English

• As two important groups of polyphenolic compounds, flavanols and stilbenes are largely present in plants as monomers, oligomers and polymers [1-5]. They react as renewable and sustainable sources in the synthesis of polyphenol polymers with fused or bridged ring systems. Flavanostilbenes constitute a class of natural polyphenols with diverse scaffolds, and are formed by adduction of flavanols and polyhydroxystilbene units [6-8]. Recently, some flavanostilbenes have been isolated and identified, such as cajanusflavanols A–C [9], rhamnoneuronal D–N [10] and polygolflavanol A [11]. These flavanostilbenes exhibited potent biological activities, such as antioxidant, antidiabetic, antitumor and anti-inflammatory activities, and are therefore highly attractive targets for synthesis [10-13]. Jezonocinol C (1) is a flavanostilbene isolated from the bark of Picea jezoensis var. jezoensis by Tanaka and coworkers in 2007, and it features a unique 2-cycloheptenone subunit. It was shown to have radical-scavenging and antitumor initiating activities [14,15]. Since its isolation, further systemic biological activities in vitro and in vivo have remained ambiguous because of the scarcity of natural samples. The unprecedented 7/6/5/6 fused ring system inspired us to investigate the synthesis of this family of compounds.

  Structurally, the 2-cyclohepten-1-one core and contiguous hexavalent carbon stereocenters are synthetic challenges [16,17]. Hence, a biomimetic strategy through a stereoselective route is highly desirable to generate 1 and its analogs. Tanaka and coworkers speculated that 1 is biosynthesized via the radical coupling of catechin and piceatannol, resulting in formation of the C6-C8′′ and C10-C7′′ bonds (Fig. 1a). Next, bridged ring opened, followed by decarboxylation to obtain 4 with a 2-cycloheptenone core and a second radical coupling between C4-C2′′. However, given the common dimerization in polyphenol units, the two-step radical coupling is not a compelling pathway. Based on our earlier work on the synthesis of flavanostilbenes [18], the C4-C2′ bond was connected first. The cycloheptenone core of 1 reminded us of the potential that the cyclization possessed a C4-C2′ linkage and a [5 + 2] cycloaddition.

  Figure 1

  

  Figure 1.  (a) Tanaka's proposed biogenesis of 1. (b) Our proposed biogenesis of 1.

  DownLoad: Full-Size Img PowerPoint

  [5 + 2] Cycloadditions have been applied in the total synthesis of natural products with complex bridged ring systems [19-22]. Natural bridged-ring products have been successfully synthesized via this type of cycloaddition to date (e.g., α-pipitzol, perezoperezone, and epicolactone) [23-27]. Hence, our proposed biosynthesis commences with nucleophilic substitution between C-4 of catechin and C-2′′ of piceatannol to form intermediate 5 (Fig. 1b). Ring A in 5 undergoes intramolecular [5 + 2] cycloaddition with a piceatannol unit to provide 6, followed by decarboxylation to afford 1. Herein, flavanostilbenes with a cycloheptenone system were synthesized in only three steps without complex protection/deprotection of the phenolic hydroxy groups, and the reaction mechanism of [5 + 2] cycloaddition was clarified (Scheme 1).

  Scheme 1

  

  Scheme 1.  Biomimmetic synthesis of flavanostilbenes 10a with a 2-cyclohepten-1-one skeleton.

  DownLoad: Full-Size Img PowerPoint

  To promote atom economy, our synthetic strategy initially focused on the non-use of protecting groups. Considering that the biomimetic substrate of 1 is limited by the need for sufficient purified catechin units, we selected (-)-epicatechin-3-gallate (ECG) as a facile unit from natural sources [28,29]. We then investigated nucleophilic substitution to generate key intermediate 9a. An abundance of ECG polymers (ECGp) was extracted from Rhodiola crenulate [30,31], and can be used as ECG substrates via depolymerization [32,33]. Benzyl mercaptan was employed to react with ECGp and HBr in MeOH [34], affording 4-benzylthio-substituted ECG monomers 7. In our previous work, silver(Ⅰ)-mediated stereoselective nucleophilic substitution of 7 at C-4 has been applied [18]. Delightfully, by direct treatment of 7 with piceatannol in the same condition of silver trifluoroacetate (AgTFA), the key intermediate 9a was obtained in 40% yield, and the absolute configuration of C-4 was assigned as 4R, similar to our report [18].

