Nonheme Iron-Catalyzed Enantioselective cis -Dihydroxylation of Aliphatic Acrylates as Mimics of Rieske Dioxygenases
2022; Chinese Chemical Society; Volume: 4; Issue: 7 Linguagem: Inglês
10.31635/ccschem.022.202201780
ISSN2096-5745
AutoresJie Chen, Xiu Luo, Ying Sun, Si Si, Yuankai Xu, Yong‐Min Lee, Wonwoo Nam, Bin Wang,
Tópico(s)Synthetic Organic Chemistry Methods
ResumoOpen AccessCCS ChemistryRESEARCH ARTICLE14 Jul 2022Nonheme Iron-Catalyzed Enantioselective cis-Dihydroxylation of Aliphatic Acrylates as Mimics of Rieske Dioxygenases Jie Chen, Xiu Luo, Ying Sun, Si Si, Yuankai Xu, Yong-Min Lee, Wonwoo Nam and Bin Wang Jie Chen School of Chemistry and Chemical Engineering, University of Jinan, Jinan 250022 , Xiu Luo School of Chemistry and Chemical Engineering, University of Jinan, Jinan 250022 , Ying Sun School of Chemistry and Chemical Engineering, University of Jinan, Jinan 250022 , Si Si School of Chemistry and Chemical Engineering, University of Jinan, Jinan 250022 , Yuankai Xu School of Chemistry and Chemical Engineering, University of Jinan, Jinan 250022 , Yong-Min Lee Department of Chemistry and Nano Science, Ewha Womans University, Seoul 03760 , Wonwoo Nam *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] Department of Chemistry and Nano Science, Ewha Womans University, Seoul 03760 and Bin Wang *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] School of Chemistry and Chemical Engineering, University of Jinan, Jinan 250022 https://doi.org/10.31635/ccschem.022.202201780 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Enantioselective cis-dihydroxylation of alkenes represents an ideal route to synthesize enantioenriched syn-2,3-dihydroxy esters that are important structural motifs in numerous biologically and pharmaceutically relevant molecules. Bioinspired nonheme iron-catalyzed enantioselective cis-dihydroxylation meets the requirement of the modern synthetic chemistry from the atomic economy, green chemistry, and sustainable development perspectives. However, nonheme iron-catalyzed enantioselective cis-dihydroxylation is much underdeveloped because of the formidable challenges of controlling chemo- and enantioselectivities and product selectivity caused by the competitive epoxidation, cis-dihydroxylation, and overoxidation reactions. Herein, we disclose the fabrication of a biologically inspired nonheme iron complex-catalyzed enantioselective cis-dihydroxylation of multisubstituted acrylates using hydrogen peroxide (H2O2) as the terminal oxidant by controlling the non-ligating or weakly ligating counterions of iron(II) complexes, demonstrating a dramatic counteranion effect on the enantioselective cis-dihydroxylation of olefins by H2O2 catalyzed by nonheme iron complexes. A range of structurally disparate alkenes were transformed to the corresponding syn-2,3-dihydroxy esters in practically useful yields with exquisite chemo- and enantioselectivities (up to 99% ee). Given the mild and benign nature of this biologically inspired oxidation system as well as the ubiquity and synthetic utility of enantioenriched syn-2,3-dihydroxy esters as pharmaceuticals candidates and natural products, we expect that this strategy could serve as a promising complement to the well-known Sharpless asymmetric dihydroxylation, which is the chemical reaction of an alkene with OsO4 to produce a vicinal diol. Download figure Download PowerPoint Introduction Optically active syn-2,3-dihydroxy esters, especially aliphatic syn-2,3-dihydroxy esters, are of great significance due to their ubiquitous relevance in a diverse range of biologically active molecules, pharmaceuticals, and natural products.1,2 Moreover, enantioenriched syn-2,3-dihydroxy esters are frequently used as synthons to prepare a variety of chiral compounds since the presence of the ester groups renders the transformation of cis-diols regioselectively.3–7 With respect to their distinct structural features and significant roles in synthetic organic chemistry, diverse methodologies have been developed to access enantiopure syn-2,3-dihydroxy esters.8–19 To this end, enantioselective cis-dihydroxylation