First total synthesis of xestobergsterol A and active structural analogues of the xestobergsterols

TETRAHEDRON

Pergamon

Tetrahedron 57 (2001) 1449±1481

First total synthesis of xestobergsterol A and active structural

analogues of the xestobergsterols

Michael E. Jung^pand Ted W. Johnson

Department of Chemistry and Biochemistry, University of California at Los Angeles, Los Angeles, CA 90095-1569, USA

Received 14 September 2000; accepted 16 November 2000

AbstractÐThe novel pentacyclic polyhydroxylated sterol, xestobergsterol A 1a, a strong inhibitor of histamine release from rat mast cells

induced by anti-IgE, has been synthesized in 24 steps and good overall yield from stigmasterol 7. The Breslow remote functionalization

process has been extended to several more highly functionalized steroid derivatives, especially those with oxygen functionality in the B-ring.

The key steps of the synthesis of xestobergsterol A 1a and its analogues, 7-deoxyxestobergsterol A 1d and 16,23-seco-23-deoxyxestoberg-

sterol A 73, are the Breslow remote functionalization of oxygenated steroids and for compounds 1a and 1d, a novel base-catalyzed

epimerization-aldol condensation of a dione to give the desired CD-cis ring structure of the xestobergsterols. Thus the known alcohol 75,

prepared from stigmasterol 7, was taken to the tetraacetate 107 which was then converted via a Breslow remote functionalization into the

alkene aldehyde 114 which was transformed in 5 steps to xestobergsterol A 1a. Testing of the synthetic materials showed that the two

analogues, 7-deoxyxestobergsterol A 1b and 16,23-seco-23-deoxyxestobergsterol A 73, are also potent inhibitors of histamine release with

IC₅₀values (IC₅₀500 nM and 750 nM, respectively) being only 1015 times less than that of xestobergsterol A itself (50 nM). q 2001

1. Introduction

releases Ca²¹from the calcium-sequestering compartment,

and diacylglycerol, which activates protein kinase C (PKC).

Consequently, xestobergsterol A also inhibits the initial

rise of Ca²¹and the generation of inositol 1,4,5-trispho-

sphate (InsP₃). Moreover, xestobergsterol A and xestoberg-

sterol C are also mildly cytotoxic against L1210 murine

The xestobergsterols are intriguing natural products posses-

sing a unique pentacyclic steroid skeleton and exhibiting

outstanding biological activity. Umeyama and coworkers

®rst isolated xestobergsterols A 1a and B 1b from the

Okinawan marine sponge, Xestospongia bergquistia

Fromont¹in 1992. Three years later, isolation of xestoberg-

sterols A 1a, B 1b and C 1c from the marine sponge Ircinia

sp. led to a revision of the C23 stereochemistry and subse-

quent conformation of the CD cis ring juncture in xestoberg-

sterols A and B.²Xestobergsterols A and B have

demonstrated strong inhibition of histamine release in rat

mast cells induced by anti-immunoglobulin B with IC₅₀

values of 50 nM and 100 nM, respectively.²These

compounds 3000±5000 times more potent inhibitors of

histamine release than the well-known anti-allergy drug

disodium chromoglycate (IC₅₀262 mM). In the case of

xestobergsterol A, it was recently shown that the mecha-

nism of action proceeds via the strong inhibition of

phosphatidyl inositol phospholipase C (PI-PLC).³PI-PLC

functions as a signal ampli®er in the inositol phospho-

lipid-dependent trans-membrane signal induction of eukary-

otic cells by generating the intracellular secondary

messengers, inositol 1,4,5-triphosphate (InsP₃), which

leukemia cells with IC₅₀values of 4.1 and 4.0 mg mL²¹

,

respectively.

Other naturally occurring steroids possessing a 15-keto, cis

CD ring junction include contignasterol 2,⁴haliclostanone

sulfate 3,⁵and 15-dehydro-14b-anosmagenin 4.⁶Although

15-dehydro-14b-anosmagenin, a steroidal aglycon isolated

from the saponins of the plant Solanum vespitilio, also has

the 14b proton con®guration, there was considerable doubt

whether the 14b con®guration exists in the natural product

or was formed by epimerization during workup. Although

biological activities have not been reported for hali-

clostanone sulfate or 15-dehydro-14b-anosmagenin, con-

tignasterol strongly inhibited histamine release from rat

mast cells induced by anti-IgE in a dose-dependent manner,

similar to xestobergsterol A, with an IC₅₀value of 800 nM.⁴

Keywords: xestobergsterols; antihistamines; remote functionalization; ster-

oids; regioselective Mitsunobu reaction; analogues; biological activity.

Corresponding author. Tel.: 11-310-825-7954; fax: 11-310-206-3722;

e-mail: jung@chem.ucla.edu

p

PII: S0040-4020(00)01086-3

1450

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

In addition to our work on the xestobergsterols,⁸only three

other reports related to their synthesis, all by Krafft and

coworkers,⁹have appeared in the literature. The ®rst two

papers^9a,butilized an intramolecular Pauson±Khand reac-

tion to generate the D and E rings of the xestobergsterol

skeleton. In the third paper,^9cKrafft abandoned the Pauson±

Khand route in favor of ours, namely an intramolecular aldol

condensation to form the additional E ring. To date, no

synthetic work has been reported on compounds 2, 3, or 4.

Since the xestobergsterols are b-hydroxy ketones, it is

likely that the biogenic precursor is a diketone which has

undergone an intramolecular aldol condensation to form the

additional ®ve-membered E ring. Therefore we designed

our synthesis to use this proposed biosynthesis and to

determine if the aldol product 1a would be produced upon

acid or base treatment of diketone 6. Since twelve aldol

isomers are possible from the diketone 6, molecular

mechanics (MM2) strain energy calculations were

performed using Macromodel 5.0 to ensure that the desired

aldol isomer (that found in the natural products) corre-

sponded to the lowest energy isomer. Of the twelve isomers,

eight are bridged and have much higher strain energies than

the four fused isomers shown. These calculations clearly

demonstrated that the desired aldol isomer was the most

stable by at least 3 kcal mol²¹, which suggested that the

desired isomer should be formed to the exclusion of the

remaining eleven under thermodynamic aldol condensation

conditions (Scheme 1).

It has also been shown to be potent in both in vivo and in

vitro models of allergen-induced bronchoconstriction and

airway smooth muscle contraction. Consequently, it is

very effective as an anti-asthma agent.⁷Interestingly, the

contignasterol reduction product 5 did not inhibit histamine

release from rat mast cells (IC₅₀approx. 300 mM),^7b

suggesting that either the 15-keto group and/or the hemi-

acetal is necessary for the inhibition of histamine release.

The authors represent the reduction product as having the

trans C/D ring junction, yet it is doubtful that the reduction

conditions would cause epimerization of the 14b hydrogen.

Furthermore, although one isomer was puri®ed and tested,

the stereochemistry at C15 was not reported.

Herein we describe in detail our successful synthesis of

xestobergsterol A, la.^8b

2. Background

2.1. Retrosynthesis

We believed that the xestobergsterols could be produced by

an intramolecular aldol condensation of the diketone 6 to

Scheme 1.

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

1451

Scheme 2.

form the additional E ring with the desired stereochemistry

(Scheme 2). Although stigmasterol 7 lacked the 15-keto

group needed for the key aldol reaction, it nevertheless

seemed to be the most appropriate choice for a starting

material since the functionality present in the A and B

rings of stigmasterol could be used to functionalize both

the B and D rings. More importantly, the side-chain alkene

of stigmasterol would allow preparation of the desired 23-

ketocholesterol side chain via ozonolysis and extension of

the resulting aldehyde. We hoped to install the 15-keto

group in ring D by a remote functionalization process,

using either the well-known Barton reaction¹⁰or the

Breslow photooxidation process.¹¹

radical species is located nearer to the desired hydrogen

atom. Mechanistic studies^11bshowed that photolysis of the

benzophenone ester generates a diradical, the oxygen radi-

cal of which then abstracts the tertiary hydrogen atom at

C14 to provide the tertiary radical at C14. Elimination to

provide the C14 alkene occurs exclusively by stereospeci®c

intramolecular abstraction of the 15a-deuterium.

Although there are other remote functionalization tech-

niques available,^12,13the one using 3a-linked 4-benzo-

phenone acetate best suited our synthetic interests in terms

of good yield, excellent selectivity, ease of preparation and

tolerance of other functional groups. Conversion of the

alkene to the desired ketone at C15 would presumably be

straightforward. Hydrolysis of the ester and protection of the

alcohol would give the protected 14,15-ole®n which on

hydroboration±oxidation followed by oxidation should

yield the trans CD 15-ketosteroid. Epimerization under

basic conditions would then provide the desired cis CD

15-ketosteroid in very few steps from the photolysis pre-

cursor. Accordingly, we decided to investigate both the

Barton and Breslow functionalization of steroids as a way

of introducing the 15-keto functionality necessary for the

xestobergsterols.

Barton and coworkers developed a very useful technique for

functionalizing unfunctionalized carbon atoms usually, but

not exclusively, in steroids.^10a±cIn addition to the well-

known oxidation of the C19 methyl group by photolysis

of the C6b nitrite ester, the analogous conversion of an

11a-nitrite ester (derived from the alcohol) via photolysis

and hydrolysis provided the Cl ketone in 50% yield.^10bIn

order to produce the desired 15-keto functionality in the

natural products, photolysis of the 7b-nitrite ester might

functionalize C15 of the steroid via an identical mechanism

to give the 15-oxime. Acidic hydrolysis of the oxime would

provide the 15-ketone, which could possibly be epimerized

to the desired cis CD 15-ketosteroid.

3. Results and discussion

3.1. The Barton reaction

The Breslow remote functionalization technique¹¹also

allows selective oxidation of C15 in a steroidal skeleton

whereby a tertiary hydrogen atom is abstracted by a radical

species which is attached usually at C3a of the steroid back-

bone. Regioselective hydrogen atom abstraction is achieved

by choosing a tether of the appropriate length so that the

Since the xestobergsterols have a 15-keto group, we

believed that the Barton oxidation would be ideal for the

oxidation of C15 using the 7b-hydroxyl group, which is also

present in the natural products. We decided to test the validity

Scheme 3. a) HCI (cat), MeOH, re¯ux, 86%; b) TBSCI, imidazole, DMF, 82%; c) PCC, CH₂CI₂, 98%; d) KOH, MeOH:water, re¯ux, 93%; e) Li, NH₃, 85%

(3:1, 7b:7a); f) CH₂N₂, ether, 98%; g) NOCI, pyridine, 97%; h) h

p

.

1452

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

Figure 1.

of this approach by photolysis of the nitrite ester of the

alcohol 13. Commercially available cholic acid 8 (Scheme

3) was converted into the known diol carboxylic acid 9a in

®ve steps and 35% overall yield.¹⁴The methyl ester 9b,

prepared in 86% yield, was selectively protected to give

the mono silyl ether 10 in 82% yield. Oxidation of the

remaining 7a-alcohol of 10 with pyridinium chloro-

chromate (PCC) afforded the 7-ketone 11a in 98% yield.

Several methods for the reduction of the C7-ketone to the

required 7b-alcohol were attempted. Treatment of the

ketone 11a with sodium borohydride or under Meerwein±

Pondorf±Verley conditions¹⁵gave the 7a-alcohol 10 exclu-

sively, in excellent yields, while attempted Mitsunobu

inversion of the alcohol 10 provided only starting material.

Zhou and coworkers, however, reported that dissolving

metal reduction of the 7-keto functionality of cholic acid

derivatives gave mainly the 7b-alcohol.¹⁶Hydrolysis of the

ester 11a produced in 93% yield the carboxylic acid 11b,

which was treated with lithium in ammonia to give an 85%

yield of a 3:1 mixture in which the 7b-alcohol 12 predomi-

nated over the 7a-alcohol. Re-esteri®cation with diazo-

methane gave the methyl ester 13, which was treated with

nitrosyl chloride in pyridine to give the nitrite ester 14 in

93% overall yield. The nitrite ester was surprisingly stable,

even to silica gel chromatography. Unfortunately, attempts

to photolyze the nitrite ester 14 provided only the unfunc-

tionalized 7b-alcohol 13. Although photochemical cleavage

of the OZNO bond occurred, no intramolecular hydrogen

atom abstraction from C15 was observed. This approach to

xestobergsterol A using the Barton reaction was therefore

abandoned.

Further analysis of the steroid skeleton and bond distances

using MM2 calculations provided a possible explanation for

this ineffectual methodology. Comparison of the successful

Barton reactions¹⁰and our unsuccessful reaction showed

signi®cant differences in the calculated atomic distances

between the oxygen radical and the hydrogen atom to be

abstracted (Fig. 1). For example, the furthest successful

distance is 2.40 A, which is the distance between the

equatorial 11b-oxygen atom and the equatorial 1b-hydro-

gen atom. Since the distance between the 7b-oxygen atom

Ê

and the 15a-hydrogen atom is 2.68 A, presumably the

Ê

additional 0.28 A required for abstraction of the 15a-hydro-

Ê

gen atom by the oxygen radical is too great for the reaction

to occur.

3.2. The Breslow remote functionalization: model studies

and analogue preparation

Although the Barton approach for C15 functionalization

was unsuccessful, the Breslow remote functionalization

technique seemed promising for making the desired

D-ring functionalized steroid. We ®rst wanted to show

that the C14 ole®n produced by the remote functionalization

could be transformed into the desired 14b-H 15-ketosteroid.

Treatment of cholestanol 15 under Mitsunobu conditions

with the known acid 16¹⁷provided the inverted ester 17

with the required 3a stereochemistry in 85% yield

(Scheme 4). Photolysis of 17 using a 450 W mercury arc

lamp with a pyrex ®lter and subsequent cleavage of the

ester with base followed by tert-butyldimethylsilyl protec-

tion provided an approximately 2:1 mixture of the desired

Scheme 4. a) hn, PhH; 10% KOH, THF:ethanol; b) TBSOTf, 2,6-lutidine, CH₂Cl₂, 75% (three steps); c) BH₃-THF, THF, 08C; NaOH, H₂O₂, 17% (19), 3%

(20); d) PCC, CH₂Cl₂, 100%; e) 10% HCI, EtOH, 100%.

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

1453

methyl group and the side chain. Hence, the bigger the

side chain, the more unstable the compounds with cis CD

ring junctions. Both the carbomethoxy group in 23 and the

pyrone in 24 are ¯at and apparently can twist out of the way

of the C18 methyl group in the cis CD isomers.

Molecular mechanics (MM2) strain energy calculations of

model systems show that simple hydrindanone derivatives

favor the cis CD ring junction by approximately 3 kcal

mol²¹(Fig. 3). However, substitution of an isopropyl

group at C17 to represent a simpli®ed cholesterol side

chain, reverses the stabilities of the trans and cis CD ring

junctions resulting in minimal difference between them.

These results suggest that for simple cholesterol derivatives,

the trans CD ring junction should be favored.

Figure 2.

cholestan-14-en-3a-ol 18a and the saturated cholestan-3a-

ol 18b in 75% combined yield. Hydroboration±oxidation of

the mixture led to an approximately 10:1 mixture of the

14a-H 15a-alcohol 19 and the 14b-H 15b-alcohol 20,

which was separated by ¯ash column chromatography in

poor yield. The unreactive TBS-protected cholestan-3a-ol

18b could also be separated at this stage. Based on the steric

hindrance of the approach from the b face of the steroid, the

hydroboration±oxidation provided not surprisingly the

undesired 14a-H 15a-alcohol 19 as the major isomer. It is

unclear why the yield of this hydroboration±oxidation is so

low, since thin layer chromatography showed a very clean

reaction, and other hydroborations on C14 alkenes have

proceeded in good yield in our laboratories (see later

schemes). Oxidation of the 14a-H 15a-alcohol 19 provided

the 14a-H 15-ketone (trans CD ring junction) 21 in excel-

lent yield. Interestingly, treatment of the ketone with base

did not provide the desired 14b-H 15-ketosteroid, but

instead gave only starting material. Acidic hydrolysis of

the tert-butyldimethylsilyl ether 21 afforded the pure

14a-H 15-keto-3a-alcohol (trans CD ring junction) 22 in

quantitative yield.

Synthesis of the 14b-H 15-ketosteroid (cis CD ring

junction) consequently proved to be more dif®cult than

anticipated, since simple equilibration could not be used

as originally planned. Our next approach involved the

Lewis acid mediated rearrangement of a 14a,15a-epoxide

to give the 14b-H 15-ketone directly.²⁰Morisaki and

coworkers have shown that steroidal 14a,15a-epoxides do

not give the expected 14b-H 15-ketone directly (Scheme 5).

They reported that treatment of the epoxide 25 with boron

tri¯uoride etherate in benzene at room temperature provided

the homoallylic alcohol 26 in 75% yield, presumably via the

intermediate I.^20bReichstein and coworkers, however,

showed that regioselective and stereoselective reductive

opening of steroidal 14a,15a-epoxides could be achieved

using hydrogen over platinum catalyst in glacial acetic

acid.²¹Reduction of the a-epoxide 27 under these condi-

tions gave a mixture of products (Scheme 6). The desired

14b-H 15a-alcohol 28a was isolated in 33% yield along

with the fully reduced product 28b in 12% yield, the isomer-

ization product 28c in 7% yield, and starting material in 4%

yield. Although the yields of the desired 14b-H 15a-alcohol

28a were low, we nevertheless decided to pursue this route

since it was precedented to give the product with the desired

stereochemistry.

Two other 15-ketosteroids have also been prepared by other

groups and their CD cis:trans ratios determined. Unlike our

simple 15-ketosteroid 22 (Scheme 4), compounds 23¹⁸and

24¹⁹both favor the cis CD ring junction (Fig. 2). These

results imply that the side chain at C17 is one important

factor in determining the CD cis:trans ratio. In the cis CD

compounds, there is a steric interaction between the C18

Figure 3.

Scheme 5.

1454

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

Scheme 6.

Scheme 7. a) Ac₂O, pyr, 97%; b) mCPBA, CHCl₃, 100% c) H₂, PtO₂, AcOH, 42%; d) PCC, CH₂Cl₂, 100%; e) KOH.

Protection of the mixture of alcohols 29a and 29b (Scheme

7) gave the acetate esters in 97% yield. These were

subjected to epoxidation using m-CPBA to provide in

quantitative yield a 10:1 mixture of a:b epoxidation

products, which were separated by ¯ash column chroma-

tography. The unreactive saturated steroid was also sepa-

rated at this stage. Reduction of the a-epoxide 30 under

Reichstein's conditions, namely, hydrogenation over

platinum oxide in acetic acid, provided the 14b-H

15a-alcohol 31 in 50% yield. Oxidation of the alcohol 31

with PCC gave the desired 14b-H 15-ketosteroid (cis CD

ring junction) 32 in quantitative yield. Interestingly, treat-

ment of the ketone 32 with base led to acetate cleavage, but

more importantly, epimerization of the 14b hydrogen

occurred in the ketone 33 to give an approximately 10:1

ratio of the trans CD ketone 22 to the cis CD steroid 33.

Thus our results are in agreement with the calculations

(Fig. 3).

It should be noted that the stereochemistry of the 14-hydro-

gen (and thus the CD ring junction) is easily established by

the chemical shifts and coupling constants of the

7b-protons. In the trans CD 15-ketone 22, the 7b-proton

resonates at low ®eld (d 2.64, dddd, J13.1, 3.2, 3.2,

3.2 Hz), whereas in the cis CD 15-ketone 33, the 7a-proton

resonates at low ®eld (d 2.51, dddd, J13.2, 13.2, 13.2,

4.6 Hz). These low-®eld resonances are due to the deshield-

ing effect of the 15-ketone which has a pseudo 1,3-diaxial

interaction with the 7a- or 7b-proton depending on the C14

stereochemistry. The coupling pattern indicates whether the

affected proton is equatorial or axial and this, in turn,

determines the stereochemistry of the CD ring junction.

The CD stereochemistry can also be corroborated by the

chemical shift of the C18 methyl group. [Analysis of a

variety of 15-ketosteroids from the work of Suginome (see

Ref. 13) and also our own work (see Experimental section)

showed that the C18 methyl in the cis CD 15-ketosteroids is

Scheme 8.

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

1455

1

further down®eld in the H NMR than in the trans CD

exercised. Finally, protection of the free alcohol at C6

as its tert-butyldimethylsilyl ether provided the desired

photolysis substrate 37 in 92% yield. Photolysis of 37 afforded

the desired ole®n 38 in only 30% yield, along with the two

diastereomeric lactones 39a (20%) and 39b (20%).^23,24

15-ketosteroids.] The compounds with the trans CD ring

junctions have a resonance for the C18 methyl singlet at

0.70±1.00 ppm, whereas those with the cis CD ring junc-

tions have a resonance for the C18 methyl singlet at 1.10±

1.30 ppm. For example, in the trans CD 15-ketone 22, C18

resonates at 0.78 ppm, while in the cis CD 15-ketone 33, it

resonates at 1.16 ppm. We have used both methods to

assign unambiguously the stereochemistry of the CD ring

junctions.

These undesired photocyclization products could easily be

avoided by protection of the C6 alcohol as its acetate

ester prior to photolysis. Hydroboration±oxidation of the

cholesteryl benzyl ether 40²⁵provided the alcohol 41 in

good yield (Scheme 9). Protection of the free alcohol with

acetic anhydride gave the ester 42 in 84% yield, which was

deprotected at C3 using hydrogen over palladium on carbon

catalyst to give 43 in 92% yield. A Mitsunobu reaction was

used once again to install the 3a-linked benzophenone

tether, which afforded the ester 44 in 86% yield. Photolysis

of the steroidal benzophenone 44 followed by complete

ester hydrolysis gave an approximately 2:1 mixture of the

desired ole®nic diol 45a and the saturated diol 45b, which

were inseparable by conventional ¯ash column chroma-

tography conditions. Protection of the mixture of diols

with acetic anhydride proceeded in 68% yield to give a

mixture of the unsaturated steroid 46a and the saturated

steroid 46b. Epoxidation gave the desired a-epoxide 47 in

91% yield (approximately 10:1 ratio of a:b epoxides),

which was separable from the saturated diacetate 46b and

the b-epoxide. Reichstein's conditions were again used to

reductively open the epoxide 47 regio- and stereoselectively

to provide the 14b-H 15a-alcohol 48 in 44% yield. This

alcohol was oxidized in 84% yield to the desired 14b-H

15-ketosteroid 49. Thus the presence of a 6a-alcohol func-

tionality, if properly protected, did not interfere with the

remote functionalization process, and the 14b-H 15-ketos-

teroid 49 could be prepared from the cholesterol derivative

40 in a straightforward manner.

We next determined if substitution at C6 or C7 would have

any deleterious effects on the Breslow remote functionali-

zation. The diol 35 was prepared in 80% yield by the

hydroboration±oxidation of the commercially available

cholesteryl acetate 34 (Scheme 8). A regioselective

Mitsunobu reaction²²with the previously described

carboxylic acid 16 provided the ester 36 in 60% yield.

Even though both additions are axial, inversion proceeded

much faster at C3 than at C6 due to severe steric hindrance

encountered in the inversion of the 6a-alcohol, e.g. the

1,3-diaxial interaction of the carboxylate nucleophile with

the C18 methyl group, as shown in the intermediate II.

Addition of more than one equivalent of the carboxylic

acid 16 in this regioselective Mitsunobu reaction results in

the inversion of the C6 alcohol as well, so care must be

Scheme 9. a) BH₃-THF, THF, 08C; NaOH, H₂O₂, 89%; b) Ac₂O, pyr, 84%; c) H₂, Pd/C, THF:EtOH, 92%; d) 16, PPh₃, DEAD, THF, 86%; e) hn, PhH; f) 10%

KOH, THF:EtOH:water (1:1:1); g) Ac₂O, pyr, 68% (3 steps, combined yield); h) mCPBA, CHCl₃, 91%; i) H₂, PtO₂, AcOH, 44%; j) PCC, CH₂Cl₂, 84%.

1456

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

Scheme 10. a) tBuOOH, RuCl₃-H₂O, C₆H₁₂:water, 63%; b NaBH₄, CeCl₃-7H₂O, CH₂Cl₂:MeOH (4:1), 2788C to rt, 93%; c) TBSCl, imidazole, DMF, 98%; d)

BH₃-THF, 08C; 10% NaOH, 30% H₂O₂.

In order to ensure that the photolysis methodology could be

used to synthesize xestobergsterol A, we attempted the

photolysis with a substrate having oxygen substitution at

both C6 and C7, thereby more closely resembling the

natural products. After several attempts using catalysts

such as copper,^26achromium,^26b±druthenium,^26eand other

methods,^26fwe found that the ruthenium-catalyzed allylic

oxidation was the most desirable one for the C7 oxidation of

protected cholesterol derivatives. Cholesteryl acetate 34

was treated with catalytic ruthenium trichloride and tert-

butyl hydroperoxide to provide the enone 50 in 63% yield

(Scheme 10). Luche reduction of the enone 50 gave reduc-

tion exclusively from the a face to afford in 93% yield the

7b-allylic alcohol 51 which was protected with TBSCl to

give the silyl ether 52 in 98% yield. Unfortunately, hydro-

boration±oxidation of the allylic silyl ether 52 gave not only

the desired alcohol 53 in 41% yield but also the product 54

in 39% yield in which the silyl group had migrated from the

7b-oxygen to the newly formed 6a-oxygen. Although the

1,2-diol is trans, the silyl group is presumably able to

migrate via the ®ve-membered transition state III to provide

the silyl-migrated product 54.²⁷Separation of the two

compounds and teatment of the C7 silyl ether 53 under

basic conditions gave the more stable C6 silyl ether 54 in

quantitative yield. The driving force for the silyl migration

is almost certainly alleviation of the severe pseudo

1,3-diaxial interaction between the C7 equatorial silyl

ether and the C14±C15 carbon±carbon bond in the D ring

of the steroid.²⁸[Friesen and coworkers have reported a very

similar migration of a silyl group from a more hindered

to a less hindered equatorial oxygen. Treatment of the ole®n

Friesen and co-workers avoided this undesirable migration

by oxidative workup with pH 7 phosphate buffer in place of

the sodium hydroxide solution.]

