The Stereochemistry of an Elimination Reaction accompanying the Hydroboration of a Steroidal Allylic Alcohol

Article abstract of DOI:10.1039/a708181b

Deuterium labelling studies have shown that the facile elimination of the 5β-hydroxy group observed in the course of hydroboration of a 5β-hydroxyandrost-3-ene may involve a trans diaxial borane-borinate elimination coupled with a syn transfer of hydrogen from the borinate.

Full text of DOI:10.1039/a708181b

View Online / Journal Homepage / Table of Contents for this issue
126 J. CHEM. RESEARCH (S), 1998
J. Chem. Research (S),
1998, 126±127$
The Stereochemistry of an Elimination Reaction
accompanying the Hydroboration of a Steroidal
Allylic Alcohol$
James R. Hanson,* Mansur D. Liman and Cavit Uyanik
School of Chemistry, Physics and Environmental Science, University of Sussex, Brighton,
Sussex BN1 9QJ, UK
Deuterium labelling studies have shown that the facile elimination of the 5ꢀ-hydroxy group observed in the course of
hydroboration of a 5ꢀ-hydroxyandrost-3-ene may involve a trans diaxial borane borinate elimination coupled with a syn
transfer of hydrogen from the borinate.
Derivatives of the allylic hydroxy group of crotyl alcohol
have been shown1 to have a signi®cant e€ect on the regio-
chemistry of hydroboration in directing the addition of a
signi®cant proportion of the borane to the carbon atom
adjacent to the hydroxy group. In cyclohex-2-en-1-ols the
addition is directed to the face anti to the hydroxy group.2
In the more rigid steroid series we have compared3,4 the
magnitude of these e€ects with other directing e€ects of
the steroid skeleton such as the axial b-methyl group at
C-10. In simpler systems such as the allyl alcohol deriva-
tives,1,5 the hydroboration may also be accompanied by
an elimination reaction in which a mono-ol is formed as a
consequence of a borane±borinate elimination reaction and
further hydroboration of the resultant alkene. Both cis
and trans mechanisms [see Scheme 1 (a) and (b)] were
considered5 for the elimination but at the time a distinction
was not made between them. The 5b-hydroxy group of a
5b-hydroxyandrost-3-ene readily undergoes elimination, for
example in the course of attempted methylation with silver
oxide and methyl iodide and silylation with trimethylsilyl
chloride. Products from an elimination reaction also formed
signi®cant components of the hydroboration of 17b-acetoxy-
5b-hydroxyandrost-3-ene (2). A comparison of these with
those obtained from the hydroboration of 17b-acetoxy-5b-
androst-3-ene (1) has enabled us to examine the stereo-
chemistry of the hydroboration and elimination reaction in
this particular situation.
C
B
C
O
(a)
C
C
R
+
B
O R
+
R2O
B
R2O + X
C
B
••
R2O
+
+
B
(b)
C
C
C
C
C
X
X–
OH
C
B
B
H
H
O
O
C
(c)
C
C
C
+
D
C
C
C
C
–
D
D
BH3
BH3
+
+
B
O
H
C
C
C
BH3
D
Scheme 1
were separated by chromatography on silica. The results are
given in Table 1.
The structures of the products were readily established
from the multiplicity of their CH(OH) resonances in their
1H NMR spectra10 and by comparison with literature
data.11 In addition the 4a,17b-dihydroxy-5a-androstane (3)
was readily distinguished from the 4b,17b-dihydroxy-5b-
androstane (4) by an NOE experiment in which the signal
from the 4b-hydrogen (dH 3.45) in 3 was enhanced (6.8%)
by irradiation of the C-10b methyl group. There was no
enhancement of the comparable CH(OH) signal in 4.
Hydroboration of the 17b-acetoxy-5b-androst-3-ene took
place predominantly from the b-face with a small pro-
portion adding from the a-face. The amount of 4b-addition
remained essentially the same between the two alkenes.
However despite the fact that there is an established
propensity2±4 for an allylic hydroxy group to direct the
incoming borane to the anti face, no 4a,5b-diols were
detected from the hydroboration of 17b-acetoxy-5b-
hydroxyandrost-3-ene (2). There was in their place a sub-
stantial amount of material that had arisen by elimination
of the 5b-hydroxy group and rehydroboration of the
resultant 4-ene. This would indicate that it is a 4a-borane
adduct, which possesses a trans diaxial relationship to the
17b-Acetoxy-5b-androst-3-ene (1) was obtained from
testosterone by a Wol€±Kishner reduction followed by
acetylation.6,7 17b-Acetoxy-5b-hydroxyandrost-3-ene (2) was
obtained by a Wharton reaction8,9 with 17b-acetoxy-4b,5b-
epoxyandrostan-3-one. The hydroboration±oxidation was
carried out with alkaline hydrogen peroxide. The products
*To receive any correspondence (e-mail: j.r.hanson@sussex.ac.uk).
$This is a Short Paper as de®ned in the Instructions for Authors,
Section 5.0 [see J. Chem. Research (S), 1998, Issue 1]; there is there-
fore no corresponding material in J. Chem. Research (M).

