Total synthesis of auripyrone a using a tandem non-aldol aldol/ paterson aldol process as a key step

Article abstract of DOI:10.1002/anie.200904607

To aldol or non-aldol: The titled reaction sequence generates the polypropionate 3 from the epoxy alcohol 1 and the ketone 2 as a single diastereomer. Compound 2 was used for an efficient synthesis of auripyrone A using a highly regioselective hemiketaliz

Full text of DOI:10.1002/anie.200904607

Communications
DOI: 10.1002/anie.200904607
Natural Products
Total Synthesis of Auripyrone A Using a Tandem Non-Aldol Aldol/
Paterson Aldol Process as a Key Step**
Michael E. Jung* and Ramin Salehi-Rad
In 1996 Yamada and co-workers reported the isolation of
auripyrones A (1) and B (2) from the methanol extracts of the
sea hare Dolabella auricularia (Aplysiidae; Figure 1).[1]
Figure 1.
Extensive NMR investigation of the compounds revealed a
complex spiroketal core capped at one end by a tetrasub-
stituted g-pyrone, residing in an anomerically favored con-
figuration wherein all but the substituents are positioned
equatorially except for the C10 methyl and the C11 acyloxy
groups. Auripyrones A (1) and B (2) showed cytotoxicity
against HeLa S3 cells with IC50 values of 260 and 480 ngmLꢀ1,
respectively. To date one total synthesis[2] of 1 has been
Scheme 1. Retrosynthetic analysis.
reported by Perkins et al., which utilizes an elegant biomim-
etic cyclization of an acyclic triketone intermediate to
generate the spiroketal moiety.[3] The major drawback of
this approach, however, is the late-stage formation of the
g-pyrone in the presence of the sensitive spiroketal moiety,
which proceeded with poor yield (39% based on recovered
starting material). Nonetheless, Perkinsꢀ convergent total
synthesis of 1 constitutes the only established synthetic route,
and led to the determination of the absolute stereochemistry
of the natural product. Herein, we report our convergent
approach for the total synthesis of auripyrone A (1) as a single
diastereomer in high chemical yield.
cyclization of the C17 ketone onto the hemiketal 3, which
should be available from the aldolate 4. The key intermediate
4 could be obtained from a fully matched[4] double stereo-
differentiating[5] anti-aldol reaction of the boron enolate of
the ketone 6 with the aldehyde 5. The g-pyrone moiety of 5
would result from the aldehyde 7 by using the protocol of
Gillingham and Hoveyda.[6] Finally, the stereopentad 7 was
envisioned to arise from the known epoxide 8[7] by a novel
tandem non-aldol aldol[8]/Paterson-lactate-derived aldol[9]
reaction with the ketone 9.
In our retrosynthetic analysis (Scheme 1), the sensitive
spiroketal moiety of 1 could be derived from a late-stage
The synthesis commenced with the assembly of 7, using a
highly convergent tandem non-aldol aldol/Paterson-lactate-
derived aldol reaction (Scheme 2). Epoxidation of the allylic
alcohol 10 (synthesized in five steps from (S)-Roche ester),[10]
under either reagent controlled Sharpless conditions[11] or
substrate controlled reaction conditions with mCPBA,[12]
furnished the epoxide 8 in 85% yield and 20:1 diastereomeric
ratio (d.r.), or 90% yield and 16:1 d.r., respectively. Protec-
tion of 8 with TESCl provided the corresponding silyl ether
which was then treated with TESOTf at ꢀ458C to give, by the
non-aldol aldol reaction, the syn-aldol adduct 11 in 86% yield
and 20:1 d.r. Unlike conventional auxiliary-based aldol meth-
ods which require a protection step and subsequent removal
of the chiral auxiliary to generate the protected aldehyde, the
[*] Prof. M. E. Jung, R. Salehi-Rad
Department of Chemistry and Biochemistry
University of California, Los Angeles
405 Hilgard Ave, Los Angeles, CA 90095-1569 (USA)
Fax: (+1)310-206-3722
E-mail: jung@chem.ucla.edu
[**] This work was supported by the National Science Foundation (CHE
0614591). R.S.-R. was an NIH Chemistry-Biology Interface trainee,
2005–2008, and thanks the NIH Medical Scientist Training Program
(MSTP) at UCLA for generous financial support.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200904607.
