Enantioselective total syntheses of (−)-clasto-lactacystin β-lactone and 7-epi-(−)-clasto-lactacystin β-lactone

Abstract

An alkylidene carbene 1,5-CH insertion has been used as a key step in an efficient enantioselective total synthesis of (−)-clasto-lactacystin β-lactone, and its C7-epimer. An additional noteworthy feature of the synthesis is the use of a novel oxidative deprotection procedure, utilizing DMDO, for the conversion of a late-stage benzylidene acetal into a primary alcohol and a secondary benzoate ester.

---

(+)-Lactacystin 1 was isolated from the bacterial strain, Streptomyces sp. OM-6519 by Õmura in 1991 during a screening program for small molecule mimics of nerve growth factors.¹ Lactacystin was later found to be a specific inhibitor of the 20S proteasome found in mammalian and bacterial cells.² The proteasome is responsible for the normal turnover of cellular proteins and degradation of damaged and mutated proteins, and 1 has played a vital role in the study of its function. The remarkable biological activity and intriguing structure of 1 has sparked continued interest over the last 13 years and a number of syntheses have been published to date.³ Lactacystin 1 spontaneously and reversibly forms (−)-clasto-lactacystin β-lactone (also known as omuralide) 2 in the extracellular medium, and it is this β-lactone that penetrates the cell and acylates an active site threonine residue, leading to inhibition of the proteasome (Scheme 1). As well as serving as the biologically active form of lactacystin 1, clasto-lactacystin β-lactone 2 has also served as a synthetic precursor to the natural product 1 itself, and the enantioselective synthesis of 2 has become an important synthetic challenge in its own right.

Scheme 1

In 2002 we reported a formal synthesis of (+)-lactacystin 1,⁴ which used a highly stereoselective alkylidene carbene 1,5-CH insertion reaction as a method to construct the key quaternary stereocentre.^5,6 In this first generation route we utilized TBS-protecting groups during the CH-insertion step (vis3 → 4, R = TBS), but in order to mitigate unexpected problems during scale-up we chose to explore an alternative protecting group strategy in a second generation route to 1 and 2. As the TBS-protecting groups on 3 (Scheme 1) created a highly hindered environment at the site of CH-insertion, we chose to explore the possibility of protecting the 1,3-diol moiety in 3 as a benzylidene acetal. During the preparation of this manuscript, Wardrop independently reported an alkylidene carbene 1,5-CH-insertion approach to 1,⁷ following an identical approach to that outlined in Scheme 1, which also uses a benzylidene acetal to protect the diol moiety in 3 and 4. We now wish to report our own results in this area, which have led to a second generation enantioselective total synthesis of clasto-lactacystin β-lactone 2 as well as its C7-epimer 22.

The key vinyl bromide CH-insertion precursor 7 was readily prepared in 4 steps from epoxy alcohol 5 according to the route shown below (Scheme 2).

Scheme 2 Reagents and conditions: (a) NaN₃, B(OMe)₃, DMF, 50 °C (77%); (b) PhCH(OMe)₂, PPTS, C₆H₆, reflux (78%); (c) H₂, Lindlar cat., MeOH, 20 °C (100%); (d) MnO₂, 8, DCM, 4 Å mol. sieves, reflux, then NaCNBH₃, AcOH, MeOH, 0 °C (75%); (e) KHMDS, THF, −30 °C (83%).

Thus treatment of the known epoxide 5 under Miyashita's conditions⁸ gave 2-azido-1,3-diol 6 in 77% yield. Acetal protection and hydrogenation next gave the amine 9 in good yield. One-pot MnO₂ oxidation, imine formation and reduction, as described by Taylor,⁹ finally provided the (E)-vinyl bromide cyclisation precursor 7 as a single isomer in 59% yield over 3 steps. After careful optimization, exposure of 7 to an excess (2 equiv.) of KHMDS in THF at −30 °C gave the desired 3-pyrroline 10 in a very pleasing 83% isolated yield. The cyclisation of 7 to 10 was also accompanied by the formation of the corresponding but-2-ynyl-amine 11 (13% yield), but this byproduct could be easily removed by column chromatography.