  With the successful synthesis of 9a, its cyclization was attempted. In view of the application of Fe(Ⅲ) catalysts in dimerization and their commercial availability, FeCl₃·6H₂O was tested first [35,36]. To our delight, despite the low yield, cycloheptenone product 10a was separated as the main product in MeOH-H₂O (1:1, v/v) in 22 h (Table 1, entry 2). We then evaluated the ability of other transition metals to increase the yield. From this screening, it was found that CuCl₂·2H₂O (0.5 equiv.) as the additive in MeOH-H₂O (1:1, v/v) afforded 10a in 23% yield in 2 h (entry 3). The higher yield and shorter reaction time indicated that Cu(Ⅱ) as the additive outperformed Fe(Ⅲ) in this cyclization [37]. Subsequent tests focused on different types of solvents. Fortunately, water as a solvent contributed to the formation of 10a in 56% yield (entry 4), and decomposition reaction occurred in nonaqueous solvents such as acetonitrile to afford monomers like piceatannol (entries 1, 6 and 7). Under the above optimized conditions, Fe(Ⅲ) could increase the yield only up to 43% and required a longer time (entry 13). Some other transition-metal catalysts (e.g., Mn(OAc)₂, FeCl₂, and CuBr) were also attempted, but no better promotion of the reaction was discovered (entries 8–12). Additionally, no significant improvement was observed by changing other reaction parameters, such as light, temperature, and concentration (Table S1 in Supporting information). With the optimal reaction conditions in hand, we achieved the biomimetic synthesis of 10a on gram scale (Scheme 1).

  Table 1

  

  Table 1.  Optimization of the reaction conditions.^a

  DownLoad: CSV

  In Tanaka′s work, the absolute configuration of 1 was not determined. In light of this, the remaining task was to verify the stereochemistry of the contiguous hexavalent carbons of 10a. A rotating frame overhauser enhancement spectroscopy (ROESY) experiment was performed to determine the relative configuration of 10a. The ROESY correlations of H-3/H-2, H-8′′ and H-10/H-4, H-7′′ revealed that H-2, H-3, and H-8′′ were positioned on the same face. Given the correlations from H-10 to H-4 and H-7′′ and the rigidity of the 5/6/7 fused-ring system, it was concluded that H-4, H-10, and H-7′′ must all be in the same orientation. Therefore, together with the fact that the absolute configurations of C-2 and C-3 of (-)-epicatechins from natural products were 2R and 3R, the absolute configuration (2R, 3R, 4R, 10R, 7′′S, 8′′R) was unambiguously verified. The absolute configuration of 10a was also confirmed by measurement of the electronic circular dichroism (ECD) spectrum and comparison with calculated ECD data by time-dependent density functional theory (TDDFT) performed at the CAM-B3LYP/6-31+g(d,p) level of theory (Fig. S1 in Supporting information).

  Furthermore, the mechanism of cyclization of 9a was investigated, and some control experiments were initially carried out. Interestingly, radical inhibitors such as 5,5-dimethyl-1-pyrroline N-oxide (DMPO) were first tested, but no inhibition was observed, precluding the possibility of a free radical mechanism (Scheme 2b and Table S3 in Supporting information) [38]. Notably, direct cycloaddition product 12 with a bicyclo[3.2.1]oct-3-ene-2,8-dione skeleton provided more convincing evidence. High-resolution mass spectrometry (HRMS) was successfully employed to characterize 12, showing [M + H]⁺ ions at m/z 685.1554. However, water-mediated fast decarboxylation at room temperature increased the difficulty of purification. To avoid decarboxylation, a hypervalent iodine reagent was utilized to replace the Fe(Ⅲ) or Cu(Ⅱ) additive in a nonaqueous solvent [39]. Treating 9a under the conditions of PhI(OAc)₂/MeOH produced 12 in 43% yield (Scheme 2a), and its structure was then determined by 1D and 2D NMR.

  Scheme 2

  

  Scheme 2.  Synthesis of intermediate 12 and mechanistic exploration of the cycloaddition.

