Synthesis of novel arylpyrazolo corticosteroids as potential ligands for imaging brain glucocorticoid receptors

Article abstract of DOI:10.1016/S0039-128X(02)00171-X

Corticosteroids regulate a variety of essential physiological functions, such as mineral balance and stress. The great interest in these steroids, especially the glucocorticoids, stems from roles they are thought to play in neuropsychiatric disorders, suc

Full text of DOI:10.1016/S0039-128X(02)00171-X

Steroids 68 (2003) 177–191
Synthesis of novel arylpyrazolo corticosteroids as potential ligands for
imaging brain glucocorticoid receptors
Frank Wüsta,b,∗, Kathryn E. Carlsonc, John A. Katzenellenbogenc
a
Institut für Bioanorganische und Radiopharmazeutische Chemie, FZ-Rossendorf e.V., Postfach 510119, Dresden 01314, Germany
b
Institut für Interdisziplinäre Isotopenforschung an der Universität Leipzig, Leipzig, Germany
c
Department of Chemistry, University of Illinois, Urbana, IL 61801, USA
Received 15 July 2002; received in revised form 8 November 2002; accepted 11 November 2002
Abstract
Corticosteroids regulate a variety of essential physiological functions, such as mineral balance and stress. The great interest in these
steroids, especially the glucocorticoids, stems from roles they are thought to play in neuropsychiatric disorders, such as severe depression
and anxiety.
Thedevelopmentofglucocorticoidreceptor(GR)ligandswhichareappropriatelylabeledwithshort-livedpositron-emittingradioisotopes
would allow the non-invasive in-vivo imaging and mapping of brain GRs by means of positron emission tomography (PET). In this context
we have synthesized a series of novel arylpyrazolo steroids exhibiting different substitution patterns at the D-ring of the steroid skeleton,
as ligands for brain GRs. Special attention was given to 4-fluorophenyl pyrazolo steroids, which are known to display high binding
affinity toward the GR. The compounds were evaluated in a competitive radiometric receptor binding assay to determine their relative
binding affinities (RBA) to the GR. Some compounds show good binding affinities of up to 56% in comparison to dexamethasone
(100%). In initial experiments, selected candidates were labeled with the positron emitter fluorine-18 and in one case with the ␥-emitter
iodine-131.
© 2002 Elsevier Science Inc. All rights reserved.
Keywords: Arylpyrazolo corticosteroids; Glucocorticoid receptor binding; Radiolabeling; Positron emission tomography (PET)
1. Introduction
GRs are also distributed in other parts of the brain, and
the highest concentrations are found in regions involved in
the feedback regulation of the hormonal stress response,
such as the paraventricular hypothalamus, hippocampus
and pituitary [4,5]. There is accumulating evidence that
corticosteroids, in addition to noradrenaline, corticotropin
releasing factor and serotonin, play an important role in
stress and the pathophysiology of anxiety disorders and
depression. It has been suggested that a disturbed HPA axis
regulation may be a primary causal factor in depression.
Since feedback inhibition of glucocorticoids is decreased
in many depressive patients, it has been hypothesized that
disturbances in the MR/GR balance may have deleterious
consequences for regulation of the stress response, and it
may increase vulnerability to depression.
Endogenous corticosteroids exert a wide range of phys-
iologic, biochemical, and behavioral effects in the central
nervous system where the hormonal response is orga-
nized [1–3]. Corticosteroid hormones are produced in the
adrenal gland, and their production is controlled via the
hypothalamo-pituitary-adrenocortical (HPA) axis, a classi-
cal closed-loop endocrine system. They are released into
the circulation particularly in high amounts after periods
of stress. Due to their lipophilic nature, corticosteroids not
only act on peripheral organs but also readily enter the
brain. In neurons, they bind to two different intracellular
receptors, the mineralocorticoid receptor (MR) and the glu-
cocorticoid receptor (GR). MRs and GRs are co-located
in brain regions that are involved in the regulation of fear
and anxiety, such as hippocampus, septum, and amygdala.
The development of GR ligands which are appropriately
labeled with short-lived positron-emitting radioisotopes
would allow the non-invasive in vivo imaging and mapping
of brain GRs by means of positron emission tomography
(PET). In this context, PET studies of brain GRs would
provide important information on the neurobiological
∗
Corresponding author. Tel.: +49-351-260-3694;
fax: +49-351-260-2915.
E-mail address: wuest@fz-rossendorf.de (F. Wüst).
0039-128X/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved.
doi:10.1016/S0039-128X(02)00171-X

178
F. Wüst et al. / Steroids 68 (2003) 177–191
Fig. 1. High-affinity arylpyrazolo glucocorticoids (a) K.E. Carlson and J.A. Katzenellenbogen, unpublished results; (b) Rat liver cytosol, [14].
basis of GR-mediated functional abnormalities of HPA axis
function in depression [6,7].
(0.040–0.063 mm). 1H-NMR, 13C-NMR and 19F-NMR
spectra were recorded on a Varian Inova-400 at 400, 100
and 376 MHz, respectively, with CDCl3 and DMSOd6. Low
resolution mass spectra were recorded on a Micromass
tandem quadrupole mass spectrometer system (1100 se-
ries Hewlett-Packard) and on Mariner mass spectrometer
system (Applied Biosystems). Analytical HPLC was con-
ducted using LaChrom systems from Merck–Hitachi. The
chemical purity was determined with a PHENOMENEX
LUNA C18 column (250 mm × 4.6 mm, 5 ␮m). The mo-
bile phase was MeOH/water (90/10) (A) and acetonitrile/
water (70/30) (B). The system was run with a flow rate of
1.0 ml/min.
Several attempts have been made to synthesize glu-
cocorticoids labeled with the short-lived positron emitter
fluorine-18 (t1/2 = 109.6 min) to image brain GRs [8–13].
Some of the compounds exhibited excellent in vitro GR
binding, in some cases even significantly higher than the
high-affinity glucocorticoid dexamethasone. However, none
of the investigated compounds were suitable for in vivo
visualization of brain GRs with PET. Rapid in vivo deflu-
orination and the inability to cross the blood–brain-barrier
are possible reasons for the lack of any accumulation of
these compounds in the brain.
A unique set of synthetic arylpyrazolo steroids, exempli-
fied by deacylcortivazol, nivazol and WIN 44577 (Fig. 1),
have been shown to be very high-affinity ligands for the GR
[14,15]. The strong GR binding affinity of compounds like
WIN 44577, as well as their stable aryl-bound fluorine and
their higher lipophilicity (for an anticipated enhanced brain
uptake), make this class of compounds attractive for the
development of novel fluorine-18 labeled ligands for PET
studies of brain GR.
Solvents and reagents were purchased from Sigma, Fluka,
or Aldrich. THF was distilled from sodium/benzophenone
ketyl prior to use. Other reagents were used as received.
Iodophenylhydrazine [17] and 4-hydroxy(tosyloxy)iodo-
toluene [18] were prepared according to literature
procedures.
2.2. Organic chemistry
Intrigued by the high GR binding affinity of several
arylpyrazolo steroids, we have synthesized a series of novel
pyrazolo steroids exhibiting different substitution patterns
at the D-ring of the steroid skeleton. The binding affin-
ity of these compounds for the GR was determined in
a competitive radiometric receptor binding assay, and in
initial experiments potential candidates were labeled with
the positron emitter fluorine-18 and in one case with the
␥-emitter, iodine-131.
2.2.1. 11β-Hydroxy-17α,20:20,21-bis-(methylenedioxy)-
pregn-4-ene-3-one (2)
To a suspension of hydrocortisone 1 (2.54 g, 7 mmol)
in CH2Cl2 (100 ml) was added formalin (40 ml, 40%) and
concentrated HCl (40 ml). The bi-layer system was vigor-
ous stirred at room temperature for 3 h. The organic layer
was separated, washed with saturated NaHCO3 solution
and flushed through a silica-gel plug. After evaporation
of the solvent under reduced pressure, 2.9 g (102%) of
2, contaminated with the 11␤-MOM-ether (20%), was
obtained as a wax. 1H-NMR (400 MHz, CDCl3): δ 1.11
(s, 3H; 18-CH3), 1.43 (s, 3H; 19-CH3), 3.98 (m, 2H;
21-CH2), 4.47 (m, 1H; 11␣-H), 4.98 (m, 4H; O–CH2–O
from bis(methylenedioxy)), 5.66 (s, 1H; 4-H), LRMS (ESI
positive) 405 [M + H].
2. Experimental
2.1. General
Melting points were determined on a BOËTIUS melt-
ing point apparatus and are uncorrected. Analytical
thin-layer chromatography (TLC) was performed on Merck
silica-gel F-254 plastic plates, with visualization un-
der UV (254 nm). Flash chromatography was performed
as described by Still et al. [16] using Merck silica-gel
2.2.2. 11β-Hydroxy-2-hydroxymethylene-17,20:20,21-
bis-(methylenedioxy)-pregn-4-ene-3-one (3)
Steroid 2 (2.8 g, contaminated with 11␤-MOM-ether) was
dissolved in dry toluene (15 ml) and methyl formate (2 ml).
Then, NaH (600 mg, 15 mmol, 60% dispersion in mineral