View Online
J. CHEM. RESEARCH (S), 1998 127
Table 1 Hydroboration of 5b-androst-3-enes
Substrate
Product
Yield (%)
17b-Acetoxy-5b-androst-3-ene (1)
3a,17b-dihydroxy-5b-androstane
3b,17b-dihydroxy-5b-androstane
4a,17b-dihydroxy-5b-androstane
4a,17b-dihydroxy-5b-androstane
17b-hydroxy-5a-androstane
4a,17b-dihydroxy-5a-androstane (3)
4b,17b-dihydroxy-5b-androstane (4)
3a,5b,17b-trihydroxyandrostane
4b,5b,17b-trihydroxyandrostane
11
29
5.5
31.5
3
41
6
22
27
17b-Acetoxy-5b-hydroxyandrost-3-ene (2)
ꢂH (CDCl3) 0.74 (3 H, s, 18-H), 0.81 (3 H, s, 19-H), 3.45 (1 H, dt,
J 4.6 and 10.6 Hz, 4B-H), 3.63 (1 H, t, J 8.6 Hz, 17ꢁ-H); (ii) 3ꢀ,17ꢀ-
dihydroxy-5ꢀ-androstane (541 mg), needles, mp 162±164 8C (lit.,11
165±167 8C), ꢂH (CDCl3) 0.71 (3 H, s, 18-H), 0.96 (3 H, s, 19-H),
3.61 (1 H, t, J 8.6 Hz, 17ꢁ-H), 4.08 (1 H, pent, J 4.3 Hz, 3ꢁ-H);
(iii) 4ꢀ,17ꢀ-dihydroxy-5ꢀ-androstane (4) (580 mg), needles, mp
176±178 8C (lit.,12 177±178 8C), ꢂH (CDCl3) 0.73 (3 H, s, 18-H), 0.99
(3 H, s 19-H), 3.64 (1 H, t, J 8.6 Hz, 17ꢁ-H), 3.87 (1 H, dt, J 5.1
and 10.7 Hz, 4ꢁ-H); and (iv) 3ꢁ,17ꢀ-dihydroxy-5ꢀ-androstane
(204 mg), prisms, mp 236±238 8C (lit.,11 237±238 8C), ꢂH (CDCl3)
0.73 (3 H, s, 18-H), 0.94 (3 H, s, 19-H), 3.64 (2 H, m, 3ꢀ- and
17ꢁ-H).
(b) Under similar conditions 17b-acetoxy-5b-hydroxyandrost-3-
ene (1 g) gave successively (i) 17b-hydroxy-5a-androstane (31 mg),
plates, mp 163±165 8C (lit.,11 164±166 8C), dH (CDCl3) 0.73 (3 H,
s, 18-H), 0.91 (3 H, s, 19-H), 3.64 (1 H, t, J 8.6 Hz, 17a-H);
(ii) 4a,17b-dihydroxy-5a-androstane (3) (364 mg), needles, mp
231±233 8C (lit.,12 235±237 8C), dH (CDCl3) 0.73 (3 H, s, 18-H), 0.81
(3 H, s, 19-H), 3.42 (1 H, dt, J 4.6 and 10.7 Hz), 3.63 (1 H, t,
5b-hydroxy group, that is participating in the elimination
reaction.
The stereochemistry of this process was studied further
by examining the fate of
a deuterium atom at C-3
in the substrate. 3-Deuterio-5b,17b-dihydroxyandrost-3-ene
was prepared by carrying out the Wharton reaction with
deuteriohydrazine. The 1H NMR spectrum of the product
established the presence of deuterium at C-3. The 4-H
resonance at dH 5.53 now appeared as a singlet whilst there
was no signal at dH 5.81 corresponding to H-3. The hydro-
boration and oxidation were repeated. In the 1H NMR
spectra of the resultant deuteriated C-4 alcohols 3 and 4,
the 4-H signal in 3 (dH 3.42) had collapsed from a triplet
(J 10.7 Hz) of doublets (J 4.5 Hz) to a doublet (J 10.7 Hz) of
doublets (J 4.5 Hz), whilst in 4 the 4-H signal (dH 3.87) had
changed from a triplet (J 10.7 Hz) of doublets (J 5.1 Hz) to
a triplet (J 10.7 Hz). In both cases the C-3 deuterium
atom had taken up the C-3a con®guration. A plausible
explanation for these results is given in Scheme 1(c). A bori-
nate ester is formed from the 5b-hydroxy group. The borane
then adds to the anti face at C-4 but undergoes a facile
trans diaxial borane±borinate elimination with the internal
transfer of hydride from the borinate group. Thus the
hydrogen atom which was introduced at C-3 was on the
same side of the molecule as the departing 5b-hydroxy
group. Part of the driving force for this particularly facile
elimination may be the relief of interactions between the
4a-substituent and the a-face of ring B in a 5b-steriod.
The hydration that is observed at C-4 has taken place from
the b-face of the molecule where these steric factors do not
apply. In this particular situation the steric constraints of
the ring system have dominated the anti-directing e€ect of
