Use of a palladium(II)-catalyzed oxidative kinetic resolution in synthetic efforts toward bielschowskysin

Abstract

Progress toward the cyclobutane core of bielshowskysin is reported. The core was thought to arise from a cyclopropane intermediate via a furan-mediated cyclopropane fragmentation, followed by a 1,4-Michael addition. The synthesis of the cyclopropane intermediate utilizes a Suzuki coupling reaction, an esterification with 2-diazoacetoacetic acid, and a copper catalyzed cyclopropanation. An alcohol intermediate within the synthetic route was obtained in high enantiopurity via a highly selective palladium(II)-catalyzed oxidative kinetic resolution (OKR).

1. Introduction

In 2003, several novel natural products were isolated from the Caribbean gorgonian octocoral Pseudopterogorgia kallos. The structure of one of these compounds, bielschowskysin (1), eluded characterization until an X-ray crystal structure was obtained approximately one year later.¹ Bielschowskysin belongs to the pseudopterane sub-class of molecules, and is one of several structurally and biosynthetically related diterpenes (Figure 1).² Furthermore, bielschowskysin has been identified as a potent inhibitor of EKVX lung cancer cells (GI₅₀ = 10 nM).

Figure 1.

Bielschowskysin and related natural products.

A characteristic feature of these molecules is a cis-fused [5,5] oxa-bicycle, as well as a dihydrofuran unit that is sometimes disguised as a 1,4 diketone moiety.³ Bielschowskysin contains a polycyclic ring system bearing eleven stereocenters, one of which is a quaternary stereocenter contained in a hexasubstituted cyclobutane ring that is also fused to a substituted oxocane ring, resulting in a highly strained cage-like structure.⁴ A proposed biosynthesis by Trauner theorizes that the bielschowskysin core arises from a simpler cembranoid natural product, bipinnatin J (3, Figure 1), via olefin hydration, followed by a [2+2] cycloaddition to form the cyclobutane core.⁵

Several partial syntheses of bielschowskysin’s cyclobutane core have been reported;⁶ yet only one of these reports includes the stereogenic quaternary center contained in the cyclobutane ring.^6d Toward planning an initial synthetic route to bielschowskysin (3), a suitable model system was designed to investigate the key formation of the cyclobutane core as well as the quaternary stereocenter. Hence, our synthetic target was cyclobutane 4 (Scheme 1). We predicated our synthetic strategy upon the formation of cyclobutane 4 via a Lewis acid–mediated cyclopropane fragmentation of ketone 6, leading to intermediate 5, followed by a 1,4 Michael addition (Scheme 1). Ihara has shown that a domino sequence involving a Michael addition followed by an aldol reaction providing strained bicyclo[3.2.0] heptane systems is achievable.⁷ We envisioned cyclopropane 6 arising from α-diazo-β-ketoester 7 by a cyclopropanation reaction using either a copper or rhodium catalyst. In turn, α-diazo-β-ketoester 7 could be obtained from allylic alcohol 8 via esterification and diazotization. The latter compound, alcohol 8, can be prepared via Suzuki coupling between iodide 9 and boronic acid 10. Moreover, we postulated that allylic alcohol 8 could be used as a branch point for an enantioselective approach to this molecule. We anticipated that the stereocenter in alcohol 8 would provide a basis for the application of a substrate-focused palladium(II)-catalyzed oxidative kinetic resolution, developed independently by the Sigman and Stoltz laboratories.^8,9 Accomplishing the latter would provide access to an enantioselective synthesis of cyclopropane 6, as well as eventually bielschowskysin (1).

Scheme 1.

Retrosynthetic analysis of cyclobutane 4.

2. Results and Discussion

The synthesis began with furan 11,¹⁰ which underwent magnesium-halogen exchange using conditions developed by Knochel,¹¹ followed by a sequential quench of trimethyl borate and 1 M HCl to furnish boronic acid 10 in 85% yield (Scheme 2). Coupling of iodoenone 9 with boronic acid 10 under similar conditions reported by Johnson provided compound 12 in excellent yield.^12,13 At this stage, cleavage of the TBS group under acidic conditions provided allylic alcohol 8 in 98% yield. Exposure of allylic alcohol 8 to diketene and a catalytic amount of 4-(N,N-dimethylamino)pyridine (DMAP) afforded unstable β-ketoester 13. Unfortunately, immediate treatment of β-ketoester 13 with p-ABSA and Et₃N did not yield the desired diazo product; instead, a complex mixture of byproducts, likely a result of the nucleophilic β-ketoester undergoing a Michael addition into the enone, was obtained. After several attempts to form α-diazo-β-ketoester 7 from intermediate 13 under neutral diazo transfer conditions failed to provide the desired product, we decided to move in a different direction.

Scheme 2.

Attempted synthesis of α-diazo-β-ketoester 7.

At this stage, we began to postulate the feasibility of attaching the β-ketoester and the diazo group in a single operation. This thought, coupled with the observation that the conditions to form β-ketoester 13 (diketene/cat. DMAP/THF) are essentially neutral, led us to attempt a coupling between alcohol 8 and a novel reagent, 2-diazoacetoacetic acid (14, Scheme 3). This reagent would not only reduce the synthesis by one step, but also prevent any undesired Michael reactions from occurring by having the diazo moiety remove the nucleophilicity of the β-ketoester. Gratifyingly, the synthesis of 2-diazoacetoacetic acid (14) was accomplished in a straightforward manner,¹⁴ and it was found that using acid 14 with DCC and catalytic DMAP provided α-diazo-β-ketoester 7 in 75% yield.¹⁴

Scheme 3.

Syntheis of α-diazo-β-ketoester 7 using 2-diazoacetoacetic acid (14).

With α-diazo-β-ketoester 7 now synthesized, we then explored numerous Cu and Rh catalysts known to effect cyclopropanation, including Cu(TBSal)₂, Cu bronze, Rh₂(oct)₄, and Cu(acac)₂. Unfortunately, formation of cyclopropane 6 was never observed using these catalysts (Scheme 4).¹⁵ We hypothesized that the enone functionality contributed to a withdrawal of electron density from the olefin, and thus inhibited cyclopropanation.

Scheme 4.

Initial cyclopropanation results and synthesis of alcohol 15

As a result, we elected to modify the synthesis by reducing the ketone functionality in enone 12 (Scheme 4) to alcohol 15. We theorized that this would increase the electron density and the corresponding nucleophilicity of the olefin to facilitate cyclopropanation. Furthermore, we have previously shown 2-aryl cyclic allylic alcohols to be exceptional substrates in our palladium(II)-catalyzed oxidative kinetic resolution.^9d Therefore, diastereoselective 1,2-reduction of the enone functionality in compound 12 provided the necessary substrate (15) with which to explore the palladium(II)-catalyzed oxidative kinetic resolution and allow us to establish the absolute stereochemistry at C(3).

