Asymmetric synthesis of the tricyclic core of cyathin diterpenoids via intramolecular Heck reaction

Article abstract of DOI:10.1016/j.tetlet.2005.07.157

The enantioselective synthesis of the ketones 3 which displays the carbon core of NGF-inducing cyathane diterpenes is described. The key tricyclic trienone 22 was assembled in 13 steps from Michael adduct (R)-8a via intramolecular Heck cyclization of the

Full text of DOI:10.1016/j.tetlet.2005.07.157

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 7263–7266
Asymmetric synthesis of the tricyclic core of cyathin
diterpenoids via intramolecular Heck reaction
a
a,b
Emmanuelle Drege, Georges Morgant and Didier Desmae¨lea,
*
`
a
ˆ
Unite´ de Chimie Organique Associe´e au CNRS, Faculte´ de Pharmacie, 5, rue J.-B. Cle´ment, 92290 Chatenay-Malabry, France
b
ˆ
Laboratoire de Cristallographie Bioinorganique, Faculte´ de Pharmacie, 5 rue J.-B. Cle´ment, 92296 Chatenay-Malabry, France
Received 13 July 2005; accepted 29 July 2005
Available online 6 September 2005
Abstract—The enantioselective synthesis of the ketones 3 which displays the carbon core of NGF-inducing cyathane diterpenes is
described. The key tricyclic trienone 22 was assembled in 13 steps from Michael adduct (R)-8a via intramolecular Heck cyclization
of the chiral triflate 21. The trienone 22 was further elaborated into ketone 3 through trimethylaluminum-promoted expansion of the
C-ring with trimethylsilyldiazomethane.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Cyathane diterpenes exemplified by erinacine A 1 are of
significant medicinal interest because they are potent
stimulators of nerve growth factor (NGF) synthesis.1
NGF is a potential treatment for degenerative neural
disorders such as Alzheimerꢀs disease and for promoting
peripheral nerve regeneration.2 However, NGFꢀs inabil-
ity to cross the blood–brain barriers limits its utility,
hence drugs that stimulate NGF synthesis by brain cells
show great promise as new medicines. Considering the
therapeutic potential of the erinacines, combined with
their scarce availability from natural sources, a general
synthetic route to these molecules and derivatives would
be of great interest. As a result, the cyathins diterpe-
noids have drawn much attention from the synthetic
community.3 We recently described an enantioselective
synthetic route to keto esters 2 in which the crucial con-
trol of the anti-relative stereochemistry between the
rings A and C is achieved by means of an intramolecular
Heck-type cyclization.4
Herein, we disclose an extension of our earlier work
allowing the implementation of the requisite isopropyl
group at C-3 and suitable functionalities in the seven-
membered C-ring, culminating in the synthesis of ketone
3. Our initial plan was to set the isopropyl group at the
outset of the synthetic scheme using keto ester (R)-8a as
the starting material. We assumed that the keto ester 8b
so obtained could be further elaborated into alkenyl tri-
flate 6, which in turn can undergo Heck cyclization to
give the C-nor-cyathane derivative 4. Finally, a regiose-
lective one-carbon ring expansion reaction should com-
plete the synthesis (Scheme 1).
The requisite keto ester 8b was synthetized in five steps
and 47% overall yield from (R)-8a (ee = 91%).5a Thus,
reaction of 8a with TMSOTf led to the expected silyl-
ated enol ether 9. Mukaiyama condensation with acetal-
dehyde, followed by sequential mesylate formation and
b-elimination with DBU produced enone 11 as a 5:1
E/Z mixture in 75% overall yield from 8a. This com-
pound was converted into the desired keto ester 8b (as
a 1:1 diastereomeric mixture at C-5) upon treatment
with lithium dimethylcuprate.
The starting material in this approach was the keto ester
(R)-8a, which is easily accessible in high optical purity
via the deracemizing Michael addition involving the chi-
ral imine derived from 2-methylcyclopentanone and (S)-
1-phenylethylamine.5
The next task was to further functionalize compound 8b
to allow the connection with the C-ring synthon 5.6 Our
initial plan involved activation of the keto group into
triflate enol ester and iododecarboxylation of the
propanoic side chain. Unfortunately, formation of
the enol triflate from the highly hindered keto group
of 8b required drastic conditions and furnished the
Keywords: Terpenes and terpenoids; Michael reactions; Heck reac-
tions; Ring transformation; Diazo compounds.
