An unusual proximity effect observed in a quinol bis-epoxide series

Article abstract of DOI:10.1016/S0040-4039(00)02177-8

The favorable intramolecular relationship between a benzyl ether group and a chiral quinol bis-epoxide moiety is shown to cause a peculiar type of rearrangement followed by a cascade of acid-catalyzed events that result in the formation of a new, highly strained polycyclic system.

Full text of DOI:10.1016/S0040-4039(00)02177-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 1011–1015
An unusual proximity effect observed in a quinol bis-epoxide
series
Guenter Neef,* Siegfried Baesler, Gisbert Depke and Harry Vierhufe
Preclinical Drug Research, Schering AG, D-13353 Berlin, Germany
Received 20 October 2000; accepted 23 November 2000
Abstract—The favorable intramolecular relationship between a benzyl ether group and a chiral quinol bis-epoxide moiety is shown
to cause a peculiar type of rearrangement followed by a cascade of acid-catalyzed events that result in the formation of a new,
highly strained polycyclic system. © 2001 Elsevier Science Ltd. All rights reserved.
Several compounds of pharmacological interest such as
aranorosin1 and diepoxin s2 contain the quinol epoxide
moiety as a core element and it appears fairly probable
that LL-C10037a,3 manumycin A4 and some tyrosine
derived metabolites of the agelorin5 and fistularin6
series are formed from quinol bis-epoxide precursors
(Fig. 1).
The regio- and stereoselective manipulation of quinol
bis-epoxides represents a major challenge to synthetic
organic chemists, although a considerable number of
terrestrial and marine natural products demonstrate
that nature has developed an admirable virtuosity in
dealing selectively with the five adjacent electrophilic
centers characteristic of the bis-epoxyquinol element.
Figure 1. A, Aranorosin. B, Diepoxin s. C, LL-C10037a. D, Manumycin A. E, Agelorin A.
Keywords: neighboring group effect; quinol derivatives; microwave-assisted thermolysis.
* Corresponding author.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)02177-8

1012
G. Neef et al. / Tetrahedron Letters 42 (2001) 1011–1015
In order to prepare key precursor 2 we had to modify
our recently disclosed protocol7 by replacing bro-
mobenzene with tetrahydropyranyl-protected 4-bro-
mophenol. As a further slight variation, the secondary
alcohol function was protected as a p-methoxycar-
bonylbenzyl ether instead of the unsubstituted benzyl
ether group. It must be pointed out that only this
combination of protecting groups ensured an unevent-
ful course for the above sequence of reactions. The
BF3–etherate catalyzed rearrangement8 of epoxide 2
and subsequent THP–ether cleavage resulted in the
formation of ring-contracted product 3 that was con-
verted to an E/Z mixture of oximes. The crude oxime
was treated with iodobenzene bis-trifluoroacetate
(PIFA)9 in aqueous acetonitrile to yield the cylohexadi-
enyl isoxazoline 4. Model considerations suggested a
hemiglobal shape for the dispiro compound 4. Hence, it
was no surprise to see epoxidation of cyclohexadienone
4 proceed with formation of a single isomer, the abso-
lute configuration of which was determined later by
X-ray crystal structure analysis.10
Figure 2. Crystal structure of 5.
Recently,7 we described a rearrangement of carvone-
derived epoxides that resulted in the stereoselective
formation of highly substituted cyclopentane deriva-
tives of type 3. The newly formed quaternary center
bearing a phenyl residue and an acetyl group offered
itself to be transformed into a spirocyclohexadienyl
isoxazoline intermediate 4, which after appropriate
epoxidation would give rise to quinol bis-epoxide 5
(Fig. 2). The secondary alcohol function originating
from the carvone part of the molecule was then
expected to act as an internal nucleophile that, for
steric reasons, would discriminate between the epoxide
moieties and select the b-position of the epoxyketone
for nucleophilic attack (Scheme 1).
Not unexpectedly, all our initial attempts to cleave the
benzyl ether group of compound 5 in the presence of
the quinol bis-epoxide met with failure. Nonetheless, we
continued our search for appropriate reaction condi-
tions because the X-ray structure revealed a uniquely
favorable distance and trajectory for the benzylic ether
oxygen to attack one of the epoxide moieties in an SN2
fashion. A breakthrough was achieved with a solid
phase microwave-assisted thermolysis11 of compound 5
on silica gel. After some optimization work a 60% yield
of a nicely crystalline product was obtained to which
NMR spectroscopy assigned the structure depicted
below (Scheme 2). Again, crystal structure analysis was
Scheme 1. Reaction conditions: (a) p-tetrahydropyranyloxy-phenylmagnesium bromide, THF, 55%; (b) p-methoxycarbonylbenzyl
bromide, NaH, DMF, 92%; (c) BF3–etherate, MeCl2, −78°C; (d) aqueous HOAc (70%), ambient temp., 67%; (f) PIFA,
MeCN–water (4:1), 58%; (g) benzyltrimethylammonium hydroxide, 30% H2O2, MeOH, 87%.