F. Wüst et al. / Steroids 68 (2003) 177–191
179
oil) was added. After stirring at room temperature for 4 h
1N HCl (100 ml) was added and the mixture was extracted
several times with methylene chloride. The methylene chlo-
ride was extracted several times with 1N NaOH. After acid-
ification of the NaOH extract with HCl and subsequent
re-extraction with methylene chloride the solvent was re-
0.23 mmol). After 1 h, acetone (1 ml) and water (3 ml)
was added followed by NaIO4 (244 mg, 1.14 mmol) and
the mixture was stirred at room temperature overnight.
Water (50 ml) was added and the mixture was extracted
with EtOAc. The EtOAc layer was flushed through a
silica-gel plug and the solvent was evaporated to give
177 mg (92%) of 6 as a solid. HPLC-analysis mobile phase
B: tR = 10.15 min, 97% chemical purity, m.p. 223–226 ◦C.
1H-NMR (400 MHz, CDCl3): δ 1.17 (s, 3H; 18-CH3),
1.33 (s, 3H; 19-CH3), 2.82 (AB quartet, ν = 123.8 Hz,
J = 15.2 Hz, 2H; 1␣/␤-H), 4.50 (m, 1H; 11␣-H), 6.07 (d,
1
moved to give 1.8 g (60%) of 3 as yellow foam. H-NMR
(400 MHz, CDCl3): δ 1.12 (s, 3H; 18-CH3), 1.33 (s, 3H;
19-CH3), 4.00 (m, 2H; 21-CH2), 4.40 (m, 1H; 11␣-H), 5.03
(m, 4H; O–CH2–O from bis(methylenedioxy)), 5.73 (d, J =
=
1.8 Hz, 1H, 4-H), 7.35 (s, 1H CHOH). LRMS (ESI posi-
1
tive) 433 [M + H].
J = 2.2 Hz, H, 4-H), 7.13–7.17 (m, 2H: Ar–H), 7.41 (s,
1H; H-pyrazole), 7.43–7.47 (m, 2H; Ar–H). LRMS (ESI
positive) 421.
2.2.3. 2ꢀ-(4-Fluorophenyl)-11β-hydroxy-2-
hydroxymethylene-17,20:20,21-bis-(methylenedioxy)-
pregn-4-eno[3,2-c]pyrazole (4)
2.2.6. 2ꢀ-(4-Fluorophenyl)-17α(1-propyn-1-yl)-11β,
Compound 3 (1.04 g, 2.4 mmol) in HOAc (20 ml) was
added to a solution of NaOAc (197 mg, 2.4 mmol) and
4-fluorophenylhydraziniumchloride (390 mg, 2.4 mmol) in
HOAc (10 ml) and water (5 ml). The reaction mixture was
stirred at room temperature overnight. 1N HCl (100 ml)
was added followed by extraction with methylene chloride.
The organic layer was washed with NaHCO3-solution fol-
lowed by water. Drying (Na2SO4) and evaporation of the
solvent gave 1.06 g (84%) of steroid 4 as a brown foam.
1H-NMR (400 MHz, CDCl3): δ 1.14 (s, 3H; 18-CH3),
1.31 (s, 3H; 19-CH3), 2.83 (AB quartet, ν = 128.7 Hz,
17β-dihydroxy-androst-4-eno[3,2-c]-pyrazole (7)
To a solution of ketone 6 (193 mg, 0.46 mmol) in dry THF
(10 ml) was added 1-propynylmagnesium bromide (9.14 ml,
4.57 mmol, 0.5 M in THF). The solution was stirred at
room temperature under an nitrogen atmosphere overnight.
Then, saturated NH4Cl-solution (100 ml) was added and the
mixture was thoroughly extracted with EtOAc. The solvent
was removed under reduced pressure and subsequent purifi-
cation by flash chromatography (50% EtOAc/hexane) af-
forded 212 mg (57%) of steroid 7 as a solid. HPLC-analysis
mobile phase A: tR = 5.56 min, 95% chemical purity,
m.p. 95–98 ◦C. 1H-NMR (400 MHz, CDCl3): δ 1.13 (s, 3H;
J
= 15.46 Hz, 2H; 1␣/␤-H), 4.00 (m, 2H; 21-CH2),
≡
4.47 (m, 1H; 11␣-H), 5.03 (m, 4H; O–CH2–O), 6.06 (d,
J = 2.2 Hz, 1H, 4-H), 7.12–7.17 (m, 2H: Ar–H), 7.41 (s,
1H; H-pyrazole), 7.43–7.47 (m, 2H; Ar–H). LRMS (ESI
positive) 524 [M + H].
2.2.4. 2ꢀ-(4-Fluorophenyl)-11β,17α,21-trihydroxy-
20-oxo-pregn-4-eno[3,2-c]pyrazole (5)
18-CH3), 1.32 (s, 3H; 19-CH3), 1.85 (s, 1H; C CCH3),
2.85 (AB quartet, ν = 115.5 Hz, J = 15.2 Hz, 2H;
1␣/␤-H), 4.49 (bs, 1H; 11␣-H), 6.05 (d, J = 2.2 Hz, 1H,
4-H), 7.13–7.17 (m, 2H: Ar–H), 7.41 (s, 1H; H-pyrazole),
7.43–7.47 (m, 2H; Ar–H). LRMS (ESI positive) 461.
2.2.7. 2ꢀ-(4-Fluorophenyl)-17α[(3-hydroxy)-1-
Steroid 4 (1.06 g, 2.02 mmol) was refluxed in 50% formic
acid (100 ml) for 45 min. The formic acid was evaporated
and the residue was re-dissolved in MeOH (75 ml) and 1N
NaOH (15 ml). After stirring for 1 h at room temperature
the solution was acidified with 1N HCl (150 ml). Extrac-
tion with methylene chloride, evaporation of the solvent
and subsequent purification by flash chromatography (80%
EtOAc–hexane) gave 780 mg (80%) of the desired product 5.
HPLC-analysis mobile phase B: tR = 5.61 min, 98% chemi-
cal purity, m.p. 187–189 ◦C. 1H-NMR (400 MHz, CDCl3): δ
0.96 (s, 3H; 18-CH3), 1.30 (s, 3H; 19-CH3), 2.82 (AB quar-
tet, ν = 123.8 Hz, J = 15.2 Hz, 2H; 1␣/␤-H), 4.48 (AB
quartet, ν = 137.1, J = 19.9 Hz, 2H; CH2–OH). 4.51
(m, 1H; 11␣-H), 6.05 (d, J = 2.2 Hz, 1H, 4-H), 7.13–7.17
(m, 2H: Ar–H), 7.40 (s, 1H; H-pyrazole), 7.43–7.46 (m, 2H;
Ar–H). LRMS (ESI positive) 481 [M + H].
propyn-1-yl]-11β,17β-dihydroxy-androst-4-eno
[3,2-c]-pyrazole (8)
Tetrahydro-2-(2-propynyloxy)-2H-pyran (1.05 g, 7.5
mmol) in THF (15 ml) was treated with n-BuLi (4.68 ml,
7.5 mmol, 1.6 M in hexane) at 0 ◦C. After 1 h at 0 ◦C, ke-
tone 6 (315 mg, 0.75 mmol) in THF (8 ml) was added and
the mixture was stirred at room temperature overnight.
1N HCl (1 ml) was added and stirring was continued for
10 min at 60 ◦C. After the addition of water (50 ml) the so-
lution was extracted with EtOAc. The EtOAc was washed
with NaHCO3-solution, dried (Na2SO4) and evaporated.
The residue was purified by flash chromatography (10%
MeOH–CH2Cl2) to afford steroid 8 (222 mg, 62%) as a
solid. HPLC-analysis mobile phase B: tR = 5.21 min, 98.5%
chemical purity, m.p. 147–150 ◦C. 1H-NMR (400 MHz,
CDCl3): δ 1.14 (s, 3H; 18-CH3), 1.32 (s, 3H; 19-CH3), 2.85
(AB quartet, ν = 115.3 Hz, J = 15.4 Hz, 2H; 1␣/␤-H),
4.32 (s, 2H; CH2–OH), 4.49 (m, 1H; 11␣-H), 6.05 (d,
J = 2.1 Hz, 1H, 4-H), 7.13–7.17 (m, 2H: Ar–H), 7.42 (s,
1H; H-pyrazole), 7.44–7.47 (m, 2H; Ar–H). LRMS (ESI
positive) 477.
2.2.5. 2ꢀ-(4-Fluorophenyl)-11β-hydroxy-androst-
4-ene-17-one[3,2-c]pyrazole (6)
To a solution of steroid 5 (220 mg, 0.457 ␮mol) in
EtOH–CH2Cl2 (10 ml, 1:1) was added NaBH4 (9 mg,