*
Corresponding author. Fax: +33 (0)146835752; e-mail: didier.
desmaele@cep.u-psud.fr
0040-4039/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.07.157

`
E. Drege et al. / Tetrahedron Letters 46 (2005) 7263–7266
7264
Scheme 2. Reagents and conditions: (a) 1.1 equiv TMSOTf, E3N,
CH2Cl2, 4 h, 20 ꢁC, 89%; (b) TiCl4, CH3CHO, À78 ꢁC, 2 h, 89%; (c)
MsCl, Et3N, DMAP, 0 ꢁC, 3 h, 95%; (d) DBU, toluene, 110 ꢁC, 1.5 h,
90%; (e) Me2CuLi, Et2O, À78 ꢁC, 2 h, 70%; (f) Tf2O, 2,6-(t-
Bu)2C5H3N, C2H4Cl2, 80 ꢁC, 10 h, 30%; (g) KOH, H2O, MeOH,
20 ꢁC, 12 h, 80%; (h) Pb(OAc)4, I2, visible light, CCl4, 80 ꢁC, 1.5 h,
65%; (i) NaBH4, CeCl3Æ7H2O, MeOH, 0 ꢁC, 3 h, 45%.
Scheme 1.
corresponding triflate in low yield. Furthermore, after
saponification of the ester group, the Barton modifica-
tion of the Kochi reaction7 (Pb(OAc)4/I2, refluxing
CCl4) failed to give the desired iodo triflate 7. On the
other hand, saponification of 8b followed by Kochi reac-
tion produced the iodo ketone 13 in 52% overall yield.
Unfortunately, this sensitive material was found to
decompose upon treatment with Tf2O.
the critical union of subunits A and C, we turned our
attention to the closing of the tricyclic ring system
through the Heck reaction-anion capture process we
previously described.4,8 Preparation of the alkenyl tri-
flate (S)-6 occurred smoothly by treatment of 17 with
triflic anhydride in refluxing 1,2-dichloroethane. How-
ever, the presence of the isopropyl group deeply affected
the reactivity of triflate 6 in Heck reactions. For exam-
ple, treatment of 6 with Pd2(dba)3ÆCHCl3 in the presence
of an acetate ion source as previously reported4 let the
starting material unchanged. All attempts at improving
this result using various palladium catalysts and reac-
tion conditions failed. We felt that the lack of reactivity
of 6 might be overcome by adding a carbonyl group at
the C-4 position of the cyclohexadienyl appendage.9
Consequently, we postponed the formation of the tri-
flate enol ether to a latter stage of the synthesis and
decided to explore the temporary shielding of the keto
group of 8b. Since we have previously shown that pro-
tection of the ketone with a dioxolane group impeded
the coupling with cyclohexadiene ester 5,4 we turned
to a reduction–protection sequence. However, new
problems stemmed from the steric hindrance around
the ketonic carbonyl group. Most usual reducing
reagents such as NaBH4 or NaBH4-CeCl3 reacted very
sluggishly with 8b, giving a mixture of alcohols 14, along
the corresponding lactones and over-reduction diols.
Therefore, this cumbersome route was not pursued fur-
ther (Scheme 2).
Allylic oxidation of (S)-6 with CrO3Æ3,5-DMP10 gave the
expected dienone (S)-19, albeit in very low yield.
Instead, products with an aromatic C-ring whose struc-
tures were not readily formulated were produced. By
contrast, allylic oxidation of the pivalate 20 (obtained
in 72% yield upon treatment of 6 with DIBAL-H, fol-
lowed by esterification with t-BuCOCl) provided the
desired dienone (S)-21 in 60% yield (Scheme 3). Treat-
ment of alkenyl triflate 21 with Pd(OAc)2/PPh3/
n-Bu4NBr in toluene at 120 ꢁC proceeded smoothly
affording trienone (+)-22 (70% isolated yield) with high
diastereoselectivity (95:5).11 The anti-relative configura-
After considerable experimentation, it was finally found
that the alkylation of the lithium enolate of cyclohexadi-
ene ester 5 with iodo ketone 13, provided the desired
keto ester 17 in 60% yield along with a small amount
of the volatile ether 18. Remarkably, the presence of
the unprotected keto group in 13 was not prejudicial to
the efficiency of the alkylation step. Having secured

`
E. Drege et al. / Tetrahedron Letters 46 (2005) 7263–7266
7265
Scheme 3. Reagents and conditions: (a) (i): LDA, THF, À78 ꢁC,
40 min, (ii): HMPA, 13, 1.5 h, 50 ꢁC, 60%; (b) Tf2O, 2,6-(t-Bu)2-
C5H3N, C2H4Cl2, 80 ꢁC, 8 h, 65%; (c) CrO3Æ3,5-DMP, CH2Cl2, 6 h,
20 ꢁC, 20%; (d) DIBAL-H, CH2Cl2, À78 ꢁC to 0 ꢁC, then 2 h, 0 ꢁC,
80%; (e) t-BuCOCl, py, DMAP, CH2Cl2, 15 h, 20 ꢁC, 90%; (f) CrO3,
3,5-DMP, CH2Cl2, 6 h, 20 ꢁC, 60%.
