Enantioselective total synthesis of (+)-colletoic acid via catalytic asymmetric intramolecular cyclopropanation of an α-diazo-β-keto diphenylphosphine oxide

Article abstract of DOI:10.1021/ol303459x

The enantioselective total synthesis of (+)-colletoic acid, a potent naturally occurring 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) inhibitor, is described. This total synthesis features a highly enantioselective catalytic asymmetric intramolecular cyclopropanation of an α-diazo-β-keto diphenylphosphine oxide and five highly stereoselective reactions (cyclopropane opening, Diels-Alder reaction, iodolactonization, alkene formation, and reduction of α,β-unsaturated carboxylic acid).

Full text of DOI:10.1021/ol303459x

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Enantioselective Total Synthesis
of (þ)-Colletoic Acid via Catalytic Asymmetric
Intramolecular Cyclopropanation of an
r‑Diazo-β-keto Diphenylphosphine Oxide
Takashi Sawada and Masahisa Nakada*
Department of Chemistry and Biochemistry, School of Advanced Science and
Engineering, Waseda University, 3-4-1 Ohkubo, Shinjuku-ku, Tokyo 169-8555, Japan
mnakada@waseda.jp
Received December 18, 2012
ABSTRACT
The enantioselective total synthesis of (þ)-colletoic acid, a potent naturally occurring 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1)
inhibitor, is described. This total synthesis features a highly enantioselective catalytic asymmetric intramolecular cyclopropanation of an R-diazo-β-keto
diphenylphosphine oxide and five highly stereoselective reactions (cyclopropane opening, DielsꢀAlder reaction, iodolactonization, alkene formation,
and reduction of R,β-unsaturated carboxylic acid).
(þ)-Colletoic acid (Figure 1) was isolated from a fungus
identified as Colletotrichum gloeosporioides SANK 21404
by a research group at Daiichi Sankyo Co., Ltd.1 The relative
stereochemistry of (þ)-colletoic acid was confirmed by X-ray
crystallographic analysis, and its absolute structure was
elucidated by a modified Mosher’s method.1 (þ)-Colletoic
acid is a potent naturally occurring inhibitor of human 11β-
hydroxysteroid dehydrogenase type 1 (11β-HSD1) with an
IC50 of 13 nM and a Ki value of 3.9 nM but exhibits almost no
effect on 11β-HSD2 (IC50 = >10000 nM).
Although 11β-HSD1 is an NADP(H)-dependent oxi-
doreductase, it primarily acts as a reductase in intact cells,
converting inactive cortisone to cortisol, a potent bioactive
hormone in humans. Cortisol elevates blood glucose levels
by increasing glucose production in the liver as well as by
inhibiting uptake and disposal of glucose in muscle and
adipose tissues.2
Consequently, it has been suggested that metabolic
syndrome may be attributed to increased levels of cortisol,
which may be derived from elevated 11β-HSD1 activity,
and that inhibition of 11β-HSD1 may reduce cortisol. Thus,
11β-HSD1 has been a promising target for the treatment of
diseases associated with metabolic syndrome.3
(þ)-Colletoic acid is a member of the acorane-type
sesquiterpenes,4 a class of molecules featuring a spiro-
[4.5]decane core, which consists of a six-membered ring
containing a methallyl alcohol and a five-membered ring
bearing cis-oriented isopropyl and carboxyl groups. The
core of (þ)-colletoic acid possesses four stereogenic centers,
three of which are successive and incorporate the all-carbon
quaternary stereogenic spiro center.
(3) (a) Wang, M. Curr. Opin. Invest. Drugs 2006, 7, 319. (b) Barf, T.;
Williams, M. Drugs Future 2006, 31, 231. (c) Fotsch, C.; Askew, B. C.;
Chen, J. J. Expert Opin. Ther. Patents 2005, 15, 289. (d) Chrousos, G. P.
Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 6329. (e) Walker, B. R.; Seckl,
J. R. Expert Opin. Ther. Targets 2003, 7, 771. (f) Masuzaki, H.; Flier,
J. S. Curr. Drug Targets Immune Endocr. Metabol. Disord. 2003, 3, 255.
(g) Stewart, P. M. Eur. J. Endocrinol. 2003, 149, 163.
(4) For recently reported acorane-type sesquiterpenes, see: (a) Sun,
L.; Li, D.-L.; Tao, M.-H.; Dan, F.-J.; Zhang, W.-M. Helv. Chim. Acta
2012, 95, 157. (b) Song, T.-F.; Zhang, W.-D.; Xia, X.-H.; Shen, Y.-H.;
Liu, C.-M.; Lin, S.; Jin, H.-z.; Li, H.-L. Arch. Pharm. Res. 2009, 32, 1233.
(c) Nagashima, F.; Suzuki, M.; Takaoka, S.; Asakawa, Y. J. Nat. Prod.
2001, 64, 1309.
(1) Aoyagi, A.; Ito-Kobayashi, M.; Ono, Y.; Furukawa, Y.; Takahashi,
M.; Muramatsu, Y.; Umetani, M.; Takatsu, T. J. Antibiot. 2008,
61, 136.
(2) (a) Draper, N.; Stewart, P. M. J. Endocrinol. 2005, 186, 251.
(b) Thieringer, R.; Hermanowski-Vosatka, A. Expert Rev. Cardiovasc.
2005, 3, 911. (c) Morton, N. M.; Paterson, J. M.; Masuzaki, H.; Holmes,
M. C.; Staels, B.; Fievet, C.; Walker, B. R.; Flier, J. S.; Mullins, J. J.;
Seckl, J. R. Diabetes 2004, 53, 931.
r
10.1021/ol303459x
XXXX American Chemical Society