8766
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 8766 –8769

Angewandte
Chemie
possessed no optical rotation and displayed only twelve
13C NMR resonances indicating a symmetrical structure,
confirmed the assigned stereochemistry of 7.
Next, we turned to the synthesis of the g-pyrone
moiety[2a,17] (Scheme 3). By using the protocol of Gillingham
and Hoveyda,[6] we obtained, through the aldol reaction of the
lithium enolate of the silyloxy enone 15[16] with the aldehyde
7, the aldolate 16 in 94% yield as a mixture of isomers.[18]
Oxidation of the isomeric mixture of 16 with DMP[19] and
subsequent heating of the resulting diketone in DMF[20]
provided the desired g-pyrone 17 in 68% yield over two
steps. Acid-promoted removal of the TES ether furnished the
alcohol 18, which was subjected to Yamaguchi esterifica-
tion[21] with isovaleric acid to give the ester 19 in 98% yield.
Treatment of the silyl ether 19 with HF·pyridine provided the
primary alcohol 20[22] which was then oxidized with DMP to
afford the aldol precursor 5. The other component of the aldol
reaction, the a-methyl-b-hydroxy ketone 6, was also readily
available from protecting the known ketone 21[2a] as the TES
ether (96% yield). The E-boron enolate of the ketone 6
underwent a highly diastereoselective anti-aldol reaction with
5 to provide the Felkin–Ahn product 4 in 94% yield and 21:1
diastereomeric ratio. The excellent diastereoselectivity of this
double stereodifferentiating[5] aldol reaction could be attrib-
uted to a fully matched[4] reactant pair, where the stereoin-
duction from both the b-hydroxy[23] and the a-methyl[14]
substituent of the aldehyde, and the a-methyl stereocenter
of the ketone[24] are reinforcing.
Scheme 2. Reagents and conditions: a) Ti(OiPr)4, tBuOOH, (+)-DIPT,
CH2Cl2, ꢀ108C, 85%, 20:1 d.r. or mCPBA, K2HPO4, CH2Cl2, ꢀ108C,
90%, 16:1 d.r.; b) TESCl, imidazole, CH2Cl2, 98%; c) TESOTf, DIPEA,
CH2Cl2, ꢀ458C, 86%, 20:1 d.r.; d) 9, cHex2BCl, Me2NEt, Et2O,
ꢀ788C!08C, 2 h; 08C!ꢀ788C; 11, ꢀ788C!ꢀ258C, 15 h; H2O2,
=
MeOH, pH 7 buffer, 08C, 1 h, 86%, one isomer; e) PMBOC( NH)-
CCl3, Sc(OTf)3, toluene 93%; f) LiBH4, THF, 97%, 8:1 d.r.; g) NaIO4,
MeOH/pH 7 buffer (2:1), 86%. h) NaBH4, EtOH, 08C, 81%;
i) TBDPSCl, imidazole, CH2Cl2; j) CAN, CH3CN/H2O (9:1), 78% over
two steps. DIPT=diisopropyltartrate, mCPBA=meta-chloroperbenzoic
acid, TES=triethylsilyl, Tf=trifluoromethanesulfonyl, DIPEA=diiso-
propylethylamine, Bz=benzoyl, PMB=para-methoxybenzyl,
TBDPS=tert-butyldiphenylsilyl, CAN=ceric ammoinuim nitrate.
non-aldol aldol reaction provides direct access
to pure silyl-protected aldehydes without the
flash column chromatography purification[13] for
an iterative aldol process. To that end, the
subsequent anti-aldol reaction of the aldehyde
11 with the E-boron enolate of Patersonꢀs
lactate-derived ketone 9[9c] furnished the desired
anti-aldolate 12 in 86% yield as a single
diastereomer. The remarkable stereoselectivity
of this reaction is a result of double stereodif-
ferentiation,[5] where the stereoinduction from
the enolate[9c] and the
a-methyl substituent of the aldehyde[14] are
reinforcing. Mild Lewis acid catalyzed[15] pro-
tection of the alcohol 12 as the PMB ether
proceeded smoothly, without removal of the
acid sensitive TES group, to afford the ketone
13 in 93% yield. Reduction of 13 and concom-
itant removal of the a’-benzoate with LiBH4 and
periodate cleavage of the resulting diol[16]
afforded the desired aldehyde 7 in 83% yield
Scheme 3. Reagents and conditions: a) 15, LDA, ꢀ788C, 7, 94%; b) DMP, CH2Cl2,
NaHCO3; c) DMF, 558C mw, 6 h, 68% over two steps; d) PPTS, CH2Cl2/MeOH (3:1),
96%; e) 2,4,6-trichlorobenzoyl chloride, DMAP, Et3N, isovaleric acid, 98%; f) HF·py,
CH3CN/py (7:1), 94%; g) DMP, CH2Cl2, NaHCO3, 98%; h) TESCl, imidazole, CH2Cl2,
96%; i) 6, cHex2BCl, Me2NEt, Et2O, ꢀ788C!08C, 2 h, 08C!ꢀ788C, 5, ꢀ788C!