tion between rings A and C of the main isomer was
unambiguously established by X-ray crystallography
analysis of the epoxide 2312 derived from 22 by base
catalyzed epoxidation reaction. Interestingly, in this
process, attack of the hydroperoxide anion occurred
mainly (5:1) on the less hindered face of the molecule,
namely on the face opposite to the bulky pivalate
group.13 The remaining task was the one-carbon expan-
sion of the C-ring of trienone 22. However, this densely
functionalized material was found to have a tendency to
give aromatic decomposition products with a variety of
reagents, presumably through cleavage of the pivalate
group followed by retro-aldol type process. To over-
come that trend, we decided to reduce the less hindered
double bond of 22. Thus, homogeneous hydrogenation
of the trienone 22 using Wilkinson catalyst furnished
the desired dienone 24 in 75% yield.
Scheme 4. Reagents and conditions: (a) 7 mol % Pd(OAc)2, 15 mol %
PPh3, K2CO3, n-Bu4NBr, toluene, 2 h, 120 ꢁC, 70%; (b) NaOH 5%,
H2O2, MeOH, 25 ꢁC, 24 h, 50%; (c) (PPh3)3RhCl, H2, EtOH, 25 ꢁC,
36 h, 75%; (d) (i): TMSCHN2, AlMe3, THF, À78 ꢁC, then 36 h, 30 ꢁC,
(ii): HCl 3 N, acetone 2 h, 60%.
In summary, we have successfully assembled the tricyclic
carbon core of the cyathins in optically active form in 18
steps from 2-methylcyclopentanone. The key steps of our
synthesis were the enantioselective Michael addition to
settle the absolute configuration, intramolecular Heck
reaction to establish the crucial anti-stereochemistry of
the C-nor-cyathane 22, and the organoaluminum-pro-
moted diazoalkane mediated ring expansion of the C-
ring. Further elaboration of ketone (+)-3 into erinacine
A and other cyathane diterpenes, requires the reduction
of the angular hydroxymethyl group into a methyl group,
manipulation of the C-12 carbonyl to introduce the req-
uisite carbaldehyde and functionalization at C-14. Epox-
ide 23, which displays an oxygen atom with the correct b-
configuration at the future C-14 center might serve as a
suitable starting material for this purpose.
At this stage, it was felt that aluminum catalyzed carbe-
noid insertion was the most expeditious way to address
the challenge of regioselective expansion of the dienone
24 to the cyathane core. Thus, when the latter com-
pound was treated with trimethylsilyldiazomethane in
the presence of trimethylaluminum14 followed by hydro-
lysis of the resulting enol ether, the expanded ketone
(+)-3 was obtained in 60% isolated yield along
with 15% of the regioisomeric enone 25 (Scheme 4).15
To our knowledge, the preferential migration of the
unsaturated a-side of enones in such a process is
unprecedented, although a similar trend was previously
observed with 1-tetralone derivatives.16

`
E. Drege et al. / Tetrahedron Letters 46 (2005) 7263–7266
7266
J = 10.0, 1.3 Hz, 1H), 6.18 (d, J = 1.3 Hz, 1H), 4.27 (d,
Acknowledgements
J = 10.5 Hz, 1H), 4.05 (d, J = 10.5 Hz, 1H), 2.85 (hept,
J = 6.8 Hz, 1H), 2.50–2.40 (m, 2H), 1.94 (dt, J = 12.8,
2.7 Hz, 1H), 1.88 (ddd, J = 12.6, 6.8, 2.7 Hz, 1H), 1.84–
1.67 (m, 4H), 1.11 (s, 9H), 1.05 (d, J = 6.8, 3H), 0.96 (d,
J = 6.8 Hz, 3H), 0.96 (s, 3H); 13C NMR (CDCl3, 100
MHz) d 186.9 (C), 177.9 (C), 157.5 (C), 152.1 (CH), 147.1
(C), 136.5 (C), 129.7 (CH), 127.6 (CH), 66.2 (CH2), 49.9
(C), 45.3 (C), 39.7 (C), 39.6 (CH2), 36.1 (CH2), 31.8 (CH2),
29.0 (CH2), 27.1 (CH), 26.9 (3CH3), 23.5 (CH3), 21.5
(CH3), 21.4 (CH3).
