  4. [4]

     F. Ndikuryayo, B. Moosavi, W.C. Yang, et al., J. Agric. Food Chem. 65 (2017) 8523–8537. doi: 10.1021/acs.jafc.7b03851

  5. [5]

     A. Santucci, G. Bernardini, D. Braconi, et al., J. Med. Chem. 60 (2017) 4101–4125. doi: 10.1021/acs.jmedchem.6b01395

  6. [6]

     A. Wójcik, M. Radoń, T. Borowski, J. Phys. Chem. A 120 (2016) 1261–1274. doi: 10.1021/acs.jpca.5b12311

  7. [7]

     P. He, G.R. Moran, Curr. Opin. Chem. Biol. 13 (2009) 443–450. doi: 10.1016/j.cbpa.2009.06.012

  8. [8]

     N.P. Crouch, M.H. Lee, T.I. Bueno, et al., Methods Enzymol. 324 (2000) 342–355.

  9. [9]

     M. Kohlmeier, Amino Acids and Nitrogen Compounds, Academic Press, San Diego, 2015.

  10. [10]

      R. Beaudegnies, A.J.F. Edmunds, T.E.M. Fraser, et al., Bioorg. Med. Chem. 17 (2009) 4134–4152. doi: 10.1016/j.bmc.2009.03.015

  11. [11]

      H. Ahrens, G. Lange, T. Muller, et al., Angew. Chem. Int. Ed. 52 (2013) 9388–9398. doi: 10.1002/anie.201302365

  12. [12]

      D.W. Wang, H.Y. Lin, R.J. Cao, et al., J. Agric. Food Chem. 62 (2014) 11786–11796. doi: 10.1021/jf5048089

  13. [13]

      D.W. Wang, H.Y. Lin, R.J. Cao, et al., J. Agric. Food Chem. 63 (2015) 5587–5596. doi: 10.1021/acs.jafc.5b01530

  14. [14]

      D.W. Wang, H.Y. Lin, R.J. Cao, et al., Pest Manag. Sci. 71 (2015) 1122–1132. doi: 10.1002/ps.3894

  15. [15]

      D.W. Wang, H.Y. Lin, B. He, et al., J. Agric. Food Chem. 64 (2016) 8986–8993. doi: 10.1021/acs.jafc.6b04110

  16. [16]

      S. Lindstedt, E. Holme, E.A. Lock, et al., Lancet 340 (1992) 813–817. doi: 10.1016/0140-6736(92)92685-9

  17. [17]

      J. Larochelle, F. Alvarez, J.F. Bussieres, et al., Mol. Genet. Metab. 107 (2012) 49–54. doi: 10.1016/j.ymgme.2012.05.022

  18. [18]

      D.C. Bartlett, C. Lloyd, P.J. McKiernan, et al., J. Inherit. Metab. Dis. 37 (2014) 745–752. doi: 10.1007/s10545-014-9683-x

  19. [19]

      E.I. Solomon, T.C. Brunold, M.I. Davis, et al., Chem. Rev. 100 (2000) 235–350. doi: 10.1021/cr9900275

  20. [20]

      T. Borowski, A. Bassan, P.E. Siegbahn, Biochemistry 43 (2004) 12331–12342. doi: 10.1021/bi049503y

  21. [21]

      X.N. Wang, H.Y. Lin, J.J. Liu, et al., Chin. Chem. Lett. 32 (2021) 1920–1924. doi: 10.1016/j.cclet.2021.02.041

  22. [22]

      J.C. Price, E.W. Barr, B. Tirupati, et al., Biochemistry 42 (2003) 7497–7508. doi: 10.1021/bi030011f

  23. [23]

      M.R.A. Blomberg, T. Borowski, F. Himo, et al., Chem. Rev. 114 (2014) 3601–3658. doi: 10.1021/cr400388t

  24. [24]

      I.A. Topol, A.V. Nemukhin, K. Salnikow, et al., J. Phys. Chem. A 110 (2006) 4223–4228. doi: 10.1021/jp055633k

  25. [25]

      S.P. de Visser, Chem. Commun. (2007) 171–173.

  26. [26]

      T. Borowski, A. Bassan, P.E.M. Siegbahn, Chem. Eur. J. 10 (2004) 1031–1041. doi: 10.1002/chem.200305306

  27. [27]

      T. Borowski, A. Bassan, P.E. Siegbahn, Inorg. Chem. 43 (2004) 3277–3291. doi: 10.1021/ic035395c

  28. [28]

      A.R. Diebold, C.D. Brown-Marshall, M.L. Neidig, et al., J. Am. Chem. Soc. 133 (2011) 18148–18160. doi: 10.1021/ja202549q

  29. [29]

      H.Y. Lin, X. Chen, J. Dong, et al., J. Am. Chem. Soc. 143 (2021) 15674–15687. doi: 10.1021/jacs.1c06227

  30. [30]

      T.T. Liu, R.H. Duan, Y.Y. Wang, et al., Chin. Chem. Lett. 33 (2022) 4281–4286. doi: 10.1016/j.cclet.2022.01.041

  31. [31]

      S. Ye, C. Riplinger, A. Hansen, et al., Chemistry 18 (2012) 6555–6567. doi: 10.1002/chem.201102829

  32. [32]

      J.N. Harvey, M. Aschi, H. Schwarz, et al., Theor. Chem. Acc. 99 (1998) 95–99. doi: 10.1007/s002140050309

  33. [33]

      H.Y. Lin, X. Chen, J.N. Chen, et al., Research 2019 (2019) 2602414.

  34. [34]

      C. Raspail, M. Graindorge, Y. Moreau, et al., J. Biol. Chem. 286 (2011) 26061–26070. doi: 10.1074/jbc.M111.227595

  35. [35]

      J.F. Lin, Y.L. Sheih, T.C. Chang, et al., PLoS ONE 8 (2013) e69733. doi: 10.1371/journal.pone.0069733

  36. [36]

      M. Gunsior, J. Ravel, G.L. Challis, et al., Biochemistry 43 (2004) 663–674. doi: 10.1021/bi035762w

• 

  Scheme 1  Spin states for each reactant and product for the un-catalyzed conversion reaction of HPP to HG.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  Relative energies and Gibbs free energies of stationary points and MECPs located in HPPD-catalyzed dioxygen activation reaction obtained at the QM/MM B3LYP-D3/6-311++G(d, p): AMBER level. The reaction coordinate was set as R_C2′-C3′ - R_Oa-C2′ (the difference of the interatomic distance between atom pairs C2′-C3′ and Oa-C2′) for the steps covering I1 to I3 and was set as R_Oa-Ob for the step covering I3 to I4. Profile color codes: yellow for triplet, red for quintet, green for septet and blue for MECP.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Key geometrical parameters and spins for atoms in the catalytic site of ^5-7M2.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  The transmission of electrons for the decarboxylation reaction run in different spin state.

  下载: 全尺寸图片 幻灯片
•