6-(2-Adamantan-2-ylidene-hydroxybenzoxazole)-O-sulfamate: A potent non-steroidal irreversible inhibitor of human steroid sulfatase

Article abstract of DOI:10.1016/j.bmcl.2003.09.050

We report the synthesis and results from the in vitro evaluation of 6-(adamantan-2-ylidene-hydroxybenzoxazole)-O-sulfamate 1 as an irreversible inhibitor of human steroid sulfatase (STS). Highly straightforward, condensation of 2-methyl-6-hydroxybenzoxazole with 2-adamantanone, subsequent elimination of water and sulfamoylation provide the title compound in 45% overall yield from the inexpensive 2,4-dihydroxyacetophenone. 1 was found to be a potent irreversible inhibitor of purified human steroid sulfatase (STS) and specific for this enzyme relative to human arylsulfatases A and B. In cellular assays with human keratinocytes, sebocytes and fibroblasts, 1 blocked STS activity with IC₅₀ values in the range of 0.15-0.8 nM, and in MCF-7 breast cancer cells with IC₅₀=2.3 nM, while it did not bind to estrogen receptors α and β. Thus, 1 is a candidate for further investigation of its potential as a drug to be used in androgen- and estrogen-dependent diseases.

Full text of DOI:10.1016/j.bmcl.2003.09.050

Bioorganic & Medicinal Chemistry Letters 13 (2003) 4313–4316
6-(2-Adamantan-2-ylidene-hydroxybenzoxazole)-O-sulfamate:
A Potent Non-steroidal Irreversible Inhibitor of Human
Steroid Sulfatase
Erwin P. Schreiner,* Barbara Wolff, Anthony P. Winiski and Andreas Billich
Novartis Forschungsinstitut, Brunner Strasse 59, A-1235 Vienna, Austria
Received 17 July 2003; revised 19 September 2003; accepted 23 September 2003
Abstract—We report the synthesis and results from the in vitro evaluation of 6-(adamantan-2-ylidene-hydroxybenzoxazole)-O-sul-
famate 1 as an irreversible inhibitor of human steroid sulfatase (STS). Highly straightforward, condensation of 2-methyl-6-
hydroxybenzoxazole with 2-adamantanone, subsequent elimination of water and sulfamoylation provide the title compound in
45% overall yield from the inexpensive 2,4-dihydroxyacetophenone. 1 was found to be a potent irreversible inhibitor of purified
human steroid sulfatase (STS) and specific for this enzyme relative to human arylsulfatases A and B. In cellular assays with human
keratinocytes, sebocytes and fibroblasts, 1 blocked STS activity with IC50 values in the range of 0.15–0.8 nM, and in MCF-7 breast
cancer cells with IC50=2.3 nM, while it did not bind to estrogen receptors a and b. Thus, 1 is a candidate for further investigation
of its potential as a drug to be used in androgen- and estrogen-dependent diseases.
# 2003 Elsevier Ltd. All rights reserved.
Steroid sulfatase (E.C. 3.1.6.2., STS) catalyses the
hydrolysis of steroid-3-O-sulfates to yield the corre-
sponding steroids. The enzyme has emerged as an
attractive drug target, because it is crucial for the local
production of active estrogens and androgens from their
systemic circulating sulfated precursors, such as estrone
sulfate and dehydroepiandrosterone sulfate (DHEAS),
in diseased tissues.1 In particular, there is evidence that
the STS pathway is a major source of estrogens in breast
cancer tissue,1,2 suggesting inhibitors of STS as potential
therapeutic agents for the treatment of breast cancer.