To that end, treatment of racemic alcohol 15 under conditions developed for the palladium(II)-catalyzed oxidative kinetic resolution afforded enantioenriched (3R,5S)-alcohol 15 in ≥ 95% ee and high selectivity (s = 23, Scheme 5), as well as (3S)-ketone 12 in 57% conversion after 24 hours.¹⁶ Furthermore, in a separate run it was found that after 18 hours, (3S)-ketone 12 was obtained in 40% yield and 93% ee. With these results, if (3S)-ketone 12 was subjected to the diastereoselective Luche reduction shown in Scheme 4, (3S,5R)-alcohol 15 would be obtained, and thus provide access to either enantiomer of alcohol 15 in high enantioselectivity (Scheme 5). Since the absolute configuration of bielschowskysin is unknown, accessing both enantiomers of alcohol 15 in this fashion is important, as it provides a route to potentially synthesize either enantiomer of the natural product.

Scheme 5.

Results of Pd(II)-catalyzed OKR.

Given the success of the palladium(II)-catalyzed oxidative kinetic resolution, our efforts shifted to the synthesis of cyclopropane 6, and subsequent investigations into the formation of cyclobutane 4. Beginning with furan 15, acetate protection of the alcohol and subsequent cleavage of the silyl ether unveiled allylic alcohol 16 in 85% overall yield (Scheme 6). The latter compound underwent coupling with 2-diazoacetoacetic acid (14) to provide α-diazo-β-ketoester 17 in 62% yield. Initially, attempts to form the cyclopropane ring using several rhodium and copper catalysts did not provide any of the desired product. Gratifyingly, formation of the cyclopropane was accomplished using 2 mol% Cu(TBSal)₂ in either toluene at reflux temperature or heating in a microwave using 1,2-dichloroethane as solvent to provide cyclopropane 18 in 50–55% yield.¹⁷ This result is consistent with our hypothesis that the olefin in enone 7 is simply too electron deficient to undergo cyclopropanation.

Scheme 6.

Formation of cyclopropane intermediate 18.

Although the synthesis of cyclopropane 18 was a significant accomplishment, this intermediate still needed to be elaborated to ketone 6. Initially, it was thought that cyclopropane 18 would undergo acetate cleavage using K₂CO₃/MeOH conditions to provide alcohol 19, followed by Dess-Martin periodinane oxidation to arrive at ketone 6 (Scheme 7, top). We hoped that exposing ketone 6 to a Lewis acid would lead to a furan-assisted fragmentation of the cyclopropane ring to provide zwitterionic intermediate 5, followed by a Michael addition to afford the highly functionalized cyclobutane 4. Thus, we treated acetate 18 with a sequence of reagents consisting of K₂CO₃/MeOH (to cleave the acetate group), DMP (to oxidize the resulting alcohol to the ketone), and La(OTf)₃ in MeOH at 55 °C (to effect the desired skeletal rearrangement). Through this sequence, a single major product was obtained in each step (first an alcohol (20), then a ketone (21), then a rearranged product (22)). Although this rearranged compound had some of the characteristics of the desired product, it clearly was not cyclobutane 4, as it was unmistakably an alcohol (determined by IR spectroscopy). Since we were unable to confirm the complete structure of the final product of this sequence via routine spectroscopic methods (i.e., IR and NMR), we chose to esterify compound 22 to provide ester 23, which enabled us to grow crystals suitable for analysis by X-ray crystallography.¹⁸ To our surprise, the X-ray data provided an unexpected structure from the sequence, β-ketoester 22. Upon examination of the connectivity of β-ketoester 22, it was apparent that the oxidation states at C(3) and C(5) were opposite to those of ketone 6. In light of the data, we suspected that an undesired translactonization had occurred during acetate removal, and that the position of the alcohol in intermediate 20 was at C(3) rather than C(5).

Scheme 7.

Synthesis of unexpected β-ketoester 22.

To confirm this undesired rearrangement, we partially incorporated deuterium at C(5) using a 1:1 mixture of NaBH₄:NaBD₄ in the Luche reduction that produced alcohol d-15, allowing us to follow the chemical shifts of the protons on both C(3) and C(5) during the course of acetate removal (Scheme 8). As anticipated, the C(3) proton in intermediate d-20 exhibited a slight upfield shift, while the C(5) proton underwent a downfield shift in the ¹H NMR. Interestingly, cyclopropyl proton H_A exhibits a change in resonance by 0.50 ppm that we attribute to conformational changes that result from transesterification. These results helped verify that the translactonization reaction does in fact occur during acetate removal.

Scheme 8.

Verifying the translactonization by partial deuterium labeling

In an effort to circumvent the translactonization reaction, we synthesized a variety of donor-acceptor cyclopropanes similar to cyclopropane 18, differing only in the hydroxyl protecting group at C(5) (Figure 2).^19,20 We postulated that these protecting groups may be removed under conditions that suppress the formation of alcohol 20. Unfortunately, all attempts to deprotect compounds 24–28 in Figure 2 led solely to undesired alcohol 20. This result led us to conclude that the undesired lactone (20) is likely the thermodynamic product.²¹

Figure 2.

Donor-acceptor cyclopropanes with various protecting groups for the C(5) alcohol.

Further analysis can be made from the Lewis acid–mediated fragmentation reaction of cyclopropyl ketone 21 to provide unexpected alcohol 22. Beginning with ketone 21 (Scheme 9), it is thought that Lewis acidic activation weakens the cyclopropane ring to allow the furan ring to assist in fragmentation to form an enolate and an extended oxocarbenium ion in intermediate 29, both of which are quenched respectively by the net addition of methanol to provide ketone 30. Thus, fragmentation of the cyclopropane ring occurred as anticipated; however, the addition of methanol preserved the aromaticity in the furan ring rather than undergoing a 1,2-addition into the oxocarbenium ion. Rearomatization is likely the driving force for the observed chemoselectivity in the methanol addition. Furthermore, under these conditions, an additional molecule of methanol can add into ketone 30 from the less-hindered α-face to form a hemiketal at C(3). Intermediate 31 can now undergo transesterification to produce alcohol 22. Notably, formation of the hemiketal at C(3) can occur at any point in the reaction; however, transesterification to generate the fused lactone is not favorable until the cyclopropane ring undergoes fragmentation.

Scheme 9.

Proposed mechanism of the formation of bicycle 22.

To examine this hypothesis, we subjected alcohol 20 to identical Lewis acidic conditions that resulted in the formation of alcohol 32 in 65% yield as a single diastereomer (Scheme 10). Alcohol 32 can undergo subsequent functionalization to provide compound 33, allowing for suitable X-ray analysis.²² We initially believed that the hemiketal in intermediate 31 might force the alcohol into a pseudoaxial position and promote transesterification. However, by forming alcohol 32 without the ability for ketalization, we have shown that transesterification is facile once the cyclopropane ring has fragmented, regardless of the C(3) oxidation state.

Scheme 10.

Formation and confirmation of alcohol 32.

