Synlett 2023; 34(11): 1235-1240
DOI: 10.1055/a-2030-7082
letter

Preparation of New Chiral Building Blocks by a Mukaiyama–Michael Reaction of 2-(Phenylsulfonyl)cyclopent-2-en-1-one

Ryoji Sugiyama
,
Masahisa Nakada

This work was financially supported in part by JSPS KAKENHI Grants Numbers JP19H02725 and JP22H02087, the Nagase Science and Technology Foundation, and a Waseda University Grant for Special Research Projects.
 


Abstract

A highly enantio- and diastereoselective Mukaiyama–­Michael reaction of 2-(phenylsulfonyl)cyclopent-2-en-1-one by using an enol silane of tert-butyl thiopropionate is described. The product was formed in 87% yield with a dr of 27:1 and 91% ee under stoichiometric conditions, whereas the yield, dr, and ee were 89%, 49:1, and 88% ee, respectively, under catalytic conditions. A highly stereoselective epimerization of the product of the Mukaiyama–Michael reaction which proceeds in 77% yield with a dr of 22:1 is also described. Because both enantiomers of the ligand for this Mukaiyama–Michael reaction are available, a method for the synthesis of all four stereoisomers of the product as useful chiral building blocks has been established.


#

The structures of cotylenin A,[1] which is an anticancer drug candidate; calcitriol lactone,[2] which is a vitamin-D3 metabolite; damsin,[3] which exhibits significant cytotoxicity against various human tumor cell lines;[4] and ergolide,[5] which exhibits strong cytotoxicity against human tumor cell lines, including the vinblastine-resistant cell lines,[6] each contain two contiguous asymmetric carbons (Figure [1]). Furthermore, the structures of cotylenin A and damsin contain two similar asymmetric carbons, and those of calcitriol lactone and ergolide contain two contiguous asymmetric carbons with different relative configurations.

Thus, structures 1a and 1b (Figure [1]), which contain two contiguous asymmetric carbons, are important chiral scaffolds. Such chiral scaffolds can be constructed through a catalytic asymmetric Mukaiyama–Michael reaction. For example, compound 2 might be prepared by the Mukaiyama–Michael reaction of 3 and 4 (Scheme [1]).[7] However, although there are many reports on catalytic asymmetric Mukaiyama–Michael reactions,[8] there are almost none on successful reactions of the enol silanes of ethyl ketones or propionates with cyclopentenones containing an electron-withdrawing group at the α-position.[9]

Zoom Image
Figure 1 Structures of cotylenin A, calcitriol lactone, damsin, ergolide, 1a, and 1b
Zoom Image
Scheme 1 Preparation of chiral building block 2 containing the chiral scaffold 1a by a catalytic asymmetric Mukaiyama–Michael reaction

Asymmetric catalysis of the Mukaiyama–Michael reaction of the enol silane of an ethyl ketone or a propionate is challenging because it requires the simultaneous control of enantioselectivity and diastereoselectivity. Although we have previously reported catalytic asymmetric Mukaiyama–Michael reactions that afforded high yields and enantioselectivities[10] [11] when an enol silane of a propionate was used therein, neither the enantioselectivity nor the diastereoselectivity could be controlled.

Here, we report a highly enantio- and diastereoselective Mukaiyama–Michael reaction of 2-(phenylsulfonyl)cyclopent-2-en-1-one with enol silanes of a propionate, together with the successful epimerization of the product, which permitted the construction of four possible diastereomers each containing two contiguous asymmetric carbons.

The reaction of 2-(phenylsulfonyl)cyclopent-2-en-1-one (5)[12] with the enol silane 6 [13] (Table [1]) was first investigated. Initially, Cu(OTf)2 (20 mol%) and ligand L3 (25 mol%) were used for the reaction of 5 and 6 (1.5 equiv) in dichloromethane (DCM);[14] however, the reaction proceeded slowly, even at room temperature. Therefore, a stoichiometric amount of the copper complex was used. After 21 hours of reaction at –78 °C, product 7 was obtained in an 80% yield with a dr of 5.5:1, and the ee of the major product was 36% (Table [1], entry 1). In the reaction at –40 °C, 5 was consumed in 16 hours, and the yield, dr, and ee were 99%, 6.6:1, and 34%, respectively (entry 2). In the reaction at room temperature, the yield was 81% with a dr of 7.4:1, but the ee was lower (entry 3). The reaction did not proceed in toluene at a low temperature because of the poor solubility of 5 (entry 4).

