PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Tetrahedron. Author manuscript; available in PMC 2010 August 1.
Published in final edited form as:
Tetrahedron. 2009 August 1; 65(31): 5904–5907.
doi:  10.1016/j.tet.2009.05.059
PMCID: PMC2719863
NIHMSID: NIHMS120348

Preparation of the Major Urinary Metabolite of (-)-Prostaglandin E2

Abstract

The best way to measure whole body production of the locally acting hormone prostaglandin E2 (PGE2) is to assess the accumulation of the major urinary metabolite, PGE2 U M. A practical preparation of this delicate diacid is described. This synthetic PGE2UM will enable production of the antibodies that will be used to quantify this key metabolite.

Introduction

Prostaglandin E21 (PGE2) (Eq. 1) displays a wide array of biological activity.1 Recently, it was observed that some lung cancers are associated with an elevated production of PGE2.2 The best way to measure whole body production of the locally acting hormone is to assess the accumulation of the major urinary metabolite, PGE2 U M (2). We had reported the preparation of the ethyl ester 3 of racemic PGE2UM.3 We now describe a practical preparation of the enantiomerically pure diacid 2, requisite for the production of the PGE2UM antibodies that will be used to quantify this key metabolite.

equation image
(1)

Result and Discussion

For the preparation of 2 (Eq. 2), there were three problems to be addressed: improvement of the production of the racemic allylic alcohol 4 having the alkyl substituents trans on the cyclopentane ring, resolution4 of 4, and the conversion of the enantiomerically pure acetate R-5 to (-)-PGE2UM2. Throughout, we had to be mindful of the sensitivity of the easily-dehydrated β-hydroxy ketone.

equation image
(2)

The starting materials for the synthesis of 4 were, as before,3a the diazo ketone 8 and the diene aldehyde 12 (Scheme 1). As previously, the diazoketone 8 was prepared by Michael addition of benzoylacetone to ethyl acrylate, to give diketone 7. The diketone 7 was smoothly converted5 to the diazo ketone 8 on exposure to p-nitrobenzenesulfonyl azide (p-NBSA) and DBU.

We have made several improvements in the preparation of the diene aldehyde 12. Ozonolysis of cyclohexene followed by overnight (vs. 45 min) exposure to triethylamine and EtOH6 led to ethyl 6-oxohexanoate 9 in better yield (78% vs 45%). Wittig reaction of the phosphonium salt 10 with the replacement of 1 equiv KHMDS (original) with t-BuOK proceeded smoothly to give the 6E, 8Z-dienol 11 with a similar improvement (82% vs 46%). Oxidation of 11 with 15 mol equiv MnO2, which is less expensive than the Dess-Martin periodinane previously used, delivered the dienal 12 with a slightly improved yield (86% vs 81%).

With the two components 8 and 12 in hand, we used the original procedure to prepare the aldol product 16 (Scheme 2). Condensation of the potassium enolate of the diazoketone 8 with the dienal 12 in the presence of triethylchlorosilane (TESCl) in toluene gave the TES-protected aldol 13 together with the free aldol product 14. Several other bases were tried, but without positive results. The recovered dienal was found to be an E/Z and E/E mixture, so it could not be recycled. Removal of the TES group from the protected aldol product 13 with TBAF in THF delivered an additional quantity of the free aldol product 14. Silylation of the combined lots of 14 with TBSCl led to the desired cyclization precursor 15.

The diazoketone 15 was cyclized with catalytic rhodium octanoate in CH2Cl2 to provide the bicyclic ketones 16 and 17 in a ratio of 65:35. We were not able to separate these two substances by chromatography, so we developed an improved protocol for preparing pure 16. We found that while the alcohol from reduction of 16 had about the same TLC Rf as 16 itself, the alcohol from reduction of 17 was much more polar. Further, 17 appeared to be significantly more reactive toward Dibal than was 16. Thus, reduction of the mixture of 16 and 17 with a limited amount of DIBAL led to total reduction of 17 to its alcohol, which was easily separated. The recovered mixture of 16 and its derived alcohol was then oxidized to give pure 16.

We then addressed the opening of the cyclopropyl ketone 16 in the presence of the TBS group. In the past,3 we had run the opening at low temperature, to be able to isolate the kinetic product from ring opening, having the two alkyl substituents cis on the cyclopentane ring. To our delight, we found that allowing the reaction mixture to warm to -30 to -10 °C for 6 h gave the pure free trans thioether 19 and some free bicyclic ketone 18, that could be recycled with TBSCl to 16. Silylation of 19 gave 20, that on oxidation and Mislow rearrangement7 gave the desired racemic allylic alcohol 4.