3. Conclusion

Although the cyclopropane route did not achieve its ultimate goal in synthesizing the cyclobutane ring of bielschowskysin, several interesting aspects of our research can be extracted. First, we have successfully implemented the palladium(II)-catalyzed oxidative kinetic resolution early in the synthesis to provide alcohol 15 in ≥95% ee and 43% yield. This method allows for rapid access to either enantiomer of allylic alcohol 15, an important intermediate in synthesizing advanced compounds within our synthetic efforts. Second, we synthesized and used 2-diazoacetoacetic acid (14) to provide α-diazo-β-ketoester 7, which was unattainable through known methods. Finally, we discovered two interesting transesterification processes that led to a highly functionalized, fused bicylic scaffold. By studying these separate transesterifications and understanding why they occur, we have gained an understanding of the reactivity of these molecules, and anticipate that continued efforts will pave the way for the formation of the cyclobutane core of bielschowskysin. Current efforts are focused on manipulating intermediate 32 into cyclobutane 4, as well as synthesizing the macrocyclic portion of bielschowskysin.

4. Experimental section

Unless otherwise stated, reactions were performed at ambient temperature (typically 20–22 °C) in flame-dried glassware under an argon or nitrogen atmosphere using dry solvents. Solvents were dried by passage through an activated alumina column under argon. Et₃N, iPr₂NH, iPr₂NEt, and pyridine were freshly distilled from CaH₂. All other commercially obtained reagents were used as received. Reaction temperatures were controlled by an IKAmag temperature modulator. All microwave reactions were performed with a Biotage Initiator 2.5 microwave. Thin-layer chromatography (TLC) was performed using E. Merck silica gel 60 F254 precoated plates (0.25 mm) and visualized by UV fluorescence quenching, anisaldehyde, KMnO₄, or CAM staining. ICN silica gel (particle size 0.032–0.063 mm) was used for flash chromatography. Optical rotations were measured with a Jasco P-1010 polarimeter at 589 nm. ¹H and ¹³C NMR spectra were recorded on a Varian Mercury 300 (at 300 MHz and 75 MHz respectively), or a Varian Inova 500 (at 500 MHz and 125 MHz respectively) and are reported relative to solvent for ¹H NMR (CHCl₃ = 7.27 ppm, C₆H₆ = 7.16 ppm, CH₂Cl₂ = 5.30 ppm, DMSO = 2.51 ppm) and ¹³C NMR (CDCl₃ = 77.0 ppm, C₆D₆ = 128.4 ppm). Data for ¹H NMR spectra are reported as follows: chemical shift (δ ppm) (multiplicitiy, coupling constant (Hz), integration). Multiplicities are reported as follows: s = singlet, d = doublet, t = triplet, q = quartet, dt = doublet of triplets, td = triplet of doublets, dd = doublet of doublets, m = multiplet, comp. m = complex multiplet, app. = apparent, bs = broad singlet. IR spectra were recorded on a Perkin Elmer BX-11 FT-IR spectrometer and are reported in frequency of absorption (cm⁻¹). High resolution mass spectra were obtained from the California Institute of Technology Mass Spectral Facility. Crystallographic analyses were performed at the California Institute of Technology Beckman Institute X-Ray Crystallography Laboratory. Crystallographic data have been deposited at the CCDC, 12 Union Road, Cambridge CB2 1EZ, UK and copies can be obtained on request, free of charge, from the CCDC by quoting the publication citation and the deposition number (see Appendix 3 for deposition numbers).

methyl-5-(3-((tert-butyldimethylsilyl)oxy)-5-oxocyclopent-1-en-1-yl)-3-methylfuran-2-carboxylate (12)

To a solution of furan 11 (609 mg, 2.78 mmol, 1.00 equiv) in THF (15 mL) at –35 °C, was added a 1.84 M solution (THF) of iPrMgCl (1.96 mL, 3.61 mmol, 1.30 equiv). The reaction was stirred for 45 minutes and the temperature was maintained between −40 and −30 °C. At this stage, trimethylborate (1.24 mL, 11.1 mmol, 4.00 equiv) was added rapidly at −30 °C, and the reaction was allowed to warm to ambient temperature and stir overnight. The reaction was cooled to 0 °C, 1 M HCl (aq.) was added, and the mixture stirred for 30 minutes. The reaction was diluted with EtOAc and H₂O, and the layers were separated. The aqueous layer was then extracted 2× with EtOAc. The organic layers were combined, transferred into an erlenmeyer flask with stirring, and saturated K₂CO₃ (aq.) was carefully added until the aqueous layer remained basic. The basic aqueous layer was then acidified using 1 M HCl (aq.), and then extracted 5× with EtOAc. The latter batch of EtOAc was washed with brine, dried with Na₂SO₄, filtered, and concentrated in vacuo to afford boronic acid 10 (375 mg, 73%) as a light-brown solid. This compound was used directly in the following Suzuki coupling reaction.

To a solution of vinyl iodide 9 (87.0 mg, 0.260 mmol, 1.00 equiv), boronic acid 10 (71.0 mg, 0.390 mmol, 1.50 equiv), Ag₂O (180 mg, 0.770 mmol, 3.00 equiv), and Ph₃As (8.00 mg, 0.0260 mmol, 0.100 equiv) in THF (1.30 mL) was added PdCl₂(PhCN)₂ (5.00 mg, 0.0130 mmol, 0.0500 equiv) and H₂O (50.0 µL). The reaction was stirred until complete consumption of vinyl iodide 9, as indicated by TLC (ca. 1.5 h), and filtered through a pad of Celite. The filtrate was concentrated in vacuo, and the residue was purified by flash chromatography using 3:97→5:95 EtOAc:hexanes to afford furan 12 as a white solid (74 mg, 80%). R_F 0.15 (90:10 hexanes:EtOAc, UV); ¹H NMR (500 MHz, CDCl₃) δ 7.74 (d, J = 2.7 Hz, 1H), 7.03 (s, 1H), 5.03 (m, 1H), 3.90 (s, 3H), 2.88 (dd, J = 5.9, 18.3 Hz, 1H), 2.42 (dd, J = 2.2, 18.6 Hz, 1H), 2.34 (s, 3H), 0.91 (s, 9H), 0.15 (s, 3H), 0.14 (s, 3H); ¹³C NMR (125 MHz, CDCl₃) δ 201.8, 160.1, 156.0, 147.4, 140.0, 133.7, 132.6, 116.1, 68.9, 51.8, 45.9, 25.8, 18.3, 11.5, - 4.8; IR (film) 3133, 2949, 2928, 2891, 2857, 1721, 1703, 1594, 1509, 1448 cm⁻¹; HRMS (FAB⁺) calc’d for [C₁₈H₂₆SiO₅]⁺: m/z 350.1550, found 350.1542.

methyl-5-(3-hydroxy-5-oxocyclopent-1-en-1-yl)-3-methylfuran-2-carboxylate (8)

