Enantioselective total syntheses of belactosin A, belactosin C, and its homoanalogue

Article abstract of DOI:10.1021/ol049409+

Enantioselective total syntheses of belactosin A, belactosin C, and its homoanalogue have been accomplished in high overall yields (32% for belactosin A from the amino acid 10, and 35 and 36% for belactosin C and its homoanalogue, respectively). This conc

Full text of DOI:10.1021/ol049409+

ORGANIC
LETTERS
2004
Vol. 6, No. 13
2153-2156
Enantioselective Total Syntheses of
Belactosin A, Belactosin C, and Its
Homoanalogue†
Oleg V. Larionov and Armin de Meijere*
Institut fu¨r Organische und Biomolekulare Chemie der Georg-August-UniVersita¨t
Go¨ttingen, Tammannstrasse 2, D-37077 Go¨ttingen, Germany
armin.demeijere@chemie.uni-goettingen.de
Received March 31, 2004
ABSTRACT
Enantioselective total syntheses of belactosin A, belactosin C, and its homoanalogue have been accomplished in high overall yields (32% for
belactosin A from the amino acid 10, and 35 and 36% for belactosin C and its homoanalogue, respectively). This concise approach comprises
a novel sequential acylation/â-lactonization reaction and allows a facile alteration of the substituents, thus providing a flexible route to a new
family of highly active belactosin-based proteasome inhibitors.
Belactosins A (1) and C (2a) (Figure 1) were isolated from
a fermentation broth of Streptomyces sp. UCK14 and
exhibited antitumor activity,1 which was shown to be
increased significantly upon acetylation of the free amino
group and esterification or amidation of the carboxyl group,
as well as displacement of the ornithine moiety in 2a with
lysine to give 2b, thus providing IC50 against human
pancreoma and colon cancer on as low as the nanomolar
level.2 More intriguingly, the high antitumor activities of
these derivatives appeared to be attributed to the proteasome
inhibition.3 This powerful mechanism of cell growth and
* Fax: +49-(0)551/399475.
† Part 99 in the series Cyclopropyl Building Blocks for Organic Synthesis.
For Part 98, see: de Meijere, A.; Bagutski, V.; Zeuner, F.; Fischer, U. K.;
Rheinberger, V.; Moszner, N. Eur. J. Org. Chem. 2004, in press. For Part
97, see: Wiedemann, S.; Rauch, K.; Savchenko, A.; Marek, I.; de Meijere,
A. Eur. J. Org. Chem. 2004, 631-635.
(1) (a) Mizukami, T.; Asai, A.; Yamashita, Y.; Katahira, R.; Hasegawa,
A.; Ochiai, K.; Akinaga, S. U.S. Patent 5 663 298, 1997; Chem. Abstr.
1997, 126 (26), 79. (b) Asai, A.; Hasegawa, A.; Ochiai, K.; Yamashita, Y.;
Mizukami, T. J. Antibiotics 2000, 53, 81-83. (c) Yamaguchi, H.; Asai,
A.; Mizukami, T.; Yamashita, Y.; Akinaga, S.; Ikeda, S.-i.; Kanda, Y. EP
Patent 1 166 781 A1, 2000; Chem. Abstr. 2000, 133 (9), 751.
Figure 1. Belactosin A (1), belactosin C (2a), and its homo-
analogue 2b.
10.1021/ol049409+ CCC: $27.50 © 2004 American Chemical Society
Published on Web 05/27/2004