The most potent STS inhibitors described so far are aryl
sulfamates which act as irreversible inhibitors. These
include steroidal compounds such as estrone sulfamate
(EMATE, Fig. 1), the first potent inhibitor of STS,6 and
non-steroidal inhibitors featuring partial replacements of
the steroid skeleton (see refs. 7 and 8 for reviews). While
EMATE as the prototype compound is highly estrogenic,9
a variety of compounds devoid of estrogenicity have been
describedwhichmaybebettersuited for therapeutic use.10
In the skin, DHEAS is cleaved to DHEA, which is fur-
ther converted into the androgens testosterone and
dihydrotestosterone.3 An important role of STS in the
pathogenesis of androgen-dependent skin diseases, such
as androgenetic alopecia and acne, has been proposed.4
Finally, STS may regulate the availability of DHEA as
a neurohormone in the brain: treatment of rats with
inhibitors of STS resulted in enhanced learning and
spatial memory.5
*Corresponding author. Tel.: +43-1-866-349004; fax: +43-1-866-
34354; e-mail: erwin.schreiner@pharma.novartis.com
Figure 1.
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.09.050

4314
E. P. Schreiner et al. /Bioorg. Med. Chem. Lett. 13 (2003) 4313–4316
We aimed at identifying non-estrogenic irreversible
inhibitors of STS as a new therapeutic approach for the
treatment of acne. In addition to our efforts to identify
inhibitors by replacement of the steroid skeleton with
6,6-bicyclic ring systems,11 we also investigated benzox-
azoles as a mimicry for the steroidal A- and B-rings, as
exemplified by 1 (Fig. 1).12
While this sequence provided 1 in an overall yield of
39%, however, the building blocks 2 and 4 are quite
expensive and used early in the synthesis. Furthermore,
three chromatographic purification steps are involved and
the potential need for a replacement of the reagents TPP/
2,20-dipyridyldisulfide and TPP/DEAD during large-scale
synthesis made this approach unattractive for up-scaling.
Therefore, we developed an improved synthetic route
towards 1, devoid of these disadvantages (Scheme 2).
We devised a general synthetic approach for substituted
6-hydroxy-benzoxazoles featuring a hydrophobic resi-
due in the 2-position, starting from the appropriate
2-aminophenol and the corresponding acids:
Starting from the inexpensive 2,4-dihydroxy-
acetophenone (8), treatment with an aqueous solution
of hydroxylammonium chloride/sodium acetate gave
the crystalline oxime 9 in 86% yield.17 Beckmann rear-
rangement of 9, induced by means of POCl3 in a mix-
ture of MeCN/dimethylacetamide (DMA), gave crude
6-hydroxy-2-methylbenzoxazole (10)18 (81%), which
was used in the next step without further purification. In
situprotection of the phenolic hydroxyl group as tri-
methylsilyl ether, subsequent deprotonation with
sodium bis(trimethylsilyl)amide and condensation with
2-adamantanone at ꢀ78 ꢁC produced a mixture (1:2.5)
of the desired olefin 6 and the primarily formed addition
product 11 in almost quantitative yield. To achieve
complete conversion of 11 into 6, the crude reaction
For the preparation of 1, 2-adamantanone (2) was
transformed into adamantylidene acetic acid (3) either
by addition of the sodium enolate of EtOAc and sub-
sequent acid-induced elimination of water (A)13 or by
Horner–Emmons olefination (B)14 in 72 and 46% yield,
respectively. Coupling of 3 with 2,4-dihydroxyaniline (4)
was achieved via the 2-thiopyridyl ester obtained from
treatment with triphenylphosphane (TPP) and 2,20-
dipyridyl-disulfide in a one-pot reaction producing the
anilide 5 in 91% yield. Cyclisation of 5 performed with
TPP/diethyl azodicarboxylate (DEAD) gave the corre-
sponding benzoxazole 6 in 67% yield. Finally, reaction
of 6 with amidochlorosulfonic acid15 (7) in the presence
of 2,6-di-tert-butyl-4-methylpyridine (DBMP) produced
the sulfamate 1 in 88% yield (Scheme 1).16
Scheme 1.
Scheme 2.

E. P. Schreiner et al. /Bioorg. Med. Chem. Lett. 13 (2003) 4313–4316
4315
mixture was refluxed with TFA in toluene for 7 h. At
this stage purification was achieved by crystallisation of
the trifluoroacetic acid salt of 6 from ethanol to give
pure 7 (75%).19 Finally, heating of 6 with 7 in DMA at
50 ꢁC for 1 h and crystallization after aqueous workup
cleanly provided 1 in 87% yield. Within this sequence
the most expensive synthone 2 is introduced late in the
synthesis and no chromatographic purification is
needed. Overall yield amounted to 45%.
