Unusual Cyclization Products Derived from Photolysis of Breslow's Steroidal Benzophenone Esters

Full text of DOI:10.1021/jo990800w

J . Org. Chem. 1999, 64, 7651-7653
7651
lactones are formed by the mechanism shown in Scheme
1. Photolysis generates the diradical I, which abstracts
a hydrogen atom from the tert-butyl group, generating
the primary radical II stabilized by the â-silicon atom.7-9
Subsequent coupling of the benzhydryl radical and the
primary radical produces an equal mixture of the 16-
membered macrocyclic lactones 6a and 6b. We have not
determined the structures of the individual diastereoi-
somers. Although we can find no examples in the
literature of this novel cyclization to generate large ring
lactones, there is a report of a similar photocyclization
involving the primary radical from a tert-butyldimeth-
ylsilyl ether, namely, the photocyclization of the buteno-
lide silyl ether 7 to form the strained eight-membered
ring 8 which has been reported by Mann and co-
Un u su a l Cycliza tion P r od u cts Der ived
fr om P h otolysis of Br eslow ’s Ster oid a l
Ben zop h en on e Ester s
Michael E. J ung* and Ted W. J ohnson1
Department of Chemistry and Biochemistry, University of
California, Los Angeles, Los Angeles, California 90085-1569
Received May 17, 1999
In 1973, Breslow and co-workers developed a method
for “remote functionalization” of steroids,2 namely, func-
tionalization of C9, C14, and C17 by photolysis of
benzophenone esters R-linked to the steroid backbone at
C3. They observed selective abstraction of tertiary hy-
drogens by the diradical of the benzophenone carbonyl
and, by choosing tethers of appropriate length, were able
to achieve good regioselectivity of hydrogen atom ab-
straction, e.g., selectivity for abstraction of C9, C14, or
C17.
Recently, our efforts have been directed at expanding
the Breslow concept to natural product synthesis.3 In a
projected total synthesis of xestobergsterol A (1), a potent
workers.10 As far as we can tell, these are the only two
photocyclizations involving the primary radical formed
from the tert-butyldimethylsilyl group. Both lactones 6a ,b
exhibit slow rotation of the para-disubstituted benzene
ring, and therefore each has six resonances for the six
aromatic carbons in the 13C NMR (e.g., the symmetry due
to rotation is lost). In addition it is noteworthy that one
of the two lactones (stereochemistry undetermined)
underwent transannular silicon transfer on storage at
25 °C for 1 year to produce the five-membered ring 9.
The other lactone was stable on storage. We can offer no
good rationalization for this difference in behavior.
inhibitor of histamine release from rat mast cells,4 we
investigated the possibility of using Breslow’s remote
functionalization as a means of producing the required
14â 15-ketone present in this molecule. Before using the
substrate functionalized on the side chain, we first
carried out model studies using the benzophenone ester
4b derived from cholesterol, e.g., with an unfunctional-
ized side chain. The diol 2 was prepared by hydrobora-
tion-oxidation of the commercially available cholesteryl
acetate. A regioselective Mitsunobu reaction5 with the
known carboxylic acid 36 provided the ester 4a in good
yield. Inversion proceeded much faster at C3 than at C6
due to severe steric hindrance encountered in the inver-
sion of the 6R alcohol, e.g., the 1,3-diaxial interaction of
the carboxylate with the C-18 methyl group. Finally,
protection of the free alcohol at C6 as its tert-butyldim-
ethylsilyl ether provided the desired substrate 4b.
Photolysis of the benzophenone ester 4b gave the
desired olefin 5 in modest yield but also afforded two
diastereomeric lactones, 6a and 6b . Presumably, the
In our successful synthesis of 7-deoxyxestobergsterol
A3 and recently xestobergsterol A11 itself, we were able
(7) Although there is some evidence that a primary radical â to a
silicon atom is somewhat stabilized,8 there are also theoretical calcula-
tions that indicate that R-silyl stabilization of alkyl radicals is greater
than â-silyl stabilization.9 We have no good argument for why the
primary radical â to silicon (e.g., abstraction of a hydrogen atom from
the tert-butyl group) II is formed rather than the corresponding
primary radical R to the silicon atom (e.g., abstraction of a hydrogen
atom from the methyl group). One possible explanation is that the
transition state for abstraction of the hydrogen atom from the methyl
group R to silicon suffers from larger steric interactions than the
corresponding transition state for abstraction of the hydrogen atom â
to the silicon.
(1) Saul Winstein Fellow, UCLA, 1998-9; Dissertation Award
Fellow, UCLA, 1998-9.
