Formal Total Synthesis of Salvinorin A.

The generation of the quaternary stereocenter at the C9 position of salvinorin A precursors by the Claisen rearrangement was investigated. The required allyl alcohol was prepared from a Wieland-Miescher ketone using a known γ-hydroxylation, reduction of the enone double bond, cyanohydrin formation, and elimination, yielding an unsaturated nitrile. A two-step reduction led to the required allyl alcohol. The subsequent Johnson-Claisen rearrangement provided a mixture of two diastereomeric 1,4-unsaturated esters in a ratio of around 2.6 : 1. The major isomer could be converted to a key intermediate of the Hagiwara synthesis of salvinorin A.

Introduction

Among the family of clerodane diterpenes, salvinorin A (1) clearly stands out (Figure 1). This has to do with the fact that it is a potent natural hallucinogen. This came as a surprise since it is not an alkaloid. Its isolation from leaves of Salvia divinorum and its structure were reported in 1982. This plant is used by indigenous people in Mexico for ritual purposes. The discovery that salvinorin A (1) is a potent and selective Κ opioid receptor agonist triggered a lot of interest among the biology and chemistry communities. Accordingly, a SciFinder search on salvinorin A generates around 500 hits. Continuing efforts led to the isolation of around 20 salvinorin A analogs from Salvia divinorum.

Figure 1

Structure of salvinorin A (1), including an alternative depiction showcasing the spatial substituent arrangement (bottom), and some related compounds.

Some related diterpenes, such as 8‐epidiosbulbin E acetate (2) from Dioscorea bulbifera L. (Figure 1), have also been isolated. It was reported that 2 has plasmid‐curing activity against multidrug‐resistant bacteria by rendering them sensitive to antibiotics again. Collybolide (3) was isolated from the mushroom Collybia maculata. Even though the furan ring in collybolide (3) points to the α‐face of the molecule, it was reported that it is also an agonist for K opioid receptors with an activity comparable to 1.

In general, agonists of opioid receptors show analgesic effects but are typically also associated with side effects. Agonists for the kappa opioid receptor (KOR) hold promise for the treatment of Central Nervous System (CNS) disorders. A recent review article summarizes the effects of structural modifications, mostly on the periphery of the core structure, on the biological activity. For example, replacement of the acetate group at C2 by alkoxymethyl ethers can lead to analogs with smaller K i values. Modifications at C4 are generally detrimental. Small changes on the furan ring seem to be tolerated. Synthetic routes that allow for serious modifications on the core structure have been published by Shenvi et al. and Prisinzano et al.

Even though many routes to clerodane diterpenes are known, the structure of salvinorin A (1) presents special challenges. These include seven stereocenters, the C ring lactone with a 3‐furyl substituent and the two axial methyl groups a C5 and C9. As has been pointed out by Shenvi et al., the 1,3‐diaxial interaction between 9‐Me and 12‐H facilities epimerization at C8, leading to loss in activity.

So far, six total syntheses for salvinorin A (1) were reported by four groups. Key issues are the synthesis of the decalin core and the time of the furan installation (Scheme 1). In the Evans synthesis, the furan was introduced early on into the macrolactone 4 which could be cyclized to decalin 5 through a domino Michael reaction. The Hagiwara group developed two routes to 1 from Wieland‐Miescher ketone derivative 6. In a key step, the quaternary center at C9 was created by an enolate alkylation. The problem in this step is the unwanted elimination of the hydroxyl group from the intermediate enolate. The two routes differ in the way the functional groups at C4 and C8 were introduced. In the first route, hydroboration on the exocyclic double bonds on diene 8 came to use, whereas in the second route, carbonylation on the enol triflates 9 served this purpose. Both routes feature late‐stage introduction of the 3‐furyl group. A key step in the synthesis of Forsyth et al. was an intramolecular Diels‐Alder reaction of triene 11. This substrate originated from d‐tartrate. A Tsuji allylation on enol carbonate 12 set the stereochemistry at C9. The Metz group published two similar routes to salvinorin A (1). In a key step, triene 14, which already contained the C‐ring, underwent an intramolecular Diels‐Alder reaction to the salvinorin A core 15. Remarkably, substrate 13 also had been prepared by an intramolecular Diels‐Alder reaction.

Scheme 1

Key steps in previous syntheses of salvinorin A (1). TES=triethylsilyl, BOM=benzyloxymethyl, MPM=(4‐methoxyphenyl)methyl (para‐methoxybenzyl), dppf=1,1′‐bis(diphenylphosphino)ferrocene, PIFA=(bis(trifluoroacetoxy)iodo)benzene (phenyliodine bis(trifluoroacetate).

Results and Discussion

We wondered whether other methods would allow for a concise introduction of the C9 quaternary center on substrates that would be easily available from Wieland‐Miescher ketone derivatives. In this context, we envisioned a Claisen rearrangement on compounds of type 16. Such reactions had previously been performed on less functionalized compounds (Scheme 2). Kakisawa et al. reported the formation of the two enals 19 a and 19 b in a ratio of 85 : 15 (95 % total yield) upon heating allyl vinyl ether 18 to 200 °C. In a related study, Terashima et al. described the Johnson‐Claisen rearrangement on cis‐decalin derivative 20. This led to the preferential rearrangement on the α‐face, giving unsaturated esters 21 a and 21 b (ratio=3 : 2, 50 % combined yield). In the first example, it was argued that the axial methyl group interferes with the transition state of the rearrangement on the α‐face. In the second example, which is described for a cis‐decalin, it seems that rearrangement is favored syn to the allylic C−H. At the outset it was not clear what an influence a hydroxy substituent at C4 (decalinone numbering) would have on the facial selectivity of a Claisen rearrangement of substrates 16. One might speculate that the axial positioned OP group directs the rearrangement to the β‐face by interfering with the rearrangement transition state on the α‐face. In this paper, we describe our results of this study.

Scheme 2

Plan for creation of the stereochemistry at C9 (salvinorin A numbering) together with some known related examples from the literature. HQ=hydroquinone, DCB=ortho‐dichlorobenzene (1,2‐dichlorobenzene).