We decided instead to protect the C7 alcohol as the acetate

which would be cleaved during the hydroboration±oxi-

dation process. Protection of the allylic alcohol 51 went

smoothly to provide the allylic acetate 55 in 93% yield

(Scheme 11). Hydroboration±oxidation of the ole®n gave

the triol 56 with a small amount of the triol derived from b

addition of the borane, which could easily be separated by

¯ash column chromatography. Attempts at protecting the

triol 56 as the triacetate using acetic anhydride/DMAP

resulted in slow reaction times and poor yields. However,

treatment of the triol 56 with acetic anhydride catalyzed by

trimethylsilyl tri¯ate gave the fully protected steroid 57 in

79% yield from 55. We found that the acetate protecting

groups at C6 and C7 were suf®ciently hindered relative to

the one at C3 so that treatment of the triacetate 57 with mild

base gave excellent yields of the mono-deprotected product

(e.g. treatment with 10% KOH in 1:1 THF:95% ethanol

gave the alcohol 58 in 92% yield). Once again, the

Mitsunobu inversion of the C3 alcohol 58 with the

carboxylic acid 16 proceeded smoothly to provide

the inverted ester 59 in 87% yield. Photolysis of the ester

59 also produced the remote functionalized steroid 60 in fair

yield. Flash column chromatography prior to hydrolysis

allowed for the facile separation of the unsaturated and

the saturated steroids and provided the pure ole®n. The

ole®nic steroid 60 was treated with lithium aluminum

hydride to afford the ole®nic triol 61 in 58% yield from

58. In addition, the reduced benzophenone tether 62 was

also isolated.

i under standard hydroboration±oxidation conditions

provided the silyl migrated product ii in good yield.

In order to determine the extent of asymmetric induction

in forming the benzhydryl alcohol 62 during photolytic

reduction of the benzophenone, we selectively protected

the diol 62 as the pivaloate ester to give the alcohol 63 in

74% yield. Analysis for enantiomeric purity using the

method of Alexakis²⁹using ³¹P NMR revealed an interesting

4:1 mixture of diastereomers (60% de during the photo-

lytic process). We have not yet determined the absolute

con®guration of the major enantiomer. Breslow also

Presumably, migration occurs to give the more stable

isomer, which lacks the severe steric interaction between

the bulky silyl ether at C4 and the bulky silyl ether at C3.

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

1457

Scheme 11. Ac₂O, DMAP (cat), pyridine, 93%; b) BH₃-THF; 10% NaOH, 30% H₂O₂; c) Ac₂O, TMSOTf (cat), 79% (two steps); d) 10% KOH, THF:95%

EtOH (1:1), 92%; e) 16, PPh₃, DEAD, THF, 87%; f) hn, PhH; g) LiAlH₄, THF, 58% (two steps); h) PvCl (1.5 eq), pyr, rt, 74%.

looked at asymmetric induction during the reduction of the

benzophenone carbonyl with the hydrogens from the steroid

backbone and found very little asymmetric induction

(approximately 55:45 ratio of diastereomers) with simple

steroidal benzophenone esters derived from cholestanol.^11b

In the same paper, Breslow examined asymmetric induction

during reduction of the benzophenone carbonyl and found

that varying solvents had only slight effects on the amount

of asymmetric induction. Thus, the presence of the 6a- and

7b-acetates greatly improves asymmetric induction during

reduction.

dimethylsilyl chloride, which is known to protect hindered

or tertiary alcohols, failed to give the tris-protected

products.³⁰Because the C7 equatorial alcohol is pseudo

1,3-diaxial to the C14±C15 carbon±carbon double bond,

it is therefore much more hindered than one might assume.

Moreover, protection of the C6 oxygen with a bulky group

increases the steric hindrance around the C7 alcohol, thus

making it much more dif®cult to protect. Hence a protecting

group small enough to protect all three alcohols and one that

could be easily removed was required. We tried to fully

protect the ole®nic triol 61 by treating it with acetic

anhydride and catalytic trimethylsilyl tri¯ate, which had

been used successfully in our labs to protect hindered

alcohols. The desired ole®nic triacetate 64a and tribenzoate

64b were produced in 93% and 91% yield, respectively

(Scheme 12). Since we did not want to cleave the ester

protecting groups upon hydroboration±oxidation of the

ole®n, we used very mild conditions for the oxidation

of the intermediate organoborane. The ole®nic triacetate

Our ®rst attempts at protecting the ole®nic triol 61 were

unsuccessful. Treatment of the triol with either tert-butyl-

dimethylsilyl tri¯ate or triethylsilyl tri¯ate led to the

formation of the 3,6-bis-protected products in moderate

yields but, even at elevated temperatures, none of the

desired tris-protected products was detected. Likewise,

treatment of the triol with 18-Crown-6/KH/tert-butyl-

Scheme 12.

1458

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

Scheme 13. a) CH₂(OME)₂, P₂O₅, 63%; b) BH₃-THF; NaOH, H₂O₂; c) PCC, CH₂Cl₂, 80% d) 10% NaOH, THF/EtOH, .95%; e) 10% HCl, THF, 50%.

64a was hydroborated and oxidized with pyridinium

chlorochromate to give the 7,15-dione 65, rather than the

desired 15-ketosteroid. Similarly, the benzoate 64b was

hydroborated and oxidized with sodium perborate,³¹but

again loss of the protecting group at C7 provided the

7b,15a-diol 66, rather than the desired 15a-alcohol. This

loss is presumably facilitated by the intramolecular chela-

tion of the oxygen atom at C7 to the Lewis acidic boron

atom forming the zwitterionic six-membered intermediate

IV. Selective oxidation of the 7b,15a-diol 66 using pyridi-

nium chlorochromate provided the C7 ketone 67 exclu-

sively rather than the desired C15 ketone. It is unclear

why the C7 alcohol is oxidized faster than the C15 alcohol

although it might be assumed that this is due to a combina-

tion of steric and electronic effects.

to the more stable 14b-H 15-ketosteroid alleviates the

severe steric interaction between 7b oxygen and both the

C14±C15 carbon±carbon bond and especially the carbonyl

group at C15. Finally, deprotection under acidic conditions

provided the cis CD keto-triol, 16,23-seco-23-deoxyxesto-

bergsterol A, 73 in 50% yield.

Molecular mechanics (MM2) strain energy calculations on

model structures are in agreement with our experimental

results. As shown in Fig. 4, calculations show that the cis

CD ketone 74a is more stable than the trans CD ketone 74b

by approximately 1.60 kcal mol²¹

.

Presumably, the

7b-methoxy substituent has a strong unfavorable steric

interaction with the C14±C15 carbon±carbon bond and

the C15-keto group. This severe steric interaction can be

alleviated in the cis CD isomer, because the C14±C15

carbon±carbon bond is no longer pseudo 1,3-diaxial with

the C7±oxygen bond and the C15-keto group interacts

with the 7a-hydrogen rather than the bulky methoxy sub-

stituent.

The pure ole®nic triol 61 was successfully protected using

methylal and phosphorous pentoxide³²to give the fully

protected ether 68 in 63% yield (Scheme 13). Surprisingly,

instead of protecting the three alcohols as methoxymethyl

(MOM) ethers, these conditions gave the MOM ether at C3

and the methylene acetal between the C6 and C7 alcohols.

Although the C6 and C7 acetonide was found to be highly

unstable toward mildly acidic conditions, the methylene

acetal in contrast was very robust toward mildly acidic

and strongly basic conditions. Hydroboration±oxidation of

the protected ole®n 68 gave an approximately 6:1 mixture of

the 14a-H 15a-alcohol 69 (75% yield) and the 14b-H

15b-alcohol 70 (13% yield), which were separable by

¯ash column chromatography. The 14a-H 15a-alcohol 69

was oxidized³³to the corresponding ketone 71, which upon

treatment with base, epimerized completely to give the more

stable 14b-H 15-ketosteroid 72. Presumably, epimerization

3.3. Synthesis of 23-ketocholesterol: a model for side-

chain formation

With the cholesterol side-chain analogues of the xestoberg-

sterols fully developed and a well-tested method in place for

their synthesis, we set about synthesizing the xestoberg-

sterol skeleton. It only remained for us to ®nd a starting

steroid that not only had functionality in the A and B

rings of the steroid, but also on the side chain. The inex-

pensive steroid stigmasterol 7 proved to be the best choice.

It was converted into the known alcohol 75 in ®ve steps and

66% overall yield by the following sequence (Scheme 14).³⁴

Protection of stigmasterol 7 with mesyl chloride gave the

mesylate which on re¯uxing with methanol and pyridine

provided the cyclopropyl carbinyl ether via a stabilized

homoallylic cation. Ozonolysis of the side-chain ole®n

and Wittig ole®nation of the resulting aldehyde, with ®nal

hydroboration±oxidation gave the alcohol 75. Oxidation of

the alcohol 75 with pyridinium chlorochromate proceeded

smoothly to afford in 88% yield the aldehyde 75, which was

treated with isobutylmagnesium bromide to give an equal

Figure 4.

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

1459

Scheme 14. MsCl, pyr, rt, 99% b) MeOH, pyr, re¯ux, 89%; c) O₃; DMS, 100%; d) CH₂PPh₃, THF, 93%; e) BH₃-THF; NaOH, H₂O₂, 81%; f) PCC, KOAc,

DCM, 88% g) iBuMgBr, ether, 79%; h) PCC, pyridine:DCM (2%), 100% i) TsOH (cat), dioxane:water (4:1), re¯ux, 94%.

mixture of diastereomeric alcohols in 79% yield. At this

point, all of the carbons for the xestobergsterols were in

place. Oxidation of the mixture of alcohols provided the

23-ketone 77 in quantitative yield. Acid-catalyzed hydroly-

sis of the cyclopropylmethyl methyl ether gave the homo-

allylic alcohol 78 in 94% yield. Unfortunately, all attempts

to protect the C23 carbonyl proved unsuccessful.³⁵

Therefore, instead of fully developing the side chain at the

beginning of the synthesis, we decided to postpone

the formation of the fully oxidized side chain until near

the end of the synthesis in order to avoid protection of the

23-ketone.

3.4. Synthesis of 7-deoxyxestobergsterol A

Having developed a general method for the synthesis of the

14b-H 15-ketone and a good substrate for the side-chain

elaboration identi®ed, we decided to continue with the

synthesis of xestobergsterol A. The alcohol 75, derived in

®ve steps and 66% overall yield from stigmasterol, was

Scheme 15. a) PvCl, pyr, 92%; b) TsOH, water:dioxane, re¯ux, 97%; c) NaH, BnBr, TBAl (cat), THF, 83%; d) BH₃-THF; 10% NaOH, 30% H₂O₂, 60%; e)

Ac₂O, pyr, 100%; f) H₂, Pt, THF/EtOH, 96%; g) 16, PPh₃, DEAD, 90%; h) hn, PhH; i) KOH (aq), THF/EtOH; j) TBSOTf, 2,6-lutidine, CH₂Cl₂, 48% (3 steps);

k) AcOH, THF/H₂O, 86%; l) PCC, CH₂Cl₂, 83%; m) i-BuMgBr, ether, 08C, 90%; n) BH₃-THF; 10% NaOH, 30% H₂O₂; o) Pcc, CH₂Cl₂, 79% (two steps); p)

10% HCl, THF, 48 hr, .95%; q) 10% NaOH, EtOH, .95%.

1460

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

Scheme 16. AcOH, re¯ux, 94%; b) RuCl₃, TBHP, C₆H₁₂, 65%; c) NaBH₄, CeCl₃, DCM:MeOH, 97%; d) Ac₂O, DMAP, pyr, 98%; e) BH₃-THF; NaOH, H₂O₂,

57%; f) Ac₂O, TMSOTf, 95%; g) 10% NaOH, THF:EtOH, 89%; h) 16, PPh₃, DEAD, THF, 93%; i) hn, PhH, 61%; j) PCC, DCM, 87%; k) LiAlH₄, THF, 79%

l) PvCl, pyr, 83%.

protected with pivaloyl chloride to provide the ester 79 in

92% yield (Scheme 15). The cyclopropylmethyl methyl

ether was opened with catalytic acid to give the homoallylic

alcohol 80 in 97% yield, which was protected as the benzyl

ether 81 in 83% yield. Hydroboration±oxidation of the

ole®n gave the 6a-alcohol 82 in 60% yield. During the

oxidation of the organoborane under basic hydrogen perox-

ide conditions, it was necessary to maintain low temperature

to prevent the removal of the pivaloate ester by hydrogen

peroxide anion. Partial loss of the protecting group may

account for the somewhat lower yield. Protection of the

alcohol provided the acetate 83 in 100% yield. Debenzyl-

ation using hydrogen gas with a platinum catalyst gave the

alcohol 84 in 96% yield. The alcohol was subjected to the

Mitsunobu reaction with the acid 16 to provide the inverted

ester 85 in 90% yield. Photolysis of 85 followed by

complete basic hydrolysis and exhaustive protection with

tert-butyldimethylsilyl tri¯ate provided in 48% yield for

the three steps the fully protected C14 ole®n 86, which

was separated from the saturated steroid by silver nitrate-

impregnated silica gel ¯ash column chromatography.³⁶

Mono-deprotection of the primary silyl ether with aqueous

acetic acid gave the primary alcohol 87 in 86% yield.

Oxidation of the alcohol 87 provided in 83% yield the

aldehyde 88, to which was added isobutylmagnesium

bromide to give a 1:1 mixture of the diastereomeric alcohols

Scheme 17. a) Ac₂O, pyr, 97%; b) AcOH, re¯ux, 97%; c) RuCl₃, TBHP, DCE, 65%; d) NaBH₄, CeCl₃, DCM:MeOH, 98%; e) Ac₂O, pyr, 97%; f) BH₃-THF;

NaOH, H₂O₂, 50%; g) Ac₂O, TMSOTf, 97%; h) 10% NaOH, THF:EtOH, 95%.

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

1461

Scheme 18. a) TlPSCl, imidazole, DMF, 76%; b) 59, PPh₃, DEAD, 74%; c) hn, benzene, 48%; d) PCC, DCM, 63%; e) LiAlH₄, THF, 91%; f) MOMCl, Nal,

DCE, re¯ux, 80%; g) TBAF, THF, 92%; h) PCC, DCM, 82%; i) iBuMgBr, ether, 80%; j) BH₃-THF; NaOH, H₂O₂; k) Dess-Martin, DCM; l) 10% NaOH,

EtOH, 73% (3 steps); m) 50% HCl, THF:EtOH, 100%.

in 90% yield. Hydroboration±oxidation of the alkene

gave an approximately 10:1 ratio of the 14a-H 15a,23-

diol and the 14b-H 15b,23-diol, which were oxidized to

the corresponding 14a-H 15,23-dione 89a and the 14b-H

15,23-dione 89b, respectively, and separated by ¯ash

column chromatography. Acidic hydrolysis of the silyl

ethers of the major ketone 89a gave the diol 90, which

undewent base-catalyzed intramolecular aldol condensation

giving 7-deoxyxestobergsterol A 1d as the only aldol isomer

in quantitative yield for the two steps. Even though the trans

CD ring junction is more stable in the tetracyclic system,

once the new E ring has been formed, the cis CD ring junc-

tion is even more stable. Thus the calculations in the model

system were borne out in the full steroidal skeleton (Scheme

1).

relieved by epimerization to the more stable cis CD

isomer.]

3.5. Synthesis of xestobergsterol A

The pivaloate ester 79 was again prepared from stigmasterol

7 as described before. Opening of the cyclopropylmethyl

methyl ether with glacial acetic acid afforded the homo-

allylic acetate 91 in 94% yield (Scheme 16). Allylic oxi-

dation under the previously described conditions (RuCl₃,

t-BuOOH) yielded in 65% yield the enone 92, which was

reduced under Luche conditions to provide the 7b-alcohol

93 as the only detectable isomer in 97% yield. Protection of

the allylic alcohol with acetic anhydride and catalytic

DMAP gave the allylic acetate 94 in 98% yield, which

was hydroborated and oxidized to give only a modest

yield (57%) of the desired triol 95. The somewhat low

yield of the triol can be attributed to loss of the pivaloate

ester under the strongly basic hydrogen peroxide conditions

giving the tetrol. Nonetheless, the triol was carried on. Acid-

catalyzed acetylation of the triol 95 led to the triacetate 96 in

a gratifying yield of 95%. Selective deacetylation at C3 with

mild base, as described before, provided the C3 alcohol 97

with superb selectivity in 89% yield. Reaction of the C3

alcohol of 97 with the acid 16 under Mitsunobu conditions

provided the inverted ester 98 in 93% yield. Photolysis of

this benzophenone ester gave the ole®nic steroid in 61%

yield, which was oxidized with pyridinium chlorochromate

to aid in puri®cation by ¯ash column chromatography.

Reductive cleavage of all the esters with lithium aluminum

hydride afforded the pure ole®nic tetrol 99 in 42% overall

yield from the benzophenone 98. The ole®nic tetrol was

selectively protected to give the primary pivaloate ester 99

in 83% yield. Unfortunately, attempted protection of the

triol 99 with phosphorous pentoxide and methylal, as

described before, led to decomposition of the substrate.

This effectively ended all routes using the pivaloate protect-

ing group for the C23 alcohol. Although this ®rst route for

The structure was assigned by comparison of the chemical

shifts and coupling constants to those reported for xesto-

bergsterol A and by a NOESY experiment which showed

correlation between the angular 16-H and the hydrogens at

C24 and C25, thus establishing the stereochemistry at C23.

Accordingly, 7-deoxyxestobergsterol A 1d is available in 17

steps and 10% overall yield from the known alcohol 75 (22

steps and 7% overall yield from stigmasterol 7). [Another

pentacyclic steroid similar to 7-deoxyxestobergsterol A was

prepared and its CD cis:trans ratio determined. Upon

treatment with methanolic potassium hydroxide, the

15-ketosteroid iii gave exclusively the cis CD ring

junction.

Evidently the 5,5-cis fusion of the D and E rings places

additional strain on the D ring of steroids which can be

1462

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

the synthesis of xestobergsterol A was unsuccessful, it did

provide valuable information which helped in designing the

synthesis which eventually led to the ®rst total synthesis of

xestobergsterol A.

the natural product. [For characterization data for xestoberg-

sterol A, please see Refs 1 and 2. Natural xestobergsterol A:

25

D

a 237.78 (c 2.8, CHCl₃); natural xestobergsterol B:

25

D

a 235.08 (c 0.89, CHCl₃), private communication

from Dr. Maseo Takei (Japan National Cancer Research

Institute).]

Treatment of the known alcohol 75 with acetic anhydride

provided the acetate ester 101 in 97% yield (Scheme 17).

The cyclopropylmethyl methyl ether of 101 was opened

with glacial acetic acid to give in 97% yield the homoallylic

acetate 102, which was subjected to the same allylic oxi-

dation conditions as previously described (RuCl₃, tBuOOH)

to give the enone 103 in 65% yield. Once again the enone

103 was reduced under Luche conditions to give the allylic

7b-alcohol 104 as the only detectable diastereomer in 98%

yield. Protection of the alcohol as the acetate provided the

allylic acetate 105 in 97% yield, which was hydroborated

and treated with basic hydrogen peroxide to yield the

pure tetrol 106 (after column chromatography to remove

the minor isomer produced from addition of the borane to

the b-face of the alkene). Simple acetylation of the tetrol

gave mainly the triacetate in which the 7b-alcohol was

unprotected. Since the 7b-alcohol is quite sterically

hindered, it was necessary to use acetic anhydride and

catalytic trimethylsilyl tri¯ate in order to produce the

tetraacetate 107 in excellent yield (97%). The C6 and C7

acetates are again suf®ciently hindered relative to the

acetates at C3 and C23 to allow selective deprotection

using mildly basic conditions to afford the diol 108 in

95% yield.

3.6. Biological activity of analogues

Since the cis CD 15-keto functionality appears to be an

important factor in inhibition of histamine release, it was

important to prepare analogues to prove their ef®cacy. Inter-

estingly, when 16,23-seco-23-deoxyxestobergsterol A 73

(see Scheme 13) was tested for its inhibitory activity against

histamine release in rat peritoneal mast cells, it was shown

to be quite active (IC₅₀750 nM). These results suggest that

the simple cholesterol side chain is tolerated quite well and

that the presence of both the E ring and the additional

hydroxyl group at C23 are not extremely important for

activity. Since analogues containing the cholesterol side

chain are easier and cheaper to make, it is likely that

compounds based on the keto triol structure 73 will be

useful as antihistamine drugs. 7-Deoxyxestobergsterol A

1d (Scheme 15) was also tested for its inhibitory activity

against histamine release in rat peritoneal mast cells. Like

the ketotriol 73, 7-deoxyxestobergsterol A 1d was also quite

active (IC₅₀500 nM). Although these compounds are 10±15

times less active than xestobergsterol A la, they have some-

what better activity than contignasterol 2 (IC₅₀800 nM),

which is being explored as a potential lead compound for

inhibiting histamine release. Further testing of these ana-

logues for cytotoxicity is underway.

Protection of the C23 primary alcohol, in preference to the

C3 secondary alcohol, as its triisopropylsilyl (TIPS) ether

produced the alcohol 109 in 76% yield, which was reacted

with the known acid 16 using Mitsunobu conditions to give

the photolysis precursor 110 in 74% yield (Scheme 18).

Photolysis of the fully protected steroid provided the remote

functionalized product in 48% yield. The lower yield may

be attributed to loss of the triisopropylsilyl ether via hydro-

lysis or oxidation since photolysis of the substrate with the

C23 alcohol protected as the pivaloate gave much better

yields of the remote functionalized product. Nonetheless,

the photolysis product was oxidized with PCC to facilitate

puri®cation by conventional ¯ash column chromatography

and reduced to yield the triol 111 in 28% overall yield from

the benzophenone 110. Since we had encountered dif®-

culties protecting a similar triol with methylal and phos-

phorous pentoxide, we decided to use basic conditions to

introduce three methoxymethyl (MOM) ether protecting

groups. Treatment of the triol 111 with excess chloromethyl

methyl ether and sodium iodide³⁷(to generate the iodo-

methyl methyl ether in situ) gave the differentially fully

protected ether 112 in 80% yield. Removal of the primary

silyl ether gave in 92% yield the alcohol 113, which was

oxidized with pyridinium chlorochromate to provide the

aldehyde 114 in 82% yield. Addition of isobutylmagnesium

bromide gave an equal mixture of diastereomeric alcohols

in 80% yield. Hydroboration±oxidation followed by oxi-

dation with the Dess±Martin periodinane and ®nally base-

catalyzed intramolecular aldol condensation gave the

b-hydroxy ketone 115 in 73% yield for the three steps.

Final deprotection of the three MOM ethers was accom-

plished using acidic conditions to give xestobergsterol A

1a in 100% yield, which was identical in all respects to

4. Conclusion

In summary, we have utilized the Breslow remote func-

tionalization process to achieve the ®rst total synthesis of

xestobergsterol A as well as the synthesis of several simpler

analogues. In addition to completing the synthesis of xesto-

bergsterol A, we have determined that the Breslow remote

functionalization process can be carried out on substrates

with functionality in the A and B rings of steroids without

any deleterious effects or loss in yields compared to the

simpler substrates used by Breslow. Furthermore, side-

chain structure, substitution or lack of substitution at C7,

and formation of the 5,5-fused DE ring system in steroids

are all important factors determining the relative stabilities

of the trans and cis CD ring junctions. Finally, biological

testing provided valuable information suggesting that the

structural moiety necessary for biological activity is the

cis CD 15-keto functionality. Other changes in the structure,

such as replacement of the additional E ring in the xesto-

bergsterols with a simple cholesterol side chain, also

produce active analogues, which may be potentially useful

in the development of potent antihistamine agents.

5. Experimental section

5.1. General

All reactions were carried out under argon with the

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

1463

exclusion of moisture. Reagents were purchased from

Aldrich Chemical Company and were used without further

puri®cation unless otherwise noted. The following solvents

and reagents were distilled from the indicated agent

under argon: tetrahydrofuran (THF) and diethyl ether from

sodium benzophenone ketyl; dichloromethane, benzene,

and toluene from calcium hydride; and triethylamine from

potassium hydroxide. Flash column chromatography was

carried out in the indicated solvent system on 230±400

mesh silica gel. Analytical thin layer chromatography was

done on Merck silica gel F₂₅₄0.2 mm precoated plates. The

proton nuclear magnetic resonance spectra (¹H NMR) were

recorded on a Bruker ARX-400 spectrometer operating at

400.132 MHz. The ¹³C NMR spectra were also recorded on

a Bruker ARX-400 operating at 100.622 MHz. Spectra were

taken in the indicated solvent at ambient temperature unless

otherwise speci®ed, and the chemical shifts are reported in

parts per million (ppm) relative to the lock of the solvent

used. Resonance patterns are reported with the following

notations: app (apparent), b (broad), s (singlet), d (doublet),

t (triplet), q (quartet), quin (quintet), and m (multiplet). The

infrared (IR) spectra were recorded on a Nicolet 510 PCIR

spectrometer and are reported in reciprocal centimeters

(cm²¹). High-resolution mass spectra (MS) were recorded

on a VG Autospec at the UCLA Mass Spectrometry Labora-

tory and are reported in m/z units for the most abundant

peaks.

28.2, 26.0, 23.7, 22.8, 20.6, 18.3, 18.2, 11.8, 24.5, 24.6.

FTIR (thin ®lm): 3521, 2930, 2857, 1742, 1254, 1090,

835 cm²¹

.

5.1.3. Methyl 3a-[((1,1-dimethyl)ethyl)dimethylsilyloxy]-

7-ketocholan-24-oate (11a). To a stirring solution of the

alcohol 10 (328 mg, 0.630 mmol) in dichloromethane

(10 mL) at 258C was added anhydrous potassium acetate

followed by pyridinium chlorochromate (406 mg, 1.892

mmol) and the reaction was stirred overnight. Celite

(400 mg) was added and the solution was diluted with

ether (10 mL). The mixture was ®ltered through Celite

and the volatile components were removed under reduced

pressure. The residue was dissolved in ether and passed

through a short plug of silica gel, eluting with ether. The

ether was removed under reduced pressure to give the

ketone 11a (318 mg, 98%) as a white solid (mp 89±908C).