To a solution of acetyl chloride (50.5 µL, 0.710 mmol, 5.00 equiv) in MeOH (570 µL) at 0 °C, was added enone 12 (52.5 mg, 0.150 mmol, 1.00 equiv) in one portion. Enone 12 dissolved slowly over 20–30 minutes at 0 °C, and once completely in solution, the reaction stirred for 45 minutes at 0 °C. The reaction was then diluted with brine and EtOAc, and these layers were separated. The aqueous layer was extracted 2× with EtOAc. The organic layers were combined, washed with brine, dried with Na₂SO₄, filtered, and concentrated in vacuo. This semi-solid was diluted with toluene, and then concentrated under vacuum (repeated 3×). Upon final concentration in vacuo, alcohol 8 was obtained as a white solid (35.3 mg, 99%). ¹H NMR (500 MHz, CDCl₃) δ 7.86 (d, J = 2.9 Hz, 1H), 7.02 (s, 1H), 5.21 (br, 1H), 3.90 (s, 3H), 2.96 (dd, J = 6.1, 18.8 Hz, 1H), 2.71 (d, J = 5.40 Hz, 1H), 2.49 (dd, J = 2.2, 18.8 Hz, 1H), 2.34 (s, 3H); ¹³C NMR (125 MHz, CDCl₃) δ 201.8, 160.0, 154.9, 147.1, 139.9, 134.1, 132.4, 116.1, 68.3, 51.7, 45.2, 11.6; IR (film) 3447, 2920, 2850, 1716, 1506, 1440, 1405, 1295, 1167, 1102 cm⁻¹; HRMS (FAB⁺) calc’d for [C₁₂H₁₃O₅]⁺: m/z 237.0763, found 237.0763.

methyl-5-(3-((2-diazo-3-oxobutanoyl)oxy)-5-oxocyclopent-1-en-1-yl)-3-methylfuran-2-carboxylate (7)

To a solution of alcohol 8 (71.0 mg, 0.300 mmol, 1.00 equiv), acid 14 (92.1 mg, 0.720 mmol, 2.40 equiv), and DMAP (3.67 mg, 0.0300 mmol, 0.100 equiv) in CH₂Cl₂ (2.00 mL) at 23 °C, was added DCC (206 mg, 1.00 mmol, 2.0 equiv) in one portion. After the addition of DCC, the reaction was heated to 40 °C for 90 minutes. The reaction was allowed to cool to ambient temperature, filtered through a plug of Celite (Et₂O eluent), and the resulting filtrate was partitioned between H₂O and Et₂O. The aqueous layer was extracted 2× with Et₂O. The organic extracts were combined, washed once with H₂O, and then once with brine. The organic layer was dried over MgSO₄, filtered, and concentrated in vacuo. The product was purified by flash chromatography using 25:75 EtOAc:Hexanes to afford compound 7 as a light-yellow solid (78 mg, 75% yield). ¹H NMR (500 MHz, C₆D₆) δ 7.30 (s, 1H), 7.23 (s, 1H), 5.24 (m, 1H), 3.46 (s, 3H), 2.33 (dd, J = 6.6, 18.6 Hz, 1H), 2.24 (s, 3H), 2.18 (s, 3H), 1.90 (d, J = 18.6 Hz, 1H); ¹³C NMR (125 MHz, C₆D₆) δ 199.3, 188.6, 160.8, 159.9, 149.1, 147.3, 141.4, 136.4, 132.7, 117.5, 71.2, 51.5, 42.0, 28.4, 11.8; IR (film) 2954, 2145, 1715, 1660, 1366, 1312, 1155, 1062 cm⁻¹; HRMS (FAB⁺) calc’d for [C₁₆H₁₄N₂O₇]⁺: m/z 346.0801; found, 346.0817.

methyl-5-(3-((tert-butyldimethylsilyl)oxy)-5-hydroxycyclopent-1-en-1-yl)-3-methylfuran-2-carboxylate (15)

To a solution of furan 12 (100 mg, 0.290 mmol, 1.00 equiv) and CeCl₃•7H₂O (430 mg, 1.10 mmol, 4.00 equiv) in MeOH:CH₂Cl₂ (4.90 mL, 1:1) at −25 °C, was added sodium borohydride (43.0 mg, 1.10 mmol, 4.00 equiv) portionwise while maintaining the reaction between −25 and −20 °C. Upon consumption of the ketone as visualized by TLC (ca. 30 min.), the reaction was diluted with EtOAc and H₂O. The solution was allowed to warm to ambient temperature and the layers were separated. The aqueous layer was extracted 3× with EtOAc, and the organic layers were combined and washed with brine. The EtOAc layer was dried with Na₂SO₄, filtered, and concentrated in vacuo to provide allylic alcohol 15 as a yellow oil (96 mg, 95%). R_F = 0.20 (20:80 EtOAc:hexanes, UV); ¹H NMR (500 MHz, CDCl₃) δ 6.58 (s, 1H), 6.39 (s, 1H), 4.84 (m, 1H), 4.79 (m, 1H), 3.91 (s, 3H), 2.76 (dt, J = 6.9, 14.2 Hz, 1H), 2.36 (s, 3H), 1.98 (d, J = 9.3 Hz, 1H), 1.77 (d, J = 14.2 Hz, 1H), 0.92 (s, 9H); ¹³C NMR (75 MHz, CDCl₃) δ 160.0, 151.7, 139.5, 136.2, 133.0, 132.6, 113.7, 74.8, 74.2, 51.5, 44.7, 25.8, 18.1, 11.6, −4.6, −4.7; IR (film) 3420, 2954, 2930, 2857, 1711, 1440, 1298, 1101 cm⁻¹; HRMS (FAB⁺) calc'd for [C₁₈H₂₇O₅Si]⁺: m/z 351.1628; found 351.1632.

methyl-5-((3R,5S)-3-((tert-butyldimethylsilyl)oxy)-5-hydroxycyclopent-1-en-1-yl)-3-methylfuran-2-carboxylate (3R,5S-15)

A 10 mL Schlenk tube equipped with a magnetic stir bar was charged with powdered molecular sieves (3ÅMS, 150 mg) and flame-dried under vacuum. After cooling under dry N₂, Pd(nbd)Cl₂ (0.380 mg, 1.42 µmol, 0.0500 equiv) was added, followed by toluene (285 µL). To this yellow suspension was added (−)-sparteine (1.33 mg, 5.67 µmol, 0.200 equiv), and the top of the Schlenk tube was then fitted with an O₂ balloon. The reaction was purged under vacuum and replaced with O₂, and this process was repeated 4×'s. The light yellow suspension was stirred at ambient temperature for 10 minutes, then heated to 80 °C for an additional 10 minutes. The reaction mixture darkened to an orange suspension. Upon completion of the 10 minutes, racemic alcohol 15 (10.0 mg, 28.4 µmol, 1.00 equiv, in 100 µL of PhMe) was added to the reaction mixture. The suspension was then heated to 80 °C for 24 hours. The reaction was allowed to cool to ambient temperature, and was then filtered through a small pad of Celite using toluene as eluent. The filtrate was concentrated in vacuo, and the residue was purified by flash chromatography using 10:90→15:85 EtOAc:hexanes to afford (3R,5S)-alcohol 15 as an oil (4.3 mg, 43% yield, 95% ee). [α]_D^24.98 +26.4 (c = 0.15, EtOAc).

methyl-5-(5-acetoxy-3-hydroxycyclopent-1-en-1-yl)-3-methylfuran-2-carboxylate (16)