We started the synthesis of Wieland‐Miescher ketone 23 from triketone 22, using conditions developed by Theodorakis et al. where d‐phenylalanine (Phe; 1 equiv.) and d‐camphorsulfonic acid (CSA; 0.5 equiv.) served as chirality inducers (Scheme 3).[ , ] Subsequent acetalization of the nonconjugated keto function provided enone 24 in excellent yield as well. Hydroxylation of the γ‐position of enone 24 using oxygen in presence of KOtBu and the copper‐aluminium mixed oxide (Cu−Al Ox) furnished alcohol 6. The method used by the Hagiwara group to prepare compound 6 from enone 24 (O2, KOH, MeOH, 2.7 d, r.t.) was less efficient in our hands.[ , ] The best yield we could get under these conditions was 47 % of 6. Based on literature precedence, the stereochemistry at C4 was assigned as shown, which means that hydroxylation took place on the α‐face.[ , ] Accordingly, hydrogenation of the enone double bond on the derived silyl ether 25 was expected to occur from the top‐face, giving the trans decalinone 26 in good yield as a single isomer, which was indeed the case. In order to establish the allyl alcohol functionality, ketone 26 was converted to cyanohydrin 27 with trimethylsilyl cyanide in a DMSO/H2O mixture. The structure of 27 was unambiguously confirmed by an X‐ray analysis. It clearly showed the all‐cis arrangement of the substituents at C4, C5 and C8a. Treatment of cyanohydrin 27 with thionyl chloride and pyridine led to unsaturated nitrile 28 with the correct double bond position. A two‐step reduction of the nitrile, first with DIBAL−H to enal 29, second with sodium borohydride, gave rise to allyl alcohol 30. A similar sequence to convert a ketone to an allyl alcohol had been used by Kakisawa et al. in the synthesis of a trans‐clerodane diterpene.

Scheme 3

Synthesis of allyl alcohol 30 as substrate for the Claisen rearrangement. ORTEP plot of cyanohydrin 27 (hydrogen atoms omitted for clarity). Displacement ellipsoids are drawn at the 50 % probability level.

We first tried the Ireland‐Claisen variant on the acetate 31, easily obtained from alcohol 30 (Scheme 4). Deprotonation of the acetate, followed by enolate trapping with trimethylsilyl chloride and subsequent heating of the silyl ketene acetal led to a complex mixture, which contained none of the acid 32. Therefore, we switched to the Johnson‐Claisen rearrangement. Thus, stirring allyl alcohol 30 with an excess of triethyl orthoacetate in ortho‐dichlorobenzene at 180 °C for 18 h led to the rearranged esters 33 a and 33 b. The conditions for the rearrangement were those reported by Terashima et al. The desired isomer 33 a was obtained in 53 % yield. The minor diastereomer 33 b was formed in around 20 % yield. Major differences in the 13C NMR spectra of the two isomers are the different chemical shift values for the C4a carbon atoms. In 33 a, C4a resonates at δ=46.5 ppm, whereas the corresponding resonance in 33 b appears at δ=54.2 ppm. Separation of the isomers was possible by flash chromatography on silica gel. The structure of the minor isomer 33 b could be confirmed by X‐ray analysis (Scheme 4).

Scheme 4

Study on the Claisen rearrangement on allyl alcohol 30. ORTEP plot of unsaturated ester 33 b. Displacement ellipsoids are drawn at the 50 % probability level.

One might speculate that the moderate selectivity results from two effects counteracting each other (Scheme 5). Thus, attack on the α‐face would be favored due to a Fürst‐Plattner type situation, since the transition state would have a chair‐like conformation. However, since the α‐face is sterically hindered due to the axial methyl and the OSiEt3 group, the β‐attack, overall, seems to be favored. In a recent example from the literature, both the stereoelectronic effect (Fürst‐Plattner rule) and steric effects seem to act synergistically. Thus, the Johnson‐Claisen rearrangement on allyl alcohol 34 gave only the desired α‐isomer 35 (Scheme 5). We believe that this simple model might help to predict or to rationalize the diastereoselectivity of a Claisen rearrangement on related substrates.

Scheme 5

Model to explain the observed diastereoselectivity on decalin systems having an allyl alcohol in the substrate. The Fürst‐Plattner rule and steric effects may counteract each other or act synergistically as in the rearrangement of 34 to 35.

For chemical correlation, acetic ester derivative 33 a was converted to known 37 in two steps (Scheme 6). Thus, our work constitutes a formal total synthesis of salvinorin A (1). We note that the modern catalysts used for enantioselective Robinson annulation do tolerate a range of substituents on the substrates. Thus, this strategy might be used to make salvinorin A analogs with different alkyl groups at C10, C5 or C9.

Scheme 6

Conversion of 1,4‐unsaturated ester 33 a to the known salvinorin A intermediate 37.

Conclusion

Wieland‐Miescher ketone 23 was converted in 12 steps to the known diketo ester 37, a key intermediate in the Hagiwara synthesis of salvinorin A (1). After γ‐hydroxylation and hydrogenation of the enone double bond, cyanohydrin formation opened the way to allyl alcohol 30. In an Ireland‐Claisen rearrangement, the quaternary stereocenter at C9 (salvinorin A numbering) was introduced. The desired isomer was obtained in 53 % yield. This work might help to predict the stereochemical outcome of related rearrangements on decalin systems.

Experimental Section

General. All reactions were performed under nitrogen atmosphere. All solvents used in the reactions were purified before use. The progress of the reactions was followed by TLC (POLYGRAM SIL G/UV254). Flash chromatography was performed on silica gel Silica M, 0.04–0.63 mm, from Machery‐Nagel GmbH & Co. KG, Germany. Distilled petroleum ether with a boiling range of 40–60 °C was used. Dry tetrahydrofuran and toluene were distilled from sodium and benzophenone, whereas CH2Cl2 was distilled from CaH2. Methanol and Ethanol were used in HPLC grade quality. All commercially available compounds (abcr, Acros, Aldrich, Fluka, Merck and TCI) were used without purification. 1H (400.160 MHz) and 13C (100.620 MHz) spectra were recorded on a Bruker Avance 400 III HD spectrometer. CDCl3 was used as solvent at room temperature. The 1H NMR spectra were referenced to the residual signal of the non‐deuterated solvent component (CDCl3, 7.25 ppm) and the 13C NMR spectra to the signal of the deuterated solvent (CDCl3, 77.0 ppm). Peak assignments were made by NMR spectroscopy (1H, 13C, DEPT‐135, H,H‐COSY, HSQC, NOESY and HMBC). HRMS (ESI‐TOF) analyses were performed on a Bruker maXis 4G system.

Deposition Numbers 2124065 (for contain the supplementary crystallographic data for this paper. These data are provided free of charge by the joint Cambridge Crystallographic Data Centre and Fachinformationszentrum Karlsruhe Access Structures service.