180
F. Wüst et al. / Steroids 68 (2003) 177–191
2.2.8. 2ꢀ-(4-Fluorophenyl)-17α[(3-
2H: Ar–H), 7.41 (s, 1H; H-pyrazole), 7.43–7.47 (m, 2H;
methansulfonyl)oxy-1-propyn-1-yl]-11β,
Ar–H). LRMS (ESI positive) 530 [M + H].
17β-dihydroxy-androst-4-eno[3,2-c]-pyrazole (9)
To a solution of alcohol 8 (47.8 mg, 0.1 mmol) in THF
(5 ml) and Et3N (30.4 ␮l, 0.22 mmol) was added MsCl
(15.7 ␮l, 0.2 mmol) at 0 ◦C. After stirring overnight the sol-
vent was partially removed and the residue was purified by
flash chromatography (80% EtOAc–hexane) to give 48 mg
(86.5%) of mesylate 9. 1H-NMR (400 MHz, CDCl3): δ 1.16
(s, 3H; 18-CH3), 1.32 (s, 3H; 19-CH3), 2.86 (AB quartet,
ν = 112.4 Hz, J = 15.3 Hz, 2H; 1␣/␤-H), 3.08 (s, 3H;
OSO2CH3), 4.51 (m, 1H; 11␣-H), 4.89 (s, 2H; CH2-OMs),
6.06 (d, J = 1.8 Hz, 1H, 4-H), 7.13–7.18 (m, 2H: Ar–H),
7.42 (s, 1H; H-pyrazole), 7.44–7.47 (m, 2H; Ar–H). LRMS
(ESI positive) 555.
2.2.12. 2ꢀ-(4-Fluorophenyl)-17α[(3-N-piperidino)-
1-propyn-1-yl]-11β,17β-dihydroxy-androst-4-eno
[3,2-c]-pyrazole (12)
Yield: 77%. HPLC-analysis mobile phase A: tR
=
6.24 min, 96.5% chemical purity, m.p. 200–203 ◦C.
1H-NMR (400 MHz, CDCl3): δ 1.15 (s, 3H; 18-CH3), 1.32
(s, 3H; 19-CH3), 1.61 (quint., J = 5.4 Hz, 2H; piperidine),
1.94 (m, 4H; piperidine), 2.49 (m, 4H; piperidine), 2.84
(AB quartet, ν = 129.8, J = 15.2 Hz, 2H; 1␣/␤-H),
≡
3.31 (s, 2H; C CCH2-piperidine), 4.50 (bs, 1H; 11␣-H),
6.06 (bs, 1H; 4-H), 7.13–7.17 (m, 2H: Ar–H), 7.48 (s,
1H; H-pyrazole), 7.43–7.47 (m, 2H; Ar–H). LRMS (ESI
positive) 544 [M + H].
2.2.9. 2ꢀ-(4-Fluorophenyl)-17α[(3-fluoro)-1-
propyn-1-yl]-11β,17β-dihydroxy-androst-4-eno
[3,2-c]-pyrazole (10)
2.2.13. 2ꢀ-(4-Fluorophenyl)-17α[(3-N-morpholino)-1-
propyn-1-yl]-11β,17β-dihydroxy-androst-4-eno
[3,2-c]-pyrazole (13)
To a solution of mesylate 9 in THF (1.5 ml) was added
TBAF (150 ␮l, 150 ␮mol, 1 M in THF) and the resulting
mixture was heated to 60 ◦C for 30 min. The solvent was
evaporated and the residue was purified by flash chromatog-
raphy (60% EtOAc–hexane) to give 7.5 mg (38%) of fluoride
10. HPLC-analysis mobile phase A: tR = 4.85 min, 95.5%
chemical purity, m.p. 125–127 ◦C. 1H-NMR (400 MHz,
CDCl3): δ 1.16 (s, 3H; 18-CH3), 1.32 (s, 3H; 19-CH3),
2.86 (AB quartet, ν = 113, J = 15.4 Hz, 2H; 1␣/␤-H),
4.50 (bs, 1H; 11␣-H), 5.00 (d, JHF = 45.9 Hz, 2H; CH2-F),
6.06 (bs, 1H, 4-H), 7.13–7.17 (m, 2H: Ar–H), 7.41 (s,
1H; H-pyrazole), 7.44–7.47 (m, 2H; Ar–H). 19F-NMR
(376 MHz, CDCl3): δ −214.24 (t, JHF = 47 Hz), −115.33
(m). LRMS (ESI positive) 479 [M + H].
Yield: 70%. HPLC-analysis mobile phase B: tR
=
1
6.29 min, 96.5% chemical purity, m.p. 96–99 ◦C. H-NMR
(400 MHz, CDCl3): δ 1.15 (s, 3H; 18-CH3), 1.32 (s, 3H;
19-CH3), 2.55 (m, 4H; morpholine), 2.82 (AB quartet,
ν = 126.4, J = 15.2 Hz, 2H; 1␣/␤-H), 3.34 (s, 2H;
≡
C CCH2-morpholine), 3.74 (m, 4H; morpholine), 4.50 (bs,
1H; 11␣-H), 6.05 (d, J = 1.9 Hz, 1H; 4-H), 7.13–7.17 (m,
2H: Ar–H), 7.40 (s, 1H; H-pyrazole), 7.43–7.47 (m, 2H;
Ar–H). LRMS (ESI positive) 546 [M + H].
2.2.14. 2ꢀ-(4-Fluorophenyl)-21-(methansulfonyl)oxy-20-
oxo-11β,17α-dihydroxy-pregn-4-eno[3,2-c]pyrazole (14)
A solution of steroid 5 (200 mg, 0.416 mmol) in dry
pyridine (4 ml) was cooled to 0 ◦C and MsCl (65.4 ␮l,
0.832 mmol) was slowly added. After stirring at 0 ◦C for 3 h,
the solution was poured into 1N HCl (100 ml) followed by
extraction with EtOAc. After drying (Na2SO4) the solvent
was evaporated and the residue was purified by flash chro-
matography (80% EtOAc/hexane) to yield 182 mg (78%)
of mesylate 14 as a pale yellow foam. 1H-NMR (400 MHz,
CDCl3): δ 0.97 (s, 3H; 18-CH3), 1.30 (s, 3H; 19-CH3),
2.82 (AB quartet, ν = 128, J = 15.3 Hz, 2H; 1␣/␤-H),
3.24 (s, 3H; OSO2CH3), 4.52 (m, 1H; 11␣-H), 5.16 (AB
quartet, ν = 108.6, J = 17.8 Hz, 2H; CH2-OMs), 6.05
(d, J = 1.7 Hz, 1H, 4-H), 7.13–7.18 (m, 2H: Ar–H), 7.41
(s, 1H; H-pyrazole), 7.43–7.46 (m, 2H; Ar–H). LRMS (ESI
positive) 559 [M + H].
2.2.10. General procedure for the preparation of
amine-containing steroids 11, 12 and 13
To a solution of Na2CO3 (50 mg, 0.47 mmol) in wa-
ter (250 ␮l) was added the amine (1.0 mmol) followed by
mesylate 9 (45 mg, 81 ␮mol) in acetone (1.5 ml). The so-
lution was stirred for 2 h at room temperature. Saturated
NaHCO3-solution (20 ml) was added and EtOAc was used
for extraction. After drying (Na2SO4) and evaporation of
the solvent the residue was purified by flash chromatog-
raphy (10% MeOH–CH2Cl2) to yield the corresponding
amine.
2.2.11. 2ꢀ-(4-Fluorophenyl)-17α[(3-N-pyrrolidino)-1-
propyn-1-yl]-11β,17β-dihydroxy-androst-4-eno
[3,2-c]-pyrazole (11)
2.2.15. Oxetan-3ꢀ-one (15)
Yield: 47%. HPLC-analysis mobile phase A: tR
=
Mesylate 14 (31 mg, 55.5 ␮mol) was dissolved in dry THF
(750 ␮l) and potassium tert-butoxide (9.34 mg, 83.2 ␮mol)
in THF (400 ␮l) was added at room temperature. A clear
yellow solution occurs. After 5 min, 1N HCl (100 ␮l) was
added to quench the reaction followed by water (10 ml).
Extraction with CHCl3, drying (Na2SO4), evaporation of
the solvent and subsequent flash chromatography (30%
5.85 min, 98.5% chemical purity, m.p. 196–198 ◦C.
1H-NMR (400 MHz, CDCl3): δ 1.15 (s, 3H; 18-CH3), 1.32
(s, 3H; 19-CH3), 1.82 (m, 4H; pyrrolidine), 2.67 (m, 4H;
pyrrolidine), 2.84 (AB quartet, ν = 127.9, J = 15.2 Hz,
≡
2H; 1␣/␤-H), 3.48 (s, 2H; C CCH2-pyrrolidine), 4.50 (bs,
1H; 11␣-H), 6.05 (d, J = 1.8 Hz, 1H; 4-H), 7.13–7.17 (m,