We had planned to effect the resolution of 4 by acetylation with the (R)-selective Amano lipase AK and vinyl acetate 21. In the past,4 we have observed that such acetylations produce high ee acetate initially, but were sometimes difficult to drive to completion. It was therefore important in the synthesis design to have a Z alkene in 20 and its precursors, so that Mislow rearrangement would give the diastereomer 4, such that the enantiomer of 4 having the R alcohol at C-15 would have the same ring absolute configuration as the natural prostaglandins. That way, the high ee acetate R-5 initially produced could be carried on directly to the enantiomerically pure urinary metabolite 2. The success of this approach was confirmed by the optical rotation of the final product 2.

We initially planned (Scheme 4) to convert (R)-5 to the diacid 2 by reduction of the ketone to the alcohol, saponification, and oxidation of the diacid diol so prepared to the diacid diketone. The preparation of the diacid diol 22 was successful, but we were not able to find conditions for the purification of the derived diacid diketone. We found it more practical to alkylate the diacid with benzyl bromide, leading to the bis benzyl ester 23. Oxidation8 of 23 proceeded smoothly to give the enone 24.

The β-hydroxy ketone 25 was quite sensitive, quickly deteriorating in CDCl3. Nevertheless, deprotection of 24 with HF in pyridine was sufficiently mild to deliver intact 25. Hydrogenation then completed the synthesis of (-)-PGE2UM2.

There had been three previous syntheses of (-)-2, by groups at Lilly,9a at Merck9b and at Upjohn.9c Our observed [α]D = -9.1(EtOH) agreed well with the reported9a [α]D = -8 (EtOH). The 1H NMR of our synthetic 2 was also congruent with those previously observed.9

Conclusion

We have developed a practical stereoselective synthetic approach to the major urinary metabolite of (-)-prostaglandin E2. This synthesis has made this metabolite available in sufficient quantity to allow continuation of the ongoing physiological studies.

Experimental

Diethyl (13S*)-(14E)-(1,2,3,4-Tetranor)-9-oxo-11-hydroxy-13-phenylthio-Δ14-prostanedioate 19

To a stirred solution of the bicyclic ketone 16 (118 mg, 0.244 mmol) and thiophenol (81 mg, 0.733 mmol) in CH2Cl2 (2.5 mL) at -78 °C was added a solution of BF3·OEt2 (139 mg, 2.13 mmol) in CH2Cl2 (0.5 mL) over 5 min. The mixture was maintained at -30 to -10 °C for an additional 6 h. The reaction mixture was partitioned between CH2Cl2 and, sequentially, saturated aqueous NaHCO3 and brine. The organic extract was dried (Na2SO4) and concentrated, and the residue was chromatographed to afford the free thioether 19 (60 mg, 51% yield from 16) as a colorless oil; TLC Rf (MTBE/ petroleum ether = 1/1) = 0.24; 1H NMR δ 7.44-7.47 (m, 2H), 7.27-7.30 (m, 3H), 5.35-5.48 (m, 2H), 4.36 (q, 1H, J = 6.8 Hz), 4.21 (dd, 1H, J = 5.2 and 9.6 Hz), 4.13 (m, 4H), 3.14 (bs, 1H), 2.71 (dd, 1H, J = 7.2 and 18.4 Hz), 2.42-2.51 (m, 1H), 2.11-2,42 (m, 7H), 1.93-2.03 (m, 1H), 1.81-1.85 (m, 1H), 1.63-1.67 (m, 1H), 1.43-1.48 (m, 2H), 1.21-1.27 (m, 6H), 1.07-1.19 (m, 2H); 13C NMR δ u:10 215.9, 173.9, 173.2, 133.7, 60.43, 60.40, 47.0, 33.9, 31.3, 28.6, 26.9, 25.3, 24.5; d: 133.9 (2), 132.7, 128.9 (2), 127.9, 127.4, 70.9, 53.0, 50.3, 48.6, 14.2 (2); IR (film) 3467, 2931, 1729, 1374, 1179 cm-1; ESI MS (m/z), 499.2 (M++Na); HRMS calcd for C26H36O6SNa 499.2130 found 499.2126. Further elution gave the free bicyclic ketone 18 (22 mg, 24% yield from 16) as a colorless oil, TLC Rf (MTBE/petroleum ether = 1/1) = 0.11. The analytical data for 18 were found to be identical to those previously reported.