Having secured an efficient route to the 3-pyrroline 10, we next needed to install the additional oxygenation and remaining stereocentres present in 1 and 2. Thus, oxidation with TPAP–NMO, and then sodium chlorite first gave the N-chlorolactam 12 and treatment with sodium borohydride then provided the desired 3-pyrrolinone 14 in 76% over 3 steps.¹⁰ Dihydroxylation was then accomplished using Sharpless's modified UpJohn conditions¹¹ which cleanly afforded the diol 13 as a single, crystalline diastereoisomer in 96% yield. Formation of the cyclic thiocarbonate, followed by selective deoxygenation with ⁿBu₃SnH and AIBN¹² then gave a separable 1 : 3.6 mixture of the 15 and 16 in 94% combined yield (Scheme 3).

Scheme 3 Reagents and conditions: (a) TPAP, NMO, MeCN, 4 Å mol. sieves (93%); (b) NaClO₂, NaH₂PO₄·2H₂O, 2-methyl-2-butene, ^tBuOH, H₂O, 0 °C; (c) NaBH₄, MeOH, 20 °C (84% over 2 steps); (d) K₂OsO₂(OH)₄, NMO, citric acid, ^tBuOH–H₂O (1 : 1), 20 °C (96%); (e) thiocarbonyl diimidazole, THF, reflux (100%); (f) ⁿBu₃SnH, AIBN, toluene, reflux (94%); (g) NaOMe, MeOH; (h) H₂, Pd/C, HCl, MeOH (93%).

Although base-mediated epimerisation of the undesired diastereoisomer 16 proved difficult to achieve in high yield, we found that 16 could be transformed into the 3-pyrrolinone 14 by treatment with NaOMe–MeOH, hence providing a high yielding (90%) method of recycling material through to the desired cis-isomer 15.

Once we had access to 15, we were able to complete a formal synthesis of 1 by conversion to the known triol 17 by standard catalytic hydrogenation in 93% yield (Scheme 3). Whilst we were pleased with this initial success, we were aware that 7 steps were still required to convert 17 into the natural product 1, and we therefore hoped to shorten this sequence by developing a selective deprotection–oxidation strategy from lactam 15.

Scheme 4 Reagents and conditions: (a) DMDO (3 equiv., 0.06 M), acetone, 0 °C (100%); (b) Dess-Martin periodinane, DCM, 20 °C (63%); (c) NaClO₂, NaH₂PO₄·2H₂O, HOSO₂NH₂, ^tBuOH, H₂O, 20 °C (81%); (d) NaOMe, MeOH, 20 °C (78%); (e) BOPCl, Et₃N, DCM, 20 °C (38%).

As we had access to both 15 and 16, we first decided to investigate this new ‘end-game’ approach on the more abundant trans-isomer 16 (Scheme 4). Thus, treatment of the benzylidene acetal 16 with DMDO (dimethyldioxirane) caused a selective, oxidative–deprotection to give the alcohol-ester 18, as a single regioisomer, in quantitative yield.¹³ This remarkable deprotection reaction gives both a free primary alcohol and protected secondary alcohol in one convenient transformation. Selective oxidation of the primary alcohol in 18 to the acid 19 was achieved using sequential Dess-Martin periodinane (63%) and sodium chlorite (81%) oxidations. Deprotection of 19 and formation of the β-lactone by treatment with BOPCl (bis(2-oxo-3-oxazolidinyl)phosphinic chloride) finally afforded 7-epi-clasto-lactacystin β-lactone 22. The spectroscopic data of 22 were identical to those reported by Corey for this compound.¹⁴

Having completed the synthesis of 22, we then turned our attention to repeating the same chemical steps on the correct diastereoisomer isomer 15 required for the synthesis of (+)-lactacystin 1 (Scheme 5). Thus, selective deprotection according to our newly developed procedure cleanly gave the alcohol-ester 23 in 75% yield. Sequential Dess-Martin periodinane (35%) and sodium chlorite (95%) oxidations then gave the acid 24. Pleasingly, the ester 24 could be saponified to afford the previously reported acid 26, which was then converted to clasto-lactacystin β-lactone 2 using the known procedure.¹⁵

Scheme 5 Reagents and conditions: (a) DMDO (3 equiv. 0.06 M), acetone, 0 °C (90%); (b) Dess-Martin periodinane, DCM, 20 °C (35%); (c) NaClO₂, NaH₂PO₄·2H₂O, HOSO₂NH₂, ^tBuOH, H₂O, 20 °C (95%); (d) NaOMe, MeOH, 20 °C (74%); (e) BOPCl, Et₃N, DCM, 20 °C (45%).

