A formal synthesis of (+)-lactacystin

Article abstract of DOI:10.1039/b508300a

A synthetic route to the neurotrophic agent (+)-lactacystin has been developed utilizing a base-promoted intramolecular alkylidenecarbene C-H insertion as the key transformation. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b508300a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
A formal synthesis of (+)-lactacystin{
Duncan J. Wardrop* and Edward G. Bowen
Received (in Bloomington, IN, USA) 13th June 2005, Accepted 9th August 2005
First published as an Advance Article on the web 19th September 2005
DOI: 10.1039/b508300a
The formation of 3-pyrrolines via alkylidenecarbene C–H
A synthetic route to the neurotrophic agent (+)-lactacystin has
been developed utilizing a base-promoted intramolecular
alkylidenecarbene C–H insertion as the key transformation.
insertion was first reported in 1970 by Walsh and Bottini, who
found that treatment of simple 3-bromopropenyldialkyl amines
with potassium tert-butoxide afforded 3-pyrrolines in low yield.7a
While Gilbert,7b Shiori7c and Baird7d have subsequently reported
similar cyclizations employing alternative strategies for carbene
generation, application of this transformation to natural product
synthesis has, until recently, been limited, despite its obvious
potential. While the work presented therein was in progress, Hayes
and co-workers reported a related route to 1 using alkylidene-
carbene C–H insertion.8
(+)-Lactacystin (1), a secondary metabolite of the bacterium
Streptomyces sp. OM-6519, was discovered by Omura during a
screen for natural product inhibitors of mouse neuronal cell
differentiation.1 Studies by Fenteany and co-workers2 have
revealed that lactacystin covalently modifies and irreversibly-
inhibits the 20S proteasome, whose function within the cell is to
process and eliminate damaged, misfolded and ubiquitinated
proteins. The universal presence of this enzyme complex ensures
that 1 exhibits a host of biological activities that, in conjunction
with its complex structure, have stimulated considerable synthetic
interest.3 Herein we report the asymmetric synthesis of compound
2, which Smith and co-workers4 have previously converted to the
natural product 1 in two steps (Scheme 1).5
Our route to 2 commenced (Scheme 2) with the reduction of
a,b-unsaturated ester 4 using alane (AlH3). Sharpless asymmetric
epoxidation of the resulting allylic alcohol then provided epoxide
5.9 Acid-catalyzed ring-opening of 5 with sodium azide in aqueous
2-methoxyethanol gave an inseparable 2 : 1 mixture of 6 and the
1,2-diol arising from C-3 ring opening.10 Treatment of this mixture
with aqueous sodium periodate selectively cleaved the 1,2-diol to
the corresponding a-azido aldehyde, which was then separated by
chromatography on silica gel. Protection of the 1,3-diol moiety of
6 as a benzylidene acetal and chemoselective catalytic hydrogena-
tion of the azide over Pd(CaCO3) then provided amine 7 as a single
diastereomer. Alkylation of 7 with bromoacetone under standard
conditions gave 8, which was found to be unstable, decomposing
within a matter of hours at room temperature (Scheme 2).
Nonetheless, when this a-amino ketone was immediately treated
with lithium (trimethylsilyl)diazomethane11 at 278 uC, the desired
While many of the existing routes to 1 utilize aldol chemistry to
set the key quaternary C-5 stereocenter, we envisioned that this
goal could be achieved through intramolecular C–H insertion of
the alkylidenecarbene intermediate 3. Since alkylidene insertion is
known to proceed with retention of configuration,6 we would
parlay the asymmetry of an existing and readily accessible tertiary
center into a quaternary carbon center. With regards to the C-6
and C-7 stereocenters, we anticipated that the equatorially-
positioned isopropyl group of the 1,3-dioxane ring of 9 would
effectively shield the a-face of the 3-pyrroline ring and ensure
that dihydroxylation or epoxidation proceeded exclusively from
the less hindered face to form 12, thereby setting the correct
stereochemistry at C-6.