(3‐Oxopentyl)‐cyclohexane‐1,3‐dione (22). To a stirred solution of 2‐methylcyclohexane‐1,3‐dione (1.89 g, 15.0 mmol, 1 equiv.) in ethyl acetate (100 mL) under N2 were added ethyl vinyl ketone (1.63 mL, 16.5 mmol, 1.1 equiv.) and triethylamine (2.70 mL, 19.5 mmol, 1.3 equiv.) at room temperature. The resulting suspension was heated to 75 °C for 18 h. After cooling to room temperature, most of the solvent was evaporated. The residue was purified by flash chromatography (petroleum ether/ethyl acetate, 3 : 2) to obtain triketone 22 as yellowish clear oil (2.64 g, 12.5 mmol, 84 %). Rf=0.24 (petroleum ether/ethyl acetate, 2 : 1); 1H NMR (400 MHz, CDCl3): δ (ppm)=2.78–2.71 (m, 4H, 4‐H, 6‐H), 2.39 (q, J=7.3 Hz, 2H, 4′‐H), 2.33 (t, J=7.5 Hz, 2H, 2′‐H), 2.11–2.01 (m, 1H, 5‐H), 2.07 (t, J=7.5 Hz, 2H, 1′‐H), 1.95–1.85 (m, 1H, 5‐H), 1.24 (s, 3H, CH3), 1.03 (t, J=7.3 Hz, 3H, 11‐H); 13C NMR (101 MHz, CDCl3): δ (ppm)=210.3 (C‐3’), 210.1 (C‐1, C‐3), 64.4 (C‐2), 37.7 (C‐4, C‐6), 37.0 (C‐2′), 36.0 (C‐4′), 29.8 (C‐1′), 19.8 (C‐5), 17.6 (CH3), 7.7 (C‐11).

(8aR)‐5,8a‐Dimethyl‐1,2,3,4,6,7,8,8a‐octahydronaphthalene‐1,6‐dione (23). To a stirred solution of (3‐oxopentyl)‐cyclohexane‐1,3‐dione 22 (1.50 g, 7.1 mmol, 1 equiv.) in dry DMF (105 mL) was added under N2 d‐phenylalanine (1.18 g, 7.1 mmol. 1 equiv.) and d‐camphorsulfonic acid (0.83 g, 3.6 mmol, 0.5 equiv.). The resulting mixture was stirred for 24 h at room temperature. Thereafter, the mixture was heated to 40 °C for another 24 h. The temperature was raised 10 °C each day until 70 °C was reached. The mixture was allowed to reach room temperature and quenched with cold saturated aqueous NaHCO3 (60 mL). The layers were separated and the aqueous layer was extracted with Et2O (3×60 mL). The combined organic layers were washed with water (3×60 mL) and saturated NaCl solution (60 mL), then dried over MgSO4, filtered and concentrated in vacuo. The resulting orange oil was purified by flash chromatography (petroleum ether/ethyl acetate, 2 : 1) and subsequently by recrystallization (hexane/Et2O, 5 : 1) at −20 °C to obtain diketone 23 (1.14 g, 6.0 mmol, 83 %, 98 % ee) as thin white needles. Rf=0.33 (petroleum ether/ethyl acetate, 4 : 1); [α]D 25=−137.3 (c 1.0, MeOH), lit. [α]D 25=−139.0 (c 1.0, MeOH); 1H NMR (400 MHz, CDCl3): δ (ppm)=2.92–2.85 (m, 1H, 4‐H), 2.73–2.65 (m, 1H, 2‐H), 2.55–2.41 (m, 4H, 7‐H, 4‐H, 2‐H), 2.19–2.04 (m, 3H, 3‐H, 8‐H), 1.82 (d, J=1.3 Hz, 3H, 5‐CH3), 1.80–1.72 (m, 1H, 3‐H), 1.43 (s, 3H, 8a‐CH3); 13C NMR (101 MHz, CDCl3): δ (ppm)=212.2 (C‐1), 197.7 (C‐6), 158.1 (C‐4a), 130.8 (C‐5), 50.7 (C‐8a), 37.4 (C‐2), 33.3 (C‐7), 29.6 (C‐8), 27.3 (C‐4), 23.4 (8a‐CH3), 21.5 (C‐3), 11.3 (5‐CH3).

(8′aR)‐5′,8′a‐Dimethyl‐3′,4′,6′,7′,8′,8′a‐hexahydro‐2′H‐spiro[1,3‐dioxolane‐2,1′‐naphthalen]‐6′‐one (24). In a flask equipped with a Dean‐Stark trap, a stirred solution of ethlene glycol (6.33 mL, 11.3 mmol, 14.5 equiv.) in dry benzene (55 mL) was heated to reflux for 2 h under N2 atmosphere. The trap was removed and the mixture allowed to reach room temperature. In a separate flask, diketone 23 (1.50 g, 7.8 mmol, 1 equiv.) was dissolved in benzene (15 mL) and added to the ethylene glycol solution over 10 min, followed by the addition of pTsOH (29.6 mg, 0.16 mmol, 0.02 equiv.). The resulting mixture was heated to reflux for an additional 2 h. After reaching room temperature, the reaction was quenched by addition of 1 % aqueous NaHCO3 (50 mL). The layers were separated and the aqueous layer was extracted with Et2O (3×50 mL). The combined organic layers were washed with water and saturated NaCl solution (50 mL each), dried over MgSO4, filtered and concentrated in vacuo. The resulting yellow oil was purified by flash chromatography (petroleum ether/ethyl acetate, 2 : 1) to yield ketal 24 (1.73 g, 7.33 mmol, 94 %) as colorless viscous oil. Rf=0.31 (petroleum ether/ethyl acetate, 2 : 1); 1H NMR (400 MHz, CDCl3): δ (ppm)=4.02–3.91 (m, 4H, OCH2CH2O), 2.77–2.72 (m, 1H, 8‐H), 2.52–2.36 (m, 2H, 7‐H), 2.29–2.07 (m, 2H, 4‐H), 1.95–1.81 (m, 2H, 2‐H, 3‐H), 1.80 (dd, 3H, 5‐CH3), 1.73–1.61 (m, 3H, 2‐H, 3‐H, 8‐H), 1.35 (s, 3H, 8a‐CH3); 13C NMR (101 MHz, CDCl3): δ (ppm)=198.7 (C‐6), 160.1 (C‐4a), 130.2 (C‐5), 112.8 (C‐1), 65.3 (OCH2CH2O), 65.1 (OCH2CH2O), 45.3 (C‐8a), 33.7 (C‐7), 29.7 (C‐2), 26.5 (C‐8), 26.4 (C‐4). 21.4 (C‐3), 20.9 (8a‐CH3), 11.5 (5‐CH3).