To a solution of alcohol 15 (96.0 mg, 0.270 mmol, 1.00 equiv) in pyridine (2.70 mL, 0.100 M) was added acetic anhydride (0.270 mL, 2.50 mmol, 9.00 equiv) and DMAP (3.30 mg, 0.0270, 0.100 equiv) at ambient temperature. The reaction was stirred until consumption of allylic alcohol 15 as monitored by TLC (ca. 30 min). The solution was cooled to 0 °C and diluted with EtOAc:brine (1:1). The layers were separated, and the aqueous solution was extracted 2× with EtOAc. The combined organic layer was successively washed with sat. CuSO₄ (aq.), H₂O, sat. NaHCO₃ (aq.), and brine. The organic layer was then dried with Na₂SO₄, filtered, and concentrated in vacuo to yield an allylic acetate that was taken directly on to the next reaction (101 mg, 95%). ¹H NMR (300 MHz, CDCl₃) δ 6.49 (dd, J = 0.9, 2.1 Hz, 1H), 6.26 (s, 1H), 5.85 (dd, J = 4.2, 7.5 Hz, 1H), 4.84 (m, 1H), 3.90 (s, 3H), 2.96 (dt, J = 7.5, 14.4 Hz, 1H), 2.33 (s, 3H), 2.10 (s, 3H), 1.73 (m, 1H), 0.91 (s, 9H), 0.11 (s, 3H), 0.11 (s, 3H). A solution of TBAF (0.290 mL of 1M solution in THF, 0.289 mmol, 1.20 equiv) was added to the crude allylic acetate (95 mg, 0.241 mmol, 1.00 equiv) in THF (4.8 mL) at 0 °C. The reaction was allowed to warm to ambient temperature and monitored by TLC. Once complete, the reaction was diluted with H₂O, and the reaction was extracted 3× with EtOAc. The combined organic extracts were washed with brine, dried with Na₂SO₄, filtered, and concentrated in vacuo. The crude solid was purified by flash chromatography using 50:50, EtOAc:hexanes to provide allylic alcohol 16 as a white solid (64 mg, 85% from alcohol 15).: R_F 0.2 (40:60, EtOAc:hexanes, UV); ¹H NMR (500 MHz, CDCl₃) δ 6.60 (d, J = 2.4 Hz, 1H), 6.29 (s, 1H), 5.88 (dd, J = 2.9, 7.3 Hz, 1H), 4.83 (m, 1H), 3.88 (s, ,3H), 2.88 (app dt, J = 14.7, 7.1, 1H), 2.32 (s, 3H), 2.07 (s, 3H), 1.81 (app dt, J = 12.2, 2.9 Hz, 1H); ¹³C NMR (125 MHz, CDCl₃) δ 170.8, 160.0, 150.6, 140.1, 134.6, 133.6, 132.6, 113.8, 75.7, 74.4, 51.7, 41.5, 21.3, 11.7; IR (film) 3424, 2953, 1712, 1598, 1511, 1440, 1374, 1297, 1240, 1197, 1104, 1035 cm⁻¹; HRMS (EI⁺) calc’d for [C₁₄H₁₆O₆]⁺: m/z 280.2804, found 280.2800.

methyl-5-(5-acetoxy-3-((2-diazo-3-oxobutanoyl)oxy)cyclopent-1-en-1-yl)-3-methylfuran-2-carboxylate (17)

To a solution of allylic alcohol 16 (15.0 mg, 0.0540 mmol, 1.00 equiv), 2-diazoacetoacetic acid (14, 16.7 mg, 0.130 mmol, 2.40 equiv), and DMAP (0.660 mg, 0.00540 mmol, 0.100 equiv) in CH₂Cl₂ (360 µL) at 23 °C, was added in one portion DCC (22.3 mg, 0.108 mmol, 2.00 equiv). The reaction was heated to 40 °C, monitored by TLC, and the reaction was complete in 60–90 minutes. The reaction was filtered through a pad of Celite (Et₂O eluent), and the resulting filtrate was partitioned between H₂O and Et₂O. The aqueous layer was extracted 2× with Et₂O. The organic extracts were combined, washed with NaHCO₃ (sat’d), and then with brine. The organic layer was dried over Na₂SO₄, concentrated in vacuo, and the resulting crude oil was purified by silica gel chromatography (80:20→75:25, hexanes:EtOAc) to yield ester 17 as a light-yellow solid (338 mg, 65% average). ¹H NMR (500 MHz, CDCl₃) δ 6.60 (d, J = 2.2 Hz, 1H), 6.30 (s, 1H), 6.19 (m, 1H), 6.07 (m, 1H), 3.88 (s, 3H), 2.47 (s, 3H), 2.44 (m, 2H), 2.33 (s, 3H), 2.08 (s, 3H); ¹³C NMR (125 MHz, CDCl₃) δ 190.0, 170.8, 161.2, 159.8, 149.9, 140.5, 136.2, 132.5, 130.5, 114.7, 79.0, 76.2, 51.8, 39.1, 28.4, 21.2, 11.7; IR (film) 2142, 1713, 1657, 1295, 1237 cm⁻¹; HRMS (EI⁺) calc’d for [C₁₈H₁₈N₂O₈]⁺: m/z 390.1063, found 390.1061.

methyl-5-(3-acetoxy-2a-acetyl-2-oxohexahydro-2H-1-oxacyclopropa[cd]pentalen-2b-yl)-3-methylfuran-2-carboxylate (18)

microwave conditions

The microwave was pre-warmed by undergoing an actual run with just DCE in the reaction tube. A solution of compound 17 (45.0 mg, 0.120 mmol, 1.00 equiv) and Cu(TBSal)₂ (5.00 mg, 0.0120 mmol, 0.100 equiv) in DCE (3.00 mL) was heated for 20 min at 105 °C in the microwave. The following parameters for the microwave are as follows: Power = 150 W, Temperature = 105 °C, Pressure = 200 atm, Ramp = 1 minute, Run Time = 20 minutes. The reaction was concentrated in vacuo, and the resulting crude oil was purified by silica gel chromatography (40:60 EtOAc:hexanes) to yield cyclopropane 18 as a white solid (21 mg, 55%). ¹H NMR (500 MHz, C₆D₆) δ 6.20 (s, 1H), 5.47 (dd, J = 1.7, 7.3 Hz, 1H), 4.01 (m, 1H), 3.53 (d, J = 4.6 Hz, 1H), 3.39 (s, 3H), 2.39 (s, 3H), 2.13 (s, 3H), 1.73 (d, J = 14.4 Hz, 1H), 1.71 (s, 3H), 1.29 (ddd, J = 3.4, 7.6, 14.6 Hz, 1H); ¹³C NMR (125 MHz, C₆D₆) δ 195.7, 170.8, 170.5, 160.0, 150.8, 141.5, 132.8, 116.2, 81.6, 78.4, 52.8, 51.9, 50.4, 46.0, 45.8, 29.7, 20.8, 12.0; IR (film) 2954, 1770, 1747, 1714, 1230 cm⁻¹; HRMS (EI⁺) calc’d for [C₁₈H₁₉O₈]⁺: m/z 363.1080, found 363.1085.