death control relies on the basic role of the proteasome in
most cellular processes that are mediated by small peptide
molecules produced by the proteasome.3 Therefore, further
studies of these substances exhibiting such remarkable
biological activities4 can ultimately lead to the development
of new treatment options against cancer and inflammatory
diseases.
propyl)alanine 5, which has recently attracted considerable
attention.6 An additional challenge was to develop an
approach that will tolerate a variety of substituents in the
amino- and carboxyl-terminal amino acid moieties of the
dipeptide 3, thus implying that the final acetylation of the
side chain amino group of 3 and subsequent â-lactone ring
construction should be performed under very mild conditions.
First, the dipeptide components 3a and 3b were synthe-
sized in four steps (Scheme 2). Benzylation of the com-
In an effort to establish a flexible access to the belactosins
and their derivatives, we embarked on a modular approach,
starting from four distinct building blocks (Scheme 1). Their
Scheme 2. Synthesis of the Dipeptides 3a and 3ba
Scheme 1. Retrosynthetic Analysis of Belactosin A
a Conditions: (a) CbzCl, DMAP, DIEA, CH2Cl2, 0 °C, 6 h; (b)
Et2NH, THF, rt, 3 h; (c) Cbz-Ala-OH, EDC, HOAt, TMP, CH2Cl2,
0 °C to rt, 16 h; (d) 3 M HCl in EtOAc, iPr3SiH, rt, 17 h.
mercial diprotected ornithine and lysine derivatives 7 with
Cbz-Cl/DIEA/DMAP furnished the esters 8, which after
removal of the Fmoc group, were coupled with Cbz-Ala-
OH. The Boc groups were cleaved off the resulting fully
protected dipeptides 9 by treatment with 3 M hydrochloric
acid to afford the hydrochlorides 3a and 3b in yields of 91
and 94%, respectively, over four steps.
variation can give rise to a large set of analogues, and their
efficient synthesis can provide a solid basis for further studies
of their biological activities.
The attempted preparation of the diprotected trans-(2-
aminocyclopropyl)alanine derivative 7c as a prerequisite for
the synthesis of belactosin A (1) by the recently published
two-step general procedure,7 which proved to be feasible for
large-scale preparations of the analogous lysine and ornithine
derivatives 7a and 7b, gave only a very low yield (27%) of
7c (Scheme 2). This must be attributed to the experimental
difficulties inevitably encountered on applying that procedure
to subgram quantities of the enantiomerically pure trans-(2-
aminocyclopropyl)alanine (5). To get around these difficul-
ties, advantage was taken of the intrinsic differentiation of
the two nitrogen atoms in trans-(2-nitrocyclopropyl)alanine
(10), an established precursor to 5.6b Therefore, compound
10 was converted into the respective Boc derivative 11 (84%
yield). The nitro group in 11 was then reduced to an amino
group. At first, this also presented a problem, since hydro-
genation of 11 over Pd/C in methanol was shown to give
rise to an extensive reductive cyclopropane ring cleavage.8
The belactosin compounds comprise an interesting trans-
disubstituted â-lactone5 with an adjacent stereogenic center
in the side chain, thus calling for a special anti-selective
aldol-addition strategy. Additionally, belactosin A contains
the new cyclopropane amino acid trans-(2-aminocyclo-
(2) Asai, A.; Tsujita, T.; Sharma, S. V.; Yamashita, Y.; Akinaga, S.;
Funakoshi, M.; Kobayashi, H.; Mizukami, T., Asahi-machi, M.-s. Biochem.
Pharmacol. 2004, 67, 227-234.
(3) (a) Gillessen, S.; Groettrup, M.; Cerny, T. Onkologie 2002, 25, 534-
539. (b) Almond, J. B.; Cohen, G. M. Leukemia 2002, 16, 433-443. (c)
Elliott, P. J.; Zollner, T. M.; Boehncke, W.-H. J. Mol. Med. 2003, 81, 235-
245. (d) A few weeks after submission of this manuscript, a very recent
publication on the total synthesis of belactosin A came to our attention:
Armstrong, A.; Scutt, J. N. Chem. Commun. 2004, 510-511.
(4) For some recent advances in isolation and syntheses of â-lactone
proteasome inhibitors, see: (a) Corey, E. J.; Li, W. D. Z. Chem. Pharm.
Bull. 1999, 47, 1-10. (b) Masse, C. E.; Morgan, A. J.; Adams, J.; Panek
J. S. Eur. J. Org. Chem. 2000, 14, 2513-2528. (c) Crane, S. N.; Corey, E.
J. Org. Lett. 2001, 3, 1395-1397. (d) Feling, R. H.; Buchanan, G. O.;
Mincer, T. J.; Kauffman, C. A.; Jensen P. R.; Fenical, W. Angew. Chem.,
Int. Ed. 2003, 42, 355-357. (e) Saravanan, P.; Corey, E. J. J. Org. Chem.
2003, 68, 2760-2764. (f) Brennan, C. J.; Pattenden, G.; Rescourio, G.
Tetrahedron Lett. 2003, 44, 8757-8760. (g) Berthelot, A.; Piguel, S.; Le
Dour, G.; Vidal, J. J. Org. Chem. 2003, 68, 9835-9838.
(6) (a) Armstrong, A.; Scutt, J. N. Org. Lett. 2003, 5, 2331-2334. (b)
Larionov, O. V.; Kozhushkov, S. I.; Brandl, M.; de Meijere, A. MendeleeV
Commun. 2003, 5, 199-200. (c) Jain, R. P.; Vederas, J. C. Org. Lett. 2003,
5, 4669-4672.
(7) Masiukiewicz, E.; Wiejak, S.; Rzeszotarska, B. Org. Prep. Proced.
Int. 2002 34, 531-537 and references therein.
(8) Zlatopolskiy, B. Dissertation, Georg-August-Universita¨t, Go¨ttingen,
Germany, 2003.
(5) Recent enantioselective approaches to â-lactones: (a) Yang, H. W.;
Romo, D. Tetrahedron 1999, 55, 6403-6434. (b) Nelson, S. G.; Zhu, C.;
Shen, X. Q. J. Am. Chem. Soc. 2004, 126, 14-15. (c) Schmidt, J. A. R.;
Mahadevan, V.; Getzler, Y. D. Y. L.; Coates, G. W. Org. Lett. 2004, 6,
373-376.
2154
Org. Lett., Vol. 6, No. 13, 2004