of α,β-unsaturated esters represents the most efficient and straightforward strategy for the synthesis of dihydroxy esters. These include metal-catalyzed (Mn,15 Ru,16 Os,1 and Pd17,18) and organocatalyzed19 enantioselective cis-dihydroxylation with various oxidants. Among these, Sharpless asymmetric dihydroxylation (SAD) stands out as the most reliable, practical, and general methodology to date for obtaining syn-2,3-dihydroxy esters with high stereoselectivity.1 However, SAD uses osmium as a catalyst which is toxic and expensive; SAD can also lead to the formation of by-products due to the overoxidation and cleavage of the diol products. In addition, when a stoichiometric amount of K3[Fe(CN)6] is used as an oxidant, ∼8.1 kg of iron salts per kg of the desired dihydroxylation product is produced as a by-product.9 Thus, it is evident that SAD does not meet the requirement of modern synthetic chemistry from the atomic economy, green chemistry, and sustainable development perspectives. One notable example utilized for biomimetic enantioselective cis-dihydroxylation is the naturally occurring Rieske dioxygenases that catalyze the cis-dihydroxylation of aromatic C=C double bonds to initiate biodegradation of arenes in soil (Scheme 1a).20,21 In this reaction, cis-diols are yielded with excellent stereo- and enantioselectivities; O2 is used as a terminal oxidant, and the two oxygen atoms of O2 are incorporated into the diol product at the same time (Scheme 1a).20,21 Naphthalene dioxygenase (NDO) contains a mononuclear nonheme iron(II) center that features a 2-histidine-1-carboxylate facial triad motif with two cis labile sites available for catalysis (Scheme 1a).20,21 The structural features of the active site and the key chemical principles that underlie their efficacy as catalysts for aerobic enantioselective cis-dihydroxylation of arene and olefin double bonds have inspired chemists to design synthetic models for Rieske dioxygenases.22,23 The first nonheme iron complex capable of performing enantioselective cis-dihydroxylation of olefins was reported by Que and co-workers,24 who extended the cis-dihydroxylation chemistry by replacing the previous tripodal tetradentate aminopyridine ligands with linear ones based on chiral trans-cyclohexane-1,2-diamine ( 1)24 and 2,2′-bipyrrolidine ( 2)25 backbone (Scheme 1b), providing cis-diols with as high as 97% enantiomeric excess (ee).25 However, the substrate must be present in a large excess amount (cat./H2O2/substrate = 1/10/500) to achieve a high diol/epoxide ratio and turnover number (TON),24,25 thereby limiting their practical applicability in synthetic chemistry. Only recently have elegant examples of enantioselective cis-dihydroxylation by nonheme iron complexes under substrate-limiting conditions been accomplished by Che and co-workers26,27; linear tetradentate aminoquinoline ligands ( 3 and 4) were demonstrated to be essential for providing cis-diols in synthetically useful yields with up to 99% ee (Scheme 1c).26,27 In spite of the significant recent advances, the discovery of a new catalytic system for enantioselective cis-dihydroxylation focuses on the ligand tuning through the laborious synthesis or modification of the supporting ligand (Schemes 1b and 1c).24–27 Moreover, compared with the numerous examples of nonheme iron-catalyzed enantioselective epoxidation of olefins,23,28 nonheme iron-catalyzed enantioselective cis-dihydroxylation reactions are much underdeveloped, with only two examples able to proceed in valuable preparative yields under substrate-limiting conditions due to the competitive epoxidation, cis-dihydroxylation, and overoxidation reactions.26,27 Scheme 1 | Enzymatic and bioinspired iron-catalyzed enantioselective cis-dihydroxylation of olefins. Download figure Download PowerPoint As part of our continuous interest in developing biologically inspired oxidation catalysis,28–32 we envisioned that the tuning of non-ligating or weakly ligating counterions of nonheme iron(II) complexes would facilitate the formation of iron(V)-oxo-hydroxo as a putative cis-dihydroxylating agent or the