Thermal conditions

A solution of Cu(TBSal)₂ (1.30 mg, 0.00300 mmol, 0.0200 equiv) in toluene (2.40 mL) was heated under Ar to 108 °C. A warm solution of diazo 17 (62.0 mg, 0.150 mmol, 1.00 equiv) in toluene (0.400 mL) was added dropwise over 6 minutes. The reaction was heated to 112 °C, and monitored by TLC (40:60 EtOAc:hexanes). After 100 minutes, the reaction was complete, and allowed to cool to ambient temperature. The solvent was removed in vacuo, and the dark residue was purified by flash chromatography (40:60 EtOAc:hexanes) to provide cyclopropane 18 (35 mg, 56%).

methyl-5-(2a-acetyl-3-hydroxy-2-oxohexahydro-2H-1-oxacyclopropa[cd]pentalen-2a1-yl)-3-methylfuran-2-carboxylate (20)

K₂CO₃ (32.0 mg, 0.240 mmol, 2.00 equiv) was added to a solution of cyclopropane 18 (43.0 mg, 0.120 mmol, 1.00 equiv) in MeOH (2.30 mL) at 0 °C. The reaction was stirred at 0 °C and monitored by TLC. Once complete, the reaction was diluted with saturated NH₄Cl (aq.) and EtOAc, the layers were then separated, and the aqueous layer was further extracted 2× with EtOAc. The organic layers were combined, washed with H₂O and brine, dried with Na₂SO₄, filtered, and concentrated in vacuo to afford a crude oil. The latter oil was purified by silica gel chromatography (50:50 EtOAc:Hexanes) to provide alcohol 20 as a glue (30 mg, 80%). ¹H NMR (500 MHz, C₆D₆) δ 5.66 (s, 1H), 4.69 (m, 1H), 4.03 (app. q, J = 1.7, 5.6 Hz, 1H), 3.42 (s, 3H), 3.34 (d, J = 5.6 Hz, 1H), 2.28 (s, 3H), 2.12 (s, 3H), 1.53 (d, J = 14.4 Hz, 1H), 1.13 (ddd, J = 3.4, 7.3, 14.4 Hz, 1H); ¹³C NMR (125 MHz, C₆D₆) δ 209.4, 196.2, 171.1, 159.6, 148.7, 141.6, 132.3, 114.6, 84.0, 71.5, 58.4, 51.4, 46.8, 45.9, 29.8, 11.8; IR (film) 3469, 1770, 1706, 1299, 1081 cm⁻¹; HRMS (EI⁺) calc’d for [C₁₆H₁₆O₇Na]⁺: m/z 343.0794, found 343.0811.

methyl-5-(2a-acetyl-2,3-dioxohexahydro-2H-1-oxacyclopropa[cd]pentalen-2a¹-yl)-3-methylfuran-2-carboxylate (21)

To a solution of alcohol 20 (62.0 mg, 0.190 mmol, 1.00 equiv) in CH₂Cl₂ (3.90 mL) was added Dess-Martin periodinane (160 mg, 0.390 mmol, 2.00 equiv), and the resulting slurry was stirred until consumption of starting material (ca. 1 h). The reaction was diluted with EtOAc and a 1:1 ratio of sat’d Na₂S₂O₄ (aq.): sat’d NaHCO₃ (aq.), and stirred vigorously for 5 minutes. The layers were separated and the organic layer was washed with brine, dried with Na₂SO₄, filtered, and concentrated in vacuo to yield crude ketone 21 as a colorless oil (60 mg, 97%). Ketone 21 was carried on crude as attempts toward further purification resulted in decomposition. ¹H NMR (500 MHz, C₆D₆) δ 5.53 (s, 1H), 4.59 (dd, J = 1.2, 4.4 Hz, 1H), 3.49 (d, J = 1.2 Hz, 1H), 3.43 (s, 3H), 2.07 (s, 3H), 2.00 (s, 3H), 1.90 (d, J = 17.6 Hz, 1H), 1.66 (dd, J = 4.6, 17.6 Hz, 1H); ¹³C NMR (125 MHz, C₆D₆) δ 202.0, 193.2, 168.7, 159.6, 146.2, 140.0, 142.4, 132.3, 115.9, 78.4, 54.4, 52.6, 51.6, 49.9, 45.3, 29.5, 11.8; IR (film) 2924, 2854, 1180, 1750, 1715, 1559, 1438, 1294, 1102 cm⁻¹; HRMS (EI⁺) calc’d for [C₁₆H₁₄O₇]⁺: m/z 318.0740, found 318.0732.

methyl-5-(3-acetyl-5-hydroxy-4,6a-dimethoxy-2-oxohexahydro-2H-cyclopenta[b]furan-4-yl)-3-methylfuran-2-carboxylate (22)

Crude ketone 21 (11.0 mg, 0.0340 mmol, 1.00 equiv) and La(OTf)₃ (20.0 mg, 0.0340 mmol, 1.00 equiv) were dissolved in MeOH (0.670 mL) at ambient temperature, then heated to 55 °C and the reaction was stirred until the majority of starting material appeared consumed by TLC. The reaction was diluted with EtOAc and H₂O and the aqueous layer was extracted 3× with EtOAc. The combined organic layers were washed with brine, dried with Na₂SO₄, filtered, and concentrated in vacuo. The residue was purified by flash chromatography using 66:34 EtOAc:Hexanes to afford alcohol 22 as an oil (8.0 mg, 55%).; ¹H NMR (500 MHz, C₆D₆) δ 5.90 (s, 1H), 4.26 (dd, J = 1.7, 5.6 Hz, 1H), 4.24 (m, 1H), 4.04 (d, J = 5.8 Hz, 1H), 3.38 (s, 3H), 3.20 (s, 3H), 2.80 (s, 3H), 2.56 (dd, J = 4.2, 14.2 Hz, 1H), 2.26 (s, 3H), 2.20 (d, J = 14.2 Hz, 1H), 2.15 (s, 3H), 1.84 (d, J = 2.7 Hz, 1H); ¹³C NMR (125 MHz, C₆D₆) δ 199.3, 170.6, 159.8, 154.1, 140.9, 132.0, 119.4, 115.9, 87.4, 76.9, 60.5, 52.0, 51.5, 51.4, 48.0, 43.7, 29.5, 11.8; IR (film) 3465, 2955, 1771, 1722, 1441, 1297, 1197, 1102 cm⁻¹; HRMS (EI⁺) calc’d for [C₁₈H₂₂O₉]⁺: m/z 382.1264, found 382.1259.

methyl-5-((E)-5-((4-bromobenzoyl)oxy)-3-(1-((4-bromobenzoyl)oxy)ethylidene)-4,6a-dimethoxy-2-oxohexahydro-2H-cyclopenta[b]furan-4-yl)-3-methylfuran-2-carboxylate (23)