In conclusion, we identified 1 as the first potent non-
estrogenic irreversible inhibitor of human steroid sulfa-
tase featuring a 5,6-bicyclic ring system as a mimicry for
the steroidal A- and B-ring. 1 is a candidate for further
evaluation of its potential as a drug in the treatment of
estrogen- and androgen-dependent diseases.
References and Notes
The target compound 1 inactivated purified human STS
irreversibly. From the kinetics of inactivation, followed
by measuring residual enzyme activity after incubation
with various concentrations of the inhibitor,11,20 an
inhibition constant (Ki) of 6 mM and a rate constant of
inactivation (kinact) of 0.097 msꢀ1 was obtained. These
values may be compared with Ki=0.47 mM and
kinact=0.009 sꢀ1 reported for EMATE as a reference
compound.20 While 1 shows weaker binding to STS
than EMATE, the efficiency of inactivation, as expres-
1. (a) Pasqualini, J. R.; Gelly, C.; Nguyen, B. L.; Vella, C. J.
Steroid Biochem. 1989, 34, 155. (b) Reed, M. J.; Purohit, A.
Rev. Endocrine Related Cancer 1993, 45, 51.
2. Utsumi, T.; Yoshimura, N.; Takeuchi, S.; Maruta, M.;
Maeda, K.; Harada, N. J. Steroid Biochem. Mol. Biol. 2000,
73, 141.
3. Voigt, W.; Sawaya, M. E.; Hsia, S. L.; Amthor, C. IRSC
Med. Sci. 1982, 10, 529.
4. (a) Hoffmann, R.; Rot, A.; Niiyama, S.; Billich, A. J.
Invest. Dermatol. 2001, 117, 1342. (b) Billich, A.; Rot, A.;
Lam, C.; Schmidt, J. B.; Schuster, I. Horm. Res. 2000, 532, 92.
5. (a) Jagannathan, S.; Johnson, D. A. Cognitive Brain Res.
1995, 2, 251. (b) Rhodes, M. E.; Li, P. K.; Burke, A. M.;
Johnson, D. A. Brain Res. 1997, 773, 28. (c) Li, P. K.; Rhodes,
M. E.; Johnson, D. A.; Wu, T.; Li, P.; Maher, T. J. Brain Res.
2000, 865, 286.
6. Howarth, N. M.; Purohit, A.; Reed, M. J.; Potter, B. V. L.
J. Med. Chem. 1994, 37, 219.
7. Poirier, D.; Ciobanu, L. C.; Maltais, R. Exp. Opin. Ther.
Pat. 1999, 9, 1083.
sed by the ratio of kinact/Kꢀi 1, is about equal (1.6 vs
4
1.9 10 sꢀ1 Mꢀ1) for the two compounds.
.
To assess the selectivity of 1 for STS over human aryl-
sulfatases A (ASA) and B (ASB), both enzymes were
isolated from human placenta,21 and assayed in pre-
sence of 1.22 Whereas ASA was not inhibited by 1 u p to
the highest test concentration of 10 mM, mASB was
weakly inhibited (by 25% at 10 mM) in a time-indepen-
dent fashion, suggesting reversible binding to the active
site of this enzyme.
8. Nussbaumer, P.; Billich, A. Exp. Opin. Ther. Pat. 2003, 13,
605.
9. Elger, W.; Schwarz, S.; Hedden, A.; Reddersen, G.;
Schneider, B. J. Steroid Biochem. Mol. Biol. 1995, 55, 395.
10. (a) Ciobanu, L. C.; Boivin, R. P.; Luu-The, V.; Poirier, D.
