Total synthesis and absolute configuration of (-)-berkeleyamide A

Article abstract of DOI:10.1021/ol902525k

[Chemical equation presented] A chiral-pool approach to (-)-berkeleyamide A 1 based on a diastereoselective nitrile oxide [3 + 2]-cycloaddition completes the first total synthesis establishing the absolute stereochemistry of the natural product.

Full text of DOI:10.1021/ol902525k

ORGANIC
LETTERS
2010
Vol. 12, No. 3
420-423
Total Synthesis and Absolute
Configuration of (-)-Berkeleyamide A
Jonathan Sperry, Eric B. J. Harris, and Margaret A. Brimble*
Department of Chemistry, UniVersity of Auckland, 23 Symonds Street,
Auckland, New Zealand
m.brimble@auckland.ac.nz
Received November 1, 2009
ABSTRACT
A chiral-pool approach to (-)-berkeleyamide A 1 based on a diastereoselective nitrile oxide [3 + 2]-cycloaddition completes the first total
synthesis establishing the absolute stereochemistry of the natural product.
The Berkeley Pit Lake was formed in 1982 when mining in
the Berkeley open-pit copper mine in Butte, Montana was
Scheme 1. Proposed Synthesis of Berkeleyamide A 1
abandoned. At present, the 1.5 mile long and 1800-foot deep
lake is part of the largest United States EPA Superfund site
in North America and consists of over 30 billion gallons of
acidic (pH 2.0), metal-laden water.1 The lake has subse-
quently proven to be a rich source of biota that thrive under
hostile conditions, and the secondary metabolites2 derived
from these extremophilic organisms provide an unusual
source of bioactive molecules with medicinal potential.3
One such molecule is berkeleyamide A 1, a novel
metabolite recently isolated from a Penicillium rubrum Stoll
species from a water sample taken from a depth of 885 feet.4
Berkeleyamide A 1 was subsequently shown to possess low
micromolar activity against caspase-15 and matrix metallo-
proteinase-3 (MMP-3).6 A total synthesis of berkeleyamide
A is essential due to its medicinal potential and for structure
(1) http://www.mbmg.mtech.edu/env/env-berkeley.asp (last accessed 10/
09).
(2) For the previous isolation of secondary metabolites from this lake,
see: (a) Stierle, A. A.; Stierle, D. B.; Parker, K.; Goldstein, E.; Bugni, T.;
Baarson, C.; Gress, J.; Blake, D. J. Nat. Prod. 2003, 66, 1097. (b) Stierle,
D. B.; Stierle, A. A.; Hobbs, J. D.; Stokken, J.; Clardy, J. Org. Lett. 2004,
6, 1049. (c) Stierle, A. A.; Stierle, D. B.; Kemp, K. J. Nat. Prod. 2004, 67,
1392. (d) Stierle, A. A.; Stierle, D. B.; Kelly, K. J. Org. Chem. 2006, 71,
5357. (e) Stierle, D. B.; Stierle, A. A.; Patacini, B. J. Nat. Prod. 2007, 70,
1820.
confirmation; the absolute stereochemistry was not unam-
biguously determined in the isolation report.4 Furthermore,
the planned bioremediation of Berkeley Pit Lake would
eliminate the natural source of this compound.
(5) (a) Matter, H.; Schudok, M. Curr. Opin. Drug DiscoVery DeV. 2004,
7, 513. (b) Whittaker, M.; Floyd, C. D.; Brown, P.; Gearing, A. J. H. Chem.
ReV. 1999, 99, 2735. (c) Overall, C. M.; Kleifeld, O. Nat. ReV. Cancer
2006, 6, 227.
(3) Wilson, Z. E.; Brimble, M. A. Nat. Prod. Rep. 2009, 26, 44.
(4) Stierle, A. A.; Stierle, D. B.; Patacini, B. J. Nat. Prod. 2008, 71,
856.
10.1021/ol902525k  2010 American Chemical Society
Published on Web 12/23/2009