fast binding of the terminal oxidant (e.g., H2O2) at the iron center in a controlled manner; thus, addressing the selectivity challenges from competitive epoxidation, cis-dihydroxylation, and overoxidation associated with this reaction and/or improving the catalytic activity and selectivity toward cis-dihydroxylation. Herein, we report the establishment of a nonheme iron complex-catalyzed enantioselective cis-dihydroxylation of multisubstituted acrylates as mimics of Rieske dioxygenases. Given the mild and benign nature of this biologically inspired oxidation system, the ubiquity and synthetic utility of enantioenriched aliphatic syn-2,3-dihydroxy esters as pharmaceuticals candidates and natural products, we deduced that this protocol developed for the enantioselective cis-dihydroxylation of acrylates by H2O2, catalyzed by bioinspired nonheme iron catalysts would be of significant interest to the communities of bioinorganic, synthetic organic, and medicinal chemistry in both academic and pharmaceutical industry settings. Experimental Methods General procedure for enantioselective cis-dihydroxylation of olefins Under aerobic conditions, 30% H2O2 (1.5 mmol, 3.0 equiv, diluted in 0.50 mL CH3OH) was added dropwise using a syringe pump to a solution of complex 11 (11 mg, 0.015 mmol) and olefin 15a– 48a (0.50 mmol) in CH3OH (1.5 mL) for 30 min. After stirring the reaction mixture for 1 h at room temperature, the resulting solution was concentrated, and the residue was purified by column chromatography on silica gel with a gradient eluent of petroleum ether/ethyl acetate to afford the desired cis-diol product 15b– 48b (see Supporting Information for detailed procedures). ee values were determined by high-performance liquid chromatography (HPLC) or gas chromatography (GC) equipped with a chiral column (Agilent CP-ChirasilDexCB, United States). The absolute configuration of the diols was determined by X-ray crystallographic analysis and comparison of the optical rotation with literature values. Caution: Perchlorate salts are potentially explosive and should be handled with care. Results and Discussion Synthesis and characterization of nonheme iron(II) complexes First, a series of linear tetradentate nitrogen-donor ligands with central chirality (BQCN, 2-Me2-BQCN, 2-(MeO)2-BQCN, and 2-Ph2-BQCN) or axial chirality (BQNAN and 2-Me2-BQNAN) were obtained from Buchwald–Hartwig cross-coupling between a chiral diamine and substituted 8-bromoquinoline derivatives, followed by reductive amination (Figure 1, see Figures S1–S2 and Tables S1–S2 for detailed procedures).26,27 The complexation of iron(II) triflate with the above tetradentate ligands in dichloromethane provided the corresponding neutral iron(II) triflate complexes formulated as [Fe(R,R-BQCN)(OTf)2] ( 5), [Fe(R,R-2-Me2-BQCN)(OTf)2] ( 6), [Fe(R,R-2-(MeO)2-BQCN)(OTf)2] ( 7), [Fe(R,R-2-Ph2-BQCN)(OTf)2] ( 8), [Fe(R-BQNAN)(OTf)2] ( 9), and [Fe(R-2-Me2-BQNAN)(OTf)2] ( 10) (Figure 1). In addition, the dicationic ferrous complexes with other non-ligating or weakly ligating anionic ligands such as ClO4– and SbF6–, were prepared (Figure 1). The reactions of (R,R)-2-Me2-BQCN, R-BQNAN, and R-2-Me2-BQNAN with iron(II) perchlorate hydrate afforded iron(II) complexes with formula [Fe(R,R-2-Me2-BQCN)(CH3CN)2](ClO4)2 ( 11), [Fe(R-BQNAN)(CH3CN)2](ClO4)2 ( 13), and [Fe(R-2-Me2-BQNAN)2](ClO4)2 ( 14), respectively. For [Fe(R,R-2-Me2-BQCN)(CH3CN)2](SbF6)2 ( 12), an iron(II) chloride complex, [Fe(R,R-2-Me2-BQCN)(Cl)2], was synthesized by reacting anhydrous iron(II) chloride with R,R-2-Me2-BQCN, followed by treating the iron(II) chloride complex with 2 equiv of silver hexafluoroantimonate (AgSbF6) in acetonitrile (see Supporting Information for details). Figure 1 | Stereogenic-at-iron complexes bearing tetradentate nitrogen-donor ligands used in this study. Download figure Download PowerPoint Single crystals suitable for X-ray crystallographic analysis were obtained by slow diffusion of anhydrous diethyl ether into an acetonitrile solution of complexes 6 and 11 at room temperature (Figure 2; see Supporting Information Figures S3 and