  DownLoad: Full-Size Img PowerPoint

  Subsequently, under the optimized conditions, we could not obtain 10a in an oxygen-free atmosphere (Table 2, entry 3), and reaction in oxygen atmosphere afforded 10a in 9% yield without the use of additive, indicating that oxidative dearomatization by oxygen was essential to cyclization and adding Cu(Ⅱ) or Fe(Ⅲ) accelerated the cycloaddition (Table 2, entry 2) [40,41]. Additionally, the use of Tris buffer (pH 7.5) and acetate buffer (pH 7.5) as weakly basic solvents led to the oxidation and decomposition of 9a, eliminating the possibility of carbanion formation (Table S1 in Supporting information) [42]. When (-)-epicatechin gallate and piceatannol monomers were mixed, no [5 + 2] cycloaddition product was detected under the same conditions (Scheme 2c), indicating that the biosynthetic pathway of 10a started with a connection between C4 and C2′ followed by cycloaddition. On the basis of the results of our control experiments, we proposed a mechanism of cyclization (Scheme 2d). The presence of oxidants leads to the dearomatization of ring A in the ECG units. Subsequently, dearomatized ring A undergoes intramolecular [5 + 2] cycloaddition with the double bond of the piceatannol unit. Finally, the unstable bridged-ring system of 12 undergoes ring-opening at C-10 by hydrolysis to relieve the steric hindrance, and decarboxylation of β-ketonic acid through a six-membered ring transition state with an intramolecular hydrogen bond gives 10a. Water plays a key role in promoting the reversible [5 + 2] cycloaddition toward the final product and increasing the yield of product [43,44].

  Table 2

  

  Table 2.  Investigations of oxygen.^a

  DownLoad: CSV

  Confusingly, resveratrol as a substitution alternative for piceatannol afforded 11b as the major product instead of 10b. Structure of 11b has been reported in our previous work and features a hexahydrocyclopenta[c]furan skeleton [18]. The yield of 10b could be increased only to 16% by heating under Fe(Ⅲ)-mediated conditions (Table 3). This discovery prompted us to investigate the substrate scope and explain the difference of chemoselectivity between 9b and 9a in intramolecular cyclization. Notably, Fe(Ⅲ) contributed to the formation of both 10b and 11b, and thus it was more appropriate to reflect the competitiveness of the cyclization. Analogs of resveratrol substituted by electron-donating or electron-withdrawing groups were synthesized via Arbuzov and Wittig-Horner reactions followed by demethylation (see Supporting information for details) [45,46]. The connection of 7 with these stilbenes afforded 9d~9j, which cyclized to generate products with a hexahydrocyclopenta[c]furan or 2-cycloheptenone core (Table 3).

  Table 3

  

  Table 3.  Chemoselectivity and substrate scope.^a

  DownLoad: CSV

  After analyzing the association between the selectivity and structures of the substrates, it was presumed that stilbenes substituted by more electron-donating groups tended to provide cycloheptenone products 10, in which 9d and 9e equipped with five or six hydroxy‑substituted stilbenes even cyclized directly in the presence of water (Table 3). Less production of cycloheptenone products 10 was observed with a decrease in electron-donating groups. For instance, resveratrol-substituted intermediate 9b could only cyclize to afford 10b as a minor product in low yield, despite the change in the substitution position of the 4′-OH moiety. In addition, intermediate 9i, substituted with electron-withdrawing groups, afforded neither 10i nor 11i under the same reaction conditions. The reactivity difference of [5 + 2] cycloaddition might be explained by frontier molecular orbital (FMO) theory. Oxidation and dearomatization of ring A led to the electron-deficient diene (delocalized 4-π-electron). The HOMO energy of the dienophile was raised through the π-conjugated framework from electron-donating groups such as hydroxyl groups, and their frontier orbitals were therefore closer in energy and able to interact [47,48].

  Finally, compound 10c, a diastereoisomer of 1, was obtained via a concise strategy of tannase-catalyzed hydrolysis (Scheme 3a) [49-51]. Hydrolysis of intermediate 9a with tannase in water in an Ar atmosphere provided 9c in 88% yield. Cyclization of 9c under standard Cu(Ⅱ)-mediated conditions afforded 10c in 49% yield. Direct hydrolysis of 10a could not be carried out because of its poor water solubility. In addition, we found that (-)-epigallocatechin gallate polymers (EGCGp) isolated from Rhodiola kirilowii could also react as the source of flavans, and cycloheptenone products 10k with was synthesized (Scheme 3b).