F. Wüst et al. / Steroids 68 (2003) 177–191
181
EtOAc/hexane) gave 4 mg (16%) of oxetanone 15 as a white
foam. HPLC-analysis mobile phase B: tR = 15.13 min, 96%
chemical purity, m.p. 217–220 ◦C. 1H-NMR (400 MHz,
CDCl3): δ 1.13 (s, 3H; 18-CH3), 1.31 (s, 3H; 19-CH3),
2.83 (AB quartet, ν = 129.3, J = 15.1 Hz, 2H; 1␣/␤-H),
4.55 (m, 1H; 11␣-H), 4.98 (AB quartet, ν = 51.4,
J = 14.5 Hz, 2H; oxetanone), 6.05 (d, J = 2.1 Hz, 1H,
4-H), 7.13–7.18 (m, 2H: Ar–H), 7.41 (s, 1H; H-pyrazole),
7.43–7.46 (m, 2H; Ar–H). LRMS (ESI positive) 463
[M + H].
tion was stirred at room temperature overnight followed
by the addition of water (100 ml), extraction with CHCl3
and flushing through a silica-gel plug. After evaporation
of the solvent 877 mg (100%) of ketone 18 was iso-
lated a white solid. 1H-NMR (400 MHz, CDCl3): δ 1.11
(s, 3H; 18-CH3), 1.42 (s, 3H; 19-CH3), 4.49 (bs, 1H;
11␣-H), 5.64 (s, 1H, 4-H). 13C-NMR (100 MHz, CDCl3):
δ 219.71, 199.82, 172.29, 122.26, 67.36, 56.39, 52.10,
46.57, 40.46, 39.08, 35.05, 34.57, 33.49, 31.58, 31.21,
30.67, 21.35, 20.68, 15.49. LRMS (ESI positive) 303
[M + H].
2.2.16. 2ꢀ-(4-Fluorophenyl)-21-iodo-20-oxo-
2.2.19. 3-(N-Pyrrolidinyl)-3,5-androstadien-11β-ol-
17-one (19)
11β,17α-dihydroxy-pregn-4-eno[3,2-c]pyrazole (16)
To a solution of mesylate 14 (129 mg, 86 ␮mol) in
acetone (1 ml) was added NaI (129 mg, 860 ␮mol) in ace-
tone (5 ml). After being stirred overnight while protected
from light the solution was poured into water (20 ml) and
0.1N Na2S2O3-solution (20 ml). Extraction with EtOAc,
drying (Na2SO4), evaporation of the solvent and subse-
quent flash chromatography (40% EtOAc/hexane) afforded
Ketone 18 (877 mg, 2.9 mmol) in toluene (70 ml) was
refluxed with pyrrolidine (2.5 ml) and TsOH·H2O (45 mg)
while removing the formed water. After 4 h the solvent was
evaporated and ether was added. Further evaporation of the
solvent gave 1 g (97%) of 19 as a yellowish-brown foam
1
which was efficiently pure. H-NMR (400 MHz, CDCl3):
1
δ 1.15 (s, 3H; 18-CH3), 1.25 (s, 3H; 19-CH3), 3.14 (m,
4H; CH2), 4.47 (m, 1H; 11␣-H), 4.74 (s, 1H; 4-H), 4.96
(m, 1H; 6-H). 13C-NMR (100 MHz, CDCl3): δ 220.15,
144.03, 142.82, 111.01, 95.32, 68.15, 53.63, 52.37, 46.98,
46.86, 40.52, 35.25, 34.92, 33.47, 30.47, 27.75, 24.79,
24.25, 21.94, 21.52, 15.60. LRMS (ESI positive) 356
[M + H].
43 mg (85%) of iodide 16 as a pale yellow foam. H-NMR
(400 MHz, CDCl3): δ 1.00 (s, 3H; 18-CH3), 1.31 (s, 3H;
19-CH3), 2.83 (AB quartet, ν = 121.8, J = 15.3 Hz, 2H;
1␣/␤-H), 4.10 (AB quartet, ν = 75.4, J = 11.3 Hz, 2H;
CH2-I). 4.52 (m, 1H; 11␣-H), 6.05 (d, J = 2.1 Hz, 1H,
4-H), 7.13–7.17 (m, 2H: Ar–H), 7.41 (s, 1H; H-pyrazole),
7.43–7.47 (m, 2H; Ar–H). LRMS (ESI positive) 591
[M + H].
2.2.20. 17α(1-Propyn-1-yl)-androst-4-ene-11β,
17β-diol-3-one (20)
2.2.17. 2ꢀ-(4-Fluorophenyl)-21-fluoro-20-oxo-11β,
17α-dihydroxy-pregn-4-eno[3,2-c]pyrazole (17)
To a solution of enamine 19 (194 mg, 0.54 mmol) in
dry THF (2 ml) was added 1-propynylmagnesium bro-
mide (5.7 ml, 2.86 mmol, 0.5 M in THF). The solution was
stirred at room temperature under an nitrogen atmosphere
overnight. Saturated NH4Cl-solution (150 ml) was added
and the mixture was thoroughly extracted with EtOAc.
The EtOAc was removed under reduced pressure and the
residue was redissolved in a solution of NaOAc (3 g) in
water (3 ml), HOAc (2 ml) and MeOH (20 ml). The re-
sulting mixture was refluxed for 1 h. Afterwards, water
(200 ml) was added followed by extraction with EtOAc.
Evaporation of the solvent and subsequent purification by
flash chromatography (75% EtOAc/hexane) gave 96 mg
(52%) of steroid 20 as a pale yellow foam. 1H-NMR
(400 MHz, CDCl3): δ 1.12 (s, 3H; 18-CH3), 1.45 (s, 3H;
19-CH3), 1.84 (s, 3H; CH3), 4.44 (m, 1H; 11␣-H), 5.68
(s, 1H, 4-H). 13C-NMR (100 MHz, CDCl3): δ 200.05,
172.84, 122.24, 82.46, 82.24, 80.05, 68.09, 55.88, 51.20,
45.83, 42.31, 39.15, 38.70, 34.72, 33.67, 32.03, 31.92,
31.82, 22.99, 20.77, 15.22, 3.53. LRMS (ESI positive) 343
[M + H].
To a solution of mesylate 14 (182 mg, 0.325 mmol) in
acetonitrile (10 ml) was added TBAF (450 ␮l, 450 mmol,
1 M in THF). After stirring for 15 min, at room temperature
the solution was heated to 60 ◦C and an additional 450 ␮l
of TBAF was added after 40 min. Addition of saturated
NaHCO3-solution, extraction (EtOAc), drying (Na2SO4),
evaporation of the solvent and subsequent preparative TLC
(60% EtOAc–hexane) gave 9 mg (5.8%) of fluoride 17
as a white solid. HPLC-analysis mobile phase B: tR
=
9.04 min, 93% chemical purity, m.p. 165–168 ◦C. 1H-NMR
(400 MHz, CDCl3): δ 0.99 (s, 3H; 18-CH3), 1.31 (s, 3H;
19-CH3), 2.82 (AB quartet, ν = 125, J = 13.9 Hz, 2H;
1␣/␤-H), 4.53 (m, 1H; 11␣-H), 5.20 (AB quartet, ν = 77,
JHH = 16.5 Hz, with additional doublet splitting due to
JHF = 47.5 Hz, 2H; CH2-F), 6.06 (s, 1H, 4-H), 7.13–7.17
(m, 2H: Ar–H), 7.40 (s, 1H; H-pyrazole), 7.43–7.46 (m,
2H; Ar–H). 19F-NMR (376 MHz, CDCl3): δ −231.13 (t,
JHF = 48 Hz), −115.28 (m). LRMS (ESI positive) 483
[M + H].
2.2.18. 11β-Hydroxy-androst-4-ene-3,17-dione (18)
A solution of cortisol 1 (1.06 g, 2.92 mmol) in EtOH–
CH2Cl2 (20 ml, 1:1) was treated with NaBH4 (44.2 mg,
1.17 mmol). After 2 h acetone (5 ml) was added followed
by water (5 ml) and NaIO4 (1.56 g, 7.3 mmol). The solu-
2.2.21. General procedure for the synthesis of
steroids 21 and 22
To a solution of steroid 20 (103 mg, 0.3 mmol) in toluene
(5 ml), pyridine (1.5 ml) and methyl formate (300 ␮l) was

182
F. Wüst et al. / Steroids 68 (2003) 177–191
added 60% NaH (72 mg, 1.8 mmol) followed by methanol
(150 ␮l). The solution was stirred overnight while the
color changed to dark red. Addition of 1N HCl (50 ml)
and extraction with methylene chloride gave a yellow
solution. The methylene chloride was extracted several
times with 1N NaOH. After acidification of the NaOH
extract with HCl and subsequent re-extraction with methy-
lene chloride the solvent was removed to give the cor-
responding 2-hydroxymethylene compound of 20 as a
yellow foam. The foam was dissolved in HOAc (10 ml)
8.2 Hz, 2H, Ar–H), 7.57 (AAꢀBBꢀ, JAB = 8.2 Hz, 2H,
Ar–H). LRMS (ESI positive) 607 [M + H].
2.2.25. General procedure for the synthesis of iodonium
salts 24 and 25
A solution of steroid 23 (115 mg, 0.19 mmol) and [hy-
droxy(tosyloxy)iodo]benzene or [hydroxy(tosyloxy)iodo]
toluene (0.19 mmol) [18] in methylene chloride (1 ml) was
stirred at room temperature for 2 h. The solvent was evapo-
rated by means of a nitrogen stream. The residue was treated
with ether resulting in the formation of a white solid. The
solid was filtered off, washed with ether and dried under
vacuo to yield the iodonium salt 24 and 25.
and
a solution of 4-bromophenylhydraziniumchloride
or 4-iodophenylhydrazine (0.3 mmol) [17] and NaOAc
(0.3 mmol) in HOAc (5 ml) and water (2 ml) was added.
The dark red colored solution was stirred at room tem-
perature overnight. Addition of water, extraction with
CHCl3 and subsequent purification by flash chromatog-
raphy (60% EtOAc/hexane) gave compounds 21 or 22,
respectively.
2.2.26. Diaryliodonium tosylate (24)
1
Yield: 28%. H-NMR (400 MHz, DMSOd6): δ 0.29 (s,
3H; CH3), 0.50 (s, 3H; CH3), 1.04 (s, 3H; CH3), 1.53 (s,
3H; CH3), 2.05 (AB quartet, ν = 150, J = 15.5 Hz, 2H;
1␣/␤-H), 3.60 (bs, 1H; 11␣-H), 5.35 (d, J = 1.8 Hz, 1H,
4-H), 6.42 and 6.89 (2d of AAꢀBBꢀ system, JAB = 8.3 Hz,
4H; Ar–H), 6.75 (m, 3H; Ar–H), 6.68 (s, 1H; H-pyrazole),
7.64 and 7.13 (2d of AAꢀBBꢀ system, JAB = 9 Hz, 4H,
Ar–H), 7.4–7.5 (m, 2H; Ar–H). LRMS (ESI positive) 645
[M-OTs].
2.2.22. 2ꢀ-(4-Bromophenyl)-17α(1-propyn-
1-yl)-11β,17β-dihydroxy-androst-4-eno[3,2-c]-
pyrazole (21)
Yield: 48%. HPLC-analysis mobile phase A: tR
=
8.12 min, 97% chemical purity, m.p. 130–132 ◦C. 1H-NMR
(400 MHz, CDCl3): δ 1.13 (s, 3H; 18-CH3), 1.32 (s,
≡
3H; 19-CH3), 1.85 (s, 3H; C C CCH3), 2.85 (AB quar-
2.2.27. Diaryliodonium tosylate (25)
Yield: 57%. H-NMR (400 MHz, DMSOd6): δ 0.29 (s,
1
tet, ν = 114.9, J = 15.3 Hz, 2H; 1␣/␤-H), 4.49 (bs,
1H; 11␣-H), 6.08 (d, J = 2.2 Hz, 1H, 4-H), 7.43 (s,
1H; H-pyrazole), 7.38 and 7.58 (2d of AAꢀBBꢀ sys-
tem, J = 8.8 Hz, 4H; Ar–H). LRMS (ESI positive) 522
[M + H].
3H; CH3), 0.50 (s, 3H; CH3), 1.00 (s, 3H; CH3), 1.55 (s, 3H;
CH3), 1.60 (s, 3H; CH3), 2.04 (AB quartet, ν = 150.4,
J = 15.5 Hz, 2H; 1␣/␤-H), 3.59 (bs, 1H; 11␣-H), 5.34 (d,
J = 1.8 Hz, 1H, 4-H), 6.41 and 6.56 (2d of AAꢀBBꢀ system,
JAB = 7.9 Hz, 4H, Ar–H), 6.67 (s, 1H; H-pyrazole), 6.82
and 7.45 (2d of AAꢀBBꢀ system, JAB = 9.0 Hz, 4H, Ar–H),
6.88 and 7.4 (2d of AAꢀBBꢀ system, JAB = 8.3 Hz, 4H,
Ar–H). LRMS (ESI positive) 659 [M-OTs].
2.2.23. 2ꢀ-(4-Iodophenyl)-17α(1-propyn-1-yl)-11β,
17β-dihydroxy-androst-4-eno[3,2-c]-pyrazole (22)
Yield: 50%. HPLC-analysis mobile phase A: tR
=
8.79 min, 96.5% chemical purity, m.p. 104–106 ◦C.
1H-NMR (400 MHz, CDCl3): δ 1.13 (s, 3H; 18-CH3),
2.3. Radiochemistry
≡
1.32 (s, 3H; 19-CH3), 1.85 (s, 3H; C C CCH3), 2.85
(AB quartet, ν = 115, J = 15.2 Hz, 2H; 1␣/␤-H), 4.49
(bs, 1H; 11␣-H), 6.08 (d, J = 1.8 Hz, 1H, 4-H), 7.43 (s,
1H; H-pyrazole), 7.25 and 7.78 (2d of AAꢀBBꢀ system,
J = 8.7 Hz, 4H; Ar–H). LRMS (ESI positive) 569 [M+H].
2.3.1. General
18F was produced by the 18O(p, n)18F reaction on an
enriched water target using an IBA CYCLONE 18/9 cy-
clotron. The [18O] water containing [18F]fluoride (ca. 5 mCi,
185 MBq) was transferred to a 5 ml Vacutainer® contain-
ing a solution of Kryptofix [2.2.2] (6.0 mg, 16.2 ␮mol)
and K2CO3 (1.5 mg, 10.9 ␮mol). Water was azeotropically
evaporated using HPLC grade acetonitrile (3× 0.5 ml)
in a 100 ◦C oil bath under a stream of nitrogen. Potas-
sium [18F]fluoride was resolubilized into 500–1000 ␮l of
anhydrous solvent (acetonitrile or DMF) and transferred
to a glass vial pre-charged with the labeling precursor
(2–10 mg). Progress of the reaction was monitored by
means of radio-TLC. A pre-purification was performed by
passing the diluted reaction mixture through a silica-gel
plug (a Pasteur pipet with ca. 150 mg silica-gel). Radioiod-
ination with iodine-131 was performed using a solution of
2.2.24. 2ꢀ-(4-Trimethylstannylphenyl)-
17α(1-propyn-1-yl)-11β,17β-dihydroxy-
androst-4-eno[3,2-c]-pyrazole (23)
A solution of steroid 22 (160 mg, 0.28 mmol) in toluene
(3 ml) was refluxed with Sn2Me6 (250 mg, 0.76 mmol) and
Pd[PPh3]4 (4.8 mg, 4.15 ␮mol) for 3 h. After removal of the
solvent the residue was purified by flash chromatography
(60% EtOAc/hexane) to afford 100 mg (59%) of steroid 23.
1H-NMR (400 MHz, CDCl3): δ 0.32 (s, 9H; Sn(CH3)3),
1.13 (s, 3H; 18-CH3), 1.32 (s, 3H; 19-CH3), 1.85 (s, 3H; C
≡
C CCH3), 2.86 (AB quartet, ν = 108.8, J = 15.2 Hz,
2H; 1␣/␤-H), 4.50 (bs, 1H; 11␣-H), 6.14 (d, J = 1.9 Hz,
1H, 4-H), 7.43 (s, 1H; H-pyrazole), 7.46 (AAꢀBBꢀ, JAB
=
[
131I]NaI (Nycomed-Amersham) in NaOH.