¹H NMR (CDCl₃, 400 MHz) d: 3.65 (3H, s), 3.54 (1H, m),

2.81 (1H, dd, J12.6, 6.0 Hz), 2.50±0.50 (25H, m), 1.16

(3H, s), 0.90 (3H, d, J6.2 Hz), 0.85 (9H, s), 0.63 (3H, s),

0.02 (6H, s). ¹³C NMR (CDCl₃, 50 MHz) d: 211.9, 174.6,

71.6, 54.8, 51.5, 49.5, 48.9, 46.2, 45.5, 42.7, 42.5, 38.9,

37.7, 35.2, 35.2, 34.3, 31.1, 30.5, 28.3, 25.9, 24.9, 23.1,

21.7, 18.4, 18.2, 12.1 (3 high-®eld carbons unresolved).

FTIR (thin ®lm): 2934, 2861, 1742, 1713, 1462, 1377,

1252, 1167, 1092 cm²¹. High-resolution MS (EI, m/z):

519.3867, calcd for C₃₁H₅₅O₄Si (M1H)¹519.3870.

5.1.1. Methyl 3a,7a-dihydroxycholan-24-oate (9b). To a

stirring solution the carboxylic acid 9a (348 mg,

0.886 mmol) in re¯uxing methanol (15 mL) was added 4

drops of concentrated hydrochloric acid. The solution

was cooled to 258C and neutralized with saturated sodium

bicarbonate. The volatile components were removed under

reduced pressure and water was added. The mixture was

extracted three times with ethyl acetate (3£5 mL). The

combined organic extracts were dried over magnesium

sulfate, ®ltered, and concentrated under reduced pressure

to give the methyl ester 9b (310 mg, 86%) as a white

solid, which was used without further puri®cation. ¹H

NMR (CDCl₃, 200 MHz) d: 3.91 (1H, m), 3.69 (3H, s),

3.50 (1H, m), 2.50±0.50 (28H, m), 0.92 (3H, d,

J7.0 Hz), 0.91 (3H, s), 0.66 (3H, s). FTIR (thin ®lm):

5.1.4. 3a-[((1,1-Dimethyl)ethyl)dimethylsilyloxy]-7-keto-

cholan-24-oic acid (11b). To a stirring solution of the

methyl ester 11a (318 mg, 0.613 mmol) in methanol

(8 mL) was added 10% KOH (aqueous, 8 mL) and the

mixture was re¯uxed for 1 h. The mixture was cooled to

258C, acidi®ed to pH 3, and extracted with ether

(310 mL). The combined organic extracts were dried over

magnesium sulfate, ®ltered, and concentrated under reduced

pressure to give the pure carboxylic acid 11b (289 mg,

1

93%). H NMR (CDCl₃, 200 MHz) d: 3.56 (1H, m), 2.81

(1H, dd, J12.6, 6.0 Hz), 2.50±0.50 (26H, m), 1.16 (3H, s),

0.91 (3H, d, J5.7 Hz), 0.85 (9H, s), 0.63 (3H, s), 0.01 (6H,

s). FTIR (thin ®lm): 3500±2500, 2934, 2857, 1705, 1254,

1092, 909, 835 cm²¹

.

3391, 2832, 2867, 1740, 1437, 1167, 1078 cm²¹

.

5.1.5. Methyl 3a-[((1,1-dimethyl)ethyl)dimethylsilyloxy]-

7b-hydroxycholan-24-oate (13). To a stirring solution of

the ketone 11b (289 mg, 0.572 mmol) in ammonia (25 mL),

tetrahydrofuran (5 mL), and methanol (3 mL) cooled to

2788C was added lithium metal (250 mg) in small chunks.

The solution was stirred for 20 min, quenched with ammo-

nium chloride, and then allowed to warm to 258C. Water

was added, the mixture was acidi®ed to a pH of 3, and

extracted with ether (3£10 mL). The combined organic

extracts were dried over magnesium sulfate, ®ltered, and

concentrated under reduced pressure to give a 3:1 mixture

of the 7b-alcohol 12 and the 7a-alcohol 10 (247 mg, 85%).

The crude products were dissolved in ether (10 mL) at 08C

and a solution of diazomethane in ether was added dropwise

until the starting material had disappeared by TLC. The

reaction was quenched with ammonium chloride, water

was added, and the mixture was extracted with ether

(3£10 mL). The combined organic extracts were dried

over magnesium sulfate, ®ltered, and concentrated under

reduced pressure. Flash column chromatography (silica

5.1.2. Methyl 3a-[((1,1-dimethyl)ethyl)dimethylsilyl-

oxy]-7a-hydroxycholan-24-oate (10). To a stirring solu-

tion of the diol 9b (87 mg, 0.214 mmol) in DMF (1.5 mL) at

258C was added imidazole (36 mg, 0.535 mmol) followed

by tert-butyldimethylsilyl chloride (39 mg, 0.257 mmol)

and the reaction was stirred overnight. Water was added

and the mixture was extracted with ethyl acetate

(3£5 mL). The combined organic extracts were dried over

magnesium sulfate, ®ltered, and concentrated under reduced

pressure. Flash column chromatography of the crude residue

(silica gel, 4:1 hexanes:ethyl acetate) provided the silyl

1

ether 10 (92 mg, 82%) as a clear glass. H NMR (CDCl₃,

400 MHz) d: 3.81 (1H, m), 3.64 (3H, s), 3.42 (1H, m), 1.85

(1H, ddd, J15.3, 10.0, 5.0 Hz), 2.20 (1H, m), 2.10±0.50

(25H, m), 0.90 (3H, d, J6.4 Hz), 0.87 (3H, s), 0.86 (9H, s),

0.63 (3H, s), 0.02 (6H, s). ¹³C NMR (CDCl₃, 100 MHz) d:

174.7, 72.9, 68.5, 55.8, 51.5, 50.4, 42.7, 41.6, 40.1, 39.6,

39.5, 35.5, 35.4, 35.1, 34.6, 32.7, 31.1, 31.00, 30.98,

1464

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

gel, 4:1 hexanes:ethyl acetate) of the residue gave the pure

7b-alcohol 13 (57% from the ketone 11b) as a clear glass.

¹H NMR (CDCl₃, 200 MHz) d: 3.66 (3H, s), 3.55 (1H, m),

2.50±0.50 (28H, m), 0.92 (3H, s), 0.90 (3H, d, J 6.2 Hz),

0.88 (9H, s), 0.66 (3H, s), 0.05 (6H, s). FTIR (thin

®lm): 3391, 2932, 2859, 1742, 1453, 1373, 1252, 1082,

32.4, 31.9, 30.8, 29.7, 28.2, 28.0, 25.9, 23.8, 22.8, 22.5,

20.4, 19.0, 18.1, 13.0, 11.3, 24.8, 24.9. FTIR (thin ®lm):

2953, 2928, 2855, 1742, 1462, 1385, 1372, 1252,

1057 cm²¹. High-resolution MS (EI, m/z): 517.4453, calcd

for C₃₃H₆₁O₂Si (M1H)¹517.4441.

835 cm²¹

.

5.1.9. 3a-Hydroxy-5a,14a-chlolestan-15-one (22). To a

stirring solution of the silyl ether 21 (4 mg, 0.008 mmol) in

ethanol (4 mL) at 258C was added concentrated hydro-

chloric acid (several drops) and the reaction was stirred

for 1 h. The ethanol was removed under reduced pressure

and water was added. The mixture was extracted with ether

(3£5 mL). The combined organic extracts were dried over

magnesium sulfate, ®ltered, and concentrated under reduced

pressure to yield the pure keto alcohol 22 (4 mg, 100%). ¹H

NMR (CDCl₃, 400 MHz) d: 4.04 (1H, m), 2.65 (1H, dddd,

J13.1, 3.2, 3.2, 3.2 Hz), 2.43 (1H, dd, J18.8, 8.6 Hz),

2.11 (1H, ddd, J12.4, 3.3, 3.3 Hz), 1.90±0.50 (27H, m),

0.97 (3H, d, J6.4 Hz), 0.863 (3H, d, J6.6 Hz), 0.859

(3H, d, J6.6 Hz), 0.78 (3H, s), 0.74 (3H, s). ¹³C NMR

(CDCl₃, 100 MHz) d: 216.4, 66.5, 66.0, 53.9, 51.5, 42.4,

42.0, 39.9, 39.3, 39.1, 36.2, 36.0, 35.8, 35.4, 32.1, 31.9,

30.6, 29.7, 28.9, 28.1, 28.0, 23.8, 22.8, 22.5, 20.3, 19.0,

13.0, 11.1 (1 additional low-®eld carbon). FTIR (thin

5.1.6. Methyl 3a-[((1,1-dimethyl)ethyl)dimethylsilyloxy]-

7b-nitrosyloxycholanoate (14). Through a stirring solution

of the alcohol 13 (27 mg, 0.052 mmol) in pyridine (6 mL) at

258C was bubbled excess nitrosyl chloride gas. The excess

pyridine was removed by high-vacuum rotovaporator and

the crude material was dissolved in ether and passed through

a short plug of silica gel, eluting with ether. Concentration

under reduced pressure gave the pure nitrite ester 14 (28 mg,

1

97%) as a clear glass. H NMR (CDCl₃, 200 MHz) d: 5.37

(1H, m), 3.65 (3H, s), 3.57 (1H, m), 2.50±0.50 (26H, m),

1.02 (3H, s), 0.93 (3H, d, J6.2 Hz), 0.92 (9H, s), 0.63 (3H,

s), 0.06 (6H, s).

5.1.7. 3a-[((1,1-Dimethyl)ethyl)dimethylsilyloxy]-5a,14a-

cholestan-15a-ol (19). To a stirring solution of the steroidal

ole®n 18a (35 mg, 0.070 mmol) in 5 mL of THF cooled to

08C was added borane tetrahydrofuran complex (1.0 M,

279 mL, 0.279 mmol). The reaction mixture was stirred at

258C for 12 h. The reaction mixture was cooled to 08C and

279 mL of 10% NaOH was added dropwise followed by

279 mL of 30% hydrogen peroxide. The reaction was stirred

vigorously for 1 h. Water was added and the products were

extracted with ether (3£5 mL). The organic layers were

combined and dried over anhydrous magnesium sulfate,

®ltered, and evaporated under reduced pressure to give a

white solid. Flash column chromatography of the crude

solid (silica gel, 20:1 hexanes:ether) gave the alcohol 19

®lm): 3289, 2923, 2851, 1736, 1449, 1366, 999 cm²¹

.

High-resolution MS (EI, m/z): 402.3497, calcd for

C₂₇H₄₆O₂402.3498.

5.1.10. 3a-Acetyloxy-14a,15a-epoxy-5a-cholestane (30).

To a stirring solution of the ole®nic alcohol 29a and the

saturated alcohol 29b (55 mg, 0.142 mmol, combined) in

pyridine at 258C was added acetic anhydride (29 mg,

0.284 mmol). The solution was stirred for 18 h and then

was quenched with ice chips. Water was added and the

mixture was extracted with ether (3£5 mL). The combined

organic extracts were dried over magnesium sulfate,

®ltered, and concentrated under reduced pressure. Column

chromatography (silica gel, 3:1 hexanes:ethyl acetate) of the

crude product gave a mixture of the unsaturated and satu-

rated acetates (59 mg, 97%). To a stirring solution of the

ole®n (30 mg, 0.070 mmol) in chloroform (5 mL) at 08C

was added m-chloroperbenzoic acid (18 mg, 0.105 mmol)

and the reaction was stirred for 1 h. Water was added and

the mixture was extracted with ether (3£5 mL). The

combined organic extracts were dried over magnesium sul-

fate, ®ltered, and concentrated under reduced pressure.

Flash column chromatography of the crude residue (silica

gel, 10:1 hexanes:ether) provided the epoxide 30 (31 mg,

100%) as a clear glass. ¹H NMR (CDCl₃, 400 MHz) d: 4.99

(1H, m), 3.32 (1H, s), 2.30±0.60 (27H, m), 2.03 (3H, s), 1.85

(1H, m), 0.856 (3H, d, 6.6 Hz), 0.851 (3H, d, J6.5 Hz),

0.847 (3H, s), 0.839 (3H, d, J6.4 Hz), 0.81 (3H, s). ¹³C

NMR (CDCl₃, 100 MHz) d: 170.8, 74.1, 70.0, 58.1, 50.1,

48.6, 41.1, 39.5, 39.0, 35.8, 35.8, 35.7, 33.7, 32.8, 32.7,

32.1, 32.0, 28.0, 27.4, 26.1, 25.3, 23.7, 22.8, 22.5, 21.5,

20.6, 18.7, 14.6, 11.2. FTIR (thin ®lm): 2936, 2867, 1736,

1468, 1456, 1364, 1250, 1238, 1020 cm²¹. High-resolution

MS (EI, m/z): 444.3607, calcd for C₂₉H₄₈O₃444.3603.

1

(6 mg, 17%) as a clear glass. H NMR (CDCl₃, 400 MHz)

d: 3.95 (2H, m), 2.00±0.50 (30H, m), 0.91 (3H, d,

J6.2 Hz), 0.88 (9H, s), 0.862 (3H, d, J6.6 Hz), 0.858

(3H, d, J6.6 Hz), 0.77 (3H, s), 0.68 (3H, s), 0.01 (6H, s).

¹³C NMR (CDCl₃, 100 MHz) d: 74.0, 66.8, 64.0, 54.3, 53.8,

44.1, 40.5, 40.2, 39.5, 38.8, 36.7, 36.1, 36.0, 35.3, 35.1,

32.5, 32.4, 29.7, 28.5, 28.0, 25.0, 23.7, 22.8, 22.6, 20.7,

18.5, 18.1, 13.4, 11.5, 24.8 (1 high-®eld carbon unre-

solved). FTIR (thin ®lm): 3335, 2930, 2855, 1472, 1252,

1059 cm²¹. High-resolution MS (EI, m/z): 517.4428, calcd

for C₃₃H₆₂O₂Si 517.4441.

5.1.8. 3a-[((1,1-Dimethyl)ethyl)dimethylsilyloxy]-5a,14a-

cholestan-15-one (21). To a stirring solution of the alcohol

19 (4 mg, 0.008 mmol) in dichloromethane (5 mL) was

added pyridinium chlorochromate (5 mg, 0.023 mmol) and

the reaction was stirred for 2 h. The solution was diluted

with an equal volume of ether and Celite (50 mg) was

added. The solution was ®ltered and concentrated under

reduced pressure to provide the ketone 21 (4 mg, 100%)

1

as a clear glass. H NMR (CDCl₃, 400 MHz) d: 3.95 (1H,

m), 2.51 (1H, dddd, J13.2, 3.1, 3.1, 3.1 Hz), 2.42 (1H, dd,

J18.8, 8.7 Hz), 2.10 (1H, ddd, J12.5, 3.4, 3.4 Hz), 1.90±

0.60 (25H, m), 1.76 (1H, dd, J18.7, 9.4 Hz), 0.98 (3H, d,

J6.4 Hz), 0.88 (9H, s), 0.864 (3H, d, J6.6 Hz), 0.860

(3H, d, J6.6 Hz), 0.75 (3H, s), 0.73 (3H, s), 0.01 (6H, s).

¹³C NMR (CDCl₃, 100 MHz) d: 216.4, 66.8, 66.0, 53.9,

51.6, 42.4, 42.0, 39.9, 39.3, 39.0, 36.6, 36.1, 36.0, 35.4,

5.1.11. 3a-Acetyloxy-5a,14b-cholestan-15a-ol (31). To

a stirring solution of the epoxide 30 (107 mg, 241 mmol) in

glacial acetic acid (5 mL) at 258C was added PtO₂(107 mg)

and the ¯ask was evacuated using a high-vacuum pump. The

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

1465

¯ask was back-®lled with hydrogen gas from a balloon. This

process of evacuating and back-®lling was repeated two

times and the reaction was stirred for 1 h. The solution

was ®ltered through Celite and evaporated under reduced

pressure to yield a crude oil. Flash column chromatography

(silica gel, 6:1 hexanes:ether) of this oil provided the

alcohol 31 (45 mg, 42%) as a clear glass. ¹H NMR

(CDCl₃, 400 MHz) d: 5.01 (1H, m), 4.26 (1H, dd, J3.6,

3.7 Hz), 2.05 (3H, s), 1.50±0.80 (30H, m), 1.00 (3H, s), 0.92

(3H, d, J6.3 Hz), 0.867 (3H, d, J6.6 Hz), 0.864 (3H, d,

J6.6 Hz), 0.74 (3H, s). ¹³C NMR (CDCl₃, 100 MHz) d:

170.8, 74.3, 70.2, 56.0, 55.6, 49.5, 43.2, 40.4, 39.7, 39.5,

39.2, 36.1, 35.3, 33.8, 33.5, 32.9, 32.8, 31.6, 28.8, 28.0,

26.0, 24.5, 22.8, 22.6, 21.8, 21.6, 21.2, 20.3, 10.3. FTIR

(thin ®lm): 3497, 2932, 2857, 1736, 1373, 1366, 1269,

1256, 1237, 1020 cm²¹. High-resolution MS (EI, m/z):

446.3760, calcd for C₂₉H₅₀O₃446.3760.

5.1.14. 3a-((4-Benzoyl)phenylacetyloxy)-6a-[((1,1-di-

methyl)ethyl)dimethylsilyloxy]-5a-cholestane (37). To a

stirring solution of the alcohol 36 (143 mg, 0.228 mmol) in

dry N,N-dimethylformamide (5 mL) at 258C was added

imidazole (31 mg, 0.456 mmol) followed by tert-butyl-

dimethylsilyl chloride (38 mg, 0.251 mmol) and the reac-

tion was stirred for 4 h. Water was added and the mixture

extracted with ether (3£10 mL). The combined organic

layers were dried over magnesium sulfate, ®ltered, and

concentrated under reduced pressure. Flash column chroma-

tography of the residue (silica gel, 8:1 hexanes:ether)

provided the silyl ether 37 (155 mg, 92%) as a clear glass.

¹H NMR (CDCl₃, 400 MHz) d: 7.90±7.70 (4H, m), 7.65±

7.30 (5H, m), 5.09 (1H, m), 3.69 (2H, s), 3.29 (1H, ddd,

J10.6, 10.6, 4.4 Hz), 1.90±0.60 (29H, m), 0.88 (3H, d,

J6.7 Hz), 0.856 (3H, d, J6.6 Hz), 0.852 (3H, d,

J6.6 Hz), 0.82 (9H, s), 0.75 (3H, s), 0.63 (3H, s), 20.01

(3H, s), 20.02 (3H, s). ¹³C NMR (CDCl₃, 100 MHz) d:

195.8, 169.9, 139.4, 137.7, 136.0, 132.2, 130.4, 129.9,

129.3, 128.2, 70.5, 70.2, 56.1, 53.7, 46.8, 42.5, 42.0, 41.9,

39.8, 39.6, 36.4, 36.2, 35.8, 34.1, 33.2, 28.2, 28.0, 27.5,

25.9, 25.8, 24.1, 23.9, 22.9, 22.7, 22.6, 20.7, 18.7, 18.0,

12.5, 12.1, 23.8, 24.7. FTIR (thin ®lm): 2936, 2857, 1732,

1661, 1609, 1472, 1447 cm²¹. High-resolution MS (CI, m/z):

741.5276, calcd for C₄₈H₇₃O₄Si (M1H)¹741.5278.

5.1.12. 3a-Acetyloxy-5a,14b-cholestan-15-one (32). To

a stirring solution of the alcohol 31 (30 mg, 0.067 mmol)

in dichloromethane (5 mL) at 258C was added pyridinium

chlorochromate (29 mg, 0.134 mmol) and the reaction was

stirred for 3 h. The solution was diluted with an equal

volume of ether and Celite (50 mg) was added and the solu-

tion was ®ltered through Celite and concentrated under

reduced pressure. Flash column chromatography of the

residue (silica gel, 6:1 hexanes:ether) provided the ketone

32 (30 mg, 100%) as a clear glass. ¹H NMR (CDCl₃,

400 MHz) d: 5.00 (1H, m), 2.51 (1H, dddd, J13.2, 13.2,

13.2, 4.6 Hz), 2.34 (1H, dd, J19.8, 10.0 Hz), 2.17 (1H, m),

2.12 (1H, m), 2.05 (3H, s), 1.92 (1H, m), 1.80±0.60 (27H,

m), 1.16 (3H, s), 0.85 (9H, d, J6.6 Hz), 0.74 (3H, s). ¹³C

NMR (CDCl₃, 100 MHz) d: 220.8, 170.8, 70.0, 57.9, 48.4,

47.2, 42.1, 39.3, 39.1, 38.3, 37.6, 35.8, 33.5, 33.3, 32.9,

32.6, 31.8, 29.0, 28.5, 28.0, 25.9, 25.5, 22.7, 22.5, 21.6,

21.2, 19.2, 19.1, 11.2. FTIR (thin ®lm): 2951, 2930, 2869,

1736, 1462, 1453, 1248, 1237 cm²¹. High-resolution MS

(EI, m/z): 444.3605, calcd for C₂₉H₄₈O₃444.3603.

5.1.15. 6a-[((1,1-Dimethyl)ethyl)dimethylsilyloxy]-3a-

((4-hydroxyphenylmethyl)-phenylacetyloxy)-5a-cholest-

14-ene (38), Lactone (39a), and Lactone (39b). The ester

37 (604 mg, 0.815 mmol) was photolyzed (450 W mercury

arc lamp, pyrex ®lter) in degassed benzene (815 mL) for

10 h at 258C. The solvent was removed under reduced pres-

sure and the residue subjected to column chromatography to

give the ole®n 38 (181 mg, 30%), the lactone 39a (120 mg,

20%), and lactone 39b (120 mg, 20%) as clear glasses.

1

39a: H NMR (CDCl₃, 400 MHz) d: 7.80±7.00 (9H, m),

5.10 (1H, s), 4.99 (1H, m), 3.63 (1H, d, J13.1 Hz), 3.49

(1H, d, J13.1 Hz), 3.28 (1H, ddd, J10.6, 10.6, 4.4 Hz),

2.43 (1H, d, J14.9 Hz), 2.20 (1H, d, J14.9 Hz), 2.10±

0.60 (29H, m), 0.91 (3H, d, J5.3 Hz), 0.886 (3H, d, J

6.2 Hz), 0.881 (3H, d, J6.5 Hz), 0.878 (3H, s), 0.73 (3H,

s), 0.70 (3H, s), 0.67 (3H, s), 0.27 (3H, s), 20.07 (3H, s). ¹³C

NMR (CDCl₃, 100 MHz) d: 170.7, 149.2, 145.1, 133.3,

131.3, 128.8, 128.0, 127.0, 126.3, 126.0, 125.3, 78.1, 71.8,

69.0, 57.2, 56.3, 56.0, 53.8, 46.6, 42.9, 42.6, 41.8, 39.8,

39.5, 36.7, 36.2, 35.9, 34.6, 33.3, 31.6, 28.8, 28.2, 28.1,

28.0, 27.7, 25.8, 24.2, 23.9, 22.9, 22.8, 22.7, 22.6, 20.7,

18.7, 14.2, 12.3, 12.1, 22.0, 24.3 (3 additional high-®eld

carbons). FTIR (thin ®lm): 3409, 2946, 2861, 1732, 1468,

1381, 1250 cm²¹. High-resolution MS (EI, m/z): 741.5262,

calcd for C₄₈H₇₃O₄Si (M1H)¹741.5278.

5.1.13. 3a-((4-Benzoyl)phenylacetyloxy)-5a-cholestan-

6a-ol (36). To a stirring solution of the diol 35 (300 mg,

0.741 mmol) in THF (10 mL) at 258C was added triphenyl-

phosphine (389 mg, 1.482 mmol) and the carboxylic acid 16

(196 mg, 1.482 mmol) followed by the dropwise addition of

diethyl azodicarboxylate (233 mL, 1.482 mmol). The reac-

tion was stirred for 0.5 h and then quenched with water. The

mixture was extracted with ether (3£10 mL). The combined

organic layers were dried over magnesium sulfate, ®ltered,

and concentrated under reduced pressure. Flash column

chromatography of the residue (silica gel, 3:1 hexanes:ethyl

1

acetate) provided the ester 36 (280 mg, 60%). H NMR

(CDCl₃, 400 MHz) d: 7.90±7.30 (9H, m), 5.11 (1H, m),

3.68 (2H, s), 3.29 (1H, ddd, J10.6, 10.6, 4.3 Hz), 2.20±

0.50 (30H, m), 0.88 (3H, d, J6.5 Hz), 0.857 (3H, d,

J6.6 Hz), 0.855 (3H, d, J6.6 Hz), 0.74 (3H, s), 0.62

(3H, s). ¹³C NMR (CDCl₃, 100 MHz) d: 196.0, 170.0,

139.3, 137.4, 136.1, 132.3, 130.3, 129.9, 129.3, 128.2,

70.2, 69.3, 56.1, 56.0, 53.6, 46.8, 42.4, 41.8, 41.6, 39.6,

39.4, 36.3, 36.0, 35.6, 34.0, 32.9, 28.0, 27.9, 27.0, 25.6,

24.0, 23.7, 22.7, 22.5, 20.6, 18.6, 12.4, 11.9. FTIR (thin

®lm): 3497, 2944, 2869, 1732, 1659, 1607, 1279, 1468,

1447 cm²¹. High-resolution MS (EI, m/z): 626.4323, calcd

for C₄₂H₅₈O₄626.4335.