Triethylamine (19.0 µL, 0.140 mmol, 2.40 equiv) was added at ambient temperature to a solution of alcohol 22 (22.0 mg, 0.0580 mmol, 1.00 equiv) in CH₂Cl₂ (310 µL). DMAP (0.700 mg, 0.00580 mmol, 0.100 equiv) and p-bromobenzoyl bromide (28.0 mg, 0.130 mmol, 2.20 equiv) were then added, and the reaction was stirred for 1 hour at ambient temperature. The reaction was diluted with EtOAc and H₂O, the layers were separated, and the aqueous layer was extracted once with EtOAc. The EtOAc layers were combined, washed with saturated NaHCO₃ (aq.), washed with brine, dried with Na₂SO₄, filtered, and concentrated in vacuo. Crystals suitable for X-ray analysis were obtained by crystallization of crude furan 23 (30 mg, yield 70%) from EtOAc:heptane. ¹H NMR (500 MHz, C₆D₆) δ 7.97 (m, 2H), 7.39 (m, 2H), 7.25 (m, 2H), 7.10 (m, 2H), 5.99 (s, 1H), 5.73 (m, 1H), 3.93 (s, 1H), 3.24 (s, 3H), 3.13 (s, 3H), 2.88 (s, 3H), 2.54 (s, 3H), 2.49 (m, 2H), 2.15 (s, 3H); ¹³C NMR (125 MHz, C₆D₆) δ 168.1, 164.6, 161.9, 161.6, 159.2, 152.9, 141.0, 132.1, 132.0, 132.0, 131.8, 131.4, 129.3, 128.8, 128.6, 127.7, 114.9, 114.7, 114.2, 87.4, 78.1, 55.5, 52.3, 51.6, 50.1, 39.3, 17.4, 11.4; IR (film) 2948, 1762, 1725, 1589, 1227, 1101, 1070, 1010 cm⁻¹; HRMS (FAB⁺) calc’d for [C₃₂H₃₀O₁₁Br⁸¹Br]⁺: m/z 750.0135, found 750.0159. mp = 175–177 °C (heptane/EtOAc).

methyl-5-(2a-acetyl-3-((benzyloxy)methoxy)-2-oxohexahydro-2H-1-oxacyclopropa[cd]pentalen-2b-yl)-3-methylfuran-2-carboxylate (24)

Cyclopropane 24 was obtained from allylic alcohol 15 using the same generic route employed to obtain cyclopropane 18. For the cyclopropanation reaction, cyclopropane 24 was obtained as an oil (3.5 mg, 74 %). ¹H NMR (500 MHz, C₆D₆) δ 7.30 (d, J = 8.3 Hz, 2H), 7.13 (d, J = 7.6 Hz, 1H), 7.06 (t, J = 7.3 Hz, 2H), 5.93 (s, 1H), 4.93 (d, J = 7.1 Hz, 1H), 4.63 (d, J = 7.1 Hz, 1H), 4.60 (s, 2H), 4.11 (m, 1H), 3.42 (d, J = 4.6 Hz, 1H), 3.39 (s, 3H), 2.42 (s, 3H), 2.12 (s, 3H), 1.86 (d, J = 14.2 Hz, 1H), 1.20 (dd, J = 3.2, 14.2 Hz, 1H); ¹³C NMR (125 MHz, C₆D₆) δ 195.0, 169.4, 158.9, 150.4, 140.4, 138.0, 131.4, 128.2, 114.2, 93.3, 80.7, 69.7, 51.4, 50.6, 50.3, 44.8, 28.6, 11.1; IR (film) 2952, 1769, 1713, 1297, 1102, 1037 cm⁻¹; HRMS (EI⁺) calc’d for [C₂₄H₂₃DO₈]⁺: m/z 441.1534, found 441.1523.

methyl-5-(2a-acetyl-3-(methoxymethoxy)-2-oxohexahydro-2H-1-oxacyclopropa[cd]pentalen-2b-yl)-3-methylfuran-2-carboxylate (25)

Cyclopropane 25 was obtained from allylic alcohol 15 using the same generic route employed to obtain cyclopropane 18. For the cyclopropanation reaction, cyclopropane 25 was obtained as an oil (2.3 mg, 62%). ¹H NMR (500 MHz, C₆D₆) δ 5.95 (s, 1H), 4.78 (d, J = 7.1 Hz, 1H), 4.48 (d, J = 7.1 Hz), 4.09 (m, 1H), 3.43 (d, J = 4.9, 1H), 3.39 (s, 3H), 3.23 (s, 3H), 2.43 (s, 3H), 2.13 (s, 3H), 1.82 (d, J = 14.2, 1H), 1.20 (dd, J = 3.4, 14.2 Hz, 1H); ¹³C NMR (125 MHz, C₆D₆) δ 196.0, 170.3, 159.8, 151.3, 141.2, 132.3, 115.0, 96.1, 81.5, 56.3, 52.3, 51.5, 51.1, 45.8, 45.7, 29.4, 12.0; IR (film) 2926, 1767, 1711, 1402, 1297, 1100 cm⁻¹; HRMS (EI⁺) calc’d for [C₁₈H₁₉DO₈]⁺: m/z 365.1221, found 365.1216.

methyl-5-(2a-acetyl-2-oxo-3-((2-(trimethylsilyl)ethoxy)methoxy)hexahydro-2H-1-oxacyclopropa[cd]pentalen-2b-yl)-3-methylfuran-2-carboxylate (26)

Cyclopropane 26 was obtained from allylic alcohol 15 using the same generic route employed to obtain cyclopropane 18. For the cyclopropanation reaction, cyclopropane 26 was obtained as an oil (2.9 mg, 62%). ¹H NMR (500 MHz, C₆D₆) δ 6.02 (s, 1H), 4.94 (d, J = 7.1Hz, 1H), 4.65 (d, J = 7.1 Hz, 1H), 4.13-4.11 (m, 1H), 3.78-3.73 (m, 1H), 3.46 (d, J = 4.6Hz, 1H), 3.42 (s, 3H), 2.44 (s, 3H), 2.15 (s, 3H), 1.93 (d, J = 14.2 Hz, 1H), 1.27 (dd, J = 3.2, 13.9 Hz, 1H), 0.94-0.90 (m), 0.01 (s, 9H); ¹³C NMR (125 MHz, C₆D₆) δ 196.0, 170.3, 159.8, 151.5, 141.2, 132.4, 115.2, 94.3, 81.6, 66.5, 56.2, 51.6, 51.1, 45.9, 46.0, 29.5, 18.6, 12.0, - 0.80; IR (film) 2953, 1770, 1713, 1440, 1297 cm⁻¹; HRMS (FAB⁺) calc’d for [C₂₂H₂₉DO₈SiNa]⁺: m/z 474.1773, found 474.1675.

methyl-5-(2a-acetyl-3-((tert-butyldimethylsilyl)oxy)-2-oxohexahydro-2H-1-oxacyclopropa[cd]pentalen-2b-yl)-3-methylfuran-2-carboxylate (27)