(2) (a) Breslow, R. Chemtracts 1988, 1, 333. (b) Breslow, R.; Baldwin,
S.; Flechtner, T.; Kalicky, P.; Liu, S.; Washburn, W. J . Am. Chem. Soc.
1973, 95, 3251. (c) Breslow, R.; Wife, R. L. Tetrahedron Lett. 1976,
517.
(3) J ung, M. E.; J ohnson, T. W. J . Am. Chem. Soc. 1997, 119, 12412.
(4) (a) Shoji, N.; Umeyama, A.; Shin, K.; Takeda, K.; Arihara, S.;
Kobayashi, J .; Takei, M. J . Org. Chem. 1992, 57, 2996. (b) Kobayashi,
J .; Shinonaga, H.; Shigemori, H.; Umeyama, A.; Shoji, N.; Arihara, S.
J . Nat. Prod. 1995, 58, 312.
(5) Mitsunobu, O. Synthesis 1981, 1.
(6) Zderic, J . A.; Kubitschek, M. J .; Bonner, W. A. J . Org. Chem.
1961, 26, 1635.
(8) (a) Wilt, J . W.; Lusztyk, J .; Peeran, M.; Ingold, K. U. J . Am.
Chem. Soc. 1988, 110, 281. (b) J ackson, R. A.; Ingold, K. U.; Griller,
D.; Nazran, A. S. J . Am. Chem. Soc. 1985, 107, 208.
(9) Ibrahim, M. R.; J orgensen, W. L. J . Am. Chem. Soc. 1989, 111,
819.
(10) Brown, D.; Cardin, C. J .; Mann, J . J . Chem. Soc., Chem.
Commun. 1995, 825.
(11) J ung, M. E.; J ohnson, T. W. Submitted for publication.
10.1021/jo990800w CCC: $18.00 © 1999 American Chemical Society
Published on Web 09/10/1999

7652 J . Org. Chem., Vol. 64, No. 20, 1999
Notes
(CDCl3, 400 MHz): δ 7.90-7.30 (9H, m), 5.11 (1H, m), 3.68 (2H,
s), 3.29 (1H, ddd, J ) 10.6, 10.6, 4.3 Hz), 2.20-0.50 (30H, m),
0.88 (3H, d, J ) 6.5 Hz), 0.857 (3H, d, J ) 6.6 Hz), 0.855 (3H, d,
J ) 6.6 Hz), 0.74 (3H, s), 0.62 (3H, s). 13C NMR (CDCl3, 100
MHz): δ 196.02, 170.00, 139.28, 137.45, 136.09, 132.26, 130.29,
129.88, 129.26, 128.16, 70.17, 69.31, 56.06, 56.01, 53.65, 46.80,
42.40, 41.82, 41.58, 39.65, 39.39, 36.28, 36.02, 35.63, 34.04, 32.92,
28.05, 27.90, 27.00, 25.56, 24.01, 23.70, 22.74, 22.48, 20.55, 18.55,
12.37, 11.92. FTIR (thin film): 3497, 2944, 1732, 1659, 1607,
1468, 1447 cm-1. High-resolution MS (EI, m/z): 626.4323, calcd
for C42H58O4 626.4335.
Sch em e 1
3r-((4-Ben zoyl)p h en yla cetyloxy)-6r-[((1,1-d im eth yl)eth -
yl)d im eth ylsilyloxy]-5r-ch olesta n e (4b). To a stirring solu-
tion of the alcohol 4a (143 mg, 0.228 mmol) in dry DMF (5 mL)
at 25 °C was added imidazole (31 mg, 0.456 mmol) followed by
tert-butyldimethylsilyl chloride (38 mg, 0.251 mmol), and the
reaction was stirred for 4 h. Water was added and the mixture
extracted with ether. The combined organic extracts were dried
over magnesium sulfate, filtered, and concentrated in vacuo.
Flash column chromatography of the residue (silica gel, 8:1
hexanes/ether) provided the ether 4b (155 mg, 92%) as a clear
1
glass. H NMR (CDCl3, 400 MHz): δ 7.90-7.70 (4H, m), 7.65-
7.30 (5H, m), 5.09 (1H, m), 3.69 (2H, s), 3.29 (1H, ddd, J ) 10.6,
10.6, 4.4 Hz), 1.90-0.60 (29H, m), 0.88 (3H, d, J ) 6.7 Hz), 0.856
(3H, d, J ) 6.6 Hz), 0.852 (3H, d, J ) 6.6 Hz), 0.82 (9H, s), 0.75
(3H, s), 0.63 (3H, s), -0.01 (3H, s), -0.02 (3H, s). 13C NMR
(CDCl3, 100 MHz): δ 195.75, 169.86, 139.43, 137.70, 136.04,
132.20, 130.45, 129.94, 129.32, 128.22, 70.49, 70.21, 56.14, 53.73,
46.83, 42.52, 42.03, 41.86, 39.81, 39.55, 36.37, 36.19, 35.80, 34.13,
33.25, 28.15, 28.03, 27.52, 25.93, 25.81, 24.13, 23.88, 22.91, 22.68,
22.63, 20.70, 18.71, 18.03, 12.54, 12.09, -3.85, -4.72. FTIR (thin
film): 2936, 1732, 1661, 1609, 1472, 1447 cm-1. High-resolution
MS (CI, m/z): 741.5276, calcd for C48H73O4Si (M + H)+ 741.5278.