Our initial retrosynthetic analysis of berkeleyamide A
assumed that the natural product originates from L-leucine.
The synthesis itself hinges on a diastereoselective [3 +
2]-cycloaddition between chiral alkene 3 and nitrile oxide
4.
Scheme 3. Kinetic Protonation
Equipped with the knowledge that the dipolarophile
component dictates the stereochemical course of [3 +
2]-cycloadditions,7 it was anticipated that the chiral alkene
3 (C11) would exert significant stereocontrol upon the newly
formed C10 stereocenter. Therefore, successful cycloaddition
would afford an isoxazoline 2 which upon reductive cleavage
would deliver the 1,3-ketoalcohol present in the natural pro-
duct. This convergent strategy relies on the L-leucine stereo-
center providing an anchor on which the absolute stereo-
chemistry of the entire structure is established (Scheme 1).
Scheme 2. Synthesis of Lactam 6
difficulties in their manipulation are not uncommon.13
A
kinetic isomerization route14 was therefore deemed more
practical (Scheme 3). Thus, generation of the enolate of
lactam 6 with lithium hexamethyldisilazide followed by
alkyation with acetaldehyde and subsequent mesylation
delivered the diastereomeric mesylates 12. Elimination with
DBU afforded the conjugated internal alkenes 5 as a 3:1
mixture of regioisomers. The key kinetic protonation was
then effected using potassium hexamethyldisilazide followed
by careful quenching of the extended enolate 13 with acetic
acid as the sterically unencumbered proton source. Gratify-
ingly, the terminal alkene 3 was isolated as an 8:1 mixture
of trans:cis isomers, as determined by a clear absence of
NOE correlation between 3-H and 5-H and a strong NOE
correlation between 3-H and the CH2 of the leucine side chain
(Scheme 3, also see Supporting Information).
Our first consideration was to devise a scalable route to
lactam 6. Thus, NaBH4-H2SO4 mediated reduction8 of
L-leucine followed by dibenzylation delivered 8,9 which was
subjected to Swern oxidation followed by Horner-Wad-
sworth-Emmons olefination with phosphonate 9 affording
ester 10.10 Next, hydrogenation of 10 under acidic conditions
effected lactamisation, delivering the 2-pyrrolidinone 1111
in moderate yield. Smooth Boc-protection then afforded the
key lactam 612 (Scheme 2).
Scheme 4. Failed Cycloaddition
Initial thoughts for the synthesis of alkene 3 from lactam
6 were straightforward; it was assumed formylation of 6
followed by Wittig olefination would provide a facile route
to alkene 3. However, upon surveying the literature it became
apparent that 2-pyrrolidinone-3-carbaldehydes often exist as
mixtures of keto-enol tautomers, and reports detailing
(6) Caspase-I is a cysteine protease with an essential mediator role in
inflammation, see: Ghayur, T.; Banerjee, S.; Hugunin, M.; Butler, D.;
Herzog, L.; Carter, A.; Quintal, L.; Sekut, L.; Talanian, R.; Paskind, M.;
Wong, W.; Kamen, R.; Tracey, D.; Allen, H. Nature 1997, 386, 619.
(7) Gothelf, K. V.; Jørgensen, K. A. Chem. ReV. 1998, 98, 863.
(8) Abiko, A.; Masamune, S. Tetrahedron Lett. 1992, 33, 5517.
(9) Cooke, W. B.; Davies, S. G.; Naylor, A. Tetrahedron 1993, 49, 7955.
(10) Concello´n, J. M.; Me´jica, C. Eur. J. Org. Chem. 2007, 5250.
(11) Craven, A. P.; Dyke, H. J.; Thomas, E. J. Tetrahedron 1989, 45,
2417.
With the alkene 3 in hand, attention turned to the key
cycloaddition step. Initially, we sought to employ the
dehydration of a nitroalkane precursor 7 to generate the nitrile
oxide 4,15 based on a similar strategy to that reported by
Kozlowski during recent studies toward the spiroacetal
(12) Smrcina, M.; Majer, P.; Majerova`, E.; Guerassina, T. A.; Eissenstat,
M. A. Tetrahedron 1997, 53, 12867.
Org. Lett., Vol. 12, No. 3, 2010
421