The authors thank MENRT for providing financial sup-
port to E.D. We also thank Mrs S. Mairesse-Lebrun for
performing elemental analyses.
References and notes
1. (a) Kawagishi, H.; Shimada, A.; Shirai, R.; Okamoto, K.;
Ojima, F.; Sakamoto, H.; Ishiguro, Y.; Furukawa, S.
Tetrahedron Lett. 1994, 35, 1569–1572; (b) Kawagishi, H.;
Shimada, A.; Hosokawa, S.; Mori, H.; Sakamoto, H.;
Ishiguro, Y.; Sakemi, S.; Bordner, J.; Kojima, N.; Furu-
kawa, S. Tetrahedron Lett. 1996, 37, 7399–7402; (c) Ohta,
T.; Kita, T.; Kobayashi, N.; Obara, Y.; Nakahata, N.;
Ohizumi, Y.; Takaya, Y.; Oshima, Y. Tetrahedron Lett.
1998, 39, 6229–6232; (d) Obara, Y.; Nakahata, N.; Kita,
T.; Takaya, Y.; Kobayashi, H.; Hosoi, S.; Kiuchi, F.;
Ohta, T.; Oshima, Y.; Ohizumi, Y. Eur. J. Pharmacol.
1999, 370, 79–84; (e) Obara, Y.; Kobayashi, N.; Ohta, T.;
Ohizumi, Y.; Nakahata, N. Mol. Pharmacol. 2001, 59,
1287–1296.
12. Compound 23: Colorless crystals; mp 159 ꢁC (i-Pr2O); IR
(neat, cmÀ1) 1727, 1668, 1461, 1410, 1280; 1H NMR
(CDCl3, 400 MHz) d 5.86 (s, 1H), 4.22 (d, J = 11.2 Hz,
1H), 4.06 (d, J = 11.2 Hz, 1H), 3.50–3.45 (m, 2H), 2.76
(hept, J = 6.8 Hz, 1H), 2.45–2.40 (m, 2H), 2.29 (td,
J = 13.8, 4.8 Hz, 1H), 1.91 (dt, J = 12.8, 2.7 Hz, 1H),
1.86–1.60 (m, 4H), 1.13 (s, 9H), 1.03 (d, J = 6.8, 3H), 1.02
(s, 3H), 0.92 (d, J = 6.8 Hz, 3H); 13C NMR (CDCl3,
100 MHz) d 187.0 (C), 181.0 (C), 163.0 (C), 159.6 (C),
145.2 (C), 124.0 (CH), 66.8 (CH2), 60.7 (CH), 55.3 (CH),
47.9 (C), 46.0 (C), 39.0 (CH2), 38.9 (C), 36.1 (CH2), 29.7
(CH2), 29.2 (CH2), 27.0 (3CH3), 26.7 (CH), 23.5 (CH3),
21.5 (2CH3); Crystallographic data: crystal of
0.23 · 0.25 · 0.29 mm. C23H32O4, M = 372.50: ortho-
rhombic, space group P 21 21 21 (No. 19), Z = 4,
2. (a) Carswell, S. Exp. Neurol. 1993, 124, 36–42; (b) Riaz, S.
S.; Tomlinson, D. R. Prog. Neurobiol. 1996, 49, 125–143;
(c) Conner, J. M.; Tuszynski, M. H. Ment. Retard Dev.
Dis. 1998, 4, 212–222.
˚
a = 10.511(5), b = 12.617(5), c = 15.818(5) A, a = b =
c = 90ꢁ, V = 2097.9(15) A , d = 1.179 g cmÀ3, F(000) =
3
˚
˚
3. For a review see: (a) Wright, D. L.; Whitehead, C. R. Org.
Prep. Proced. Int. 2000, 32, 307–330; For more recent
approaches see: (b) Takeda, K.; Nakane, D.; Takeda, M.
Org. Lett. 2000, 2, 1903–1905; (c) Ward, D. E.; Gai, Y.;
Qiao, Q. Org. Lett. 2000, 2, 2125–2127; (d) Piers, E.;
Gilbert, M.; Cook, K. L. Org. Lett. 2000, 2, 1407–1410; (e)
Wender, P. A.; Bi, F. C.; Brodney, M. A.; Gosselin, F.
Org. Lett. 2001, 3, 2105–2108; (f) Takano, M.; Umino, A.;
Nakada, M. Org. Lett. 2004, 6, 4897–4900; (g) Ward, D.
E.; Gai, Y.; Qiao, Q.; Shen, J. Can. J. Chem. 2004, 82,
254–267; (h) Trost, B. M.; Dong, L.; Schroeder, G. M.