La cton e 6a a n d La cton e 6b. Ester 4b (604 mg, 0.815 mmol)
was photolyzed (450 W mercury arc lamp, Pyrex filter) in
degassed benzene (815 mL) for 10 h at room temperature. The
solvent was evacuated in vacuo and the residue subjected to flash
column chromatography (silica gel, 4:1 hexanes/ethyl acetate)
to provide olefin 5 (181 mg, 30%), lactone 6a (120 mg, 20%), and
lactone 6b (120 mg, 20%).
to avoid these undesired cyclization products by protect-
ing the C6 alcohol as the acetate and observed no unusual
products from that substrate. However, the formation of
these lactones represents an unusual photocyclization
reaction involving the participation of a tert-butyldim-
ethylsilyl protecting group and therefore implies that this
normally extremely stable protecting group has unrec-
ognized photochemical reactivity. We have also shown
that regioselective Mitsunobu reactions can be carried
out on steroidal diols with excellent regioselectivity and
good yield. Further work in this area is in progress.11
1
Da ta for La cton e 6a . H NMR (CDCl3, 400 MHz): δ 7.80-
7.00 (9H, m), 5.10 (1H, s), 4.99 (1H, m), 3.63 (1H, d, J ) 13.1
Hz), 3.49 (1H, d, J ) 13.1 Hz), 3.28 (1H, ddd, J ) 10.6, 10.6, 4.4
Hz), 2.43 (1H, d, J ) 14.9 Hz), 2.20 (1H, d, J ) 14.9 Hz), 2.10-
0.60 (29H, m), 0.91 (3H, d, J ) 5.3 Hz), 0.886 (3H, d, J ) 6.2
Hz), 0.881 (3H, d, J ) 6.5 Hz), 0.878 (3H, s), 0.73 (3H, s), 0.70
(3H, s), 0.67 (3H, s), 0.27 (3H, s), -0.07 (3H, s). 13C NMR (CDCl3,
100 MHz): δ 170.74, 149.16, 145.08, 133.29, 131.28, 128.79,
127.99, 126.95, 126.28, 125.96, 125.31, 78.09, 71.82, 69.04, 57.17,
56.27, 55.98, 53.76, 46.55, 42.94, 42.55, 41.82, 39.76, 39.52, 36.66,
36.18, 35.90, 34.60, 33.30, 31.61, 28.77, 28.16, 28.10, 28.03, 27.72,
25.79, 24.15, 23.93, 22.93, 22.85, 22.68, 22.59, 20.70, 18.70, 14.15,
12.26, 12.08, -2.03, -4.29. FTIR (thin film): 3409, 2946, 1732,
1468 cm-1. High-resolution MS (EI, m/z): 741.5262, calcd for
C48H73O4Si (M+H)+ 741.5278.
Exp er im en ta l Section
Gen er a l Meth od s. All reactions were carried out under
argon with the exclusion of moisture. Reagents were purchased
from commercial sources and used without further purification
unless otherwise specified. The following solvents and reagents
were distilled from the indicated agent under argon: tetrahy-
drofuran (THF) and diethyl ether from sodium benzophenone
ketyl; dichloromethane, benzene, and toluene from calcium
hydride; and triethylamine from calcium hydride. Flash column
chromatography was carried out in the indicated solvent system
on 230-400 mesh silica gel. The proton nuclear magnetic
resonance (1H NMR) spectra were recorded at 400 MHz, and
the carbon nuclear magnetic resonance (13C NMR) spectra were
recorded at 100 MHz. Spectra were taken in the indicated
solvent at ambient temperature, and the chemical shifts are
reported in parts per million relative to the solvent used. The
infrared (IR) spectra were recorded on a Fourier transform IR
spectrometer.