Cyclopropane 27 was obtained from allylic alcohol 15 using the same generic route employed to obtain cyclopropane 18. For the cyclopropanation reaction, cyclopropane 27 was obtained as an oil (6.5 mg, 71%). ¹H NMR (500 MHz, C₆D₆) δ 5.71 (s, 1H), 4.15-4.13 (m, 1H), 3.42 (s, 3H), 3.33 (d, J = 4.6 Hz, 1H), 2.45 (s, 3H), 2.13 (s, 3H), 1.63 (d, J = 13.7 Hz, 1H), 1.21 (dd, J = 3.4, 13.7 Hz, 1H), 1.09 (s, 9H), 0.29 (s, 3H), 0.10 (s, 3H); ¹³C NMR (125 MHz, C₆D₆) δ 196.3, 170.2, 159.7, 151.2, 141.4, 132.0, 114.4, 81.6, 53.6, 52.6, 51.4, 48.2, 45.3, 29.4, 26.3, 18.8, 11.9, −4.5, −4.6; IR (film) 2953, 2930, 2857, 1770, 1714, 1297, 1113 cm⁻¹; HRMS (FAB⁺) calc’d for [C₂₂H₂₉DO₇SiNa]⁺: m/z 459.1800, found 458.1749.

methyl-5-(2a-acetyl-3-((2-methoxyethoxy)methoxy)-2-oxohexahydro-2H-1-oxacyclopropa[cd]pentalen-2b-yl)-3-methylfuran-2-carboxylate (28)

Cyclopropane 28 was obtained from allylic alcohol 15 using the same generic route employed to obtain cyclopropane 18. For the cyclopropanation reaction, thermal conditions were used, and cyclopropane 28 was obtained as an oil (9.1 mg, 42%). ¹H NMR (500 MHz, CDCl₃) δ 6.23 (s, 1H), 5.11 (m, 1H), 5.02 (d, J = 7.1 Hz, 1H), 4.92 (d, J = 7.1 Hz, 1H), 3.99 (d, J = 4.9 Hz, 1H), 3.86 (s, 3H), 3.83 (q, J = 4.4 Hz, 2H), 3.58 (t, J = 4.4 Hz, 2H), 3.39 (s, 3H), 2.46 (s, 3H), 2.34 (d, J = 14.2 Hz, 1H), 2.29 (s, 3H), 2.20 (dd, J = 3.2, 14.2 Hz, 1H); ¹³C NMR (125 MHz, CDCl₃) δ 196.2, 170.1, 159.4, 149.8, 140.2, 131.9, 114.7, 94.3, 81.5, 71.6, 67.4, 59.0, 51.6, 51.4, 50.9, 45.4, 45.2, 29.1, 11.6; IR (film) 2925, 1767, 1711, 1613, 1548, 1440, 1297, 1198, 1099, 1037 cm⁻¹; HRMS (EI⁺) calc'd for [C₂₀H₂₃DNaO₉]⁺: m/z 432.1381, found 432.1365.

methyl-5-(3-acetyl-5-hydroxy-4-methoxy-2-oxohexahydro-2H-cyclopenta[ b]furan-4-yl)-3-methylfuran-2-carboxylate (32)

Furan 20 (49.0 mg, 0.153 mmol, 1.00 equiv) and La(OTf)₃ (89.5 mg, 0.153 µmol, 1.00 equiv) were dissolved in MeOH (3.00 mL) at ambient temperature, and then heated to 55 °C and the reaction was stirred for 1–2 hours. The reaction was diluted with EtOAc and H₂O and the aqueous layer was extracted 3× with EtOAc. The combined organic layers were washed with brine, dried with Na₂SO₄, filtered, and concentrated in vacuo. The residue was purified by flash chromatography using 60:40 EtOAc:Hexanes to afford alcohol 32 as an oil (35 mg, 65%). ¹H NMR (300 MHz, CDCl₃) δ 6.34 (s, 1H), 5.30 (br, 1H), 5.17 (t, J = 7.3 Hz, 1H), 4.64 (d, J = 3.8 Hz, 1H), 4.10 (t, J = 6.7 Hz, 1H), 3.91 (d, J = 5.5 Hz, 1H), 3.89 (s, 3H), 3.03 (s, 3H) 2.50 (ddd, J = 4.1, 7.1, 15.4 Hz, 1H), 2.43 (s, 3H), 2.47 (s, 3H), 2.20 (d, J = 15.4 Hz, 1H); ¹³C NMR (125 MHz, C₆D₆) δ 200.0, 172.4, 159.9, 154.3, 140.9, 132.1, 116.2, 88.5, 83.5, 77.0, 58.6, 51.4, 51.4, 45.9, 40.0, 29.6, 11.8; IR (film) 3474, 2952, 1769, 1717, 1297, 1195, 1098 cm⁻¹; HRMS (EI⁺) calc'd for [C₁₇H₂₁O₈Na]⁺: m/z 353.1236, found 353.1246.

methyl-5-((E)-3-(1-((4-bromobenzoyl)oxy)ethylidene)-5-hydroxy-4-methoxy-2-oxohexahydro-2H-cyclopenta[b]furan-4-yl)-3-methylfuran-2-carboxylate (33)

Triethylamine (43.0 µL, 306 µmol, 4.00 equiv) was added at 0 °C and under N₂, to a solution of alcohol 32 (27.0 mg, 76.6 µmol, 1.00 equiv) in CH₂Cl₂ (955 µ L). After 1 minute, p-bromobenzoyl bromide (42.0 mg, 192 µmol, 2.50 equiv) was added, and the reaction was stirred for 30 minutes at 0 °C. The reaction was allowed to warm to ambient temperature and stir for an additional 1 hour. The reaction was then diluted with Et₂O and H₂O, the layers were separated, and the aqueous layer was extracted once with Et₂O. The Et₂O layers were combined, washed with sat’d NaHCO₃ (aq.), washed with brine, dried with MgSO₄, filtered, and concentrated in vacuo. Crystals suitable for X-ray analysis were obtained by crystallization of crude furan 33 (54 mg, 98% yield) from EtOAc:heptane. ¹H NMR (500 MHz, CDCl₃) δ 7.78 (d, J = 8.5 Hz, 2H), 7.62 (d, J = 8.5 Hz, 2H), 6.38 (s, 1H) 5.06 (t, J = 6.4 Hz, 1H), 4.46 (q, J = 7.2, 3.5 Hz, 1H), 3.80 (m, 4H), 3.18 (s, 3H), 2.45 (s, 3H), 2.37 (m, 1H), 2.33 (s, 3H), 2.28 (d, J = 3.9 Hz, 1H); ¹³C NMR (125 MHz, CDCl₃) δ 169.5, 162.0, 159.9, 153.5, 139.9, 132.0, 131.9, 131.6, 129.2, 129.0, 127.6, 115.8, 115.3, 88.7, 79.8, 52.6, 52.3, 51.5, 38.5, 17.1, 11.6; IR (film) 3445, 2950, 1749, 1705, 1589, 1439, 1399, 1300, 1231, 1070, 1010 cm⁻¹; HRMS (EI⁺) calc'd for [C₂₄H₂₃BrO₉]⁺: m/z 557.0423, found 557.0427. mp = 158–159 °C (EtOAc:Heptane).