J. Am. Chem. Soc. 2005, 127, 2844–2845.
4. Thominiaux, C.; Chiaroni, A.; Desmae¨le, D. Tetrahedron
Lett. 2002, 43, 4107–4110.
5. (a) Pfau, M.; Revial, G.; Guingant, A.; dꢀAngelo, J. J. Am.
Chem. Soc. 1985, 107, 273–274; For a review, see: (b)
dꢀAngelo, J.; Desmae¨le, D.; Dumas, F.; Guingant, A.
Tetrahedron: Asymmetry 1992, 3, 459–505.
808, k = 0.710693 A(Mo Ka), l = 0.079 mmÀ1
;
5788
reflections measured (0 6 h 6 14, 0 6 k 6 17, 0 6 l 6 22)
on a Nonius CAD4 diffractometer. The structure was
solved with SIR92 and refined with CRYSTALS. Hydro-
gen atoms riding. Refinement converged to R(gt) = 0.0493
for the 1716 reflections having I P 2r(I), and
wR(gt) = 0.1041, goodness-of-fit S = 0.9117. Residual
3
˚
electron density: À0.25 and 0.29 eA . Crystallographic
results have been deposited (CIF file), in the Cambridge
Crystallographic Data Centre, UK, and allocated the
deposition number CCDC 267231.
13. The opposite trend was observed in the absence of the
C-5–C-10 double bond.4
14. Maruoka, K.; Concepcion, A. B.; Yamamoto, H. Synthe-
sis 1994, 1283–1290.
15. Compound 3: [a]D +285 (EtOH, c = 0.8); IR (neat, cmÀ1
)
1
2957, 2934, 2865, 1727, 1667, 1479, 1458, 1150; H NMR
(CDCl3, 400 MHz) d 5.30 (t, J = 6.5 Hz, 1H), 3.97 (d,
J = 11.2 Hz, 1H), 3.93 (d, J = 11.2 Hz, 1H), 3.18 (dd,
J = 14.4 Hz, J = 6.2 Hz, 1H), 3.48 (dd, J = 14.4 Hz, J =
6.7 Hz, 1H), 2.71 (hept, J = 6.8 Hz, 1H), 2.72–2.60 (m,
1H), 2.52 (ddd, J = 18.8 Hz, J = 8.9 Hz, J = 2.3 Hz, 1H),
2.35–2.25 (m, 3H), 1.95–1.50 (m, 6H), 1.38 (dt,
J = 13.8 Hz, J = 3.3 Hz, 1H), 1.79 (s, 9H), 0.94 (d,
J = 6.8, 3H), 0.92 (s, 3H), 0.90 (d, J = 6.8 Hz, 3H); 13C
NMR (CDCl3, 100 MHz) d 209.2 (C), 178.2 (C), 142.3 (C),
139.9 (C), 139.7 (C), 117.8 (CH), 66.8 (CH2), 48.7 (C), 44.8
(C), 41.6 (CH2), 38.9 (C), 38.7 (CH2), 37.8 (CH2), 36.4
(CH2), 31.7 (CH2), 30.7 (CH2), 28.3 (CH2), 27.3 (3CH3),
26.4 (CH), 23.4 (CH3), 21.6 (CH3), 21.3 (CH3); Anal. Calcd
for C24H36O3: C, 77.38; H, 9.74; Found: C, 77.15; H, 9.84.
16. Smalley, T. L., Jr. Synth. Commun. 2004, 34, 1973–1980.
6. Rabideau, P. W.; Huser, D. L.; Nyikos, S. J. Tetrahedron
Lett. 1980, 21, 1401–1404.
7. Barton, D. H. R.; Faro, H. P.; Serebryakov, E. P.;
Woosley, N. F. J. J. Chem. Soc. 1965, 2438–2443.
8. (a) Kagechika, K.; Shibasaki, M. J. Org. Chem. 1991, 56,
4093–4094; (b) Kagechika, K.; Ohshima, T.; Shibasaki, M.
Tetrahedron 1993, 49, 1773–1782.
9. Kondo, K.; Sodekoa, M.; Mori, M.; Shibasaki, M.
Synthesis 1993, 920–930.
10. Salmond, W. G.; Barta, M. A.; Havens, J. L. J. Org.
Chem. 1978, 43, 2057–2059.
11. Compound 22: Colorless oil; [a]D +82.5 (EtOH, c = 2); IR
(neat, cmÀ1) 1734, 1658, 1624, 1450, 1239, 1209; 1H NMR
(CDCl3, 400 MHz) d 6.79 (d, J = 10.0 Hz, 1H), 6.34 (dd,