  Scheme 3

  

  Scheme 3.  (a) Synthesis of a diastereoisomer of Jezonocinol C. (b) Synthesis of compound 10k.

  DownLoad: Full-Size Img PowerPoint

  In conclusion, flavanostilbenes with a 2-cyclohepten-1-one core have been synthesized on gram scale for the first time through oxidative dearomatization followed by a Cu-mediated [5 + 2] cycloaddition/decarboxylation cascade. The first intramolecular [5 + 2] cycloaddition of the phloroglucinol unit as the diene was reported. The new synthesis route features contiguous hexavalent carbon stereocenters constructed in three steps under mild conditions without sophisticated protection/deprotection of the phenolic hydroxy groups. The significant intermediate of [5 + 2] cycloaddition 12 was synthesized, and the mechanism of cyclization was proven. Water as a green solvent provides better chemoselectivity, shorter reaction time and higher yield. In addition, investigation of the substrates demonstrated that the substitution of stilbenes also affected the regioselectivity of the cyclization reaction.

  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.

  CRediT authorship contribution statement

  Gangsheng Li: Writing – review & editing, Writing – original draft, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Xiang Yuan: Validation, Supervision, Methodology, Formal analysis, Data curation. Fu Liu: Validation, Investigation, Formal analysis, Conceptualization. Zhihua Liu: Validation, Supervision. Xujie Wang: Validation, Methodology. Yuanyuan Liu: Validation, Formal analysis. Yanmin Chen: Methodology. Tingting Wang: Formal analysis. Yanan Yang: Supervision, Project administration, Funding acquisition, Conceptualization. Peicheng Zhang: Supervision, Project administration, Funding acquisition, Conceptualization.

  Acknowledgments

  We are grateful for the financial surpport from the CAMS Innovation Fund for Medical Sciences (CIFMS, No. 2021-I2M-1-028). This research was supported by Biomedical High Performance Computing Platform, Chinese Academy of Medical Sciences.

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.109880.

  
  
• 1. [1]

     S. Quideau, D. Deffieux, C. Douat-Casassus, L. Pouységu, Angew. Chem. Int. Ed. 50 (2011) 586–621. doi: 10.1002/anie.201000044

  2. [2]

     M.N. Lund, Trends Food Sci. Technol. 112 (2021) 241–251. doi: 10.1016/j.tifs.2021.03.056

  3. [3]

     F.F. de Araujo, D. de Paulo Farias, I.A. Neri-Numa, G.M. Pastore, Food Chem. 338 (2021) 127535. doi: 10.1016/j.foodchem.2020.127535

  4. [4]

     C. Papuc, G.V. Goran, C.N. Predescu, V. Nicorescu, G. Stefan, Compr. Rev. Food Sci. Food Saf. 16 (2017) 1243–1268. doi: 10.1111/1541-4337.12298

  5. [5]

     A. Durazzo, M. Lucarini, E.B. Souto, et al., Phytother. Res. 33 (2019) 2221–2243. doi: 10.1002/ptr.6419

  6. [6]

     L. Panzella, A. Napolitano, J. Agric. Food Chem. 70 (2022) 751–758. doi: 10.1021/acs.jafc.1c07229

  7. [7]

     T. Teka, L. Zhang, X. Ge, et al., Phytochemistry 197 (2022) 113128. doi: 10.1016/j.phytochem.2022.113128

  8. [8]

     S. Chinnabattigalla, R.K. Dakoju, S. Gedu, J. Heterocycl. Chem. 58 (2020) 415–441. doi: 10.1002/jhet.4176

  9. [9]

     Q. He, Z. Wu, X. Huang, et al., Org. Lett. 20(2018) 876–879. doi: 10.1021/acs.orglett.8b00010

  10. [10]

      H.M. Cho, M. Zhang, E.J. Park, et al., J. Nat. Prod. 85 (2022) 70–82. doi: 10.1021/acs.jnatprod.1c00689

  11. [11]

      L. Chen, X. Huang, M. Li, et al., Phytochem. Lett. 5 (2012) 756–760. doi: 10.1016/j.phytol.2012.08.007

  12. [12]