In summary, we have developed a second generation synthesis of 1 and 2 using an alkylidene carbene 1,5-CH insertion reaction as a key step, and during this work we have also completed a total synthesis of the recently described proteasome inhibitor 7-epi-clasto-lactacystin β-lactone 22. In addition to developing novel methodology for the construction of quaternary stereocentres, we have also utilised a novel procedure for the selective oxidative deprotection of benzylidene acetals using DMDO and we are currently exploring the wider synthetic utility of this particular transformation.

Acknowledgements

We thank the EPSRC and Pfizer Central Research for financial support of this work.

References

1. S. Õmura, T. Fujimoto, K. Otoguro, K. Matsuzaki, R. Moriguchi, H. Tanaka and Y. Sasaki, J. Antibiot., 1991, 44, 113.
2. G. Fenteany, R. F. Standaert, W. S. Lane, S. Choi, E. J. Corey and S. L. Schreiber, Science, 1995, 268, 726.
3. (a) For a review of this area see: J. S. Panek, C. E. Masse, A. J. Morgan and J. Adams, Eur. J. Org. Chem., 2000, 2513 , and references therein; (b) T. J. Donohoe, H. O. Sintim, L. Sisangia, K. W. Ace, P. M. Guyo, A. Cowley and J. D. Harling, Chem. Eur. J., 2005, 11, 4227; (c) T. J. Donohoe, H. O. Sintim, L. Sisangia and J. D. Harling, Angew. Chem., Int. Ed., 2004, 43, 2293; (d) H. Ooi, N. Ishibashi, Y. Iwabuchi, J. Ishihara and S. Hatekeyama, J. Org. Chem., 2004, 69, 7765.
4. M. P. Green, J. C. Prodger and C. J. Hayes, Tetrahedron Lett., 2002, 43, 6609.
5. M. P. Green, J. C. Prodger, C. J. Hayes and A. E. Sherlock, Org. Lett., 2001, 3, 3377.
6. (a) J. M. A. Auty, I. Churcher and C. J. Hayes, Synlett, 2004, 1443; (b) D. M. Bradley, R. Mapitse, N. M. Thomson and C. J. Hayes, J. Org. Chem., 2002, 67, 7613; (c) S. M. Worden, R. Mapitse and C. J. Hayes, Tetrahedron Lett., 2002, 43, 6011; (d) R. Mapitse and C. J. Hayes, Tetrahedron Lett., 2002, 43, 3541; (e) R. Gabaitsekgosi and C. J. Hayes, Tetrahedron Lett., 1999, 40, 7713; (f) R. S. Grainger and R. B. Owoare, Org. Lett., 2004, 6, 2961; (g) A. Sasakai, T. Aoyama and T. Shiori, Tetrahedron Lett., 2000, 41, 6859; (h) D. F. Taber, J.-L. Liang, B. Chen and L. Cai, J. Org. Chem., 2005, 70, 8739; (i) D. F. Taber and P. H. Storck, J. Org. Chem., 2003, 68, 7768; (j) D. F. Taber, T. D. Neubert and A. L. Rheingold, J. Am. Chem. Soc., 2002, 124, 12416; (k) D. J. Wardrop and W Zhang, Tetrahedron Lett., 2002, 43, 5389.
7. J. J. Wardrop and E. G. Bowen, Chem. Commun., 2005, 5106.
8. M. Sasaki, K. Tanino, A. Hirai and M. Miyashita, Org. Lett., 2003, 5, 1789.
9. H. Kanno and R. J. K. Taylor, Tetrahedron Lett., 2002, 43, 7337.
10. M. P. Green, J. C. Prodger and C. J. Hayes, Tetrahedron Lett., 2002, 43, 2649.
11. P. Dupau, R. Epple, A. A. Thomas, V. V. Pokin and K. B. Sharpless, Adv. Synth. Catal., 2002, 344, 421.
12. H. Uno, J. E. Baldwin and A. T. Russell, J. Am. Chem. Soc., 1994, 116, 2139.
13. Z. F. Akbalina, S. S. Zlotskii, N. N. Kabal'nova, I. A. Grigor'ev, S. A. Kotlyar and V. V. Shereshovets, Russ. J. Appl. Chem., 2002, 75, 1120.
14. L. R. Reddy, P. Saravan, J.-F. Fournier, S. V. Reddy and E. J. Corey, Org. Lett., 2005, 7, 2703.
15. E. J. Corey, W. Li and G. A. Reichard, J. Am. Chem. Soc., 1998, 120, 2330.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental procedures and characterization data for 10, 11, 17, 18 and 23. See DOI: 10.1039/b516311k

---