Table 1 Optimization of the Solvent for the Reaction of 5 with 6

Entry

Solvent

Temp (°C)

Time (h)

Yielda (%)

drb

eeb (%)

1

DCM

–78

21

80

5.5:1

36

2

DCM

–40

16

99

6.6:1

34

3

DCM

  rt

15

81

7.4:1

29

4

toluene

  rt

46

trace

N/A

N/A

5

Ac

–40

13

94

8.8:1

57

6

B d

–40

40

80

9:1

76

7

THF

  rt

46

trace

N/A

N/A

8

DCE

–40

15

95

5.2:1

32

a Isolated yield of 7.

b Dr and ee were determined by 1H-NMR and HPLC of the enol phosphate of 7, respectively [see the Supporting Information (SI)].

c A: 1:1 toluene–DCM.

d B: 2:1 toluene–DCM.

However, 5 was soluble in a 1:1 mixture of toluene and DCM; therefore, the reaction was carried out in the mixed solvent, which afforded the product in 94% yield with a dr of 8.8:1 and 57% ee (Table [1], entry 5). The reaction in a 2:1 mixture of toluene and DCM also proceeded at –40 °C without the precipitation of 5, yielding the product in 80% yield with a dr ratio of 9:1 and 76% ee (entry 6). The reaction did not proceed in THF (entry 7), and the use of 1,2-dichloroethane did not improve the dr or ee over those obtained by using DCM (entry 8).

Next, the solvent was fixed at a 2:1 mixture of toluene and DCM, and the reactions of 5 and 6 were examined by using various bisoxazoline ligands L112 [15] (Figure [2]). The dr ratios obtained by using L1, L4, L9, and L11 were higher than that obtained with L3; however, the ee values were lower (Table [2], entries 1, 4, 9, and 11, respectively). The dr ratios and ee obtained by using L2, L5, L6, L10, and L12 (entries 2, 5, 6, 10, and 12) were lower than those obtained by using L3. When L7 and L8 were used, the reaction did not proceed, possibly owing to the bulkiness of the ligand (entries 7 and 8).

Zoom Image
Figure 2 Structures of ligands L112

Table 2 Optimization of the Ligand for the Reaction of 5 with 6

Entry

Ligand

Time (h)

Yielda (%)

drb

eeb (%)

 1

L1

41.5

84c

10.5:1

–21

 2

L2

17

75c

 4.7:1

 44

 3

L3

40

80

  9:1

76

 4

L4

85

68

 12:1

 72

 5

L5

43.5

90

 8.8:1

 73

 6

L6

35.5

79 c

 7.5:1

 75

 7

L7

40

NR 

  N/A

N/A

 8

L8

2 (12)d

NR

  N/A

N/A

 9

L9

40

75 c

 9.4:1

 5.5

10

L10

64

77

 6.4:1

 72

11

L11

65

23

 18:1

 56

12

L12

64

23

 8.4:1

  8

a Isolated yields.

b Dr and ee were determined by 1H-NMR and HPLC of the enol phosphate of 7, respectively [see SI].

c Calculated from the isolated amount, and the ratio was determined by 1H NMR because it could not be separated from the ligand.

d Reaction at –40 °C for 2 h and at rt for 12 h.

The results in Table [2] show that the ee exceeded 70% when L36 (Table [2], entries 3–6) or L10 (Table [2], entry 10) was used. Thus, ligands bearing bulkier substituents at their bisoxazoline linkages afforded better results. Among the ligands considered, L3 was the most suitable for this reaction.

The Mukaiyama–Michael reactions of 5 with various enol silanes of propionates were studied by using L3 in the mixed solvent (2:1 toluene–DCM) (Table [3]). The reaction with 8a was completed to afford 9a in 12 hours at –40 °C in 89% yield with a dr of 10.6:1 and 78% ee (Table [3], entry 1). When the reaction was carried out at –60 °C, an improvement in the dr and ee was observed (entry 2), whereas at –78°C, a decrease in the dr and ee occurred (entry 3).