1

39b: H NMR (CDCl₃, 400 MHz) d: 7.80±7.00 (9H, m),

5.04 (1H, m), 3.64 (1H, d, J12.7 Hz), 3.42 (1H, d, J

12.7 Hz), 3.23 (1H, ddd, J10.3, 10.3, 4.5 Hz), 2.95 (1H,

d, J14.6 Hz), 2.20±0.70 (30H, m), 2.07 (1H, d, J

14.6 Hz), 1.08 (3H, s), 0.92 (3H, d, J6.5 Hz), 0.874 (3H,

d, J6.6 Hz), 0.870 (3H, d, J6.6 Hz), 0.75 (3H, s), 0.67

(3H, s), 0.44 (3H, s), 0.03 (3H, s), 20.19 (3H, s). ¹³C NMR

(CDCl₃, 100 MHz) d: 170.7, 150.2, 145.5, 132.4, 129.9,

128.8, 128.3, 127.4, 127.0, 125.9, 125.6, 80.6, 70.0, 69.3,

56.3, 56.3, 54.0, 47.5, 46.3, 42.8, 42.6, 42.4, 39.9, 39.5,

1466

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

36.5, 36.2, 35.9, 34.1, 33.5, 28.2, 28.0, 27.7, 27.4, 26.0,

24.2, 23.9, 22.8, 22.6, 21.2, 20.7, 20.6, 18.7, 12.4, 12.1,

22.9, 26.3. FTIR (thin ®lm): 3505, 2944, 2869, 1732,

1466, 1381, 1248 cm²¹. High-resolution MS (EI, m/z):

739.5114, calcd for C₄₈H₇₁O₄Si (M±H)¹739.5122.

¹³C NMR (CDCl₃, 50 MHz) d: 170.7, 72.6, 70.7, 60.4, 56.2,

53.7, 48.7, 42.6, 39.8, 39.5, 37.7, 37.2, 36.6, 36.1, 35.7,

34.1, 32.1, 31.1, 28.2, 28.0, 24.1, 23.8, 22.8, 22.6, 21.2,

18.7, 14.2, 13.2, 12.0. FTIR (thin ®lm): 3395, 2946, 2869,

1740, 1242. High-resolution MS (EI, m/z): 446.3763, calcd

for C₂₉H₅₀O₃446.3760.

5.1.16. 3b-Phenylmethoxy-5a-cholestan-6a-ol (41). To a

stirring solution of the ole®n 40 (401 mg, 0.841 mmol) in

THF (5 mL) at 08C was added borane tetrahydrofuran

complex (1.0 M, 4.2 mL, 4.2 mmol) and the solution was

stirred overnight at 258C. The solution was cooled to 08C

and 10% sodium hydroxide (4.2 mL) was added dropwise

followed by 30% hydrogen peroxide (4.2 mL). The mixture

was stirred vigorously for several hours. Water was added

and the mixture was extracted with ether (3£15 mL). The

combined organic extracts were dried over magnesium

sulfate, ®ltered, and concentrated under reduced pressure.

Flash column chromatography of the residue (silica gel, 4:1

hexanes:ethyl acetate) gave the alcohol 41 (380 mg, 89%) as

a white solid. ¹H NMR (CDCl₃, 200MHz) d: 7.50±7.10 (5H,

m), 4.61 (1H, d, J23.7 Hz), 4.53 (1H, d, J23.7 Hz),

3.60±3.10 (2H, m), 2.50±2.25 (1H, m), 2.30±0.50 (38H,

m), 0.89 (3H, s), 0.65 (3H, s). ¹³C NMR (CDCl₃, 50 MHz)

d: 139.1, 128.4, 127.6, 127.4, 76.5, 70.0, 69.5, 56.2, 53.9,

51.7, 42.6, 41.7, 39.9, 39.6, 37.4, 36.6, 35.8, 28.8, 28.2,

28.2, 28.1, 24.3, 23.9, 22.9, 22.8, 22.6, 21.2, 18.7, 13.5,

12.1 (2 high-®eld carbons unresolved). FTIR (thin ®lm):

5.1.19. 3a-((-4-Benzoyl)phenylacetyloxy)-6a-acetyloxy-

5a-cholestane (44). To a stirring solution of the alcohol

43 (250 mg, 0.560 mmol) in THF (10 mL) at 258C was

added triphenyl-phosphine (294 mg, 1.120 mmol) and the

carboxylic acid 16 (161 mg, 0.672 mmol) followed by

the dropwise addition of DEAD (176 mL, 1.120 mmol)

and the reaction was stirred for 1 h. Water was added and

the mixture was extracted with ether (3£5 mL). The

combined organic extracts were dried over magnesium

sulfate, ®ltered, and concentrated under reduced pressure.

Flash column chromatography of the crude residue (silica

gel, 6:1 hexanes:ethyl acetate) provided the benzophenone

ester 44 (322 mg, 86%) as a clear glass (mp 56±578C).

a

Â Ã

25

D

118.38 (c 0.75, CH₂Cl₂). ¹H NMR (CDCl₃,

400 MHz) d: 7.85±7.63 (4H, m), 7.60±7.48 (1H, m),

7.48±7.30 (4H, m), 5.05 (1H, m), 4.57 (1H, ddd, J10.4,

10.4, 4.3 Hz), 3.69 (2H, s), 2.10±0.40 (29H, m), 1.95

(3H, s), 0.873 (3H, d, J5.4 Hz), 0.841 (6H, d,

J6.4 Hz), 0.79 (3H, s), 0.60 (3H, s). ¹³C NMR (CDCl₃,

100 MHz) d: 195.8, 170.7, 169.9, 139.4, 137.6, 136.2,

132.4, 130.4, 129.9, 129.3, 128.3, 72.5, 70.0, 56.1, 56.1,

53.6, 44.0, 42.6, 42.0, 39.7, 39.5, 37.6, 36.7, 36.1, 35.7,

33.9, 32.9, 28.1, 28.0, 27.1, 25.6, 24.0, 23.8, 22.8, 22.6,

21.2, 20.6, 18.7, 12.4, 12.0. FTIR (thin ®lm): 2948, 2869,

1732, 1661, 1607, 1447, 1375, 1244, 1148, 756 cm²¹. High-

resolution MS (CI, m/z): 669.4515, calcd for C₄₄H₆₁O₅

(M1H)¹669.4519.

3546, 2946, 1497, 1464, 1375, 1111, 1042, 909, 734 cm²¹

.

5.1.17. 6a-Acetyloxy-3b-phenylmethoxy-5a-cholestane

(42). To a stirring solution of the alcohol 41 in pyridine

(20 mL) at 258C was added acetic anhydride (310 mL,

3.24 mmol) and the reaction was stirred overnight. Water

was added and the mixture was extracted with ether

(3£20 mL). The organic layers were dried over magnesium

sulfate, ®ltered, and concentrated under reduced pressure.

Excess pyridine was removed by high-vacuum rotovapo-

ration. Flash column chromatography of the residue (silica

gel, 100% CHCl₃) provided the ester 42 (1.21 g, 84%) as a

5.1.20. 3a,6a-Diacetyloxy-14a,15a-epoxy-5a-cholestane

(47). The benzophenone ester 44 (444 mg, 0.664 mmol) in

puri®ed, degassed benzene (664 mL) at 258C was photo-

lyzed using a 450 W mercury arc lamp for 10 h. The mixture

was concentrated under reduced pressure to give a crude oil.

The crude oil was dissolved in 10% KOH (4 mL), ethanol

(12 mL) and THF (4 mL) and the mixture stirred at 258C for

18 h. Water was added and the mixture was extracted with

ether (3£20 mL). The combined organic extracts were dried

over magnesium sulfate, ®ltered, and concentrated under

reduced pressure to give a pale yellow solid. The crude

ole®nic diol was dissolved in pyridine (5 mL) at 258C and

excess acetic anhydride was added and the mixture was

stirred overnight. Water was added and the mixture was

extracted with ether (3£5 mL). The combined organic

extracts were dried over magnesium sulfate, ®ltered, and

concentrated under reduced pressure. Flash column chroma-

tography (silica gel, 8:1 hexanes:ethyl acetate) of the crude

solid gave a mixture of the ole®nic diacetate 46a and the

saturated diacetate 46b (220 mg, 68% combined yield). To a

stirring solution of the ole®n mixture (116 mg, 0.238 mmol)

in chloroform (5 mL) cooled to 08C was added m-chloro-

perbenzoic acid (62 mg, 0.357) and the reaction was stirred

for approximately 1 h. Saturated sodium thiosulfate

(0.1 mL) was added followed by water and the mixture

was extracted with dichloromethane (2£10 mL). The

combined organic layers were dried over magnesium

sulfate, ®ltered, and evaporated under reduced pressure to

1

white solid (mp 146±1488C). H NMR (CDCl₃, 200 MHz)

d: 7.50±7.10 (5H, m), 4.67 (1H, ddd, J10.4, 10.4, 4.6 Hz),

4.57 (1H, d, J23.8 Hz), 4.51 (1H, d, J23.8 Hz), 3.31 (1H,

m), 2.30±0.50 (41H, m), 2.02 (3H, s), 0.66 (3H, s). ¹³C

NMR (CDCl₃, 50 MHz) d: 170.6, 139.1, 128.4, 127.6,

127.4, 77.9, 72.6, 70.1, 56.3, 53.8, 48.7, 42.7, 39.9, 39.6,

37.8, 37.2, 36.9, 36.2, 35.8, 34.2, 29.2, 28.2, 28.0, 27.9,

24.2, 24.0, 22.9, 22.7, 21.3, 21.2, 18.8, 13.4, 12.1. FTIR

(thin ®lm): 3031, 2944, 2869, 1740, 1455, 1375, 1242,

1026 cm²¹. High-resolution MS (EI, m/z): 536.4232, calcd

for C₃₆H₅₅O₃(M±H)¹536.4229.

5.1.18. 6a-Acetyloxy-5a-cholestan-3b-ol (43). The ben-

zyl ether 42 (492 mg, 0.917 mmol) was dissolved in

THF:ethanol (10:1, 12 mL) at 258C and activated palladium

on carbon catalyst (49 mg) was added. The ¯ask was evac-

uated and back-®lled with hydrogen gas from a balloon.

This process of evacuation and back-®lling was repeated

two times and the reaction was stirred for 2 h. The mixture

was ®ltered through a small plug of Celite and concentrated

under reduced pressure to give the pure alcohol 43 (375 mg,

92%) as a white solid (mp 848C). ¹H NMR (CDCl₃,

200 MHz) d: 4.67 (1H, ddd, J10.4, 10.4, 4.6 Hz), 3.55

(1H, m), 2.30±0.50 (42H, m), 2.03 (3H, s), 0.64 (3H, s).

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

1467

yield a crude oil. Flash column chromatography (silica gel,

10:1 hexanes:ethyl acetate) of the crude oil provided the

epoxide 47 (109 mg, 91% based on the amount of ole®n

the mixture was allowed to stir for an additional 18 h.

Sodium sul®te (2.158 g) was added and the mixture allowed

to stir for 1 h. Water was added and the mixture was

extracted with ether (3£30 mL). The combined organic

extracts were dried over magnesium sulfate, ®ltered, and

concentrated under reduced pressure. Flash column chroma-

tography (silica gel, 7:1 hexanes:ethyl acetate) of the resi-

due provided the enone 50 (1.409 g, 63%) as a white solid.

¹H NMR (CDCl₃, 400 MHz) d: 5.67 (1H, d, J1.0 Hz), 4.69

(1H, dddd, J11.4, 11.4, 4.6, 4.6 Hz), 2.60±2.30 (3H, m),

2.20 (1H, dd, J10.7, 10.7 Hz), 2.10±0.50 (22H, m), 2.02

(3H, s), 1.19 (3H, s), 0.90 (3H, d, J6.5 Hz), 0.840 (3H, d,

J6.6 Hz), 0.836 (3H, d, J6.6 Hz), 0.66 (3H, s). ¹³C NMR

(CDCl₃, 100 MHz) d: 201.9, 170.2, 163.8, 126.7, 72.2, 54.8,

49.9, 49.8, 45.4, 43.1, 39.5, 38.6, 38.3, 37.7, 36.2, 36.0,

35.7, 28.5, 28.0, 27.4, 26.3, 23.8, 22.8, 22.6, 21.2, 21.2,

18.9, 17.2, 12.0. FTIR (thin ®lm): 2951, 1728, 1672,

1

as judged by H NMR in the previous step) as a colorless

Â Ã

25

D

1

glass. a 151.38 (c 0.75, CH₂Cl₂). H NMR (CDCl₃,

400 MHz) d: 4.99 (1H, m), 4.57 (1H, ddd, J11.3, 11.3,

4.6 Hz), 3.25 (1H, m), 2.24 (1H, ddd, J12.1, 12.1, 4.1 Hz),

2.10±0.60 (40H, m), 2.00 (3H, s), 1.94 (3H, s). ¹³C NMR

(CDCl₃, 100 MHz) d: 170.9, 170.4, 73.5, 72.3, 68.8, 58.2,

49.4, 48.5, 43.1, 41.0, 39.4, 36.8, 35.7, 35.4, 33.6, 32.7,

31.9, 31.2, 30.8, 27.9, 27.1, 25.6, 23.6, 22.8, 22.5, 21.5,

21.2, 20.4, 18.7, 14.5, 12.3. FTIR (thin ®lm): 2951, 2872,

1738, 1240, 1022 cm²¹

.

5.1.21. 3a,6a-Diacetyloxy-5a,14b-cholestan-15a-ol (48).

To a stirring solution of the epoxide 47 (95 mg, 197 mmol)

in glacial acetic acid (5 mL) was added Pt (95 mg) and the

¯ask was evacuated using a high-vacuum pump. The ¯ask

was back-®lled with hydrogen gas from a balloon. This

process of evacuating and back-®lling was repeated two

times and the reaction was allowed to stir at 258C for

12 h. The solution was ®ltered through Celite and evapo-

rated under reduced pressure to yield a crude oil. Flash

column chromatography (silica gel, 6:1 hexanes:ethyl

1636, 1468, 1248, 1038, 911, 731 cm²¹

.

5.1.24. 3b-Acetyloxycholest-5-en-7b-ol (51). To a stirring

solution of the enone 50 (1.150 g, 2.598 mmol) in dichloro-

methane:methanol (4:1, 30 mL) cooled to 2788C was added

cerium trichloride heptahydrate (1.936 g, 5.196 mmol)

followed by sodium borohydride (0.197 g, 5.196 mmol).

The solution was stirred and allowed to warm to 258C

slowly. The solution was acidi®ed with 10% HCl and

extracted with ether (3£100 mL). The combined organic

extracts were dried over magnesium sulfate, ®ltered, and

concentrated under reduced pressure to yield the allylic

alcohol 51 (1.070 g, 93%) as a white solid. ¹H NMR

(CDCl₃, 400 MHz) d: 5.28 (1H, s), 4.59 (1H, m), 3.81

(1H, d, J7.4 Hz), 2.50±2.20 (2H, m), 2.20±0.50 (27H,

m), 2.00 (3H, s), 1.03 (3H, s), 0.89 (3H, d, J6.3 Hz),

0.84 (6H, d, J6.6 Hz), 0.66 (3H, s). ¹³C NMR (CDCl₃,

100 MHz) d: 170.5, 142.2, 126.4, 73.4, 73.1, 55.9, 55.4,

48.2, 42.9, 40.7, 39.52, 39.50, 37.6, 36.7, 36.5, 36.2, 35.7,

28.5, 28.0, 27.7, 26.4, 23.8, 22.8, 22.6, 21.4, 21.0, 19.1,

18.8, 11.8. FTIR (thin ®lm): 3328, 2944, 2870, 2853,

acetate) of the crude oil provided the alcohol 48 (42 mg,

Â Ã

25

D

1

44%) as a clear glass. a 175.88 (c 1.2, CH₂Cl₂). H

NMR (CDCl₃, 400 MHz) d: 5.05 (1H, m), 4.64 (1H, ddd,

J8.6, 8.6, 4.5 Hz), 4.22 (1H, dd, J3.5, 3.5 Hz), 2.10±

0.60 (28H, m), 2.05 (3H, s), 2.01 (3H, s), 0.98 (3H, s),

0.90 (3H, d, J6.4 Hz), 0.85 (3H, d, J5.2 Hz), 0.84 (3H,

d, J6.6 Hz), 0.79 (3H, s). ¹³C NMR (CDCl₃, 100 MHz) d:

171.0, 170.7, 74.1, 73.3, 69.1, 55.6, 55.2, 48.3, 43.7, 43.6,

42.8, 40.4, 39.5, 39.4, 37.0, 37.0, 35.2, 33.4, 32.8, 32.2,

28.0, 27.3, 25.6, 24.5, 22.8, 22.6, 21.7, 21.3, 21.1, 20.2,

11.7. FTIR (thin ®lm): 3530, 2951, 2870, 1736, 1466,

1248, 1022 cm²¹

.

5.1.22. 3a,6a-Diacetyloxy-5a,14b-cholestan-15-one (49).

To a stirring solution of the alcohol 48 (31 mg, 0.061 mmol)

in dichloromethane (5 mL) at 258C was added pyridinium

chlorochromate (26 mg, 0.122 mmol) and the reaction was

stirred for 3 h. The solution was diluted with an equal

volume of ether and Celite (50 mg) was added. The solution

was ®ltered and concentrated under reduced pressure. Flash

column chromatography of the crude residue (silica gel, 8:1

1732, 1468, 1375, 1248, 1138, 1036 cm²¹

.

5.1.25. 3b-Acetyloxy-7b-[((1,1-dimethyl)ethyl)dimethyl-

silyloxy]-cholest-5-ene (52). To a stirring solution of the

alcohol 51 (219 mg, 0.491 mmol) in DMF (10 mL) at

258C was added imidazole (50 mg, 0.737 mmol) followed

by tert-butyl dimethylsilyl chloride (185 mg, 1.228 mmol)

and the reaction was stirred for 18h. Water was added and

the mixture was extracted with ether (3£5 mL). The

combined organic extracts were dried over magnesium sul-

fate, ®ltered and concentrated under reduced pressure. Flash

column chromatography of the crude residue (silica gel, 20:1

hexanes: ether) afforded the ole®n 52 (269 mg, 98%) as a

hexanes:ethyl acetate) provided the ketone 49 (31 mg, 84%)

Â Ã

25

D

1

as a clear glass. a 137.28 (c 1.2, CH₂Cl₂). H NMR

(CDCl₃, 400 MHz) d: 5.04 (1H, m), 4.52 (1H, ddd, J8.6,

8.6, 4.5 Hz), 2.54 (1H, bddd, J11.4, 8.6, 8.6 Hz), 2.10±

0.60 (29H, m), 2.06 (3H, s), 2.00 (3H, s), 1.14 (3H, s), 0.85

(9H, d, J6.7 Hz), 0.80 (3H, s). ¹³C NMR (CDCl₃,

100 MHz) d: 220.4, 171.2, 170.7, 73.1, 68.9, 57.0, 48.3,

46.4, 43.0, 42.0, 39.3, 38.0, 37.4, 36.8, 34.5, 33.3, 32.6,

31.9, 31.7, 27.9, 27.2, 25.5, 22.7, 22.5, 21.6, 21.3, 21.0,

19.2, 19.0, 12.3 (1 high-®eld carbon unresolved). FTIR

(thin ®lm): 2950, 2870, 1734 (broad), 1466, 1373, 1437,

1

clear glass. H NMR (CDCl₃, 400 MHz) d: 5.29 (1H, s),

4.61 (1H, m), 3.93 (1H, d, J7.8 Hz), 2.40±2.20 (2H, m),

2.20±0.50 (24H, m), 2.01 (3H, s), 1.04 (3H, s), 0.90 (3H, d,

J 6.5 Hz), 0.86 (9H, s), 0.852 (3H, d, J6.6 Hz), 0.848

(3H, d, J6.6 Hz), 0.66 (3H, s), 0.05 (3H, s), 0.04 (3H, s).

¹³C NMR (CDCl₃, 100 MHz) d: 170.4, 140.9, 127.4, 74.6,

73.5, 56.2, 55.5, 48.4, 43.0, 40.5, 39.6, 39.5, 37.8, 36.8,

36.4, 36.2, 35.8, 28.5, 28.0, 27.8, 26.8, 26.3, 25.7, 23.8,

22.8, 22.6, 21.4, 21.1, 18.9, 18.8, 18.2, 11.9, 22.6, 23.3.

FTIR (thin ®lm): 2932, 2855, 1736, 1470, 1443, 1375, 1248,

1057, 1088 cm²¹. High-resolution MS (EI, m/z): 558.4469,

calcd for C₃₅H₆₂O₃Si 558.4468.

1364, 1244, 1022 cm²¹

.

5.1.23. 3b-Acetyloxycholest-5-en-7-one (50). A mixture of

the ole®n 34 (2.158 g, 5.034 mmol) and ruthenium tri-

chloride monohydrate (7 mg) in cyclohexane:water (5:1,

30 mL) was stirred at 258C. tert-Butyl hydroperoxide

(70% aqueous, 6.52 mL) was added dropwise over 6 h and

1468

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

5.1.26. 7b-[((1,1-Dimethyl)ethyl)dimethylsilyloxy]-5a-

cholestan-6a,3b-diol (53) and 6a-[((1,1-dimethyl)-

ethyl)dimethylsilyloxy]-5a-cholestan-3b,7b-diol (54).

solution of the ole®n 55 (1.010 g, 2.075 mmol) at 08C in

THF (10 mL) was added borane±tetrahydrofuran complex

(1.0 M, 8.3 mL, 8.300 mmol) and the mixture was allowed

to warm to 258C and stir overnight. The solution was cooled

to 08C and 10% NaOH (8 mL) was added followed by 30%

hydrogen peroxide (8 mL) and the mixture was stirred

vigorously for 2 h. Water was added and the mixture was

extracted with ether (3£30 mL). The combined organic

extracts were dried over magnesium sulfate, ®ltered and

concentrated under reduced pressure to yield a white

solid. Flash column chromatography of the crude residue

(silica gel, 12:1 chloroform:methanol) provided the triol

56 (700 mg, 80%) as a white solid. ¹H NMR (acetic

acid-d₄, 400 MHz) d: 3.63 (1H, m), 3.39 (1H, dd, J8.8,

10.8 Hz), 3.21 (1H, dd, J9.5, 9.1 Hz), 2.30±0.50 (30H, m),

0.95 (3H, d, J6.5 Hz), 0.875 (3H, s), 0.875 (3H, d, J

6.6 Hz), 0.872 (3H, d, J6.5 Hz), 0.70 (3H, s). ¹³C NMR

(CDCl₃with acetic acid-d₄, 100 MHz) d: 80.3, 74.6, 70.9,

55.9, 55.3, 52.1, 47.6, 43.4, 40.8, 39.8, 39.5, 37.2, 36.2,

35.7, 35.6, 31.9, 30.2, 28.6, 28.0, 26.8, 23.8, 22.8, 22.5,

21.4, 18.8, 13.4, 12.1. FTIR (thin ®lm): 3339, 2950, 2869,

To

a stirring solution of the ole®n 52 (265 mg,

0.474 mmol) in THF (10 mL) at 258C was added borane

tetrahydrofuran complex (1.0 M, 2.37 mL, 2.371 mmol)

and the reaction was stirred for 18 h. 3 mL of 10% sodium

hydroxide was added followed by the addition of 30%

hydrogen peroxide and the mixture was stirred for 1 h.

Water was added and the mixture was extracted with ether

(3£5 mL). The combined organic extracts were dried over

magnesium sulfate, ®ltered, and concentrated under reduced

pressure. Flash column chromatography (silica gel, 3:1

hexanes:ethyl acetate) of the crude solid provided the diol

53 (104 mg, 41%) and the diol 54 (100 mg, 39%) as clear

glasses.

53: ¹H NMR (CDCl₃, 400 MHz) d: 3.54 (1H, m), 3.29 (1H,

dd, J8.8, 8.4 Hz), 3.23 (1H, dd, J10.3, 8.3 Hz), 2.30±

0.50 (32H, m), 0.89 (9H, s), 0.89 (3H, d, J6.5 Hz), 0.851

(3H, d, J6.6 Hz), 0.846 (3H, d, J6.6 Hz), 0.82 (3H, s),

0.65 (3H, s), 0.14 (3H, s), 0.12 (3H, s). ¹³C NMR (CDCl₃,

100 MHz) d: 83.4, 74.8, 71.0, 56.5, 55.4, 52.8, 48.2, 43.6,

42.0, 40.0, 39.5, 37.3, 36.2, 35.8, 35.1, 32.8, 30.8, 28.4,

28.0, 27.5, 26.8, 23.9, 22.8, 22.6, 21.8, 19.0, 18.8, 13.6,

12.4, 21.4, 22.3. FTIR (thin ®lm): 3391, 2853, 1472,

1385, 1252, 1096 cm²¹. High-resolution MS (EI, m/z):

535.4546, calcd for C₃₃H₆₃O₃Si (M1H)¹535.4546.

1470, 1445, 1375, 1368, 1090, 1053, 737 cm²¹

.

