Production of 16β-(acetoxy)acetoxy derivatives by reaction of 17-keto steroid enol acetates with lead (IV) acetate

Steroids 66 (2001) 743–748

Production of 16␤-(acetoxy)acetoxy derivatives by reaction of 17-keto steroid enol acetates with lead (IV) acetate Mitsuteru Numazawaa,*, Momoko Shelangouskia, Masamichi Nakakoshib a

Tohoku Pharmaceutical University1, 4-4-1, Komatsushima, Aobaku, Sendai 981-8558, Japan Research Institute of Life Science, Snow Brand Milke Products Co. Ltd., Ishibashimachi, Tochigi, Japan

b

Received 18 September 2000; received in revised form 6 December 2000; accepted 15 December 2000

Abstract Treatment of enol acetates of 3␤-acetoxyandrost-5-en-17-one and its 5␣-reduced analog, 5␣-androstan-17-one, and estrone acetate, 1– 4, with Pb(OCOCH3)4 in acetic acid and acetic anhydride gave the previously unreported products, 16␤-(acetoxy)acetoxy-17-ketones 8 –10 and 12, in 9 –15% yields along with the known major products, 16␤-acetoxy-17-ketones 5–7 and 11. Similar treatment of the 16␤-acetoxy17-ketones with the lead reagent did not yield the corresponding (acetoxy)acetates. Reaction of the enol acetate 3 with Pb(OCOCD3)4 in CD3COOD yielded principally the labeled (acetoxy)acetate 10-d 3 , which had a CD3COOCH2COO moiety at C-16␤. In contrast, when the deuterated enol acetate 3-d 3 , which was obtained by treatment of the 17-ketone 14 with (CD3CO)2O in the presence of LDA and which had a CD3COO moiety at C-17, was reacted with Pb(OCOCH3)4, the resulting product was the labeled compound 10-d 2 . This product had a CH3COOCD2COO function at C-16␤. Based on these results, along with further isotope-labeling experiments, it seems likely that the (acetoxy)acetate is produced through a lead (IV) acetate-catalyzed migration of the 17-acetyl function of the enol acetate to the C-16␤-position followed by attack of an acetoxy anion of the lead reagent. © 2001 Elsevier Science Inc. All rights reserved. Keywords: 17-Enol acetoxy steroid; Lead (IV) acetate; 16␤-(Acetoxy)acetoxylation; 16␤-Acetoxylation; Reaction mechanism; Isotope-labeling

1. Introduction Reaction of 17-keto steroid enol acetates with lead (IV) acetate has been used for stereospecific introduction of an acetoxy group at the 16␤-position [1–3]. This reaction is thought to be ionic as it occurs at room temperature [4]. The products are 16␤-acetoxy-17-ketones in contrast to the 16␣configuration [5–9] of bromination products obtained from the same enol acetates. A recent study using deuteriumlabeling experiments has unambiguously established that the 16␤-acetoxy derivative is not an isomerized product of the 16␣-isomer initially formed by ␣-attack of the reagent, but a product of direct ␤-attack [3,10]. In connection with our study of 16␤-hydroxy steroids, we treated the 17-ketone enol acetates 1– 4 with lead (IV) acetate and found, for the first time, the production of the 16␤-(acetoxy)acetoxy-17-ketones 8 –10 and 12. The mech* Corresponding author. Tel.: ⫹81-22-234-4181; fax: ⫹81-22-2752013. E-mail address: [email protected] (M. Numazawa). 1 Formerly Tohoku College of Pharmacy.

anism by which these 16␤-(acetoxy)acetoxyketones are produced was then studied using deuterium-labeling experiments.

2. Experimental Melting points were measured on a Yanagimoto melting point apparatus (Kyoto, Japan) and are uncorrected. Infrared (IR) spectra were recorded in KBr pellets on a Perkin Elmer FT-IR 1725X spectrophotometer (Norwalk, CT, USA). Proton and 13C nuclear magnetic resonance (1H NMR and 13C NMR) spectra in CDCl3 were obtained with a JEOL EX 270 (270 MHz for 1H and 67.5 MHz for 13C) spectrometer (Tokyo, Japan) using tetramethylsilane (␦ ⫽ 0.00) as an internal standard, and mass (MS) spectra were determined with a JEOL JMS-DX 303 spectrometer. Thin-layer chromatography (TLC) was performed on E. Merck precoated TLC plates, silica gel 60 –254, layer thickness 0.25 mm (Darmstadt, Germany). Silica gel column chromatography was conducted with E. Merck Kieselgel 60 (70 –230 mesh). CD3COOD and (CD3CO)2O were obtained from Acros Or-