Diethyl (13S*)-(14E)-(1,2,3,4-Tetranor)-9-oxo-11-dimethyl(dimethylethyl)silyloxy-13-phenylthio-Δ14-prostanedioate 20

To a stirred solution of thioether 19 (84 mg, 0.18 mmol) in CH2Cl2 (1.5 mL) was added imidazole (34 mg, 0.53 mmol) and TBSCl (53 mg, 0.35 mmol), then DMAP (3 mg, cat) at room temperature. After an additional 30 h, the mixture was partitioned between CH2Cl2 and H2O. The organic extract was dried (Na2SO4) and concentrated, and the residue was chromatographed to afford the TBS-protected thioether 20 (79 mg, 76% yield from 19) as a colorless oil; TLC Rf (MTBE/petroleum ether = 3/7) = 0.58; 1H NMR δ 7.34-7.36 (m, 2H), 7.18-7.20 (m, 3H), 5.27-5.34 (m, 2H), 4.40 (q, 1H, J = 1.6 Hz), 4.00-4.06 (m, 4H), 3.94 (dd, 1H, J = 6.0 Hz and 10.4 Hz), 2.54 (dd, 1H, J = 6.0 Hz and 18.0 Hz), 2.31-2.38 (m, 2H), 2.07-2.17 (m, 5H), 1.91-1.97 (m, 2H), 1.63-1.72 (m, 1H), 1.50-1.60 (m, 1H), 1.32-1.39 (m, 2H), 1.14-1.17 (m, 6H), 1.01-1.11 (m, 1H), 0.84-0.96 (m, 1H), 0.81 (s, 9H), 0.04 (s, 3H), 0.01 (s, 1H); 13C NMR δ u: 217.1, 173.4, 173.0, 134.0, 60.3, 60.2, 47.4, 34.0, 31.8, 28.4, 27.1, 26.0, 24.5, 17.8; d: 133.9 (2), 132.5, 128.9 (2), 127.8, 127.6, 71.2, 54.1, 49.4, 48.4, 25.7 (3), 14.25, 14.22, -4.4, -4.7; IR (film) 2933, 2858, 1737, 1180, 836 cm-1; ESI MS (m/z), 613.3 (M++Na); HRMS calcd for C32H50O6SSiNa 613.2995 found 613.2990.

Diethyl (15R*)-(1,2,3,4-Tetranor)-9-oxo-11-dimethyl(dimethylethyl)silyloxy-15-acetoxyprostanedioate 4

To a stirred solution of the thioether 20 (154 mg, 0.26 mmol) in CH2Cl2 (12 mL) at -78 °C was added a solution of 3-chloroperoxybenzoic acid (90 mg, 0.52 mmol) in CH2Cl2 (2.0 mL) over a period of 10 min. The mixture was stirred for an additional 50 min, then a solution of trimethyl phosphite (324 mg, 2.61 mmol) in EtOH (4 mL) was added. The mixture was stirred at -78 °C for an additional 10 min and then the cooling bath was removed and the mixture was stirred for an additional 6.5 h at room temperature. The mixture was partitioned between EtOAc and, sequentially, saturated aqueous NaHCO3 and brine. The organic extract was dried (Na2SO4) and concentrated, and the residue was chromatographed to afford the allylic alcohol 4 (132 mg, 99% yield from 20) as a colorless oil; TLC Rf (MTBE/petroleum ether = 1/1) = 0.18; 1H NMR δ 5.64 (dd, J = 6.8 Hz and 15.2 Hz), 5.50 (dd, J = 8.4 Hz and 15.2 Hz), 4.05-4.12 (m, 4H), 4.00 (q, 1 H, J = 8.4 Hz), 2.61 (ddd, 1 H, J = 1.2 Hz, 7.2 Hz and 18.4 Hz), 2.32-2.52 (m, 3H), 2.27 (t, 2 H, J = 7.6 Hz), 2.16 (dd, 1H, J = 8.8 Hz and18.0 Hz), 1.87-2.02 (m, 3 H), 1.77-1.81 (m, 1 H), 1.58-1.65 (m, 2 H), 1.36-1.56 (m, 3 H), 0.83 (s, 9 H), 0.00 (s, 6 H); 13C NMR δ u: 214.9, 173.7, 173.2, 60.5, 60.2, 47.2, 36.7, 34.2, 31.7, 25.0, 24.8, 23.4, 18.0; d: 136.7, 131.0, 72.7, 72.1, 54.5, 52.7, 25.7 (3), 14.24, 14.22, -4.6, -4.7; IR (film) 3467, 2933, 1736, 1101, 838 cm-1; HRMS: calcd for C26H46O7SiNa 521.2911 found 521.2908.