5.1.29. 3b,6a,7b-Triacetyloxy-5a-cholestane (57). To a

stirring solution of the triol 56 (21 mg, 0.050 mmol) in

dichloromethane (1 mL) at 258C was added acetic

anhydride (21 mL, 0.225 mmol) followed by trimethylsilyl

tri¯ate (1 mL) and the reaction mixture was stirred for

30 min. Saturated sodium bicarbonate was added and the

mixture allowed to stir for an additional 30 min. The

mixture was extracted with dichloromethane (3£10 mL)

and the combined organic extracts were dried over mag-

nesium sulfate, ®ltered, and concentrated under reduced

pressure. Flash column chromatography of the residue

(silica gel, 5:1 hexanes:ethyl acetate) yielded the triacetate

57 (27 mg, 100%) as a white solid. ¹H NMR (CDCl₃,

400 MHz) d: 4.84 (1H, dd, J10.0, 10.0 Hz), 4.66 (1H,

dd, J10.0, 10.0 Hz), 4.53 (1H, dddd, J11.3, 11.3, 4.9,

4.9 Hz), 2.20±0.50 (27H, m), 2.01 (3H, s), 1.99 (3H, s), 1.95

(3H, s), 0.95 (3H, s), 0.89 (3H, d, J6.5 Hz), 0.852 (3H, d,

J6.6 Hz), 0.847 (3H, d, J6.6 Hz), 0.66 (3H, s). ¹³C NMR

(CDCl₃, 100 MHz) d: 170.9, 170.8, 170.5, 77.9, 74.5, 72.7,

55.0, 54.7, 51.7, 46.0, 43.6, 39.4, 39.4, 39.1, 36.7, 36.1,

35.8, 35.5, 28.4, 28.3, 28.0, 27.0, 24.8, 23.7, 22.8, 22.6,

21.5, 21.4, 21.3, 20.9, 18.8, 13.3, 12.0. FTIR (thin ®lm):

2953, 2869, 1746, 1468, 1449, 1375, 1248, 1240,

54: ¹H NMR (CDCl₃, 400 MHz) d: 3.53 (1H, m), 3.28 (1H,

dd, J10.3, 8.2 Hz), 3.08 (1H, dd, J8.3, 8.2 Hz), 2.20±

0.50 (32H, m), 0.91 (9H, s), 0.90 (3H, d, J6.5 Hz), 0.857

(3H, d, J6.6 Hz), 0.852 (3H, d, J6.6 Hz), 0.83 (3H, s),

0.67 (3H, s), 0.12 (3H, s), 0.09 (3H, s). ¹³C NMR (CDCl₃,

100 MHz) d: 80.6, 71.3, 56.2, 55.4, 52.3, 48.1, 43.6, 40.9,

40.0, 39.5, 37.5, 36.2, 35.8, 35.7, 33.4, 31.0, 28.6, 28.0,

27.0, 26.2, 23.9, 22.8, 22.6, 21.5, 18.8, 18.4, 13.7, 12.3,

23.4, 23.8. FTIR (thin ®lm): 3613, 3384, 2951, 2857,

1471, 1253, 1125, 1092 cm²¹

.

5.1.27. 3b,7b-Diacetyloxycholest-5-ene (55). To a stirring

solution of the allylic alcohol 51 (1.314 g, 2.955 mmol) in

pyridine (10 mL) at 258C was added acetic anhydride

(0.418 mL, 4.432 mmol) followed by DMAP (4 mg). The

reaction was stirred overnight and then ice chips were

added. Water was added and the mixture was extracted

with ether (3£20 mL). The combined organic extracts

were dried over magnesium sulfate, ®ltered and concen-

trated under reduced pressure. The excess pyridine was

removed by a high-vacuum rotovaporator to provide the

1034 cm²¹

.

5.1.30. 6a,7b-Diacetyloxy-5a-cholestan-3b-ol (58). To a

stirring solution of the triacetate 57 (10 mg, 0.018 mmol) in

THF:ethanol (1:2, 1 mL) was added 10% NaOH solution (4

drops, 0.055 mmol) and the reaction mixture was stirred at

08C for 1 h. 10% HCl was added and the mixture was

extracted with ether (3£2 mL). The combined organic

extracts were dried over magnesium sulfate, ®ltered, and

concentrated under reduced pressure. Flash column chroma-

tography of the residue (silica gel, 1:1:0.25 hexanes:ethyl

acetate:chloroform) yielded the alcohol 58 (9 mg, 92%). ¹H

NMR (CDCl₃, 400 MHz) d: 4.83 (1H, dd, J11.2, 9.3 Hz),

4.72 (1H, dd, J9.9, 9.3 Hz), 3.51 (1H, dddd, J11.1, 11.1,

4.6, 4.6 Hz), 2.30±0.50 (28H, m), 1.98 (3H, s), 1.94 (3H, s),

0.93 (3H, s), 0.88 (3H, d, J6.5 Hz), 0.840 (3H, d, J

6.6 Hz), 0.835 (3H, d, J6.6 Hz), 0.65 (3H, s). ¹³C NMR

1

allylic acetate 55 (1.343 g, 93%) as a clear glass. H NMR

(CDCl₃, 400 MHz) d: 5.19 (1H, bs), 4.98 (1H, d, J8.6 Hz),

4.54 (1H, m), 2.40±2.20 (2H, m), 2.10±0.50 (24H, m), 1.97

(3H, s), 1.96 (3H, s), 1.03 (3H, s), 0.86 (3H, d, J6.5 Hz),

0.814 (3H, d, J6.6 Hz), 0.810 (3H, d, J6.6 Hz), 0.64 (3H,

s). ¹³C NMR (CDCl₃, 100 MHz) d: 170.9, 170.2, 144.1,

122.3, 75.4, 73.2, 55.5, 55.4, 48.1, 42.8, 39.4, 39.3, 37.5,

36.5, 36.4, 36.1, 35.6, 28.4, 28.0, 27.6, 25.1, 23.8, 22.8,

22.5, 21.6, 21.3, 21.0, 19.0, 18.7, 11.8 (1 high-®eld carbon

unresolved). FTIR (thin ®lm): 2948, 2870, 2857, 1736,

1732 cm²¹

.

5.1.28. 5a-Cholestan-3b,6a,7b-triol (56). To a stirring

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

1469

(CDCl₃, 100 MHz) d: 171.0, 170.8, 78.0, 74.6, 70.7, 55.0,

54.8, 51.8, 46.2, 43.6, 39.44, 39.43, 39.1, 37.0, 36.1, 35.8,

35.5, 32.1, 31.0, 28.4, 28.0, 24.9, 23.7, 22.8, 22.6, 21.5,

21.3, 20.8, 18.8, 13.4, 12.0. FTIR (thin ®lm): 3509, 3195,

62 (93 mg, 0.407 mmol) in 4 mL of pyridine cooled to

08C was added pivaloyl chloride (50 mL, 0.407 mmol) drop-

wise. The reaction mixture was warmed to 258C and stirred

for an additional 4 h. Sodium bicarbonate solution was

added and the mixture was extracted with ether

(3£10 mL). The organic layers were combined, dried over

anhydrous magnesium sulfate, ®ltered, and evaporated

under reduced pressure. Excess pyridine was removed by

a high-vacuum rotovaporator. Flash column chromato-

graphy of the crude solid (silica gel, 6:1 hexanes:ethyl

acetate) gave the alcohol 63 (94 mg, 74%) as a white

2952, 2869, 1757, 1740, 1468, 1447, 1377, 1235 cm²¹

.

5.1.31. 6a,7b-Diacetyloxy-3a-((4-benzoyl)phenylacetyl-

oxy)-5a-cholestane (59). To a stirring solution of the

bis-protected triol 58 (37 mg, 0.073 mmol) in 4 mL of

THF at 258C was added triphenylphosphine (39 mg,

0.147 mmol) and the carboxylic acid 16 (21 mg,

0.088 mmol) followed by the dropwise addition of DEAD

(23 mL, 0.147 mmol) and the solution was stirred for 5 min.

The reaction mixture was quenched with water and

extracted with ether (3£5 mL). The organic layers were

combined and dried over anhydrous magnesium sulfate,

®ltered, and evaporated under reduced pressure to give a

pale orange oil. Flash column chromatography of the

crude oil (silica gel, 3:1 hexanes:ethyl acetate) gave the

1

solid. H NMR (CDCl₃, 400 MHz) d: 7.50±7.00 (9H, m),

5.79 (1H, s), 4.24 (2H, t, J6.9 Hz), 2.92 (2H, t, J6.9 Hz),

2.61 (1H, bs), 1.16 (9H, s). ¹³C NMR (CDCl₃, 100 MHz) d:

178.6, 144.0, 142.2, 137.3, 129.1, 128.5, 127.5, 126.7,

126.6, 76.0, 64.8, 38.7, 34.8, 27.2. FTIR (thin ®lm): 3451,

3029, 2973, 2936, 2909, 2872, 1727, 1482, 1287,

1159 cm²¹

.

1

ester 59 (46 mg, 87%) as a white solid. H NMR (CDCl₃,

5.1.34. 3a,6a-Bis[((1,1-dimethyl)ethyl)dimethylsilyloxy]-

5a-cholest-14-en-7b-ol. To a stirring solution of the triol

61 (24 mg, 0.057 mmol) in dichloromethane (1 mL) cooled

to 08C was added 2,6-lutidine (53 mL, 0.459 mmol)

followed by tert-butyldimethylsilyl tri¯ate (79 mL,

0.342 mmol). The mixture was warmed to 258C and stirred

overnight. Water was added and the mixture was extracted

with dichloromethane (3£2 mL). The combined extracts

were dried over magnesium sulfate, ®ltered, and concen-

trated under reduced pressure to yield an oil. Flash column

chromatography (silica gel, 4:1 hexanes:benzene) of the

crude oil provided the bis-silyl ether (24 mg, 65%) as a

400 MHz) d: 7.79 (4H, m), 7.57 (1H, m), 7.47 (2H, m), 7.40

(2H, m), 5.06 (1H, m), 4.80 (1H, dd, J10.9, 9.3 Hz), 4.73

(1H, dd, J9.8, 9.3 Hz), 3.70 (2H, s), 1.97 (3H, s), 1.95 (3H,

s), 1.90±0.60 (27H, m), 0.89 (3H, s), 0.88 (3H, d,

J5.4 Hz), 0.854 (3H, d, J6.6 Hz), 0.850 (3H, d,

J6.6 Hz), 0.64 (3H, s). ¹³C NMR (CDCl₃, 100 MHz) d:

196.0, 170.85, 170.76, 170.0, 139.0, 137.6, 136.4, 132.4,

130.4, 130.0, 129.3, 128.3, 77.8, 74.6, 69.5, 55.0, 54.8,

52.0, 43.5, 41.8, 43.6, 39.5, 39.4, 39.0, 36.1, 36.0, 35.5,

32.8, 28.5, 28.0, 27.2, 25.5, 24.9, 23.7, 22.8, 22.6, 21.5,

20.8, 20.7, 18.8, 12.5, 12.0. FTIR (thin ®lm): 2951, 2869,

1742, 1659, 1607, 1377, 1277, 1246, 1026 cm²¹

.

clear glass. H NMR (CDCl₃, 400 MHz) d: 5.57 (1H, s),

1

4.02 (1H, m), 3.49 (1H, dd, J9.7, 8.3 Hz), 3.30 (1H, dd,

J11.1, 8.2 Hz), 2.50±0.50 (37H, m), 0.91 (9H, s), 0.86

(9H, s), 0.80 (3H, s), 0.12 (3H, s), 0.06 (3H, s), 0.01 (3H,

s), 0.00 (3H, s). ¹³C NMR (CDCl₃, 100 MHz) d: 151.8,

119.1, 76.3, 66.1, 58.4, 52.0, 47.6, 42.7, 41.7, 41.5, 39.6,

36.2, 36.1, 35.8, 33.9, 32.6, 31.5, 29.2, 28.0, 26.1, 25.8,

23.7, 22.8, 22.6, 22.1, 19.0, 18.4, 18.1, 17.0, 12.6, 23.6,

23.9, 24.9, 25.0 (1 high-®eld carbon unresolved). FTIR

(thin ®lm): 3615, 1930, 2892, 2857, 1472, 1464, 1252, 1059,

5.1.32. 4-(2-Hydroxyethyl)-a-phenylbenzenemethanol (62).

A solution of the benzophenone ester 59 (800 mg,

1.100 mmol) at 258C in degassed benzene (1100 mL) was

photolyzed using a 450 W mercury arc lamp for 10 h. The

solution was concentrated under reduced pressure and

subjected to ¯ash column chromatography (silica gel, 3:1

hexanes:ethyl acetate) to give the epimeric alcohols 60

(545 mg, 68%), which were carried on to the next step as

a mixture. To a stirring mixture of alcohols 60 (500 mg,

0.688 mmol) in THF (10 mL) at 258C was added lithium

aluminum hydride (261 mg, 6.88 mmol) in portions. The

mixture was stirred for 1 h. Water (261 mL) was added

slowly at 08C followed by 10% NaOH (261 mL), which

was followed by addition of water (1 mL) again. The

mixture was warmed to 258C and then stirred for 15 min.

Magnesium sulfate was added and the mixture was ®ltered

and concentrated under reduced pressure. Flash column

chromatography (silica gel, 5:4 hexanes:ethyl acetate

followed by 20:1 chloroform:methanol) gave the pure diol

62 (no yield recorded) followed by the desired ole®nic triol

61 (263 mg, 91%), which was not characterized at this step.

¹H NMR (CDCl₃, 400 MHz) d: 7.50±7.00 (9H, m), 5.76

(1H, s), 3.75 (2H, t, J6.9 Hz), 2.79 (2H, t, J6.9 Hz),

2.27 (1H, bs). ¹³C NMR (CDCl₃, 100 MHz) d: 143.9,

142.1, 137.8, 129.2, 128.5, 127.5, 126.8, 126.5, 76.0, 63.5,

38.8. FTIR (thin ®lm): 3335, 3090, 3031, 2930, 2874, 1453,

835, 775 cm²¹

.

5.1.35. 3a,6a-Bis(triethylsilyloxy)-5a-cholest-14-en-7b-

ol. To a stirring solution of the triol 61 (242 mg,

0.578 mmol) in dichloromethane (5 mL) cooled to 08C

was added 2,6-lutidine (404 mL, 3.468 mmol) followed by

triethylsilyl tri¯ate (523 mL, 2.312 mmol). The mixture was

stirred at 258C overnight. Water was added and the mixture

extracted with dichloromethane (3£10 mL). The combined

organic layers were dried over magnesium sulfate, ®ltered,

and concentrated under reduced pressure to yield an oil.

Flash column chromatography (silica gel, 3:1 hexanes:

benzene) of the crude oil provided the bis-silyl ether

(352 mg, 80%) as a clear glass. ¹H NMR (CDCl₃,

400 MHz) d: 5.53 (1H, s), 4.05 (1H, m), 3.48 (1H, dd,

J9.7, 8.3 Hz), 3.35 (1H, dd, J11.1, 8.2 Hz), 2.29 (1H,

ddd, J15.2, 7.8, 2.1 Hz), 2.22 (1H, m), 2.15±0.50 (47H,

m), 0.867 (3H, d, J6.6 Hz), 0.865 (3H, d, J6.6 Hz), 0.79

(3H, s), 0.65 (6H, q, J 7.8 Hz), 0.55 (6H, q, J8.0 Hz). ¹³C

NMR (CDCl₃, 100 MHz) d: 151.7, 119.0, 77.2, 76.5, 65.9,

58.4, 52.0, 47.6, 42.8, 41.8, 41.5, 39.6, 36.3, 36.1, 35.8,

33.9, 32.6, 31.2, 29.3, 28.0, 23.8, 22.8, 22.6, 22.1, 19.0,

1493, 1043, 1017 cm²¹

.

5.1.33. 4-(2-(1,1-Dimethylpropanoyloxy)ethyl)-a-phenyl-

benzenemethanol (63). To a stirring solution of the diol

1470

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

17.0, 12.6, 7.1, 7.0, 5.5, 4.9. FTIR (thin ®lm): 3615, 2955,

2932, 2911, 2876, 1466, 1372, 1238, 1100, 1007 cm²¹

stirred for 15 h. Celite (150 mg) was added followed by

ether (5 mL). The mixture was ®ltered through Celite and

the ®ltrate was concentrated under reduced pressure. Flash

column chromatography (silica gel, 2:1 hexanes:ethyl

acetate) of the crude residue afforded the diketone 65

(29 mg, 33%) as a clear glass. ¹H NMR (CDCl₃,

400 MHz) d: 5.089 (1H, d, J12.5 Hz), 5.087 (1H, m),

2.70 (1H, dd, J11.9, 11.0 Hz), 2.57 (1H, dd, J17.1,

7.6 Hz), 2.43 (1H, d, J10.6 Hz), 2.20±0.60 (22H, m),

2.14 (3H, s), 2.02 (3H, s), 1.15 (3H, s), 0.97 (3H, d, J

6.5 Hz), 0.853 (3H, d, J6.6 Hz), 0.848 (3H, d, J

6.6 Hz), 0.70 (3H, s). ¹³C NMR (CDCl₃, 100 MHz) d:

211.4, 203.41, 170.4, 170.0, 76.7, 68.4, 58.7, 54.7, 52.0,

46.8, 42.9, 41.2, 41.1, 39.2, 38.5, 36.9, 35.9, 35.5, 32.7,

28.3, 27.9, 25.6, 23.8, 22.8, 22.5, 21.4, 21.0, 20.7, 19.0,

12.8, 12.1. FTIR (thin ®lm): 2951, 2869, 1754, 1732,

.

5.1.36. 3a,6a,7b-Triacetyloxy-5a-cholest-14-ene (64a).

To a stirring solution of the triol ole®n 61 (90 mg,

0.215 mmol) and acetic anhydride (91 mL, 967 mmol) in

dichloromethane (5 mL) at 258C was added trimethylsilyl

tri¯ate (1 mL) and the reaction was allowed to stir for

10 min. Saturated sodium bicarbonate was added and the

mixture was extracted with dichloromethane (3£5 mL).

The organic layers were dried over magnesium sulfate,

®ltered, and concentrated under reduced pressure. Flash

column chromatography (silica gel, 6:1 hexanes:ethyl

acetate) of the residue provided the triacetate 64a (94 mg,

93%) as a white solid. ¹H NMR (CDCl₃, 400 MHz) d: 5.08

(1H, s), 5.05 (2H, m), 4.87 (1H, dd, J11.6, 9.2 Hz), 2.37

(1H, dd, J11.9, 11.9 Hz), 2.24 (1H, ddd, J15.8, 7.2,

2.2 Hz), 2.15±0.50 (22H, m), 2.00 (3H, s), 1.98 (3H, s),

1.93 (3H, s), 0.91 (3H, s), 0.88 (3H, s), 0.87 (3H, d,

J7.2 Hz), 0.829 (3H, d, J6.6 Hz), 0.826 (3H, d,

J6.6 Hz). ¹³C NMR (CDCl₃, 100 MHz) d: 170.91,

170.86, 170.4, 149.4, 118.1, 75.5, 74.6, 68.5, 58.3, 51.9,

47.6, 42.5, 40.7, 40.0, 39.4, 36.1, 35.9, 33.8, 32.7, 28.0,

27.1, 25.4, 23.8, 22.8, 22.5, 22.2, 21.4, 21.2, 20.8, 19.0,

16.8, 12.3 (1 high-®eld carbon unresolved). FTIR (thin

1366, 1375, 1242, 1211, 1022 cm²¹

.

5.1.39. 3a,6a-Benzoyloxy-5a-cholestane-7b,15a-diol

(66). To a stirring solution of the tribenzoate ole®n 64b

(48 mg, 0.066 mmol) in THF (2 mL) at 08C was added

borane±tetrahydrofuran

complex

(1.0 M,

263 mL,

0.263 mmol) and the reaction was warmed to 258C and

stirred for 3 h. The solution was cooled again to 08C and

water was added dropwise. The mixture was extracted with

ether (3£5 mL) and the combined organic extracts were

concentrated under reduced pressure. The crude oil obtained

was re-dissolved in THF (3 mL) and water (1 mL) at 258C

and sodium perborate (20 mg, 0.132 mmol) was added and

the mixture stirred overnight. Water was added and the

mixture was extracted with ether (3£5 mL). The combined

organic extracts were dried over magnesium sulfate,

®ltered, and concentrated under reduced pressure. Flash

column chromatography (silica gel, 3:1 hexanes:ethyl

acetate) of the crude residue afforded the diol 66 (no yield

®lm): 2953, 2934, 2870, 1744, 1364, 1246, 1024, 733 cm²¹

.

5.1.37. 3a, 6a,7b-Tribenzoyloxy-5a-cholest-14-ene (64b).

To a stirring solution of the triol ole®n 61 (31 mg,

0.074 mmol) and benzoic anhydride (75 mL, 333 mmol) in

dichloromethane (2 mL) at 258C was added trimethylsilyl

tri¯ate (1 mL) and the reaction was allowed to stir for 1 h.

Saturated sodium bicarbonate was added and the mixture

was extracted with dichloromethane (3£5 mL). The organic

layers were dried over magnesium sulfate, ®ltered, and

concentrated under reduced pressure. Flash column

chromatography (silica gel, 6:1 hexanes:ethyl acetate) of

the residue provided the tribenzoate 64b (49 mg, 91%) as

1

recorded) as a clear glass. H NMR (CDCl₃, 400 MHz) d:

8.20±7.90 (4H, m), 7.30±7.70 (6H, m), 5.31 (1H, m), 5.04

(1H, dd, J11.3, 9.1 Hz), 4.12 (1H, m), 3.62 (1H, bt), 2.10±

0.75 (41H, m), 1.02 (3H, s), 0.91 (3H, d, J5.9 Hz), 0.864

(3H, d, J6.6 Hz), 0.861 (3H, d, J6.6 Hz), 0.76 (3H, s).

¹³C NMR (CDCl₃, 100 MHz) d: 168.1, 165.7, 133.3, 132.9,

130.9, 129.8, 129.7, 129.5, 128.5, 77.8, 72.8, 69.4, 62.6,

53.4, 51.3, 44.4, 42.5, 41.9, 39.0, 39.4, 38.5, 36.4, 36.0,

35.5, 33.3, 29.7, 28.0, 27.8, 25.8, 23.7, 22.8, 22.6, 20.6,

18.7, 13.6, 12.7. FTIR (thin ®lm): 3366, 2951, 2870, 1717,

1

a white solid. H NMR (C₆D₆, 400 MHz) d: 8.40±8.00

(6H, m), 7.30±6.80 (9H, m), 6.07 (1H, dd, J10.2,

9.5 Hz), 5.81 (1H, dd, J11.3, 10.1 Hz), 5.77 (1H, s),

5.32 (1H, m), 2.86 (1H, m), 2.15±0.50 (23H, m), 1.01

(3H, d, J5.9 Hz), 0.99 (3H, s), 0.98 (6H, d, J6.6 Hz),

0.86 (3H, s). ¹³C NMR (CDCl₃, 100 MHz) d: 166.65,

166.55, 165.6, 149.3, 132.9, 132.8, 132.6, 130.8, 130.4,

129.8, 129.6, 129.5, 128.4, 128.2, 128.1, 118.7, 75.9,

75.3, 69.1, 58.4, 52.4, 47.8, 42.6, 41.9, 40.5, 39.4,

36.6, 36.0, 35.9, 33.8, 33.2, 28.0, 27.7, 25.6, 23.8,

22.8, 22.6, 22.4, 19.1, 16.9, 12.7 (1 aromatic carbon

unresolved). FTIR (thin ®lm): 2953, 2932, 2869, 1721,

1314, 1279, 1113, 710 cm²¹

.

5.1.40. 3a,6a-Benzoyloxy-15a-hydroxy-5a-cholestan-

7-one (67). To a stirring solution of the diol 66 (10 mg,

0.013 mmol) in dichloromethane (3 mL) at 08C was added

pyridinium chlorochromate (0.003, 0.013 mmol) in smaller

portions over 1 h. Ether (3 mL) was added followed by

Celite (10 mg) and the reaction was ®ltered through Celite.

The ®ltrate was concentrated under reduced pressure and the

crude oil was subjected to ¯ash column chromatography

(silica gel, 5:1 hexanes:ethyl acetate) to provide the pure

ketone 67 (no yield recorded, although the reaction was

1451, 1275, 1113, 710 cm²¹

.

5.1.38. 3a,6a-Diacetyloxy-5a-cholestane-7,15-dione (65).

To a stirring solution of the triacetate ole®n 64a (94 mg,

0.173 mmol) in THF (5 mL) at 08C was added borane±tetra-

hydrofuran complex (1.0 M, 692 mL, 0.692 mmol) and the

reaction was warmed to 258C and stirred for 3 h. The solu-

tion was cooled again to 08C and water was added dropwise.

The mixture was extracted with ether (3£5 mL) and the

combined organic extracts were concentrated under reduced

pressure. The crude oil obtained was re-dissolved in

dichloromethane (5 mL) at 258C and pyridinium chloro-

chromate (149 mg, 0.692 mmol) was added and the mixture

1

very clean by TLC) as a clear glass. H NMR (CDCl₃,

400 MHz) d: 8.20±7.90 (4H, m), 7.30±7.70 (6H, m), 5.40

(1H, m), 5.29 (1H, d, J13.0 Hz), 3.93 (1H, ddd, J8.6,

8.6, 2.9 Hz), 2.83 (1H, dd, J11.1, 11.0 Hz), 2.10±0.75

(39H, m), 1.25 (3H, s), 0.91 (3H, d, J6.5 Hz), 0.859

(3H, d, J6.6 Hz), 0.856 (3H, d, J6.6 Hz), 0.75 (3H, s).

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

1471

5.1.41. 3a-Methoxymethoxy-6a,7b-methylenedioxy-5a,

14a-cholestan-15a-ol (69) and 3a-methoxymethoxy-

6a,7b-methylenedioxy-5a,14b-cholestan-15b-ol (70).

To a stirring solution of the ole®nic triol 61 (86 mg,

0.205 mmol) in dimethoxymethane (5 mL) at 258C was

added excess phosphorus pentoxide. The solution was

stirred overnight, or until the reaction was judged to be

complete by TLC. Saturated sodium carbonate solution

was added and the mixture was extracted with ether

(3£10 mL). The combined organic extracts were dried

over magnesium sulfate, ®ltered, and concentrated under

reduced pressure. Flash column chromatography (silica

gel, 8:1 hexanes:ethyl acetate) of the crude product provided

the protected ole®n 68 (62 mg, 63%). To a stirring solution

of the steroidal ole®n 68 (62 mg, 0.131 mmol) in 3 mL of

THF cooled to 2788C was added borane±tetrahydrofuran

complex (1.0 M, 393 mL, 0.393 mmol). The reaction

mixture was allowed to warm to 258C and stirred for 12 h.

The reaction mixture was cooled to 08C and 393 mL of 10%

NaOH was added dropwise followed by 393 mL of 30%

hydrogen peroxide. The reaction was stirred vigorously

for 1 h. Water was added and the solution was extracted

with ether (3£5 mL). The organic layers were combined,

dried over anhydrous magnesium sulfate, ®ltered, and

evaporated under reduced pressure to give a white solid.

