Toward a practical synthesis of morphine. The first several generations of a radical cyclization approach

SYNTHESIS

April 1998

665

Toward a Practical Synthesis of Morphine. The First Several Generations of a Radical

Cyclization Approach¹

Gabor Butora, Tomas Hudlicky,* Stephen P. Fearnley, Michele R. Stabile, Andrew G. Gum, David Gonzalez

Department of Chemistry, University of Florida, Gainesville, Florida 32611–7200, USA

Fax +1(352)8461203; E-mail: hudlicky@chem.ufl.edu

Received 4 September 1997; revised 3 November 1997

Abstract: A radical cyclization approach to the complete skeleton

of morphine was investigated in several iterations. The ﬁrst attempt

at a radical cascade via a Bergman-type intermediate derived from

ene-diyne 10 failed during a model study in which 10-membered

silicon-tethered ene-diyne 17 proved inert to Bergman cyclization

conditions. A second model study involving ene-diyne 27, function-

alized with an allyl group, underwent Claisen rearrangement to 32

in preference to a Bergman-type cyclization. Several simple model

studies were performed with bromophenols appended to protected

diols 40 and 50, respectively, to determine the feasibility of

C12–C13 bond formation in the former case and the cascade clo-

sure C12–C13/C14–C9 in the latter via radical species generated

from the aryl halides. The second-generation approach employed

the diene diol 7a derived biocatalytically from β-bromoethylben-

zene via oxidation with E. coli JM109(pDTG601), its conversion to

cyclization precursor 55, and the radical cyclization to 56a,b. The

conditions and the outcome of this process are discussed in detail

along with the rationalization of stereochemistry of the cyclization,

which furnished C14-epi conﬁguration in 56a in low yield.

future. Equally elusive has been the effort by the synthetic

community to furnish morphine in a manner that would ri-

val isolation in terms of economics. Morphine serves as a

convenient starting material for other medicinally impor-

tant substances (Figure 1) such as codeine (2), as well as

for the (illicit) production of heroin (3). Various antago-

nists, such as naloxone (4) and naltrexone (5), are chemi-

cally synthesized via noroxymorphone (6) and its

derivatives, themselves produced from morphine by semi-

synthesis.

The third-generation synthesis relied on stepwise radical cyclization

of vinyl bromide 67 derived from o-bromo-β-ethylbenzene (also by

biocatalytic means) and equipped with an oxazolidone as the radi-

cal acceptor group. Isoquinoline derivatives 68a and 68b were

obtained as a mixture of isomers, the major of which, 68a, was con-

verted via a second tin-mediated cyclization to the pentacyclic com-

pound 78, also possessing C14-epi conﬁguration. The stepwise

radical cyclization proceeded in higher yields, produced cleaner

reaction mixtures, and was also performed with the more ﬂexible

alcohol 87, whose tin-mediated closure produced a 1:1 mixture of

C14 epimers, tetracyclic compounds 81 and 89. Finally, tetracycle

80 or pentacycle 79 was converted to oxo aldehyde 83 and cyclized

to the complete morphinan skeleton, 84, in the ent-C14-epi series.

Additionally, preliminary studies were performed on direct closures

of chloride 82 to 85, via a C10/C11 alkylation of a sp³-hybridized

center. The three generations of synthetic effort are discussed in

detail and physical and spectral data are provided for all new com-

pounds. The relative merits of tandem vs. stepwise radical cycliza-

tion are evaluated and projections for future work are indicated.

Figure 1. Common Morphine-Derived Alkaloids in Use Today

Since Gates’ first synthesis^8aof morphine in 1956 there

have been fewer than twenty total or formal total synthe-

ses of morphine⁸reported in the literature, all recently re-

viewed.⁹The most efficient to date, that of Rice,⁸ⁱ

provided the title alkaloid in 29% overall yield, still not

sufficiently competitive with isolation. There have also

been many ingenious approaches to the architecturally

difficult skeleton of morphine.¹⁰It is clear from the com-

bined synthetic experience of those organic chemists who

have together committed 40+ years of serious effort to-

ward a concise and practical synthesis that morphine

stands out as a target of difficulty. In this manuscript, we

report the first few iterations of an approach based on a

cascade of bond-forming reactions which furnished the

skeleton of 1 in 13 steps: an accomplishment far removed

from the goal of achieving the entire synthetic sequence in

under 8 steps; nevertheless, it is a valuable first step.

