Glucuronide and sulfate conjugates of ICI 182,780, a pure anti- estrogenic steroid. Order of addition, catalysis and substitution effects in glucuronidation

Article abstract of DOI:10.1016/S0040-4039(99)02067-5

The 3-sulfate 4 and 3- and 17-glucuronide conjugates 5 and 6 of the pure anti-estrogenic steroid ICI 182,780 1, which is expected to be an effective agent for the treatment of breast cancer, have been prepared. The synthesis of 6 could only be satisfactorily achieved using an inverse addition technique, not previously employed in the glucuronic acid series: the value of this technique for some other aglycones is discussed.

Full text of DOI:10.1016/S0040-4039(99)02067-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 389–392
Glucuronide and sulfate conjugates of ICI 182,780, a pure
anti-estrogenic steroid. Order of addition, catalysis and
substitution effects in glucuronidation
John R. Ferguson,a John R. Harding,b Keith W. Lumbard,a Feodor Scheinmann a and
Andrew V. Stachulski a,∗
aUltrafine UFC Ltd, Synergy House, Guildhall Close, Manchester Science Park, Manchester M15 6SY, UK
bAstraZeneca Pharmaceuticals, Safety of Medicines Department, Mereside, Alderley Park, Macclesfield,
Cheshire SK10 4TG, UK
Received 6 August 1999; revised 26 October 1999; accepted 2 November 1999
Abstract
The 3-sulfate 4 and 3- and 17-glucuronide conjugates 5 and 6 of the pure anti-estrogenic steroid ICI 182,780 1,
which is expected to be an effective agent for the treatment of breast cancer, have been prepared. The synthesis
of 6 could only be satisfactorily achieved using an inverse addition technique, not previously employed in the
glucuronic acid series: the value of this technique for some other aglycones is discussed. © 2000 Elsevier Science
Ltd. All rights reserved.
The 7α-substituted estradiol derivative, ICI 182,780 (7α-[9-(4,4,5,5,5-pentafluoropentylsulfinyl)-
nonyl]estra-1,3,5,(10)-triene-3,17β-diol) 1, is a pure receptor antagonist of estrogen1 and should be
an effective treatment for estrogen dependent breast cancer2 and non-malignant estrogen dependent
conditions.
As part of a continuing programme to define the metabolism and distribution of 1 in animals and
man, a number of putative metabolites, in particular the 3-sulfate, 3-glucuronide and 17-glucuronide,
∗
Corresponding author.
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(99)02067-5
tetl 15995

390
were required as metabolic standards. This paper both describes the efficient syntheses of the above
metabolites and presents a study of the scope of an inverse addition technique applied to glucuronidation.
Acid-catalysed transesterification of 1 (EtOAc, tosic acid)3 delivered the 17-acetate 2 which was a
suitable protected version for the synthesis of the 3-sulfate 4 (using Et3N·SO3)4 and the 3-glucuronide
5,6
5 (using imidate 7);
full details will be published later. Synthesis of the 17-glucuronide 6, however,
from the 3-benzoate 3 (1+PhCOCl, CH2Cl2, aq. KOH) proved much more difficult.
Reaction of the 3-benzoate 3 with 7 using 1.5 equiv. of BF3·Et2O (this unusually large quantity of
catalyst was required during the synthesis of 5 also: we assume the first equivalent binds strongly to the
sulfoxide) gave almost entirely the 17-acetate 10 with no detectable glucuronide. It had been reported7
that reaction of a 3-protected androstane-3,17-diol derivative with 7 gave only 8% of the desired 17-
glucuronide, though in that case a Koenigs–Knorr reaction with bromosugar 8 and Ag2CO3 gave a 34%
8
yield. We obtained no glucuronide on reaction of 3 with 8 and Ag2CO3; when CdCO3 was used as
catalyst, an orthoester was produced in 75% yield [δH, inter alia, 1.8 (3H, s) and 5.9 (1H, d)]. Attempted
acid-catalysed rearrangement of the orthoester led only to decomposition.
Even when the tri-isobutyryl imidate 9 9,10 was used, the transacylated compound 11 was still by far the
major product, with just 7% of the desired conjugate 12 being formed. At this point, however, we found
that the order of addition had a profound influence on the course of this reaction. When 9 was added
slowly to a solution of 3 and 1.5 equiv. BF3·Et2O in 1,2-dichloroethane at −15°C, rather than adding
BF3·Et2O to 3+9, the desired conjugate 12 was obtained in 50% yield after chromatography with 11 now