S4). The results showed that 6 and 11 had the topology of cis-α coordination (Figures 2b and 2c), although the linear tetradentate nitrogen-donor ligands, in principle, could coordinate to an octahedral ferrous center in three different topologies such as cis-α, cis-β, and trans conformations (Figure 2a).33–36 The geometry was C2-symmetric, in which two quinoline groups coordinated in trans fashion to each other, and two N–Me groups were oriented anti to each other, leaving two labile cis-coordinating sites at the ferrous center, occupied by triflate anions for 6 (Figure 2b) and two molecules of acetonitrile for 11 (Figure 2c); those triflate anions and acetonitrile ligands could readily be replaced by exogenous ligands such as H2O2 and dioxygen (O2). This arrangement is reminiscent of the active site of the mononuclear nonheme ferrous center for the Rieske dioxygenases, as illustrated in Scheme 1a20,21; O2 or H2O2 could bind to the metal center and be activated to generate high-valent iron-oxygen species as an active oxidant for cis-dihydroxylation.20–23 Moreover, the absolute configuration of the iron center of the above octahedral stereogenic-at-metal complexes was designated as Δ (Figure 2a),33–36 which, in turn, was determined by the chirality of the (R,R)-1,2-diaminocyclohexane backbone. The Fe–Ndiamine bond lengths for 6 and 11 were similar (∼2.207 Å) to each other, but the average Fe–Nquinoline bond lengths were different, viz, 2.290 Å for 6 and 2.255 Å for 11 (for details, see Supporting Information Table S4). In addition, the replacement of the much weaker triflate ligand with acetonitrile resulted in an increase in the average Fe–Otriflate bond length from 2.105 Å for 6 to the average Fe–Nacetonitrile bond length of 2.186 Å for 11 ( Supporting Information Table S4). These distances were indicative of high-spin iron(II) centers.37,38 In contrast, the average Fe–Ndiamine bond length (2.236 Å) was a bit longer, and the average Fe–Nquinoline bond length (2.219 Å) was slightly shorter for the closely related structure 3 (Scheme 1c).26 The distortion from octahedral symmetry was expressed by the angle value (N1–Fe–N4) involving one axial ligand, the Fe center, and the second axial ligand that deviated from an ideal value of 180°. Notably, the angles of N1–Fe–N4 of 6 (173.4°) and 11 (174.5°) were much larger than that of 3 (164.3°), indicating a stronger geometrical distortion of the octahedral coordination sphere in 3 ( Supporting Information Table S4).26 Figure 2 | (a) Coordination modes for octahedral stereogenic-at-iron complexes featuring linear tetradentate-nitrogen ligand. X-ray crystal structures of (b) [Fe(2-Me2-BQCN)(OTf)2] (6) and (c) [Fe(2-Me2-BQCN)(CH3CN)2](ClO4)2 (11) as an Oak Ridge thermal ellipsoid plot (ORTEP) drawing with 50% probability ellipsoids. Perchlorate counterions are omitted for clarity (Fe, sea green; N, blue; O, red; F, medium orchid; S, yellow; C, black). Crystallographic and structural data for 6 and 11 are listed in Supporting Information Tables S3 and S4. Download figure Download PowerPoint Optimization of reaction conditions Using (E)-benzyl crotonate 15a as a model substrate, nonheme iron(II) triflate complexes were screened for enantioselective cis-dihydroxylation in methanol with H2O2 as an oxidant (Table 1, entries 1–6). First, we found that the desired dihydroxylation product 15b was not produced when 5 bearing BQCN was employed as a catalyst (Table 1, entry 1). Interestingly, when iron complexes ( 6– 8) bearing different sterically hindered substituents on the quinoline ring at the C-2 position of BQCN were tested in the enantioselective cis-dihydroxylation (Table 1, entries 2–4), we observed that the reaction catalyzed by 6 with methyl groups (2-Me2-BQCN) afforded the cis-dihydroxylation product 15b in 47% yield with 98% ee, but half of 15a was unconsumed (Table 1, entry 2).26 However, an attempt to further improve the reaction with steric bulkiness on the quinoline ring was ineffective (Table 1, entries 3 and 4). Moreover, complexes 9 and 10 bearing tetradentate nitrogen-donor ligands with an axial chirality