Key words: enzymatic dioxygenation, radical cyclization, iso-

quinoline alkaloids, approach to (+)-morphine, ent-morphinans

Introduction

The consumption of morphine¹(1) in the United States is

approaching one hundred metric tons annually.²The

world’s oldest drug on record,^3atoday used routinely as

both an analgesic and an anesthetic,^3bis available by com-

mercial processing of raw opium from Papaver somnifer-

um grown in those latitudes that favor the biogenesis of

morphinans.⁴As of this writing, all morphine for medici-

nal as well as illicit use originates in natural sources. De-

spite years of research focused on plant tissue culture

production and microbial transformations of morphine

alkaloids⁵as well as the enzymology of morphine biogen-

esis,⁶the probability of large scale production by biologi-

cal means remains uncertain, at least for the foreseeable

Results and Discussion

We selected several topologically independent strategies

to approach the morphine synthesis in a practical manner.

Biomimetic in principle continues to be our approach via

the intramolecular Diels–Alder reaction, the initial

results¹¹of and recent progress¹²of which, have been dis-

closed. The approach discussed herein evolved from con-

siderations of a radical cascade as the most efficient

means of the rapid assembly of the carbon skeleton (Fig-

SYNTHESIS

Papers

666

ure 2). The question of stereoselectivity in such a cascade

was unclear, especially with respect to the C14 stereo-

center, which has proved notoriously difficult to control in

many of the reported syntheses. Nevertheless, we initially

placed the overall brevity of the assembly above stereose-

lectivity issues in ultimate importance.

The Bergman Cascade Approach. The idea of simulta-

neously creating ring A and a phenyl radical for a cycliza-

tion cascade¹³is shown in Figure 2. The original concept

assumed that diene 7a^14awould be derived via fermenta-

tion of β-bromoethylbenzene with P. putida¹⁵and its re-

combinant organisms available through the efforts of

Gibson.¹⁶This particular diene diol and others derived

from β-substituted ethyl benzenes, have recently been

used in several synthetic ventures.^12,17,18

After alkylation of oxazolidone 8 with 7b protected at

the distal hydroxyl, the free alcohol was envisioned as a

nucleophilic partner for the coupling with diyne epoxide

9 in an overall equivalent of the Williamson ether syn-

thesis of the intermediate aryl ether 11. Note that the

electronics of this design are opposite to those of tradi-

tional Williamson syntheses, i.e., the alkyl, not the aryl

unit bears the nucleophilic oxygen. The oxidation of the

trans-diol derived from 9 would provide the hydrogen

bonded species 10 whose tendency toward Bergman cy-

clization should be the function of interatomic distance

c–d between the acetylenic termini. The aryl diradical 11

would undergo cyclization and yield 12 after protodesi-

lylation of the silicon tethers introduced to control c–d

distance in 9 and 10. The precedent for cyclization of 10-

membered ene-diynes that possess the structural features

of 9, found in compounds 13–16, was indeed favorable

as shown in Figure 3. Molecular dynamics (Cache) per-

formed on a truncated version of 10, namely 17, predict-

ed a cyclization temperature of 60–75°C with c–d

distance of 3.38 Å. From the literature data it is expected

that the ene-diynes having a c–d distance less than

3.20 Å suffer spontaneous closure while those with a c–d

distance greater than 3.20–3.31Å are stable at ambient

temperature.²¹

Figure 3. Cyclization Parameters of some Bergman Precursors

Reagents: i LiOH, THF, H₂O, ii BuLi, THF, iii bis(chlorodimethylsi-

lyl)ethane, iv EtBr, Mg, v H₃O⁺, vi CrO₃, vii SmI₂, viii Bu₂SnO, allyl

bromide, ix Tf₂O, pyridine, CH₂Cl₂then DBU, r.t., x CBr₄, PPh₃,

CH₂Cl₂, xi Na₂S, xii MCPBA, xiii a) SO₂Cl₂, CH₂Cl₂, –78°C b)

MCPBA, CH₂Cl₂, 0°C, xiv MeLi.

