Nucl. Med. Biol. Vol. 20, No. 3, pp. 351-358, 1993 0883-2897/93 $6.00 + 0.00 Printed in Great Britain. All rights reserved Copyright 0 1993 Pergamon Press Ltd

Synthesis, Receptor Binding and Tissue Distribution of

17a -E[125 Iliodovinyl- 1 lb -ethyl-estradiol

ROBERT N. HANSON’*, MITALI GHOSHAL’, FRANK G. MURPHY’, CATHERINE ROSENTHAL’, RAYMOND E. GIBSON*t,

NELSON FERRIERA’, VIRENDAR SOOD’ and JENNIFER RUCH’

’ Department of Pharmaceutical Sciences, College of Pharmacy and Allied Health Professions, Northeastern University, Boston, MA 02115 and ‘Section of Radiopharmaceutical Chemistry,

Department of Radiology, The George Washington University Medical Center, Washington, DC 20052, U.S.A.

(Received 24 June 1992)

In this study we prepared and evaluated a derivative of estradiol with an ethyl group at the 1 lb-position and an E-iodovinyl group at the 17a-position. This new ligand binds to the estrogen receptor with an affinity slightly less than estradiol (RBA = 43%) at 0°C but much greater (RBA = 890%) at 25°C. The 1251-labeled derivative was obtained by radioiododestannylation of the tri-n-butylstannyl precursor in good radiochemical yield with a specific activity exceeding 1500 Ci/mmol. The tissue distribution in immature female rats was evaluated over a 48 h period to determine uterine uptake and selectivity. Peak uterine uptake at 2 h was 6% ID/g and was significantly greater than that of [3H]estradiol, 2.4% ID/g. Substantial uptake in the uterus was still present at 48 h (2.4% ID/g). Co-administration of estradiol reduced the uptake at 2 and 24 h by 85%. Uterus-to-plasma ratios increased with time, from about 25: 1 at 2 h to nearly 9O:l at 48 h. The affinity, ease of radiosynthesis and tissue distribution of the 17u-E-[‘*‘I]iodovinyl-1 l/3-ethyl-estradiol suggest that further evaluation of this agent as an imaging agent for estrogen-receptor-positive breast cancer is warranted.

Introduction

As part of our program to develop radioiodinated imaging estrogen-receptor-positive tumors we have focused on derivatives of estradiol that possess recep- tor affinities comparable to estradiol. In our early publications (Hanson and Franke, 1984; Hanson et al., 1982, 1984, 1989a, b, 1991) we described the conversion of 17a-ethynyl estradiols to their corre- sponding E-[‘251]iodovinyl products via the tri- n-butylstannylvinyl intermediates. Although the compounds demonstrated good in vivo target tissue uptake, retention and selectivity, their affinity for the estrogen receptor was only equal to or slightly greater than estradiol itself (Hanson et al., 1991; Napolitano et al., 1991).

More recently our work has focused on derivatives which may possess significantly greater receptor affinities, a property that should translate into im- proved target tissue retention and selectivity. Based upon an earlier study (Belanger et al., 1981) that

*Author for correspondence. tPresent address: Research Imaging, Merck, Sharp &

Dohme Research Laboratories, WP44C-2, West Point, PA 19486, U.S.A.

reported that 1 l/I-alkyl substituents enhanced estro- gen receptor affinities, we prepared and evaluated the receptor binding properties of a number of novel 1 l/?-substituted estradiols (Hanson et al., 1990b; Napolitano et al., 1989, 1990). In this series the 1 lb-ethyl estradiol demonstrated a very high affinity compared to the other derivatives. This property, combined with the relative ease of its synthesis, suggested its use in future radiochemical design. Because our experience with the 17a-E-iodovinyl estradiols indicated that they behaved much like their 17a-ethynyl analogs, it appeared that the 17a-E[‘2SI]iodovinyl-l l/?-ethyl estradiol would be a high affinity ligand. The 1 l/I-ethyl substituent would impart high receptor affinity and the 17a-E-iodovinyl group would provide radionuclide stability. In this paper we describe the successful synthesis and recep- tor binding of the unlabeled ligand as well as the radiosynthesis and in vivo tissue distribution of the ‘25 I-labeled radioligand.

Materials and Methods

Melting points (m.p.) were determined on an elec- trotherm apparatus using open capillaries and are uncorrected. Starting materials and reagents were

351

352 ROBERT N. HANSON et al.

obtained commercially and were used without further purification. Solvents were obtained commercially and, except for the THF which was distilled from sodium-benzophenone ketyl, were used without fur- ther purification. Thin layer chromatography (TLC) was conducted on glass backed plates with a 0.25 mm layer of either silica gel or alumina containing a fluorescent indicator. Infrared spectra were taken with a Perkin-Elmer Model 527 spectrophotometer. The proton NMR spectra were recorded on a Varian XL-300 nuclear magnetic spectrometer using tetra- methyl silane (TMS) as an internal standard. Low resolution mass spectra (EI) were determined using a Finnegan 402B apparatus at 70 eV ionization voltage. Optical rotations were determined with a Perkin- Elmer Model 141 polarimeter using a 1 dm cell. Ultraviolet spectra were obtained using a Perkin- Elmer Lambda 3 UVjVIS spectrophotometer.