Scheme 2 Reagents and conditions. a: AlH3, Et2O, 0 uC, 1 h (85%);
b: (+)-DIPT, Ti(OiPr)4, CH2Cl2, 4 s sieves, 220 uC, 7 h (76%, 96% ee);
c: (i) NaN3, NH4Cl, MeOCH2CH2OH, H2O, reflux, 14 h, (ii) NaIO4,
THF, H2O, rt, 24 h (47%); d: PhCHO, p-TsOH?H2O, PhCH3, reflux, 1 h
(78%); e: H2 (1 atm), Pd/CaCO3, EtOH, rt, 3 h (98%); f: bromoacetone,
K2CO3, acetone, reflux, 7 h (56%); g: Me3SiCLiN2, THF, 278 uC, 1 h
(35%). h: 1,3-dibromo-2-methyl-propene (E : Z 5 3 : 2), K2CO3, CH3CN,
reflux, 16 h (76%).
Scheme 1 Retrosynthetic analysis of (+)-lactacystin (1).
Department of Chemistry, University of Illinois at Chicago, Chicago,
USA. E-mail: wardropd@uic.edu; Fax: +1 312 996 0431;
Tel: +1 312 355 1035
{ Electronic supplementary information (ESI) available: Experimental
procedures and characterisation data for all compounds prepared. See
http://dx.doi.org/10.1039/b508300a
5106 | Chem. Commun., 2005, 5106–5108
This journal is ß The Royal Society of Chemistry 2005

View Article Online
3-pyrroline 9 was isolated, albeit in a rather modest yield. Analysis
of the 2D-NOESY spectrum of 9 revealed correlations between
Hax-6, Hax-9 and the amine proton, thus confirming that insertion
proceeded with retention of configuration.
Encouraged by this preliminary result but suspecting that
further pursuit of a-amino ketone substrates would be fraught
with difficulties, we now decided to examine a-elimination as an
alternate method for alkylidenecarbene generation. Thus, heating
an equimolar mixture of 7 and 1,3-dibromo-2-methyl-propene
(E : Z 5 3 : 2)12 in acetonitrile, generated the corresponding
mixture of alkenyl bromides 10, which proved to be stable
(Table 1). Upon treatment of an ethereal solution of 10 with
potassium hexamethyldisilazide (KHMDS), cyclization proceeded
rapidly at room temperature to provide 3-pyrroline 9 and
propargyl amine 11, which arise through a competitive 1,2-
rearrangement of the intermediate carbene (Table 1, entry 1). In an
effort to ascertain the effect of alkenyl bromide geometry upon
insertion, the individual isomers of 10 were separated by careful
flash chromatography and submitted to the same cyclization
conditions as before (Table 1, entries 2 and 3). Intriguingly, there
appears to be a correlation between the efficiency of the insertion
and the geometry of the alkenyl bromide precursor, with E-10
giving 9 in 67% yield. This observation is rather surprising, given
that Taber has previously demonstrated, as least for carbocyclic
systems, that the geometry of the alkenyl halide precursor does not
influence the efficiency of C–H insertion.13
Scheme 3 Reagents and conditions. a: (Boc)2O, NH2OH?HCl, Et3N,
CH2Cl2, rt, 96 h (93%); b: m-CPBA, BHT, CH2Cl2, rt, 18 h (90%); c:
BF3?Et2O, CH2Cl2, rt, 15 min (90%); d: H2 (2300 psi), Pd(OH)2, EtOAc,
rt, 24 h (94%); e: (i) TEMPO, trichloroisocyanuric acid, CH2Cl2, rt, 2 h, (ii)
Having established the C-5 center, we now turned our attention
to installing the remaining stereocenters and adjusting the
oxidation state at C-4 (Scheme 3). In view of the sensitivity of 9
to aerial oxidation, we opted now to acylate the nitrogen atom.
However, introduction of N-Boc protection proved to be rather
troublesome, primarily because of the highly hindered nature of
this substrate. The most effective conditions identified involved
treating 9 with tert-butyl aminocarbonate, generated in situ from
(Boc)2O and hydroxylamine.14 Under these conditions,
N-acylation was slow but ultimately proceeded in excellent yield.
Epoxidation with m-CPBA in the presence of 2,6-di-tert-butyl-4-
methylphenol (BHT)15 then proceeded smoothly to furnish 12.
Attempts to accomplish this transformation without a radical
inhibitor failed to provide any epoxide, but instead generated the
a,b-unsaturated lactam resulting from allylic oxidation at C-2.