(4′S,8a′R)−4′‐Hydroxy‐5′,8a′‐dimethyl‐3′,4′,8′,8a′‐tetrahydro‐2′H‐spiro[[1,3]dioxolane‐2,1′‐naphthalen]‐6′(7′H)‐one (6). Catalyst preparation: To a stirred solution of CuCl2 (10.0 g, 74.4 mmol) and AlCl3 ⋅ 6H2O (8.0 g, 60.0 mmol) in water (100 mL) was added a solution of Na2CO3 (2.54 g, 24.0 mmol) and NaOH (10.4 g, 260.0 mmol) in water (200 mL) over a period of 1.5 h. The resulting mixture was stirred at 77 °C for 22 h and subsequently filtered. The filter cake was washed with warm water. The black catalyst was dried at 95 °C overnight, finely crushed and left exposed to air for an additional 72 h.

The Cu−Al Ox catalyst (28.6 mg) was suspended in EtOH (1.7 mL) and the mixture stirred for 10 min open to air. Ketal 24 (100.0 mg, 0.42 mmol, 1 equiv.) in EtOH (0.3 mL) and tBuOK (59.4 mg, 0.53 mmol, 1.25 equiv.) were added followed by stirring of the mixture for 22 h at room temperature. Thereafter, the mixture was filtered through a pad of Celite, the filter cake was rinsed with MeOH and the filtrate was concentrated in vacuo. The residue was purified by flash chromatography (petroleum ether/ethyl acetate, 1 : 1) to give allylic alcohol 6 (62.8 mg, 0.25 mmol, 59 %) as colorless oil. Rf=0.17 (petroleum ether/ethyl acetate, 2 : 1); 1H NMR (400 MHz, CDCl3): δ (ppm)=4.94 (t, J=3.1 Hz, 1H, 4‐H), 4.03–3.92 (m, 4H, OCH2CH2O), 2.59–2.46 (m, 2H, 7‐H), 2.32–2.20 (m, 2H, 2‐H, 8‐H), 1.99–1.94 (m, 2H, 3‐H), 1.88 (s, 3H, 5‐CH3), 1.72–1.66 (m, 1H, 8‐H), 1.61–1.56 (dt, 1H, 2‐H), 1.53 (s, 3H, 8a‐CH3); 13C NMR (101 MHz, CDCl3): δ (ppm)=199.7 (C‐6), 158.1 (C‐4a), 132.9 (C‐5), 112.4 (C‐1), 66.0 (C‐4), 65.3 (OCH2CH2O), 65.0 (OCH2CH2O), 44.2 (C‐8a), 33.6 (C‐7), 28.9 (C‐3), 27.3 (C‐8), 24.6 (C‐2), 22.5 (8a‐CH3), 11.0 (5‐CH3).

(4′S,8a′R)‐5′,8a′‐Dimethyl‐4′‐((triethylsilyl)oxy)‐3′,4′,8′,8a′‐tetrahydro‐2′H‐spiro[[1,3]dioxolane‐2,1′‐naphthalen]‐6′(7′H)‐one (25). To a stirred solution of alcohol 6 (1.90 g, 7.5 mmol, 1 equiv.) in dry CH2Cl2 (20 mL) under N2 were sequentially added imidazole (3.06 g, 45.0 mmol, 6 equiv.) and TESCl (3.78 mL, 22.5 mmol, 3 equiv.). The resulting mixture was stirred for 45 min at room temperature, then diluted with CH2Cl2 (20 mL) and finally treated with aqueous saturated NaHCO3 (40 mL). The layers were separated and the aqueous layer was extracted with CH2Cl2 (3×20 mL). The combined organic layers were dried over MgSO4, filtered and concentrated in vacuo to yield silyl ether 25 a colorless oil (3.18 g) which was used in the next step without any further purification. Rf=0.73 (petroleum ether/ethyl acetate, 2 : 1); 1H NMR (400 MHz, CDCl3): δ (ppm)=4.88 (t, J=2.8 Hz, 1H, 4‐H), 4.06–3.86 (m, 4H, OCH2CH2O), 2.60–2.41 (m, 2H, 7‐H), 2.41–2.17 (m, 2H, 2‐H, 8‐H), 1.85 (s, 3H, 5‐CH3), 1.84–1.74 (m, 2H, 3‐H), 1.73–1.55 (m, 1H, 2‐H, 8‐H), 1.53 (s, 3H, 8a‐CH3), 1.04–0.87 (m, 9H, Si(CH2CH 3)3), 0.70–0.50 (m, 6H, Si(CH 2CH3)3); 13C NMR (101 MHz, CDCl3): δ (ppm)=200.2 (C‐6), 157.8 (C‐4a), 131.4 (C‐5), 112.8 (C‐1), 66.5 (C‐4), 65.3 (OCH2CH2O), 65.0 (OCH2CH2O), 45.0 (C‐8a), 33.7 (C‐7), 30.7 (C‐3), 27.7 (C‐8), 24.9 (C‐2), 22.3 (8a‐CH3), 11.4 (5‐CH3), 6.8 (Si(CH2CH3)3), 5.0 (Si(CH2CH3)3); HRMS (ESI‐TOF): calcd. for C20H34O4Si 389.21186 [M+Na]+, found 389.21184.

(4′S,4a′R,5′S,8a′R)‐5′,8a′‐Dimethyl‐4′‐((triethylsilyl)oxy)hexahydro‐2′H‐spiro[[1,3]dioxolane‐2,1′‐naphthalen]‐6′(7′H)‐one (26). Crude enone 25 (3.18 g) and 10 % Pd/C (320 mg) in ethyl acetate (55 mL) were stirred under a H2 atmosphere (1 atm) for 17 h. The resulting mixture was filtered through a pad of Celite, the filter cake was washed with ethyl acetate, and the filtrate was concentrated in vacuo. The residue was purified by flash chromatography (petroleum ether/ethyl acetate, 12 : 1) to provide ketone 26 (2.41 g, 6.5 mmol, 87 % over two steps) as a colorless oil. Rf=0.59 (petroleum ether/ethyl acetate, 3 : 1); 1H NMR (400 MHz, CDCl3): δ (ppm)=4.07–3.75 (m, 5H, OCH2CH2O, 4‐H), 2.72 (td, J=14.6 Hz, 6.1 Hz, 1H, 7‐Ha), 2.56 (quint, J=13.5, 7.0 Hz, 1H, 5‐H), 2.28–2.08 (m, 2H, 2‐H, 7‐H), 2.00–1.85 (m, 2H, 4a‐H, 8‐H), 1.83–1.66 (m, 2H, 3‐H), 1.64–1.60 (m, 1H, 8‐H), 1.57 (s, 3H, 8a‐CH3), 1.48 (t, J=3.0 Hz, 1H, 2‐H), 1.43 (d, J=7.7 Hz, 3H, 5‐CH3), 1.04–0.92 (m, 9H, Si(CH2CH 3)3), 0.63 (quart., J=8.0  Hz, 6H, Si(CH 2CH3)3); 13C NMR (101 MHz, CDCl3): δ (ppm)=215.8 (C‐6), 112.6 (C‐1), 72.3 (C‐4), 65.1 (OCH2CH2O), 64.9 (OCH2CH2O), 50.3 (C‐5), 46.3 (C‐4a), 42.6 (C‐8a), 34.3 (C‐7), 32.6 (C‐3), 31.5 (C‐8), 26.2 (C‐2), 19.2 (8a‐CH3), 15.9 (5‐CH3), 7.0 (Si(CH2CH3)3), 5.2 (Si(CH2CH3)3); HRMS (ESI‐TOF): calcd. for C20H36O4Si 391.22751 [M+Na]+, found 391.22738.