0039-128X/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S 0 0 3 9 - 1 2 8 X ( 0 1 ) 0 0 1 0 3 - 9

744

M. Numazawa et al. / Steroids 66 (2001) 743–748

ganics (New Jersey, USA), and CD3COOH was obtained by treatment of (CD3CO)2O with H2O. Androsta-5,16-diene-3␤,17-yl diacetate (1) [2], 3␤,17diacetoxy-5␣-androsta-16-ene (2) [1], and 3,17-diacetoxyestra-1,3,5(10),16-tetraene (4) [11] were synthesized according to the methods previously reported. Pb(OCOCD3)4 was prepared from Pb3O4 according to the known method [12] using CD3COOD and (CD3CO)2O. 2.1. 5␣-Androstan-16-en-17-yl acetate (3) Catalytic solution [0.4 ml of H2SO4:isopropenyl acetate (0.05:1, v/v)] was added to a solution of 5␣-androstan-17one (1 g, 3.6 mmol) in 15 ml of isopropenyl acetate, and the resulting mixture was distilled over a period of 5 h. The solution was then cooled and diluted with ether (100 ml). The ether solution was washed sequentially with 5% NaHCO3 solution and water and then dried with Na2SO4. Evaporation of the solvent yielded a brown solid that was dissolved in 50 ml of a mixture of hexane and ethyl acetate (4:1, v/v) and passed through a column packed with 30 g of silica gel. Evaporation of the solvent from the eluate gave the crude product, which was recrystallized from acetone to give compound 3 as colorless needles; mp 73–75°C (lit. 76 – 80°C [13]). 1H NMR ␦ 0.82 (s, 3H), 0.88 (s, 3H), 2.13 (s, 3H), 5.45 (m, 1H); FT-IR 1620 cm⫺1 (C¢O). 2.2. Reaction of the enol acetates 1– 4 with lead (IV) acetate A solution of 800 mg (2.1–2.5 mmol) of each enol acetate in 15 ml of acetic acid and 5 ml of acetic anhydride was treated with a 1.1 mol equivalent of lead (IV) acetate at room temperature for 10 or 20 h with stirring in the dark. After this time, the reaction mixture was diluted with ethyl acetate (300 ml), washed sequentially with 5% NHCO3 solution and water, and dried with Na2SO4. Evaporation of the solvent gave a brown oil, which was purified by silica gel column chromatography (hexane/ethyl acetate, 25:1, v/v, for isolation of 7 and 10, 10:1, v/v, for isolation of 5 and 8 or 6 and 9, and 4:1, v/v, for isolation of 11 and 12), and recrystallization afforded the 16␤-acetoxy steroids 5–7 and 11 (47–70% yield) and the 16␤-acetoxyacetates 8 –10 and 12 (9 –15% yield), respectively. 2.3. 3␤,16␤-Diacetoxyandrost-5-en-17-one (5) Yield: 62%. mp 167–170°C (lit. 168 –170°C [14]). 1H NMR ␦ 0.98 (s, 3H), 1.06 (s, 3H), 2.02 (s, 3H), 4.61 (m, 1H), 5.00 (t, J ⫽ 8.7 Hz, 1H), 5.41 (d, J ⫽ 5.1 Hz, 1H), FT-IR 1726, 1752 cm⫺1 (C¢O).

