Published on Web 05/19/2005

Total Synthesis of Salinosporamide A

Atsushi Endo and Samuel J. Danishefsky*
Laboratory for Bioorganic Chemistry, Sloan-Kettering Institute for Cancer Research, 1275 York AVenue, New York,
New York 10021, and Department of Chemistry, Columbia UniVersity, HaVemeyer Hall, 3000 Broadway, New York,
New York 10027
Received April 8, 2005; E-mail: s-danishefsky@mskcc.org

Recently, Fenical and associates at the Scripps Institute of Scheme 1. Global Strategy toward Salinosporamide A
Oceanography reported on the cultivation and phylogenetic char-
acterization of a new group of actinomycete bacteria, widely
distributed in oceanic sediments.1 The term Salinospora was
advanced to correlate the strains. Following preliminary screening,
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

a highly active metabolite, termed salinosporamide A (1, Figure

1), was identified and isolated from these sediments. Salinospor-
Downloaded via Keren Cohen on February 1, 2023 at 08:02:04 (UTC).

practiced in our recent synthesis of jiadifenin,4 a structurally

differentiated malonate moiety is created with complete stereo-
Figure 1. Structures of Salinosporamide A (1) and Omuralide (2).
chemical definition. In time, the substituent at C3 is presented as
amide A displays remarkable in vitro cytotoxicity (IC50 of ap- an exo-methylene group (cf. 4 f 5). An acetaldehyde residue,
proximately 10 nM), and its activity appears to be directed to the derivable at C2, is used to differentiate the faces of this exo-
inhibition of the 20S proteasome. Thus, salinosporamide A is methylene group (cf. 5 f 6), thereby ensuring the properly
approximately 35 times more potent than is omuralide (2), which configured β-lactone moiety. Adaptation of the Corey concept in
is directed to the same molecular target. Our fascination with the context of addition of the allylic zinc reagent 83 to constrained
this target was first provoked by still another natural product, aldehyde 7 provides remarkable stereoselection at both C6 and C5.
TMC-95A, which we synthesized in a manner that allowed us to We now describe the orchestration of these general concepts en
conduct some telling SAR experiments.2 Thus, when salinospora- route to salinosporamide A. The bicyclic enamide 35 was treated
mide A came along, it seemed to us an appropriate target to broaden with divinyl cuprate under mediation by TMSCl,6 affording 9 as a
the involvement of our laboratory in the exciting field of naturally single product (Scheme 2). In a subsequent step, alkylation of 9,
occurring 20S proteasome inhibitors. as shown, furnished the lactam 11 in 77% yield as a 14:1 mixture
At this writing, there is a single reported total synthesis of of diastereomers.7 We next turned to the conversion of the vinyl
salinosporamide A (i.e., that of E. J. Corey and associates).3 A group to a carbonate ester acylating agent. Ozonolysis followed
remarkably enabling feature of that synthesis was the solution it by reductive treatment with sodium borohydride afforded 12. The
offered to what might otherwise have been a most difficult problem, derived ethyl carbonate was subjected to cleavage of the N,O-acetal
that is, that of providing stereochemical control at carbons 6 and
Scheme 2. Synthesis of Intermediate 15a
5. The coordinated Corey solution to both of these stereogenic
centers involves the action of a cyclohexenyl zinc agent with an
appropriately presented aldehyde function corresponding to C5 of
salinosporamide.
Indeed, the route described herein exploits use of the cyclohex-
enyl zinc methodology to solve the stereochemical issues at both
C5 and C6. However, we first focused on solving the internal
stereochemical issues associated with the building of the cis-fused
pyrrolidone-β-lactone ensemble.
In Scheme 1, the overall stereochemical gestalt of our program
is described. The strong facial bias of the pyroglutamate derivative,
3, served to direct attack at C3 (originally conducted by 1,4-addition a Key: (a) vinylmagnesium bromide, TMSCl, CuI, THF, -78 °C (75%);
of a vinyl cuprate nucleophile) from its R-face. Correspondingly, (b) 10,10 LDA, THF, room temperature (rt) (77%, dr ) 14:1); (c) O3,
alkylation at C2 proceeds with high selectivity from its β-face. The CH2Cl2-MeOH (3:1), -78 °C then NaBH4, 0 °C (86%); (d) ClCO2Et,
R-substituent, introduced at C3, in time is presented as a carbonate pyridine, rt (96%); (e) TfOH, THF-H2O (9:1), rt (quant); (f) Jones reagent,
acetone, rt; (g) Me2NCH(Ot-Bu)2, toluene, reflux (72% in two steps); (h)
ester. To enable the strategic C-acylation, a novel imidate ensemble Et3OBF4, K2CO3, CH2Cl2, rt (88%); (i) LHMDS, THF, -20 °C (82%); (j)
(see formal structure 4) was devised to direct lithiation to C4. 1 M HCl aq, THF, 0 °C (90%); (k) PMBCl, NaH, DMF, rt (61%); (l)
Following intramolecular acylation by the carbonate ester, as Pd(OH)2-C, H2, EtOH, rt (quant).
8298 9 J. AM. CHEM. SOC. 2005, 127, 8298-8299 10.1021/ja0522783 CCC: $30.25 © 2005 American Chemical Society
 COMMUNICATIONS