the hydroxy group.
J
8.6 Hz); (iii) 4b,17b-dihydroxy-5b-androstane (4) (52 mg), mp
178±181 8C, identical with the material described above; (iv)
4b,5b,17b-trihydroxyandrostane (255 mg) prisms, mp 210±212 8C
(Found: C, 73.3; H, 10.4. C19H32O3 requires C, 74.0; H, 10.5%),
1
ꢃ
max/cm 3403, 3375; dH (CDCl3) 0.74 (3 H, s, 18-H), 0.97 (3 H, s,
19-H), 3.64 (1 H, t, J 8.5 Hz, 17a-H), 4.02 (1 H, dd, J 6.5
and 11.2 Hz, 4a-H); (v) 3a,5b,17b-trihydroxyandrostane, (203 mg),
prisms, mp 237±239 8C (lit.,11 237±238 8C), dH (CDCl3) 0.73 (3 H, s,
18-H), 0.96 (3 H, s, 19-H), 3.62 (1 H, t, J 8.6 Hz, 17a-H), 4.14 (1 H,
broad s, 3b-H).
We thank Professor Sir John Cornforth for very helpful
discussions. C. U. wishes to thank Kocaeli University,
Izmit, Turkey, for study leave and for ®nancial assistance.
Received, 13th November 1997; Accepted, 27th November 1997
Paper E/7/08181B
Experimental
References
General experimental details have been described previously.3
Steroids were recrystallized from ethyl acetate±petrol mixtures. 17b-
Acetoxy-5b-androst-3-ene (1), prepared from 17b-hydroxyandrost-4-
en-3-one by reduction with hydrazine hydrate and acetylation had
mp 139±142 8C (lit.,6 138±140 8C) whilst 17b-acetoxy-5b-hydroxy-
androst-3-ene (2) prepared by reduction of 17b-acetoxy-4b,5b-
epoxyandrostan-3-one with hydrazine hydrate had mp 119±121 8C
(lit.,9 118±119 8C). 3-Deuterio-5b,17b-dihydroxyandrost-3-ene had
mp 157±159 8C; dH (CDCl3) 0.75 (3 H, s, 18-H), 1.00 (3 H, s,
19-H), 3.63 (1 H, t, J 8.5 Hz, 17a-H), 5.53 (1 H, s, 4-H).
Hydroboration Experiments.Ð(a) 17ꢀ-Acetoxy-5ꢀ-androst-3-ene
(2 g) in dry tetrahydrofuran (50 cm3) was treated with 1 M borane
in tetrahydrofuran (40 cm3) at 0 8C under nitrogen for 4 h. Water
(20 cm3) was added carefully and the solution was then maintained
at 0 8C whilst 10% aqueous sodium hydroxide (40 cm3) was added
dropwise followed by 30% hydrogen peroxide (50 cm3). The mixture
was left to stir overnight. Sodium sul®te (2 g), acetic acid (1 cm3),
water (50 cm3), dilute hydrochloric acid (50 cm3) and ethyl acetate
(100 cm3) were then added. The stirring was continued for a further
15 min. The organic layer was washed with water and brine and
dried over sodium sulfate. The solvent was evaporated to give a
gum which was chromatographed on silica. Elution with 25% ethyl
acetate±light petroleum gave successively (i) 4ꢁ,17ꢀ-dihydroxy-5ꢀ-
androstane (102 mg), prisms, mp 232±234 8C (lit.,12 235±237 8C),
1 H. C. Brown and R. M. Gallivan, J. Am. Chem. Soc., 1968, 90,
2906.
2 E. Dunkelblum, R. Levene and J. Klein, Tetrahedron, 1972, 28,
1009.
3 J. R. Hanson, P. B. Hitchcock, M. Liman and S. Nagaratnam,
J. Chem. Soc., Perkin Trans. 1, 1995, 2183.
4 M. Alam, J. R. Hanson, M. Liman and S. Nagaratnam,
J. Chem. Res. (S), 1997, 56.
5 H. C. Brown and O. J. Cope, J. Am. Chem. Soc., 1964, 86,
1801.
6 A. Crastes de Paulet and J. Bascoul, Bull. Soc. Chim. Fr., 1966,
939.
7 J. McKenna, J. K. Norymberski and R. D. Stubbs, J. Chem.
Soc., 1959, 2502.
8 P. S. Wharton and D. H. Bohlen, J. Org. Chem., 1961, 26, 3615.
9 M. Gorodetsky, N. Danieli and Y. Mazur, J. Org. Chem., 1967,
32, 760.
10 J. E. Bridgeman, P. C. Cherry, A. S. Clegg, J. M. Evans,
Sir Ewart Jones, A. Kasal, V. Kumar, G. D. Meakins,
Y. Morisawa, E. E. Richards and P. D. Woodgate, J. Chem.
Soc. C, 1970, 250.
11 Dictionary of Steroids, ed. R. A. Hill, D. N. Kirk, H. L. Markin
and G. M. Murphy, Chapman and Hall, London, 1992.
12 D. Marcos and H. Rojas, Acta Cient. Venezolana, 1974, 25, 195.