Scheme 1

With this information we undertook the synthesis of a

model system with structural features of 10 as shown in

Scheme 1. The ene-diyne 21 was synthesized in yields not

exceeding ~20% by two rather laborious routes. In the

first of these, Vollhardt’s method of coupling cis-dichlo-

roethylene with trimethylsilyl acetylene²⁵was adapted to

produce 18 which was subjected to desilylation in aque-

Figure 2. Bergman Cyclization Approach to Morphine

SYNTHESIS

April 1998

667

ous LiOH and the free bisacetylene 19 metalated with

BuLi. The dianion 20 was quenched to produce the ten-

membered ene-diyne 21 in low yields, with recovery of

the material complicated by its volatility. The second

route, Scheme 1, relied on the SmI₂-mediated McMurry-

type coupling²⁶of the known dialdehyde²⁷24 derived by

oxidation of the corresponding diol. This material was

available in reasonable yield by a known procedure from

protected propargylic alcohol 23. Diol 25 was initially

subjected to several procedures for elimination, including

a Corey–Winter²⁸reaction which had shown some success

for Semmelhack.²⁴No ene-diyne was detected in any of

these reactions. We chose an alternate route from diol 24,

which was converted, via dibromide 28, to a cyclic sulfox-

ide, and ultimately taken to the Ramberg–Backlund pre-

cursor 31.²⁹Treatment of the chlorosulfone with MeLi at

–78°C in diethyl ether afforded a low yield of 21 (<20%).

Disappointingly, it has been shown to be stable to ther-

molysis up to 225°C and no cycloaromatization to the ex-

pected and independently synthesized product 21a was

detected. Despite the predicted geometry and c–d dis-

tance, which bode favorably for a low reaction tempera-

ture, the material proved inert and this was attributed, at

least for the moment, to the electronic effect of tethering

silicon atoms. Several cases reported in the literature de-

scribe long-range effect of silicon tethers on the cycliza-

tion rates and attendant temperatures occasionally

requiring > 600°C.²⁷The final evidence for the probable

lack of utility of this cascade approach became available

by thermolysis of enol ether 27, whose cyclization param-

eters portrayed, quite accurately, those foreseen in com-

pound 10. The enol ether was prepared from diol 25 by

monoallylation³⁰and subsequent elimination of the allyl

mesylate. At the predicted cyclization temperature of

65°C this material was found completely inert. After

60 hours at 80°C the starting material was consumed. No

evidence of Bergman-type products was found but ketone

32 was isolated in 43% yield, indicating that the energet-

ics for the Claisen rearrangement in 27 are more favorable

than Bergman cyclization. The project was abandoned at

this point until such time that a solution to the detrimental

influence of silicon tethers is found.

Figure 4. Tandem Radical Cyclization Approach

chol dehydrogenase to a catechol which then suffers an ortho

cleavage and is ultimately channeled through further enzy-

matic steps to acetate as a carbon source for the organism.

Gibson developed not only blocked mutants¹⁵of Pseudomo-

nas but also recombinant strains¹⁶of E. coli in which further

enzyme expression beyond either the first step [P. putida 39/

D; E. coli JM109(pDTG601)] or the second step [E. coli

(pDTG602)]¹⁶is arrested. Thus the required catechol 33 can

be obtained from bromobenzene by the use of the latter mu-

tant organism and one further chemical step, namely selec-

tive alkylation of the “distal” phenol group.

Model Study: C12–C13 bond. To validate the chemistry

contained in this approach, we chose the simplest system,

capable of only a single radical closure, as shown in

Scheme 2. The required catechol 33 was prepared from

bromobenzene or 2-methoxyphenol as previously report-

ed by Hoshino³²and used as a nucleophile in the second

Mitsunobu inversion³³of the alcohol 42 derived from tol-

uene diol via alcohol 40.¹¹For the initial model study the

thexyl derivative 38 was chosen because we had not yet

manufactured the catechol enzymatically and this material

was prepared by a protection/oxidation sequence from 36.

The Tandem Radical Cyclization. Because of the failure

of the Bergman cyclization approach, in which the aryl

radical would have been generated in situ, we chose a

more direct route where the required radical species orig-

inated in the reduction of an aryl halide. One of the appar-

ent advantages of this approach, depicted in Figure 4, was

the possibility of incorporation of several enzymatic

transformations into the synthetic pathway. Diol 7a re-

mains accessible from β-bromoethylbenzene through the

first oxidation step in the natural degradation pathway of

arenes in the wild strains of Pseudomonas putida.³¹

Reagents: i P. putida TG02C, ii JM109 (pDTG601), iii Jones oxidati-

on, iv MeI, acetone, v TMSI, vi THSCl, imidazole, DMF, vii oxalyl

chloride, DMSO, Et₃N, -78°C to 0°C, viii potassium azodicarboxyla-

te, HOAc, ix benzoic acid, Bu₃P, DEAD, THF, x NaOMe, MeOH, xi

38, Bu₃P, DEAD, THF, xii H₃O⁺, xiii benzyl bromide, K₂CO₃, ace-

tone, xiv Bu₃SnH, AIBN, toluene, reflux.