(4′S,4a′R,5′S,6′R,8a′R)−6′‐Hydroxy‐5′,8a′‐dimethyl‐4′‐((triethylsilyl)oxy)octahydro‐2′H‐spiro[[1,3]dioxolane‐2,1′‐naphthalene]‐6′‐carbonitrile (27). To a stirred solution of ketone 26 (2.40 g, 6.5 mmol, 1 equiv.) in a mixture of DMSO/H2O (5 : 1, 75 mL) was added TMSCN (2.46 mL, 19.5 mmol, 3 equiv.). The resulting mixture was stirred at room temperature for 19 h, after which time the solution was diluted with H2O (150 mL). The aqueous layer was extracted with ethyl acetate (3×200 mL). The combined organic layers were washed with water and saturated NaCl solution (200 mL each), dried over MgSO4, filtered, and concentrated in vacuo. The residue was purified by flash chromatography (petroleum ether/ethyl acetate, 5 : 1) to obtain cyanohydrin 27 (2.09 g, 5.3 mmol, 81 %) as white solid. Rf=0.50 (petroleum ether/ethyl acetate, 3 : 1); 1H NMR (400 MHz, CDCl3): δ (ppm)=4.15–3.76 (m, 5H, OCH2CH2O, 4‐H), 2.37 (ddd, J=7.3 Hz, 3.9, 1.5 Hz, 1H, 5‐H), 2.14–2.04 (m, 1H, 2‐H), 2.03–1.91 (m, 3H, 3‐H, 4a‐H), 1.91–1.79 (m, 2H, 7‐H, 8‐H), 1.66 (dq, J=3.4 Hz, 13.8 Hz, 1 H, 7‐H), 1.45 (td, J=12.9, 3.3 Hz, 1H, 2‐H), 1.36 (s, 3H, 8a‐CH3), 1.32 (d, J=7.3 Hz, 3H, 5‐CH3), 1.29–1.23 (m, 1H, 8‐H), 1.04–0.91 (m, 9H, Si(CH2CH3)3), 0.69–0.54 (m, 6H, Si(CH2CH3)3); 13C NMR (101 MHz, CDCl3): δ (ppm)=122.7 (CN), 112.1 (C‐1), 74.0 (C‐6), 72.8 (C‐4), 65.2 (OCH2CH2O), 64.8 (OCH2CH2O), 44.1 (C‐4a), 43.8 (C‐5), 42.4 (C‐8a), 32.5 (C‐7), 28.8 (C‐8), 28.7 (C‐3), 26.1 (C‐2), 19.4 (8a‐CH3), 10.6 (5‐CH3), 7.0 (Si(CH2CH3)3), 5.1 (Si(CH2CH3)3); HRMS (ESI‐TOF): calcd. for C21H37NO4Si 418.23841 [M+Na]+, found 418.23870.

(4′S,4a′R,8a′R)‐5′,8a′‐Dimethyl‐4′‐((triethylsilyl)oxy)‐3′,4′,4a′,7′,8′,8a′‐hexahydro‐2′H‐spiro[[1,3]dioxolane‐2,1′‐naphthalene]‐6′‐carbonitrile (28). To a stirred solution of cyanohydrin 27 (1.00 g, 2.5 mmol, 1 equiv.) in dry toluene (60 mL) was added pyridine (0.82 mL, 10.1 mmol, 4 equiv.) under N2 at room temperature. The solution was cooled to 0 °C and treated with SOCl2 (0.37 mL, 5.1 mmol, 2 equiv.) over a period of 15 min. The resulting mixture was heated to reflux for 24 h and allowed to reach room temperature before being quenched by addition of water (60 mL). The layers were separated and the aqueous layer was extracted with CH2Cl2 (3×60 mL). The combined organic layers were washed with saturated NaCl solution (60 mL), dried over MgSO4, filtered and concentrated in vacuo. The orange residue was purified by flash chromatography (petroleum ether/ethyl acetate, 10 : 1) to yield unsaturated nitrile 28 (740.2 mg, 2.0 mmol, 78 %) as yellow oil. Rf=0.55 (petroleum ether/ethyl acetate, 3 : 1); 1H NMR (400 MHz, CDCl3): δ (ppm)=4.40–4.33 (m, 1H, 4‐H), 4.02–3.80 (m, 4H, OCH2CH2O), 2.37–2.31 (m, 1H, 4a‐H), 2.30–2.15 (m, 3H, 2‐H, 7‐H), 2.12 (s, 3H, 5‐CH3), 1.81–1.76 (m, 2H, 3‐H), 1.58–1.44 (m, 3H, 2‐H, 8‐H), 1.14 (s, 3H, 8a‐CH3), 1.02–0.90 (m, 9H, Si(CH2CH3)3), 0.68–0.55 (m, 6H, Si(CH2CH3)3); 13C NMR (101 MHz, CDCl3): δ (ppm)=154.1 (C‐5), 119.6 (CN), 112.2 (C‐1), 106.7 (C‐6), 67.0 (C‐4), 65.3 (OCH2CH2O), 65.1 (OCH2CH2O), 49.2 (C‐4a), 40.6 (C‐8a), 31.5 (C‐3), 27.3 (C‐8), 25.9 (C‐2), 24.7 (C‐7), 20.0 (5‐CH3), 17.6 (8a‐CH3), 7.0 (Si(CH2CH3)3), 5.3 (Si(CH2CH3)3); HRMS (ESI‐TOF): calcd. for C21H35NO3Si 400.22784 [M+Na]+, found 400.22718.