Flash column chromatography of the crude solid (silica

gel, 3.5:1 hexanes:ethyl acetate) gave the alcohol 69

(30 mg, 75%) as a white solid and the alcohol 70 (6 mg,

13%) as a white solid.

dried over anhydrous magnesium sulfate, ®ltered and

concentrated under reduced pressure. Flash column chroma-

tography of the crude solid (silica gel, 3:1 hexanes:ethyl

acetate) gave the ketone 71 (40 mg, 80%) as a white solid.

¹H NMR (CDCl₃, 400 MHz) d: 5.19 (1H, s), 5.04 (1H, s),

4.63 (1H, d, J 6.9 Hz), 4.61 (1H, d, J6.9 Hz), 3.88 (1H,

m), 3.33 (3H, s), 3.19 (1H, dd, J11.5, 8.5 Hz), 2.94 (1H,

dd, J9.3, 8.7 Hz), 2.50 (1H, dd, J18.4, 9.3 Hz), 2.20±

0.50 (24H, m), 0.96 (3H, d, J6.5 Hz), 0.837 (3H, d,

J6.5 Hz), 0.840 (3H, s), 0.832 (3H, d, J6.4 Hz), 0.73

(3H, s). ¹³C NMR (CDCl₃, 100 MHz) d: 212.3, 95.6, 94.4,

85.8, 78.6, 69.8, 64.5, 55.2, 52.3, 51.6, 42.1, 41.4, 41.1,

39.7, 39.2, 37.1, 35.8, 35.5, 34.5, 33.3, 28.1, 27.9, 26.3,

23.9, 22.8, 22.5, 20.0, 19.2, 13.4, 13.1. FTIR (thin ®lm):

2944, 2938, 1752, 1468, 1366, 1040 cm²¹

.

5.1.43. 3a-Methoxymethoxy-6a,7b-methylenedioxy-5a,

14b-cholestan-15-one (72). To a stirring solution of the

14a,15-ketosteroid 71 (40 mg, 0.082 mmol) in 1:1

THF:ethanol (4 mL) at 258C was added 10% NaOH

(0.5 mL) and the reaction mixture was stirred for 4 h.

Water was added and the mixture was extracted with ether

(3£5 mL). The combined organic extracts were washed

with brine, dried over anhydrous magnesium sulfate,

®ltered, and evaporated under reduced pressure to yield

the 14b,15-ketosteroid 72 (40 mg, 100%) as a clear glass,

which was used without further puri®cation. ¹H NMR

(CDCl₃, 400 MHz) d: 5.16 (1H, s), 5.11 (1H, s), 4.65 (1H,

d, J6.9 Hz), 4.63 (1H, d, J6.9 Hz), 4.26 (1H, dd, J11.2,

8.8 Hz), 3.88 (1H, m), 3.36 (3H, s), 2.97 (1H, dd, J11.4,

8.8 Hz), 2.57 (1H, m), 2.37 (1H, dd, J20.0, 10.1 Hz), 2.21

(1H, d, J20.0 Hz), 2.10±0.70 (22H, m), 1.18 (3H, s), 0.87

(3H, d, J6.5 Hz), 0.856 (3H, d, J6.4 Hz), 0.852 (3H, d,

J6.8 Hz), 0.82 (3H, s). ¹³C NMR (CDCl₃, 100 MHz) d:

219.4, 96.2, 94.6, 80.2, 79.9, 70.0, 55.4, 53.1, 48.8, 46.6,

41.5, 40.4, 39.3, 38.4, 37.9, 37.4, 36.4, 33.2, 32.9, 31.9,

28.3, 28.1, 26.4, 25.7, 22.9, 22.7, 21.1, 19.4, 19.1, 13.8.

FTIR (thin ®lm): 2934, 2872, 1736, 1468, 1368, 1385,

69: ¹H NMR (CDCl₃, 400 MHz) d: 5.118 (1H, d, J0.6 Hz),

5.036 (1H, d, J0.6 Hz), 4.64 (2H, s), 4.24 (1H, bs), 3.92

(2H, m), 3.35 (3H, s), 3.13 (1H, dd, J11.0, 8.7 Hz), 3.06

(1H, dd, J8.7, 8.7 Hz), 2.20±0.50 (25H, m), 0.88 (3H, d,

J6.2 Hz), 0.87 (3H, s), 0.849 (3H, d, J6.6 Hz), 0.846

(3H, d, J6.6 Hz), 0.72 (3H, s). ¹³C NMR (CDCl₃,

100 MHz) d: 95.3, 94.5, 84.7, 79.2, 72.3, 69.9, 61.9, 55.2,

53.7, 52.2, 43.4, 40.9, 40.0, 39.4, 38.0, 37.8, 37.3, 36.0,

35.5, 33.2, 28.0, 28.0, 26.2, 23.6, 22.8, 22.6, 20.3, 18.6,

13.7, 13.3. FTIR (thin ®lm): 3503, 2946, 2940, 2878,

1089 cm²¹

.

1101, 1080, 1042 cm²¹

.

5.1.44. 3a,6a,7b-Trihydroxy-5a,14b-cholestan-15-one

(73). To a stirring solution of the 14b,15-ketosteroid 72

(40 mg, 0.082 mmol) in THF (4 mL) was added 50%

concentrated HCl (0.5 mL) and the reaction mixture was

stirred at 258C for 1 h. Water was added and the products

were thoroughly extracted with ethyl acetate (3£10 mL).

The combined organic extracts were dried over anhydrous

magnesium sulfate, ®ltered, and evaporated under reduced

pressure. Flash column chromatography (silica gel, 14:1

chloroform:methanol) of the residue yielded the ketone

1

70: H NMR (CDCl₃, 400 MHz) d: 5.12 (1H, s), 5.05 (1H,

s), 4.66 (1H, d, J6.9 Hz), 4.64 (1H, d, J6.9 Hz), 4.23

(1H, m), 3.91 (1H, m), 3.37 (3H, s), 3.13 (1H, dd, J11.2,

8.6 Hz), 3.04 (1H, dd, J10.3, 8.6 Hz), 2.99 (1H, bs), 2.20±

0.50 (25H, m), 0.94 (3H, s), 0.91 (3H, d, J7.7 Hz), 0.90

(3H, s), 0.864 (3H, d, J6.6 Hz), 0.860 (3H, d, J6.6 Hz).

¹³C NMR (CDCl₃, 100 MHz) d: 95.7, 94.5, 84.9, 79.4, 70.8,

69.9, 60.0, 56.2, 55.2, 53.0, 42.0, 40.9, 40.7, 39.4, 38.6,

37.5, 36.0, 35.1, 35.0, 33.2, 28.0, 26.2, 23.7, 22.8, 22.6,

20.7, 18.7, 14.0, 13.7 (1 high-®eld carbon unresolved).

FTIR (thin ®lm): 3542, 2934, 2890, 1468, 1443, 1385,

1

triol 73 (15 mg, 50%) as a white solid. H NMR (CDCl₃,

400 MHz) d: 4.27 (1H, dd, J9.9, 9.6 Hz), 4.13 (1H, m),

3.16 (1H, dd, J10.1, 9.4 Hz), 2.67 (1H, m), 2.41 (1H, dd,

J19.8, 10.2 Hz), 2.20 (1H, bd, J19.6 Hz), 2.10±0.70

(25H, m), 1.15 (3H, s), 0.87 (3H, d, J7.4 Hz), 0.857

(3H, d, J6.7 Hz), 0.853 (3H, d, J6.6 Hz), 0.79 (3H, s).

¹³C NMR (CDCl₃, 100 MHz) d: 222.1, 75.1, 74.8, 65.8,

52.5, 49.0, 46.4, 41.8, 41.7, 40.0, 38.7, 38.2, 38.1, 36.8,

33.3, 32.2, 31.9, 30.2, 28.3, 28.1, 25.7, 23.0, 22.7, 20.9,

19.5, 19.4, 12.6. FTIR (thin ®lm): 3453, 3359, 2951,

2936, 2872, 2859, 1736, 1453, 1385, 1020 cm²¹. High-reso-

lution MS (EI, m/z): 435.3476, calcd for C₂₇H₄₇O₄(M1H)¹

435.3474.

1042 cm²¹

.

5.1.42. 3a-Methoxymethoxy-6a,7b-methylenedioxy-5a,

14a-cholestan-15-one (71). To a stirring solution of the

alcohol 69 (50 mg, 0.101 mmol) in 4 mL of dichloro-

methane at 258C was added Dess±Martin periodinane

(86 mg, 0.202 mmol) in portions. The reaction mixture

was stirred for 5 h. Sodium bisulfate solution (10%

aqueous) was added and the mixture was extracted with

ether (3£10 mL). The organic layers were combined,

1472

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

5.1.45. 3a,5-Cyclo-6b-methoxy-24-norcholan-23-al (76).

To a stirring solution of the alcohol 75 (830 mg,

2.300 mmol) in 2% pyridine:dichloromethane (25 mL) at

258C was added pyridinium chlorochromate (1.751 g,

8.150 mmol) and the reaction mixture was stirred for 1 h.

Ether (25 mL) was added followed by Celite (25 g) and the

mixture was ®ltered. The ®ltrate was concentrated under

reduced pressure and ¯ash column chromatography (silica

gel, 14:1 hexanes:ethyl acetate) of the residue yielded the

to give a white solid. Flash column chromatography (silica

gel, 3:1:0.5 hexanes:ethyl acetate:chloroform) of the crude

solid gave the homoallylic alcohol 78 (860 mg, 94%) as a

1

white solid. H NMR (CDCl₃, 400 MHz) d: 5.34 (1H, d,

J10.0 Hz), 3.51 (1H, m), 2.50±0.60 (26H, m), 2.41 (1H,

dd, J16.1, 2.6 Hz), 1.01 (3H, s), 0.917 (3H, d, J6.2 Hz),

0.912 (3H, d, J6.5 Hz), 0.901 (3H, d, J6.5 Hz), 0.72 (3H,

s). ¹³C NMR (CDCl₃, 100 MHz) d: 211.3, 140.9, 121.3,

71.5, 56.8, 56.0, 52.5, 50.4, 50.0, 42.4, 42.2, 39.6, 37.3,

36.4, 32.5, 31.8 (2 carbons), 31.5, 28.4, 24.5, 24.2, 22.6,

22.5, 21.0, 19.8, 19.4, 11.9. FTIR (thin ®lm): 3351, 2936,

2870, 1709, 1468, 1368 cm²¹. High-resolution MS (EI,

m/z): 400.3343, calcd for C₂₇H₄₄O₂400.3341.

1

aldehyde 101 (1.280 g, 88%) as a clear glass. H NMR

(CDCl₃, 400 MHz) d: 9.69 (1H, d, J2.2 Hz), 3.27 (3H,

s), 2.72 (1H, dd, J2.5, 2.5 Hz), 2.40 (1H, dd, J15.7,

2.3 Hz), 2.10±0.70 (21H, m), 0.97 (3H, s), 0.96 (3H, d,

J5.9 Hz), 0.80 (3H, s), 0.60 (1H, dd, J4.5, 4.5 Hz),

0.38 (1H, dd, J7.9, 5.1 Hz). ¹³C NMR (CDCl₃,

100 MHz) d: 203.3, 82.3, 56.5, 56.4, 56.0, 50.9, 47.9,

43.3, 42.9, 40.1, 35.2, 35.0, 33.3, 31.6, 30.5, 28.5, 24.9,

24.1, 22.7, 21.4, 20.0, 19.2, 13.1, 12.2. FTIR (thin ®lm):

2936, 2869, 2712, 1727, 1456, 1383, 1098 cm²¹. High-reso-

lution MS (EI, m/z): 358.2869, calcd for C₂₄H₃₈O₂

328.2872.

5.1.48. 3a,5-Cyclo-23-(1,1-dimethylpropanoyloxy)-6b-

methoxy-24-norcholane (79). To a stirring solution of the

alcohol 75 (310 mg, 0.860 mmol) in pyridine (13 mL) at

258C was added pivaloyl chloride (381 mL, 2.580 mmol)

dropwise. The mixture was stirred for 24 h. Water was

added and the mixture was extracted with hexanes

(3£15 mL) and the combined extracts were washed with

saturated sodium bicarbonate. The organic layer was dried

over magnesium sulfate, ®ltered, and concentrated under

reduced pressure. Flash column chromatography (silica

5.1.46. 3a,5-Cyclo-6b-methoxycholestan-23-one (77). To

a stirring solution of the aldehyde 76 (1.151 g, 3.21 mmol)

in diethyl ether (20 mL) at 08C was added isobutylmagne-

sium bromide (2.0 M, 4.82 mL, 9.63 mmol) and the reaction

was stirred for 1.5 h. The mixture was quenched with a

saturated solution of ammonium chloride and extracted

with ether (3£20 mL). The combined organic extracts

were dried over magnesium sulfate, ®ltered, and concen-

trated under reduced pressure. Flash column chromato-

graphy (silica gel, 10:1 hexanes:ethyl acetate) of the

residue provided a 1:1 mixture of the diastereomeric alco-

hols (1.060 g, 79%). To a stirring solution of the alcohols

(976 mg, 2.342 mmol) in 2% pyridine:dichloromethane

(50 mL) at 258C was added pyridinium chlorochromate

(1.509 g, 7.028 mmol) and the reaction mixture was stirred

for 3 h. Ether (50 mL) was added followed by Celite (25 g)

and the mixture was ®ltered. The ®ltrate was concentrated

under reduced pressure and the crude oil was passed through

a short column of silica gel, eluting with ether. Concentra-

tion yielded the ketone 77 (0.970 g, 100%) as a clear glass.

¹H NMR (CDCl₃, 400 MHz) d: 3.25 (3H, s), 2.69 (1H, bdd,

J2.6, 2.6 Hz), 2.41 (1H, dd, J15.5, 2.5 Hz), 2.24 (1H, d,

J6.9 Hz), 2.15±0.65 (23H, m), 1.02 (3H, s), 0.840 (6H, d,

J6.6 Hz), 0.829 (3H, d, J6.6 Hz), 0.69 (3H, s), 0.57 (1H,

dd, J4.5, 4.5 Hz), 0.35 (1H, dd, J7.9, 5.1 Hz). ¹³C NMR

(CDCl₃, 100 MHz) d: 210.9, 82.3, 56.5 (2 carbons), 56.2,

52.4, 50.5, 47.9, 43.3, 42.8, 40.1, 35.2, 35.0, 33.3, 32.5,

30.4, 28.4, 24.9, 24.4, 24.1, 22.7, 22.6, 22.5, 21.4, 19.7,

19.2, 13.1, 12.2. FTIR (thin ®lm): 2953, 2870, 1713,

1470, 1372, 1100 cm²¹. High-resolution MS (EI, m/z):

414.3496, calcd for C₂₈H₄₆O₂414.3498.

gel, 10:1 hexanes:ether) of the residue provided the ester

Â Ã

25

79 (353 mg, 92%) as a clear glass. a 149.48 (c 1.5,

D

CH₂Cl₂). H NMR (CDCl₃, 400 MHz) d: 4.06 (2H, m),

1

3.28 (3H, s), 2.73 (1H, dd, J2.6, 2.6 Hz), 2.00±0.70

(22H, m), 1.16 (9H, s), 0.99 (3H, s), 0.93 (3H, d,

J6.6 Hz), 0.68 (3H, s), 0.61 (1H, dd, J4.3, 4.3 Hz),

0.39 (1H, dd, J5.1, 8.0 Hz). ¹³C NMR (CDCl₃,

100 MHz) d: 178.6, 82.4, 62.5, 56.5, 56.5, 56.2, 48.0,

43.4, 42.8, 40.2, 38.7, 35.3, 35.0, 34.6, 33.3, 33.2, 30.4,

28.3, 27.2, 25.0, 24.1, 22.7, 21.5, 19.3, 18.8, 13.1, 12.1.

FTIR (thin ®lm): 2953, 2870, 1732, 1100 cm²¹. High-reso-

lution MS (EI, m/z): 444.3594, calcd for C₂₉H₄₈O₃

444.3603.

5.1.49. 23-(1,1-Dimethylpropanoyloxy)-24-norcholan-5-

en-3b-ol (80). A mixture of the ether 79 (324 mg,

0.729 mmol) and toluenesulfonic acid monohydrate

(30 mg, 0.073 mmol) was re¯uxed in dioxane:water (3:1,

15 mL) for 3 h. The solution was cooled to 258C and satu-

rated sodium bicarbonate was added followed by water. The

mixture was extracted with ether (3£15 mL). The combined

organic extracts were dried over magnesium sulfate,

®ltered, and concentrated under reduced pressure. Flash

column chromatography of the residue (silica gel, 4:1:1

hexanes:ethyl acetate:chloroform) provided the homoallylic

alcohol 80 (305 mg, 97%) as a white solid (mp 1488C).

[a]²_D⁵238.08 (c 1.0, CH₂Cl₂). ¹H NMR (CDCl₃,

400 MHz) d: 5.27 (1H, d, J4.9 Hz), 4.02 (2H, m), 3.42

(1H, m), 2.64 (1H, s), 2.30±0.50 (23H, m), 1.12 (9H, s), 0.94

(3H, s), 0.91 (3H, d, J6.2 Hz), 0.62 (3H, s). ¹³C NMR

(CDCl₃, 100 MHz) d: 178.6, 140.9, 121.4, 77.4, 71.5,

62.6, 56.7, 56.0, 50.0, 42.4, 42.2, 39.7, 38.6, 37.3, 36.4,

34.5, 33.1, 31.8, 31.5, 28.2, 27.2, 24.2, 21.0, 19.4, 18.8,

11.7. FTIR (thin ®lm): 3482, 2969, 2930, 2896, 2863,

2836, 1707, 1292, 1177, 1067 cm²¹. High-resolution MS

(EI, m/z): 430.3444, calcd for C₂₈H₄₆O₃430.3447.

5.1.47. 3b-Hydroxycholest-5-en-23-one (78). To a stirring

solution of the ketone 77 (950 mg, 2.29 mmol) in 30 mL of

dioxane and 10 mL of water at 08C was added toluene-

sulfonic acid monohydrate (50 mg) and the solution was

re¯uxed for 3 h. The reaction mixture was allowed to cool

to 258C and saturated sodium bicarbonate was added and the

mixture was extracted with ether (3£5 mL). The organic

layers were combined, dried over anhydrous magnesium

sulfate, ®ltered and concentrated under reduced pressure

5.1.50. 23-(1,1-Dimethylpropanoyloxy)-3b-phenylmethoxy-

24-norchol-5-ene (81). To a stirring solution of the alcohol

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

1473

80 (278 mg, 0.646 mmol) in tetrahydrofuran (5 mL) at 258C

was added sodium hydride (104 mg, 2.584 mmol) and

127.4, 77.8, 72.6, 70.0, 62.5, 56.2, 56.0, 53.7, 48.6, 42.7,

39.7, 38.7, 37.7, 37.1, 36.9, 34.5, 34.1, 33.1, 29.1, 28.2,

27.8, 27.2, 24.0, 21.3, 21.1, 18.8, 13.3, 11.9. FTIR (thin

tetrabutylammonium

iodide

(48 mg,

0.130 mmol)

followed by benzyl bromide (184 mL, 1.550 mmol). The

reaction was stirred for 72 h. Water was added and the

mixture was extracted with ether (3£5 mL). The combined

organic layers were dried over magnesium sulfate,

®ltered, and concentrated under reduced pressure. Column

chromatography (silica gel, 15:1 hexanes:ether) of the resi-

®lm): 2942, 2869, 1736, 1242, 1159 cm²¹

.

5.1.53. 6a-Acetyloxy-23-(1,1-dimethylpropanoyloxy)-5a-

24-norcholan-3b-ol (84). To a stirring solution of the

benzyl ether 83 (165 mg, 0.284 mmol) in ethanol (12 mL)

at 258C was added Pt (5 mg) and the ¯ask was evacuated

using a high-vacuum pump. The ¯ask was back-®lled with

hydrogen gas from a balloon. This process of evacuating

and back-®lling was repeated two times and the reaction

was stirred for 7 h. The solution was ®ltered through Celite

and evaporated under reduced pressure to yield a crude oil.

Flash column chromatography (silica gel, 1:1 hexanes: ethyl

acetate) of the residue provided the alcohol 84 (134 mg,

96%) as a white solid (mp 1488C). ¹H NMR (CDCl₃,

400 MHz) d: 4.62 (1H, ddd, J10.9, 10.9, 4.6 Hz), 4.04

(2H, m), 3.49 (1H, m), 2.10±0.50 (24H, m), 2.03 (1H, bs),

1.98 (3H, s), 1.14 (9H, s), 0.91 (3H, d, J6.6 Hz), 0.82 (3H,

s), 0.61 (3H, s). ¹³C NMR (CDCl₃, 100 MHz) d: 178.7,

170.8, 72.5, 70.8, 62.5, 56.2, 56.0, 53.6, 48.6, 42.7, 39.7,

38.7, 37.6, 37.2, 36.6, 34.5, 34.1, 33.1, 32.2, 31.1, 28.2,

27.2, 24.0, 21.2, 21.1, 18.7, 13.3, 11.9. FTIR (thin ®lm):

due yielded the benzyl ether 81 (280 mg, 83%) as a white

Â Ã

25

D

solid (mp 1088C). a 223.98 (c 0.95, CH₂Cl₂). ¹H NMR

(CDCl₃, 400 MHz) d: 7.40±7.20 (5H, m), 5.36 (1H, d,

J5.2 Hz), 4.57 (2H, s), 4.11 (2H, m), 3.29 (1H, m), 2.45

(1H, m), 2.30 (1H, m), 2.10±0.50 (21H, m), 1.21 (9H, s),

1.03 (3H, s), 0.99 (3H, d, J6.6 Hz), 0.70 (3H, s). ¹³C NMR

(CDCl₃, 100 MHz) d: 178.6, 140.9, 139.1, 128.4, 127.6,

127.4, 121.6, 78.6, 70.0, 62.6, 56.8, 56.0, 50.2, 42.4, 39.8,

39.2, 38.7, 37.3, 36.9, 34.6, 33.2, 32.0, 31.9, 28.5, 28.3,

27.3, 24.3, 21.1, 19.4, 18.9, 11.8. FTIR (thin ®lm): 2928,

1727 cm²¹

.

5.1.51. 23-(1,1-Dimethylpropanoyloxy)-3b-phenylmethoxy-

5a-24-norcholan-6a-ol (82). To a stirring solution of the

ole®n 81 (265 mg, 0.509 mmol) in 5 mL of THF cooled to

08C was added borane±tetrahydrofuran complex (1.0 M,

1.018 mL, 1.018 mmol). The reaction mixture was stirred

at 258C for 12 h. The reaction mixture was cooled to 08C

and 1 mL of 10% NaOH was added dropwise followed by

1 mL of 30% hydrogen peroxide. The reaction was stirred

vigorously for 1 h. Ether and water were added and the

mixture was extracted with ether (3£10 mL). The organic

layers were combined, dried over anhydrous magnesium

sulfate and evaporated under reduced pressure to give a

white solid. Flash column chromatography of the crude

solid (silica gel, 20:1 hexanes:ether) gave the alcohol 82

(162 mg, 60%) as a white solid. ¹H NMR (CDCl₃,

400 MHz) d: 7.40±7.20 (5H, m), 4.58 (1H, d, J

11.8 Hz), 4.52 (1H, d, J11.8 Hz), 4.07 (2H, m), 3.34

(1H, m), 3.31 (1H, m), 2.40 (1H, m), 2.04 (1H, bs), 2.00±

0.50 (23H, m), 1.14 (9H, s), 0.94 (3H, d, J6.5 Hz), 0.79

(3H, s), 0.63 (3H, s). ¹³C NMR (CDCl₃, 100 MHz) d: 178.7,

139.1, 128.3, 127.6, 127.4, 78.1, 69.9, 69.3, 62.6, 56.2, 56.1,

53.8, 51.6, 42.6, 41.6, 39.8, 38.7, 37.3, 36.5, 34.6, 34.3,

33.2, 28.8, 28.2, 28.1, 27.2, 24.2, 21.1, 18.8, 13.4, 12.0.

3440, 2944, 2870, 1732 cm²¹

.

5.1.54. 6a-Acetyloxy-3a-((4-benzoyl)phenylacetyloxy)-

23-(1,1-dimethylpropanoyloxy)-5a-24-norcholane (85).

To a stirring solution of the alcohol 84 (112 mg,

0.228 mmol) in THF (6 mL) at 258C was added the

carboxylic acid 16 (66 mg, 0.274 mmol), triphenylphos-

phine (120 mg, 0.456 mmol), and DEAD (72 mL,

0.456 mmol). After stirring for 1 h, the reaction was

quenched with water and the mixture was extracted with

ether (3£10 mL). The combined organic extracts were

dried over magnesium sulfate, ®ltered, and concentrated

under reduced pressure to yield a crude oil which was

subjected to ¯ash column chromatography (silica gel, 4:1

hexanes:ethyl acetate) to provide the ester 85 (145 mg,

1

90%) as a white solid. H NMR (CDCl₃, 400 MHz) d:

7.90±7.70 (4H, m), 7.55 (1H, m), 7.50±7.35 (4H, m), 5.05

(1H, m), 4.57 (1H, ddd, J10.6, 10.6, 4.6 Hz), 4.06 (2H, m),

3.68 (2H, s), 2.00±0.50 (24H, m), 1.94 (3H, s), 1.16 (9H, s),

0.92 (3H, d, J 6.5 Hz), 0.78 (3H, s), 0.60 (3H, s). ¹³C NMR

(CDCl₃, 100 MHz) d: 195.9, 178.6, 170.8, 170.0, 139.4,

137.6, 136.2, 132.4, 129.9, 129.3, 128.3, 128.3, 72.4, 70.0,

62.5, 56.1, 56.0, 53.6, 44.0, 42.6, 42.0, 39.6, 38.7, 37.5,

36.7, 34.6, 33.9, 32.9, 28.2, 27.2, 27.1, 25.6, 24.05, 23.95,

21.2, 20.6, 18.8, 12.4, 11.9. FTIR (thin ®lm): 2944, 2869,

FTIR (thin ®lm): 3422, 2938, 2870, 2851, 1727 cm²¹

.