The second step in the pathway utilized by the wild strain

consists of further oxidation of the arene-cis-diol by cate-

Scheme 2

SYNTHESIS

Papers

668

The first Mitsunobu inversion provided the benzoate 41, idation. [Note: The compounds synthesized in the course

from which transesterification with NaOMe in MeOH lib- of the model studies as well as those prepared during the

erated the alcohol 42. The second Mitsunobu reaction of study of the Bergman cyclization (21, 25–32, 41–47) were

alcohol 42 and thexyloxy bromophenol 38 gave ultimate- characterized only to permit the ultimate judgment as to

ly derivative 44, after the labile thexyl group in the con- the future merits of that particular approach. All of the

densation product had been replaced by a benzyl moiety compounds have either led to a dead end investigation or

in 43. The protected ether 44 was exposed to AIBN/ validated the completion of a preliminary model study and

Bu₃SnH in refluxing toluene to afford a 50 % yield of tri- therefore were not characterized as fully as the intermedi-

cycle 45 containing three of the five stereogenic centers of ates in the subsequent application (see Experimental Sec-

morphine in the correct absolute configuration. With this tion for partial characterization of these intermediates)].

promising result we turned to the assembly of the entire

Tandem Radical Cyclization: First Generation. Based on

morphine skeleton.

the results of the simple model studies we turned to the ob-

Model Study: C14–C9 bond. To provide initial informa- jective of assembling the entire morphine skeleton. β-Bro-

tion about the relative stereochemistry at C14–C9, tradi- moethylbenzene was subjected to enzymatic dihy-

tionally the pitfall of most syntheses of morphine, we droxylation to yield the homochiral diol 7a^14awhich was

designed a simple model cyclization shown in Scheme 3. reduced to give 49,^14aprotected as a TBS-ether at the dis-

The less substituted alkene in diene diol 7b was reduced tal hydroxyl group (50) and further transformed via a Mit-

with diimide and the C6-hydroxyl protected as a TBS- sunobu inversion to benzoate 51, in 85 % yield (Scheme

ether (See structures 49 and 50, Scheme 4). Mitsunobu re- 4). The alkylation of this material with oxazolidone 8 did

action with o-bromophenol, followed by displacement of not proceed without problems. As reported in a prelimi-

the primary alkyl bromide with the sodium salt of oxazoli- nary communication, elimination to diene 53 took place

done 8, furnished the protected trans-diol 46 which was quite readily; in principle this diene could be “recycled”

exposed to Bu₃SnH/AIBN in refluxing toluene to furnish via a hydroboration/oxidation/displacement sequence to

1

pentacyclic product 47 in approximately 10% yield. H furnish better yields of 52 (See Scheme 6 for recycling of

NMR analysis established the relationship of the C14–H a similar compound, 71^14b). Hydrolysis of the benzoate

and C9–H as trans; however, it was not possible to discern and Mitsunobu inversion of the alcohol with bromophenol

the configuration relative to C5 or C6 and the product was 33 gave the penultimate precursor for the radical cycliza-

assigned as either 47a or 47b.

tion, ether 55, which was exposed to (TMS)₃SiH/AIBN to

afford a complex mixture of more than six products. La-

borious chromatographic separation provided no more

than a modest 15% yield of the pentacyclic material 56a,

whose analysis by NMR techniques established the epi-

C14 configuration as shown. Although not identified un-

ambiguously, the isomer 56b was also isolated as a minor

Reagents: i potassium azodicarboxylate, AcOH, ii TBSOTf, iii o-bro-

mophenol, Bu₃P, DEAD, THF, iv NaH, 2-oxazolidone, v Bu₃SnH,

AIBN.

Scheme 3

The results of the two aforementioned model studies val-

idated the premise of the tandem radical cyclization con-

sidered for the assembly of the morphine skeleton.

Conceptually similar to the strategy employed by Parker^8p

(except for the connectivity of C11–C10–C9 atoms) the

proposed sequence would create the two crucial bonds

(C12/C13 and C14/C9) in a cascade manner and it was as-

sumed that the C14/C9 stereochemistry would be subject

to some degree of control (vis-à-vis Parker’s experience).

The two novel issues in our approach were the anticipated

C10/C11 closure of an otherwise intact skeleton and the

incorporation of chirality into ring C by the enzymatic ox-

Reagents: i JM109(pDTG601), ii potassium azodicarboxylate, AcOH,

iii TBDMSOTF, CH₂Cl₂, iPr₂NEt, iv benzoic acid, nBu₃P, DEAD,

THF, v 2-oxazolidone, NaH, DMSO, vi aq NaOH, vii 33, nBu₃P,

DEAD, THF, viii (TMS)₃SiH, AIBN, benzene, 140°C, sealed tube

Scheme 4

SYNTHESIS

April 1998

669

product. Among the various byproducts isolated from the

reaction mixture we also identified the enol ether 58

which probably originated in the hydrogen abstraction

route shown in Figure 5. The stereochemistry of the two

protons in 58, corresponding to C14 and C9 of morphine,

was established as trans, but the relative stereochemistry

with respect to C6 remained undetermined.

