Synthesis of the tetracyclic core of berkelic acid using gold(I)-catalyzed hydroarylation and oxidative radical cyclizations

Article abstract of DOI:10.1021/ol302536j

A synthetic approach to the tetracyclic core of berkelic acid is reported using gold(I)-catalyzed intramolecular hydroarylation and oxidative radical cyclizations to effect the key ring-forming steps. The carboxylic acid was introduced via a late-stage palladium-catalyzed carbonylation to afford the core tetracycle with the correct relative stereochemistry for the natural product.

Full text of DOI:10.1021/ol302536j

ORGANIC
LETTERS
2012
Vol. 14, No. 23
5820–5823
Synthesis of the Tetracyclic Core of
Berkelic Acid Using Gold(I)-Catalyzed
Hydroarylation and Oxidative Radical
Cyclizations
Margaret A. Brimble,* Isabell Haym, Jonathan Sperry, and Daniel P. Furkert
School of Chemical Sciences, The University of Auckland, 23 Symonds Street, Auckland,
New Zealand
m.brimble@auckland.ac.nz
Received September 16, 2012
ABSTRACT
A synthetic approach to the tetracyclic core of berkelic acid is reported using gold(I)-catalyzed intramolecular hydroarylation and oxidative radical
cyclizations to effect the key ring-forming steps. The carboxylic acid was introduced via a late-stage palladium-catalyzed carbonylation to afford
the core tetracycle with the correct relative stereochemistry for the natural product.
Extremophiles are microorganisms which grow under
harsh conditions such as at high temperature, high pres-
sure, high salt concentrations, and low pH.1 An example of
a low pH environment is the Berkeley Pit lake in Butte,
MT, that formed upon infiltration of groundwater into
an abandoned copper mine. The acidic water (pH ∼2.5) is
contaminated with high concentrations of metal sulfates,
and only certain microorganisms can survive in these
extreme conditions. A Penicillium species isolated from
the lake by Stierle et al.2 produces berkelic acid (1), a novel
Figure 1. (ꢀ)-Berkelic acid, 1. Stereochemistry as assigned by
Snider.4
metabolite with a highly substituted tetracyclic spiroacetal
core. Enzyme assays established that berkelic acid inhibits
the endopeptidase MMP-3 (GI50 = 1.87 μM) as well as the
cysteine protease caspase-1 (GI50 = 0.098 nM). Screening
at the National Cancer Institute revealed that berkelic acid
exhibits selective activity against the ovarian cancer cell
line OVCAR-3 in the nanomolar range (GI50 = 91 nM).
published the first total synthesis of (ꢀ)-berkelic acid (1),
confirmingthereassignedstereochemistryandestablishing
the absolute configuration (Figure 1). De Brabander et al.5
synthesized berkelic acid by combining the two natural
product inspired fragments spicifernin and pulvilloric acid
using a [4 þ 2] cycloaddition. Pettus et al.6 have also
Prompted by a recent synthesis-driven structure revision
€
3
by Furstner et al. the stereochemistry of berkelic acid at
C-18 and C-19 was reassigned. Shortly after, Snider et al.4
(4) Wu, X.; Zhou, J.; Snider, B. B. Angew. Chem., Int. Ed. 2009, 48,
1283–1286.
(1) Wilson, Z. E.; Brimble, M. A. Nat. Prod. Rep. 2009, 26, 44–71.
(2) Stierle, A. A.; Stierle, D. B. J. Org. Chem. 2006, 71, 5357–5360.
(3) Buchgraber, P.; Snaddon, T. N.; Wirtz, C.; Mynott, R.; Goddard,
(5) Bender, C. F.; Yoshimoto, F. K.; Paradise, C. L.; De Brabander,
J. K. J. Am. Chem. Soc. 2009, 131, 13350–13352.
(6) Wenderski, T. A.; Marsini, M. A.; Pettus, T. R. R. Org. Lett.
2011, 13, 118–121.