Resolution of racemic allylic alcohol 4

To a stirred solution of racemic alcohol 4 (230 mg, 0.46 mmol) in vinyl acetate 21 (11 ml) was added R-selective Amano lipase AK (1.8 g). The reaction mixture was stirred at 55 °C (oil bath) for an additional 48 h. The suspension was filtered through a pad of celite with MTBE, and the filtrate was concentrated and chromatographed to afford the acetate R-5 (87 mg, 70% based on the R-allylic alcohol charged) and recovered alcohol 4 (125 mg, 54% based on the racemic allylic alcohol) as colorless oils. For R-5: TLC Rf (MTBE/petroleum ether = 1/1) = 0.59; [α]D -4.9 (c 0.83, CH2Cl2); 1H NMR δ 5.52-5.63 (m, 2 H), 5.22-5.29, 4.04-4.12 (m, 4 H), 4.00 (q, 1 H), 2.61 (ddd, 1 H, J = 1.2 Hz, 7.2 Hz and 18.4 Hz), 2.31-2.43 (m, 3 H), 2.26 (t, 3 H, J = 7.6 Hz), 2.16 (dd, 1 H, J = 8.8 Hz and 18.4 Hz), 1.98-2.03 (m, 4 H), 1.85-1.89 (m, 1 H), 1.75-1.78 (m, 1 H), 1.57-1.65 (m, 4 H), 1.28-1.40 (m, 2 H), 1.20-1.24 (m, 6 H), 0.84 (s, 9 H), 0.01 (s, 6 H); 13C NMR δ u: 214.5, 173.4, 173.0, 170.2, 60.3, 60.2, 47.1, 34.11, 34.06, 24.73, 24.70, 23.3, 18.0 d: 133.1, 131.7, 72.7, 72.5, 54.5, 52.6, 25.7 (3), 21,2, 14.24, 14.21, -4.7, -4.8; IR (film) 2935, 1735, 1373, 1240, 840 cm-1; ESI MS (m/z), 563.3 (M++Na); HRMS: calcd for C28H48O8SiNa 563.3016 found 563.3012.

Dibenzyl (1,2,3,4-Tetranor)-9,15-dioxo-11-dimethyl(dimethylethyl)silyloxyprostanedioate 24

To stirred solution of acetate R-5 (21.2 mg, 0.039 mmol) in ethanol (1 mL) was added NaBH4 (2.2 mg). After an additional 40 min, the reaction mixture was partitioned between EtOAc and H2O. The organic extract was dried (Na2SO4) and concentrated to afford the crude mixture of diastereomeric alcohols.

To a stirred solution of the crude diols in THF (1 mL) was added a 1 M solution of aqueous LiOH (1 mL, 1 mmol) at room temperature. After stirring for 24 h, the mixture was acidified with 1 M aqueous HCl to pH = 3~4, and partitioned between CHCl3 and saturated brine. The organic extract was dried (Na2SO4) and concentrated to afford the bisacid crude 22.

The crude diacid 22 was dissolved in DMF (1 mL), then powdered K2CO3 (39 mg, 0.28 mmol) and benzyl bromide (48 mg, 0.14 mmol) were added. After stirring for 24 h, the reaction mixture was partitioned between EtOAc and H2O. The organic extract was dried (Na2SO4) and concentrated, and the residue was chromatographed to afford the bisbenzyl ester 23 (12.1 mg, 50% yield for 3 steps from the acetate R-5). TLC Rf (MTBE/petroleum ether = 1/1) = 0.28 and 0.19; 1H NMR (diastereomeric mixture, typical signal peaks were recorded) δ 7.31-7.36 (m, 10 H), 5.30-5.53 (m, 2 H), 5.10-5.11 (m, 4 H), 3.94-4.04 (m, 3 H).