(4′S,4a′R,8a′R)‐5′,8a′‐Dimethyl‐4′‐((triethylsilyl)oxy)‐3′,4′,4a′,7′,8′,8a′‐hexahydro‐2′H‐spiro[[1,3]dioxolane‐2,1′‐naphthalene]‐6′‐carbaldehyde (29). A stirred solution of nitrile 28 (6.2 g, 16.3 mmol, 1 equiv.) in dry toluene (250 mL) was treated with DIBAL−H (1.0 m in hexane, 35 mL, 2.1 equiv.) under N2 at −78 °C followed by stirring of the mixture for 4 h. Thereafter, the reaction was quenched at −78 °C by slow addition of 10 % NaOH (150 mL). The layers were separated and the aqueous layer was extracted with Et2O (3×150 mL). The combined organic layers were washed with saturated NaCl solution, dried over Na2SO4, filtered and concentrated in vacuo. The residue was purified by flash chromatography (petroleum ether/ethyl acetate, 8 : 1) to yield aldehyde 29 (4.8 g, 12.6 mmol, 77 %) as a colorless oil. Rf=0.74 (petroleum ether/ethyl acetate, 3 : 1); 1H NMR (400 MHz, CDCl3): δ (ppm)=10.21 (s, 1H, CHO), 4.50 (q, J=2.5 Hz, 1H, 4‐H), 4.01–3.85 (m, 4H, OCH2CH2O), 2.44 (s, 1H, 4a‐H), 2.40–2.34 (m, 1H, 7‐Ha), 2.26 (s, 3H, 5‐CH3), 2.22–2.08 (m, 2H, 2‐H, 7‐H), 1.86–1.76 (m, 2H, 3‐H), 1.55–1.47 (m, 3H, 2‐H, 8‐H), 1.11 (s, 3H, 8a‐CH3), 1.01–0.92 (m, 9H, Si(CH2CH3)3), 0.69–0.53 (m, 6H, Si(CH2CH3)3); 13C NMR (101 MHz, CDCl3): δ (ppm)=191.6 (CO), 156.6 (C‐5), 133.9 (C‐6), 112.5 (C‐1), 66.8 (C‐4), 65.3 (OCH2CH2O), 65.1 (OCH2CH2O), 50.8 (C‐4a), 40.7 (C‐8a), 31.7 (C‐3), 27.1 (C‐8), 26.1 (C‐2), 19.9 (C‐7), 17.8 (8a‐CH3), 14.2 (5‐CH3), 7.0 (Si(CH2CH3)3), 5.4 (Si(CH2CH3)3); HRMS (ESI‐TOF): calcd. for C21H36NO4Si 403.22751 [M+Na]+, found 403.22744.

((4′S,4a′R,8a′R)‐5′,8a′‐Dimethyl‐4′‐((triethylsilyl)oxy)‐3′,4′,4a′,7′,8′,8a′‐hexahydro‐2′H‐spiro[[1,3]dioxolane‐2,1′‐naphthalen]‐6′‐yl)methanol (30). To a stirred solution of aldehyde 29 (108 mg, 0.28 mmol, 1 equiv.) in a mixture of THF/H2O (10 : 1, 3 mL) was added NaBH4 (32 mg, 0.9 mmol, 3 equiv.) portion‐wise at 0 °C over a period of 15 min. The reaction mixture was stirred for 2 h at 0 °C and subsequently quenched by addition of aqueous NH4Cl (5 mL). The layers were separated and the aqueous layer was extracted with ethyl acetate (3×10 mL). The combined organic layers were washed with saturated NaCl solution, dried over Na2SO4, filtered and concentrated in vacuo. The residue was purified by flash chromatography (petroleum ether/ethyl acetate, 3 : 1) to yield alcohol 30 (74 mg, 0.2 mmol, 68 %) as white solid. Rf=0.38 (petroleum ether/ethyl acetate, 3 : 1); 1H NMR (400 MHz, CDCl3): δ (ppm)=4.37 (d, J=2.6 Hz, 1H, 4‐H), 4.26–4.09 (m, 1H, CH 2OH), 4.04–3.85 (m, 5H, CH 2OH, OCH2CH2O), 2.26 (br s, 1H, 4a‐H), 2.21–2.12 (m, 3H, 2‐H, 7‐H), 1.89–1.69 (m, 5H, 3‐H, 5‐CH3), 1.60–1.41 (m, 3H, 2‐H, 8‐H), 1.15 (s, 3H, 8a‐CH3), 1.02–0.89 (m, 9H, Si(CH2CH3)3), 0.70–0.52 (m, 6H, Si(CH2CH3)3); 13C NMR (101 MHz, CDCl3): δ (ppm)=131.4 (C‐6), 130.0 (C‐5), 113.0 (C‐1), 67.4 (C‐4), 65.2 (OCH2CH2O), 65.0 (OCH2CH2O), 63.6 (CH2OH), 48.6 (C‐4a), 40.7 (C‐8a), 31.9 (C‐3), 28.2 (C‐8), 26.2 (C‐2), 25.4 (C‐7), 17.6 (8a‐CH3), 14.9 (5‐CH3), 7.0 (Si(CH2CH3)3), 5.4 (Si(CH2CH3)3); HRMS (ESI‐TOF): calcd. for C21H38O4Si 405.24316 [M+Na]+, found 405.24363.