G. Neef et al. / Tetrahedron Letters 42 (2001) 1011–1015
1013
Severe doubts arose, however, when, as a control exper-
iment, an isomer of compound 5 was subjected to the
same reaction conditions. After 5 min of microwave
irradiation isomer 7 had not undergone a defined trans-
formation (Scheme 4). Only starting material (45%)
could be reisolated with no sign of benzyl ether cleaved
products in the remainder.
Although the distinctly altered spatial relationship
between the benzylic ether oxygen and the epoxy
ketone moiety could account for the different chemical
reactivity of isomer 7, benzylic ether cleavage as the
initial step in the formation of polycycle 6 became
rather questionable.
Scheme 2. Reaction conditions: 5 (300 mg) on silica gel (5.0
g), 5 min irradiation, 60%.
As a consequence, we started a new series of experi-
ments with bis-epoxy ketone 5 aiming no longer at a
selective benzyl ether cleavage, but rather trying to
activate the epoxide function with Lewis acids in a
nucleophile-depleted environment.
Though unsatisfactory from a preparative point of
view, the treatment of compound 5 with BF3–etherate
in methylene chloride at −78°C resulted in the forma-
tion of an unusual product 8 (Scheme 5). The favorably
located epoxide moiety feeling the vicinity of the ben-
zylic oxygen had undergone
rearrangement.
a
new type of
The microwave-assisted thermolysis of 8 on silica gel
led to the formation of product 6, thus making it highly
probable that the cascade of product-forming steps in
the transformation of 5 and 6 does not start with a
benzylic ether cleavage but rather with the rearrange-
ment observed in the BF3-catalyzed process (5–8).
Fig. 3. Crystal structure of 6.
used to confirm the preliminary assignment for product
612 (Fig. 3).
These conclusions are further supported by the com-
pletely different behavior of isomer 7, which showed a
remarkable resistance to boron trifluoride treatment
remaining unchanged for several hours at ambient tem-
perature. With the use of trimethylsilyl triflate as a
Lewis acid, it was even possible to isomerize the isopro-
Although far from being predicted, the formation of
product 6 seemed to be easily explainable by assuming
an initial benzylic ether cleavage followed by a cascade
of mechanistically plausible events as described by the
hypothetical sequence of Scheme 3.
Scheme 3. Hypothetical steps in the formation of 6. (a) Benzyl ether cleavage. (b) SN2 epoxide opening. (c) Ene–diol
tautomerization. (d) Payne rearrangement. (e) 1,2-H shift.

1014
G. Neef et al. / Tetrahedron Letters 42 (2001) 1011–1015
In conclusion, we have been able to demonstrate that
a directed intramolecular attack at a quinol bis-epox-
ide moiety can affect all the electrophilic centers in a
way to make them chemically distinguishable and
amenable to further structural variation. In that
respect, our work can be seen as a complement to the
careful analysis reported by Wipf et al.13 who investi-
gated the intermolecular thiophilic ring-opening and
rearrangement reactions of epoxyketone natural prod-
ucts.
Scheme 4.
It is fairly obvious that the unusual rearrangement
product 6 bears the potential of being easily trans-
formed into a number of derivatives that contain the
spirocyclohexenone isoxazoline subunit which is a
pharmacophoric group of considerable recent inter-
est.14
penyl double bond of the molecule 7 without the
quinol bis-epoxide element being affected (Scheme
6).
Scheme 5.
Scheme 6.