Table 3 Enantioselective Mukaiyama–Michael Reactions of 5 with 8ag

Entry

Reactant

R

R′

Temp (°C)

Time (h)

Product

Yielda (%)

drb

eeb (%)

 1

8a

OMe

TMS

–40

12

9a

89

10.6:1

78

 2

8a

OMe

TMS

–60

 0.7

9a

82

18.5:1

80

 3

8a

OMe

TMS

–78

17.5

9a

94

 3.4:1

40

 4

8b

OMe

TBS

–60

12.5

9a

67

 34:1

68

 5

8c

O t Bu

TMS

–60

 1

9c c

80

 21:1

82

 6

8d

SEt

TMS

–60

 3

9d c

84

 63:1

85

 7

8e

SEt

TBS

–60

15

9d c

86

 31:1

88

8

8f

S t Bu

TMS

–60

19

9f

87

27:1

91

9 d

8f

S t Bu

TMS

–60

42

9f c

89

49:1

88

10

8g e

S t Bu

TBS

–60

18

9f

75

 52:1

88

a Isolated yields.

b Dr and ee were determined by 1H-NMR and HPLC of the enol phosphates or TBS enol ethers of products 9a–f, respectively [see SI].

c Calculated from the isolated amount and the NMR ratio because the product could not be separated from the ligand.

d Cu(OTf)2 (20 mol%), L3 (25 mol%), 3 Å MS, 2:1 toluene–DCM, –60 °C, 42 h.

e The E-isomer of 8g was used.

The reason for the effects of changes in the reaction temperature is not certain at this stage, but the reaction at –60°C was found to be suitable. In the reaction with the bulky enol silane 8b, the dr ratio improved to 34:1 (Table [3], entry 4).

Therefore, the reaction of the enol silane of the tert-butyl ester 8c was carried out and the product was obtained in 80% yield with a dr ratio of 21:1 and 82% ee (Table [3], entry 5). The reaction with the enol silane of the ethyl thioester 8d resulted in an 84% yield with a 63:1 dr and an improved ee (85% ee) (entry 6). The reaction with bulkier 8e resulted in an 86% yield, a 31:1 dr, and an improved ee of 88% (entry 7). The reaction of the enol silane of the tert-butyl thioester 8f gave the best results (87% yield, 27:1 dr, and 91% ee) (entry 8).[16]

When this reaction was carried out with a catalytic amount of Cu(OTf)2 (20 mol%) and L3 (25 mol%) (Table [3], entry 9), the reaction time was 42 hours and the product was obtained in 89% yield with a dr of 49:1 and 88% ee,[17] indicating that an asymmetric catalysis of this reaction had been successfully developed. No improvement in ee was observed for the reaction with bulkier reactant 8g (Table [3], entry 10).

The product 9f obtained by using the conditions shown in Table [3], entry 8 was converted into a crystalline derivative 9f′(Scheme [2]), which was subjected to single-crystal X-ray structural analysis (Figure [3]);[18] this confirmed that the absolute configuration of 9f is that shown in Table [3].

Zoom Image
Scheme 2 Transformation of 9f into 9f′
Zoom Image
Figure 3 Structure of 9f′ and its X-ray crystal structure

The enantioselectivity of the Mukaiyama–Michael reaction of 5 can be well explained using the model (Figure [4]) previously proposed by us for a similar reaction.[11] As a result of the coordination of 5 to a complex composed of the bisoxazoline ligand L3 and Cu(OTf)2, the resulting planar tetracoordinated complex exhibits a bent shape owing to steric repulsion between the ligand and substrate. Therefore, the enol silane 8f can favorably approach the less-hindered convex face of the bent complex and react with 5 at the Re-face, which explains the configuration of the asymmetric carbon generated in the cyclopentanone of 9f. The better results with bisoxazoline ligands bearing bulky substituents at their bisoxazoline linkages (Table [2]) might be explained by their steric effect, which hinders the reaction from the concave side in the bent model shown in Figure [4].