F. Wüst et al. / Steroids 68 (2003) 177–191
183
2.3.2. 2ꢀ-([18F]-4-Fluorophenyl)-17α(1-propyn-
2.4. Biological methods
1-yl)-11β,17β-dihydroxy-androst-4-eno[3,2-c]-
pyrazole ([18F]-7)
2.4.1. Determination of the receptor binding affinity
Compounds tested in the GR-binding assay were analyzed
by HPLC indicating a chemical purity of 93% for com-
pound 17 and a chemical purity ≥ 95% for compounds 5–8,
10–13, 15 and 21–22. The binding affinity of the arylpyra-
zolo steroids for the GR was determined by a modification
of the method reported for the estrogen receptor [8,19]. The
radiotracer was [3H]dexamethasone and adrenalectomized
male rat liver cytosol was used as the protein. The cytosol
was incubated with buffer or several concentrations of unla-
beled competitor together with [3H]dexamethasone at 0 ◦C
for 18–24 h. The unlabeled arylpyrazolo steroid competitor
was prepared in 1:1 dimethylformamide-buffer to ensure
solubility. The free steroid was adsorbed on dextran-coated
charcoal as described by Katzenellenbogen et al. [20].
[3H]Dexamethasone was [1,2,4,-3H]dexamethasone (Amer-
sham Biosciences, Piscataway, NJ).
The reaction was employed using 10 mg of iodonium
salts 24 or 25. After resolubilization the radioactivity (ca.
3–4 mCi, 110–150 MBq) was transferred to the labeling
precursor 24 or 25 in DMF (500 ␮l). The vial was sealed
and heated for 40 min at 120 ◦C. Radio-TLC (Rf = 0.38;
80% EtOAc/hexane) showed the formation of 0.2% (with
iodonium salt 24) and 2% (with iodonium salt 25) of the
desired compound [18F]-7 together with large amounts
of non-reacted [18F]fluoride and [18F]fluorobenzene and
[18F]fluorotoluene, respectively.
2.3.3. 2ꢀ-(4-Fluorophenyl)-17α-(3-[18F]fluoro)-1-
propyn-1-yl)]-11β,17β-dihydroxy-androst-4-eno
[3,2-c]-pyrazole ([18F]-10)
The reaction was employed using 2–3 mg of mesylate 9.
After resolubilization the radioactivity (4.0 mCi, 150 MBq)
was transferred to mesylate 9 in acetonitrile (500 ␮l). The
vial was sealed and heated for 20 min at 65 ◦C. After cool-
ing and addition of ethyl acetate (2 ml) the solution was
passed through a silica-gel plug. The eluate (11 MBq, 9%,
decay-corrected) was analyzed by radio-TLC (Rf = 0.43;
80% EtOAc/hexane) indicating a radiochemical purity of
>93%.
3. Results
3.1. Synthesis of 4-fluorophenylpyrazolo steroids
containing 17α-alkynyl side chains
Natural corticosteroids possess the characteristic 17␣,21-
dihydroxy-20-keto substitution pattern of the pregnane side
chain, as exemplified by cortisol 1. However, several re-
ports on glucocorticoids have shown that the D-ring is
tolerant toward different substitution patterns at position
C17 of the steroid skeleton while retaining high binding
affinity to the GR [21]. Moreover, D-ring modified deriva-
tives often possess greatly reduced binding affinities for
the mineralocorticoid receptor, thus constituting steroids
called “pure glucocorticoids” [21,22]. Since 17␣-alkynyl
substituted steroids are easily accessible by 1,2 addition
of lithium or magnesium based organometallic acetylides
to 17-keto steroids [23,24], we undertook the synthesis of
a series of novel 4-fluorophenylpyrazole steroids in which
the characteristic corticosteroid 1,3-dihydroxyacetone side
chain was replaced by a 17␣-alkynyl-17␤-hydroxy moiety.
The resulting steroids (7, 8, 10–13) combine the beneficial
effects of the 17␣-alkynyl-17␤-hydroxy and 4-fluorophenyl
pyrazole moieties for high and selective binding to the GR.
Furthermore, the modification of the 17␣-alkynyl side chain
by either neutral (Me, OH, F) or basic (cyclic aliphatic
amines) substituents results in interesting effects on bio-
logically important parameters like receptor binding and
lipophilicity (vide infra).
2.3.4. 2ꢀ-(4-Fluorophenyl)-21-[18F]fluoro-
20-oxo-11β,17α-dihydroxy-pregn-4-eno
[3,2-c]pyrazole ([18F]-17)
The reaction was employed using 2–4 mg of mesylate 14
or iodo precursor 16. After resolubilization the radioactivity
(3–5.4 mCi, 110–200 MBq) was transferred to the labeling
precursor 14 or 16 in acetonitril (500 ␮l). The vial was sealed
and heated for 20 min at 65 ◦C. After cooling and addition
of ethyl acetate (2 ml) the solution was passed through a
silica-gel plug. The eluate (3.1 MBq, 4%, decay-corrected
for precursor 14 and 21.8 MBq, 14%, decay-corrected for
precursor 16) was analyzed by radio-TLC (Rf = 0.55; 80%
EtOAc/hexane) indicating a radiochemical purity of >95%.
2.3.5. 2ꢀ-([131I]-4-Iodophenyl)-17α(1-propyn-
1-yl)-11β,17β-dihydroxy-androst-4-eno[3,2-c]-pyrazole
([131I]-22)
The reaction was employed using 1.5 mg of organostan-
nane 23. [131I]NaI (2 ␮Ci, 74 kBq) was added to a solution
consisting of 50 ␮l buffer solution (5% NaOAc in HOAc)
and 50 ␮l oxidizing solution (30% H2O2/HOAc 2:1). Tin
precursor 23 in 200 ␮l MeOH was added and the solution
was stirred at room temperature for 30 min. Then, 70 ␮l of
a 0.1 M Na2S2O3-solution was added to stop the reaction.
Radio-TLC (Rf = 0.47; 80% EtOAc/hexane) revealed the
formation of 78% of the desired product [131I]-22 besides
non-reacted [131I]iodide. After the addition of ethyl acetate
(2 ml) and silica-gel purification the radiochemical purity of
the eluate was >95%.
The synthesis of 17-keto steroid 6 as prerequisite
starting material for 17␣-alkynyl substituted steroids via
organometallic 1,2 addition is shown in Fig. 2.
Starting from commercially available cortisol 1, the
synthesis commenced by protecting the 1,3-dihydroxyace-
tone side chain of 1 by treatment with formaldehyde and