391
a minor by-product (15%). Schmidt has noted11 a similarly advantageous inverse addition when using a
much more reactive donor (sc. a fucosyl imidate). Finally, hydrolysis of 12 was performed using NaOH
in aq. i-PrOH and 6 was isolated as its sodium salt, followed by reverse-phase silica chromatography to
give material of >98% purity as assayed by analytical reverse-phase HPLC. Characterisation by 1H NMR
(300 MHz), negative-ion mode electrospray (ES) and positive-ion mode fast-atom bombardment (FAB)
MS confirmed the structure.12
The value of the inverse addition technique was further studied using other alcohols. In the case of
cyclohexanol 13, using either imidate 7 or 9 with 0.5 equiv. BF3·Et2O, the inverse addition gave a slightly
but consistently higher yield, although not nearly so dramatic as with 3 (Table 1). Using either normal or
inverse addition, 9 gave double the yield of 7. Another steroidal aglycone, 11α-hydroxyprogesterone 14,
gave a vastly better yield by the inverse technique. It appears likely that the effect is most pronounced
with aglycones bearing groups such as C_O and S_O which can strongly complex the Lewis acid: we
note this was the case in Schmidt’s example cited above.
Table 1
Glucuronidation of alcohols using the normal (method A) and inverse (method B) procedures (see
text). aUsing 1.5 equiv. imidate. bUsing 1 equiv. imidate
In the case of a primary alcohol, 2-phenylethanol 16, the inverse method offered no advantage but
again the tri-isobutyryl imidate 9 was greatly superior to the tri-acetyl imidate 7 by virtue of reduced
transacylation.10
When only 0.2 equiv. BF3·Et2O was used in the reaction of 13 with 9, considerable amounts of the
orthoester 15 resulted in addition to the glucuronide even when the inverse technique was used. Prolonged
reaction (40 h) with addition of further catalyst gave steady conversion of 15 to glucuronide in a virtually
identical yield (51%) to that obtained before. Using ZnCl2 (0.5 equiv.), 15 was the major product in the
early stages: after prolonged reaction the glucuronide was obtained in satisfactory (39%) yield. With the
stronger Lewis acid catalyst trimethylsilyl trifluoromethanesulfonate (TMSOTf, 0.25 equiv.), ‘inverse’
reaction of 13 with 9 was complete in <1 h at −10°C and the glucuronide was isolated in 52% yield with
no orthoester seen.
We conclude that order of addition, catalyst and acyl substitution are all important variables in the
trichloroacetimidate glucuronidation of alcohols and that the inverse addition technique may offer a
significant advantage, particularly for complex aglycones.

392
Acknowledgements
We are grateful to Dr. G. Robinson (Process Development Department, AstraZeneca Pharmaceuticals)
for valuable technical discussions on the chemistry of 1, to Mr. I. Cumpstey (St. John’s College, Oxford)
for skilled experimental assistance and to Mrs. V. Boote (University of Manchester) for high-resolution
MS in the FAB mode.
References
1. Wakeling, A. E.; Bowler, J. J. Steroid Biochem. Mol. Biol. 1992, 43, 173.
2. DeFriend, D. J.; Howell, A.; Nicholson, R. I.; Anderson, E.; Dowsett, M.; Mansell, R. E.; Blamey, A. W.; Bundred, N. J.;
Robertson, J. F.; Saunders, C.; Baum, W.; Walton, P.; Sutcliffe, F.; Wakeling, A. E. Cancer Res. 1994, 54, 408.
3. Tsuneda, K.; Yamada, J.; Yasuda, K.; Mori, H. Chem. Pharm. Bull. Jpn. 1963, 11, 510.
4. Dusza, J. P.; Joseph, J. P.; Bernstein, S. Steroids 1968, 12, 49.
5. Schmidt, R. R.; Kinzy, W. Adv. Carbohydr. Chem. Biochem. 1994, 50, 21.
6. Brown, R. T.; Scheinmann, F.; Stachulski, A. V. J. Chem. Res. (S) 1997, 370.
7. Rao, P. N.; Rodriguez, A. M.; Miller, D. W. J. Steroid Biochem. 1986, 25, 417.
8. Conrow, R. B.; Bernstein, S. J. Org. Chem. 1971, 36, 863.
9. Brown, R. T.; Mayalarp, S. P.; McGown, A. T.; Hadfield, J. A. J. Chem. Res. (S) 1993, 496.
10. Brown, R. T.; Carter, N. E.; Lumbard, K. W.; Scheinmann, F. Tetrahedron Lett. 1995, 36, 8661.
11. Schmidt, R. R.; Toepfer, A. Tetrahedron Lett. 1991, 32, 3353.
12. Selected spectroscopic data for 6: δH (300 MHz, CD3OD) 0.88 (3H, s, 18-CH3), 3.65 (1H, d, 50-H), 3.85 (1H, t, 17-H), 4.40
(1H, d, J 8 Hz, 10-H), 6.46 (1H, d, 4-H), 6.53 (1H, dd, 2-H) and 7.08 (1H, d, 1-H); m/z (ES −ve mode) 781 (M−Na, 100%).
1
Found (FAB +ve mode): m/z, 827.3207. C38H54F5Na2O9S requires: MNa+, 827.3191. Consistent 300 MHz H NMR and
high resolution FAB mass spectra were also obtained for 4 and 5.