      M. Boozari, S. Nejad Ebrahimi, S. Soltani, et al., Bioorg. Chem. 85 (2019) 498–504. doi: 10.1016/j.bioorg.2019.01.051

  13. [13]

      J. Kwon, S. Basnet, J.W. Lee, et al., Bioorg. Med. Chem. Lett. 25 (2015) 3314–3318. doi: 10.1016/j.bmcl.2015.05.062

  14. [14]

      S.I. Wada, Y. Yasui, T. Hitomi, R. Tanaka, J. Nat. Prod. 70 (2007) 1605–1610. doi: 10.1021/np070104o

  15. [15]

      S.I. Wada, Y. Yasui, H. Tokuda, R. Tanaka, Bioorg. Med. Chem. 17 (2009) 6414–6421. doi: 10.1016/j.bmc.2009.07.016

  16. [16]

      P. Maity, S.D. Lepore, J. Am. Chem. Soc. 131 (2009) 4196–4197. doi: 10.1021/ja810136m

  17. [17]

      M. Wang, A. Wu, X. Pan, H. Yang, J. Org. Chem. 67 (2002) 5405–5407. doi: 10.1021/jo011077o

  18. [18]

      X. Wang, F. Liu, J. Yun, et al., Angew. Chem. Int. Ed. 57 (2018) 10127–10131. doi: 10.1002/anie.201804329

  19. [19]

      K. Gao, J. Hu, H. Ding, Acc. Chem. Res. 54 (2021) 875–889. doi: 10.1021/acs.accounts.0c00798

  20. [20]

      L. Min, X. Liu, C.C. Li, Acc. Chem. Res. 53 (2020) 703–718. doi: 10.1021/acs.accounts.9b00640

  21. [21]

      K. Gao, Y.G. Zhang, Z. Wang, H. Ding, Chem. Commun. 55 (2019) 1859–1878. doi: 10.1039/c8cc09077g

  22. [22]

      H. Pellissier, Adv. Synth. Catal. 360 (2018) 1551–1583. doi: 10.1002/adsc.201701379

  23. [23]

      J.C. Green, T.R. Pettus, J. Am. Chem. Soc. 133 (2011) 1603–1608. doi: 10.1021/ja109925g

  24. [24]

      G. Zhu, C. Zhou, S. Chen, S. Fu, B. Liu, Org. Lett. 21 (2019) 7809–7812. doi: 10.1021/acs.orglett.9b02791

  25. [25]

      P. Ellerbrock, N. Armanino, M.K. Ilg, R. Webster, D. Trauner, Nat. Chem. 7 (2015) 879–882. doi: 10.1038/nchem.2336

  26. [26]

      K. Yu, Z.N. Yang, C.H. Liu, et al., Angew. Chem. Int. Ed. 58 (2019) 8556–8560. doi: 10.1002/anie.201903349

  27. [27]

      S. Chen, T. Chen, G. Liu, et al., Org. Biomol. Chem. 17 (2019) 4711–4714. doi: 10.1039/c9ob00661c

  28. [28]

      H.M. Diaz-Mula, F.A. Tomas-Barberan, R. Garcia-Villalba, J. Agric. Food Chem. 67 (2019) 9160–9167. doi: 10.1021/acs.jafc.8b07155

  29. [29]

      P.V. Gadkari, M. Balaraman, Food Bioprod. Process. 93 (2015) 122–138. doi: 10.1016/j.fbp.2013.12.004

  30. [30]

      H.M. Chiang, H.C. Chen, C.S. Wu, P.Y. Wu, K.C. Wen, J. Food Drug Anal. 23 (2015) 359–369. doi: 10.1016/j.jfda.2015.04.007

  31. [31]

      A. Panossian, G. Wikman, J. Sarris, Phytomedicine 17 (2010) 481–493. doi: 10.1016/j.phymed.2010.02.002

  32. [32]

      L. Rouméas, G. Billerach, C. Aouf, É. Dubreucq, H. Fulcr, A.C.S. Sustain. Chem. Eng. 6 (2017) 1112–1120. doi: 10.1021/acssuschemeng.7b03409

  33. [33]

      L. Roumeas, C. Aouf, E. Dubreucq, H. Fulcr, Green Chem. 15 (2013) 3268–3275. doi: 10.1039/c3gc41281d

  34. [34]