184
F. Wüst et al. / Steroids 68 (2003) 177–191
Fig. 2. Synthesis of 2ꢀ (4-fluorophenyl)-11␤-hydroxy-androst-4-ene-17-one[3,2-c]pyrazole 6.
concentrated HCl in chloroform to afford cortisol bis(methy-
lenedioxy)ether 2. It was observed that a portion of the
product was also converted to the 11␤-MOM ether of 2
under these reaction conditions.
pathway for the introduction of 17␣-alkynyl side chains and
their further functionalization is depicted in Fig. 3.
Treatment of ketone 6 with 1-propynylmagnesium bro-
mide in THF at room temperature gave steroid 7 in
57% yield. The 1,2 addition of the lithium derivative of
tetrahydro-2-(2-propynyloxy)-2H-pyran to ketone 6 af-
forded alcohol 8 in 62% yield after removal of the THP-ether
protecting group. Alcohol 8 was used as a suitable starting
material for further modifications of the 17␣-alkynyl side
chain. For this purpose the primary hydroxyl group in 8 was
first converted into the corresponding mesylate to obtain a
good leaving group. This approach allows the convenient
and flexible terminal functionalization of the alkynyl side
chain with a variety of substituents by displacement of the
mesylate with several nucleophiles. Thus, treatment of me-
sylate 9 with TBAF in THF afforded the fluoride-substituted
steroid 10 in 38% yield. In another set of reactions, mesy-
late 9 was treated with several aliphatic cyclic amines in
the system acetone-Na2CO3 to provide steroids 11–13 in
47, 77 and 70% yield, respectively.
The amount of the 11␤-MOM ether contamination
was determined to be 20% by means of 1H-NMR us-
ing the 11␣-H signals at 4.47 ppm (OH) and 4.25 ppm
(MOM). However, the following steps are not affected
by the 11␤-MOM ether group, and 2 could be used as a
mixture of the 11␤-OH and 11␤-MOM compound. The
introduction of the 2-hydroxymethylene functionality in
3 was achieved by reacting the sodium enolate of 2 with
methyl formate in toluene. Subsequent condensation of 3
with 4-fluorophenylhydrazine hydrochloride in the pres-
ence of NaOAc in acetic acid gave 4-fluorophenyl pyrazole
4 in 84% yield. By using these reaction conditions, the
desired 2ꢀ-fluorophenylpyrazole steroid 4 was formed ex-
clusively, and the corresponding 1ꢀ-fluorophenylpyrazole
isomer was not observed. The bis(methylenedioxy) ether
protecting group and the partially present 11␤-MOM ether
were removed in boiling 50% formic acid. The resulting
1,3-dihydroxyacetone side chain in 5 was transformed into
the 17-keto steroid 6 by reduction of the 20-keto group in
5 with NaBH4 and subsequent oxidative cleavage of the in-
termediate 17,20,21-trihydroxy side chain with NaIO4. The
total yield of 6 was 37%, based on cortisol 1. The reaction
The steroids obtained according to the reaction scheme in
Fig. 3 can be subdivided into two classes: one set of com-
pounds (7, 8, and 10) possess neutral substituents (Me in 7,
OH in 8, F in 10) in the 17␣-alkynyl side chain, whereas
compounds 11–13 contain pyrrolidine, piperidine or mor-
pholine as basic amine substituents.

F. Wüst et al. / Steroids 68 (2003) 177–191
185
Fig. 3. Synthesis of 17␣-alkynyl-17␤-hydroxy steroids.
3.2. Synthesis of 4-fluorophenylpyrazolo steroids
containing an oxetan-3ꢀ-one moiety and modified
1,3-dihydroxyacetone side chains
ment reaction was also observed as a competitive reaction
in the synthesis of several fluorine-18 containing cortico-
steroids starting from the corresponding ␣-keto mesylates
[8,11].
The formation of 21-fluoro steroid 17 proved to be chal-
lenging, and ␣-keto mesylate 14 could be converted into
21-fluoro steroid 17 using TBAF in acetonitrile only in very
a low yield (5.8%). Oxetanone 15 was always inevitably
formed as a side product, because of the strong basic charac-
ter of the fluoride anion under these conditions. Adaptation
of this procedure for the formation of the corresponding
F-18 labeled compound also gave disappointing results.
Better results in the F-18 fluorination could be achieved
using the 21-iodo compound 16 rather than the ␣-keto me-
sylate 14. The required 21-iodo steroid 16 was easily pre-
pared by the reaction of mesylate 14 with NaI in acetone in
85% yield.
As another set of compounds, we have synthesized sev-
eral 4-fluorophenyl pyrazolo steroids containing an intact
and modified 1,3-dihydroxyacetone side chain, as well as
a spiro-oxetanone. Steroid 5 possessing the 1,3-dihydroxy-
acetone side chain was obtained as an intermediate in
the reaction sequence for the synthesis of ketone 6
(vide supra, Fig. 2). Steroid 5 was used to convert the
1,3-dihydroxyacetone side chain into the oxetan-3ꢀ-one 15
and to form iodide 16 and fluoride 17 (Fig. 4).
Although several methods are known for preparing
oxetan-3-ones (e.g. by oxidation [25], by acid-catalyzed
decomposition of diazo ketones [26], by [2 + 2] cycload-
dition [27]), only intramolecular displacement reactions
[28] have been used to prepare steroidal oxetane-3ꢀ-ones.
In the case of C17 spiro-oxetan-3ꢀ-ones, the reaction seems
especially challenging, since forcing reaction conditions
are required, and the reaction only proceeds in poor yields.
However, Pons and Simons [29] have described a general
method to overcome these problems and to provide the
desired C17 spiro-oxetan-3ꢀ-ones in good yields. Following
this procedure, first mesylate 14 was prepared from triol
5 using MsCl in pyridine in 78% yield. An intramolec-
ular oxetan-3ꢀ-one ring closure reaction of mesylate 14
using potassium tert-butoxide in THF at room tempera-
ture provided the desired spiro-oxetan-3ꢀ-one 15 in 16%
yield. It is noteworthy that this intramolecular displace-
3.3. Synthesis of 4-bromophenylpyrazolo steroid
21 and 4-iodophenylpyrazolo steroid 22
Within the series of 17␣-alkynyl-17␤-hydroxy steroids,
we also wanted to study the influence of 4-bromophenyl
and 4-iodophenyl derivatives in terms of their GR bind-
ing, since bromine and iodine also have radioisotopes
(Br-76, I-123 or I-124, respectively) suitable for in vivo
imaging. Therefore, we set up the synthesis of two addi-
tional 17␣-(1-propyn-1-yl)-17␤-hydroxy steroids 21 and
22, containing bromine or iodine in the arylpyrazolo moiety
(Fig. 5).

186
F. Wüst et al. / Steroids 68 (2003) 177–191
Fig. 4. Synthesis of steroids containing an oxetan-3ꢀ-one moiety and modified 1,3-dihydroxyacetone side chains.
Employing the same oxidative cleavage reaction as exem-
plified by the transformation of steroid 5 to 17-keto steroid 6,
cortisol 1 was converted into steroid 18 in quantitative yield.
Further reaction of compound 18 requires the selective pro-
tection of the C3 keto group in the presence of the C17 keto
group. Such a selective protection of the 3-keto group in 18
could conveniently be achieved through formation of the cor-
responding 3-enamine by refluxing 3,17-dienone 18 in ben-
zene with pyrrolidine and a trace of TsOH, while removing
the water generated by this reaction. Subsequent 1,2-addition
of 1-propynylmagnesium bromide to the 17-keto group in
19, followed by restoration of the 3-enone structure by de-
protection with NaOAc/HOAc, gave compound 20 in a to-
tal yield of 50% based on cortisol 1. Pyrazole formation
was performed according to the described procedure (vide
supra, Fig. 2), using the condensation of the correspond-
ing 1,3-dicarbonyl compound (derived from 20 by treatment
with NaH and methyl formate in toluene) with 4-bromo- and
4-iodophenylhydrazine [17] in acetic acid. This procedure
afforded the desired arylpyrazolo steroids 21 and 22 in 48%
and 50% yield, respectively.
3.4. Synthesis of diaryliodonium salts 24 and 25
The use of [18F]fluoride in nucleophilic aromatic sub-
stitution reactions represents a very important and widely
applied reaction in fluorine-18 chemistry [30,31]. Nucle-
ophilic aromatic substitutions with [18F]fluoride proceed
efficiently when a suitable good leaving group (nitro, tri-
alkylammonium, etc.) is attached ortho- or para to an
Fig. 5. Synthesis of 4-bromo- and 4-iodophenylpyrazolo steroids 21 and 22.

F. Wüst et al. / Steroids 68 (2003) 177–191
187
Fig. 6. Synthesis of diaryliodonium salts 24 and 25.
electron-withdrawing group (nitro, CHO, CN, etc.) essen-
tial to activate the aromatic ring. In the case of arylpyrazolo
steroids, the pyrazole moiety is not well suited to activate
the aromatic ring efficiently for a nucleophilic aromatic
substitution with [18F]fluoride. However, several diaryliodo-
nium salts have been shown to react with [18F]fluoride
to generate the corresponding aryl fluorides in reasonable
yield, even from non-activated aromatic rings [32]. There-
fore, we have synthesized diaryliodionium tosylates 24 and
25 as precursors for the introduction of [18F]fluoride into
the aromatic ring. The synthesis of aryliodonium salts 24
and 25 is given in Fig. 6.
Several methods have been reported for the preparation
of diaryliodonium salts that comprise the reaction of either
aryltrialkylsilanes [33], aryltrialkylstannanes [34] or aryl-
boronic acids [35] with hydroxy(tosyloxy)iodobenzenes, as
well as electrochemical methods [36]. We choose the re-
action employing aryltrialkylstannanes. For this purpose,
the required trimethylstannyl-containing steroid 23 was pre-
pared via a Pd-catalyzed cross-coupling reaction between
iodoaryl steroid 22 and hexamethyldistannane. Treatment
of steroid 23 with hydroxy(tosyloxy)iodobenzene (Koser’s
reagent) or hydroxy(tosyloxy)-iodotoluene [18] resulted in
an electrophilic aromatic substitution (ipso-destannylation)
reaction that provided the desired diaryliodonium tosylates
24 and 25 in 28 and 57% yield, respectively.
ceeded by [18F]fluoride ion displacement of mesylate pre-
cursors 14 and 9 or iodo precursor 16. For the synthesis of
[18F]-17, resolubilized [18F]fluoride (Kryptofix®/potassium
carbonate, [37]) in acetonitrile was transferred to a glass vial
containing the appropriate precursor 14 or 16, and the re-
action mixture was heated at 65 ◦C for 20 min. To remove
non-reacted [18F]fluoride, the reaction was diluted with ethyl
acetate (2000 ␮l) and applied to a silica “plug” (a Pasteur
pipet with ca. 150 mg silica-gel) and eluted into a flask.
Analysis of the product by radio-TLC (Rf = 0.55; 80%
EtOAc/hexane) indicated a radiochemical purity >95%. The
average radiochemical yield (decay-corrected) was 4% start-
ing from mesylate precursor 14. However, the radiochemical
yield could be increased up to 14% (decay-corrected) when
iodo precursor 16 was used.
Employing the same protocol, steroid [18F]-10 was ob-
tained in 9% decay-corrected radiochemical yield after the
reaction of mesylate 9 with resolubilized [18F]fluoride.
Radio-TLC (Rf = 0.43; 80% EtOAc/hexane) indicated a
radiochemical purity of >93% after purification by silica-gel
filtration.
3.6. Radiochemistry II—18F-labeling using
diaryliodonium salts 24 and 25
In order to incorporate the F-18 label into the metabol-
ically stable position of the aromatic ring, we made use
of diaryliodonium salts 24 and 25 as suitable labeling pre-
cursors (Fig. 8). Diaryliodonium salts 24 and 25 differ in
the functionalization of the aryl ring, being phenyl and
tolyl, respectively. Radiochemistry was performed using
the powerful nucleophilic radiofluorinating agent K[18F]F
3.5. Radiochemistry I—18F-labeling on aliphatic
side chains
The synthesis of 18F-labeled steroids [18F]-17 and [18F]-
10 is depicted in Fig. 7. The incorporation of fluorine-18 pro-