ꢀ258C, 15 h, H2O2, MeOH, pH 7 buffer, 08C, 1 h, 94%, 21:1 d.r. TIPS=triisopropylsilyl,
LDA=lithium diisopropylamide, DMP=Dess–Martin periodinane, DMF=N,N-dime-
thylformamide, PPTS=pyridinium para-toluenesulfonate, DMAP=4-dimethylaminopyr-
over two steps. This novel tandem non-aldol
aldol/Paterson-lactate-derived aldol protocol
constitutes
a
highly efficient, convergent
approach for the synthesis of the desired stereo-
pentad 7, generating four aldol stereocenters in
two steps. Conversion of the aldehyde 7 in three
steps into the meso-polypropionate 14, which idine, py=pyridine.
Angew. Chem. Int. Ed. 2009, 48, 8766 –8769
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 8767

Communications
With the key intermediate 4 in hand, we set out to
investigate the formation of the spiroketal moiety of 1
(Scheme 4). Our initial attempts at generating the desired
unstable hemiketal 23 and recycle the acyclic diketone 22, a
variety of conditions were screened for the removal of the
ethers. Gratifyingly, oxidation of the aldolate 4 and then
treatment with CAN in acetonitrile/water
(9:1) for 15 minutes led to concurrent
removal of the PMB and TES ethers,
providing exclusively the stable hemike-
tal 24 as a single diastereomer in 74%
yield over two steps.[31] The selectivity of
this reaction is remarkable and yet diffi-
cult to explain since the removal of the
PMB and TES ethers unveils two alco-
hols which could potentially cyclize onto
the C13 ketone to form a hemiketal.
Simple thermodynamic MM2 calcula-
tions proved ineffective and higher level
calculations will be necessary to elucidate
the remarkable selectivity of this reaction
for exclusive formation of the hemiketal
24.
Having successfully prepared the
hemiketal 24 as a single diastereomer
from the aldolate 4 in two steps in
excellent yield, we next oxidized the
alcohol of 24, and the resulting dike-
tone[32] was treated with Amberlyst-15 to
afford the natural product auripyrone A
(1) as a single diastereomer in 80% yield
over two steps. The remarkably high
chemical yield of this spiroketalization
could presumably be attributed to the
lower entropic cost of cyclization onto a
conformationally limited hemiketal plat-
Scheme 4. Reagents and conditions: a) DMP, CH2Cl2, NaHCO3, 94%; b) DDQ, CH2Cl2/pH 7
buffer (9:1), 66% 23, 21% 22; c) HF·py, CH3CN/py (5:1), 86%; d) HF·py, CH3CN/py (5:1),
78%; e) DMP, CH2Cl2, NaHCO3; f) CAN, CH3CN:H2O (9:1), 74% over two steps; g) DMP,
CH2Cl2, NaHCO3; h) Amberlyst-15, CH2Cl2, 08C!258C, 80% over two steps. DDQ=2,3-
dichloro-5,6-dicyanobenzoquinone.