3r-((4-Ben zoyl)p h en yla cetyloxy)-6r-h yd r oxy-5r-ch oles-
ta n e (4a ). To a stirring solution of the diol 2 (300 mg, 0.741
mmol) in THF (10 mL) at 25 °C were added triphenylphosphine
(389 mg, 1.482 mmol) and the carboxylic acid 3 (196 mg, 1.482
mmol), followed by the dropwise addition of diethylazodicar-
boxylate (233 µL, 1.482 mmol). The reaction was stirred for 0.5
h and then quenched with water. The mixture was extracted
with ether. The combined organic layers were dried over
magnesium sulfate, filtered, and concentrated in vacuo. Flash
column chromatography of the residue (silica gel, 3:1 hexanes/
ethyl acetate) provided the ester 4a (280 mg, 60%). 1H NMR
1
Da ta for La cton e 6b. H NMR (CDCl3, 400 MHz): δ 7.80-
7.00 (9H, m), 5.04 (1H, m), 3.64 (1H, d, J ) 12.7 Hz), 3.42 (1H,
d, J ) 12.7 Hz), 3.23 (1H, ddd, J ) 10.3, 10.3, 4.5 Hz), 2.95 (1H,
d, J ) 14.6 Hz), 2.20-0.70 (30H, m), 2.07 (1H, d, J ) 14.6 Hz),
1.08 (3H, s), 0.92 (3H, d, J ) 6.5 Hz), 0.874 (3H, d, J ) 6.6 Hz),
0.870 (3H, d, J ) 6.6 Hz), 0.75 (3H, s), 0.67 (3H, s), 0.44 (3H, s),
0.03 (3H, s), -0.19 (3H, s). 13C NMR (CDCl3, 100 MHz):
δ
170.73, 150.25, 145.53, 132.39, 129.87, 128.77, 128.30, 127.45,
126.96, 125.91, 125.63, 80.58, 69.97, 69.32, 56.31, 56.26, 53.95,
47.50, 46.32, 42.81, 42.56, 42.44, 39.90, 39.53, 36.47, 36.18, 35.86,
34.06, 33.50, 28.19, 28.03, 27.67, 27.40, 26.01, 24.18, 23.88, 22.83,
22.57, 21.22, 20.71, 20.65, 18.68, 12.43, 12.06, -2.87, -6.27.
FTIR (thin film): 3505, 2944, 1732, 1466 cm-1. High-resolution
MS (EI, m/z): 739.5114, calcd for C48H71O4Si (M - H)+ 739.5122.
3r-(4-(2,2,3,3-Tetr a m eth yl-5-p h en yl-1,2-oxa sila la n e-5-yl)-
p h en yla cetyloxy)-5r-ch olesta n -6r-ol (9). Lactone 6a (20 mg,
0.027 mmol) was stored (neat) in a sample vial for approximately
1 year. Ester 9 was formed cleanly as a clear glass (20 mg, 100%).
1H NMR (CDCl3, 400 MHz): δ 7.60-7.35 (4H, m), 7.35-7.05
(5H, m), 5.09 (1H, m), 3.54 (2H, s), 3.30 (1H, ddd, J ) 10.7, 10.7,

Notes
J . Org. Chem., Vol. 64, No. 20, 1999 7653
4.5 Hz), 2.54 (1H, d, J ) 13.6 Hz), 2.49 (1H, d, J ) 13.6 Hz),
1.85-0.50 (30H, m), 0.92 (3H, d, J ) 6.5 Hz), 0.877 (3H, d, J )
6.6 Hz), 0.876 (3H, s), 0.872 (3H, d, J ) 6.6 Hz), 0.825 (3H),
0.763 (3H, s), 0.654 (3H, s), 0.22 (3H, s), 0.21 (3H, s). 13C NMR
(CDCl3, 100 MHz): δ 171.02, 148.82, 147.94, 132.34, 128.85,
127.90, 126.23, 125.48, 125.35, 84.51, 69.69 (2 carbons), 56.25
(2 carbons), 55.41, 53.73, 47.17, 42.58, 41.88, 41.48, 39.85, 39.53,
36.50, 36.17, 35.78, 34.24, 33.01, 28.20, 28.04, 27.18, 25.65, 25.46,
25.27, 24.19, 23.82, 23.04, 22.84, 22.59, 20.68, 18.70, 12.48, 12.06,
-2.03, -2.19. FTIR (thin film): 3443, 2938, 1732, 1466, 1447
Ack n ow led gm en t. We thank the National Insti-
tutes of Health and the Agricultural Research Division
of the American Cyanamid Co. for financial support.
Su p p or t in g In for m a t ion Ava ila b le: 1H and 13C NMR
spectra of all new compounds, e.g., compounds 4a , 4b, 6a , 6b,
and 9. This material is available free of charge via the Internet
at http://pubs.acs.org.
cm-1
.
J O990800W