188
F. Wüst et al. / Steroids 68 (2003) 177–191
Fig. 7. Radiosynthesis of 18F-labeled steroids [18F]-17 and [18F]-10.
(generated by treatment of cyclotron-produced [18F]fluoride
with Kryptofix/potassium carbonate) in DMF as the solvent
at 120 ◦C for 40 min. The use of phenyl-functionalized iodo-
nium salt 24 gave the desired product [18F]-7 in only 0.2%
yield. Radio-TLC analysis after silica-gel purification (vide
supra) showed the desired product [18F]-7 (Rf = 0.38; 80%
EtOAc/hexane) together with [18F]fluorobenzene. Because
of its volatility, the amount of [18F]fluorobenzene could not
be quantitated.
deficient aromatic ring, the use of the more electron-donating
tolyl-fuctionalized iodonium salt 25 should favor the forma-
tion of F-18 labeled steroid [18F]-7. This assumption was
confirmed, because using this precursor, the radiochemical
yield of [18F]-7 could be increased up to 2%. However,
radio-TLC analysis also indicated the formation of large
amounts of 4-[18F]fluoro-toluene as a by-product.
3.7. Radiochemistry III—131I-labeling with
trimethylstannyl steroid 23
The observation of [18F]fluorobenzene formation is
consistent with the findings in the literature when non-
symmetric phenyl-functionalized iodonium salts were used
for fluorinations with [18F]fluoride [32]. Since the attack of
the fluoride ion preferentially occurs on the more electron-
Several iodine isotopes such as iodine-120, -123, -124
possess decay properties suitable for in vivo imaging by
positron or single-photon emission tomography (PET or
Fig. 8. Radiosynthesis of 18F-labeled steroid [18F]-7 via diaryliodonium salts 24 and 25.

F. Wüst et al. / Steroids 68 (2003) 177–191
189
Fig. 9. Radiosynthesis of 131I-labeled steroid [18F]-22.
Table 1
SPECT). The convenient half-life of iodine-131 (t1/2
=
8.04 days) makes this isotope ideal for the elaboration of la-
beling protocols using iodine isotopes. Several radioiodina-
tion reactions using organometallic intermediates have been
shown to be very effective, rapid and site-specific methods
for the incorporation of iodine isotopes into small molecules
like steroids [38]. In this context, the use of organostannanes
has provided especially good results for the radioiodination
of aromatic rings via iodo-destannylation reactions. Thus,
we used organostannane 23 as a precursor for radiolabeling
with iodine-131 (Fig. 9). For this purpose, [131I]NaI (ca.
2 ␮Ci, 75 kBq) in a buffer solution (5% NaOAc in HOAc)
was oxidized with H2O2 in HOAc. The progress of the re-
action with tin precursor 23 was monitored by radio-TLC.
The reaction was completed after 30 min, at which point a
solution of Na2S2O3 was added to stop the reaction. After
silica-gel purification (vide supra), the eluate was analyzed
by radio-TLC (Rf = 0.47; 80% EtOAc/hexane), which
indicated a radiochemical purity of >95% for [131I]-22.
Radio-TLC analysis revealed the formation of 78% of
Relative binding affinities (RBAs) of arylpyrazolo steroids for the GR
a
Compound
RBA (0 ◦C) (%)
Log Po/w
Dexamethasone
Cortisol 1
5
6
7
8
10
11
12
13
15
17
21
22
100
8.4b
50
18
56
31
37
7.5
8.5
17.5
22
2.06
1.43, (1.61c)
3.46
3.30
4.50
3.34
4.31
4.82
5.38
3.91
3.08
11
6.5
1.5
3.87
5.22
5.48
Each value represents the mean of two determinations, with a coefficient
of variance of 0.3.
a Log Po/w values have been calculated based on ACDLabs predictions.
b Literature value [8].
c Literature value [39].
[
131I]-22 besides non-reacted [131I]iodide.
effect on the RBA value. In the case of the neutral ter-
minal substituents F (10), OH (8) and Me (7), the RBA
values vary between 37, 31 and 56%, respectively. These
values represent relatively good binding to the GR, and the
binding is significantly higher compared to steroids 11–13
containing basic groups. Thus, the morpholine-containing
steroid 13 reaches an RBA of about 17.5%, whereas the
RBA drops markedly when more basic cyclic amines are
used, being 7.5% for pyrrolidine 11 and 8.5% for piperi-
dine 12. These findings indicate that neutral side chains
support GR binding better than basic side chains. Nev-
ertheless, all measured RBA values within the series of
17␣-alkynyl substituted steroids display at least the same
(compounds 11 and 12) or better GR binding (compounds
7, 8, 10 and 13) compared to the natural occurring steroid
cortisol 1.
In the second set of compounds, the 1,3-dihydroxyacetone
side chain containing steroid 5 shows the highest RBA value,
50% relative to dexamethasone. In contrast, the correspond-
ing 21-fluoro compound 17 displays a reduced binding affin-
ity of 11%, which is comparable to that of cortisol 1. On
the other hand, structurally very different ketone 6 and oxe-
tanone 15 exhibit moderate to good binding of 18 and 22%,
respectively.
3.8. Glucocorticoid receptor binding affinities and log
Po/w values of arylpyrazolo corticosteroids
The GR binding affinities of all arylpyrazolo steroids
and reference compounds dexamethasone and cortisol 1 are
listed, along with calculated log Po/w, values in Table 1.
The GR binding affinities were assayed using liver cytosol
from adrenalectomized rats, with [3H]dexamethasone as ra-
diotracer and unlabeled dexamethasone as standard (RBA =
100%). All steroids tested in this work show lower GR bind-
ing in comparison to the synthetic and high-affinity ligand
dexamethasone. However, some compounds still exhibit ap-
preciable binding, up to 56% that of dexamethasone, which
is considerably better than the binding affinity of the natural
hormone cortisol 1 (RBA = 8.4%). Moreover, all synthe-
sized arylpyrazolo steroids show an increased lipophilicity
(expressed as calculated log Po/w values) by at least one or-
der of magnitude compared to dexamethasone or cortisol 1.
The enhanced lipophilicity might be a benefit for improved
blood–brain-barrier penetration when appropriately radiola-
beled compounds are used for brain imaging studies.
Within the series of 17␣-alkynyl substituted steroids, the
character of the terminal substitutent shows an interesting