G. Neef et al. / Tetrahedron Letters 42 (2001) 1011–1015
1015
References
13. Wipf, P.; Jeger, P.; Kim, Y. Bioorg. Med. Chem. Lett. 1998,
8, 351.
1. Fehlhaber, H. W.; Kogler, H.; Mukhopadhyay, T.;
Vijayakumar, E. K. S.; Ganguli, B. N. J. Am. Chem. Soc.
1988, 110, 8242.
2. Wipf, P.; Jung, J.-K. Angew. Chem., Int. Ed. Engl. 1997,
36, 764.
3. Wipf, P.; Kim, Y. J. Org. Chem. 1993, 58, 1649.
4. (a) Buzzetti, F.; Gaumann, E.; Hu¨etter, R.; Keller-Schier-
lein, W.; Neipp, L.; Prelog, V.; Zaehner, H. Pharm. Acta
Helv. 1963, 38, 871; (b) Wipf, P.; Coish, P. D. G.
Tetrahedron Lett. 1997, 38, 5073.
5. (a) Koenig, G. M.; Wright, A. D. Heterocycles 1993, 16,
1351; (b) Gao, H.; Kelly, M.; Hamann, M. T. Tetrahedron
1999, 55, 9717.
14. (a) Goldenstein, K.; Fendert, T.; Proksch, P.; Winterfeldt,
E. Tetrahedron 2000, 56, 4173; (b) Beil, W.; Jones, P. G.;
Nerenz, F.; Winterfeldt, E. Tetrahedron 1998, 54, 7273.
15. Selected physicochemical and spectroscopic data: 5, mp
191–193°C (ethyl acetate), [h]D +46.5 (CHCl3, c=0.505);
1H NMR (CDCl3, 400 MHz):16 l=1.55 ppm (dd, J=11
and 14 Hz, 1H, H-5); 1.68 (s, 3H, 4-C[ꢀC]-CH3); 1.78 (ddd,
J=6, 10, 15 Hz, 1H, H-3); 1.95 (s, 3H, 5%-CH3); 2.03–2.12
(m, 1H, H-3); 2.53 (dd, J=8 and 14 Hz, 1H, H-5); 2.09–2.11
(m, 1H, H-4); 3.44 (dd, J=3 and 5 Hz, 1H, H-5%%); 3.48 (dd,
J=3 and 5 Hz, 1H, H-7%%); 3.77 (t, J=5 Hz, 1H, H-1%%); 3.93
(s, 3H, COOCH3); 4.07 (t, J=5 Hz, 1H, H-3%%); 4.18 (dd,
J=6 and 7 Hz, 1H, H-2); 4.45–4.58 (AB-system, 2H,
O-CH2-Ph); 4.69 (s, 1H, C=CH[‘Z’]); 4.78 (s, 1H,
CꢀCH[‘E’]); 7.34 (d, J=8 Hz, 2H, aryl-H); 8.03 (d, J=
8 Hz, 2H, aryl-H). 6, mp 192–194°C (MeCN–H2O),
[h]D +205.5 (CHCl3, c=0.523); 1H NMR (CDCl3, 400
MHz):16 l=1.68 ppm (ddd, J=5, 12.5, 15 Hz, 1H, H-3);
1.77 (s, 3H, 2-C[ꢀC]-CH3); 1.84 (t, J=12.5 Hz, 1H, H-5);
2.00 (s, 3H, 11-CH3); 2.33 (d, J=13 Hz, 8-OH); 2.35–2.45
(m, 2H, H-3 and H-5); 2.45–2.57 (m, 1H, H-4); 2.84–2.90
(m, 2H, H-7); 4.12 (dt, J=13 and 7.5 Hz, 1H, H-8); 4.78
(s, 1H, C=CH); 4.83 (d, J=2 Hz, 1H, C=CH); 5.12 (d;
J=5 Hz, 1H, H-3a); 5.78 (s, 1H, 5-OH). 8, mp 176–179°C
(hexane–ethyl acetate), [h]D +109.0 (CHCl3, c=0.5); 1H
NMR (CDCl3, 400 MHz):16 l=1.48 ppm (ddd, J=6, 11,
14 Hz, 1H, H-3); 1.69 (t, J=14 Hz, 1H, H-1); 1.73 (s, 3H,
2-C[=C]-CH3); 1.98 (s, 3H, 11-CH3); 2.08 (ddd, J=2, 5,
14 Hz, 1H, H-3); 2.23 (ddd, J=2, 4, 14 Hz, 1H, H-1);
2.35–2.44 (m, 1H, H-2); 3.53 (dd, J=1 Hz and 4 Hz, 1H,
H-7); 3.65–3.68 (m, 1H, H-6); 3.80–3.82 (m, 1H, H-5); 3.92
(s, 3H, COOCH3); 4.07 (t, J=2 Hz, 1H, H-4a); 4.38 (d, J=6
Hz, 1H, H-3a); 4.57 (d, J=13 Hz, 1H, O-CH2-Ph); 4.73