spiroketal by acid-catalyzed cyclization of the C9 hydroxy
group onto either an g-pyrone[25] or an acyclic triketone[2a]
failed, and led to rapid 1,5-acyl migration. The high propen-
sity of our system for acyl migration presumably arises from
the inherent preference of the acyclic polyketide[26] to
populate the local conformation I where the C9 and C11
oxygen moieties are in close spatial proximity. To circumvent
this problem, we decided to mask the C9 hydroxy substituent
and attempt spiroketalization from a hemiketal platform.[27]
Oxidation of the alcohol 4 afforded the corresponding
diketone, as a single isomer,[28] which was then deprotected
with DDQ to furnish the hemiketal 23 as the major product
(determined by 1H and 13C NMR analyses of the crude
reaction mixture). However, purification on silica gel gave the
desired cyclic isomer 23 only in 66% yield as a single isomer,
along with the open chain isomer 22 in 21% yield as a mixture
of two C14 methyl epimers.[29] Removal of the TES ether of
the hemiketal 23 with HF·pyridine provided the hemiketal 24
in 86% yield as a single isomer[30] which was stable to column
chromatography. Interestingly, the removal of the TES group
on the acyclic diketone 22 also afforded the hemiketal 24
exclusively, as a mixture of C14 epimers, in 78% yield. The
mixed configuration of the C14 stereocenter in the hemiketal
24 is not critical since both diastereomers could equilibrate
under thermodynamic cyclization conditions to the natural
product. Nonetheless, to circumvent having to purify the
form, or the increase in the rate of the acid-catalyzed
cyclization, since the configuration of the C14 stereocenter
of the isomerically pure starting material 24 matches the
desired configuration in the natural product.
The chemistry described herein constitutes a highly
convergent approach for the synthesis of auripyrone A (1)
from the known epoxide 8 in 18 steps and 17% overall yield.
Our strategy employs a novel tandem non-aldol aldol/
Paterson-lactate-derived aldol to generate the stereopentad
backbone, a highly regioselective hemiketalization of a keto
diol, and a late stage spiroketalization onto the stable
hemiketal.
Received: August 19, 2009
Published online: October 12, 2009
Keywords: aldol reaction · epoxides · natural products ·
.
rearrangement · total synthesis
[1] K. Suenaga, H. Kigoshi, K. Yamada, Tetrahedron Lett. 1996, 37,
5151 – 5154.
[2] a) T. Lister, M. V. Perkins, Angew. Chem. 2006, 118, 2622 – 2626;
Angew. Chem. Int. Ed. 2006, 45, 2560 – 2564; for synthetic efforts
towards auripyrone A 1 see: b) M. V. Perkins, S. Jahangiri, M. R.
Taylor, Tetrahedron Lett. 2006, 47, 2025 – 2028; c) M. V. Perkins,
8768 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 8766 –8769

Angewandte
Chemie
R. A. Sampson, J. Joannou, M. R. Taylor, Tetrahedron Lett. 2006,
47, 3791 – 3795.
[18] Although the major isomer could be separated, the mixture of
diastereomers was taken forward since they could all lead to the
same product.
[19] a) D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155 – 4156;
b) D. B. Dess, J. C. Martin, J. Am. Chem. Soc. 1991, 113, 7277 –
7287.
[20] The original reaction conditions (THF, DBU, 608C) reported by
Gillingham and Hoveyda (see reference [6]) were slightly less
efficient for cyclization in our system (53% yield over two steps).
[21] J. Inanaga, K. Hirata, H. Saeki, T. Katsuki, M. Yamaguchi, Bull.
Chem. Soc. Jpn. 1979, 52, 1989 – 1993.
[22] Treatment with tetra-n-butylammonium flouride led to an acyl
migration providing exclusively the undesired primary ester/
secondary alcohol in 78% yield.
[23] a) D. A. Evans, M. J. Dart, J. L. Duffy, M. G. Yang, A. B.
Livingston, J. Am. Chem. Soc. 1995, 117, 6619 – 6620; b) D. A.
Evans, M. J. Dart, J. L. Duffy, M. G. Yang, J. Am. Chem. Soc.
1996, 118, 4322 – 4343.
[24] a) D. A. Evans, H. P. Ng, J. S. Clark, D. L. Rieger, Tetrahedron
1992, 48, 2127 – 2142; b) I. Paterson, M. V. Perkins, J. Am. Chem.
Soc. 1993, 115, 1608 – 1610; c) A. Vulpetti, A. Bernardi, C.
Gennari, J. M. Goodman, I. Paterson, Tetrahedron 1993, 49,
685 – 696.
[3] The minor isomer (6%) in this reaction was inseparable during
the remainder of the synthesis and was observed as a minor
impurity in the spectra of the final synthetic auripyrone A (1).