Zoom Image
Figure 4 Proposed models for the observed high enantioselectivity in the Mukaiyama–Michael reaction of 5

In contrast, enol silane 8f approaches 5, which coordinates with a complex composed of the bisoxazoline ligand L3 and Cu(OTf)2 while avoiding steric repulsion with the bulky phenylsulfonyl group (Figure [5]). Therefore, reactions at the Si-faces of 5 and 8f are favorable, as shown in Figure [5], resulting in the diastereoselective formation of 9f.

Zoom Image
Figure 5 Proposed models for the observed high diastereoselectivity in the Mukaiyama–Michael reaction of 5. The complex composed of the bisoxazoline ligand L3 and Cu(OTf)2 is omitted for clarity

The high diastereoselectivity observed in the Mukaiyama-aldol–Michael reaction indicates that the reaction did not yield other diastereomers of 9f. Therefore, to obtain the diastereomers of 9f, we investigated its epimerization.

The epimerization of 9f was first carried out in THF with an excess of LDA to ensure the formation of dienolate of 9f. A subsequent reaction with bulky t-BuOH afforded 10 and 9f in a ratio of 1.2:1 (Table [4], entry 1). Changing the proton source to pivalic acid improved the ratio to 22:1 and the yield to 77% (entry 2). The ratio decreased to 3.1:1 when the solvent was changed to 10:1 THF–N,N′-dimethylpropyleneurea (DMPU) (entry 3), indicating that THF is the optimal solvent. When 2,6-di-tert-butylphenol was used as the proton source, the ratio decreased to 8:1 (entry 4), establishing pivalic acid as the optimal proton source.

Table 4 Epimerization of 9f

Entry

Solvent

ROH

Yielda (%)

10/9f b

1

THF

t-BuOH

64

1.2:1

2

THF

PivOH

77

22:1

3

10:1 THF–DMPU

PivOH

30

3.1:1

4

THF

2,6-t-Bu2C6H3OH

53

8:1

a Isolated yield.

b Ratio determined by 1H NMR of the products (see SI).

The stereoselectivity of this epimerization can be explained by using the chelate model shown in Figure [6]. The dienolate forms chelates, and the Si-face of the thioester enolate is shielded by a phenyl group. Consequently, protonation proceeds from the less-hindered Re-face. Because the ketosulfone enolate is more stable, the thioester enolate is protonated first. The decrease in diastereoselectivity upon the addition of the polar solvent DMPU might be explained by the fact that it prevents chelate formation. Similarly, 2,6-di-tert-butylphenol was less effective as a proton source (Table [4], entry 4) because its steric bulkiness might have caused dissociation of the chelate before the protonation.

Zoom Image
Figure 6 Proposed model for the highly stereoselective protonation

As shown in Scheme [3], ent -10, which is the enantiomer of 10 and includes structure 1a, was successfully used in our enantioselective formal total synthesis of damsin[19] because ent-10 could be synthesized by using the enantiomer of ligand L3. In that enantioselective formal total synthesis, the β-ketosulfone 11, which was derived from ent-10, was effectively used for the stereoselective construction of the quaternary asymmetric carbon. That is, 11 was readily methylated to 12, and subsequent reductive removal of its phenylsulfonyl and reaction of the resultant enolate with methyl cyanoformate in a one-flask manner afforded 13 as a single isomer. The enantioselective formal total synthesis, achieved by further transformation from 13 into De Clercq and Vandewalle’s synthetic intermediate,[20] demonstrated the usefulness of ent-10 bearing a phenylsulfonyl group.

Zoom Image
Scheme 3 Outline of our enantioselective formal total synthesis of damsin

In summary, we found that the Mukaiyama–Michael reaction of 2-(phenylsulfonyl)cyclopent-2-en-1-one with the enol silane of tert-butyl thiopropionate proceeded in a highly enantio- and diastereoselective manner. The yield, dr, and ee of the product were 87%, 27:1, and 91%, respectively, under stoichiometric conditions, and 89%, 49:1, and 88%, respectively, under catalytic conditions. We also found that a highly stereoselective epimerization of the product of the Mukaiyama–Michael reaction, proceeded in 77% yield with dr of 22:1. Because both enantiomers of the ligand required for this catalytic asymmetric Mukaiyama–Michael reaction are available, we established a method for the synthesis of all four stereoisomers of the product as useful chiral building blocks. We are currently investigating the efficiency of the developed method for the enantioselective total synthesis of natural products. The results will be reported in due course.