5.1.52. 6a-Acetyloxy-23-(1,1-dimethylpropanoyloxy)-3b-

phenylmethoxy-5a-24-norcholane (83). To a stirring

solution of the alcohol 82 (157 mg, 0.291 mmol) in 5 mL

of pyridine at 258C was added acetic anhydride (83 mL,

0.873 mmol) and the reaction was stirred overnight. Ice

chips were added and the products were extracted with

ether (3£10 mL). The organic layers were combined,

dried over anhydrous magnesium sulfate, and concentrated

under reduced pressure to give a white solid. Flash column

chromatography of the crude solid (silica gel, 6:1 hexanes:

ethyl acetate) gave the acetate 83 (169 mg, 100%) as a white

1732, 1659, 1607 cm²¹

.

5.1.55. 3a,6a,23-Tris[((1,1-dimethyl)ethyl)dimethylsilyl-

oxy]-5a-24-norchol-14-ene (86). The benzophenone ester

85 (140 mg, 0.196 mmol) was dissolved in 200 mL of

degassed, puri®ed benzene. The solution was kept under

argon at 258C and irradiated with a 400 W mercury arc

lamp for 10 h. The resulting solution was concentrated

under reduced pressure. To the crude oil obtained was

added 10 mL of KOH (10% aqueous):EtOH:THF (1:3:1)

and the solution stirred overnight at 258C and extracted

thoroughly with ether (3£15 mL). The combined extracts

were washed with 10% sodium bicarbonate solution, dried

over magnesium sulfate, ®ltered, and concentrated under

1

solid. H NMR (CDCl₃, 400 MHz) d: 7.35±7.20 (5H, m),

4.65 (1H, ddd, J10.6, 10.6, 4.6 Hz), 4.56 (1H, d,

J11.8 Hz), 4.52 (1H, d, J11.8 Hz), 4.20±3.90 (2H, m),

3.30 (1H, m), 2.02 (3H, s), 2.00±0.60 (24H, m), 1.18 (9H, s),

0.95 (3H, d, J6.5 Hz), 0.86 (3H, s), 0.64 (3H, s). ¹³C NMR

(CDCl₃, 100 MHz) d: 178.6, 170.7, 139.0, 128.4, 127.6,

1474

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

reduced pressure to yield a white solid. The white solid

obtained was dissolved in dichloromethane (10 mL) and

cooled to 08C. To the stirring solution was added 2,6-

lutidine (83 mL, 0.714 mmol) followed by tert-butyl-

dimethylsilyl tri¯ate (123 mL, 0.536 mmol). The solution

was quenched with water after 1 h and extracted with

dichloromethane (3£10 mL). The combined dichloro-

methane extracts were dried over magnesium sulfate,

®ltered, and concentrated under reduced pressure. The

crude oil was subjected to ¯ash column chromatography

on 20% silver nitrate-impregnated silica gel (6:1 hexanes:

ethyl acetate) to yield the pure ole®n 86 (66 mg, 48%) as a

extracts were dried over magnesium sulfate, ®ltered, and

concentrated under reduced pressure. Flash column chroma-

tography (silica gel, 8:1 hexanes:ethyl acetate) provide an

equal mixture of the diastereomeric alcohols (0.018 mg,

90%). To a stirring solution of the ole®nic alcohols

(0.017 mg, 0.026 mmol) in THF (3 mL) at 08C was added

borane±tetrahydrofuran

complex

(1.0 M,

78 mL,

0.078 mmol). The reaction mixture was stirred at 258C for

12 h. The reaction mixture was cooled to 08C and 0.5 mL of

10% NaOH was added dropwise followed by 0.5 mL of

30% hydrogen peroxide. The reaction was stirred

vigorously for 1 h. Water was added and the aqueous

layer was extracted with ether (3£10 mL). The organic

layers were combined, dried over anhydrous magnesium

sulfate, ®ltered, and concentrated under reduced pressure

to give the crude diol. To a stirring mixture of the crude

diol in dichloromethane (4 mL) at 258C was added pyri-

dinium chlorochromate (26 mg, 0.120 mmol) and the reac-

tion was stirred for 4 h. Celite (26 mg) was then added

followed by ether and the mixture was ®ltered and concen-

trated under reduced pressure. Flash column chroma-

tography (silica gel, 8:1 hexanes:ethyl acetate) of the

residue provided the desired dione 89a (12 mg, 71% for

1

clear glass. H NMR (CDCl₃, 400 MHz) d: 5.17 (1H, s),

4.01 (1H, m), 3.65 (2H, m), 3.39 (1H, ddd, J10.6, 10.6,

4.2 Hz), 2.33 (1H, ddd, J15.3, 7.7, 2.3 Hz), 2.20±0.50

(20H, m), 0.93 (3H, d, J6.4 Hz), 0.904 (3H, s), 0.895

(9H, s), 0.876 (9H, s), 0.867 (9H, s), 0.77 (3H, s), 0.05

(6H, s), 0.04 (3H, s), 0.02 (3H, s), 0.005 (3H, s), 0.003

(3H, s).

5.1.56. 3a,6a-Bis[((1,1-dimethyl)ethyl)dimethylsilyloxy]-

5a-24-norchol-14-en-23-ol (87). The tris-silyl ether 86

(35 mg, 0.054 mmol) was stirred at 258C for 24 h in

5.5 mL of AcOH:THF:water (4:6:1). The mixture was

extracted with benzene (3£5 mL) and the combined organic

layers were dried over magnesium sulfate, ®ltered, and

concentrated under reduced pressure to provide a clear oil.

Flash column chromatography (silica gel, 5:1 hexanes:ethyl

acetate) of the oil provided the alcohol 87 (25 mg, 86%) as a

1

three steps) as a clear glass. H NMR (CDCl₃, 400 MHz)

d: 4.01 (1H, m), 3.35 (1H, ddd, J10.5, 10.5, 4.4 Hz), 2.87

(1H, ddd, J12.6, 3.2, 3.2 Hz), 2.50±0.50 (24H, m), 0.98

(3H, d, J6.4 Hz), 0.911 (3H, d, J6.5 Hz), 0.900 (3H, d,

J6.5 Hz), 0.87 (9H, s), 0.86 (9H, s), 0.77 (3H, s), 0.75 (3H,

s), 0.06 (3H, s), 0.02 (3H, s), 0.006 (6H, s). ¹³C NMR

(CDCl₃, 100 MHz) d: 214.6, 210.2, 69.8, 66.4, 65.6, 53.2,

52.7, 51.2, 49.9, 46.0, 42.4, 41.7, 40.2, 39.7, 36.6, 32.7,

31.8, 31.0, 30.5, 29.6, 25.94, 25.88, 24.5, 22.6, 22.5,

20.25, 20.19, 18.2, 18.1, 13.0, 12.5, 24.0, 24.7, 24.8,

24.9.

1

clear glass. H NMR (CDCl₃, 400 MHz) d: 5.17 (1H, s),

4.01 (1H, m), 3.70 (2H, m), 3.39 (1H, ddd, J10.6, 10.6,

4.2 Hz), 2.33 (1H, ddd, J15.2, 7.8, 2.1 Hz), 0.50±2.20

(21H, m), 0.95 (3H, d, J6.5 Hz), 0.91 (3H, s), 0.87 (9H,

s), 0.86 (9H, s), 0.76 (3H, s), 0.03 (3H, s), 0.02 (3H, s), 0.003

(3H, s), 0.000 (3H, s).

5.1.59. 7-Deoxyxestobergsterol A (1d). To a stirring solu-

tion of the diketone 89a (8 mg, 0.012 mmol) in tetrahydro-

furan (3 mL) was added 20% HCl (0.5 mL) and the reaction

allowed to stir at 258C for 48 h. Saturated sodium bicarbo-

nate was added and the mixture was extracted with ether

(3£5 mL). The combined organic extracts were dried over

magnesium sulfate, ®ltered, and concentrated under reduced

pressure to provide diol dione 90 (5 mg, 100%) as a white

solid. To a stirring solution of the diol dione 90 (4 mg,

0.009 mmol) in ethanol (3.5 mL) was added 10% NaOH

(0.5 mL) and the mixture was allowed to stir until the reac-

tion was judged to be complete by TLC (24 h). Water was

added and the mixture was extracted with ether (3£5 mL).

The organic layers were dried over magnesium sulfate,

®ltered, and concentrated under reduced pressure to yield

a white solid. The solid was subjected to ¯ash column

chromatography (silica gel, 13:1 chloroform:methanol) to

5.1.57. 3a,6a-Bis[((1,1-dimethyl)ethyl)dimethylsilyloxy]-

5a-24-norcholan-23-al (88). To a stirring solution of the

alcohol 87 (23 mg, 0.039 mmol) in dichloromethane (4 mL)

at 258C was added pyridinium chlorochromate (13 mg,

0.059 mmol) and the reaction was allowed to stir for 5 h.

Celite (13 mg) and ether (4 mL) were added and the solution

was ®ltered through Celite and concentrated under reduced

pressure. The resulting crude oil was subjected to ¯ash

column chromatography (silica gel, 15:1 hexanes:ether) to

1

provide the aldehyde 88 (19 mg, 83%) as a clear glass. H

NMR (CDCl₃, 400 MHz) d: 9.78 (1H, dd, J1.5, 1.5 Hz),

5.15 (1H, s), 4.01 (1H, m), 3.39 (1H, ddd, J10.6, 10.6,

4.3 Hz), 2.60±1.05 (21H, m), 1.02 (3H, d, J6.2 Hz), 0.95

(3H, s), 0.87 (9H, s), 0.86 (9H, s), 0.77 (3H, s), 0.03 (3H, s),

0.017 (3H, s), 0.003 (3H, s), 0.001 (3H, s). ¹³C NMR

(CDCl₃, 100 MHz) d: 203.4, 155.2, 116.6, 70.2, 66.4,

58.4, 52.9, 50.8, 47.1, 45.4, 42.2, 40.0, 36.8, 35.8, 33.9,

32.7, 31.0, 30.0, 29.5, 25.9, 25.8, 21.4, 20.3, 18.12, 18.07,

16.8, 12.3, 24.0, 24.6, 24.9, 25.0.

provide the aldol product 1d (5 mg, 100%) as a white

Â Ã

25

D

1

solid. a 223.88 (c 0.40, CH₂Cl₂). H NMR (pyridine-

d₅, 400 MHz) d: 4.37 (1H, m), 3.65 (1H, ddd, J10.8, 10.8,

4.4 Hz), 3.22 (1H, ddd, J11.7, 11.7, 11.7 Hz), 2.81 (1H,

bd, J11.6), 2.67 (1H, d, J10.0 Hz), 2.60 (1H, m), 2.56

(1H, bs), 2.33 (1H, m), 2.16 (1H, m), 2.09 (2H, m), 2.01

(1H, dd, J14.2, 4.7 Hz), 2.00±0.90 (16H, m), 1.54 (1H, dd,

J14.3, 7.2 Hz), 1.16 (3H, s), 1.12 (3H, d, J6.3 Hz), 1.01

(3H, d, J6.5 Hz), 0.97 (3H, d, J6.7 Hz), 0.84 (3H, s). ¹³C

NMR (pyridine-d₅, 100 MHz) d: 217.0, 82.0, 69.6, 65.4,

62.8, 57.7, 56.5, 52.1, 51.7, 47.4, 46.4, 40.2, 38.70, 38.66,

5.1.58. 3a,6a-Bis[((1,1-dimethyl)ethyl)dimethylsilyloxy]-

5a-cholestane-15,23-dione (89a). A solution of the alde-

hyde 88 (18 mg, 0.031 mmol) in ether (3 mL) was added to

a stirring solution of isobutylmagnesium bromide (2.0 M,

78 mL, 0.155 mmol) cooled to 08C. The reaction was stirred

for 0.5 h and then water was added and the mixture was

extracted with ether (3£5 mL). The combined organic

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

1475

37.2, 35.0, 33.2, 32.0, 31.5, 29.5, 25.2, 24.9, 24.7, 21.2,

20.9, 19.9, 12.5. FTIR (thin ®lm): 3482, 3416, 3349,

2923, 2851, 1723 cm²¹. High-resolution MS (EI, m/z):

432.3233, calcd for C₂₇H₄₄O₄432.3240.

11.7. FTIR (thin ®lm): 3474, 2950, 2909, 2870, 2853,

1730, 1285, 1161, 1036 cm²¹

.

5.1.63. 3b,7b-Diacetyloxy-23-(1,1-dimethylpropanoyl-

oxy)-24-norchol-5-ene (94). To a stirring solution of the

allylic alcohol 93 (1.000 g, 2.249 mmol) in pyridine

(10 mL) at 258C was added acetic anhydride (0.424 mL,

4.497 mmol) followed by DMAP (4 mg). The reaction

was stirred overnight and then ice chips were added.

Water was added and the mixture was extracted with ether

(3£20 mL). The combined organic layers were dried over

magnesium sulfate, ®ltered, and concentrated under reduced

pressure. The excess pyridine was removed by a high-

vacuum rotovaporator to provide the allylic acetate 94

(1.012 g, 92%) as a white solid, which was used without

5.1.60. 3b-Acetyloxy-23-(1,1-dimethylpropanoyloxy)-24-

norchol-5-ene (91). The methyl ether 79 (1.690 g,

0.380 mmol)) was dissolved in glacial acetic acid and

heated to 908C for 3 h. The solution was cooled and the

volatile components were removed under reduced pressure

to give the allylic acetate 91 (1.696 g, 94%) as a white solid.

¹H NMR (400 MHz) d: 5.34 (1H, m), 4.56 (1H, m), 4.20±

3.90 (2H, m), 2.40±0.50 (23H, m), 1.99 (3H, s), 1.16 (9H, s),

0.99 (3H, s), 0.94 (3H, d, J6.5 Hz), 0.65 (3H, s). ¹³C NMR

(CDCl₃, 100 MHz) d: 178.6, 170.4, 139.6, 122.5, 73.9, 62.5,

56.6, 56.0, 50.0, 42.4, 39.6, 38.7, 38.1, 37.0, 36.6, 34.6,

33.2, 31.8, 28.3, 27.7, 27.2, 24.2, 21.4, 21.0, 19.3, 18.8,

11.7. FTIR (thin ®lm): 2973, 2965, 2940, 2822, 1732,

1

further puri®cation. H NMR (CDCl₃, 400 MHz) d: 5.20

(1H, s), 4.98 (1H, d, J8.6 Hz), 4.54 (1H, m), 4.05 (2H,

m), 2.40±2.25 (2H, m), 2.10±0.80 (19H, m), 1.982 (3H, s),

1.975 (3H, s), 1.15 (9H, s), 1.04 (3H, s), 0.93 (3H, d,

J6.5 Hz), 0.66 (3H, s). ¹³C NMR (CDCl₃, 100 MHz) d:

178.5, 171.0, 170.2, 144.1, 122.2, 75.4, 73.2, 62.4, 55.4,

55.3, 48.0, 42.9, 39.2, 38.7, 37.5, 36.50, 36.46, 36.4, 34.6,

33.0, 28.4, 27.6, 27.2, 25.1, 21.6, 21.3, 21.0, 19.0, 18.8,

11.6. FTIR (thin ®lm): 2950, 2872, 2911, 2855, 1734,

1373, 1240, 1161, 1032 cm²¹. High-resolution MS (EI,

m/z): 529.3527, calcd for C₃₂H₄₉O₆(M±H)¹529.3529.

1478, 1466, 1368, 1157, 1040 cm²¹

.

5.1.61. 3b-Acetyloxy-23-(1,1-dimethylpropanoyloxy)-24-

norchol-5-en-7-one (92). A mixture of the ole®n 91

(54 mg, 0.120 mmol) and ruthenium trichloride mono-

hydrate (5 mg) was stirred at 258C in 9:1 cyclohexane:water

(32 mL). tert-Butyl hydroperoxide (70% aqueous, 5.53 mL)

was added dropwise to the solution over 6 h and the mixture

was stirred for an additional 18 h. Sodium sul®te (1.6 g) was

added and the mixture allowed to stir for 1 h. Water was

added and the mixture was extracted with ether (3£30 mL).

The combined organic extracts were dried over magnesium

sulfate, ®ltered, and concentrated under reduced pressure.

Flash column chromatography (silica gel, 5:1:1 hexanes:

ethyl acetate:chloroform) of the residue provided the

enone 92 (1.111 g, 65%) as a white solid. ¹H NMR

(CDCl₃, 400 MHz) d: 5.66 (1H, d, J1.6 Hz), 4.67 (1H,

m), 4.03 (2H, m), 2.80±0.50 (21H, m), 2.01 (3H, s), 1.17

(3H, s), 1.15 (9H, s), 0.93 (3H, d, J6.6 Hz), 0.65 (3H, s).

¹³C NMR (CDCl₃, 100 MHz) d: 201.7, 178.6, 170.2, 163.9,

126.6, 72.2, 62.5, 54.7, 49.9, 49.7, 45.3, 43.2, 38.7, 38.6,

38.3, 37.7, 36.0, 34.6, 33.1, 28.6, 27.3, 27.2, 26.2, 21.2,

21.1, 18.9, 17.2, 11.8. FTIR (thin ®lm): 2950, 2870, 1732,

1671, 1638 cm²¹. High-resolution MS (EI, m/z): 486.3338,

calcd for C₃₀H₄₆O₅486.3345.

5.1.64. 23-(1,1-Dimethylpropanoyloxy)-5a-24-norcholan-

3b,6a,7b-triol (95). To a stirring solution of the ole®n 94

(1.057 g, 1.992 mmol) in 20 mL of THF at 08C was added

borane±tetrahydrofuran complex (1.0 M, 6 mL, 5.976

mmol). The reaction mixture was stirred at 258C for 12 h.

The reaction mixture was cooled to 08C and 5 mL of 10%

NaOH was added dropwise followed by 2 mL of 30%

hydrogen peroxide. The reaction was stirred vigorously

for 1 h. Ether and water were added and the products were

extracted with ether (3£20 mL). The organic layers were

combined, dried over anhydrous magnesium sulfate,

®ltered, and evaporated under reduced pressure to give a

white solid. Flash column chromatography (silica gel,

12:1 chloroform:methanol) of the crude solid gave the

1

triol 95 (527 mg, 57%) as a white solid. H NMR (CDCl₃,

400 MHz) d: 4.09 (2H, m), 3.58 (1H, dddd, J10.9, 10.9,

4.7, 4.7 Hz), 3.25 (1H, dd, J10.5, 8.7 Hz), 3.11 (1H, dd,

J9.3, 8.7 Hz), 2.30±0.60 (25H, m), 1.19 (9H, s), 0.97 (3H,

d, J6.6 Hz), 0.86 (3H, s), 0.68 (3H, s). ¹³C NMR (CDCl₃,

100 MHz) d: 178.7, 80.2, 74.7, 70.8, 56.0, 55.3, 52.1, 47.9,

43.5, 40.9, 39.8, 38.7, 37.4, 35.6, 34.6, 33.0, 32.5, 32.5,

30.6, 28.6, 27.2, 26.9, 21.4, 18.9, 13.6, 12.2. FTIR (thin

®lm): 3359, 2938, 2867, 1728, 1480, 1447, 1285, 1159,

5.1.62. 3b-Acetyloxy-23-(1,1-dimethylpropanoyloxy)-24-

norchol-5-en-7b-ol (93). To a stirring solution of the

enone 92 (1.087 g, 2.234 mmol) in dichloromethane:

methanol (4:1, 50 mL) cooled to 2788C was added cerium

trichloride heptahydrate (1.1664 g, 4.467 mmol) followed

by sodium borohydride (0.169 g, 4.467 mmol). The solution

was stirred and allowed to warm to 258C slowly. The solu-

tion was acidi®ed with 10% HCl and extracted with ether

(3£100 mL). The combined organic layers were dried over

magnesium sulfate, ®ltered, and concentrated under reduced

pressure to yield the allylic alcohol 93 (1.060 g, 97%) as a

white solid, which was used without further puri®cation. ¹H

NMR (CDCl₃, 400 MHz) d: 5.27 (1H, d, J1.6 Hz), 4.57

(1H, m), 4.20±3.90 (2H, m), 3.80 (1H, d, J7.9 Hz), 2.50±

0.50 (22H, m), 1.99 (3H, s), 1.15 (9H, s), 1.02 (3H, s), 0.93

(3H, d, J6.6 Hz), 0.66 (3H, s). ¹³C NMR (CDCl₃,

100 MHz) d: 178.6, 170.4, 142.2, 126.5, 73.4, 73.0, 62.5,

55.9, 55.3, 48.1, 43.0, 40.6, 39.4, 38.7, 37.6, 36.6, 36.5,

34.6, 33.1, 28.6, 27.7, 27.2, 26.3, 21.4, 21.0, 19.0, 18.9,

1087, 735 cm²¹

.

5.1.65. 3b,6a,7b-Triacetyloxy-23-(1,1-dimethylpropa-

noyloxy)-5a-24-norcholane (96). To a stirring solution of

the triol 95 (261 mg, 0.562 mmol) in dichloromethane

(7 mL) at 258C was added acetic anhydride (239 mL,

2.528 mmol) followed by trimethylsilyl tri¯ate (5 mL) and

the reaction mixture was stirred for 30 min. Saturated

sodium bicarbonate was added and the mixture was allowed

to stir an additional 0.5 h. Water was added and the mixture

was extracted with dichloromethane (3£10 mL). The

combined organic extracts were dried over magnesium

sulfate, ®ltered, and concentrated under reduced pressure.

1476

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

Flash column chromatography of the residue (silica gel, 3:1

hexanes:ethyl acetate) yielded the triacetate 96 (316 mg,

zene (389 mL) at 258C for 10 h with a 450 W mercury arc

lamp and a pyrex ®lter. The solution was concentrated under

reduced pressure and the residue was subjected to ¯ash

column chromatography (silica gel, 3:1:1 hexanes:ethyl

acetate:chloroform) to give the ole®n (184 mg, 61%). To a

stirring solution of the ole®n (175 mg, 0.227 mmol) at 258C

in dichloromethane (7 mL) was added pyridinium chloro-

chromate (98 mg, 0.454 mmol) and the reaction was stirred

for 2 h. Celite (100 mg) and ether (7 mL) were added and

the mixture was ®ltered through Celite. Concentration under

reduced pressure gave the crude ole®n, which was subjected

to ¯ash column chromatography (silica gel, 3:1 hexanes:

ethyl acetate) to give the pure ole®n (152 mg, 87%). To a

stirring solution of the ester (142 mg, 0.185 mmol) in THF

(5 mL) cooled to 08C was added lithium aluminum hydride

in portions (70 mg, 1.850 mmol). The mixture was stirred at

258C overnight. Water (100 mL) was added, followed by

10% NaOH (100 mL), followed by 3 mL of water. The

mixture was warmed to 258C and stirred for 15 min. Mag-

nesium sulfate was added and the reaction stirred for an

additional 15 min. The solution was ®ltered and washed

thoroughly with methanol. Concentration under reduced

pressure and ¯ash column chromatography of the resulting

crude solid (silica gel, 9:1 chloroform:methanol) provided

the tetrol 99 (55 mg, 79%) as a white solid. ¹H NMR

(CD₃OD, 400 MHz) d: 5.57 (1H, s), 4.04 (1H, m), 3.75±

3.50 (2H, m), 3.41 (1H, dd, J9.6, 8.6 Hz), 2.29 (1H, dd,

J11.2, 8.6 Hz), 2.35 (1H, ddd, J14.9, 7.9, 2.3 Hz), 2.15±

0.50 (22H, m), 0.98 (3H, d, J6.5 Hz), 0.96 (3H, s), 0.85

(3H, s). ¹³C NMR (CD₃OD, 100 MHz) d: 152.6, 120.4, 78.3,

76.0, 66.6, 60.9, 60.2, 53.7, 48.8, 44.2, 43.0, 42.9, 39.9,

37.3, 37.0, 33.7, 32.4, 31.1, 29.2, 23.2, 19.8, 17.3, 12.8.

1

95%) as a clear glass. H NMR (CDCl₃, 400 MHz) d: 4.76

(1H, dd, J11.1, 9.2 Hz), 4.66 (1H, dd, J9.9, 9.2 Hz), 4.53

(1H, dddd, J11.3, 11.3, 5.0, 5.0 Hz), 4.20±3.80 (2H, m),

2.00±0.70 (22H, m), 1.92 (3H, s), 1.91 (3H, s), 1.87 (3H, s),

1.10 (9H, s), 0.88 (3H, s), 0.87 (3H, d, J5.9 Hz), 0.59 (3H,

s). ¹³C NMR (CDCl₃, 100 MHz) d: 178.5, 170.8, 170.7,

170.4, 76.9, 74.4, 72.6, 62.4, 55.0, 54.7, 51.7, 46.0, 43.6,

39.3, 39.0, 38.7, 36.7, 35.8, 34.6, 33.0, 28.5, 28.3, 27.2,

27.0, 24.8, 21.5, 21.4, 21.2, 20.9, 18.8, 13.3, 11.9. FTIR

(thin ®lm): 2955, 2870, 1744, 1375, 1248, 1161, 1036,

735 cm²¹

.