€
R.; Furstner, A. Angew. Chem., Int. Ed. 2008, 47, 8450–8454.
r
10.1021/ol302536j
Published on Web 11/27/2012
2012 American Chemical Society

executed a racemic diastereoselective formal synthesis
using a [4 þ 2]-cycloaddition and earlier this year a highly
efficient approach based on silver-catalyzed cycloaddition
of an enol ether and an o-quinonemethide to form the
Scheme 2. Synthesis of Tricycle 13
tetracyclic core was described by Fananas and Rodrı
´
guez.7
~
ꢀ
Recently, our laboratory reported an enantioselective
formal synthesis of berkelic acid via addition of a silyl enol
ether to an aryl oxonium ion.8 We now herein report
a unique approach to the tetracyclic spiroacetal core of
berkelic acid using a gold(I)-catalyzed hydroarylation and
oxidative radical cyclizations to construct the tetracyclic
spiroacetal core (Scheme 1).
Scheme 1. Synthetic Strategy for the Tetracyclic Core of
Berkelic Acid
commercially available 3-hydroxyphenylacetic acid fol-
lowed by selective bromination10 at C-4 generated phenol 7.
Subsequent propargylation provided alkyne 8 which
was then subjected to intramolecular CꢀC bond forma-
tion. Use of thermal conditions to effect Claisen rearrange-
ment did not afford chromene 9, presumably due to the
absence of gem-dimethyl groups.11 Transition metals, such
as Pd,12 Pt,13 Ru,14 Hg,15 and Au,16 have been demon-
strated to activate aryl CꢀH bonds toward hydroarylation
for cyclization.17 Sames et al.13 reported that PtCl4 was
the only suitable catalyst that delivered satisfactory
yields to effect cyclization on similar substrates. However,
use of PtCl4 on parent substrate 8 gave only poor yields
Our initial attention focused on the tetracyclic chromane
spiroacetal core structure 2 present in berkelic acid. As
outlined in our retrosynthesis, it was envisaged that an
oxidative radical cyclization could be used to form the
CꢀO bond at the benzylic C-15 position (Scheme 1). The
tetracyclic chromane structure 3 could be constructed by
oxidative radical cyclization of alcohol 4 onto the benzylic
carbon. In turn, precursor 4 could be synthesized by gold-
(I)-catalyzed hydroarylation of alkyne 6 followed by oxi-
dative radical spiroacetalization. Finally, a late-stage
palladium-catalyzed carbonylation of bromide 3 allows
introduction of the carboxylic acid 2 present in berkelic
acid, utilizing the orthogonality of palladium and gold
catalysis.9
€
(10) Lindberg, M.; Hellberg, S.; Bjork, S.; Gotthammar, B.;
Hogberg, T.; Persson, K.; Schwarcz, R.; Luthman, J.; Johansson, R.
€
Eur. J. Med. Chem. 1999, 34, 729–744.
(11) (a) Brown, P. E.; Lewis, R. A. J. Chem. Soc., Perkin Trans. 1
1992, 573–577. (b) Harfenist, M.; Thom, E. J. Org. Chem. 1972, 37, 841–
848.
(12) (a) Jia, C.; Lu, W.; Oyamada, J.; Kitamura, T.; Matsuda, K.;
Irie, M.; Fujiwara, Y. J. Am. Chem. Soc. 2000, 122, 7252–7263. (b) Jia,
C.; Piao, D.; Kitamura, T.; Fujiwara, Y. J. Org. Chem. 2000, 65, 7516–
7522.
(13) Pastine, S. J.; Youn, S. W.; Sames, D. Org. Lett. 2003, 5, 1055–
1058.
The synthesis of the tricyclic scaffold 11 from chromane
10 (Scheme 2) was our initial goal. Esterification of
(14) Chatani, N.; Inoue, H.; Ikeda, T.; Murai, S. J. Org. Chem. 2000,
65, 4913–4918.
(15) Nishizawa, M.; Takao, H.; Yadav, V. K.; Imagava, H.; Sugihara,
T. Org. Lett. 2003, 5, 4563–4565.
(16) Menon, R. S.; Findley, A. D.; Bissember, A. C.; Banwell, M. G.