#

Conflict of Interest

The authors declare no conflict of interest.

Acknowledgment

We acknowledge the support of the Materials Characterization Central Laboratory, Waseda University, in the characterization of new compounds.

Supporting Information

  • References and Notes

  • 2 Ohnuma N, Bannai K, Yamaguchi H, Hashimoto Y, Norman AW. Arch. Biochem. Biophys. 1980; 204: 387
  • 4 Bujnicki T, Wilczek C, Schomburg C, Feldmann F, Schlenke P, Müller-Tidow C, Schmidt TJ, Klempnauer K.-H. Leukemia 2012; 26: 615
  • 6 Wang Q, Zhou B.-N, Zhang R.-W, Lin Y.-Y, Lin L.-Z, Gil RR, Cordell GA. Planta Med. 1996; 62: 166
  • 7 For the first Michael reaction of 2-(tolylsulfonyl)cyclopent-2-en-1-one, see: Posner GH, Switzer C. J. Am. Chem. Soc. 1986; 108: 1239
  • 10 Oyama H, Orimoto K, Niwa T, Nakada M. Tetrahedron: Asymmetry 2015; 26: 262
  • 11 Nagatani K, Minami A, Tezuka H, Hoshino Y, Nakada M. Org. Lett. 2017; 19: 810
  • 12 Yechezkel T, Ghera E, Ostercamp D, Hassner A. J. Org. Chem. 1995; 60: 5135
  • 13 Trimitsis G, Beers S, Ridella J, Carlon M, Cullin D, High J, Brutts D. J. Chem. Soc., Chem. Commun. 1984; 1088
  • 14 The reaction of 5 and 6 (1.5 equiv) was performed under the optimized conditions shown in Table 3 by using Cu(OTf)2 (20 mol%) and L3 (25 mol%) in the presence of 3 Å MS (300 wt%) in 2:1 toluene–DCM (2:1) at –40 °C. However, 7 was obtained in only 17% yield (dr 7.4:1, 74% ee) after 40 h.
  • 15 Sawada T, Nakada M. Tetrahedron: Asymmetry 2012; 23: 350
  • 16 The reaction of 2-(2,4,6-trimethylphenylsulfonyl)cyclopent-2-en-1-one and 8f (1.5 equiv) was examined by using Cu(OTf)2 (100 mol%), ligand L3 (110 mol%), and 3 Å MS in 2:1 toluene–DCM at –60 °C. The product was obtained in 95% yield with a dr of 42:1; however, the enantioselectivity was decreased (53% ee).
  • 17 Compound 9f A mixture of Cu(OTf)2 (4.5 mg, 0.0124 mmol), L3 (6.5 mg, 0.0155 mmol), and 3 Å MS (38.7 mg) in 2:1 toluene–DCM (0.290 mL) was stirred at rt for 1.5 h. A solution of 5 (13.2 mg, 0.0594 mmol) in 2:1 toluene–DCM (0.600 mL) was added and the mixture was stirred at rt for 1.5 h. A solution of 8f (17.7 mg, 0.0810 mmol) in 2:1 toluene–DCM (0.300 mL) was added at –60 °C, and the mixture was stirred at –60 °C for 41.5 h. The reaction was quenched with sat. aq NaHCO3 (2.0 mL) and the mixture was filtered. The aqueous layer was extracted with EtOAc (3 × 5 mL), and the combined organic layer was washed with H2O (5 mL) and brine (5 mL) then dried (Na2SO4) and concentrated under reduced pressure. The residue was purified by flash column chromatography [silica gel, hexane–EtOAc (6:1)] to afford 9f containing inseparable L3 as a white solid; yield: 26.0 mg [0.0531 mmol, 89% (calcd based on 1H NMR)]; Rf = 0.70 (hexane–EtOAc, 2:3). [α]D 21 –55 (c 0.65, DCM). IR (ATR): 1747, 1666, 1310, 1151, 1084, 1063, 970, 755, 729, 686, 621, 586, 555, 533, 494 cm–1. 1H NMR (500 MHz, CDCl3): δ = 7.86 (d, J = 7.5 Hz, 2 H), 7.70 (t, J = 7.5 Hz, 1 H), 7.58 (dd, J = 7.5, 7.5 Hz, 2 H), 3.83 (d, J = 5.0 Hz, 1 H), 3.25 (ddd, J = 8.0, 5.5, 5.0 Hz, 1 H), 2.90 (ddd, J = 7.5, 7.0 , 5.0 Hz, 1 H), 2.48–2.38 (m, 1 H), 2.36–2.25 (m, 1 H), 1.88–1.79 (m, 1 H), 1.43 (s, 9 H), 1.18 (d, J = 7.5 Hz, 3 H). 13C NMR (125 MHz, CDCl3): δ = 206.4, 202.8, 137.8, 134.3, 129.2, 71.8, 50.5, 48.6, 40.5, 37.7, 29.7, 23.6, 14.6. HRMS (ESI): m/z [M + Na]+ calcd for C18H24NaO4S2: 391.1014; found: 391.1007.
  • 18 CCDC 2235289 contains the supplementary crystallographic data for compound 9f′. The data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/structures
  • 19 Sugiyama R, Nakada M. Synlett 2023; in press
  • 20 De Clercq P, Vandewalle M. J. Org. Chem. 1977; 42: 3447

Corresponding Author

Masahisa Nakada
Department of Chemistry and Biochemistry, School of Advanced Science and Engineering, Waseda University
3-4-1 Ohkubo, Shinjuku-ku, Tokyo 169-8555
Japan   

Publication History

Received: 13 December 2022

Accepted after revision: 08 February 2023

Accepted Manuscript online:
08 February 2023

Article published online:
02 March 2023

© 2023. Thieme. All rights reserved

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany

  • References and Notes

  • 2 Ohnuma N, Bannai K, Yamaguchi H, Hashimoto Y, Norman AW. Arch. Biochem. Biophys. 1980; 204: 387
  • 4 Bujnicki T, Wilczek C, Schomburg C, Feldmann F, Schlenke P, Müller-Tidow C, Schmidt TJ, Klempnauer K.-H. Leukemia 2012; 26: 615
  • 6 Wang Q, Zhou B.-N, Zhang R.-W, Lin Y.-Y, Lin L.-Z, Gil RR, Cordell GA. Planta Med. 1996; 62: 166
  • 7 For the first Michael reaction of 2-(tolylsulfonyl)cyclopent-2-en-1-one, see: Posner GH, Switzer C. J. Am. Chem. Soc. 1986; 108: 1239
  • 10 Oyama H, Orimoto K, Niwa T, Nakada M. Tetrahedron: Asymmetry 2015; 26: 262
  • 11 Nagatani K, Minami A, Tezuka H, Hoshino Y, Nakada M. Org. Lett. 2017; 19: 810
  • 12 Yechezkel T, Ghera E, Ostercamp D, Hassner A. J. Org. Chem. 1995; 60: 5135
  • 13 Trimitsis G, Beers S, Ridella J, Carlon M, Cullin D, High J, Brutts D. J. Chem. Soc., Chem. Commun. 1984; 1088
  • 14 The reaction of 5 and 6 (1.5 equiv) was performed under the optimized conditions shown in Table 3 by using Cu(OTf)2 (20 mol%) and L3 (25 mol%) in the presence of 3 Å MS (300 wt%) in 2:1 toluene–DCM (2:1) at –40 °C. However, 7 was obtained in only 17% yield (dr 7.4:1, 74% ee) after 40 h.
  • 15 Sawada T, Nakada M. Tetrahedron: Asymmetry 2012; 23: 350
  • 16 The reaction of 2-(2,4,6-trimethylphenylsulfonyl)cyclopent-2-en-1-one and 8f (1.5 equiv) was examined by using Cu(OTf)2 (100 mol%), ligand L3 (110 mol%), and 3 Å MS in 2:1 toluene–DCM at –60 °C. The product was obtained in 95% yield with a dr of 42:1; however, the enantioselectivity was decreased (53% ee).
  • 17 Compound 9f A mixture of Cu(OTf)2 (4.5 mg, 0.0124 mmol), L3 (6.5 mg, 0.0155 mmol), and 3 Å MS (38.7 mg) in 2:1 toluene–DCM (0.290 mL) was stirred at rt for 1.5 h. A solution of 5 (13.2 mg, 0.0594 mmol) in 2:1 toluene–DCM (0.600 mL) was added and the mixture was stirred at rt for 1.5 h. A solution of 8f (17.7 mg, 0.0810 mmol) in 2:1 toluene–DCM (0.300 mL) was