4.63– 4.75 (m, 1H), 4.98 (t, J ⫽ 8.6 Hz, 1H); FT-IR 1730, 1750 cm⫺1 (C¢O). 2.5. 16␤-Acetoxy-5␣-androstan-17-one (7) The crude product was recrystallized from MeOH to give compound 7 (450 mg, 54%) as colorless needles; mp 149 – 152°C (lit. 154 –155°C [13]). 1H NMR ␦ 0.81 (s, 3H), 0.95 (s, 3H), 2.11 (s, 3H), 4.98 (t, J ⫽ 8.6 Hz, 1H); FT-IR 1730, 1741 cm⫺1 (C¢O). 2.6. 3␤-Acetoxy-16␤-(acetoxy)acetoxyandrost-5-en-17-one (8) Yield: 10% (96 mg). mp 195–197°C (from acetone); 1H NMR ␦ 0.98 (s, 3H), 1.06 (s, 3H), 2.04 (s, 3H), 2.16 (s, 3H), 4.60 (m, 1H), 4.63 and 4.70 (d, J ⫽ 8.3 Hz, 2H), 5.03 (t, J ⫽ 8.6 Hz, 1H), 5.41 (d, J ⫽ 4.9 Hz, 1H); 13C NMR ␦ 14.20, 19.28, 20.02, 20.36, 21.33, 27.62, 29.22, 30.64, 30.78, 31.52, 36.77, 36.82, 37.99, 46.07, 46.78, 50.19, 60.50, 73.57, 75.20, 121.53, 139.93, 167.21, 170.14, 170.19, 213.31. FT-IR 1752 (C¢O) cm⫺1. Analysis calculated for C25H34O7: C, 67.24; H, 7.67. Found: C, 66.98; H, 7.69. 2.7. 3␤-Acetoxy-16␤-(acetoxy)acetoxy-5␣-androstan-17one (9) Yield: 9% (88 mg). mp 112–114°C (from acetone); 1H NMR ␦ 0.85 (s, 3H), 0.95 (s, 3H), 2.02 (s, 3H), 2.16 (s, 3H), 4.69 (m, 1H), 4.65 and 4.67 (d, J ⫽ 10.1 Hz, 2H), 5.01 (t, J ⫽ 8.6 Hz, 1H). FT-IR 1728, 1749 cm⫺1 (C¢O). Analysis calculated for C25H36O7: C, 66.94; H, 8.09. Found: C, 66.67; H, 8.12. 2.8. 16␤-(Acetoxy)acetoxy-5␣-androstan-17-one (10) Yield: 14% (133 mg). mp 89 –91°C (from MeOH); 1H NMR ␦ 0.81 (s, 3H), 0.95 (s, 3H), 2.17 (s, 3H), 4.62 and 4.70 (d, J ⫽ 16.0 Hz, 2H), 5.01 (t, J ⫽ 8.6 Hz, 1H). FT-IR 1759 cm⫺1 (C¢O). Analysis calculated for C23H34O5: C, 70.73; H, 8.70. Found: C, 70.76; H, 8.96. 2.9. 3,16␤-Diacetoxyestra-1,3,5(10)-trien-17-one (11) Yield: 47%. mp 143–145°C (lit. 148 –149°C [11]); 1H NMR ␦ 1.00 (s, 3H), 2.13 (s, 3H), 2.28 (s, 3H), 2.91 (m, 2H), 5.05 (t, J ⫽ 8.4 Hz, 1H), 6.82 (d, J ⫽ 2.5 Hz, 1H), 6.86 (dd, J ⫽ 2.6 and 8.4 Hz, 1H), 7.28 (d, J ⫽ 6.6 Hz, 1H).

2.4. 3␤,16␤-Diacetoxyandrostan-17-one (6)

2.10. 3-Acetoxy-16␤-(acetoxy)acetoxyestra-1,3,5(10)-trien17-one (12)

Yield: 60%. mp 159 –160°C (lit. 156 –158°C [1]). 1H NMR ␦ 0.86 (s, 3H), 0.95 (s, 3H), 2.02 (s, 3H), 2.11 (s, 3H),

Yield: 15% (152 mg). mp 161–163°C (from MeOH); 1H NMR ␦ 1.00 (s, 3H), 2.17 (s, 3H), 2.29 (s, 3H), 2.91 (m,