exhibited low reactivities (∼25% yields) with moderate to high enantioselectivities (78% and 86% ee for 9 and 10, respectively; Table 1, entries 5 and 6). These results indicated that the catalytic activity of iron(II) complexes in the cis-dihydroxylation of olefins by H2O2 was very sensitive to the ligand structure of the tetradentate BQCN ligand; iron(II) triflate complex 6 containing 2-Me2-BQCN ligand afforded the best reactivity and enantioselectivity (Table 1, entry 2).24–27 This observation led us to explore the nonheme iron(II) complexes with different counterions while preserving the same supporting ligand 2-Me2-BQCN (Table 1, entries 7 and 8) to determine the effect of counterions on the catalytic activity of the iron catalysts. Interestingly, when triflate anions in 6 were replaced by perchlorate anions ( 11), the substrate, (E)-benzyl crotonate 15a, disappeared completely, and the desired cis-diol product 15b was obtained in high yield with excellent enantioselectivity (90% yield and 98% ee; Table 1, entry 7 vs entry 2). When hexafluoroantimonate was used as the counterion, cis-dihydroxylation product was obtained with 66% yield and 98% ee (Table 1, entry 8), revealing that the catalytic activity of nonheme iron complexes in the cis-dihydroxylation of olefins by H2O2 was improved significantly by the fine-tuning of the counterions (e.g., triflate vs perchlorate). Similarly, when triflates in 9 and 10 were replaced by perchlorates, the product yields were increased significantly, but the enantioselectivities were similar (Table 1, entry 9 vs entry 5; entry 10 vs entry 6). Thus, it is interesting to note that the product yields in the cis-dihydroxylation of olefins by nonheme iron complexes and H2O2 were dependent on the counterions of the iron catalysts. One possible explanation for this behavior is that the counterions and the solvent ligands competed with H2O2 for the cis-coordinating sites at the ferrous center, deaccelerating the H2O2 activation and the formation of high-valent iron-oxygen species responsible for the cis-dihydroxylation. Therefore, the catalytic activities of the nonheme iron(II) complexes transformed inversely with the change of the binding affinity of the counterions and the solvent ligands, such that the most strongly coordinating counterions led to the lowest activity, while the most readily replaced solvent ligands yielded the highest activity.39–42 Other reaction parameters such as the catalyst loading (Table 1, entry 11), the oxidant loading (Table 1, entry 12), and the solvent (Table 1, entries 13–18), were also examined, providing us with the following optimal reaction conditions: [Fe(2-Me2-BQCN)(CH3CN)2](ClO4)2 ( 11) (3.0 mol %) and H2O2 (3.0 equiv) in MeOH at room temperature for 1.5 h, for the enantioselective cis-dihydroxylation of olefins. Table 1 | Optimization of Reaction Conditions for cis-Dihydroxylation of Olefinsa Entry Catalyst Solvent Yield (%)b ee (%)c 1 5 MeOH N.D. — 2 6 MeOH 47 98 3 7 MeOH N.D. — 4 8 MeOH N.D — 5 9 MeOH 25 78d 6 10 MeOH 28 86d 7 11 MeOH 90 98 8 12 MeOH 66 98 9 13 MeOH 58 80d 10 14 MeOH 66 86d 11e 11 MeOH 68 98 12f 11 MeOH 59 98 13 11 EtOH 51 99 14 11 TFE 36 82 15 11 t-BuOH 12 93 16 11 i-PrOH 15 98 17 11 CH3CN 40 96 18 11 THF N.D. — aReaction conditions: 30% H2O2 (1.5 mmol, diluted in 0.50 mL of MeOH) was added dropwise via a syringe pump to a solution of 15a (0.50 mmol) and the catalyst (3.0 mol %) in MeOH (1.5 mL) for 0.5 h. The resulting solution was then stirred for 1 h at room temperature. N.D., not detected; TFE, 2,2,2-trifluoroethanol; THF, tetrahydrofuran. bIsolated yield. cDetermined by HPLC analysis on a chiral stationary phase. dOther enantiomer of 15b was obtained. e2.0 mol % of 11 was used. f2.0 equiv of H2O2 was added. Scope of alkenes Enantiomerically pure syn-2,3-dihydroxybutyrates are important building blocks for the construction of complex carbohydrates, nonproteinogenic amino acids, and natural products.2,43–47 Diazotization-hydrolysis of L-threonine43,44 and the SAD of crotonic acid esters43–45 have been widely employed for the synthesis of enantioenriched