protecting arrangement to afford 13. The hydroxymethyl lactam Upon conversion of the benzyl ester to an aldehyde, intermediate
was converted to the imidate ester 4 as shown by a sequence 7 was in hand.
consisting of Jones’ oxidation, esterification, and treatment with Treatment of 7 with the cyclohexenyl zinc reagent, 8, under the
Meerwein reagent (Et3OBF4). With the lactam functionality thus Corey protocol3 proceeded with excellent diastereocontrol to afford
masked, treatment of 4 with LHMDS led to exclusive anion 19 in 88% yield (dr ) 20:1 at C6). By sharp contrast, the use of
formation at C4. Internal acylation with the pendant ethyl carbonate the corresponding imidate aldehyde derived from 1412 instead of 7
proceeded smoothly to afford lactone 14.4 Acidic treatment of 14 resulted in poor diastereoselectivity (78% yield, 4:3, configuration
led to the restoration of the lactam moiety, which was subsequently not determined). Obviously, the PMB group plays a critical role in
protected with PMBCl. Removal of the benzyl protecting group diastereoselection in the novel Corey reaction.3
afforded 15. Removal of the PMB group from 19, followed by reductive
The lactone of 15 was subjected to nucleophilic ring opening opening of the benzyl glycoside, gave rise to triol 20. Acidic
with phenylselenium anion,8 and the resultant carboxylic acid was cleavage of the tert-butyl ester was effected through treatment with
benzylated to afford the differentially esterified 16 (Scheme 3). BCl3, and the crude trihydroxy acid was then subjected to
lactonization-chlorination3 to provide 1, whose spectroscopic
properties were in complete accord with the natural material.1 In
Scheme 3. Synthesis of Salinosporamide Aa addition, the structure of fully synthetic 1 was corroborated
crystallographically.
In summary, an efficient and highly stereocontrolled enantiose-
lective synthesis of salinosporamide A has been achieved. Several
key features of our synthesis include the temporary masking of a
lactam functionality to accomplish selective anion formation at C4
(see 4), the use of a nucleophilic selenium species to open a lactone
in a regiocontrolled fashion (see 15), and the use of an unusual
cationic hemiacetal selenocyclization to install the quaternary center
at C3 in manageable form with complete stereocontrol.
Acknowledgment. This work was supported by the National
Institutes of Health (S.J.D. CA28824). A postdoctoral fellowship
a Key: (a) PhSeSePh, NaBH , EtOH, 60 °C; (b) BnBr, K CO , DMF, rt
4 2 3 is gratefully acknowledged by A.E. (Mr. William H. Goodwin and
(65% in 2 steps); (c) 30% H2O2 aq, THF, rt; (d) toluene, 100 °C (94% in
two steps, 72% 17 + 22% 5); (e) Dess-Martin periodinane, CH2Cl2, rt Mrs. Alice Goodwin and the Commonwealth Foundation for Cancer
(92%, 89% in three steps from 16); (f) PhSeBr, AgBF4, BnOH, CH2Cl2, Research, the Experimental Therapeutics Center, SKI). We also
-20 to 0 °C (74% as an anomeric mixture, 12:1); (g) AIBN, n-Bu3SnH, thank Dr. Louis Todaro (Hunter College, The City University of
toluene, 100 °C (98%); (h) NaBH4, THF-EtOH (3:1), rt (85%); (i) Dess- New York) for X-ray structural analyses and Dr. George Sukenick
Martin periodinane, CH2Cl2, rt (95%); (j) 8, THF, -78 °C (88% for 19, dr
(NMR Core Facility, SKI) for mass spectral analyses.
) 20:1); (k) ceric ammonium nitrate (CAN), CH3CN-H2O, 0 °C (90%);
(l) Na, liq NH3, -78 °C; (m) NaBH4, THF-H2O (2:1), rt (97% in two Supporting Information Available: Experimental procedures and
steps); (n) BCl3, CH2Cl2, 0 °C; (o) BOPCl, TEA, CH2Cl2, rt; (p) Ph3PCl2,