~
ꢀ
(7) Fananas, F. J.; Mendoza, A.; Arto, T.; Temelli, B.; Rodrı
´
guez, F.
Angew. Chem., Int. Ed. 2012, 51, 4930–4933.
J. Org. Chem. 2009, 74, 8901–8903.
(8) (a) McLeod, M. C.; Wilson, Z. E.; Brimble, M. A. Org. Lett. 2011,
13, 5382–5385. (b) McLeod, M. C.; Wilson, Z. E.; Brimble, M. A. J. Org.
Chem. 2012, 77, 400–416.
(17) For recent examples and reviews on CꢀH activation, see:
(a) Boormann, T. C.; Larrosa, I. Chem. Soc. Rev. 2011, 40, 1910–1925.
(b) McMurray, L.; O’Hara, F.; Gaunt, M. J. Chem. Soc. Rev. 2011, 40,
1885–1898. (c) Gutekunst, W. R.; Baran, P. S. Chem. Soc. Rev. 2011, 40,
1976–1991. (d) Lewis, J. C.; Coehl, P. S.; Arnold, F. H. Chem. Soc. Rev.
2011, 40, 2003–2021.
€
€
(9) Hashmi, A. S. K.; Lothschutz, C.; Dopp, R.; Ackermann, M.;
Becker, J.; Rudolph, M.; Scholz, C.; Rominger, F. Adv. Syn. Catal. 2012,
354, 133–147.
Org. Lett., Vol. 14, No. 23, 2012
5821

(Table 1, entry 1). Cationic gold(I) catalysts have been
shown to be excellent catalysts for chromene formation.18
The Echavarren group has reported the stable gold(I)
catalyst A19,20 (Table 1), which does not need a cocatalyst
due to weak coordination of an acetonitrile ligand,
whereas catalysts B and C need to be activated by a
cocatalyst (typically a silver salt) to generate the active
catalytic species. In the present work, conversion of 8 to 9
proceeded in good yield using all three gold catalysts AꢀC
(Table 1, entries 2ꢀ4). Olefin 9 was then hydrogenated
over Adam’s catalyst (PtO2), and subsequent reduction of
the methyl ester with LiAlH4 afforded alcohol 10.
with NBS and dibenzoyl peroxide (BPO) successfully
resulted in the direct formation of the tricyclic chromane
11 in 50% yield without any Norrish type I fragmentation
product 12 observed.
Scheme 3. Radical Cyclization of Alcohol 14
Table 1. Metal-Catalyzed Intramolecular Hydroarylationa
With tricyclic bromide 11 in hand, the late-stage carbo-
nylation was next attempted. Neither lithiumꢀhalogen
exchange nor Grignard-based halogenꢀmetal exchange
followed by introduction of carbon dioxide or Mander’s
reagent (NCCO2Me) led to formation of the desired
product.24ꢀ26 Palladium catalyzed carbonylation of bro-
mide 11 in a 2:1 mixture of DMF/MeOH in the presence of
entry
cat.
conditions
9, yield (%)
€
Pd(OAc)2/dppp and Hunig’s base under carbon monoxide
b
at high temperature or pressure did not afford the desired
carboxylic acid.27 Pleasingly however, generation of car-
bon monoxide in situ using the modified conditions re-
ported by Caille and co-workers28 afforded the desired
carboxylic acid 13 in 61% yield.
We next sought to introduce further substituents onto
tricycle 11 and needed to establish whether a secondary
alcohol was compatible with the subsequent cyclization.
Toward this end, alcohol 10 was oxidized to the aldehyde
and treated with n-pentylmagnesium bromide to afford
secondary alcohol 14 (Scheme 3). Employing NBS with
BPO successfully afforded the desired tricycle 15, showing
that a secondary alcohol precursor bearing the requisite
side chain of berkelic acid could participate in the key
oxidative cyclization. The trans-stereochemistry between
H-3a and H-5 was assigned based on the observation of an
NOE between H-3a and H-10.