      C. Fu, W. Chen, Y.L. Quek, et al., Tetrahedron Lett. 51 (2010) 6322–6324. doi: 10.1016/j.tetlet.2010.09.123

  35. [35]

      H. Shalit, A. Dyadyuk, D. Pappo, J. Org. Chem. 84 (2019) 1677–1686. doi: 10.1021/acs.joc.8b03084

  36. [36]

      I. Bauer, H.J. Knolker, Chem. Rev. 115 (2015) 3170–3387. doi: 10.1021/cr500425u

  37. [37]

      S.E. Allen, R.R. Walvoord, R. Padilla-Salinas, M.C. Kozlowski, Chem. Rev. 113 (2013) 6234–6458. doi: 10.1021/cr300527g

  38. [38]

      Y. Long, Y. Ding, H. Wu, et al., Angew. Chem. Int. Ed. 58 (2019) 17552–17557. doi: 10.1002/anie.201911978

  39. [39]

      C. Xu, A. Han, S.E. Reisman, Org. Lett. 20 (2018) 3793–3796. doi: 10.1021/acs.orglett.8b01387

  40. [40]

      J.H. George, Acc. Chem. Res. 54 (2021) 1843–1855. doi: 10.1021/acs.accounts.1c00019

  41. [41]

      C. Zhuo, C. Zheng, S. You, Acc. Chem. Res. 47 (2014) 2558–2573. doi: 10.1021/ar500167f

  42. [42]

      K. Hashida, S. Ohara, R. Makino, J. Wood Chem. Technol. 23 (2003) 227–232. doi: 10.1081/WCT-120026930

  43. [43]

      R.A. Ward, P. Bethel, C. Cook, et al., J. Med. Chem. 60 (2017) 3438–3450. doi: 10.1021/acs.jmedchem.7b00267

  44. [44]

      J. Lv, J. Li, D. Zhang-Negrerie, et al., Org. Biomol. Chem. 11 (2013) 1929–1932. doi: 10.1039/c3ob00039g

  45. [45]

      I.Y. El-Deeb, T. Funakoshi, Y. Shimomoto, R. Matsubara, M. Hayashi, J. Org. Chem. 82 (2017) 2630–2640. doi: 10.1021/acs.joc.6b03037

  46. [46]

      G. Chen, J. Wei, X. Yang, Z. Yao, Org. Lett. 18 (2016) 1502–1505. doi: 10.1021/acs.orglett.6b00010

  47. [47]

      L.R. Domingo, J.A. Saez, Org. Biomol. Chem. 7 (2009) 3576–3583. doi: 10.1039/b909611f

  48. [48]

      R. Robiette, J. Marchand-Brynaert, D. Peeters, J. Org. Chem. 67 (2002) 6823–6826. doi: 10.1021/jo025796u

  49. [49]

      A. Jana, S.K. Halder, A. Banerjee, et al., Bioresour. Technol. 157 (2014) 327–340. doi: 10.1016/j.biortech.2014.02.017

  50. [50]

      B. de Las Rivas, H. Rodriguez, J. Anguita, R. Munoz, Appl. Microbiol. Biotechnol. 103 (2019) 603–623. doi: 10.1007/s00253-018-9519-y

  51. [51]

      S. Hirose, Y.O. Kamatari, E. Yanase, Tetrahedron Lett. 61 (2020) 151601. doi: 10.1016/j.tetlet.2020.151601

• 

  Figure 1  (a) Tanaka's proposed biogenesis of 1. (b) Our proposed biogenesis of 1.

  下载: 全尺寸图片 幻灯片
  

  Scheme 1  Biomimmetic synthesis of flavanostilbenes 10a with a 2-cyclohepten-1-one skeleton.

  下载: 全尺寸图片 幻灯片
  

  Scheme 2  Synthesis of intermediate 12 and mechanistic exploration of the cycloaddition.

  下载: 全尺寸图片 幻灯片
  

  Scheme 3  (a) Synthesis of a diastereoisomer of Jezonocinol C. (b) Synthesis of compound 10k.

  下载: 全尺寸图片 幻灯片

  Table 1.  Optimization of the reaction conditions.^a

  下载: 导出CSV

  Table 2.  Investigations of oxygen.^a

  下载: 导出CSV

  Table 3.  Chemoselectivity and substrate scope.^a

  下载: 导出CSV
•