syn-2,3-dihydroxybutyrates. However, (E)-alkyl crotonates are among the most challenging substrates in the osmium-catalyzed cis-dihydroxylation yielding the corresponding cis-diols with 80‒92% ee ( Supporting Information Figure S5) due to low steric recognition. Therefore, we investigated the cis-dihydroxylation of a number of structurally disparate β-alkyl substituted acrylate esters with the optimized reaction conditions to evaluate the generality of the newly developed cis-dihydroxylation protocol (Figure 3a). As shown in Figure 3a, a variety of easily available (E)-alkyl crotonates ( 15a – 21a), including (E)-benzyl crotonate 15a, (E)-methyl crotonate 16a, (E)-ethyl crotonate 17a, (E)-tert-butyl crotonate 18a, (E)-n-hexyl crotonate 19a, (E)-4-methoxybenzyl crotonate 20a, and (E)-4-phenylbenzyl crotonate 21a, were found to be suitable substrates for this nonheme iron catalytic system, affording the corresponding cis-diols 15b– 21b in high yields (70–90%) with excellent enantioselectivities (95–99% ee). Linear alkenes with varied chain length ( 22a– 26a) such as ethyl trans-2-pentenoate 22a, ethyl trans-2-hexenoate 23a, ethyl trans-2-heptenoate 24a, ethyl trans-2-octenoate 25a, and ethyl trans-2-nonenoate 26a, readily underwent cis-dihydroxylation to produce optically pure dihydroxy esters ( 22b– 26b) with excellent enantioselectivities (97–99% ee). These dihydroxy esters have been demonstrated to be versatile building blocks for the total synthesis of natural products.1,2 Additionally, (E)-ethyl 3-cyclohexylacrylate 27a was cis-dihydroxylated to afford the desired cis-diol 27b in 90% yield with 99% ee, an important synthon for the preparation of the anti-inflammatory and human immunodeficiency virus (HIV) antagonist drug ONO-4128.48 To evaluate the chemoselectivity of this catalytic system, selective oxidation of substrates with alternative readily oxidizing benzylic C(sp3)‒H bonds, other than C=C bonds, were explored. The olefinic ester 28a housing two competing benzylic methylene (CH2) groups underwent cis-dihydroxylation with exquisite chemo- and enantioselectivity to afford the intended cis-diol 28b (70% yield, 95% ee) as an important intermediate for the total synthesis of the antifungal agent echinocandin C49; no formation of the benzylic C(sp3)–H oxidation product was observed. Methyl 6-phthalimido-2-hexenoate 29a was cis-dihydroxylated to 29b in moderate yield but with excellent enantioselectivity, which is a key intermediate for threo-3-hydroxylysine, an unnatural amino acid and putative intermediate in the biosynthesis of the most potent protein kinase A and C (PKA and PKC) inhibitor, balanol.50 Ethyl cinnamates 30a and 31a could also be employed as substrates to give the corresponding cis-diols with excellent ee values, which are essential building blocks for the construction of the anti-inflammatory agents vancomycin and synthetic tetrapeptide, aeruginosin 98B.51–53 It should be noted that benzylic C(sp3)‒H bonds in 30a and 31a were well tolerated and did not undergo C(sp3)‒H activation initiated by high-valent iron-oxygen species. The absolute stereochemistry of the cis-diol products 27b and 30b were confirmed by X-ray crystal analysis (see Supporting Information Figures S6–S7 and Tables S5–S6), revealing their absolute configurations (2S,3R). Figure 3 | Scope of the enantioselective cis-dihydroxylation of olefins. (a) β-Alkyl substituted acrylate esters. (b) α,β- and β,β-Dialkyl substituted acrylate esters. Reaction conditions: H2O2 (3.0 equiv) was added dropwise via a syringe pump to a methanol solution containing substrate (0.50 mmol) and 11 (3.0 mol %) within 30 min, then the resulting solution was stirred for an additional 1 h. Isolated yield and ee values were determined by HPLC or GC analysis on a chiral stationary phase. Download figure Download PowerPoint Trisubstituted alkenes including acyclic ( 32a– 43a), exocyclic ( 44a and 45a), and endocyclic ( 46a– 48a) trisubstituted alkenes were also tested for the generality of this method (Figure 3b). Tiglic acid is a monocarboxylic unsaturated organic acid found in croton oil and