To a stirred solution of the bisbenzyl ester 23 (12.1 mg, 0.019 mmol) in CH2Cl2 (1 mL) was added Dess-Martin periodinane (33 mg, 0.078 mmol) at room temperature. After an additional 30 min, the mixture was concentrated and the residue was chromatographed to afford the desired enone 24 (9.5 mg, 40% yield from R-5) TLC Rf (MTBE/petroleum ether = 1/1) = 0.77. [α]D -34.7 (c 0.45, CH2Cl2); 1H NMR δ 7.31-7.37 (m, 10H), 6.69 (dd, 1H, J= 15.6 and 8.8 Hz), 5.11 (s, 2H), 5.09 (s, 2H), 4.09 (q, 1 H, J = 8.4 Hz), 2.67 (ddd, 1H, J = 0.8 Hz, 7.2 Hz and 18.4 Hz), 2.44-2.57 (m, 5H), 2.47-2.49 (m, 2H), 2.14-2.28 (m, 2H), 1.78-1.96 (m, 2H), 1.65-1.67 (m, 4H), 0.85 (s, 9H), 0.02 (s, 3H), 0.00 (s, 3H); 13C NMR δ u: 213.5, 198.9, 173.2, 172.7, 135.9 (2), 66.3, 66.2, 47.1, 40.3, 34.1, 31.4, 24.5, 23.4, 23.3, 18.0 d: 144.8, 132.2, 128.6 (4), 128.3 (3), 128.22 (2), 128.20, 72.2, 54.5, 52.4, 25.6 (3), -4.65, -4.74; IR (film): 2932, 1738, 1162, 838, 744 cm-1; ESI MS (m/z), 643.2 (M++Na); HRMS: calcd for C36H48O7SiNa 643.3067 found 643.3056.

Dibenzyl (1,2,3,4-Tetranor)-9,15-dioxo-11-hydroxyprostanedioate 25

To a stirred solution of enone 24 (9.5 mg, 0.015 mmol) in CH3CN (1 mL) at 0 °C was added pyridine (0.1 mL) and 52% solution of HF in H2O(0.2 mL). After stirring for 7 h, the reaction mixture was mixed with silical gel, concentrated and the residue was chromatographed to afford the free enone 25 (6.5 mg, 84% yield from the enone 24) TLC Rf (MTBE/petroleum ether = 4/1) = 0.27; [α]D +11.2 (c 0.25, CH2Cl2); 1H NMR δ 7.31-7.38 (m, 10 H), 6.72 (dd, 1 H, J = 8.4 Hz and 15.6 Hz), 6.31 (d, 1 H, J = 15.6 Hz), 5.11 (s, 2 H), 5.09 (s, 2 H), 4.19 (q, J = 8.4 Hz), 2.79 (ddd, J = 0.8 Hz, 7.2 Hz and 18.4 Hz), 2.44-2.60 (m, 5 H), 2.20-2.41 (4 H), 1.81-1.97 (m, 2 H), 1.65-1.68 (m, 5H); 13C NMR δ u: 212.8, 199.2, 173.4, 172.7, 135.9, 135.8, 66.4, 66.3, 46.0, 40.5, 33.9, 31.3, 24.3, 23.4, 23.2; d: 144.2, 132.4, 128.6 (4), 128.3 (3), 128.2 (3), 71.5, 54.3, 53.0; IR (film): 3464, 2928, 1733, 1453, 1163 cm-1; HRMS: calcd for C30H34O7Na 529.2202 found 529.2198.

(1,2,3,4-Tetranor)-9,15-dioxo-11-hydroxy-13,14-dihydroprostanedioic Acid (PGE2UM) 2

To a stirred solution of enone 25 (4.3 mg, 0.0085 mmol) in ethanol (1 mL) was added Pd-C (40 mg). An H2 balloon was attached, and the mixture was stirred at rt for 40 min. The reaction mixture was filtered and concentrated to give the desired PGE2UM2 (2.5 mg, 90% yield from the enone 25). TLC Rf (Et2O/0.5 M pH = 4 aqueous buffer/HOAc = 90/9/1) = 0.14; [α]D -9.1 (c 0.10, MeOH); 1H NMR (CD3OD) δ 3.98 (q, 1H, J= 6.4 Hz), 2.57-2.67 (m, 3H), 2.46-2.48 (m, 2H), 2.32-2.40 (m, 2H), 2.23 (bs, 2H), 2.08 (dd, 1H, J = 18.0 Hz and 6.8 Hz), 1.70-1.84 (m, 5 H), 1.55-1.59 (m, 5 H); 13C NMR δ u: 217.4, 211.6, 176.3, 175.9, 45.9, 41.4, 38.9, 33.4, 26.9, 25.7, 25.1, 23.9 d: 71.5, 52.0, 47.9; IR (film): 3441, 1703, 1406 cm-1; HRMS: calcd for C16H24O7Na 351.1420 found 351.1414.