3-Methoxy-estra-2,5(10)-diene-17p-01, 2

To a 5 L flask containing 265 mL of dry THF was added 1.4 L of liquid ammonia and 21 g (2.9 mol) of lithium metal. A solution of 15 g (0.052 mol) estradiol 3-methyl ether 1 in 225 mL of THF was added dropwise over 1 h. The mixture was stirred at -78°C for 2 h and 400 mL of absolute ethanol was added dropwise over 1 h. The mixture was warmed to ambient temperature, allowed to stir for 16 h and then concentrated under reduced pressure. The mix- ture was diluted with water, extracted with dichloro- methane, dried over MgSO, (anhydrous), filtered and evaporated to dryness to give 14.6 g (96%, 0.050 mol) of the crude product. A small portion was recrystal- lized to give a pure sample, m.p. 108-l 12°C (lit m.p. 109-i 12°C). The crude product was used directly for the next step. IR (KBr)cm-‘: 3365, 1685, 1660. ‘H-NMR [(CD,),CO]: 4.65 (s, lH), 3.75-3.43 (m, 5H), 2.85(m, 2H), 2.55 (m, 2H), 2.1&1.10(m, 16H). MS (m/e) = 288 (M+). UV,,, (acetone) 206 nm.

17fl-Hydroxy-estra-5(10)-ene-3-one, 3

To a solution of 14.5 g (0.050 mol) of the enol ether 2 in a 1: 1 mixture of THF and methanol was added a solution of 19.5g oxalic acid in 250mL water (0.1 M oxalic acid). The solution was stirred at ambient temperature for 1 h, concentrated by evapor- ation at 40°C diluted with 500mL water and ex- tracted with dichloromethane. The organic layer was washed with a saturated sodium bicarbonate sol- ution, dried over MgSO, (anhydrous) filtered and evaporated to dryness to give 13.05 g (95%, 0.047 mol) of crude product. Sample was recrystal- lized from EtOH m.p. 180-182°C (lit m.p. 187-188°C). The IR and NMR spectra were also consistent with the proposed structure and the crude product was used directly for the next step. IR (KBr) cm-‘: 3400, 1695. ‘H-NMR [(CD,),CO] 3.70 (m, lH), 3.00-2.70 (m, 3H), 2.7G0.97 (m, 19H), 0.75 (5, 3H). UV,,, (acetone) 207 nm. MS (m/e) = 274 (M+).

17/I-Hydroxyestra-4,9-diene-3-one, 4

A solution of 10.4 g (0.037 mol) of the ene-one 3 in 120 mL of pyridine was cooled to 0°C. To the sol- ution was added 13.3 g (0.041 mol) pyridinium bro- mide perbromide in portions over 1 h. The resultant mixture was heated at 5@-6O”C for 1.5 h, cooled to ambient temperature and then poured into 375 mL of cold 6 N hydrochloric acid. The solution was ex- tracted with dichloromethane. The organic layer was washed twice with aqueous NaHCO, (saturated), water, dried over MgSO, (anhydrous), filtered and evaporated to dryness to give 9.75 g (0.035 mol, 95%) of crude product. The crude product was purified by flash chromatography, on silica gel using 3 : 1 ethyl acetate:hexane to give the pure product 4 6.83 g (70%, 0.025 mol), m.p. 182-183°C (lit m.p. 187-188°C). The NMR and IR spectra were consist- ent with the proposed structure. ‘NMR (CDCI,): 0.93(s, 3H), 1.06-3.13 (m, 18H), 3.63 (m, lH), 5.63(brdm, 1H). IR (KBr): 3500, 1640, 1590. MS (m/e) = 272 (M+). UV,,, (acetone) 327, 211 nm.

Estra-4,9-diene-3,17-dione, 5

A pre-dried solution of N-methylmorpholine-N- oxide, NM0 (9.5 g), in 75 mL of dry dichloro- methane was added to a dichloromethane solution containing 7.3 g (0.027 mol) of the alcohol 6. The mixture was stirred under nitrogen for 10min and 50 mg of tetrapropylammonium perruthenate, TPAP, was added. The reaction solution was stirred at ambient temperature for 3 h following the progress reaction by TLC. Because starting material was still present, an additional 9.5 g of NM0 and 50 mg of TPAP were added and the reaction mixture was stirred for an additional 18 h. The mixture was diluted with 200 mL dichloromethane, filtered and the filtrate was washed sequentially with aqueous NaHCO, (saturated), brine and aqueous cupric sul- fate (saturated). The organic layer was dried over MgS04, filtered and evaporated to dryness. The residue was purified by column chromatography on silica gel using ethyl acetate: hexane (2: 3) as the eluent. The solution gave 5.6 g (74%, 0.020 mol) of the product 5 and recovered starting material 4, 1.2 g (15%, 0.4 mol). The product had an m.p. 142-143°C and NMR and IR spectra consistent with the pro- posed structure. NMR (CDCI,): 1.01 (s, 3H), 1.14-3.24 (m, 18H), 5.98 (s, Br, 1H). IR (KBr): 1720, 1640cm-‘. UV,,, (CDCI,), 228 nm, 296 nm. MS (m/e = 270 (M+).