Scheme 3. Synthesis of the Dipeptide 3ca
Scheme 4. Synthesis of the â-Lactone Building Blocka
a Conditions: (a) (i) CuSO4, NaHCO3, Boc2O, H2O, Me2CO, rt,
48 h; (ii) Fmoc-OSu, 8-quinolinol, Na2CO3, rt, 3 h, 27% (over two
steps). (b) Boc2O, Na2CO3, 6 N aq KOH, aq dioxane, rt then 35
°C, 30 h, 98%. (c) Zn, AcOH, rt, 3 h. (d) Fmoc-OSu, Na2CO3,
H2O, DMF, rt, 3 h, 75% (over two steps). (e) Cbz-Cl, DMAP,
DIEA, CH2Cl2, 0 °C, 6 h, 81%. (f) TFA, iPr3SiH, CH2Cl2, rt, 4 h.
(g) Cbz-Ala-OH, EDC, HOAt, TMP, CH2Cl2, 0 °C to rt, 16 h, 94%
(over two steps). (h) Et2NH, THF, rt, 3 h.
a Conditions: (a) H2NOSO3H, NaOH, H2O, 0 °C, 6 h, reflux, 3
h, 84%; (b) DCC, PhSH, DMAP, CH2Cl2, 0 °C to rt, 7 h, 97%; (c)
LiTMP, Me3SiCl, THF, -78 °C, 16 h, 92%; (d) EtO2CCHO,
Sn(OTf)2 (10 mol %)/BOX (11 mol %), CH2Cl2, -78 °C, 20 h,
then 2 N HCl, THF, rt, 2 h, 99%; (e) 10% aq HCl, dioxane (1:6),
60 °C, 51 h, 74%.
This overhydrogenation proved to be difficult to circum-
vent even by employing more basic (e.g., DMF) or less polar
(e.g., EtOAc) solvents. On the other hand, the reduction of
an oxime containing a cyclopropane ring, to an amine, could
be efficiently accomplished by Zn dust in AcOH, even when
the hydrogenation led to reductive ring destruction.9 Gratify-
ingly, the reduction of the nitro group in 11 with Zn in AcOH
proceeded cleanly, to give the aminocyclopropyl intermedi-
ate,10 which was immediately transformed into the Fmoc-
protected derivative 12 (75% over two steps). This underwent
smooth benzylation at the carboxyl group to give the ester
13, which, after acid-catalyzed removal of the Boc group,
was in turn coupled with Cbz-Ala-OH (94% over two steps).
The Fmoc group in the dipeptide 14 was then cleaved off
by treatment with diethylamine to give the desired amino
dipeptide 3c with a terminal amino group.
Considering different approaches to the â-lactone com-
ponent 4, the necessity to establish the anti configuration
along with that of the adjacent side chain stereogenic center,
bearing substituents of a very similar size (i.e., methyl and
ethyl), had to be kept in mind. This makes the construction
of the remote stereogenic center via common synthetic
protocols unfavorable. Employing L-isoleucine (15) allows
all the merits of a chiral pool compound as a starting material
to be enjoyed. L-Isoleucine (15) was hydrodeaminated11 with
hydroxylamine-O-sulfonic acid in 84% yield (Scheme 4).
With the acid 6 in hand, the construction of 4 was eventually
accomplished. A number of noncatalytic anti-selective aldol
reaction protocols were tested, but they provided neither
satisfying yields nor diastereo- and enantioselectivities.
Attention was therefore turned to the Mukaiyama-type
protocol for an efficient catalytic asymmetric aldol reaction
as developed by Evans et al.12 The acid 6 was converted
into the phenyl thioester 16, which was subsequently
transformed into the (Z)-silylketene acetal 17. This was then
employed in the Mukaiyama-type aldol reaction with ethyl
glyoxalate, using Sn(OTf)2/BOX as the catalytic system. This
procedure afforded the substituted malic acid derivative 18
in an excellent yield (99%) and with a high degree of enantio-
(99% ee) and diastereoselectivity (syn:anti > 40:1)13 even
on a multigram scale.
To be able to proceed with the assembly of the belactosin
backbone, the ethyl ester group in 18 had to be cleaved
without affecting the phenyl thioester moiety. This did not
(12) Evans, D. A.; MacMillan, D. W. C.; Campos, K. R. J. Am. Chem.
Soc. 1997, 119, 10859-10860.
(13) Absolute configuration of 18 was unambiguously established by
an X-ray crystallographic analysis of the amide 20, which was prepared
from 18 and L-phenylethylamine.
(9) Larionov, O. V.; de Meijere, A. Manuscript in preparation.
(10) Another successful example of selective reduction of nitro- to
aminocyclopropanes by Zn in HCl has recently been described: Wurz, R.
P.; Charette, A. B. J. Org. Chem. 2004, 69, 1262-1269.
(11) Doldouras, G. A.; Kollonitsch, J. J. Am. Chem. Soc. 1978, 4, 341-
342.
For details, see: Larionov, O. V.; de Meijere, A.; Yufit, D. S.; Howard, J.
A. K. Acta Crystallogr. E 2004, 60, o681-o683.
Org. Lett., Vol. 6, No. 13, 2004
2155