Supplementary Material

01

Acknowledgments

We thank the National Institutes of Health (GM42056) for support of this work. We thank Dr. John Dykins for mass spectrometric measurements, supported by the NSF (0541775), and Dr. Glenn Yap for X-ray analysis. This work is dedicated to the memory of our friend and colleague Jason D. Morrow.

Footnotes

Supplementary Data Available. General experimental procedures, experimental procedures and spectra for all new compounds can be found in the online version.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

1. For an overview of the physiological function of PGE2, see: (a) Kidney Int. 1986;29:1110. [PubMed] (b) Tchetina EV, Di Battista JA, Zukor DJ, Antoniou J, Poole AR. Arthritis Res Ther. 2007;9:R75. [PubMed] (c) Ma W, Quirion R. Neurosci Lett. 2008;437:165. [PubMed]
2. For leading references to the elevated levels of PGE2 observed with some lung cancers, see (a) Huang M, Stolina M, Sharma S, J T, Zhu L, Miller PW, Wollman J, Herschmann H, Dubinett SM. Cancer Res. 1998;15:1208. [PubMed] (b) Dohadwala M, Batra RK, Luo J, Lin Y, Krysan K, Põld M, Sharma S, Dubinett SM. J Biol Chem. 2002;277:50828. [PubMed] (c) Murphey LJ, Williams MK, Sanchez SC, Byrne LM, Csiki JA, Oates JA, Johnson DH, Morrow JD. Anal Biochem. 2004;334:266. [PubMed]
3. We recently reported a synthesis of the diethyl ester 3: (a) Taber DF, Teng D. J Org Chem. 2002;67:1607. [PubMed]This synthesis was based on an approach to the prostaglandins and isoprostanes that we had previously reported: (b) Taber DF, Hoerrner RS. J Org Chem. 1992;57:441. (c) Taber DF, Herr RJ, Gleave DM. J Org Chem. 1997;62:194. [PubMed] (d) Taber DF, Xu M, Hartnett JC. J Am Chem Soc. 2002;124:13121. [PubMed]
4. For R-selective acetylation of allylic alcohols with Amano lipase AK, see Taber DF, Reddy GP, Arneson KO. J Org Chem. 2008;73:3467. [PubMed]
5. For selective diazo transfer to α-benzoyl ketones, see (a) Taber DF, Gleave DM, Moody K, Hennessy MJ. J Org Chem. 1995;60:2283. (b) Taber DF, Herr RJ, Pack SK, Geremia JM. J Org Chem. 1996;61:2908. [PubMed] (c) Taber DF, Kong S, Malcolm SC. J Org Chem. 1998;63:7953.
6. For the unsymmetrical ozonolysis of alkenes, see (a) Schreiber SL, Claus RE, Reagen J. Tetrahedron Lett. 1982;23:3867. (b) Taber DF, Nakajima K. J Org Chem. 2001;66:3426.
7. (a) Bickart P, Carson FW, Jacobus J, Miller EG, Mislow K. J Am Chem Soc. 1968;90:4869. (b) Tang R, Mislow K. J Am Chem Soc. 1970;92:2100. (c) Evans DA, Andrews GC, Sims CL. J Am Chem Soc. 1971;93:4956. (d) Taber DF. J Am Chem Soc. 1977;99:3513. [PubMed]
8. (a) Dess DB, Martin JC. J Org Chem. 1983;48:4155. (b) Ireland RE, Liu L. J Org Chem. 1993;58:2899.
9. Three previous syntheses of 2 had been reported: (a) Boot JR, Foulis MJ, Gutteridge NJA, Smith CW. Prostaglandins. 1974;8:439. [PubMed] (b) Taub D, Zelawski ZS, Wendler NL. Tetrahedron Lett. 1975;16:3667. (c) Lin CH. J Org Chem. 1976;41:4045. [PubMed]
10. 13C multiplicities were determined with the aid of a JVERT pulse sequence, differentiating the signals for methyl and methine carbons as “d” and for methylene and quaternary carbons as “u”.