A few comments regarding the course of this cascade se-

quence are in order. First, the reaction produced complex

mixtures which proved very difficult to separate and mon-

itor by traditional methods (TLC, HPLC). Second, the

course of the crucial C14–C9 bond formation would be-

come extremely difficult to evaluate without the accurate

knowledge of the stereochemical identity of all products

which originated in the double closure. Third, the confor-

mations of the radical species at C14 prior to the closure

will depend on the bulk of the substituent at the C6 oxygen

as well as the exact placement of the oxazolidone ring at

the time of bond formation. (See Figure 6 for the likely

possible conformation of the intermediate radical species

59). Finally, the radical closure of a species derived from

the unreduced diene 60 would produce the allylic radical

61 (Figure 7) whose fate in further bond formation would

be affected by the additional 10–12 kcal/mol of allylic sta-

bilization. The comparison of the four possible conforma-

tions in both 56 and 62 indicates that the transition state

“b”, leading in both cases to the correct C14–C9 stereo-

chemistry, is the least concave of the four conformers.

Figure 6. Conformations of the Four Possible Isomers at C14/C5

Figure 5. Formation of Enol Ether 58 by Intramolecular H-Abstrac-

tion

The possibility for stereocontrol through the manipulation

of the peripheral protecting groups could be pursued in fu-

ture studies. The conformations of 56 and 62 illustrate the

difficulty in predicting the precise outcome of C14–C9

stereochemistry. Consequently, we turned to the examina-

tion of a stepwise radical cyclization approach in which

the C9 stereochemistry would be set in the first step and

the steric fate of C14 would become dependent on the fa-

cial selectivity of a hydrogen abstraction process, rather

than the relatively slower bond-forming process. As it

turned out following this particular study, the two strate-

gies can be related analytically with regard to the confor-

mations (See Figures 6 and 7).

Figure 7. Conformation of Isomers in Which Ring C Retains Unsatu-

ration

SYNTHESIS

Papers

670

Stepwise Radical Cyclization: Second Generation. The from the base-induced elimination. This material was “re-

biooxidation of o-bromo-β-bromoethylbenzene was ex- cycled” as shown in Scheme 6 by a hydroboration/oxida-

amined with the expectation that the larger β-bromoethyl tion/mesylation sequence, to provide useful quantities of

group would direct the enzymatic placement of the cis-di- the cyclization precursor.

ol. Fortuitously, this turned out to be the case and diol

64^14bwas isolated from the broth of fermentation of 63

with recombinant organism E. coli JM109(DTG601A),

(Scheme 5).

Reagents: i 2-oxazolone, NaH, DMSO, ii 9-BBN, THF, followed by

H₂O₂/NaOH, iii MsCl, iPr₂NEt, CH₂Cl₂.

Scheme 6

Exposure of 67 to Bu₃SnH/AIBN provided octahydroiso-

quinolines 68a and 68b in a 2:1 ratio in favor of the isomer

possessing the epi-C9 configuration. The lack of selectiv-

ity in the closure may be attributed to the marginal steric

effect of the distant acetonide moiety. The combined yield

of the cyclization was high and the two compounds were

easily isolated from a clean reaction mixture. The stereo-

chemistry of 68a was determined by a single crystal X-ray

Reagents: i JM109(pDTG601A), ii potassium azodicarboxylate,

AcOH, iii 2,2-dimethoxypropane, pTSA, iv 2-oxazolone, NaH,

crystallography of the free diol 74.¹

DMSO, v nBu₃SnH, AIBN, benzene, reflux, vi Cs₂CO₃, acetone,

vii Et₄NBr, BF₃.Et₂O, CH₂Cl₂.