Scheme 1
Figure 1. Structure of (þ)-colletoic acid.
To the best of our knowledge, (þ)-colletoic acid is the
first acorane-type sesquiterpene with a carboxyl group at
the C1 position, which has been reported to be a crucial
moiety for 11β-HSD1 inhibitory activity since all other
acorane-type sesquiterpenes and (þ)-colletoic acid methyl
ester have no 11β-HSD1 inhibitory activity.1 The potent
11β-HSD1 inhibitory activity, unique structure, and struc-
tureꢀactivity relationships of (þ)-colletoic acid make it an
attractive target. Herein, we report the first enantioselec-
tive total synthesis of (þ)-colletoic acid.
We select lactone 1 (Scheme 1) as the synthetic inter-
mediatefor (þ)-colletoicacidbecause the C1ꢀC2electron-
deficient alkene would preferentially undergo nucleophilic
reduction at the less-hindered face to afford the desired
product. Lactone 1 was envisaged to be prepared from 2
via electrophile-induced lactonization and subsequent al-
kene formation. Carboxylic acid 2 would be obtained by
palladium-catalyzed carbonylation of enol triflate 3, which
could be prepared from the corresponding ketone. Con-
struction of the scaffold of 3 was envisioned via Dielsꢀ
Alder reaction of enone 4 with isoprene because it could
preferentially occur at the less-hindered side of 4. Enone 4
was thought to be obtained by the reaction of keto sulfone
5, which would be prepared from cyclopropane 6.
We reported the catalytic asymmetric intramolecular
cyclopropanation (CAIMCP) of R-diazo-β-keto sulfone 7,
which successfully affords 6 in high yield and ee (Scheme 2).5
Moreover, enantiopure 6 is easily available by a single
recrystallization owing to its highly crystalline nature.
Hence, we initially studied the total synthesis of (þ)-colletoic
acid starting from 6.
Scheme 2
β-keto aryl sulfone.7 Under the reported conditions, how-
ever, 9 was not formed. Moreover, reductive enolate forma-
tion from 5 with lithium naphthalenide or SmI2 and
subsequent reaction with formaldehyde or its equivalents8
did not afford 10, and the corresponding desulfonylated
product was formed instead. The same results were obtained
in the reaction of less bulky phenyl sulfone 5a, andattempted
JuliaꢀKociensky-type reactions of 5b9 and 5c10 also failed to
provide the desired products.
The cyclopropane-opening reaction of 6 proceeded with
several reducing reagents, but in most cases, both 5 and the
corresponding desulfonylated product were formed. The
reaction with tri-n-butyltin hydride in the presence of
AIBN, however, provided 5 as a single isomer in 67%
yield (Scheme 3). It was expected that β-keto sulfone 5
would afford compound 9 upon reaction with formalde-
hyde or an equivalent because the same reaction has been
reported for the corresponding β-keto esters6 and a simple
As no promising results were obtained in the reaction of
5, we turned our attention to β-keto phosphine oxide 11
(Scheme 3), which should undergo a Horner reaction of the
β-keto phosphine oxide with an aldehyde. Compound 11 was
thought to be prepared by the same method as 5. However, to
the best of our knowledge, no CAIMCP of R-diazo-β-keto
diphenylphosphine oxide has been reported thus far. Hence,
we decided to develop the CAIMCP of R-diazo-β-keto
(7) Yang, J.; Li, H.; Li, M.; Peng, J.; Gu, Y. Adv. Synth. Catal. 2012,
(5) (a) Honma, M.; Sawada, T.; Fujisawa, Y.; Utsugi, M.; Watanabe,
H.; Umino, A.; Matsumura, T.; Hagihara, T.; Takano, M.; Nakada, M.
J. Am. Chem. Soc. 2003, 125, 2860. (b) Honma, M.; Takeda, H.; Takano,
M.; Nakada, M. Synlett 2009, 1695.
354, 688.
(8) Deguest, G.; Bischoff, L.; Fruit, C.; Marsais, F. Org. Lett. 2007,
9, 1165.
(9) Mirk, D.; Grassot, J.-M.; Zhu, J. Synlett 2006, 1255.
(6) (a) Kumamoto, H.; Kato, K.; Haraguchi, K.; Tanaka, H.;
Nitanda, T.; Baba, M.; Dutschman, G. E.; Cheng, Y.-C. Nucleic Acids
Symp. 2004, 48, 41. (b) Bica, K.; Gaertner, P. Eur. J. Org. Chem. 2008,
20, 3453.
ꢀ
(10) (a) Alonso, D. A.; Najera, C.; Varea, M. Tetrahedron Lett. 2004,
ꢀ
45, 573. (b) Alonso, D. A.; Fuensanta, M.; Najera, C.; Varea, M. J. Org.
Chem. 2005, 70, 6404.
B
Org. Lett., Vol. XX, No. XX, XXXX