3,17-Bis(2,2-dimethyl-trimethylene-1,3-dioxy)-estra- 5(10),9(1 I)-diene, 6

To 4.8 g (17 nmol) of the diene-dione 5 in 400 mL of dry benzene was added 14.7 g of 2,2-dimethyl- propan-1,3-diol and 0.225 g of p-toluene-sulfonic acid monohydrate. The mixture was heated at reflux for 24 h with continuous azeotropal removal of water (Dean-Stark trap). The reaction was cooled to room

Study of 17a-E[‘251]iodovinyl-1 Ig-ethyl-estradiol 353

temperature, poured into 600mL of a saturated aqueous NaHCO, solution and the organic layer separated. The organic layer was washed with brine, dried over MgSO, (anhydrous), filtered and evapor- ated to dryness. The residue was purified by column chromatography on neutral alumina (activity III), using hexane : ethyl acetate (95 : 5) as the eluent. The product was isolated in an 87% yield (6.8 g) as a colorless solid, m.p. 154156°C. The ’ H-NMR, IR and electron impact mass spectra were consistent with the proposed structure. MS (m/e) = 442 (M+). ‘H-NMR (CDCI,), 0.53-3.13 (overlapping aliphatic multiplets, 18H), 3.49 (brd m, 8H), 5.43 (brd s, lH, C,,-H) IR (KBr); 2940 cm-‘.

3,17- Bis (3,2- dimethyl - trimethylene - 1,3- dioxy)- See,

I&-oxido-estra-9(11)-ene, 7a

A solution of the diketal 6 (2.05 g, 4.6 mmol) in 7.5 mL of dichloromethane was cooled to 0°C. To the cooled, magnetically stirred solution was added 23 FL of pyridine, 64 mL of hexachloroacetone and 724 PL of 30% hydrogen peroxide. The mixture was stirred at 0°C for 0.5 h and then warmed to ambient temperature. After stirring at ambient temperature for 16 h the reaction was diluted with 100 mL of dichloromethane, washed twice with 100 mL NazS,O, and once with 50 mL of brine. The organic layer was dried over magnesium sulfate (anhydrous), filtered and evaporated to dryness. The mixture was separated by column chromatography on neutral alumina (activity III) with hexane: ethyl acetate (95: 5) as the eluent to give 12% (260 mg, 0.5 mmol) of recovered starting material 6, 5% (107 mg, 0.2 mmol) of pure 5a, lOa-oxido product 7a and 29% (615 mg, 1.3 mmol) of pure 58, 10s -oxide product 7b. Another 44 mg of a 1: 1 mixture of the 5cr, 1Oa and 5b, IO/?-oxido isomers were isolated along with more polar side products. The ‘H-NMR and IR spectra were consistent for the two isomeric epoxides.

7a: (5n, lOa)(CDCl,): 0.71 (s, 3H), 0.79 (s, 3H), 0.86 (s, 3H), 102 (s, 3H), 1.14 (s, 3H), 1.162.80 (m, 12H), 3.3-3.7(m, 14H), 6.05(m, lH, C,,-H).

7b: (S/l, lO~)(CDCl,): 0.70(s, 3H), 0.81 (s, 3H), 1.05(s, 3H), 1.15(s, 3H), l.lS-1.74(m, 12H), 3.30-3.72(m, 14H), 5.88(m, lH, C,,-H).

1 l/I-Ethyl-estra-4,9-diene-3,17-dione, 8

To 5 mL of dry THF at 0°C under a nitrogen atmosphere was added 1.9 mL (3.8 mmol) of 2 M ethylmagnesium bromide in THF, followed by 10 mg of copper (I) chloride. The violet mixture was stirred at 0°C for 0.5 h before a THF solution containing 170 mg (0.38 mmol) of the 5a, lOa-epoxide 7a was added dropwise. The resultant mixture was main- tained at 0°C for 0.5 h, warmed to ambient tempera- ture and then stirred at ambient temperature for 16 h. The reaction mixture was poured into 50 mL of a saturated NH,Cl solution and extracted with dichloromethane. The organic layer was dried over magnesium sulfate (anhydrous), filtered and evapor-

ated to dryness. The residue was dissolved in 3 mL of 70% acetic acid and the resultant solution was heated at 5060°C for 1.5 h. The reaction was cooled to ambient temperature, poured into 100 mL of satu- rated aqueous NaHCO, and extracted with dichloro- methane. The organic layer was washed with brine, dried over magnesium sulfate (anhydrous), filtered and evaporated to dryness. The residue was purified by flash chromatography on silica gel using hex- ane:ethyl acetate (1: 1) as the eluent to give 43 mg (0.14 mmol, 40%) of the product 8, m.p. 118-I 19°C. The ‘H-NMR, IR and mass spectra were consistent with the proposed structure. MS (m/e) = 298 (M + ). ‘H-NMR (CDCI,): 0.93 (s, 3H), 0.99 (t, 6H), 1.13-3.09(m, 19H), 5.66(s, 1H; C,-H) IR (KBr): 1730, 1660 cm-‘.

11/I-Ethyl-estra-1,3,5(10)-triene-17-one, 9

To a solution of 178 mg (0.6 mmol) of the diene- dione 8 in dichloromethane, cooled to 0°C were added 2 mL of acetic anhydride and 0.1 mL of acetyl bromide. The reaction was stirred at 0°C for 1 h and then poured into 10 mL of aqueous NaHCO,. The mixture was extracted with dichloromethane, the organic layer separated and evaporated to dryness. The residue was dissolved in 2 mL of methanol. A 10 N sodium hydroxide solution (1 .O mL) was added and the reaction was stirred at ambient temperature for 2 h. The reaction mixture was acidified with 2 N sulfuric acid and extracted with dichloromethane. The organic layer was washed with brine, dried over magnesium sulfate (anhydrous), filtered and evapor- ated to dryness. The residue was purified by column chromatography on silica gel using hexane : ethyl acetate (1: 1) as the eluent to give 100 mg (0.3 mmol, 56%) of the product 9, m.p. 26G262”C. The ‘H- NMR, IR and mass spectra were consistent with the proposed structure. ’ H-NMR (CDCI,): 0.96 (s, 3H), l.O6(t, 3H), 1.1&2.94(m, 18H), 4.68 (brd s, lH), 6.56 (m, IH), 6.70(m, IH), 7.06 (m, 1H). IR (KBr): 1700 crn~-‘. MS (m/e) = 209 (M+).