Scheme 5. Final Assembly of the Belactosin Backbonea
a Conditions: (a) EDC, HOAt, TMP, CH2Cl2, -30 °C, 6 h, rt, 30 h, 74%; (b) H2 (1 atm), Pd/C, AcOH, rt, 15 h.
appear to be trivial, since in the majority of described cases,
thioester groups usually are highly selectively hydrolyzed
in the presence of O-ester groups. An attempted enzymatic
hydrolysis of the ethyl ester group of 18 under neutral
conditions completely failed, probably due to the high
hydrophobicity of the molecule. Fortunately, the desired
selective ethyl ester cleavage succeeded under acidic condi-
tions (10% aq HCl in dioxane at 60 °C), furnishing the acid
19 in 74% yield. It is noteworthy that strict temperature
control (i.e., (3 °C) is required to achieve a high yield, since
at lower temperatures the hydrolysis is too slow, whereas at
higher temperatures retro-aldol decomposition occurs.
Subsequent condensations of the acid 19 with the dipep-
tides 3a,b were envisaged to yield the thioesters 21 (Scheme
5), which were then supposed to give the â-lactone products
22 upon activation with thiophilic metal salts (CuI, AgI,
HgII),14 or, if this would fail, hydrolytic removal of the
phenylthio group, followed by â-lactonization, to furnish the
desired diprotected belactosin derivatives 22a,b. Quite
unexpectedly, the reaction of 19 with 3a under the peptide
coupling conditions (EDC/TMP/HOAt) directly led to the
â-lactone products 22 in 62-73% yields. This appears to
be the first example of such a domino-type acylation/â-
lactonization sequence. However, this reaction may actually
follow a more complicated pathway than just acylation with
subsequent â-lactone ring closure, since an attempt to effect
the â-lactonization of 18 under the same conditions failed.
A more detailed study of this transformation is currently
underway. The benzyl and Cbz groups of the â-lactone
compounds 22 were then cleaved off by hydrogenolysis in
AcOH, thus affording the three members 1 and 2a,b of the
belactosin family.
In conclusion, we have accomplished the first enantio-
selective total syntheses of belactosin C (2a) and its
homoanalogue (2b), along with a new total synthesis of
belactosin A (1). Our convergent approach comprises only
11 steps (overall yields of 35 and 36%, respectively) in the
cases of 2a and 2b and 14 steps (overall yield of 32% from
10) for 1 and involves a new domino-type acylation/â-
lactonization sequence, which may be employed in syntheses
of further â-lactone carboxamides, derived from carboxyl-
substituted â-hydroxy acids.
Acknowledgment. This work was supported by the
Deutsche Forschungsgemeinschaft (SFB 416, Project A3),
the Fonds der Chemischen Industrie, and by an INTAS grant
of the European Union (Project INTAS-2000-0549). O.V.L.
is indebted to the Degussa-Stiftung (Degussa AG) for a
graduate fellowship. The authors are grateful to the compa-
nies BASF AG, Bayer AG, Chemetall GmbH, and Degussa
AG for generous gifts of chemicals and to Dr. B. Knieriem,
Go¨ttingen, for his careful proofreading of the final manu-
script.
Supporting Information Available: Experimental pro-
cedures and full characterization for selected compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(14) Chan, W. K.; Masamune, S.; Spessard, G. O. Org. Synth. 1983, 61,
48-55.
OL049409+
2156
Org. Lett., Vol. 6, No. 13, 2004