Table 1. CAIMCP of R-Diazo-β-keto Diphenylphosphine
Oxide 12
Scheme 3
entry
ligand
CuX
time (h)
yielda (%)
eeb (%)
1
2
3
4
5
6
7
8
9
8a
8b
8c
8d
8e
8f
CuOTfc
CuOTfc
CuOTfc
CuOTfc
CuOTfc
CuOTfc
CuOTfc
CuOTfc
1.5
16
4
43
37
90
59
51
43
44
64
79
63
73
75
86
75
81
84
84
91
4
3
diphenylphosphine oxide because the CAIMCP of R-diazo-
β-keto arylsulfones we previously developed indicated that
the bulky diphenylphosphine oxide group could control the
enantioselectivity of the CAIMCP.
19
5
8g
8h
8d
30
12
d
CuBF4
R-Diazo-β-keto diphenylphosphine oxide 12 (Scheme 4)
was prepared through the reaction of ethyl 5-methylhex-4-
enoate11 with lithiated methyl diphenylphosphine oxide
and subsequent diazo-transfer.
a Isolated yields. b ee determined by HPLC. For conditions, see the
SupportingInformation. c (CuOTf)2 toluene was used. d CuBF4 4MeCN
was used.
3
3
The CAIMCP of R-diazo-β-keto sulfones developed by
us revealed that the enantioselectivity could be improved
by changing the structure of the bisoxazoline ligands.
Hence, the CAIMCP of 12 was examined using a variety
of bisoxazoline ligands 8aꢀh12 bearing an isopropyl group
on the oxazoline and different substituents at the bisoxazo-
line junction (Table 1). The reactions with ligands 8aꢀc
(15 mol %) and CuOTf (10 mol %) afforded desired product
13 with ee’s in the range of 63ꢀ75% (entries 1ꢀ3). Better
results were obtained from the reactions with ligands 8dꢀh
and CuOTf (entries 4ꢀ8). Finally, 13 (91% ee) was
obtained by the reaction with ligand 8d (15 mol %) and
CuBF4 (10 mol %) in 79% yield (entry 9).
the absolute configuration of 13 was as shown in Table 1
and that sense of enantioselectivity of the CAIMCP of 12
was the same as those of R-diazo-β-keto sulfones. There-
fore, the enantioselective outcome would be well explained
by our proposed model.5a
With enantiopure 13 in hand, the total synthesis of (þ)-
colletoic acid was investigated. The reductive ring-opening
reaction of cyclopropane 13 proceeded with tri-n-butyltin
hydride under the same conditions used in the reaction of 6
(Scheme 5). The reaction of 11 with paraformaldehyde
under Masamune conditions14 afforded 4 in 85% yield.
The DielsꢀAlder reaction of 4 and isoprene at elevated
temperature resulted in decomposition of 4. However, we
found that the reaction in the presence of aluminum
chloride as a Lewis acid proceeded at ꢀ20 °C to afford
14 as a single product. The structure of 14 was elucidated
by NMR studies as shown in Scheme 5. Ketone 14 was
then converted to the enol triflate and subsequent palla-
dium-catalyzed carbonylation afforded 2. The reaction of
2 with iodine and NaHCO3 in aqueous THF induced
iodolactonization of the carboxylic acid to afford the
iodolactone as a single product, which underwent regiose-
lective alkene formation to afford 1 as a single product
upon treatment with DBU in THF at 50 °C.
Scheme 4
With 1 in hand, the chemo- and stereoselective reduction
of the C1ꢀC2 alkene was examined. Several reducing
reagents for 1,4-reduction were surveyed, and L-Selectride
was found to effectively reduce 1 to afford the product as a
single isomer, which was subjected to hydrolysis under
Compound 13 was highly crystalline, and 58% of en-
antiopure 13 was obtained by a single recrystallization. Its
absolute configuration was elucidated by X-ray crystal-
lographic analysis.13 The crystal structure indicated that
(11) Cane, D. E.; Tandon, M. Tetrahedron Lett. 1994, 35, 5355.
(12) Sawada, T.; Nakada, M. Tetrahedron: Asymmetry 2012, 23, 350.
(13) See the Supporting Information.
(14) Blanchette, M. A.; Choy, W.; Davis, J. T.; Essenfeld, A. P.;
Masamune, S.; Roush, W. R.; Sakai, T. Tetrahedron Lett. 1984, 25, 2183.
Org. Lett., Vol. XX, No. XX, XXXX
C