5.1.66. 6a,7b-Diacetyloxy-23-(1,1-dimethylpropanoyl-

oxy)-5a-24-norcholan-3b-ol (97). To a stirring solution

of the triacetate 96 (296 mg, 0.501 mmol) in 1:1 THF:

ethanol (10 mL) at 08C was added 10% NaOH solution

(0.5 mL) and the reaction mixture was stirred for 1 h. HCl

(10% aqueous) was added and the mixture was extracted

with ether (3£10 mL). The organic extracts were dried over

magnesium sulfate, ®ltered, and concentrated under reduced

pressure. Flash column chromatography (silica gel, 1.5:1

hexanes:ethyl acetate) of the residue yielded the alcohol

97 (244 mg, 89%) as a white solid. ¹H NMR (CDCl₃,

400 MHz) d: 4.84 (1H, dd, J11.2, 9.3 Hz), 4.66 (1H, dd,

J10.0, 9.3 Hz), 4.20±3.80 (2H, m), 3.52 (1H, dddd, J

11.3, 11.3, 5.0, 5.0 Hz), 2.00±0.70 (23H, m), 1.99 (3H, s),

1.95 (3H, s), 1.17 (9H, s), 0.95 (3H, d, J5.9 Hz), 0.93 (3H,

s), 0.66 (3H, s). ¹³C NMR (CDCl₃, 100 MHz) d: 178.6,

170.9, 170.7, 78.0, 74.5, 70.7, 62.4, 55.0, 54.8, 51.8, 46.2,

43.6, 39.4, 39.1, 38.7, 37.0, 35.8, 34.5, 32.9, 32.1, 31.0,

28.5, 27.2, 24.8, 21.5, 21.3, 20.8, 18.8, 13.4, 11.9. FTIR

(thin ®lm): 3441, 2950, 2940, 2870, 1746, 1728, 1480,

FTIR (thin ®lm): 3316, 2928, 1437, 1368, 1094, 1016 cm²¹

.

1460, 1246, 1161 cm²¹

.

5.1.69. 23-(1,1-Dimethylpropanoyloxy)-5a-24-norchol-

14-ene-3a,6a,7b-triol (100). To a stirring solution of the

tetrol 99 (48 mg, 0.127 mmol) in pyridine (3 mL) cooled to

08C was added pivaloyl chloride (19 mL, 0.152 mmol). The

mixture was stirred at 258C overnight. Water was added and

the mixture was thoroughly extracted with ethyl acetate

(3£10 mL). The combined organic phases were dried over

magnesium sulfate, ®ltered, and concentrated under reduced

pressure. The excess pyridine was removed by high-vacuum

rotovaporator. Flash column chromatography of the residue

(silica gel, 17:1 chloroform:methanol) provided the triol

ester 100 (49 mg, 83%) as a white solid. ¹H NMR

(CD₃OD, 400 MHz) d: 5.59 (1H, s), 4.30±4.09 (2H, m),

4.06 (1H, m), 3.42 (1H, dd, J9.7, 8.7 Hz), 2.35 (1H,

ddd, J15.1, 7.8, 2.1 Hz), 2.29 (1H, dd, J11.1, 8.6 Hz),

2.20±0.70 (21H, m), 1.20 (9H, s), 1.02 (3H, d, J6.4 Hz),

0.97 (3H, s), 0.87 (3H, s). ¹³C NMR (CD₃OD, 100 MHz) d:

180.2, 152.6, 120.3, 78.2, 76.0, 66.6, 63.8, 59.8, 53.7, 48.8,

44.2, 42.93, 42.91, 39.9, 37.3, 37.0, 35.8, 33.6, 32.7, 31.1,

29.2, 27.7, 23.2, 19.7, 17.3, 12.8. IR (thin ®lm): 3339, 2957,

5.1.67. 6a,7a-Diacetyloxy-3a-((4-benzoyl)phenylacetyl-

oxy)-23-(1,1-dimethylpropanoyl-oxy)-5a-24-norcholane

(98). To a stirring solution of the alcohol 97 (240 mg,

0.437 mmol) in THF (7 mL) at 258C was added triphenyl-

phosphine (229 mg, 0.875 mmol) and the carboxylic acid 16

(126 mg, 0.524 mmol) followed by the dropwise addition of

DEAD (138 mL, 0.875 mmol) and the solution was stirred

for 0.5 h. The reaction mixture was concentrated under

reduced pressure to give a pale orange oil. Flash column

chromatography of the crude oil (silica gel, 3:1 hexanes:

ethyl acetate) gave the ester 98 (313 mg, 93%) as a white

1

solid. H NMR (CDCl₃, 400 MHz) d: 7.85±7.65 (4H, m),

7.53 (1H, dd, J7.4 Hz), 7.43 (2H, dd, J7.7, 7.4 Hz), 7.37

(2H, d, J8.1 Hz), 5.02 (1H, m), 4.76 (1H, dd, J11.2,

9.2 Hz), 4.68 (1H, dd, J9.7, 9.3 Hz), 4.20±3.70 (2H, m),

3.66 (2H, s), 1.93 (3H, s), 1.91 (3H, s), 1.85±0.50 (22H, m),

1.14 (9H, s), 0.90 (3H, d, J6.5 Hz), 0.85 (3H, s), 0.61 (3H,

s). ¹³C NMR (CDCl₃, 100 MHz) d: 196.0, 178.6, 170.8,

170.8, 170.0, 139.1, 137.6, 136.4, 132.4, 130.5, 130.0,

129.3, 128.3, 77.8, 74.6, 69.6, 62.4, 55.0, 54.8, 52.0, 43.6,

41.8, 41.6, 39.3, 39.0, 38.7, 36.0, 34.6, 33.0, 32.8, 28.5,

27.3, 27.2, 25.5, 24.8, 21.5, 20.9, 20.8, 18.9, 12.5, 11.9.

FTIR (thin ®lm): 3061, 2955, 2872, 1740, 1661, 1607,

2932, 2870, 2826, 1732, 1285, 1159, 1019 cm²¹

.

5.1.70. 23-Acetyloxy-3a,5-cyclo-6b-methoxy-24-norch-

olane (101). To a stirring solution of the alcohol 75

(3.590 g, 9.956 mmol) in pyridine (40 mL) at 258C was

added acetic anhydride (1.13 mL, 11.95 mmol) dropwise

and the mixture was stirred for 24 h. Water was added and

the mixture was extracted with hexanes (3£15 mL). The

organic layers were combined and dried over magnesium

1366, 1377, 1248, 1157 cm²¹

.

5.1.68. 5a-24-Norchol-14-ene-3a,6a,7b,23-tetrol (99).

The ester 98 (300 mg, 0.389 mmol) was photolyzed in ben-

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

1477

sulfate, ®ltered, and concentrated under reduced pressure.

Flash column chromatography (silica gel, 8:1 hexanes:

ether) of the residue provided the ester 101 (3.990 mg,

to yield the allylic alcohol 104 (1.879 g, 98%) as a white

solid. ¹H NMR (CDCl₃, 400 MHz) d: 5.26 (1H, s), 4.56 (1H,

dddd, J11.4, 11.4, 4.4, 4.4 Hz), 4.20±3.90 (2H, m), 3.79

(1H, d, J7.9 Hz), 2.50±0.50 (22H, m), 1.993 (3H, s), 1.986

(3H, s), 1.01 (3H, s), 0.92 (3H, d, J6.5 Hz), 0.66 (3H, s).

¹³C NMR (CDCl₃, 100 MHz) d: 171.1, 170.3, 142.0, 126.4,

73.3, 72.9, 62.6, 55.8, 55.2, 48.0, 42.8, 40.5, 39.3, 37.4,

36.5, 36.3, 34.4, 33.0, 28.4, 27.6, 26.2, 21.2, 20.9, 20.8,

18.9, 18.7, 11.6. FTIR (thin ®lm): 3455, 2907, 2948,

1

100%). H NMR (CDCl₃, 400 MHz) d: 4.20±3.85 (2H,

m), 3.29 (3H, s), 2.74 (1H, dd, J2.6, 2.6 Hz), 2.00±0.70

(22H, m), 2.01 (3H, s), 0.99 (3H, s), 0.93 (3H, d, J6.6 Hz),

0.70 (3H, s), 0.62 (1H, dd, J4.3, 4.3 Hz), 0.39 (1H, dd,

J5.1, 8.0 Hz). ¹³C NMR (CDCl₃, 100 MHz) d: 171.1, 82.3,

62.7, 56.5, 56.4, 56.1, 47.9, 43.3, 42.8, 40.2, 35.2, 35.0,

34.5, 33.3, 33.1, 30.4, 28.2, 24.9, 24.1, 22.7, 21.4, 21.0,

19.2, 18.7, 13.0, 12.1. FTIR (thin ®lm): 2938, 2913, 2849,

2872, 2853, 1736, 1468, 1439, 1366, 1246 cm²¹

.

2869, 1742, 1242, 1100 cm²¹

.

5.1.74. 3b,7b,23-Triacetyloxy-24-norchol-5-ene (105).

To a stirring solution of the allylic alcohol 104 (1.861 g,

4.17 mmol) in pyridine (20 mL) at 258C was added acetic

anhydride (0.590 mL, 6.25 mmol) followed by 4-(dimethyl-

amino)pyridine (10 mg). The reaction was stirred overnight

and then ice chips were added. Water was added and the

mixture was extracted with ether (3£40 mL). The combined

organic extracts were dried over magnesium sulfate,

®ltered, and concentrated under reduced pressure. The

excess pyridine was removed by high-vacuum rotovapo-

rator. Flash column chromatography of the residue (silica

gel, 4:1 hexanes:ethyl acetate) provided the allylic acetate

5.1.71. 3b,23-Diacetyloxy-24-norchol-5-ene (102). The

ether 101 (3.800 g, 9.439 mmol) was dissolved in glacial

acetic acid (90 mL) and heated to 908C for 3 h. The solution

was cooled to 258C and the acetic acid was removed under

reduced pressure to provide the homoallylic acetate 102

(3.950 g, 97%) as a white solid. ¹H NMR (CDCl₃,

400 MHz) d: 5.33 (1H, d, J4.8 Hz), 4.57 (1H, m), 4.20±

3.90 (2H, m), 2.40±2.20 (2H, m), 2.15±0.80 (21H, m), 2.01

(3H, s), 1.99 (3H, s), 0.99 (3H, s), 0.93 (3H, d, J6.5 Hz),

0.66 (3H, s). ¹³C NMR (CDCl₃, 100 MHz) d: 171.1, 170.4,

139.5, 122.4, 73.8, 62.7, 56.5, 55.9, 49.9, 42.3, 39.6, 38.0,

36.9, 36.5, 34.5, 33.1, 31.74, 31.73, 28.1, 27.6, 24.1, 21.3,

21.0, 20.9, 19.2, 18.7, 11.7. FTIR (thin ®lm): 2963, 2938,

1

105 (1.968 g, 97%) as a white solid. H NMR (CDCl₃,

400 MHz) d: 5.19 (1H, s), 4.97 (1H, d, J8.6 Hz), 4.54

(1H, dddd, J11.4, 11.4, 4.4, 4.4 Hz), 4.20±3.90 (2H, m),

2.50±0.50 (21H, m), 1.986 (3H, s), 1.974 (3H, s), 1.967 (3H,

s), 1.03 (3H, s), 0.91 (3H, d, J6.5 Hz), 0.65 (3H, s). ¹³C

NMR (CDCl₃, 100 MHz) d: 171.0, 170.9, 170.1, 144.0,

122.1, 75.3, 73.0, 62.6, 55.3, 55.2, 47.9, 42.8, 39.1, 37.4,

36.4, 36.33, 36.30, 34.4, 33.0, 28.2, 27.5, 24.9, 21.5, 21.2,

20.9, 20.9, 18.8, 18.7, 11.6. FTIR (thin ®lm): 2950, 2815,

2892, 2867, 1732, 1460, 1439, 1368, 1375, 1246 cm²¹

.

5.1.72. 3b,23-Diacetyloxy-24-norchol-5-en-7-one (103).

A mixture of the ole®n 102 (2.900 g, 6.735 mmol) and ruth-

enium trichloride monohydrate (9 mg) was stirred at 258C in

1,2-dichloroethane (34 mL) and water (7 mL). tert-Butyl

hydroperoxide (70% aqueous, 8.76 mL) was added drop-

wise over 6 h and the mixture was allowed to stir for an

additional 18 h. Sodium sul®te (2.9 g) was added and the

mixture allowed to stir for 1 h. Water was added and the

mixture was extracted with ether (3£30 mL). The combined

organic extracts were dried over magnesium sulfate,

®ltered, and concentrated under reduced pressure. Flash

column chromatography (silica gel, 5:1:2 hexanes:ethyl

2874, 2855, 1736, 1368, 1240, 1032 cm²¹

.

5.1.75. 5a-24-Norcholane-3b,6a,7b,23-tetrol (106). To

stirring solution of the triacetate 105 (1.933 g,

a

3.956 mmol) in THF (20 mL) cooled to 08C was added

borane±tetrahydrofuran complex (1.0 M, 12 mL, 11.867

mmol) dropwise and the reaction was stirred overnight at

258C. The reaction was cooled to 08C and sodium hydroxide

(10% aqueous, 12 mL) was added very slowly followed by

hydrogen peroxide (30% aqueous, 12 mL) and the mixture

was stirred vigorously for 1 h. The mixture was acidi®ed

with HCl (10% aqueous) and extracted thoroughly with

ethyl acetate (3£100 mL). The combined organic layers

were dried over magnesium sulfate, ®ltered, and concen-

trated under reduced pressure to yield the tetrol 106

(0.750 g, 50%) as a white solid, which was used without

further puri®cation. ¹H NMR (CD₃OD, 400 MHz) d:

3.70±3.40 (2H, m), 3.10 (1H, dd, J10.4, 8.6 Hz), 2.96

(1H, dd, J9.3, 8.9 Hz), 2.20±0.50 (27H, m), 0.94 (3H, d,

J6.5 Hz), 0.85 (3H, s), 0.71 (3H, s). ¹³C NMR (CH₃OD,

100 MHz) d: 80.0, 74.6, 70.4, 59.4, 56.3, 55.7, 52.3, 48.0,

43.4, 40.9, 40.0, 38.6, 37.3, 35.3, 32.8, 31.8, 30.5, 28.5,

26.5, 21.3, 18.3, 12.7, 11.5. FTIR (thin ®lm): 3386, 2938,

acetate:chloroform) of the residue provided the enone 103

Â Ã

25

(1.940 g, 65%) as a white solid. a 2106.48 (c 1.1,

D

CH₂Cl₂). H NMR (CDCl₃, 400 MHz) d: 5.57 (1H, s),

1

4.58 (1H, dddd, J11.4, 11.4, 4.4, 4.4 Hz), 4.20±3.80

(2H, m), 2.60±0.50 (21H, m), 1.93 (3H, s), 1.92 (3H, s),

1.10 (3H, s), 0.85 (3H, d, J6.5 Hz), 0.58 (3H, s). ¹³C NMR

(CDCl₃, 100 MHz) d: 201.4, 171.0, 170.0, 163.8, 126.5,

72.1, 62.6, 54.6, 49.9, 49.6, 45.2, 43.1, 38.5, 38.2, 37.6,

35.9, 34.5, 33.0, 28.5, 27.2, 26.2, 21.2, 21.05, 20.97, 18.9,

17.2, 11.8. FTIR (thin ®lm): 2950, 2894, 2874, 2857, 1732,

1673, 1636, 1466, 1368, 1246 cm²¹. High-resolution MS

(EI, m/z): 444.2873, calcd for C₂₇H₄₀O₅444.2876.

5.1.73. 3b,23-Diacetyloxy-24-norchol-5-en-7b-ol (104).

To a stirring solution of the enone 103 (1.909 g,

4.294 mmol) in dichloromethane:methanol (4:1, 100 mL)

cooled to 2788C was added cerium trichloride heptahydrate

(3.200 g, 8.588 mmol) followed by sodium borohydride

(0.325 g, 8.588 mmol). The solution was stirred and allowed

to warm to 258C slowly. The solution was acidi®ed with

10% HCl and extracted with ether (3£100 mL). The

combined organic extracts were dried over magnesium

sulfate, ®ltered, and concentrated under reduced pressure

2863, 1642, 1051 cm²¹

.

5.1.76. 3b,6a,7b,23-Tetraacetyloxy-5a-24-norcholane

(107). To a stirring solution of the tetrol 106 (20 mg,

0.053 mmol) in dichloromethane (1 mL) at 258C was

added acetic anhydride (30 mL, 0.315 mmol) followed by

trimethylsilyl tri¯ate (1 drop) and the mixture was stirred for

1 h. Saturated sodium bicarbonate was added and the

1478

M. E. Jung, T. W. Johnson / Tetrahedron 57 (2001) 1449±1481

mixture was extracted with dichloromethane (3£2 mL). The

combined organic extracts were dried over magnesium

sulfate, ®ltered, and concentrated under reduced pressure

to yield the tetraacetate 107 (28 mg, 97%) as a clear glass.

¹H NMR (CDCl₃, 400 MHz) d: 4.82 (1H, dd, J11.2,

9.1 Hz), 4.72 (1H, dd, J10.0, 9.2 Hz), 4.59 (1H, dddd,

J11.1, 11.1, 4.7, 4.7 Hz), 4.20±3.90 (2H, m), 2.10±0.50

(22H, m), 2.00 (3H, s), 1.98 (3H, s), 1.97 (3H, s), 1.92 (3H,

s), 0.93 (3H, s), 0.92 (3H, d, J6.7 Hz), 0.65 (3H, s). ¹³C

NMR (CDCl₃, 100 MHz) d: 171.2, 170.85, 170.75, 170.5,

77.8, 74.4, 72.6, 62.7, 55.0, 54.7, 51.6, 46.0, 43.6, 39.3,

39.0, 36.7, 35.8, 34.5, 33.0, 28.4, 28.3, 26.9, 24.8, 21.5,

21.3, 21.2, 21.1, 20.8, 18.8, 13.2, 11.9. FTIR (thin ®lm):

2955, 2870, 1744, 1474, 1437, 1370, 1248, 1240,

1034 cm²¹. High-resolution MS (EI, m/z): 549.3437, calcd

for C₃₁H₄₉O₈(M1H)¹549.3427.

5.1.79.

3a-((-4-Benzoyl)phenylacetyloxy)-6a,7b-dia-

cetyloxy-23-[tris((1-methyl)ethyl)-silyloxy]-5a-24-nor-

cholane (135). To a stirring solution of the alcohol 134

(284 mg, 0.457 mmol) in THF (10 mL) at 258C was added

triphenylphosphine (240 mg, 0.914 mmol) and the

carboxylic acid 40 (132 mg, 0.548 mmol) followed by the

dropwise addition of DEAD (144 mL, 0.914 mmol) and the

reaction was stirred for 1 h. Water was added and the

mixture was extracted with ether (3£10 mL). The combined

organic extracts were dried over magnesium sulfate,

®ltered, and concentrated under reduced pressure. Flash

column chromatography of the crude residue (silica gel,

4:1 hexanes:ethyl acetate) provided the benzophenone

ester 135 (286 mg, 74%) as a clear glass. ¹H NMR

(CDCl₃, 400 MHz) d: 7.80±7.70 (4H, m), 7.60±7.48 (1H,

m), 7.48±7.30 (4H, m), 5.03 (1H, m), 4.77 (1H, dd, J9.3,

11.2 Hz), 4.70 (1H, dd, J9.8, 9.3 Hz), 3.80±3.50 (2H, m),

3.68 (2H, s), 2.10±0.40 (25H, m), 1.94 (3H, s), 1.92 (3H, s),

1.03 (18H, m), 0.89 (3H, d, J6.5 Hz), 0.86 (3H, s), 0.62

(3H, s). ¹³C NMR (CDCl₃, 100 MHz) d: 195.80, 170.67,

170.57, 169.85, 138.90, 137.40, 136.18, 132.20, 130.26,

129.82, 129.12, 128.11, 77.63, 74.46, 69.36, 61.14, 55.06,

54.61, 51.87, 43.43, 41.61, 39.17, 38.86, 35.77, 32.62,

32.35, 28.30, 27.02, 25.30, 24.73, 21.36, 20.68, 20.57,

18.91, 17.90, 12.31, 11.84, 11.74. FTIR (thin ®lm): 2944,

2892, 2867, 1744, 1661, 1607, 1464, 1447, 1377,

5.1.77. 6a,7b-Diacetyloxy-5a-24-norcholane-3b,23-diol

(108). To a stirring solution of the tetraacetate 107 (350 mg,

0.626 mmol) in THF:ethanol (1:1, 10 mL) cooled to 08C

was added 10% sodium hydroxide (1 mL) dropwise and

the reaction was stirred for 1 h. HCl (10% aqueous) was

added to make the solution neutral and the mixture was

extracted with dichloromethane (3£10 mL). The combined

organic layers were dried over magnesium sulfate, ®ltered,

and concentrated under reduced pressure. Flash column

chromatography of the residue (silica gel, 19:1 chloroform:

methanol) yielded the diol 108 (276 mg, 95%) as a white

solid. ¹H NMR (CDCl₃, 400 MHz) d: 4.68 (1H, dd, J10.5,

9.6 Hz), 4.58 (1H, dd, J9.6, 9.3 Hz), 3.60±3.20 (3H, m),

2.98 (2H, bs), 2.20±0.50 (22H, m), 1.84 (3H, s), 1.81 (3H,

s), 0.79 (3H, s), 0.78 (3H, d, J6.5 Hz), 0.53 (3H, s). ¹³C

NMR (CDCl₃, 100 MHz) d: 170.9, 170.8, 77.9, 74.6, 70.2,

60.2, 55.2, 54.7, 51.7, 46.1, 43.5, 39.3, 38.9, 38.6, 36.8,

35.7, 32.6, 31.8, 30.7, 28.4, 24.7, 21.1, 21.2, 20.7, 18.8,

13.2, 11.9. FTIR (thin ®lm): 3370, 2946, 2870, 1744,

1377, 1260, 1246, 1046, 1028, 733 cm²¹. High-resolution

MS (EI, m/z): 464.3131, calcd for C₂₇H₄₄O₆464.3138.

1277 cm²¹

.

5.1.80. 3a,6a,7b-Tris(methoxymethoxy)-23-[tris((1-

methyl)ethyl)silyl]oxy-5a-24-norchol-14-ene (137). The

benzophenone ester 135 (286 mg, 0.339 mmol) was

dissolved in puri®ed, degassed benzene and photolyzed

using a 450 W mercury arc lamp with a pyrex ®lter at

258C for 10 h. The solvent was removed under reduced

pressure and the solid subjected to ¯ash column chromato-

graphy (silica gel, 3:1:1 hexanes:ethyl acetate:chloroform)

to give the epimeric alcohols (137 mg, 48%). To a stirring

solution of the epimeric alcohols (137 mg, 0.163 mmol) in

dichloromethane (7 mL) at 258C was added pyridinium

chlorochromate (70 mg, 0.325 mmol) and the mixture was

stirred for 2 h. Ether (10 mL) and Celite (100 mg) were

added and the mixture was ®ltered through Celite and

concentrated under reduced pressure to give the benzo-

phenone ole®n (86 mg, 63%). To a stirring solution of the

benzophenone ole®n (81 mg, 0.096 mmol) in THF (2 mL)

at 258C was added lithium aluminum hydride (37 mg,

0.963 mmol) and the reaction was stirred for several

hours. The mixture was cooled to 08C and water (100 mL)

was added followed by 10% NaOH (100 mL), which was

followed by the addition of water (300 mL) again. The

mixture was warmed to 258C and stirred for 15 min. Mag-

nesium sulfate was added and the mixture was ®ltered and

concentrated under reduced pressure. Flash column chroma-

tography (silica gel, 20:1 chloroform:methanol) of the crude

solid gave the ole®nic triol 136 (44 mg, 85%) as a white

solid. To a stirring solution of the triol 136 (32 mg,

0.060 mmol) in 1,2-dimethoxyethane (3 mL) was added

diisopropylamine (470 mL, 2.700 mmol) and chloromethyl

methyl ether (68 mL, 0.900 mmol) followed by sodium

iodide (0.027 mg, 0.180 mmol) and the mixture was

re¯uxed for 12 h. The mixture was allowed to cool and

saturated sodium carbonate and water were added and the

mixture was extracted with ether (3£5 mL). The combined

5.1.78. 6a,7b-Diacetyloxy-23-[tris(1-methylethyl)]silyl-

oxy-5a-24-norcholan-3b-ol (134). To a stirring solution

of the diol 133 (283 mg, 0.609 mmol) in DMF (5 mL) at

08C was added triisopropylchlorosilane (143 mL,

0.670 mmol) dropwise and the reaction was stirred for 4 h.

Water was added and the mixture was extracted with

dichloromethane (3£10 mL). The combined organic

extracts were dried over magnesium sulfate, ®ltered, and

concentrated under reduced pressure. Flash column chroma-

tography of the residue (silica gel, 2:1 hexanes:ethyl

acetate) yielded the diol 134 (286 mg, 76%) as a white

solid. ¹H NMR (CDCl₃, 400 MHz) d: 4.78 (1H, dd,

J10.5, 9.6 Hz), 4.68 (1H, dd, J9.6, 9.3 Hz), 3.67±3.52

(2H, m), 3.44 (1H, dddd, J11.3, 11.3, 5.0, 5.0 Hz), 2.23

(1H, bs), 2.00±0.50 (25H, m), 1.93 (3H, s), 1.89 (3H, s),

0.99 (18H, m), 0.88 (3H, s), 0.86 (3H, d, J6.6 Hz), 0.61

(3H, s). ¹³C NMR (CDCl₃, 100 MHz) d: 170.91, 170.73,

77.96, 74.58, 70.46, 61.24, 55.22, 54.77, 51.81, 46.14,

43.58, 39.37, 39.02, 38.95, 37.01, 35.76, 32.48, 32.04,

30.91, 28.42, 24.83, 21.49, 21.26, 20.79, 19.04, 18.01,

13.32, 11.96, 11.86. FTIR (thin ®lm): 3401, 1944, 2890,

2867, 1746, 1464, 1377, 1244, 1100, 1046 cm²¹. High-

resolution MS (EI, m/z): 621.4565, calcd for C₃₆H₆₅O₆Si

(M1H)¹621.4550.

Article Doi

First total synthesis of xestobergsterol A and active structural analogues of the xestobergsterols

DOI: 10.1016/S0040-4020(00)01086-3

Source and publish data:

Authors:

Article abstract of DOI:10.1016/S0040-4020(00)01086-3

Full text of DOI:10.1016/S0040-4020(00)01086-3

Products guided by the article

R&D Labs maybe for 336099-88-8

Relevant to this article

Hot Product