That epoxidation of the pyrroline ring had proceeded from the less
hindered face could not be immediately determined because of the
t
NaClO2, BuOH, 2-methyl-2-butene, 0 uC, 1 h, (iii) CH2N2, Et2O, rt,
10 min (96%); f: Me3SiCl, imidazole, DMF, rt, 4 h (94%); g: LDA, THF,
0 uC, 2 h (93%); h: Me3SiCl, imidazole, DMF, rt, 1 h (85%); i: NBS, H2O,
1,4-dioxane, rt, 3 h (95%); j: PDC, DMF, rt, 4 h, then: Mg(ClO4)2,
CH3CN, 40 uC, 1 h (67%); k: NH4F, MeOH, reflux, 1 h (95%); l: SmI2,
THF, rt, 15 min (80%, 1 : 2).
presence of amide rotamers in the 1H NMR spectrum.
Accordingly, the N-Boc group was removed under mild conditions
(BF3?Et2O)16 to provide the corresponding amine 13 as a single
diastereomer. Analysis of this material by
a 2D-NOESY
experiment revealed correlations between H-5 and the methyl
protons of the isopropyl group, thereby confirming that epoxida-
tion had indeed proceeded from the less hindered face of 12.
The benzylidene acetal of 12 was now removed by hydro-
genolysis over Adam’s catalyst [Pd(OH)2] at high pressure and the
resulting 1,3-diol converted to ester 14. Thus, selective oxidation of
the primary alcohol using TEMPO and trichloroisocyanuric acid17
provided the corresponding b-hydroxy aldehyde in excellent yield.
This compound proved to be quite unstable and accordingly was
immediately converted to the corresponding ester through Pinnick
oxidation and methylation with diazomethane. The secondary
hydroxyl group was then protected as a trimethylsilyl ether and the
epoxide ring opened using LDA to afford the corresponding
hydroxy enamide, which was then silylated to furnish 15. That the
exocyclic olefin product arising from deprotonation at the methyl
group was not observed in the ring opening reaction is likely a
reflection of the chelational and electronic directing effects of
the carbamate group.18 Hydrobromination of 15 with
N-bromosuccinamide (NBS) in aqueous 1,4-dioxane now pro-
ceeded with complete regioselectivity to form a mixture of
a-bromo carbinolamine diastereomers. After screening a number
Table 1 2-Pyrrolines via alkylidenecarbene C–H insertiona
Substrate
Yield of 9 (%)b
Yield of 11 (%)b
(E/Z)-10
(E)-10
(Z)-10
a
50
67
14
36
25
42
A solution of 10 in Et2O was added to 1.5 equiv. of KHMDS in
Et2O and the resulting mixture stirred at ambient temperature for
15 min. Yield of isolated product after flash chromatography.
b
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 5106–5108 | 5107

View Article Online
6 (a) D. F. Taber, R. P. Meagley and D. D. Doren, J. Org. Chem., 1996,
61, 5723; (b) Reviews covering the 1,5-C–H insertion of alkylidene-
carbenes: P. J. Stang, Chem. Rev., 1978, 78, 383; P. J. Stang, Angew.
Chem., Int. Ed. Engl., 1992, 31, 274; D. F. Taber, in Methods of Organic
Chemistry, ed. G. Helmchen, Thieme, Stuttgart, 1995, vol. E21, pp. 1127;
W. Kirmse, Angew. Chem., Int. Ed. Engl., 1997, 36, 1164.
7 (a) R. A. Walsh and A. T. Bottini, J. Org. Chem., 1970, 35, 1086; (b)
J. C. Gilbert and B. K. Blackburn, J. Org. Chem., 1986, 51, 3656; (c)
H. Ogawa, T. Aoyama and T. Shioiri, Heterocycles, 1996, 42, 75; (d)
M. S. Baird, A. G. W. Baxter, A. Hoorfar and I. Jefferies, J. Chem.
Soc., Perkin Trans. 1, 1991, 2575.
8 (a) M. P. Green, J. C. Prodger, A. E. Sherlock and C. J. Hayes, Org.
Lett., 2001, 3, 3377; (b) M. P. Green, J. C. Prodger and C. J. Hayes,
Tetrahedron Lett., 2002, 43, 6609.
9 Y. Gao, J. M. Klunder, R. M. Hanson, H. Masamune, S. Y. Ko and
K. B. Sharpless, J. Am. Chem. Soc., 1987, 109, 5765.
10 M. Caron, P. R. Carlier and K. B. Sharpless, J. Org. Chem., 1988, 53,
5185.
11 S. Ohira, K. Okai and T. Moritani, J. Chem. Soc., Chem. Commun.,
1992, 721.