Scheme 5

The structure and absolute stereochemistry of 64^14bwas

determined, along with the absolute stereochemistry of

7a,^14aby correlation with the styrene-derived cis-diol

69,³⁴whose absolute configuration has been firmly estab-

Because of the greater availability of 68a we chose to pur-

lished.³⁵The yields of 64 were low and the mass balance

sue the synthesis of the ent-morphine skeleton, since only

of 64 and starting material 63 proved extremely low. We

attributed this yield problem to the differences in the intra-

cellular transport rates for the two compounds. While the

yield of 7a is among the highest for any arene cis-diol

(>10 g/L), diol 64 became available in 200 mg/L with

over 50% of the fate of 63 unaccounted for. Nevertheless,

we proceeded with the synthesis under the assumption

a single Mitsunobu inversion is required for the introduc-

tion of the catechol moiety. An assumption was made that

the synthesis of ent-morphine would be an appropriate

model study for the approach to the natural enantiomer

from 68b at such time as the control in setting the C9 cen-

ter would either be absolute in either direction, or at least

lead to a 1:1 mixture of isomers. In the latter event the syn-

that the biooxidation would be optimized in the future,

thesis would be enantiodivergent from the same interme-

were this approach found synthetically viable. Note also

diate, mimicking the strategy of Rice.⁸ⁱ

that the presence of benzofuran derivatives 70a and 70b

A Mitsunobu inversion with phenol 33 generated the pre-

cursor for the second cyclization, ether 76,¹whose expo-

sure to Bu₃SnH/AIBN gave a clean yield of pentacycle

78, Scheme 7.¹The stereochemistry at C14 was deter-

mined to be in the epi configuration (the coupling constant

between C14 and C9 was found to be 11 Hz, see Experi-

mental Section), presumably because the final act of hy-

drogen atom abstraction occurred from the convex face of

the conformer 86a. A reasonable explanation of this stereo-

chemical outcome is to invoke the rigidity of the 6-5 sys-

tem and the reluctance to favor the conformation 86b in

which the concave face would be more accessible. That

contributed to the low yield of either 64 or 65. In the latter

case 65 was regenerated from 70b as shown in Scheme 6

and as previously described.^14b

Reduction of the diene in 64 with potassium azodicarbox-

ylate followed by the introduction of oxazolone provided

67 along with a substantial amount of styrene 71 resulting

SYNTHESIS

April 1998

671

the reason for exclusive epi-C14 quench was not the pres- of a C10–C11 bond on an intermediate in which the ben-

ence of the bulky silyl group was ascertained by perform- zofuran bridge is in place and where C10 is sp³hybrid-

ing the cyclization on the free alcohol 77 and obtaining 79 ized. A preliminary evidence [HRMS (CI/methane):

with the same result: epi-C14 configuration, Scheme 7. 302.1812, (C₁₈H₂₃NO₃+H) requires 302.1756; (FAB):

The above arguments became even more credible when 324.1574, (C₁₈H₂₃NO₃+Na) requires, 324.1575] indicated

the cyclization was performed on the more flexible system that upon exposure to AlCl₃in benzene pentacycle 85 was

such as alcohol 87, Scheme 8. The reduction of the ox- produced. The minute scale of the reaction precluded full

azolidone moiety, usually performed on 78 after the cy- characterization of products. Future work will need to ad-

clization to generate 81 and prepare the C10 center for the dress this procedure on a scale where manipulation of alu-

final connection to C11, was done with 76 instead, gener- minum chloride catalyst proceeds without substantial

ating 87 which was subjected to the radical cyclization. hydrolysis.

With this substrate, a 1:1 mixture of C14 epimers, 81 and

89, was isolated confirming the fact that the silyl group lo-

cated on the concave face of the intermediate radical 88

had no influence on the final quenching event. Rather it is

the increase in the flexibility of the system that allows 88

to adopt the less concave conformation leading to 89 in

addition to that leading to 81, Scheme 8. In future studies,

the control of stereochemistry will be enhanced by design-

ing a cyclization precursor that will favor maximization of

the conformation apparently operating in the formation of

89.

Reagents: i DIBAL-H, CH₂Cl₂, ii nBu₃SnH, AIBN, benzene, reflux.

Scheme 8

Stereochemistry of Radical Cyclization. From the results

obtained in the parallel experiments with 76, 77, and 87

several rationalizations can be extended also to the expla-

nations of results in the cascade cyclization. Figure 6

showed the four possible outcomes of a radical cyclization

that led to the formation of 56a and small amounts of 56b.

Reagents: i TBSTf, iPr₂NEt, CH₂Cl₂, –78°C, ii 33, nBu₃P, DEAD,

The transition state operating in the closure of the radical

THF, iii nBu₃SnH, AIBN, benzene, reflux, iv TBAF, THF, v DIBAL-

59 may resemble the conformations of the product with a

substantial sp²character still connected at C14. In such a

case, the course of this cyclization would parallel the

events described for the transformations of 76 to 81 and

87 to 81 and 89. If this argument is correct, then perform-

ing the tandem cyclization on a more flexible system, such

as 91, may lead to the enhancement of the tendency to as-

sume conformations that will maximize production of

56b. The “open” form, 91, can be synthesized easily by a

conversion of halide or azide of the side chain to a car-

bamate and its condensation with α-bromoacetaldehyde

(or its dimethylketal), obtained by the method of Kraus by

ozonolysis of 1,4-dibromobut-2-ene.³⁶The closure of the

radical such as 90 may lead to maximization of the correct

C14/C9 absolute stereochemistry through the less convex

shape.

