Enantioselective synthesis of the C-14 to C-5 cyclopentane segment of jatrophane diterpenes

Article abstract of DOI:10.1016/j.tetlet.2003.10.171

The enantioselective synthesis of the C-14 to C-5 cyclopentane segment of jatrophane diterpenes is reported. An Evans aldol addition, a Horner-Wadsworth-Emmons olefination and a thermal intramolecular carbonyl ene reaction of an α-keto ester served as key C/C-connecting transformations.

Full text of DOI:10.1016/j.tetlet.2003.10.171

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 289–292
Enantioselective synthesis of the C-14 to C-5 cyclopentane segment
of jatrophane diterpenes
Hannes Helmboldt, Julia Rehbein and Martin Hiersemann*
Institut f€ur Organische Chemie, Technische Universit€at Dresden, D-01062 Dresden, Germany
Received 25 September 2003; revised 18 October 2003; accepted 29 October 2003
Abstract—The enantioselective synthesis of the C-14 to C-5 cyclopentane segment of jatrophane diterpenes is reported. An Evans
aldol addition, a Horner–Wadsworth–Emmons olefination and a thermal intramolecular carbonyl ene reaction of an a-keto ester
served as key C/C-connecting transformations.
Ó 2003 Elsevier Ltd. All rights reserved.
The milky latex of plants belonging to the genus
Euphorbia (family Euphorbiaceae) is a rich source of
jatrophane diterpenes featuring the bicyclo[10.3.0]pen-
tadecane framework 1 (Fig. 1).
studies of the biological activities of jatrophane diter-
penes revealed promising results. For instance, jatro-
phane 2 from E. pubescens showed a cell type selective
growth inhibitory effect on the cancer cell line NCI-
H460 (nonsmall cell lung cancer).1 Jatrophane 3 from
E. semiperfoliata exhibited microtubule-interacting prop-
erties and influenced p53 expression and Raf-1/Bcl-2
activation.2 Jatrophane 4 from E. dendroides inhibited
the P-glycoprotein-mediated efflux of daunomycin from
human K562/R7 leukaemia cells.3
Jatrophanes are being isolated from Euphorbia species
with remarkable diversity with respect to hydroxylation
and/or acyloxylation of the carbon skeleton. Recent
O
14
AcO
1
The validated biological activities and structural diver-
sity of jatrophanes qualifies them as privileged structures
for drug discovery.4 A diversity-oriented synthetic
approach towards natural and nonnatural jatrophanes
requires a highly convergent retrosynthetic analysis. Our
synthetic plan considers the C-14 to C-5 cyclopentane
fragment 5 as a building block of central importance for
an efficient synthetic access to jatrophane diterpenes.
10
4
H
6
OAc
AcO
CH3CH2CH2COO
2
1
AcO
HO
O
HO
H
H
We envisioned that a thermal intramolecular carbonyl
ene reaction could convert the easily accessible acyclic
a-keto ester 6 to a highly substituted 1-hydroxycyclo-
pentane carboxylic acid ester 5 (Fig. 2).5 Based on a
simple qualitative analysis of the concerted transition
state geometry for the ene reaction, we expected that the
absolute configuration of C-3 of the a-keto ester 6
(jatrophane numbering is used throughout this letter)
would be decisive for the substrate-induced diastereo-
selectivity of the ene reaction (see Fig. 3).
O
OAc
OAc
BzO
HO
AcO
(CH3)2CHCOO
AcO
BzO
3
4
Figure 1. Jatrophane framework 1 and selected jatrophanes from
Euphorbia pubescens 2, E. semiperfoliata 3, E. dendroides 4.
Keywords: Diterpene; Jatrophane; Euphorbia; Total synthesis; Car-
bonyl ene reaction.
Therefore, we decided to utilize the (3R)-configured
a-keto ester 6 as substrate for the ene reaction and to
* Corresponding author. Tel.: +49-351-463-35480; fax: +49-351-463-
33661; e-mail: martin.hiersemann@chemie.tu-dresden.de
0040-4039/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2003.10.171

290
H. Helmboldt et al. / Tetrahedron Letters 45 (2004) 289–292
O
O
O
OH
O
HO
1
a,b
O
OR2
H3CO
O
N
14
2, 3, 4
3
8
9
(62%)
5
R1O
5
c,d,e
OR1
O
nBuLi, THF, (AlkO)2P
-78 °C, 5 min
then add 7
OR2
OTBS
O
O
O
11a: Alk= Me, R= Me
11b: Alk= Et, R= iPr
OR2
O
7a: R1= TPS (78%)
7b: R1= TBS (80%)
7c: R1= BOM (68%)
R1O
R1O
7
6
OR1
Figure 2.
TBSO
R2O2C
10a-f (E/Z > 10/1)
6a: R1= TPS, R2= Me (86%)
6b: R1= TPS, R2= iPr (68%)
6c: R1= TBS, R2= Me (76%)
6d: R1= TBS, R2= iPr (76%)
6e: R1= BOM, R2= Me (57%)
6f: R1= BOM, R2= iPr (74%)
subsequently invert the absolute configuration at C-3.
The a-ketoester 6 should be easily obtainable from the
aldehyde 7 by chain elongation. The synthesis of a-keto
esters 6a–f with different protecting groups R1 and ester
substituents R2 would enable an investigation concern-
ing the influence of steric effects of substituents on the
diastereoselectivity of the critical ene reaction.
TBAF, THF
- 78 °C, 10 min
O
OR1
R2O
O
6a-f, yield from 7a-c
The enantioselective synthesis of the a-keto esters 6a–f is
outlined in Scheme 1. Evansꢀ reliable aldol methodology
followed by nucleophilic cleavage of the auxiliary
afforded the diastereomerically pure ester 9.6;7 Protec-
tion of the C-3-hydroxyl group and subsequent redox
chemistry provided the aldehydes 7a–c.8 Horner–
Wadsworth–Emmons olefination of the aldehydes 7a–c
with the lithiated 2-(tert-butyldimethylsiloxy)-phospho-
noacetates 11a,b proceeded smoothly to afford the pre-
dominantly E-configured silyl enol ethers 10a–f.9;10
Desilylation of the silyl enol ethers 10a–f under carefully
optimized reaction conditions afforded the enantiomer-
ically pure a-keto esters 6a–f11 in good yields.12
Scheme 1. TPS ¼ t-BuPh2Si, TBS ¼ t-BuMe2Si, BOM ¼ PhCH2OCH2.
Reagents and conditions: (a) n-Bu2BOTf, Et3N, CH2Cl2, 0 °C, 10 min;
(E)-H3CCH@CHCHO, )78 °C to rt. (b) NaOMe, MeOH, 0 °C,
20 min. (c) (i) TPSCl, imidazole, CH2Cl2, 0 °C to rt; (ii) TBSCl,
imidazole, 0 °C to rt; (iii) BOMCl, EtiPr2N, TBAI, CH2Cl2, 0 °C to rt.
(d) DIBAH, CH2Cl2, )78 °C, 0.5 h. (e) SO3Æpyridine, DMSO, CH2Cl2,
0 °C to rt.
decane
180 ˚C
3 d
O
O
HO
HO
1
OR2
OR2
14
15
4
+
6a-f
The pivotal intramolecular type I carbonyl ene reaction
was performed under thermal conditions.13 The a-keto
esters 6a–f were heated to 180 °C in decane in a sealed
tube to provide the corresponding cyclopentanes as a
mixture of the two diastereomers 5a–f and 12a–f
(Scheme 2). We observed the exclusive formation of the
4,15-cis configuration. The two cis diastereomers were
formed as a 5/1 mixture in favor of the desired
(3R,4S,15R)-configured cyclopentanes 5a–f.14 Some-
what surprisingly, neither the nature of R2 nor the
nature of the protecting group R1 significantly influ-
enced the outcome of the ene reaction. A brief study of
the connection between the conversion of 6c to 5,12c
and the diastereoselectivity of this transformation
revealed, that the 5/1 ratio of diastereomers remains
virtually unchanged during the course of the reaction.15
In order to avoid extensive decomposition, the reaction
was stopped after 3 days without complete conversion of
the substrate 6c.
R1O
R1O
5
(3R,4S,15R)-5a-f : (3R,4R,15S)-12a-f= 5:1
5,12a: R1= TPS, R2= Me (73%)
5,12b: R1= TPS, R2= iPr (69%)
5,12c: R1= TBS, R2= Me (79%)
5,12d: R1= TBS, R2= iPr (75%)
5,12e: R1= BOM, R2= Me (69%)
5,12f: R1= BOM, R2= iPr (79%)
diastereomers
separable by
chromatography
Scheme 2.
the desired (3R,4S,15R)-configured cyclopentanes 5c–f16
are accessible as single diastereomers, even on a larger
scale.17 Heating either the separated minor or the major
diastereomer of 5c in decane to 180 °C for several days
led to the formation of a 5/1 mixture of 5c and 12c by a
retro-ene/ene reaction. This finding indicates that the
cyclopentanes 5a–f are the thermodynamically more
stable products of the thermal ene reactions of 6a–f.15
The 4,15-cis-configured diastereomers 5c–f and 12c–f
were separable by flash chromatography and, therefore,

H. Helmboldt et al. / Tetrahedron Letters 45 (2004) 289–292
291
The kinetic preference for the cis-configured cyclopen-
tanes 5,12a–f may be explained as depicted in Figure 3.
We assume that the ene reaction proceeds through a
concerted bicyclo[4.3.0]nonane-like transition state 13.
The simple diastereoselectivity is a consequence of the
favourable less strained cis-annulated bicyclic transition
state 13 (the corresponding trans-annulated transition
states are not depicted).15
absolute configuration at C-2. Work towards the
completion of jatrophane total syntheses is underway in
our laboratory and will be reported in due course.
Acknowledgements
Financial support by the DFG and the FCI is gratefully
acknowledged.
6a-f
H3C
OR1
4
H
H
O
H
H
H
15 CO2R2
OR1
15
References and Notes
O
4
CH3
CO2R2
H
1. Valente, C.; Ferreira, M. J. U.; Abreu, P. M.; Pedro, M.;
Cerqueira, F.; Nascimento, M. S. J. Planta Medica 2003,
69, 361–366.
(3R,4Si,15Re)-13
(3R,4Re,15Si)-13
2. Miglietta, A.; Gabriel, L.; Appendino, G.; Bocca, C.
Cancer Chemother. Pharmacol. 2003, 51, 67–74.
(3R,4S,15R)-5a-f
(3R,4R,15S)-12a-f
3. Corea, G.; Fattorusso, E.; Lanzotti, V.; Taglialatela-
Scafati, O.; Appendino, G.; Ballero, M.; Simon, P.-N.;
Dumontet, C.; Di Pietro, A. J. Med. Chem. 2003, 46,
3395–3402.
Figure 3.
4. (a) Aylward, J. H.; Parsons, P. G. Int. Pat. WO02/11743
A2 (A3), 2002; Chem. Abstr. 2002, 136, 177959x; (b)
Aylward, J. H.; Parsons, P. G.; Suhrbier, A.; Turner, K.
A. Int. Pat. WO01/93885 A1, 2001; Chem. Abstr. 2002,
136, 31680c; (c) Aylward, J. H.; Parsons, P. G.; Suhrbier,
A.; Turner, K. A. WO01/93884 A1, 2001; Chem. Abstr.
2002, 136, 42795v; (d) Aylward, J. H.; Parsons, P. G.;
Suhrbier, A.; Turner, K. A. WO01/93883 A1, 2001; Chem.
Abstr. 2002, 136, 31656z; (e) Aylward, J. H. WO99/08994
A1, 1999; Chem. Abstr. 1999, 130, 205113p.
5. Studies concerning the intramolecular carbonyl ene reac-
tion of a-keto esters are scarce. For examples under
thermal conditions, see: (a) Hiersemann, M. Eur. J. Org.
Chem. 2001, 483–491; (b) Hiersemann, M. Synlett 2000,
415–417; For catalytic asymmetric procedures, see: (c)
Kaden, S.; Hiersemann, M. Synlett 2002, 1999–2002;
(d) Yang, D.; Yang, M.; Zhu, N. Org. Lett. 2003, 5, 3749–
3752.
As envisioned, the (3R)-configuration of the a-keto
esters 6a–f induced the required (4R,15S)-configuration
of the ene reaction products 5a–f (Fig. 3). However, the
synthesis of the C-14 to C-5 fragment of the jatrophane
diterpenes required the (3S)-configuration. We envi-
sioned a Mitsunobu inversion as the method of choice
for a stereospecific inversion of the configuration at
C-3.18 The plan was realized using 5d as representative
starting material (Scheme 3). The cleavage of the TBS
ether on the secondary hydroxyl group proceeded
uneventfully with HF in pyridine. The succeeding
Mitsunobu inversion afforded the desired C-14 to C-5
segment (2S,3S,4S,15R)-1419 of the jatrophane diter-
penes featuring the correct absolute configuration at all
four chiral carbon atoms.
6. (a) Evans, D. A.; Nelson, J. V.; Vogel, E.; Taber, T. R.
J. Am. Chem. Soc. 1981, 103, 3099–3111; (b) Evans, D. A.;
Bartroli, J.; Shih, T. L. J. Am. Chem. Soc. 1981, 103, 2127–
2129.
7. The aldol addition was realized as described in (a) Smith,
A. B.; Minbiole, K. P.; Verhoest, P. R.; Schelhaas, M.
J. Am. Chem. Soc. 2001, 123, 10942–10953; the auxiliary
was removed as described in (b) Sunazuka, T.; Hirose, T.;
Shirahata, T.; Harigaya, Y.; Hayashi, M.; Komiyama, K.;
Omura, S.; Smith, A. B. J. Am. Chem. Soc. 2000, 122,
2122–2123.
In summary, the first enantioselective synthesis of the
C-14 to C-5 segment 14 of the jatrophane diterpenes 2–
4 is reported. The intramolecular carbonyl ene reaction
of an a-keto ester 6 afforded the desired cyclopentane
framework 5. Thereby, the two crucial chiral centres at
C-15 and C-4 were created with a complete simple
(cis/trans) and an acceptable substrate-induced diaste-
reoselectivity. Evansꢀ reliable aldol addition secured
access to the a-keto ester 6 featuring the desired
8. Parikh–Doering oxidation, see: Parikh, J. R.; Doering, W.
v. E. J. Am. Chem. Soc. 1967, 89, 5505–5507.
9. The phosphonates 11a,b were synthesized according to
(a) Horne, D.; Gaudino, J.; Thompson, W. J. Tetrahedron
Lett. 1984, 25, 3529–3532; The HWE reaction was realized
according to (b) Ouellet, L.; Langlois, P.; Deslongchamps,
P. Synlett 1997, 689–699; further examples for the
construction of an enol ether double bond by a HWE
reaction, see: (c) Grell, W.; Machleidt, H. Liebigs Ann.
Chem. 1966, 699, 53–59; (d) Ireland, R. E.; Mueller, R. H.;
Willard, A. K. J. Am. Chem. Soc. 1976, 98, 2868–2877; (e)
Ireland, R. E.; Mueller, R. H.; Willard, A. K. J. Org.
Chem. 1976, 41, 986–996; (f) Nakamura, E. Tetrahedron
1. HF py, THF
2. PPh3, DIAD
4-Br-PhCO2H
(63%)
O
14
HO
1
OiPr
5d
(3S)
5
4-Br-BzO
(2S,3S,4S,15R)-14
Scheme 3.

292
H. Helmboldt et al. / Tetrahedron Letters 45 (2004) 289–292
Lett. 1981, 22, 663–666; (g) Overman, L. E.; Rogers, B. N.;
Tellew, J. E.; Trenkle, W. C. J. Am. Chem. Soc. 1997, 119,
7159–7160; (h) Paterson, I.; McLeod, M. D. Tetrahedron
ꢀ
Lett. 1997, 38, 4183–4186; (i) Belanger, G.; Hong, F.;
Overman, L. E.; Rogers, B. N.; Tellew, J. E.; Trenkle, W.
that (3R,4Re,15Si)-13 (leading to 12a–f) is destabilized
relative to (3R,4Si,15Re)-13 (leading to 5a–f) because the
bulky substituent OR1 at C-3 is directed toward the
concave face of the bicyclo[4.3.0]nonane-like transition
state (Fig. 3).
1
C. J. Org. Chem. 2002, 67, 7880–7883.
16. (2S,3R,4S,15R)-5d: H NMR (500 MHz, CDCl3) d )0.05
10. Assignment of the double bond configuration based on the
chemical shift of the vinyl ether double bond proton, see:
Hiersemann, M. Synthesis 2000, 1279–1290, and Ref. 9f.
(s, 3H), 0.00 (s, 3H), 0.86 (s, 9H), 1.10 (d, J ¼ 6:9 Hz, 3H),
1.25 (dd, J1 ¼ J2 ¼ 6:3 Hz, 6H), 1.40 (dd, J1 ¼ 14:0 Hz,
J2 ¼ 8:0 Hz, 1H), 1.91–2.02 (m, 1H), 2.53 (dd, J1 ¼
14:0 Hz, J2 ¼ 10:2 Hz, 1H), 2.61 (dd, J1 ¼ J2 ¼ 9:5 Hz,
1H), 3.17 (br s, 1H), 3.74 (dd, J1 ¼ 9:5 Hz, J2 ¼ 8:2 Hz,
1H), 5.00–5.08 (m, 2H), 5.13 (dd, J1 ¼ 10:1 Hz,
J2 ¼ 2:2 Hz, 1H), 5.75 (dt, J1 ¼ 17:3 Hz, J2 ¼ 9:8 Hz,
1H); 13C NMR (126 MHz, CDCl3) d )4.1, )3.5, 18.0,
18.8, 21.7, 25.9, 40.0, 42.9, 61.6, 69.7, 80.1, 82.8, 119.5,
134.6, 175.6; IR (neat) 775, 835, 1100, 1250, 1720, 2860,
1
11. 6d: H NMR (300 MHz, CDCl3) d )0.02 (s, 3H), 0.00 (s,
3H), 0.86 (s, 9H), 0.88 (d, J ¼ 6:8 Hz, 3H), 1.36 (d,
J ¼ 6:2 Hz, 6H), 1.68 (dd, J1 ¼ 6:8 Hz, J2 ¼ 0:3 Hz, 3H),
2.15–2.29 (m, 1H), 2.48 (dd, J1 ¼ 17:0 Hz, J2 ¼ 8:0 Hz,
1H), 3.00 (dd, J1 ¼ 16:9 Hz, J2 ¼ 5:5 Hz, 1H), 3.93 (dd,
J1 ¼ J2 ¼ 6:0 Hz, 1H), 5.11 (sept, 1H), 5.33–5.43 (m, 1H),
5.48–5.62 (m, 1H); 13C NMR (75.5 MHz, CDCl3) d )4.9,
)4.2, 15.4, 17.6, 18.2, 21.6, 25.9, 36.0, 42.0, 70.4, 76.8,
127.2, 131.7, 161.1, 194.9; IR (neat) 1050, 1080, 1100,
1260, 1730, 2860, 2930, 2960, 2980 cmÀ1; Anal. calcd for
2930, 2960 cmÀ1; ½aꢀ25 +4.8 (c 1.14, CHCl3); Anal. calcd for
D
C18H34O4Si: C, 63.11; H, 10.00. Found: C, 62.95; H, 10.21.
17. The thermal intramolecular carbonyl ene reaction of 6c
has been realized on a 4 mmol scale. Up-scaling has no
detrimental effect on the yield or the diastereoselectivity.
18. Mitsunobu, O. Synthesis 1981, 1–28.
19. (2S,3S,4S,15R)-14: 1H NMR (500 MHz, CDCl3) d 1.02 (d,
J ¼ 6:9 Hz, 3H), 1.28 (dd, J1 ¼ 6:3 Hz, J2 ¼ 3:8 Hz, 6H),
1.85 (dd, J1 ¼ 13:7 Hz, J2 ¼ 10:9 Hz, 1H), 2.37–2.47 (m,
1H), 2.55 (dd, J1 ¼ 13:6 Hz, J2 ¼ 8:8 Hz, 1H), 2.98 (dd,
J1 ¼ 8:8 Hz, J2 ¼ 4:1 Hz, 1H), 3.23 (br s, 1H), 5.02–5.14
(m, 3H), 5.56 (dd, J1 ¼ J2 ¼ 3:9 Hz, 1H), 5.79 (ddd,
J1 ¼ 17:3 Hz, J2 ¼ 10:4 Hz, J3 ¼ 8:8 Hz, 1H), 7.58 (d,
J ¼ 8:8 Hz, 2H), 7.98 (d, J ¼ 8:5 Hz, 2H); 13C NMR
(126 MHz, CDCl3) d 14.2, 21.7, 38.2, 45.7, 57.7, 69.9, 80.9,
82.1, 119.4, 128.1, 129.0, 131.2, 131.4, 131.8, 165.4, 175.9;
IR (neat) 750, 1100, 1270, 1720, 2980 cmÀ1; Anal. calcd for
C18H34O4Si: C, 63.11; H, 10.00. Found: C, 63.39; H, 10.36;
25
½aꢀ )3.4 (c 1.11, CHCl3).
D
12. An increased reaction temperature or a prolonged reaction
time led to inferior chemical yields.
13. Attempts to catalyze the ene reaction of 6d with copper(II)
bis(oxazolines) according to the procedure of Yang et al.5d
did not afford the desired product 5d.
14. The configurational assignment is based on extensive
NOESY experiments.
15. The observation that the cyclopentane diastereomer 5c
was preferentially formed over 12c even in the initial phase
of the thermal ene reaction when the retro-ene reaction
is slow due to the low concentration of 5c and 12c
indicates in our opinion, that 5a–f are not only the
thermodynamically more stable products of the thermal
ene reaction but also kinetically favored. We suppose
C19H23BrO5: C, 55.49; H, 5.64. Found: C, 55.66; H, 5.82;
25
½aꢀ +98.2 (c 4.50, CHCl3).
D