M. Numazawa et al. / Steroids 66 (2001) 743–748

745

2H), 4.64 and 4.71 (d, J ⫽ 16.1 Hz, 2H), 5.08 (t, J ⫽ 8.3 Hz, 1H), 6.82 (d, J ⫽ 2.5 Hz, 1H), 6.86 (dd, J ⫽ 2.5 and 8.4 Hz, 1H), 7.28 (d, J ⫽ 7.1 Hz, 1H). FT-IR 1765 cm⫺1 (C¢O). Analysis calculated for C24H28O7: C, 67.27; H, 6.59. Found: C, 67.33; H, 6.60. 2.11. Treatment of the 16␤-acetates 5 and 11 with lead (IV) acetate (A) Compounds 5 and 11 were separately treated with lead (IV) acetate under the conditions described for the reaction of the enol acetate: compounds 5 and 11, 0.13 mmol; acetic acid, 1 ml; acetic anhydride, 0.3 ml; reaction time 20 h. (B) A mixture of compounds 5 and 11 (50 mg each, ca. 0.13 mmol) and the lead reagent (66 mg, 0.15 mmol) in acetic anhydride (3 ml) or benzene (3 ml) was stirred at room temperature for 20 h. After the same work-up as described above, TLC analysis of the product showed that the substrate was quantitatively recovered, and there was no production of the acetoxy acetate derivative. 2.12. Treatment of the 16␤-(acetoxy)acetate 8 with NaOH Compound 8 (50 mg, 0.11 mmol) was dissolved in MeOH (5 ml). 1% NaOH solution in MeOH was added to this solution and the mixture was stirred for 2 h at room temperature. The mixture was subsequently diluted with EtOAc (50 ml), washed sequentially with 5% HCl, 5% NaHCO3 solution, and H2O, and dried with Na2SO4. Evaporation of the solvent gave a crude product, which was recrystallized from acetone to give 3␤,17␤-dihydroxyandrost-5-en-16-one (13) (20 mg, 55%). mp 201–204°C (lit. 202–205°C [2]). 1H NMR ␦ 0.75 (3H, s, 18-Me), 1.05 (3H, s, 19-Me), 3.58 (1H, m, 3␣-H), 3.76 (1H, s, 17␣-H), 5.36 (1H, d, J ⫽ 4.9 Hz, 6-H). This was identical with the authentic sample.

3. Results and discussion Reaction of the enol acetate of 3␤-acetoxyandrost-5-en17-one, compound 1, with lead (IV) acetate in acetic acid containing acetic anhydride was initially carried out at room temperature according to a previously reported method [1]. Column chromatography followed by recrystallization gave 16␤-acetoxy-17-ketone 5 [2] as the major product along with 3␤-acetoxy-16␤-(acetoxy)acetoxyandrost-5-en-17-one (8) as the minor product (Scheme 1). The structural assignment for compound 8 was confirmed by its spectral data and elemental analysis, along with a chemical aspect of this compound. The 1H NMR spectrum showed a singlet signal for acetyl protons (␦: 2.04), two doublet signals for methylene protons (␦: 4.63 and 4.70, J ⫽ 8.3 Hz), and a triplet signal (␦: 5.41) for the 16␣-proton. The 1H decoupled 13C NMR spectrum for compound 8 disclosed the presence of 6

Scheme 1. Reaction of 17-enol acetates with lead (IV) acetate.

sp2 carbons (four carbonyls and two C¢C carbons) in the low field region. Treatment of compound 8 with NaOH in aqueous MeOH gave the 17␤-hydroxy-16-keto derivative 13, the most thermodynamically stable ketol among the four possible 16,17-ketols [4,16,17]. This derivative was produced through isomerization of its 16␤-hydroxy-17-keto isomer, which was the most unstable ketol initially produced (Scheme 2). These NMR and alkaline hydrolysis results along with IR (␯max: 1752 cm⫺1) and mass (M⫹:m/z 388) spectral data supported the assigned structure. The final proof for the structure of compound 8 was obtained from two dimensional sequences (COSY and NEOSY) in NMR spectroscopy. The other enol acetates 2– 4 (5␣-reduced, 3-deoxy, and A-ring aromatic steroids, respectively) were also subjected to reaction with the lead reagent under similar conditions and gave the corresponding 16␤-(acetoxy)acetoxy compounds 9, 10, and 12, respectively, in addition to the corresponding 16␤-acetates 6, 7, and 11, respectively (Scheme 1).

Scheme 2. Alkaline hydrolysis of 16␤-(acetoxy)acetate 8.