It was next decided to introduce the 6,5-spiroacetal unit
present in berkelic acid by incorporation of an additional
oxidative radical cyclization. In this case, a more substi-
tuted alkyne was required in the initial hydroarylation
step in order to introduce the 5-membered spiroacetal ring.
Alkylation of bromide 16 (Scheme 4) with phenol 7
afforded alkyne 6. Intramolecular hydroarylation with
Echavarren’s catalyst A and subsequent hydrogenation
with concomitant debenzylation over Adam’s catalyst
furnished chromane 5. Radical spirocyclization using
1
2
3
4
PtCl4
Ac
DCE, 80 °C, MW, 1 h
CH2Cl2, rt, 1.5 h
CH2Cl2, rt, 1.5 h
CH2Cl2, rt, 1.5 h
23
76
72
69
c
c
B/AgSbF6
C/AgSbF6
a All gold-catalyzed reactions were carried out in the dark. b 5 mol %.
c 1 mol % catalyst and cocatalyst if required.
Attention next focused on the key oxidative radical
€
cyclization using an oxa variant of the HofmannꢀLofflerꢀ
Freytag reaction.21 Applying the hypoiodite-induced
free-radical conditions successfully employed for the for-
mation of spiroacetals,22 alcohol 10 was irradiated (60 W)
in the presence of iodobenzene diacetate and iodine at
room temperature. In this case, alcohol 10 was found to
predominantly undergo a Norrish type I fragmentation
to form iodide 12 (Scheme 2).23 Treatment of alcohol 10
with DDQ, as reported by Pettus,6 did not result in the
formation of chromane 11. After a survey of reagents and
conditions, it was eventually found that treatment of 10
(18) (a) Wang, Y.; Ji, K.; Lan, S.; Zhang, L. Angew. Chem., Int. Ed.
2012, 51, 1915–1918. (b) Lykakis, I. N.; Efe, C.; Gryparis, C.; Stratakis,
M. Eur. J. Org. Chem. 2011, 2334–2338.
(19) For recent reviews on gold catalysis, see: (a) Rudolph, M.;
€
Hashmi, A. S. K. Chem. Soc. Rev. 2012, 41, 2448–2462. (a) Furstner,
A. Chem. Soc. Rev. 2009, 38, 3208–3221. (b) Hashmi, A. S. K. Chem.
Rev. 2007, 107, 3180–3213. (c) Li, Z.; Brouwer, C.; He, C. Chem. Rev.
2008, 108, 3239–3265.
(20) Herrero-Gomez, E.; Nieto-Oberhuber, C.; Lopez, S.; Benet-
Buchholz, J.; Echavarren, A. M. Angew. Chem., Int. Ed. 2006, 45,
5455–5459.
ꢀ
ꢀ
(24) Goddard, C. M.; Nowick, J. S. J. Org. Chem. 2010, 75, 1822–
ꢀ
ꢀ
(21) (a) de Armas, P.; Concepcion, J. I.; Francisco, C. G.; Hernandez,
ꢀ
1830.
(25) Walborsky, H. M.; Goedken, V. L.; Gawronsky, J. K. J. Org.
Chem. 1992, 57, 410–413.
R.; Salazar, J. A.; Suarez, E. J. Chem. Soc., Perkin Trans. 1 1989, 405–
ꢀ
411. (b) Francisco, C. G.; Freire, R.; Herrera, A. J.; Perez-Martin, I.;
Suarez, E. Org. Lett. 2002, 4, 1959–1961.
ꢀ
(26) Boudet, N.; Lachs, J. R.; Knochel, P. Org. Lett. 2007, 9, 5525–
5528.
(27) Grimm, J. B.; Wilson, K. J.; Witter, D. J. Tetrahedron Lett. 2007,
48, 4509–4513.
(22) (a) Furkert, D. P.; Brimble, M. A. Org. Biomol. Chem. 2004, 2,
3573–3583. (b) Meilert, K.; Brimble, M. A. Org. Biomol. Chem. 2006, 4,
2184–2192. (c) Sperry, J.; Liu, Y.-C.; Brimble, M. A. Org. Biomol. Chem.