(17a,2OE)21- Tri-n-butylstannyl- 11/3-ethyl- 19-nor-

pregna-1,3,5(10,20)-tetraene-3,17fi-dial, 10

To a solution of 210 mg (0.7 mmol) of the ketone 9 in 8 mL of dry THF was added 0.43 mL (0.7 mmol) of 1.6 M n-butyllithium under nitrogen. The pale yellow solution that formed was allowed to stir at ambient temperature for 1 h. In a separate flask, 20 mL of dry THF was added to 1.85 mL (3.5 mmol) of E-1,2-bis(tri-n -butylstannyl)-ethylene under nitro- gen. The solution was cooled to -78°C and 2.2 mL (3.5 mmol) of 1.5 M n-butyllithium was added drop- wise. The resultant solution was stirred at - 78°C for 1 h, then allowed to warm to -40°C and stirred at that temperature for 1 h. The solution of the vinyl- lithium reagent was retooled to -78°C and was transferred dropwise to the cooled initial solution under nitrogen. The resultant mixture was stirred at -78°C for 2 h, quenched with saturated aqueous

354 ROBERT N. HANKIN et al.

NH4Cl and extracted with dichloromethane. The organic layer was washed with brine, dried over magnesium sulfate (anhydrous), filtered and evapor- ated to dryness. The residue was purified by column chromatography on silica gel using hexane initially to elute the stannylated byproducts followed by hex- ane:ethyl acetate (8: 2) to elute 197 mg (0.31 mmol, 45%) of the desired product 10, 55 mg (0.15 mmol, 20%) of the 17a-n-butyl byproduct, and 65mg (0.21 mmol, 30%) of the starting ketone 9. The ’ H-NMR, IR and mass spectra were consistent with the proposed structures for the products. ‘H-NMR (CDCI,): 0.51-2.94 (m, 45H), 5.51 (m, lH), 6.08 (d, 2H), 6.44(s, lH), 6.61 (m, lH), 6.84(s, 1H).

(17u,20E)-21-Iodo-ll~-ethyl-19-norpregna-l,3,5(lO), 20-tetraene-3,17/3-diol, 11

To a solution of 197 mg (0.31 mmol) of the tri-n- butylstannyl derivative 10 in 4mL of freshly THF was added 89 mg (0.39 mmol) of N-iodosuccinimide. The mixture was stirred in the dark at ambient temperature for 0.5 h and then diluted with 25 mL of ethyl acetate. The organic layer was washed with water, brine, dried over magnesium sulfate (anhy- drous), filtered and evaporated to dryness. The residue was purified by column chromatography on silica gel using hexane:ethyl acetate (7: 3) as the eluent to give 15 mg (0.02 mmol, 7%) of the starting material 10, 32mg (0.05mmo1, 17%) of the 4,21- diiodo by-product and 60 mg (0.13 mmol, 41%) of the desired product 11. The i H-NMR, IR and mass spectra were consistent with the proposed structure. ’ H-NMR (CDCI,): 1.11 (s, 3H), 1.9g2.90 (m, 19H), 4.53 (s, lH), 6.65(d, J = 8Hz, lH), 6.35(d, J = 15 Hz, lH), 6.85 (d, J = 15 Hz, lH), 6.67(m, lH), 7.04 (d, J = 8 Hz, 1H). MS (m/e) = 452 (M+).

(l7c(,20E)-21[‘25I]Iodo-l9-ll~-ethyl-norpregna-1,3,5 (10),20-tetraene-3,l 7b-diol[‘251], 11

The compound was labeled using the procedure previously described (Hanson et al., 1984) and iso- lated using HPLC [C-l8 analytical column, ethanol: 0.05M (NH,),HPO,-pH 4.5 buffer-l : 1 as the elu- ent]. The specific activity was determined to be 1500 Ci/ mmol using a saturation analysis method (Jagoda et al., 1984). The product was filtered through a 0.45 pm filter and diluted to a specific concentration of 50 Ci/mL with physiological saline. The solutions were stored at 4°C in the dark until used.

In vitro competitive binding assay

Uterine tissue from immature female lambs was homogenized at 4°C in a O.OlM Tri-1.5 mm EDTA- 0.02% sodium oxide, pH 7.4 buffer (TEA), and the homogenate was centrifuged at 18,OOOg for 60min. The supernatant was removed and the cytosol was diluted to a concentration of approx. 3 mg of protein/ mL.

The competitive binding assays were per- formed using procedures previously described

(Katzenellenbogen et al., 1973, 1977). To the lamb uterine cytosol, containing the 10 nM [3H]estradiol were added aliquots of the competitor such that the range of 10-5-10-‘oM competitor was obtained. The solutions were incubated at @4 or 25°C for 24 h. After 24 h a suspension of dextran-coated charcoal (DCC) was added, followed by centrifugation. An aliquot of the supernatant was removed and the activity was assayed by liquid scintillation counting. The remaining activity associated with specific estra- diol binding was plotted against the competitor con- centration. From this plot the molar concentrations of the unlabeled estradiol and the steroidal competi- tor that reduced the specific binding of [3H]estradiol could be determined. The effectiveness of the com- petitor was established using the ratio of the un- labeled estradiol for 50% competition to competitor concentrations for 50% competition. The ratio was multiplied by 100 to obtain the relative binding affinity (RBA) at a particular temperature. Therefore, by definition, the RBA for estradiol under these conditions is 100 (Raynaud et al., 1979; Katzenellen- bogen et al., 1982).

Tissue distribution in immature female rats

The radioligand (5 PCi in 100 PL solution) was injected via the tail vein in weanling Sprague-Dawley rats (average weight 60 g). Groups of animals (at least 5 per group) were sacrificed at various times and uterus, lungs, liver and blood samples were removed. Heparinized blood was centrifuged and the plasma samples were counted. For one group at 1 h, [3 Hlestradiol (110 Ci/mmol) was coinjected and the animals were sacrificed and dissected in the same way. For two groups, at 2 h and 24 h, 20 pg of unlabeled estradiol per rat was coinjected. The animals were sacrificed and dissected as previously described. Tissue samples (20-50 mg) were counted in an auto- gamma counter with dpm values calculated using an absolute counting method or when ‘H was present, the samples were solubilized in Protosol, liquid scin- tillation cocktail added and the samples were counted for dual isotope (‘H/iz51) in a liquid scintillation counter programmed to provide corrected dpm val- ues for each isotope. The data are expressed as % g (% ID/g) injected dose per g net weight of tissue or plasma and are the average of 4-10 animals per time point.

Chemistry Results

The synthesis of the final compound, 17u-E- [‘*j Iliodovinyl-1 lfl-ethyl-estradiol involved three key steps: (1) the synthesis of the 1 lb-ethylestrone; (2) the synthesis of the 17a -E-(tri-n -butylstannyl-vinyl) intermediates; and (3) the conversion to the corre- sponding 17a-E-iodovinyl or [‘251]iodovinyl product. The procedure which was the most efficient was an adaptation of that described by Teutsch et al. (1979) (Scheme 1). The Birch reduction of estradiol methyl

Study of 17~-E[~~~I]iodovinyl-l l,!Gethyl-estradiol

OH OH

355

cH*& a cHo& b 2 3

f *

g h

0

9 / / 0

8

L

Ho& k ) Ho&ksnB’m )

9 10

OH

HO 11

11 x=1

(1251) - 11 x = ‘=I

Scheme 1. a, Li-NH,; b, oxalic acid; c, Pyr, Pyr H+Bc, ; d, TPAP, NMO; e, HOCH,C(CH,),CH,OH; f, H,02, Pyr, CCl,COCCl,; g, C2H, MgBr, 01(l); h, 70% HOAc; i, AcBr, Ac,O; j, MeOH, HO-; k,

E-Bu,Sn CH = CH - L; 1, NIS; m, lz51-, H,O,-HOAc.

356 ROBERT N. HANSON ef al.

ether followed by hydrolysis (Wilds and Nelson, 1953) and bromination-dehydrobromination (Perelman and Farkas, 1963) gave the 17p -hydroxyestra-4,9-diene- 3-one. Oxidation with N-methylmorpholine-N-oxide (NMO) and tetrapropyl-ammonium perruthenate (TPAP) (Griffith et al., 1987) followed by bis-ketal- ization with 2,2-dimethylpropane-1,3-diol gave the protected estra-5( 10),9( 1 I)-diene-dione. Epoxidation with hydrogen peroxide and hexachloroacetone (Ottow et al., 1984) gave a mixture of 51x, lOcr-oxido and 58, 10/I-oxido isomers in which the 5c(, lOa-iso- mer was the major product. The addition of ethyl- magnesium bromide using Cu (I) catalysis, followed by deketalization gave the 1 ID-ethylestra-4,9-diene- 3,17-dione (Belanger et al., 1981; Teutsch and Belanger, 1979). Aromatization with acetyl bromide and acetic anhydride followed by sodium hydroxide in methanol gave the desired llg-ethylestrone. The overall yield was 3% for the 8 steps, largely as a result of the 29% yield obtained in the epoxidation step.

The introduction of the E-t&n -butylstannylvinyl moiety was achieved via addition of E-tri-n-butyl- stannylvinyllithium (Hanson et al., 1982, 1989a). The product was isolated from the reaction mixture in a 45-55% yield accompanied by a 30-35% recovery

-10 -9 -8 -7 -6 -5 -4

Log concentration (M)

I B

I I \‘Kj. 1. , A I -10 -9 -8 -7 -6 -5 -4

Log concentration (M)

Fig. 1. Relative binding a&itv assay. Various concen- trations of 11 (0) and &radio< (0) were incubated with [‘Hjestradiol in lamb uterine cytosol at 0°C (A) or 25°C (B). Dextran-coated charcoal (DCC) was used to absorb the unbound steroid. Relative binding affinity (RBA) values

were determined relative to estradiol (RBA = 100).

Table 1. Distribution of 1’251111 in weanline female rats (% ID/e)

Time (h) n

a 10

1 8

2 8

6 IO

13 4

24 10

48 9

uterus 2.01

(0.11) 3.90

(0.22) 6.09

(0.60) 4.03

(0.17) 3.23

(0.13) 3.36

(0.29) 2.38

(0.24)

Liver

4.60 (0.16) 2.80

(0.14) 2.90

(0.23) 1.40

(0.097) 0.815

(0.035) 0.862

(0.15) 0.603

(0.16)

Lung

1.28 (0.062)

0.936 (0.11) 0.903

(0.079) 0.369

(0.026) 0.105

(0.0091) 0.0831

(0.0075) 0.025 1

(0.0018)

Plasma

0.406 (0.016) 0.248

(0.0098) 0.225

(0.0093) 0.172

(0.0092) 0.0657

(0.0048) 0.0636

(0.0084) 0.0267

10.0021~

Standard error in parentheses.

of lib-ethylestrone and a lo-20% yield of the 1 la-ethyl-17a-n-butylestradiol.

The unlabeled 17E -E-iodovinyl- 11 fi -ethylestradiol was prepared by the iododestannylation of 10 using N-iodosuccinimide (NIS) in THF. Unlike the destannylation reactions with iodine which gave tri- n-butyltin iodide as an undesirable side product, this method significantly reduced the stannylated con- taminants. At ambient temperature the reaction gave the desired compound as the major product (41% yield after chromatography) accompanied by the 4,21-diiodionated product (17%) and unreacted starting material (7%). Lowering the reaction tem- perature to 4°C reduced the diiodinated product to less than 5% of the reaction mixture.

The radioiodination was achieved in a 40% radio- chemical yield using our previously described pro- cedure (Hanson et al., 1982). Under these conditions in which the stannylvinyl precursor is in great excess compared to the iodide, diiodination was not ob- served. The 17cl-E-[‘2SI]iodovinyl-l l/3-ethylestradiol was isolated by HPLC to give a final product that was radiochemically pure (> 98%) with a specific activity of approx. 1500 Ci/mmol (2200 Ci/mmol maximum) as determined by saturation analysis (Jagoda et al., 1984).

Biological properties

The affinity of the unlabeled compound for the estrogen receptor was determined by competitive binding studies using a lamb uterus cytosol prep- aration at 0 and at 25°C. The new ligand was com- pared to unlabeled estradiol at both temperatures to determine the RBA values. Under these conditions 17a-E-iodovinyl-1 lb-ethylestradiol possessed an RBA of 43% at 0°C and 890% at 25°C. The displace- ment curves for this new ligand and estradiol are shown in Fig. l(A) and (B).

The tissue distribution data of the 17a-E-[‘251]- iodovinyl-ll/?-ethylestradiol in immature female rats are summarized in Table 1 and Figs 2 and 3. Specific receptor binding in vivo waF evaluated by coadminis- tration of 20 pg of estradiol to two groups which were sacrificed at 2 and 24 h post-injection, respectively. The level of radioactivity in uterus, liver, lungs and

Study of 17a -E[ o5 Iliodovinyl-1 lb -ethyl-estradiol 357

plasma was determined over a 0.2548 h time interval and plotted as a percent injected dose per gram (% ID/g). Selectivity for the target tissue (uterus) was evaluated by determining the target-to-non-target tissue ratios (T : NT) and plotting those values versus time.

Discussion

In this study we report the synthesis and biological evaluation of a new radioligand for the in uiuo

detection of estrogen receptor containing tissues. Although the synthesis of the final product is long and proceeds in a low overall yield, the conversion of the final radiolabeling step (40-50%) is satisfactory, particularly because such a small quantity of tin precursor is required. The synthetic route to the 1 l/l- ethyl-estrone reported here proceeds in yields com- parable to those described by others (Pomper et al.,

1990). This route has been subsequently improved upon by our laboratory (Hanson et al., 1990a). Therefore we can prepare the precursor for the target radioligand in reasonable yields.

The results of the binding studies are presented in Fig. l(A) and (B). The results suggest that the compound possesses very favorable receptor binding properties. Our earlier studies with 1 lp-substituted estradiols indicated that the presence of the 1 l/l-sub- stituent tended to lower the RBA at 0°C indicative of an impeded association with the receptor. At 25°C however, several of these 1 lb-substituted estradiols possessed RBA values much greater than estradiol, a measure of retarded dissociation. Compared to 1 lg-ethylestradiol, the 17a-E-iodovinyl derivative demonstrated slightly lower RBA values at both 0 and 25°C however, the later value (890%) still is as high or much higher than most of the other estrogenic radioligands described in the literature (Ali et al.,

1988, 1991; Hochberg, 1979; Kiesewetter et al.,

1984; Pomper et al., 1990). The tissue distribution data are consistent with a

radioligand that processes high estrogen receptor

6 i-t

5 u\

1 h :-. II b I I 1 12 13 24 48

Time (h)

Fig. 2. Time-course of the distribution of [‘251]11 in the weanling female rat. Uterus (O), liver (+), and lung (A). Uterine uptake for [‘251]11 coinjected with 20 ng estradiol

(m), Uterine uptake for [)H]estradiol (0).

75 -

T 60 -

NT

12 6 13 24 48

Time (h)

Fig. 3. Time-course for uterus-to-non-target tissue ratios (T:NT) in the weanling female rat. Uterus-to-plasma (a),

uterus-to-liver (+) and uterus-to-lung (A).

affinity. Uptake in the estrogen receptor rich uterus increases over the first 2 h post-administration to a peak of 6.1% ID/g. This peak uptake is 2f times greater than that observed for [3H]estradiol in the same animals and indicates the enhanced localization in this ligand. In addition, a high level of radioactivity is maintained in this tissue out to 48 h. High levels of activity are also observed initially (0.25 h) in the liver, lung and plasma, however, these levels decline con- tinuously over the entire 48 h time period, indicative of non-receptor mediated localization and clearance. Coadministration of cold estradiol produced an 85% decrease in uterine localization at 2 and 24 h, demon- strating that the uptake in the uterus was receptor mediated. The levels of activity in the other tissues and plasma were not significantly affected.