Eur. J. Med. Chem. 2001, 36, 659. (b) Purohit, A.; Vernon,
K. A.; Hummelinck, A. E. et al. J. Steroid Biochem. Mol. Biol.
1998, 64, 269. (c) Kolli, A.; Chu, H. H.; Rhodes, M. E.; Inoue,
K.; Selcer, K. W.; Li, P. K. J. Steroid Biochem. Molec. Biol.
1999, 68, 31. (d) Malini, B.; Purohit, A.; Ganeshapillai, D.;
Woo, L. W.; Potter, B. V.; Reed, M. J. J. Steroid Biochem.
Mol. Biol. 2000, 75, 253.
11. Nussbaumer, P.; Lehr, P.; Billich, A. J. Med. Chem. 2002,
73, 4310.
12. Billich, A.; Schreiner, E. P.; Wolff-Winiski, B., WO
0136398 A1, 2001.
13. Masters, A. P.; Sorensen, T. S.; Tran, P. M. Can. J. Chem.
1987, 65, 1499.
Measurement of enzyme inhibition in cellular systems is
one step closer to the in vivo situation. First, we asses-
sed the ability of 1 to inhibit human STS in CHO cells
over-expressing STS:23 an IC50 of ꢂ37.5 nM was
determined (Table 1). Furthermore, we tested for inhi-
bition of STS-catalysed hydrolysis of DHEAS in differ-
ent cell types of the skin, namely keratinocytes,
fibroblasts, and sebocytes (Table 1).24 IC50 values in the
range of 0.15–0.8 nM were obtained. Finally, the IC50
value for inhibition of STS-dependent cleavage of
estrone sulfate by the human breast cancer cell line
MCF-7 was 2.3 nM.25
14. Narayanan, V. L. Ger. Offen. 2,043,380 (CA 75,20180k),
1972 (according to the abstract).
15. Appel, R.; Berger, G. Chem. Ber. 1958, 91, 1339.
These IC50 values roughly follow the amount of enzyme
activity in the various cell types (data not shown), which
is in line with the irreversible action of 1 and hence a
dependence on enzyme concentration.
1
16. For 1: H NMR (250 MHz, DMSO-d6) d 8.02 (br.s, 2H),
7.76 (d, J=8.6 Hz, 1H), 7.62 (d, J=2.2 Hz, 1H), 7.26 (dd,
J=8.6 and 2.2 Hz, 1H), 4.46 (s, 2H), 6.24 (s, 1H), 4.22 (s,
1H), 2.68 (s, 1H), 1.79–2.05 (m, 12H). 13C NMR (250 MHz,
DMSO-d6) d (ppm): 2167.76, 163.44, 148.82, 147.28, 140.02,
119.44, 105.43, 103.60, 40.97, 40.03, 39.01, 36.62, 33.50,
27.45, IR (neat) nmax 1192, 1388 cmꢀ1, MS (APCI) m/z 361.1
[MH]+.
The title compound 1 did not show any measurable
affinity to the human estrogen receptors a and b
(EC50>100 mM) whereas for EMATE EC50 values of
2.6 and 3.7 mM were determined.26
17. Lindemann, H.; Koenitzer, H.; Romanoff, S. Liebigs Ann.
Chem. 1927, 456, 284.