746

M. Numazawa et al. / Steroids 66 (2001) 743–748

and/or substrate 3 under various conditions. When CD3COOD was employed as a solvent in the (acetoxy)acetoxylation reaction of non-labeled substrate 3 with either non-labeled Pb(OCOCH3)4 or labeled reagent Pb(OCOCD3)4, deuterium atoms were incorporated efficiently into the 16␤-(acetoxy)acetoxy moiety of compound 10, producing d 3 -species with a CD3COOCH2COO function as well as d 4 -species with a CD3COOCH(D)COO function in a ratio of d 3 to d 4 of 73:19, 60:18, or 61:25 (Scheme 3; Table 1, entry a, c, or d). In contrast, (acetoxy)acetoxylation reaction with deuterated Pb(IV) reagent in the non-labeled solvent (entry b) did not give rise to significant incorporation of deuterium in the (acetoxy)acetate 10. These results show that the acetoxy group in product 10 came from the acetic acid solvent. Moreover, it is likely that this occurred via exchange of an acetoxy anion of the reagent for the solvent acetic acid, although there is no direct evidence for this. Treatment of the non-labeled (acetoxy)acetate 10 with Pb(OCOCH3)4 in CD3COOD and (CD3CO)2O under the same conditions as those described for the treatment of enol acetate 3 failed to label the 16␤-substituent with deuterium (entry g). Moreover, when the deuterium-labeled product 10 (d 3 , 73 atom%; d 4 , 19 atom%) was similarly treated under the above conditions using non-labeled reagent and solvents, there was no significant difference in a deuterium content before and after the treatment. These results reveal that all of the deuterium atoms labeled in the methyl and methylene groups of the 16␤-(acetoxy)acetoxy substituent were incorporated not through enolization of carbonyl functions of the 16␤-substituent of the (acetoxy)acetate initially produced, but through the lead (IV) acetate-assisted reaction sequence. Then, to confirm the origin of the methylene moiety, we synthesized the deuterated enol acetate 3 with a 17OCOCD3 moiety (d 3 , 83 atom%) by treatment of 5␣-an-

Scheme 3. Deuterium-labeling of 16␤-(acetoxy)acetate 10 under various conditions.

Treatment of the 16␤-acetates 5, 6, 7, and 11 with lead (IV) acetate under the same conditions as those employed for the reaction of the enol acetate did not produce the corresponding 16␤-(acetoxy)acetates 9, 10, 11, and 12, respectively, indicating that the (acetoxy)acetates were not produced through acetoxylation of the 16␤-acetates initially formed. When acetic anhydride or benzene was used as a solvent instead of acetic acid for the reaction of the enol acetate 3 with lead (IV) acetate, neither the 16␤-acetate 7 nor the 16␤-(acetoxy)acetate 10 was produced, suggesting that acetic acid is essential as the reaction solvent for the production of both steroids 7 and 10. To further explore the mechanism for the production of the 16␤-(acetoxy)acetate, the (acetoxy)acetoxylation reaction was carried out using deuterated solvents, reagent,

Table 1 Deuterium-labeling of the 16␤-(acetoxy)acetate 10 and the 16␤-acetate 7 under various conditions Deuterium content, atom%a)

Conditions

Entry

a b c d e f g h

Substrate

Non-labeled 3 Non-labeled 3 Non-labeled 3 Non-labeled 3 Non-labeled 3 Labeled-d3 3 Non-labeled 10 Non-labeled 7 a

Reagent

Pb(OCOCD3)4 Pb(OCOCD3)4 Pb(OCOCH3)4 Pb(OCOCH3)4 Pb(OCOCH3)4 Pb(OCOCH3)4 Pb(OCOCH3)4 Pb(OCOCH3)4

16␤-(Acetoxy)acetate 10

Solvent

CD3COOD CH3COOH CD3COOD CD3COOD CD3COOH CH3COOH CD3COOD CD3COOD

and and and and and and and and

(CD3CO)2O (CH3CO)2O (CH3CO)2O (CD3CO)2O (CD3CO)2O (CH3CO)2O (CD3CO)2O (CD3CO)2O

Deuterium content was obtained by mass spectrometry. Deuterium labeling data was obtained by 1H NMR spectrometry.

b

Deuterium labeling of a 16␤-substituent of compound 10, %b) 16␤-Acetate 7

d0

d1

d2

d3

d4

d5

d0

d1

d2

d3

0 95 13 2 6 6 100 —

2 2 5 3 4 24 0 —

3 0 2 4 9 68 0 —

73 3 60 61 79 2 0 —

19 0 18 25 2 0 0 —

3 0 2 5 0 0 0 —

12 97 26 10 12 71 — 100

4 0 3 4 4 2 — 0

2 0 2 0 9 5 — 0

82 3 69 79 76 22 — 0

Methyl

Methylene

⬎98 0 67 95 95 0 0

20 0 20 30 30 83 0

M. Numazawa et al. / Steroids 66 (2001) 743–748

747

Scheme 5. Proposed mechanism for 16␤-acetoxylation of a 17-enol acetate.