((4′S,4a′R,8a′R)‐5′,8a′‐Dimethyl‐4′‐((triethylsilyl)oxy)‐3′,4′,4a′,7′,8′,8a′‐hexahydro‐2′H‐spiro[[1,3]dioxolane‐2,1′‐naphthalen]‐6′‐yl)methyl acetate (31). A solution of alcohol 30 (30 mg, 78.4 μmol) in a mixture of pyridine/Ac2O (1 : 2, 0.45 mL) was stirred under N2 for 2 h. After this time, the reaction was quenched by addition of water (1 mL). The layers were separated and the aqueous layer was extracted with CH2Cl2 (3×5 mL). The combined organic layers were dried over Na2SO4, filtered and concentrated in vacuo. The residue was purified by flash chromatography (petroleum ether/ethyl acetate, 10 : 1) to yield acetate 31 (23 mg, 54.2 μmol, 69 %) as a colorless oil. Rf=0.51 (petroleum ether/ethyl acetate, 5 : 1); 1H NMR (400 MHz, CDCl3): δ (ppm)=4.64 (d, J=11.8 Hz, 1H, CH2OAc), 4.50 (d, J=12.0 Hz, 1H, CH2OAc), 4.43–4.34 (m, 1H, 4‐H), 4.02–3.92 (m, 4H, OCH2CH2O), 2.29 (br s, 1H, 4a‐H), 2.21–2.09 (m, 3H, 2‐H, 7‐H), 2.05 (s, 3H, O2CCH3), 1.81–1.75 (m, 5H, 3‐H, 5‐CH3), 1.57 (td, J=12.3, 6.5 Hz, 1H, 8‐H), 1.51–1.45 (m, 1H, 2‐H), 1.44–1.39 (m, 1H, 8‐H) 1.14 (s, 3H, 8a‐CH3), 0.94 (t, J=8.0 Hz, 9H, Si(CH2CH3)3), 0.62–0.54 (qd, J=8.2, 3.2 Hz, 6H, Si(CH2CH3)3); 13C NMR (101 MHz, CDCl3): δ (ppm)=171.4 (O2 CCH3), 133.9 (C‐6), 125.4 (C‐5), 113.0 (C‐1), 67.4 (C‐4), 65.2 (OCH2CH2O), 65.0 (CH2OAc), 48.7 (C‐4a), 40.7 (C‐8a), 31.9 (C‐3), 28.1 (C‐8), 26.2 (C‐2), 25.7 (C‐7), 21.0 (O2CCH3), 17.6 (8a‐CH3), 15.3 (5‐CH3), 7.0 (Si(CH2CH3)3), 5.4 (Si(CH2CH3)3); HRMS (ESI‐TOF): calcd. for C23H40O5Si 447.25372 [M+Na]+, found 447.25366.

Ethyl 2‐((4′S,4a′R,5′R,8a′R)‐5′,8a′‐dimethyl‐6′‐methylene‐4′‐((triethylsilyl)oxy)octahydro‐2′H‐spiro[[1,3]dioxolane‐2,1′‐naphthalen]‐5′‐yl)acetate (33 a) and its epimer (33 b). To a stirred solution of alcohol 30 (500 mg, 1.31 mmol, 1 equiv.) in dry o‐DCB (4.8 mL) were sequentially added hydroquinone (72 mg, 0.65 mmol, 0.5 equiv.) and triethyl orthoacetate (4.8 mL, 26.1 mmol, 20 equiv.). The resulting mixture was heated to 180 °C for 17 h. Thereafter, the reaction mixture was allowed to reach room temperature and quenched by addition of water (10 mL). The layers were separated and the aqueous layer was extracted with ethyl acetate (3×10 mL). The combined organic layers were washed with aqueous NaHCO3 solution, saturated NaCl solution (15 mL each), dried over Na2SO4, filtered and concentrated in vacuo. The brown residue was purified by flash chromatography (petroleum ether/ethyl acetate, 30 : 1) to yield ester 33 a (312 mg, 0.69 mmol, 52 %) as a viscous oil and its C5 epimer 33 b (119 mg, 0.26 mmol, 20 %) as colorless needles.

33 a: Rf=0.51 (petroleum ether/ethyl acetate, 5 : 1); 1H NMR (400 MHz, CDCl3): δ (ppm)=4.79–4.58 (m, 2H, 6‐CH2), 4.39 (d, J=2.1 Hz, 1H, 4‐H), 4.10 (dq, J=7.1, 2.9 Hz, 2H, OCH 2CH3), 3.98–3.80 (m, 4H, OCH2CH2O), 2.66, 2.59 (2d, J=16.0 Hz, 2H, CH2CO2Et), 2.48 (td, J=13.9, 4.3 Hz, 1H, 7‐H), 2.32 (d, J=1.8 Hz, 1H, 4a‐H), 2.19–2.03 (m, 2H, 2‐H, 7‐H), 1.83–1.59 (m, 3H, 3‐H, 8‐Ha), 1.50 (s, 3H, 8a‐CH3), 1.46–1.32 (m, 2H, 2‐Hb, 8‐Hb) 1.36 (s, 3H, 5‐CH3), 1.24 (t, J=7.2 Hz, 3H, OCH2CH 3), 1.03–0.91 (m, 9H, Si(CH2CH3)3), 0.70–0.57 (m, 6H, Si(CH2CH3)3); 13C NMR (101 MHz, CDCl3): δ (ppm)=171.6 (CO2Et), 155.4 (C‐6), 113.2 (C‐1), 105.0 (6‐CH2), 68.6 (C‐4), 65.2 (OCH2CH2O), 64.9 (OCH2CH2O), 59.8 (OCH2CH3), 46.5 (C‐4a), 44.0 (C‐8a), 43.8 (C‐5), 43.1 (CH2CO2Et), 32.7 (C‐8), 32.4 (C‐3), 29.6 (C‐7), 26.6 (5‐CH3), 26.0 (C‐2), 19.7 (8a‐CH3), 14.2 (OCH2 CH3), 7.1 (Si(CH2CH3)3), 5.4 (Si(CH2CH3)3); HRMS (ESI‐TOF): calcd. for C25H44O5Si 475.28502 [M+Na]+, found 475.28491.

33 b: Rf=0.45 (petroleum ether/ethyl acetate, 5 : 1); 1H NMR (400 MHz, CDCl3): δ (ppm)=4.79–4.59 (m, 2H, 6‐CH2), 4.47 (d, J=2.3 Hz, 1H, 4‐H), 4.02 (dq, J=7.2, 3.9 Hz, 2H, OCH 2CH3), 3.96–3.80 (m, 4H, OCH2CH2O), 3.02, 2.87 (2d, J=13.1 Hz, 2H, CH2CO2Et), 2.59 (td, J=13.3, 3.9 Hz, 1H, 7‐H), 2.20–2.02 (m, 2H, 2‐H, 7‐H), 1.76–1.64 (m, 2H, 3‐H), 1.61–1.51 (m, 1H, 8‐H), 1.55 (s, 3H, 8a‐CH3), 1.47 (t, J=3.7 Hz, 1H, 8‐H), 1.43 (t, J=3.5 Hz, 1H, 2‐H), 1.39 (d, J=1.8 Hz, 1H, 4a‐H), 1.31 (s, 3H, 5‐CH3), 1.20 (t, J=7.2 Hz, 3H, OCH2CH 3), 1.03–0.95 (m, 9H, Si(CH2CH 3)3), 0.71–0.63 (m, 6H, Si(CH 2CH3)3); 13C NMR (101 MHz, CDCl3): δ (ppm)=172.6 (CO2Et), 154.4 (C‐6), 113.2 (C‐1), 106.9 (6‐CH2), 68.1 (C‐4), 65.2 (OCH2CH2O), 64.8 (OCH2CH2O), 59.6 (OCH2CH3), 54.2 (C‐4a), 44.1 (C‐8a), 42.8 (C‐5), 41.2 (CH2CO2Et), 34.0 (C‐8), 32.5 (C‐3), 29.6 (C‐7), 26.0 (C‐2), 24.7 (5‐CH3), 19.4 (8a‐CH3), 14.2 (OCH2 CH3), 7.0 (Si(CH2 CH3)3), 5.4 (Si(CH2CH3)3); HRMS (ESI‐TOF): calcd. for C25H44O5Si 475.28502 [M+Na]+, found 475.28479.