H, CH₂Cl₂, vi oxalyl chloride, DMSO, Et₃N, CH₂Cl₂, vii trifluorome-

thanesulfonic acid, viii MsCl, Et₃N, THF, ix AlCl₃, benzene, reflux.

Scheme 7

To complete the synthesis of the ent-morphinan skeleton,

the pentacyclic compound 78 was deprotected, the ox-

azolidone moiety was reduced to yield diol 80 which was

subjected to a double Swern oxidation to yield the rather

unstable oxo aldehyde 83. Exposure of this material to tri-

fluoromethanesulfonic acid led to the formation of alco-

hol 84, containing the complete morphinan skeleton. The

synthesis terminated here, even though a reduction of this

compound would furnish ent-epi-dihydrocodeine based

on the known stereochemistry of the reductions at C6. Ad-

ditionally, conversion of 81 via its mesylate to chloride 82

provided the opportunity to perform the first-ever closure

SYNTHESIS

Papers

672

Another way in which the more “planar” (or less convex) explored for the step-to-step manipulation of these syn-

conformation required for 56b might be allowed would be thons in terms of both procedures and yield. Electrochem-

by performing the cyclization on 60 in which the original ical means of oxidation, reduction, and C–C bond

double bonds of the arene cis-diol 7a have been retained, formation will specifically be looked at in this regard as a

Figure 7. Formation of radical 61 and its further behavior promising replacement technology for the more tradition-

should parallel the arguments offered above: in order to al methods. We look forward to reporting further results

maximize the conformation leading to the correct isomer of our quest for morphine synthesis in the future.

62b, the systems must attain a more “planar” configura-

tion. This can either be achieved by the introduction of the

1,1,8,8-Tetramethyl-1,8-disilacyclodeca-2,6-diyne-4,5-diol (25):

The corresponding dialdehyde derived from diol 24 (2.4 g, 9.6 mmol)

additional sp²centers into the C ring or by not restricting

the flexibility of conformers with the rigid oxazolidone

ring. It is relatively easy to see in Figures 6 and 7 that of

the four conformations, the three that lead to the incorrect

epimers are all relatively convex while the one leading to

the correct configuration at C14 and C9 is relatively flat.

was added neat to a solution of SmI₂[320 mL, 31.6 mmol (0.1 M so-

lution in THF)], under vigorous stirring at r.t. The initial deep blue

color of the SmI₂solution changed to dark orange within 30 seconds

following the addition of the dialdehyde. After stirring for 15 min, a

precipitate developed. The lanthanide species was then dissolved in

4 mL of 0.1 N aq HCl solution. Evaporation of the combined organic

layers afforded 25 (2.12 g, 87% crude) as a brown foam; mp

113–114°C (benzene/hexanes); R_f= 0.3 (2:1 hexanes/EtOAc).

At this point, the best assumption for the next generation

improvement would be the following:

IR (CCl₄): ν = 3375 (br), 3005, 2961, 1407 cm^–1.

¹H NMR (400 MHz, CDCl₃,): δ = 4.40 (d, J = 4.0 Hz, 2H), 2.25 (d, J

= 4.0 Hz, 2H), 0.74 (s, 4H), 0.17 (s, 12H).

1. Perform the stepwise cyclization on the flexible hy-

droxy methyl compound and study the influence of hy-

drogen donor groups attached to the hydroxyl.

2. Perform the tandem cyclization of the “open form” of

oxazolidone which maintains the radical recipient

functionality (i.e., enol ether 90).

¹³C NMR (100 MHz, CDCl₃,): δ = 104.5, 93.6, 68.9, 7.9, –2.5, –3.0.

MS: m/z = 252 (M⁺, 10), 235 (100), 145 (45).

1,1,8,8-Tetramethyl-1,8-disilacyclodeca-2,6-diyne-5-(2-propenyl-

oxy)-4-ol (26):

3. Devise other methods of cyclization for octahydroiso-

quinoline 68 (electrochemical or acid-catalyzed), and

study the stereochemical outcome.