Scheme 4. Synthesis and reaction of deuterium-labeled 17-enol acetate 3.

drost-17-one (14) with (CD3CO)2O in the presence of LDA at ⫺78°C (Scheme 4) [17]. Reaction of compound 3-d 3 with Pb(OCOCH3)4 in acetic acid containing acetic anhydride produced the deuterated (acetoxy)acetate 10, which had a CH3COOD2COO function at C-16␤ (d 1 , 24 atom%; d 2 , 68 atom%) (Table 1, entry f). The length of the reaction period (10 or 20 h) did not affect the deuterium content of product 10. Since the methylene hydrogens did not exchange with a hydrogen of the solvent during the reaction and isolation procedures described above, it seems likely that the formation of the d 1 -species occurred in the reaction sequence. The relative amounts of the d 1 - to d 2 -species corresponded well to those of the d 3 - to d 4 - plus d 5 -species (d 3 , 60–73 atm%; d 4 ⫹ d 5 , 20–30 atom%) obtained by reaction of the non-labeled substrate 3 with CD3COOD as solvent under various conditions (Table 1, entries a, c, and d). On the basis of these results, it is concluded that a methylene carbon of the 16␤-(acetoxy)acetoxy group was derived from a methyl carbon of a 17-acetoxy function of the enol acetate. Taken together, the acetoxylation and (acetoxy)acetoxylation sequences are thought to proceed as follows (Schemes 5 and 6). The first oxidation proceeds via an electrophilic attack of Pb(IV) acetate on an olefin produces a cyclic carbocation 15. This reacts with an acetoxy anion to complete the first oxidation, giving a 16␤-acetate 17 through a 16␤-acetoxy-17␣-Pb(OCOCH3)3 intermediate 16 (normal oxidation, Scheme 5). On the other hand, migration of a 17-acetyl group of the carbocation 15 to the 16␤position followed by attack of an acetoxy anion on the

terminal methylene of a cyclic intermediate 18 completes the first oxidation, producing an enol 17-(acetoxy)acetate (19) (alternate oxidation, Scheme 6). The second oxidation is electrophilically driven like the first, since an enol ester 19 was regenerated. Isomerization of a cyclic carbocation 20, produced by reaction of the intermediate 19 with the lead reagent, to a 17␣-lead adduct 21 followed by attack of an acetoxy anion yields an acetal 22 (path a), which is transformed into carbonyl compound 24 by attack of a nucleophile. In contrast, deprotonation of a methylene function of the adduct 21 produces an olefin 23 to which addition of acetic acid also gives the intermediate 22 (path b). This deprotonation-addition process causes the partial loss of the methylene deuteriums (Table 1, entry f) or the partial labeling of the methylene hydrogens with a deuterium (entries a, c, and d). There was no significant production of the d 4 -species of compound 10 in the reaction using CD3COOH instead of CD3COOD (d 3 , 79 atom%, d 4 , 2 atom%, entry e), which is consistent with the involvement of this process in the reaction sequence. The deuterium content of the 16␤-acetate 7, obtained as

Scheme 6. Proposed 16␤-(acetoxy)acetate.