several other natural products. Enantioenriched cis-diols obtained from tiglic acid esters have been widely employed in the construction of biologically active natural products and unnatural α,β-dimethylamino acids.54,55 Therefore, we considered testing the tolerance of tiglic acid esters under our reaction conditions. To our delight, we found that tiglic acid esters 32a– 35a underwent cis-dihydroxylation readily to afford the cis-diols 32b– 35b with quaternary carbon stereogenic centers in high yields and excellent enantioselectivities (90–99% ee). X-ray diffraction analysis of 35b established an absolute configuration of the cis-dihydroxylation product (see Supporting Information Figure S8 and Table S7). Further, we investigated the tolerance of the reaction toward trisubstituted aliphatic alkenes by conducting the cis-dihydroxylation of a series of α,β-dialkyl acrylates with disparate carbon chain lengths, including ethyl (E)-2-methyl-2-pentenoate 36a, ethyl (E)-2-methyl-2-hexenoate 37a, ethyl (E)-2-methyl-2-heptenoate 38a, ethyl (E)-2-methyl-2-octenoate 39a, ethyl (E)-2-methyl-2-nonenoate 40a, and ethyl (E)-3-cyclohexyl-2-methyl-2-propenoate 41a. Enantiomerically enriched cis-dihydroxylation products 36b‒ 41b with tetrasubstituted carbon centers were obtained from the corresponding substrates in excellent yields and remarkable enantioselectivities (93‒99% ee), showing that our method was tolerant of trisubstituted alkenes with various carbon chain lengths. Dibenzyl mesaconate 42a was oxidized to provide the corresponding cis-diol 42b in excellent yield with a good ee value. Moreover, β,β-disubstituted acrylates, such as acyclic trisubstituted alkene 43a and exocyclic trisubstituted alkenes 44a and 45a were also compatible, producing the desired products 43b‒ 45b in good yields with uniform levels of enantioselective induction (87–90% ee). Also, the catalytic enantioselective cis-dihydroxylation of endocyclic trisubstituted alkenes were thoroughly studied since the cis-diols produced have been widely used in the total synthesis of complex natural products. For instance, ester-substituted indene 46a, cyclopentene 47a, and cyclohexene 48a performed well in the reaction with high chemoselectivities, resulting in the corresponding cis-diols in 68–79% yields with 97–98% ee; the absolute configuration of the product 48b was determined by the X-ray structure analysis (see Supporting Information Figure S9 and Table S8). On the basis of the above results, we conclude that iron-oxygen intermediate(s) that prefers C=C cis-dihydroxylation to allylic C−H bond activation was generated in this catalytic system exhibiting excellent chemoselectivity, different from the preferred allylic C–H bond activation of cycloalkenes by nonheme metal-oxo complexes.56,57 Notably, the enantioselective cis-dihydroxylation of acrylates, catalyzed by 11 delivered most of the corresponding cis-diols with comparable or even higher enantioselectivities than that of the well-established osmium-based SAD ( Supporting Information Figure S5); thus, demonstrating that our biologically inspired nonheme iron-based catalysis could serve as a complementary approach to the SAD method. Gram-scale experiment and synthetic application The cis-dihydroxylation reaction was carried out on a gram-scale to yield 1.31 g of 29b and 1.09 g of 30b without loss of enantioselectivity (Scheme 2, 98% and 99% ee, respectively), demonstrating the practical and industrial potential of the synthetic utility of this biomimetic oxidation system. Enantioenriched cis-diols could be transformed into the unnatural β-hydroxylated α-amino acids of biological importance. For instance, optically active cis-diol 29b prepared by our method could be transformed into β-hydroxylysine 29c (Scheme 2a), which is an unnatural amino acid used for the synthesis of (−)-balanol.50 Notably, optically active cis-diol 30b underwent diverse functional group transformations to provide β-hydroxytyrosine 30c (Scheme 2b), which has been established as one of the seven unnatural amino acid building blocks for the total synthesis of vanc
Referência(s)