2010, 8, 29–38.
(23) Heusler, K.; Kalvoda, J. Angew. Chem. 1964, 76, 518–531.
Angew. Chem., Int. Ed. 1964, 3, 525–538.
(28) Berger, P.; Bessmernykh, A.; Caille, J.-C.; Mignonac, S. Synthesis
2006, 18, 3106–3130.
5822
Org. Lett., Vol. 14, No. 23, 2012

of an NOE correlation between H-30a and H-50 consistent
with a syn 1,3-diaxial relationship together with the char-
acteristic downfield chemical shift for H-30a at δ 4.9 ppm
(C6D6) due to the deshielding effect of the axial C2ꢀO1
bond.27 Finally, synthesis of the tetracyclic core of berkelic
acid was initiated. Alcohol 18 was oxidized to the aldehyde
with IBX followed by treatment with n-pentylmagnesium
bromide to give a separable mixture of racemic secondary
alcohol diastereomers 20a and 20b. Changing the reaction
solvent from THF to diethyl ether increased the ratio of
diastereomers from 1:1.2 to 1:6 favoring 20b.
Scheme 4. Synthesis of the Tetracyclic Core of Berkelic Acid
Each of the racemic diastereomers 20a and 20b was
individually subjected to treatment with NBS and BPO.
Diastereomer 20a selectively afforded tetracycle 21a as as
the sole product. Compound 21a exhibiteda trans relation-
ship between H-30a and H-50 as confirmed by an NOE
correlation between H-30a and H-100. In contrast, cycliza-
tion of alcohol 20b afforded an inseparable mixture of
tetracycles 21b and 21c and a small amount of 21a
(ca. 3:3:1). Tetracycle 21b, for which an NOE correlation
was observed between H-30a and H-50, represents the
stereochemistry of berkelic acid. The stereochemistry
for 21c was assigned based on a similar H-30a to H-50
NOE correlation, and the distinctive upfield shift of H-30a
due to the absence of a 1,3-diaxial oxygen.29,30 Gratifyingly,
palladium-catalyzed carbonylation of the mixture of 21aꢀc
with a 20-fold excess of acetic anhydride, lithium formate
and DIPEA to generate CO in situ afforded the correspond-
ing carboxylic acids in high yield. Chromatographic separa-
tion afforded 22b and its (2,30a)-epimer 22a.31 22b possesses
the same relative stereochemistry as the tetracyclic core of
berkelic acid.
In summary, we report a novel series of disconnections
to access the tetracyclic core of berkelic acid using a
combination of gold(I) catalyzed intramolecular hydroar-
ylation and oxidative radical cyclizations. Notably, the
stereochemistry obtained in the final free radical bromina-
tion and cyclization was dependent on the relative config-
uration of the C-200 alcohol and C-2 spiroketal. A late-stage
palladium-catalyzed carbonylation of an aryl bromide
was used to introduce the carboxylic acid group onto the
berkelic acid scaffold, affording the tetracyclic spiroketal
22b possessing the correct relative stereochemistry for the
natural product.
Acknowledgment. We thank The Universityof Auckland
and the Royal Society of New Zealand Marsden Fund for
financial support.
PhI(OAc)2/I2 and subsequent reduction afforded spiro-
acetal 18 which was then subjected to radical cyclization
using HgO and iodine to provide tetracycle 19 for which
the stereochemistry was assigned based on the observation
Supporting Information Available. Experimental pro-
cedures and spectroscopic data of new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(29) Although the mixture of 21aꢀc proved inseparable, the indivi-
dual tetracyclic systems were able to be assigned using 2D NMR and
NOE correlations.
(30) Zhou, J.; Snider, B. B. Org. Lett. 2007, 9, 2071–2074.
(31) Predominant formation of 22a and 22b indicates some epimer-
ization at C-2 during the carbonylation step.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 23, 2012
5823