mechanisms

for

the

production

of

748

M. Numazawa et al. / Steroids 66 (2001) 743–748

the major product of the reaction of the enol acetate 3 under various conditions was also analyzed (Table 1). The reaction with either Pb(OCOCH3)4 or Pb(OCOCD3)4 in deuterated acetic acid (CD3COOD or CD3COOH) gave essentially the deuterated compound 7, which had a 16␤OCOCD3 group (d 0 , 10–26 atom%; d 3 , 69–82 atom%, respectively; entries a, c, d, and e). On the other hand, treatment of the deuterated enol acetate 3-d 3 with the nonlabeled reagent and solvents produced principally non-labeled 16␤-acetate 7 (d 0 , 71 atom%; d 3 , 22 atom%, entry f). On the basis of these labeling results, it seems likely that the 16␤-acetoxy moiety of compound 7 principally came from the lead reagent through an ionic attack sequence, as previously reported [4]. In conclusion, reaction of a steroidal 17-enol acetate with Pb(IV) acetate in acetic acid gave a 16␤-(acetoxy)acetoxy17-ketone as the minor product along with a 16␤-acetoxy17-ketone as the major product. The production of the (acetoxy)acetate occurred first through Pb(IV) acetate-catalyzed migration of a 17-acetyl group to the 16␤-position and the acetoxy moiety directly attached to the steroid nucleus originated from a 17-acetyl function of the substrate. A terminal acetoxy moiety of the 16␤-(acetoxy)acetoxy function of the (acetoxy)acetate as well as the 16␤-acetoxy group of the acetate could arise from either the lead reagent or the acetic acid solvent since, presumably, there is some exchange and/or equilibration between them.

References [1] Johnson WS, Gastambide B, Pappo R. The action of lead tetraacetate on an enol acetate. The epiandrosterone acetate, their interconversion and rearrangement. J Am Chem Soc 1957;79:1991– 4.

[2] Kincl FA. An improved synthesis of 16␤-hydroxy dehydroepiandrosterone. J Steroid Biochem 1976;7:419 –20. [3] Numazawa M, Nagaoka M, Mutsumi A. Stereospecific 1,2-hydride shift in the rearrangement of 16␤-hydroxy-17-oxo steroids to 17␤hydroxy-16-ones with acid and base. Chem Pharm Bull 1987;35: 4763– 8. [4] Kirk DN, Hartshorm MP. Steroid Reaction Mechanisms, New York, Elsevier Publishing Company, 1968. pp. 176 –7. [5] Fajkos J. Steroids. XV. Configuration of 16-bromo derivatives of androstane. Collection Czechoslov Chem Communs 1955;20:312– 4. [6] Ellis B, Patel D, Petrow V. Modified steroid hormones. VIII. Some 16-bromo- and 16-chloro derivatives of testosterone. J Chem Soc 1958:800 – 4. [7] Shoppee CW, Jenkins RH, Summers GHR. Steroids and Walden inversion part XL. The configurations of the bromination products of androsta-17-one. J Chem Soc 1958:3048 –58. [8] Glazier ER. Bromination with cupric bromide. II. Bromination in the presence of an olefinic bond. J Org Chem 1962;27:4397–9. [9] Numazawa M, Osawa Y. Improved synthesis of 16␣-hydroxylated androgens: intermediates of estriol formation in pregnancy. Steroids 1978;32:519 –27. [10] Numazawa M, Nagaoka M. Intramolecular 1,2-hydride shift in the rearrangement of steroidal 16␤-hydroxy-17-ones to 17␤-hydroxy-16ones. J Chem Soc, Chem Commun 1982:530 –1. [11] Biggerstaff WR, Gallagher TF. 3,16␤-Dihydroxy-1,3,5-estratrien-17one and related compounds. J Org Chem 1957;22:1220 –2. [12] Fieser LF, Fieser M. Reagents for Organic Synthesis, Vol. 1, New York, John Wiley & Sons 1967. pp. 537– 63. [13] Brutcher FV, Bauer W. The conformations of substituted cyclopentanes. III. Rind D in the steroids. J Am Chem Soc 1962;84:2236 – 41. [14] Aoki T, Yamamura H, Takei K, Mori H. Synthesis of 16-oxygenated androst-5-en-3␤-ol derivatives. Chem Pharm Bull 1964;12:808 –12. [15] Leeds NS, Fukushima DK, Gallagher TF. Studies of steroid ring D epoxides of enol acetates; a new synthesis of enol and of androstane3␤,16␣,17␤-triol. J Am Chem Soc 1954;76:2943– 8. [16] Fishman J. Rearrangements of steroidal ring D ketols. J Am Chem Soc 1960;82:6143–7. [17] Kowalski C, Creary X, Rollin AJ, Burke MC. Reductions of ␣-substituted ketones by lithium diisopropylamide. J Org Chem 1978;43: 2601– 8.