added at –60 °C, and the mixture was stirred at –60 °C for 41.5 h. The reaction was quenched with sat. aq NaHCO3 (2.0 mL) and the mixture was filtered. The aqueous layer was extracted with EtOAc (3 × 5 mL), and the combined organic layer was washed with H2O (5 mL) and brine (5 mL) then dried (Na2SO4) and concentrated under reduced pressure. The residue was purified by flash column chromatography [silica gel, hexane–EtOAc (6:1)] to afford 9f containing inseparable L3 as a white solid; yield: 26.0 mg [0.0531 mmol, 89% (calcd based on 1H NMR)]; Rf = 0.70 (hexane–EtOAc, 2:3). [α]D 21 –55 (c 0.65, DCM). IR (ATR): 1747, 1666, 1310, 1151, 1084, 1063, 970, 755, 729, 686, 621, 586, 555, 533, 494 cm–1. 1H NMR (500 MHz, CDCl3): δ = 7.86 (d, J = 7.5 Hz, 2 H), 7.70 (t, J = 7.5 Hz, 1 H), 7.58 (dd, J = 7.5, 7.5 Hz, 2 H), 3.83 (d, J = 5.0 Hz, 1 H), 3.25 (ddd, J = 8.0, 5.5, 5.0 Hz, 1 H), 2.90 (ddd, J = 7.5, 7.0 , 5.0 Hz, 1 H), 2.48–2.38 (m, 1 H), 2.36–2.25 (m, 1 H), 1.88–1.79 (m, 1 H), 1.43 (s, 9 H), 1.18 (d, J = 7.5 Hz, 3 H). 13C NMR (125 MHz, CDCl3): δ = 206.4, 202.8, 137.8, 134.3, 129.2, 71.8, 50.5, 48.6, 40.5, 37.7, 29.7, 23.6, 14.6. HRMS (ESI): m/z [M + Na]+ calcd for C18H24NaO4S2: 391.1014; found: 391.1007.
  • 18 CCDC 2235289 contains the supplementary crystallographic data for compound 9f′. The data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/structures
  • 19 Sugiyama R, Nakada M. Synlett 2023; in press
  • 20 De Clercq P, Vandewalle M. J. Org. Chem. 1977; 42: 3447

Zoom Image
Figure 1 Structures of cotylenin A, calcitriol lactone, damsin, ergolide, 1a, and 1b
Zoom Image
Scheme 1 Preparation of chiral building block 2 containing the chiral scaffold 1a by a catalytic asymmetric Mukaiyama–Michael reaction
Zoom Image
Figure 2 Structures of ligands L112
Zoom Image
Scheme 2 Transformation of 9f into 9f′
Zoom Image
Figure 3 Structure of 9f′ and its X-ray crystal structure
Zoom Image
Figure 4 Proposed models for the observed high enantioselectivity in the Mukaiyama–Michael reaction of 5
Zoom Image
Figure 5 Proposed models for the observed high diastereoselectivity in the Mukaiyama–Michael reaction of 5. The complex composed of the bisoxazoline ligand L3 and Cu(OTf)2 is omitted for clarity
Zoom Image
Figure 6 Proposed model for the highly stereoselective protonation
Zoom Image
Scheme 3 Outline of our enantioselective formal total synthesis of damsin