(s, 1H, C=CH [Z]); 4.74 (d, J=13 Hz, 1H, O-CH2-Ph);
4.80 (s, 1H, C=CH [E]); 7.44 (d, J=8 Hz, 2H, aryl-H);
8.01 (d, J=8 Hz, 2H, aryl-H). 9, mp 172-173 (hexane–ethyl
acetate), [h]D −154.7 (CHCl3, c=0.506); 1H NMR (CDCl3,
400 MHz):16 l=1.64 ppm (s, 6H, [H3C]2C=C); 2.04 (s, 3H,
5%-CH3); 2.37–2.47 (m, 1H, H-3); 2.46 (d, J=16 Hz, 1H,
H-5); 2.52 (d, J=16 Hz, 1H, H-5); 2.85 (dd, J=17 Hz and
8 Hz, 1H, H-3); 2.38 (dd, J=3 Hz and 4 Hz, 1H, H-5¦);
3.49 (dd, J=3 Hz and 4 Hz, 1H, H-7¦); 3.60 (t, J=8 Hz,
1H, H-1¦); 3.86 (t, J=8 Hz, 1H, H-3¦); 3.92 (s, 3H,
COOCH3); 4.18 (t, J=8 Hz, 1H, H-2); 4.46 (d, J=12 Hz,
1H, O-CH2-Ph); 4.69 (d, J=12 Hz, 1H, O-CH2-Ph); 7.34
(d, J=8 Hz, 2H, aryl-H); 8.02 (d, J=8 Hz, 2H, aryl-H).
16. Atom numbering for compounds 5 and 6.
6. Mancini, I.; Guella, G.; Laboute, P.; Debitus, C.; Pietra,
F. J. Chem. Soc., Perkin Trans. 1 1993, 3121.
7. Neef, G.; Baesler, S.; Depke, G.; Vierhufe, H. Tetrahedron
Lett. 1999, 40, 7969.
8. Domagala, J. M.; Bach, R. D.; Wemple, J. J. Am. Chem.
Soc. 1976, 98, 1975.
9. Kac¸an, M.; Koyuncu, D.; McKillop, A. J. Chem. Soc.,
Perkin Trans. 1 1993, 1771.
10. Crystal data for 5:15 C25H27NO7, M=453.48, orthorhom-
bic, space group P2(1)2(1)2(1), crystal size 0.7×0.25×0.1
mm3, unit cell dimensions a=9.606(2), b=10.482(2), c=
3
,
,
22.943(3) A, h=90, i=90, k=90°, V=2310.1(7) A , Z=4,
D=1.304 Mg/m3, absorption coefficient 0.095 mm−1, T=
,
273(2) K, wavelength u=0.71073 A, 3352 reflections col-
lected, 3146 independent reflections [R(int)=0.0128], R
indices (all data) R1=0.0505, wR2=0.1210. The crystallo-
graphic data for 5 have been deposited at the CCDC, 12
Union Road, Cambridge CB2 1EZ, UK and copies can be
obtained on request, free of charge, by quoting the
publication citation and the deposition number CCDC
152704.
11. (a) Loupy, A.; Petit, A.; Hamelin, J.; Texier-Boullet, F.;
Jacquault, P.; Mathe´, D. Synthesis 1998, 1213; (b) The
equipment used was a Synthewave 402, ProLabo. After 5
min of irradiation (300 W) the internal temperature of the
probe had reached 162°C.
12. Crystal data for 6·0.5H2O:15 C16H19NO5, M=305.54,
orthorhombic, space group P2(1)2(1)2(1), crystal size 0.7×
0.5×0.2 mm3, unit cell dimensions a=6.732(1), b=
,
10.576(2), c=21.90(4) A, h=90, i=90, k=90°, V=
3
3
,
1559.3(5) A , Z=4, D=1.335 Mg/m , absorption coeffi-
cient 0.101 mm−1, T=173(2) K, wavelength u=0.71073 A,
,
2939 reflections collected, 2512 independent reflections
[R(int)=0.0204], R indices (all data) R1=0.0309, wR2=
0.0734. The crystallographic data for 6 have been deposited
at the CCDC under deposition number CCDC 152705.
.