[4] D. A. Evans, M. J. Dart, J. L. Duffy, D. L. Rieger, J. Am. Chem.
Soc. 1995, 117, 9073 – 9074.
[5] S. Masamune, W. Choy, J. S. Petersen, L. R. Sita, Angew. Chem.
1985, 97, 1 – 31; Angew. Chem. Int. Ed. Engl. 1985, 24, 1 – 30.
[6] D. G. Gillingham, A. H. Hoveyda, Angew. Chem. 2007, 119,
3934 – 3938; Angew. Chem. Int. Ed. 2007, 46, 3860 – 3864.
[7] S. H. Kang, H. S. Jun, Chem. Commun. 1998, 1929 – 1930.
[8] a) M. E. Jung, D. C. DꢀAmico, J. Am. Chem. Soc. 1993, 115,
12208 – 12209; b) M. E. Jung, B. Hoffmann, B. Rausch, J. M.
Contreras, Org. Lett. 2003, 5, 3159 – 3161.
[9] a) I. Paterson, D. J. Wallace, S. M. Velazquez, Tetrahedron Lett.
1994, 35, 9083 – 9086; b) I. Paterson, D. J. Wallace, Tetrahedron
Lett. 1994, 35, 9087 – 9090; c) I. Paterson, D. J. Wallace, C. J.
Cowden, Synthesis 1998, 639 – 652.
[10] J. Zhou, K. Burgess, Angew. Chem. 2007, 119, 1147 – 1149;
Angew. Chem. Int. Ed. 2007, 46, 1129 – 1131.
[11] T. Katsuki, K. B. Sharpless, J. Am. Chem. Soc. 1980, 102, 5974 –
5976.
[12] a) M. R. Johnson, Y. Kishi, Tetrahedron Lett. 1979, 20, 4347 –
4350; b) M. E. Jung, D. C. DꢀAmico, J. Am. Chem. Soc. 1997,
119, 12150 – 12158.
[13] No column chromatography is necessary since simple extraction
with water and diethyl ether removes the triflate salts, and the
triethylsilanol can be removed under reduced pressure to
provide pure silyl-protected aldehyde.
[14] W. R. Roush, J. Org. Chem. 1991, 56, 4151 – 4157.
[15] A. N. Rai, A. Basu, Tetrahedron Lett. 2003, 44, 2267 – 2269.
[16] I. Paterson, S. B. Blakey, C. J. Cowden, Tetrahedron Lett. 2002,
43, 6005 – 6008.
[17] Previously reported methods for the synthesis of the g-pyrone
through triketone cyclization proved inefficient for our system.
See: a) D. W. Jeffery, M. V. Perkins, J. M. White, Org. Lett. 2005,
7, 407 – 409; b) D. W. Jeffery, M. V. Perkins, J. M. White, Org.
Lett. 2005, 7, 1581 – 1584; c) H. Arimoto, S. Nishiyama, S.
Yamamura, Tetrahedron Lett. 1990, 31, 5619 – 5620; d) S. Yama-
mura, S. Nishiyama, Bull. Chem. Soc. Jpn. 1997, 70, 2025 – 2037.
[25] M. T. Crimmins, R. Omahony, J. Org. Chem. 1990, 55, 5894 –
5900.
[26] R. W. Hoffmann, Angew. Chem. 2000, 112, 2134 – 2150; Angew.
Chem. Int. Ed. 2000, 39, 2054 – 2070.
[27] For a similar spiroketalization onto a hemiketal, see: S. Favre, S.
Gerber-Lemaire, P. Vogel, Org. Lett. 2007, 9, 5107 – 5110.
[28] Prolonged reaction time or slow elution during column chro-
matography led to the epimerization of the acidic C14 proton.
[29] Slow elution during the column chromatography changed the
ratio of 22/23 from 3:1 to 1:1.
[30] The structure of this hemiketal was tentatively assigned since re-
protecting the alcohol group of the hemiketal 24 as the TES
ether afforded a compound that exhibited identical spectral data
to that of the hemiketal 23.
[31] The short reaction time is critical, since prolonged exposure led
to the formation of the ketal.
[32] This diketone was also stable to column chromatography and
existed as a single isomer.
Angew. Chem. Int. Ed. 2009, 48, 8766 –8769
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 8769