Ethyl 2‐((4′S,4a′R,5′R,8a′R)‐5′,8a′‐dimethyl‐6′‐oxo‐4′‐((triethylsilyl)oxy)octahydro‐2′H‐spiro[[1,3]dioxolane‐2,1′‐naphthalen]‐5′‐yl)acetate (36). Through a stirred solution of acetate 33 a (27.0 mg, 59.5 μmol, 1 equiv.) in a mixture of MeOH/CH2Cl2 (3 : 1) was bubbled ozone at −78 °C. After 1 h, the reaction mixture was purged with argon, quenched by addition of Me2S (0.2 mL) and subsequently concentrated in vacuo. The residue was purified by flash chromatography (petroleum ether/ethyl acetate, 5 : 1) to yield keto ester 36 (18.4 mg, 40.5 μmol, 68 %) as colorless oil. Rf=0.30 (petroleum ether/ethyl acetate, 5 : 1); 1H NMR (400 MHz, CDCl3): δ (ppm)=4.21 (d, J=1.8 Hz, 1H, 4‐H), 4.10 (dddd, J=25.5, 18.5, 7.2, 3.7 Hz, 2H, OCH 2CH3), 4.01–3.84 (m, 4H, OCH2CH2O), 2.99 (d, J=16.8 Hz, 1H, CH2CO2Et), 2.67–2.53 (m, 1H, 7‐H), 2.50–2.36 (m, 2H, CH2CO2Et, 7‐H), 2.33 (d, J=1.5 Hz, 1H, 4a‐H), 2.10 (m, 2H, 2‐H, 8‐H), 1.76–1.44 (m, 4H, 2‐H, 3‐H, 8‐H), 1.54 (s, 3H, 8a‐CH3), 1.34 (s, 3H, 5‐CH3), 1.24 (t, J=7.2 Hz, 3H, OCH2CH 3), 0.98 (t, J=8.3 Hz, 9H, Si(CH2CH 3)3), 0.65 (q, J=7.9 Hz, 6H, Si(CH 2CH3)3); 13C NMR (101 MHz, CDCl3): δ (ppm)=214.9 (C‐6), 171.7 (CO2Et), 112.8 (C‐1), 68.5 (C‐4), 65.2 (OC2CH2O), 65.1 (OCH2CH2O), 60.4 (OCH2CH3), 50.4 (C‐5), 47.0 (C‐4a), 43.0 (C‐8a), 42.7 (5‐CH2), 34.9 (C‐7), 32.7 (C‐3), 29.2 (C‐8), 26.0 (C‐2), 24.6 (5‐CH3), 19.4 (8a‐CH3), 14.2 (OCH2 CH3), 7.1 (Si(CH2 CH3)3), 5.4 (Si(CH2CH3)3); HRMS (ESI‐TOF): calcd. for C24H42O6Si 477.26429 [M+Na]+, found 477.26425.

Ethyl 2‐((1R,4aR,8S,8aS)−1,4a–dimethyl‐2,5‐dioxo‐8‐((triethylsilyl)oxy)decahydronaphthalen‐1‐yl)acetate (37). To a stirred solution of ester 36 (10.00 mg, 22.0 μmol, 1 equiv.) in EtOH (0.5 mL) was added 3 m HCl (0.17 mL). The resulting mixture was stirred at room temperature for 3 h and subsequently quenched by addition of saturated aqueous NaHCO3 solution (2 mL). The aqueous layer was extracted with Et2O (3 mL). The combined organic layers were washed with water and saturated NaCl solution (5 mL each), dried over anhydrous MgSO4, filtered and concentrated in vacuo. The yellow residue was purified by flash chromatography (petroleum ether/ethyl acetate, 6 : 1) to yield ester 37 (7.4 mg, 18.0 μmol, 82 %) as a colorless oil. Rf=0.71 (petroleum ether/ethyl acetate, 6 : 1); 1H NMR (400 MHz, CDCl3): δ (ppm)=4.36 (br s, 1H, 4‐H), 4.06 (dddd, J=17.5, 10.4, 7.2, 3.7 Hz, 2H, OCH 2CH3), 3.06 (dt, J=6.1, 14.3 Hz, 1H, 6‐H), 3.03 (d, J=17.2 Hz, 1H, CH2CO2Et), 2.67–2.54 (m, 1H, 3‐H), 2.51–2.45 (m, 2H, 3‐H), 2.46 (d, J=17.2 Hz, 1H, CH2CO2Et), 2.27 (d, J=1.1 Hz 1H, 8a‐H), 2.23 (ddd, J=14.2, 2.5, 1.7 Hz, 1H, 6‐H), 2.10 (ddd, J=14.1, 3.2, 2.6 Hz, 1H, 7‐H), 1.97 (dd, J=9.9, 3.8 Hz, 2H, 4‐H), 1.78 (ddt, J=4.3, 2.5, 1.8 Hz, 1H, 7‐H), 1.63 (s, 3H, 4a‐CH3), 1.41 (s, 3H, 1‐CH3), 1.22 (t, J=7.2 Hz, 3H, OCH2CH 3), 1.02 (t, J=8.3 Hz, 9H, Si(CH2CH 3)3), 0.70 (q, J=7.9 Hz, 6H, Si(CH 2CH3)3); 13C NMR (101 MHz, CDCl3): δ (ppm)=213.7 (C‐2), 213.3 (C‐5), 171.4 (CO2Et), 67.8 (C‐8), 60.5 (OCH2CH3), 50.6 (C‐1), 49.5 (C‐8a), 48.3 (C‐4a), 42.3 (CH2CO2Et), 35.2 (C‐7), 34.5 (C‐3), 32.9 (C‐6), 31.9 (C‐4), 24.3 (1‐CH3), 20.9 (4a‐CH3), 14.1 (OCH2 CH3), 7.1 (Si(CH2 CH3)3), 5.3 (Si(CH2CH3)3); HRMS (ESI‐TOF): calcd. for C22H38O5Si 433.23807 [M+Na]+, found 433.23783.

Conflict of interest

The authors declare no conflict of interest.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supporting Information

Click here for additional data file.