The target-to-non-target ratios, as a measure of selectivity, are shown in Fig. 3. As the results show, the radioligand demonstrates a significant degree of selectivity. Although uterus-to-non-target tissue ratios are low in the early time periods (< 2 h), by 2 h the ratios exceed 30: 1, climbing to > 80: 1 by 48 h. These values are comparable to those observed for other radioiodinated and “F-labeled estrogens pro- posed as imaging agents (Ali et al., 1988, 1991; DaSilva and van Lier, 1990; Gatley et al., 1981; Jagoda et al., 1984; Kiesewetter et al., 1984; Nakatsuka et al., 1984; Pomper et al., 1990; Zielinski et al., 1989).

In summary, we have described the synthesis of a new ligand for the estrogen receptor which can be readily prepared in its radioiodinated form for in uiuo

receptor localization studies. The ligand possesses a very high affinity for the estrogen receptor that is demonstrated in uiuo by high uterine uptake, retention and selectivity. The results suggest the future work should focus on a comparison of the localization kinetics of this compound versus other estrogenic ligands.

Acknowledgements-This research was supported in part by a grant from the Public Health Service (lROlCA41399 to Robert N. Hanson). The authors acknowledge the

358 ROBERT N. HANSON et al.

contribution of Dr John A. Katzenellenbogen, supported by PHS-5-ROI-CA-25836, for the work on the receptor binding assay.

References

Ali H., Rousseau J., Ghaffari M. A. and van Lier J. E. (1988) Synthesis, receptor binding and tissue distribution of (17a, 20E)- and (17a, 20Z)-21[1-125]-iodo-19-nor- pregna-1,3,5(10), 20-tetraene-3,17-diol. J. Med. Chem. 31, 1946.

Ali H., Rousseau J., Ghaffari M. A. and van Lier J. E. (1991) Synthesis, receptor binding and tissue distribution of 7a- and 1 l/?-substituted (17a, 2OE)- and (17a, 2OZ)-21- [I-125]-iodo-19-norpregna-1,3,5(10), 20-tetraene,3,17- diol. J. Med. Chem. 34, 854.

Belanger A., Philibert D. and Teutsch G. (1981) Regio- and stereo-specific synthesis of 1 lb-substituted-19-nor- steroids. Steroi& 37, 361.

DaSilva J. N. and van Lier J. E. (1990) In uiuo evaluation of 7a-[l I-(4-[I-125]-iodophenyl undecyl] 17/3-estradiol; a potential vector for therapy of adrenal and estrogen receptor-positive cancers. J. Steroid Biochem. Molec. Biol. 37, $7. -

Gatley S. J., Shaugnessy W. J., Inhorn L. and Lieberman L. M. (1981) Studies with 17/?(16a-[I-125]iodo)-estradiol, an estrogen receptor-binding radiophatmaceutical. in rats bearing mammary tumors. J. Nk. Med. 22, 459.

Griffith W. P., Ley S. V., Whitcombe G. P. and White A. D. (1987) Preparation and use of tetra-n-butylammonium perruthenate (TBAP reagent) and tetra-n-propylammo- nium perruthenate (TPAP reagent) as new catalytic oxi- dants for alcohols. J. Chem. Sot. Chem. Commun. 1625.

Hanson R. N. and Franke L. A. (1984) Preparation and evaluation of 17a-[I-125]-iodovinyl-I l/3-methoxyestra- diol as a highly selective radioligand for estrogen recep- tor-containing tissues: concise communication. J. Nucl. Med. 25, 998.

Hanson R. N., Franke L. A. and Kaplan M. L. (1989a) Synthesis and evaluation of (17a, 20E)-21-[‘*‘I]iodo-1 l- substituted-lPnorpregna-1,3(10), 20-tetraene-3,17/I-diols: the influence of 11-stereochemistry on tissue distribution of radioiodinated estrogens. Nucl. Med. Biol. 16, 3.

Hanson R. N., Franke L. A. and Kaplan M. L. (1990a) Radioiodinated ligands for the estrogen receptor. Tissue distribution of 17a-[‘251]iodovinylestradiol derivatives in normal and tumor-bearing adult female rats. Nucl. Med. Biol. 17, 239.

Hanson R. N., Seitz D. E. and Bottaro J. C. (1982) E-17a -[I- 125]-iodovinylestradiol: an estrogen receptor seeking radiopharmaceutical. J. Nucl. Med. 23, 43 1.

Hanson R. N., Seitz D. E. and Bottaro J. C. (1984) Radiohalodestannylation. Synthesis of [I-125]-labeled 17a-E-iodovinyl estradiol. Int. J. Appl. Radiat. Isot. 35, 810.

Hanson R. N., Napolitano E. and Fiaschi R. (1991) Structure-activity relationships for 17a-E/Z-vinyl and l l-substituted-17a-E/Z-vinyl estradiols. J. Labelled Cmpd. Radiopharm. 30, 437.

Eckelman W. C. (1984) [I-125]-17a-Iodovinyl 1 lb- methoxylestradiol: in uiuo and in vitro properties of a high-affinity estrogen-receptor radiopharmaceutical. J. Nucl. Med. 25, 472.