Scheme 5
.
Cycloaddition
Table 1. Cycloaddition Conditions
entry
conditions
reaction timea (h)
temp
solvent
dr (2a:2b:2c) yield (%)
1
2
3
4
5
6
7
8
9
16 (2 equiv) and 3, add Et3N (2 equiv) over 15 min
7
7
5
3
5
2
2
2
5
7
-78 °C CH2Cl2
-78 °C CH2Cl2
-78 °C CH2Cl2
-78 °C CH2Cl2
-45 °C CH2Cl2
-45 °C CH2Cl2
-45 °C CH2Cl2
-45 °C toluene
-45 °C CH2Cl2
-45 °C CH2Cl2
1.9:1:0
2.3:1:0
2.1:1:0
1.5:1:0
2:1:1.2
3:1.5:1
2.1:1:1
2.5:1:1
3:1:0
23
27
25
12
49
48
53
35
17
22
16 (2 equiv) and 3, add Et3N (2 equiv) over 2 h
3 and Et3N (2 equiv), add 16 (2 equiv) over 2 h
3 and Et3N (2 equiv), add 16 (2 equiv) over 4 h
3 and Et3N (2 equiv), add 16 (2 equiv) over 10 min
16 (2 equiv) and 3, add Et3N (2 equiv) over 3 h
16 (3 equiv) and 3, add Et3N (3 equiv) over 3 h
16 (3 equiv) and 3, add Et3N (3 equiv) over 3 h
16 (2 equiv) and 3, add (Bu3Sn)2O (1 equiv) over 5 min
16 (3 equiv) and 3, add (Bu3Sn)2O (1.5 equiv) over 2 h
10
2.5:1:1.5
a After addition of final reactant described in ‘conditions’.
moiety of the rubromycins.16 However, treatment of a
solution of nitroalkane17 7 and alkene 3 with phenyl
isocyanate and catalytic quantities of triethylamine disap-
pointingly effected isomerization of the alkene 3 and rapid
homodimerization of the nitrile oxide to the novel oxadiazole
N-oxide 14. Inverse addition of the base and varying the
equivalents of the reagents had no effect on the outcome of
this reaction (Scheme 4).
From these preliminary studies it became clear that there
were two problemssthe nitrile oxide 4 was undergoing facile
homodimerization at room temperature and the alkene 3 was
undergoing rapid isomerization, even in the presence of
catalytic quantities of base. It was therefore decided to generate
the nitrile oxide 4 by dehydrohalogenation of a hydroxyiminoyl
halide,18 a transformation that can be conducted at lower
temperatures hopefully suppressing homodimerization as well
as the undesired isomerization of alkene 3.
relatively unstable iminoyl chloride 16 after a short silica
gel column.20 Addition of freshly prepared alkene 3 and
triethylamine to 16 at -78 °C gratifyingly afforded a mixture
of two separable diastereomeric adducts 2a and 2b, albeit
in poor yield (Entry 1, Table 1).
It was assumed that the cycloadducts 2a and 2b predomi-
nantly adopt a conformation wherein the C10-O bond of the
isoxazoline ring is antiperiplanar to the carbonyl group of
the pyrrolidine ring thereby minimizing unfavorable dipole-
dipole interactions. A strong NOE correlation between the
H10 and H11 in cycloadduct 2b clearly established the
H10-H11 syn stereochemistry whereas the absence of a
similar NOE correlation in cycloadduct 2a established the
H10-H11 anti stereochemistry in this case. The difference
(14) For similar isomerisations, see: (a) Zlatopolskiy, B. D.; Kroll, H.-
P.; Melotto, E.; de Meijere, A. Eur. J. Org. Chem. 2004, 4492. (b)
Yamauchi, C.; Kuriyama, M.; Shimazawa, R.; Morimoto, T.; Kakiuchi, K.;
Shirai, R. Tetrahedron 2008, 64, 3133.
Thus, the revised cycloaddition was conducted as follows
(Scheme 5). Treatment of oxime 1519 with N-chlorosuccin-
imide and a catalytic quantity of pyridine afforded the
(15) Mukaiyama, T.; Hoshino, T. J. Am. Chem. Soc. 1960, 82, 5339.
(16) (a) Waters, S. P.; Fennie, M. W.; Kozlowski, M. C. Tetrahedron
Lett. 2006, 47, 5409. (b) Waters, S. P.; Fennie, M. W.; Kozlowski, M. C.
Org. Lett. 2006, 8, 3243.
(17) Sinhababu, A. K.; Boschardt, R. T. Tetrahedron Lett. 1983, 24,
227.
(13) (a) Battersby, A. R.; Westwood, S. W. J. Chem. Soc., Perkin Trans.
1 1987, 1679. (b) Tamura, N.; Matsushita, Y.; Iwama, T.; Harada, S.;
Kishimoto, S.; Itoh, K. Chem. Pharm Bull. 1991, 39, 1199. (c) Katoh, T.;
Nagata, Y.; Kobayashi, Y.; Arai, K.; Minami, J.; Terashima, S. Tetrahedron
1994, 50, 6221.
(18) Kozikowski, A. P.; Adamczyk, M. J. Org. Chem. 1983, 48, 366.
(19) Kabalka, J. W.; Gaudgaon, N. M. Synth. Commun. 1988, 18, 693.
(20) Quadrelli, P.; Mella, M.; Ivernizzi, A. G.; Caramella, P. Tetrahedron
1999, 55, 10497.
422
Org. Lett., Vol. 12, No. 3, 2010