characterization, including polarimetric data, for new compounds. In
pyridine, CH3CN, rt (51% in three steps).
addition, confirmatory crystallographic data for 1 and 19 are included.
This material is available free of charge via the Internet at http://
pubs.acs.org.
Surprisingly, the subsequent selenide oxidation elimination sequence
gave rise to a mixture of the expected alcohol 17 (72%), along References
with aldehyde 5 (22%), which was in fact a one-step advancement
(1) Feling, R. H.; Buchanan, G. O.; Mincer, T. J.; Kauffman, C. A.; Jensen,
in our planned synthetic route. Upon purification, we converted P. R.; Fenical, W. Angew. Chem., Int. Ed. 2003, 42, 355.
the bulk unoxidized material, 17, to aldehyde 5 through exposure (2) (a) Lin, S.; Danishefsky, S. J. Angew. Chem., Int. Ed. 2002, 41, 512. (b)
to Dess-Martin periodinane.9 Yang, Z.-Q.; Kwok, B. J. B.; Lin, S.; Koldobskiy, M. A.; Crews, C. M.;
Danishefsky, S. J. Chembiochem 2003, 4, 508. (c) Lin, S.; Yang, Z.-Q.;
With intermediate 5 in hand, the stage was now set for a key Kwok, B. H. B.; Koldobskiy, M.; Crews, C. M.; Danishefsky, S. J. J.
acetal-mediated cationic cyclization.11 We note that electrophilically Am. Chem. Soc. 2004, 126, 6347.
(3) Reddy, L. R.; Saravanan, P.; Corey, E. J. J. Am. Chem. Soc. 2004, 126,
induced cyclization at the aldehyde (or hemiacetal) oxidation level 6230.
was central to the success of the project. Presumably, a tetrahy- (4) Cho, Y. S.; Carcache, D. A.; Tian, Y.; Li, Y.-M.; Danishefsky, S. J. J.
Am. Chem. Soc. 2004, 126, 14358.
drofuran derived from haloetherification could not have been readily (5) (a) Thottathil, J. K.; Moniot, J. L.; Mueller, R. H.; Wong, M. K. T.; Kissick,
opened to expose the required functionalities at C2 and C3. T. P. J. Org. Chem. 1986, 51, 3140. (b) Hamada, Y.; Kawai, A.; Kohno,
Y.; Hara, O.; Shioiri, T. J. Am. Chem. Soc. 1989, 111, 1524. (c) Hamada,
Conversely, selenolactonization using an acetic acid residue at C2 Y.; Hara, O.; Kawai, A.; Kohno, Y.; Shioiri, T. Tetrahedron 1991, 47,
would have produced a lactone that would not be readily dif- 8635.
(6) Corey, E. J.; Boaz, N. W. Tetrahedron Lett. 1985, 26, 6019.
ferentiable from the bis-acyl functionality already present at C4. (7) For earlier studies of stereocontrolled access to all-trans R,β,γ-substituted
Thus, recourse to the benzyl glycoside modality for storing and pyrrolidinone from 3, see: (a) Hanessian, S.; Ratovelomanana, V. Synlett
1990, 501. (b) Baldwin, J. E.; Moloney, M. G.; Shim, S. B. Tetrahedron
unveiling the C2-C3 functionality was a unique solution to a Lett. 1991, 32, 1379. (c) Okamoto, N.; Hara, O.; Makino, K.; Hamada,
difficult problem. Upon treatment with phenylselenenyl bromide Y. Tetrahedron: Asymmetry 2001, 12, 1353.
(8) Scarborough, R. M.; Toder, B. H.; Smith, A. B. J. Am. Chem. Soc. 1980,
and AgBF4 in the presence of benzyl alcohol, an intermediate 102, 3904.
hemiacetal was generated, which presumably assisted in the (9) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1982, 104, 902.
phenylselenenylation of the exocyclic methylene to afford 18. (10) Berlage, U.; Schmidt, J.; Peters, U.; Welzel, P. Tetrahedron Lett. 1987,
28, 3091
Importantly, this reaction allowed for the introduction of the (11) Current, S.; Sharpless, K. B. Tetrahedron Lett. 1978, 51, 5075.
quaternary center at C3 with complete stereoselectiVity. Radical (12) See Supporting Information for method of synthesis.
deselenenylation provided the desired methyl functionality at C3. JA0522783

J. AM. CHEM. SOC. 9 VOL. 127, NO. 23, 2005 8299