Katzenellenbogen J. A., Johnson H. J. Jr and Myers H. M. (1973) Reagents for photoafinity labeling of estrogen binding protein. The binding affinity of some azide and diazo derivatives of estradiol, estrone and hexestrol for the estrogen receptor binding protein of rat uterus. Biochemistry 12, 4085.

Katzenellenbogen J. A., Carlson K. E., Johnson H. J. Jr and Mvers H. M. (1977) Estroeen nhotoaffinitv labels. II. Reversible binding and coval&t aitachment of photo- sensitive hexestrol derivatives of the uterine receptor. Biochemistry 16, 1970.

Katzenellenbogen J. A., Heiman D. F., Carlson K. E. and Lloyd J. E. (1982) In uiuo and in vitro steroid receptor assays in the design of estrogen radiopharmaceuticals. In Receptor Binding Radiotracers (Edited by Eckelman W. C.), Vol. I, P. 93. CRC Press. Boca Raton. Fla.

Kiesewetter D. O:, Kilboum M. R., Landvatter S. W., Heiman D. F., Katzenellenbogen J. A. and Welch M. J. (1984) Preparation of four fluorine-18-labeled estrogens and their selective uptake in target tissues of immature rats. J. Nucl. Med. 25, 1212.

Nakatsuka, I., Ferreira N. L., Eckelman W. C., Francis B. E., Rzeszotarski W. J., Gibson R. E., Jagoda E. M. and Reba R. C. (1984) Synthesis and evaluation of (17a, 20E)-21-[I-125]iodo-19-norpregna-1,3,5(10), 20-tetraene-3, 17diol and (17a, 20E)-21-B-125]-iodo-1 I-methoxy-19-nor- pregna-1,3,5(10), 20-tetraene-3,17-dial (17a-(iodovinyl- estradiol derivatives) as high specific activity potential radiopharmaceuticals. J. Med. Chem. 27, 1287.

Napolitano E., Fiaschi R. and Hanson R. N. (1989) Synthesis of 1, 1 l/3-ethanoestradiol-1,3,5(10)-triene-3, 17B-diol. J. Chem. Sot. Chem. Commun. 1330.

Napolitano E., Fiaschi R. and Hanson R. N. (1990) Estro- gen receptor binding characteristics of 1, 1 lb-ethano- estradiol: effect of a 1, 11/I-bridge on estradiol estrogens. J. Steroid Biochem. Molec. Biol. 37, 295.

Napolitano E., Fiaschi R. and Hanson R. N. (1991) Struc- ture activity relationship of estrogenic ligands. Synthesis and evaluation of (17a. 20E) and (17a. 20Z)-21-halo- genated-19-norpregna-1,3,5(lt$, 20-tetraehe-3, 17P-diols. J. Med. Chem. 34, 2754.

Ottow E., Beier S., Elger W., Henderson D. A., Neef G. and Wiechert R. (1984) Synthesis of ent-17-(prop-l-ynyl- 178- hydroxy- 1 la-(4-(N, N-dimethylamino)-phenyl)-4, 9-estradien-3-one. Steroids 44, 519.

Perelman M. and Farkas A. (1963) 19-Nor-4,9(10)-steroids and process. U.S. Pat. 3,086,077.

Pomper M. G., van Brocklin H., Thieme A. M., Thomas R. D., Kiesewetter D. O., Carlson K. E., Mathias C. J., Welch M. J. and Katzenellenbogen J. A. (1990) 1 l/7- Methoxy-, 1 lb-ethyl- and 17a-ethynyl-substituted 16a- fluoroestradiols: receptor-based imaging agents with enhanced uptake efficiency and selectivity. J. Med. Chem. 33, 3143.

Raynaud J. P., Ojasoo T., Bouton M. M. and Philibert D. (1979) Receptor binding as a tool in the development of new bioactive steroids. In Drug Design (Edited by Ariens E. J.), Vol. VIII, p. 169. Academic Press, New York.

Teutsch G. and Belanger A. (1979) Regio- and stereospecific synthesis of 1 l/I-substituted-19norsteroids. Tetrahed. L&t. 2051.

Wilds A. L. and Nelson N. A. (1953) The facile synthesis of 19-nortestosterone and 19norandro- stenedione from estrone. J. Am. Chem. Sot. 75, 5366.

Zielinski J. E., Lamer J. M., Hoffner P. B. and Hochberg R. G. (1989) The synthesis of 1 l/I-methoxy-[16a-I-123]- iodoestradiol and its interaction with the estrogen receptor in uiuo and in vitro. J. Nucl. Med. 30, 209.

Hanson R. N.,.El-Wakil H., Murphy F. and Wilbur D. S. (1989b) Synthesis and evaluation of Br-82 and Br-77 labeled (lia, 20E)-21-bromo-19-norpregna-1,3,5(10), 20- detraene-3, 17/i-diol. J. Labelled Cmpd. Radiopharm. 27, 615.

Hanson R. N., Napolitano E., Fiaschi R. and Onan K. D. (1990b) Synthesis and estrogen receptor binding of novel 116-substituted-estra-1.3.5(10)-triene-3.17-dio1s. J. Med. &em. 33, 155. ’ ’ ’

Hochberg R. (1979) Iodine-125-labeled estradiol: a gamma- emitting analog of estradiol that binds to the estrogen receptor. Science 205, 1138.

Jagoda E. M., Gibson R. E., Goodgold H., Ferreira N., Francis B. E., Reba R. C., Rzeszotarski W. J. and