Table 1. Inhibition of human STS in different cell types
18. Fujita, S.; Koyama, K.; Inagaki, Y. Synthesis 1982, 68.
19. Procedure: 60 g (0.4 mol) of 6-hydroxy-2-methyl-ben-
zoxazole (10) were dissolved in 1 L of dry THF, cooled to
0 ꢁC, and 75 mL (0.44 mol) of DIEA and 52 mL (0.42
mol) trimethylsilyl chloride were added. Then the ice bath
was removed and stirring continued for 2 h. For the
deprotonation the reaction mixture was cooled to ꢀ78 ꢁC
Cell type
IC50 (nM)
STS-expressing CHO cells
Keratinocytes
Fibroblasts
Sebocytes
MCF-7 breast cancer cells
31.8ꢃ3.6
0.77ꢃ0.32
0.75ꢃ0.15
0.15ꢃ0.01
2.3ꢃ0.8

4316
E. P. Schreiner et al. /Bioorg. Med. Chem. Lett. 13 (2003) 4313–4316
and 680 mL of a 2 M NaHMDS solution in THF were
added within 2 h. After one additional hour solid 66 g (0.44
mol) 2-adamantanone was added and kept at ꢀ78 ꢁC for
another hour. Then, the reaction mixture was allowed to
warm up to rt, poured into 3 L of an 1M aqueous NaHSO4
solution and 5 L of ethyl acetate were added. The layers are
separated, and the organic phase was washed once with 1 L
of a 1M aqueous NaHSO4 solution and with 1 L of brine,
and was dried over sodium sulfate. The solvent is evapo-
rated and the residue obtained is re-dissolved in 2.4 L of
K.; Gasa, S.; Makita, A. Biochem. Int. 1992, 26, 1025. (b)
Kumagai, A.; Tomioka, H. Methods Enzymol. 1982, 86, 17.
22. Roy, A. B. Methods Enzymol. 1987, 143, 207.
23. Wolff, B.; Billich, A.; Brunowsky, W.; Herzig, G.; Lindley,
I.; Nussbaumer, P.; Pursch, E.; Rabeck, C.; Winkler, G. Anal.
Biochem. 2003, 318, 276.
24. HaCaT keratinocytes were cultured according to: Bou-
kamp, P.; Petrussevska, R. T.; Breitkreutz, D.; Hornung, J.;
Markham, A.; Fusenig, N. E. J. Cell Biol. 1988, 106, 761.
Primary skin-derived human fibroblasts were cultured in
DMEM plus 10% FCS. Primary human sebocytes were iso-
lated and cultured according to Fujie, T.; Shikiji, T.; Uchida,
N.; Urano, Y.; Nagae, H.; Arase, S. Arch Dermatol Res. 1996,
288, 703. Cells were incubated for 24 h in serum-free media
with [3H]-labelled DHEAS (specific activity: 21 Ci/mmol, 300
nM, 0.75 mCi/mL) in the presence of graded concentrations of
inhibitor 1. The reaction product DHEA was quantified by
HPLC equipped with a radiodetector to determine the turn-
over of the substrate by STS.
25. STS activity in MCF-7 cells was measured using estrone
sulfate as substrate, as described in: Billich, A.; Nussbaumer,
P.; Lehr, P. J. Steroid Biochem. Mol. Biol. 2000, 73, 225.
26. Affinity to the receptors was measured using a commercial
assay kit obtained from PanVera (Madison, WI, USA): Par-
ker, G. J.; Law, T. L.; Lenoch, F. J.; Bolger, R. E. J. Biomole.
Screen. 2000, 5, 77.
ꢁ
toluene and 32 mL of TFA by heating at 115 C for 7 h. Fil-
tration of the precipitate formed in the course of cooling to rt
afforded 125 g of the trifluoroacetic acid salt of 6. A second-
crystallisation from ethanol gave 66 g (59%) 6-hydroxy-2-
adamantan-2-ylidenemethyl-benzoxazole (6). Concentration
of the mother liquor yielded a second crop of 18 g (16%). MP
262 ꢁC, 1H NMR (500 MHz, DMSO-d6) d 9.70 (br.s, 1H), 7.44
(d, J=8.5 Hz, 1H), 6.96 (d, J=2.2 Hz, 1H), 6.78 (dd, J=8.5
and 2.2 Hz, 1H), 6.11 (s, 1H), 4.18 (s, 1H), 2.60 (s, 1H), 1.75–
1.99 (m, 12H). 1H NMR (125 MHz, DMSO-d6) d 164.53,
160.61, 155.77, 149.99, 134.10, 119.17, 113.03, 103.081, 96.89,
40.56, 38.64, 36.41, 32.94, 27.45.
20. Purohit, A.; Williams, G. J.; Howarth, N. M.; Potter,
B. V. L.; Reed, M. J. Biochemistry 1995, 34, 11508.
21. (a) ASA and ASB were purified by a combination of
published methods, namely: Jin, T.; Kobayashi, T.; Honke,