Scheme 5
Scheme 6
to (þ)-colletoic acid in all respects (1H and 13C NMR, IR,
MS, and [R]D), thus confirming the enantioselective total
synthesis of (þ)-colletoic acid.
In summary, the first enantioselective total synthesis of
(þ)-colletoic acid has been achieved. The total synthesis
relies on an approach via the highly enantioselective
CAIMCP of R-diazo-β-keto diphenylphosphine oxide,
which has been developed in this study. Further elabora-
tion from the CAIMCP product to (þ)-colletoic acid
features highly controlled reactions: regioselective cyclopro-
pane-opening reaction of the CAIMCP product, stereose-
lective DielsꢀAlder reaction to construct the spiro[4.5]-
decane core, regioselective iodolactonization and subse-
quent alkene formation by DBU treatment, and chemo-
and stereoselective Birch reduction of the C1ꢀC2 double
bond to yield (þ)-colletoic acid. It is noteworthy that all the
above reactions afforded a single isomer and no protecting
groups were used in this total synthesis.
basic conditions to afford the final product. However, the
1H NMR spectrum of the synthesized product differed
from the reported data of (þ)-colletoic acid, which sug-
gested the formation of C1-epi-(þ)-colletoic acid.15
Acknowledgment. We thank Daiichi Sankyo RD Novare
Co., Ltd. for kindly donating (þ)-colletoic acid. This work
was financially supported in part by a Grant-in-Aid for
Scientific Research on Innovative Areas “Organic Synthesis
Based on Reaction Integration. Development of New Meth-
ods and Creation of New Substances” (No. 2105).
We speculated that the conformation of the δ-lactone in
1 would be restricted by the tricyclic ring system, resulting
in formation of the thermodynamically more stable un-
desired C1-epimer. Consequently, we examined the reduc-
tion of the C1ꢀC2 double bond of the hydroxy carboxylic
acid derived from 1 (Scheme 6). As a result, Birch reduc-
tion of the hydroxy carboxylic acid successfully afforded
the single isomer. The final product proved to be identical
Supporting Information Available. Experimental pro-
cedures and compound characterization data. This ma-
terial is available free of charge via the Internet at http://
pubs.acs.org.
(15) C1-epi-(þ)-Colletoic acid was unstable and spontanously con-
verted to the corresponding lactone in part during isolation probably
because the carboxyl group is close to the hydroxyl group.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX