A new approach to steroid ring construction based on a novel radical cascade sequence

Article abstract of DOI:10.1039/a707320h

A new approach to steroid ring construction based on sequential cascade 6-endo-trig cyclisation/macrocyclisation/transannulation reactions and exemplified in the synthesis of the cis, anti, cis, anti, cis tetracycle 17 from 15, is presented.

Full text of DOI:10.1039/a707320h

View Article Online / Journal Homepage / Table of Contents for this issue
A new approach to steroid ring construction based on a novel radical cascade
sequence
Sandeep Handa, Gerald Pattenden*† and Wan-Sheung Li
Department of Chemistry, Nottingham University, Nottingham, UK NG7 2RD
A new approach to steroid ring construction based on
sequential cascade 6-endo-trig cyclisation/macrocyclisation/
transannulation reactions and exemplified in the synthesis
of the cis, anti, cis, anti, cis tetracycle 17 from 15, is
presented.
selenyl ester group was chosen as the radical precursor because
of the known propensity for hex-5-enoyl radicals to undergo
6-endo cyclisations,3,7 and the conjugated enone group was
incorporated in the substrate 7 to facilitate the 9-endo-trig
cyclisation8 following ring-opening of the cyclopropylcarbinyl
radical intermediate (cf. 2 ? 3 ? 4, Scheme 1). The synthesis
of 7 was achieved in fifteen steps from the known bromo
alcohol 6 (Scheme 2).9,10 When a solution of the selenyl ester 7
in dry degassed benzene (4 mm), at reflux, was treated with
Bu3SnH (syringe pump addition over 4 h) in the presence of
AIBN, one major product was isolated in ca. 40% yield.
Analysis of the spectroscopic data for the compound showed
that, instead of generating the steroid dione structure 9 resulting
from the predicted cascade highlighted in Scheme 1, the acyl
radical intermediate 8 had undergone a single 14-endo-trig
cyclisation producing the macrocyclic 1,4-dione 10.11
The efficient 14-endo-trig macrocyclisation of the nucleoph-
ilic acyl radical 8, leading to 10, over the competing 6-endo-trig
cascade is no doubt a consequence of the electrophilicity of the
terminal enone in 8. In light of this we designed a new
cyclisation precursor to avoid the competing pathway. The
designed polyene selenyl ester 15 incorporates two carboxylate
ester moieties, the first of which [at C(5)] was introduced in
order to increase the rate of 6-endo-trig cyclisation of the acyl
radical precursor and the second [at C(13)] was incorporated to
lower the electrophilicity of the terminal olefin (with respect to
7). The new precursor 15 was prepared as shown in Scheme 3.
Thus formation of the Normant Grignard reagent from 3-chloro-
Since their structures were firmly established some fifty years
ago, there have been a wide range of imaginative approaches to
the total synthesis of natural steroids and to steroidal ring
systems. Perhaps paramount amongst the methods that have
been developed are those based on either Diels–Alder reactions
(inter- and intra-molecular, including transannular based)1 or
biomimetic-type electrophilic cyclisations of polyene precursor
compounds.2 In recent years our own research group has been
examining the scope for a range of cascade radical processes in
the synthesis of steroids, including those based on consecutive
6-endo-trig cyclisations from appropriate polyene acyl radical
precursors,3 and those based on radical macrocyclisations in
tandem with radical transannulations.4 Indeed, these approaches
have led to some highly efficient routes to steroids, including
aza-steroids5 and some natural products.6 We have now
conceived a new approach to steroid constructions, based on
these earlier studies, whereby the A/B ring system is elaborated
by consecutive 6-endo-trig cyclisations (1 ? 2), in concert with
formation of the C/D ring system via a macrocyclisation/
transannulation sequence (3 ? 4 ? 5), from an appropriately
functionalised cyclopropyl-based polyene precursor 1 (Scheme
1).
Thus we chose the cyclopropyl substituted trienone selenyl
ester 7 as the starting point for our cyclisation studies. The
O
consecutive
H
steps
6-endo-trig
Br
H
B
cyclisations
A
OH
PhSe
O
H
O
O
6
7
1
2
Bu3SnH
AlBn
O
H
O
H
H
H
H
H
H
9-endo-trig
cascade
H
H
macrocyclisation
H
O
O
4
3
H
transannular
cyclisation
O
O
9
8
H
C
H
D
H
H
14-endo-trig
H
H
O
O
O
O
O
5
10
Scheme 1
Scheme 2
Chem. Commun., 1998
311

View Article Online
OH
OR
Treatment of the selenyl ester 15 with Bu3SnH–AIBN under
conditions similar to those which produced 10 from 7 led to the
formation of one major product in ca. 45% yield, and some
minor products. Detailed analysis of the NMR data for the
separated major product indicated that the compound had the
hoped-for tetracyclic (steroid) ring system, but the data could
not define its stereochemistry unambiguously. Fortunately, the
tetracyclic ketone product could be reduced selectively using
NaBH4 leading to the crystalline carbinol 18, mp 126–128 °C
(Et2O–light petroleum). An X-ray crystallographic analysis of
this tetracycle showed, surprisingly, that it had the very unusual
cis, anti, cis, anti, cis-stereochemistry, viz. 17 and 18 (Fig. 1).
Thus we had not only demonstrated the scope for the new
strategy for steroid ring construction enunciated in Scheme 1,
but we have also uncovered a route to the unusual all cis-
stereochemistry for the steroid ring system produced when 15
undergoes a radical cascade cyclisation to 17. Further studies
are now in progress to develop this novel stratagem for
polycycle construction in other fused ring systems.
i
HO
Cl
11a R = H
b R = SiPh2But
ii
iii
I
CO2Me
iv, v
ButPh2SiO
ButPh2SiO
13
12
vi–viii
CO2Me
CO2Me
CO2Me
CO2Me
13
MeO2C
MeO2C
MeO2C
H
MeO2C
H
vi, vii,
ix, x
5
H
H
H
H
PhSe
ButPh2SiO
O
H
O
H
HO
14
15
17
18
Scheme 3 Reagents and conditions: i, MeMgBr, THF, 230 °C, Mg, 70 °C,
then dicyclopropyl ketone, 68%; ii, ButPh2SiCl, imidazole, 97%; iii, NIS,
PPh3, CH2Cl2 230 °C, 64%; iv, NaH, trimethyl phosphonoacetate, DMSO,
73%; v, NaH, (CH2O)3, 85%; vi, TBAF, 91–95%; vii, Dess–Martin
periodinane, 90–92%; viii, NaH, ButPh2SiO(CH2)4CH(CO2Me)PO(OMe)2
16, 0 °C, 87%; ix, KH2PO4, ButOH, H2O, NaClO2, 2-methylbut-2-ene,
95%; x, N-(phenylseleno)phthalimide, PBu3, 230 °C, 72%
We thank the EPSRC for support of this work via a
Postdoctoral Fellowships (to S. H. and W-S. L.) and purchase of
the X-ray diffractometer.
Notes and References
† E-mail: gp@nottingham.ac.uk
propanol followed by reaction with dicyclopropyl ketone first
gave the diol 11a which was then protected selectively as its
TBDPS ether 11b. Selective ring opening of one of the
cyclopropane rings in 11b was achieved by treatment with
N-iodosuccinimide (NIS)–PPh3 to produce the vinyl cyclopro-
pyl intermediate 12 as a 1:1 mixture of E- and Z-isomers.
Nucleophilic displacement of the homoallylic iodide in 12 by
the anion generated from trimethyl phosphonoacetate and
subsequent Wadsworth–Emmons reaction with paraformalde-
hyde next installed the terminal methacrylate residue in the
form of intermediate 13. The silyl ether group in 13 was
converted into the corresponding aldehyde which was then used
in a further Wadsworth–Emmons olefination with the phos-
phate 16 giving the triene 14 as a mixture of separable isomers
about the newly formed double bond. Finally, a series of
functional group interconversions then led to the target selenyl
ester 15.
‡ Crystal data for 18: C21H32O5, M = 364.47, triclinic, a = 7.499(6), b
= 10.405(6), c = 12.297(7) Å, a = 82.70(4), b = 80.93(6), g = 79.83(5)°,
U = 927.7(11) Å3, T = 150 K, space group P1 (No. 2), Z = 2, Dc = 1.305
g cm23, m(Mo-Ka) = 0.091 mm21, 5459 reflections measured, 3120
unique (Rint
= 0.019), final R1 [Fo ! 4s(Fo)] = 0.0628, wR2 (all
data) = 0.209 for 242 refined parameters. CCDC 182/715.
1 For specific examples of Diels–Alder approaches to steroidal systems,
see L. Quellet, P. Langois and P. Deslongchamps, Synlett, 1997, 689;
Y. Haruo, Y. Sakamoto and T. Takahashi, Synlett, 1995, 231;
H. Pellissier and M. Santelli, Tetrahedron, 1996, 52, 9093; T. Sugahara
and K. Ogasawara, Tetrahedron Lett., 1996, 37, 7403; T. J. Grinsteiner
and Y. Kishi, Tetrahedron Lett., 1994, 35, 8333 and references cited
therein.
2 See: E. J. Corey and S. Lin, J. Am. Chem. Soc., 1996, 118, 8765;
P. V. Fish, W. S. Johnson, G. S. Jones, F. S. Tham and R. K. Kullnig,
J. Org. Chem., 1994, 59, 6150.
3 A. Batsanov, L. Chen, G. B. Gill and G. Pattenden, J. Chem. Soc.,
Perkin Trans. 1, 1996, 45 and references cited therein. For a specific
example of a palladium-catalysed cascade approach to fused polycycles,
see T. Sugihara, C. Cope´ret, D. Owczarczyk, L. S. Harring and
E. Negishi, J. Am. Chem. Soc., 1994, 116, 7923.
4 G. Pattenden and P. Wiedenau, Tetrahedron Lett., 1997, 38, 3647 and
references cited therein.
5 P. J. Double and G. Pattenden, unpublished work.
6 G. Pattenden and L. Roberts, Tetrahedron Lett., 1996, 37, 4191.
7 C. Chatgilialoglu, C. Ferreri, M. Lucarini, A. Venturini and A. A.
Zavitsas, Chem. Eur. J., 1997, 3, 376; D. L. Boger and R. J. Mathvink,
J. Org. Chem., 1992, 57, 1429; D. Batty and D. Crich, J. Chem. Soc.,
Perkin Trans. 1, 1992, 3205.
O
O
8 For a review of free-radical mediated macrocyclisations, see S. Handa
and G. Pattenden, Contemp. Org. Synth., 1997, 4, 196.
9 S. Hatakeyama, H. Numata, K. Osanai and S. Takano, J. Chem. Soc.,
Chem. Commun., 1989, 1893.
O
O
10 Details of the synthetic route to 7 will be reported in a full paper.
11 For examples of acyl radical mediated 14-endo-trig cyclisation over the
competing 6-endo-trig pathway, see D. L. Boger and R. J. Mathvink,
J. Am. Chem. Soc., 1990, 112, 4008; M. P. Astley and G. Pattenden,
Synthesis, 1992, 101.
O
Fig. 1 X-Ray crystallographic structure of 18
Received in Liverpool, UK, 9th October 1997; 7/07320H
312
Chem. Commun., 1998