in magnitude of the observed vicinal J10,11 coupling constants
(7.5 Hz in 2a, 4.2 Hz in 2b) also supported this stereochem-
ical assignment of the newly formed isoxazoline ring, with
NOE also supporting the H11-H14 trans-relationship across
the lactam (Scheme 6, also see SI).
yield. The Boc group was removed from 17a and 17b with
TFA delivering 1a and 1b, respectively (Scheme 7).
Scheme 7
.
Synthesis of (-)-Berkeleyamide A 1a and
(-)-10-epi-Berkeleyamide A 1b
Scheme 6. NOE Correlations in Cycloadducts 2a and 2b
Comparison of the 1H and 13C NMR spectra of 1a and 1b
with the spectroscopic data provided4 for natural berkeleya-
mide A clearly showed that 1a was identical in all aspects,
whereas 1b showed several significant differences (also see
Supporting Information). This comparison conclusively
established the previously unassigned relative stereochemistry
of berkeleyamide A.
Slow addition of the alkene 3 had a negligible effect on
yield and diastereoselectivity (entry 2), as did addition of
the iminoyl chloride 16 to the alkene 3 and base (entries 3
and 4). Increasing the reaction temperature to -45 °C
increased the yield significantly as well as producing an
inseparable, third set of diastereomers 2c (entries 5-7),
presumably arising from successful cycloaddition of the
minor cis-lactam obtained during the isomerization step.
Nonetheless, epimerization of the diastereomers 2c with DBU
in boiling benzene delivered a further batch of 2a and 2b in
moderate yield as a 3:1 mixture. Overall, the best results
were obtained from the slow addition of alkene 3 and iminoyl
chloride 16 at -45 °C, giving a 53% yield of the cycload-
ducts 2a-c (entry 7). Conducting the reaction in toluene
resulted in a reduction in yield (entry 8), as did the use of
bis(tributyltin) oxide21 as the dehalogenating agent (entries
9 and 10).
Next, we set out to assign the absolute stereochemistry of
berkeleyamide A 1 using an authentic sample of the natural
product. Co-injection of synthetic 1a and the natural sample
resulted in a single peak being observed by chiral-HPLC (see
Supporting Information). Also, in our hands the optical
20
rotations of natural [R]D -15.2 (c 0.11, MeOH) and
20
synthetic [R]D -15.5 (c 0.11, MeOH) samples were in
excellent agreement. We therefore conclude that the absolute
stereochemistry of berkeleyamide A 1 is (10S), (11R), (14S)
and 1 is thus derived from L-leucine.
In conclusion, we have completed a chiral-pool based total
synthesis of (-)-berkeleyamide A 1 using a diastereoselective
[3 + 2]-cycloaddition, thus assigning the absolute stereo-
chemistry of the natural product.
Next, the key reductive cleavage was conducted. A
suspension of Mo(CO)6 and 2a in wet acetonitrile22 was
heated to 80 °C for 3 h, delivering the desired keto alcohol
17a in good yield. Cycloadduct 2b succumbed to the same
reaction conditions delivering keto alcohol 17b, also in good
Acknowledgment. We thank the Royal Society of New
Zealand Marsden fund for financial support and Donald and
Andrea Stierle (University of Montana) for providing an
authentic sample of 1.
Supporting Information Available: Characterization data
and HPLC traces. This material is available free of charge
via the Internet at http://pubs.acs.org.
(21) Moriya, O.; Takenaka, H.; Iyoda, M.; Urata, Y.; Endo, T. J. Chem.
Soc., Perkin Trans. 1 1994, 413.
(22) Baraldi, P.; Barco, A.; Benetti, S.; Manfredini, S.; Simoni, D.
Synthesis 1987, 276.
OL902525K
Org. Lett., Vol. 12, No. 3, 2010
423