12 W. Fischetti, K. T. Mak, F. G. Stakem, J. Kim, A. L. Rheingold and
R. Heck, J. Org. Chem., 1983, 48, 948.
13 D. F. Taber and T. D. Neubert, J. Org. Chem., 2001, 66, 143.
14 R. B. Harris and I. B. Wilson, Tetrahedron Lett., 1983, 24, 231.
15 Y. Kishi, M. Aratani, H. Tanino, T. Fukuyama, T. Goto, S. Inoue,
S. Sugiua and H. Kakoi, J. Chem. Soc., Chem. Commun., 1972, 64.
16 E. F. Evans, N. J. Lewis, I. Kapfer, G. Macdonald and R. J. K. Taylor,
Synth. Commun., 1997, 27, 1819.
of oxidizing reagents, pyridinium dichromate was found to most
successfully mediate the transformation to the corresponding
a-bromolactam. Removal of the N-Boc group, to form 16, was
then accomplished under mild conditions, using magnesium
perchlorate.19 Removal of the silyl ethers was effected by NH4F
in methanol to give diol 17. Finally, reduction of the a-bromo ester
with samarium diiodide gave a mixture of C-4 epimers 2 and 18
(1 : 2) which were separated to give 2. The spectral and physical
properties of 220 were in accordance with those previously
reported.4 This compound has been previously been converted
to 1 by Smith4, and hence the route described above represents a
formal, total synthesis of (+)-lactacystin.
In summary, we have developed a novel synthetic route
to (+)-lactacystin, based on the [1,5]-C–H insertion of
alkylidenecarbenes.
We thank the National Institutes of Health (GM67176) and the
University of Illinois at Chicago for their financial support of this
work.
Notes and references
1 S. Omura, K. Matsuzaki, T. Fujimoto, K. Kosuge, T. Furuya, S. Fujita
and A. Nakagawa, J. Antibiot., 1991, 44, 117.
17 L. De Luca, G. Giacomelli and A. Porcheddu, Org. Lett., 2001, 3,
3041.
2 G. Fenteany, R. F. Standaert, W. S. Lane, S. Choi, E. J. Corey and
S. L. Schreiber, Science, 1995, 268, 726.
18 P. Beak, A. Basu, Y. S. P. Gallagher and S. Thayumanavan, Acc. Chem.
Res., 1996, 29, 552.
19 J. A. Stafford, M. F. Brackeen, D. S. Karanewsky and N. L. Valvano,
Tetrahedron Lett., 1993, 34, 7873.
20 Selected physical and spectroscopic data for 2: [a]2D5 5 +60.0 (c 0.2 in
CH3OH), lit.4 [a]D24 5 +65.0 (c 0.3 in CH3OH). dH (400 MHz, CD3OD):
4.43 (1 H, d, J 5 6.2 Hz), 3.90 (1 H, d, J 5 7.0 Hz), 3.73 (3 H, s), 3.00–
2.90 (1 H, m), 1.69–1.61 (1 H, m), 1.06 (3 H, d, J 5 7.6 Hz), 0.97 (3 H, d,
J 5 7.0 Hz) and 0.83 (3 H, d, J 5 7.0 Hz). dC (100 MHz, CD3OD):
182.3, 172.9, 80.1, 76.9, 76.8, 52.5, 42.5, 32.3, 20.1, 19.5 and 8.8. HRMS-
EI: Found 245.12350, C11H19NO5 [M]+ requires 245.12363.
3 (a) For an excellent review, see: C. E. Masse, A. J. Morgan, J. Adams
and J. S. Panek, Eur. J. Org. Chem., 2000, 2513; (b) P. C. B. Page,
A. S. Hamzah, D. C. Leach, S. M. Allin, D. M. Andrews and
G. A. Rassias, Org. Lett., 2003, 5, 353; (c) C. J. Brennan, G. Pattenden
and G. Rescourio, Tetrahedron Lett., 2003, 44, 8757.
4 T. Nagamitsu, T. Sunazuka, H. Tanaka, S. Omura, P. A. Sprengeler
and A. B. Smith, J. Am. Chem. Soc., 1996, 118, 3584.
5 Presented in part at the 222nd National Meeting of the American
Chemical Society: D. J. Wardrop and E. G. Bowen, Abstracts of Papers,
200th National Meeting of the American Chemical Society, Division of
Organic Chemistry, Chicago, IL, August 26–30, 2001, abstract no. 200.
5108 | Chem. Commun., 2005, 5106–5108
This journal is ß The Royal Society of Chemistry 2005