190
F. Wüst et al. / Steroids 68 (2003) 177–191
Comparison of the RBA values within a third series of
the aromatic ring the electronic and steric characteristics
apparently prevent any sufficient binding to the GR. This
effect is especially distinct when iodine is used.
compounds reveal a strong influence of the halogen attached
to the phenyl ring on the GR binding affinity. By far the best
binding affinity is observed with 4-fluorophenyl pyrazole 7,
being 56%. With bromine in compound 21 or iodine in com-
pound 22, the RBA drops to 6.5 and 1.5%, respectively. The
low RBA values for bromine- and iodine-containing steroids
21 and 22 in comparison to fluorine-containing steroid 7
confirm the beneficial effect of the 4-fluorophenyl moiety in
terms of high binding affinity to the GR.
Selected candidates were chosen to investigate the label-
ing of arylpyrazolo steroids with the short-lived positron
emitter fluorine-18. The fluorine-18 label could be easily
introduced into aliphatic side chains to afford radiolabeled
steroids [18F]-17 and [18F]-10 in radiochemical yields of 14
and 9%, respectively. It is noteworthy that the incorpora-
tion of fluorine-18 proceeds well despite the free OH-groups
present in precursors 9, 14 or 16.
In an effort to label arylpyrazolo steroids in a metabol-
ically stable position, we attempted the synthesis of
fluorine-18 labeled steroid [18F]-7. Since the aromatic ring
is not sufficiently activated by a strong electron-withdrawing
group, we used diaryliodonium salts 24 and 25 as precursors
for the incorporation of fluorine-18. The experiments show
the influence of the structure of the iodonium salt used on
the yield and product distribution. Iodonium salt 25 contains
a more electron releasing tolyl substituent compared to the
phenyl ring in 24. This change causes an improved yield
in the attack of the [18F]fluoride to the steroidal part of the
iodonium salt. Despite the low decay-corrected radiochem-
ical yields (0.2 and 2.0%, respectively), the metabolically
stable fluorine-18 labeled glucocorticoid [18F]-7 represents
an interesting agent to investigate further for the study of
brain GR by means of PET.
4. Discussion
In our effort to develop imaging agents for brain GRs, we
have prepared several novel arylpyrazolo steroids. The com-
pounds that we have synthesized can be divided into three
classes: 4-fluorophenyl pyrazoles possessing a 17␣-alkynyl
side chain, 4-fluorophenyl pyrazoles containing a selection
of different substitution patterns in the D-ring, and com-
pounds which differ with respect to the halogen attached to
the arylpyrazole moiety. All steroids were tested for bind-
ing to the GR, and their lipophilicity (log Po/w values) was
calculated.
We found that all steroids containing a 4-fluorophenyl
pyrazole moiety possess at least the same binding capabil-
ity to the GR as the natural stress-induced hormone cortisol
1. Relatively high binding of 50 and 56%, relative to dex-
amethasone, was observed for compounds 5 and 7, which
contain the intact 1,3-dihydroxyacetone side chain or the rel-
atively small and neutral 17␣-propynyl side chain, respec-
tively. These binding affinities are 6–7 times better than the
GR binding of cortisol 1. This finding also shows that the
1,3-dihydroxyacetone side chain is not necessary for high
GR binding and can be replaced by neutral 17␣-alkynyl
side chains. By comparing the measured RBA values, it
also became evident that basic substituents like morpholine
and to a greater extent pyrrolidine and piperidine diminish
GR binding significantly. The GR seems to tolerate neutral
17␣-alkynyl side chains rather than basic side chains. Ap-
parently, the basic character of the cyclic amines seems to
have more influence on the GR binding than the steric bulk
when the RBA values of the strong basic pyrrolidine 11
(RBA = 7.5%) and piperidine 12 (RBA = 8.5%) are com-
pared with the less basic morpholine 13 (RBA = 17.5%).
Further evidence for the GR tolerance toward modifi-
cations of the steroidal D-ring was revealed by the rela-
tively good binding affinities of ketone 6 (RBA = 18%)
and oxetanone 15 (RBA = 22%). However, only mod-
erate binding affinity of 11% was found for 21-fluoro
steroid 17, which represents a compound with a modified
1,3-dihydroxyacetone side chain. The beneficial effect of
4-fluorophenyl pyrazoles on the GR was confirmed by com-
parison of the RBA values of fluorine 7 (RBA = 56%),
bromine 21 (RBA = 6.5%), and iodine 22 (RBA = 1.5%).
By introduction of the halogens bromine and iodine into
Acknowledgments
The authors thank Prof. Dr. G. Stöcklin for helpful discus-
sions. Financial support to FW by the Deutsche Forschungs-
gemeinschaft and to JAK and KEC by a grant from the
Department of Energy (DE FG02 86ER60401) is gratefully
acknowledged.
References
[1] De Kloet ER, Sutanto W, editors. Neurobiology of steroids, vol. 22.
New York: Academic Press; 1994.
[2] Joëls M, Vreugdenhil E. Corticosteroids in the brain. Mol Neurobiol
1998;17:87–108.
[3] De Kloet ER, Vreugdenhil E, Oitzl MS, Joëls M. Brain corticosteroid
receptor balance in health and disease. Endocr Rev 1998;19:269–301.
[4] Korte SM. Corticosteroids in relation to fear, anxiety and
psychopathololgy. Neurosci Biobehav Rev 2001;25:117–42.
[5] Reul JMHM, De Kloet ER. Two receptor systems for corticosterone
in rat brain: microdistribution and differential occupation. Endo-
crinology 1985;117:2505–11.
[6] Holsboer F. The rational for corticotropin-releasing hormone receptor
(CRH-R) antagonists to treat depression and anxiety. Psychiatr Res
1999;33:181–214.
[7] De Kloet ER. Brain corticosteroids receptor balance and homeostatic
control. Front Neuroendocrinol 1991;12:95–164.
[8] Pomper MG, Kochanny MJ, Thieme AM, Carlson KE, VanBrocklin
HF, Mathias CJ, et al. Fluorine-substituted corticosteroids: synthesis
and evaluation as potential receptor-based imaging agents for positron
emission tomography of the brain. Nucl Med Biol 1992;19:461–80.

F. Wüst et al. / Steroids 68 (2003) 177–191
191
[9] Hoyte RM, Labaree DC, Fede J-M, Harris C, Hochberg RB.
Iodinated and fluotinated steroid 2ꢀ-aryl-[3,2-c] pyrazoles as potential
glucocorticoid receptor imaging agents. Steroids 1998;63:595–602.
[10] DaSilva JN, Crouzel C, Stulzaft O, Khalili-Varasteh M, Hantraye
P. Synthesis, tissue distribution in rats and PET studies in baboon
brain of no-carrier-added [18F]RU 52461: in vivo evalution as a brain
glucocorticoid receptor radioligand. Nucl Med Biol 1992;19:167–73.
[11] Feliu AL, Rottenberg DA. Synthesis and evaluation of fluorine-18
21-fluoroprednisone as a potential ligand for neuro-PET studies. J
Nucl Med 1986;28:998–1005.
[24] El Amouri H, Vessiéres A, Vichard D, Top S, Gruselle M,
Jaouen G. Syntheses and affinities of novel organometallic-labeled
estradiol derivatives: a structure–affinity relationship. J Med Chem
1992;35:3130–5.
[25] Wojtowitcz JA, Polak R. 3-Substituted oxetanes. J Org Chem
1973;38:2061–6.
[26] Ducruix A, Pascard C, David S, Fischer J-C. Crystal and molecular
structure of a carbohydrate-derived oxetan-3-one. Definitive evidence
for a novel participation of a pyranose ring oxygen atom. J Chem
Soc Perkin Trans 1976;2:1678–82.
[12] Visser GM, Krugers HJ, Luurtsema G, van Waarde A, Elsinga PH,
de Kloet ER, et al. Synthesis and organ distribution of [18F]fluoro-
Org6141 in the rat: a potential glucocorticoid receptor ligand for
positron emission tomography. Nucl Med Biol 1995;22:915–20.
[13] Feliu AL. Synthetic studies with [18F]p-fluorobenzenediazonium
chloride: application to the synthesis of radiolabelled glucocorticoid
[27] Schroeter S, Orlando CM. Photocycloaddition of various ketones and
aldehydes to vinyl ethers and ketene diethyl acetals. J Org Chem
1969;34:1181–7.
[28] Rowland AT, Drawbough RS, Dalton JR. Steroid oxetanones. Part 3.
Synthesis of 5,7-alpha-epoxy-5-alpha-cholestan-6-ones. J Org Chem
1977;42:487–90.
[29] Pons M, Simons Jr SS. Facile, high-yield synthesis of spiro C-17
steroidal oxetan-3ꢀ-ones. J Org Chem 1981;46:3262–4.
[30] Berridge MS, Tewson TJ. Chemistry of fluorine-18 radiopharma-
ceuticals. Appl Radiat Isot 1986;37:685–93.
[
18F] Win 4457. J Labelled Cpd Radiopharm 1988;25:1245–54.
[14] Schlechte JA, Simons Jr SS, Lewis DA, Thompson EB. [3H]Corti-
vazol: a unique high affinity ligand for the glucocorticoid receptor.
Endocrinology 1885;117:1355–62.
[15] Winneker RC, Russell MM, Might CK, Schane HP. The interaction of
nivazol with the glucocorticoid receptor from rat and rhesus monkey
target tissues. Steroids 1985;42:447–57.
[16] Still WC, Kahn M, Mitra A. Rapid chromatographic technique
for preparative separations with moderate resolution. J Org Chem
1978;43:2923–5.
[17] Baloniak S, Thiel U, Pacholczyk M. Synthesis of 1-(4-iodophenyl)-3-
hydroxypyridazin-6-one derivatives. Acta Pol Pharm 1976;33:73–9.
[18] Koser GF, Wettach RH. [Hydroxy(tosyloxy)iodo]benzene, a versatile
reagent for the mild oxidation of aryl iodides at the iodine atom by
ligand transfer. J Org Chem 1980;45:1542–3.
[19] Katzenellenbogen JA, Johnsonm Jr HJ, Myers HN. Photoaffinity
labels for estrogen binding proteins of rat uterus. Biochemistry
1973;12:4085–92.
[31] Kilbourn RM. Fluorine-18 labeling of radiopharmaceuticals. Nuclear
Science Series NAS-NS-32203. Washington DC: National Acadamy
Press; 1990.
[32] Shah A, Pike VW, Widdowson DA. The synthesis of [18F]fluoro-
arenes from the reaction of cyclotron-produced [18F]fluoride ion
with diaryliodonium salts.
2043–6.
J Chem Soc Perkin Trans 1998;1:
[33] Koser GF, Wettach RH, Smith CS. New methodology in iodonium
salt synthesis. Reactions of [hydroxy(tosyloxy)iodo]arenes with
aryltrimethylsilanes. J Org Chem 1980;45:1543–4.
[34] Pike VW, Butt F, Shah A, Wissowson DA. Facile synthesis of
substituted diaryliodonium tosylates by treatment of aryltributyl-
stannanes with Koser’s reagent. J Chem Soc Perkin Trans 1999;1:
245–8.
[20] Katzenellenbogen JA, Johnson Jr HJ, Carlson KE, Myers HN.
Photoreactivity of some light-sensitive estrogen derivatives. Use of
an exchange assay to determine their photointeraction with the rat
uterine estrogen binding protein. Biochemistry 1974;13:2986–94.
[21] Teutsch G, Costerousse G, Deraedt R, Benzoni J, Fortin M, Philibert
D. 17␣-Alkynyl-11␤,17-dihydroxyandrostane derivatives: a new class
of potent gluco-corticoids. Steroids 1981;38:651–65.
[22] Ojasoo T, Doré J-C, Gilbert J, Raynaud J-P. Binding of steroids to the
progestin and glucocorticoid receptors analyzed by correspondence
analysis. Steroids 1988;31:1160–9.
[35] Carroll MA, Pike VW, Widdowson DA. New synthesis of diaryl-
iodonium sulfonates from arylboronic acids. Tetrahedron Lett
2000;41:5393–6.
[36] Peacock MJ, Pletcher D. The electrosynthesis of diaryliodonium
salts. Tetrahedron Lett 2000;41:8995–8.
[37] Coenen HH, Klatte B, Knöchel A, Schüller M, Stöcklin G.
Preparation of n.c.a. 17-[18F]fluoroheptodecanoic acid in high yields
via aminopolyether supported nucleophilic fluorination. J Labelled
Cpd Radiopharm 1986;23:455–67.
[38] Wilbur DS. Radiohalogenations of proteins: an overview of
radionuclides. Bioconjugate Chem 1992;3:433–70.
[23] Wüst F, Spies H, Johannsen B. Synthesis of 17␣-substituted mer-
captoalkynyl derivatives of 3,17␤-estradiol. Tetrahedron Lett 1998;
38:2931–2.
[39] Alvarez Nuñez FA, Yalkowsky SH. Correlation between log P and
Clog P for some steroids. J Pharm Sci 1997;86:1187–9.