A 250 mL flask containing diol 25 (1.78 g, 7.06 mmol) was charged

with benzene and Bu₂SnO (2.10 g, 8.48 mmol). The mixture was re-

fluxed overnight with azeotropic removal of H₂O until the Bu₂SnO

was completely dissolved, indicating the end of the reaction. After

evaporation of the solvent, 4 g of product were obtained which, with-

out further purification was dissolved in excess allyl bromide. After

2 days at reflux, the starting material was consumed and silica gel was

added to the mixture. The solvents were removed under reduced pres-

sure and the residue was purified by column chromatography (hex-

anes/EtOAc 4:1) to furnish the monoallyl ether 26 (740 mg, 36%) as

a yellow oil; R_f= 0.5 (hexanes/EtOAc 10:1).

The attainment of the correct configuration at C14 relative

to C9 constitutes one of the pivotal problems in morphine

synthesis as the majority of total syntheses attests to. The

multigeneration approach to the synthesis of this alkaloid

will hopefully resolve this problem adequately in the not

too distant future.

IR (neat): ν = 3400, 2945, 2900, 2160,1250 cm^–1.

¹H NMR (400 MHz, CDCl₃,): δ = 5.91 (m, 1H), 5.33 (d, J = 16 Hz,

1H), 5.23 (d, J = 12 Hz, 1H), 4.53 (dd, J = 7, 4 Hz, 1H), 4.29 (dd, J =

12.5, 6 Hz, 1H), 4.21 (d, J = 7 Hz, 1H), 4.03 (dd, J = 12.5, 6 Hz, 1H),

2.45 (d, J = 4 Hz, 1H), 0.74 (s, 2H), 0.72 (s, 2H), 0.16 (s, 6H), 0.15 (s,

6H).

¹³C NMR (67.5 MHz, CDCl₃,): δ = 133.7, 118.2, 103.5, 102.1, 94.3,

92.8, 75.1, 70.6, 67.2, 7.7, –2.3, –3.1.

MS: m/z = 292 (M⁺, 2%), 251 (18), 145 (100).

1,1,8,8-Tetramethyl-1,8-disilacyclodeca-2,6-diyne-5-(2-propenyl-

oxy)-4-ene (27):

To the allyl ether 26 (190 mg, 0.651 mmol) in CH₂Cl₂at 0°C was add-

ed pyridine (0.13 mL, 1.628 mmol) followed by dropwise addition of

triflic anhydride (0.16 mL, 0.976 mmol). After stirring at 0°C for

15 min, the mixture was poured into 1% HCl (10 mL), and extracted

with CH₂Cl₂. The combined organic phases were dried (Na₂SO₄), fil-

tered and evaporated. The mixture was quickly passed through a plug

of silica (hexanes/EtOAc 25:1) to give the corresponding triflate

(89.6 mg, 32%) as a colorless oil, which was used immediately in the

next reaction.

Conclusions

The first few steps that were necessary to evaluate the rad-

ical cyclization approach to the morphine skeleton have

been completed. The traditional problem of controlling

the C14 stereochemistry materialized in our approach as

well and in the future ameliorations serious attention will

be given to the events that control the conformations of ei-

ther the cyclization precursors (cascade approach) or the

cyclized radical species just before the hydrogen atom ab-

straction (stepwise approach). To achieve the promise of

a practical synthesis of this alkaloid, as insinuated in the

title of this paper, careful attention must also be given to

the overall number of operations and to the practical exe-

cution of each step. To this end, the enzymatic generation

of the key building blocks will almost certainly be re-

tained. On the other hand, more effective means will be

The triflate was dissolved in benzene (6 mL). DBU (0.035 mL,

0.212 mmol) was added dropwise at r.t. After stirring for 15 min, the

solution turned aquamarine in color, silica gel (100 mg) was added,

and the solvent was evaporated. The mixture was applied directly to

a flash silica column (hexanes/EtOAc 98:2) to yield 27 (28 mg, 48 %)

as a colorless oil; R_f= 0.3 (hexanes/EtOAc 98:2).

IR (neat): ν = 2817, 2120 cm^–1.

¹H NMR (400 MHz, CDCl₃): δ = 5.91 (m, 1H), 5.31 (s, 1H), 5.33 (d,

J = 16 Hz, 1H), 5.26 (d, J = 16, 1H), 4.39 (d, J = 5.6 Hz, 2H), 0.39 (s,

4H), 0.04 (s, 12H).

SYNTHESIS

April 1998

673

¹³C NMR (67.5 MHz, CDCl₃): δ = 148.1, 132.2, 118.5, 101.2, 98.8,

95.2, 84.1, 70.2 (di), 9.7, 7.9, –0.3, –2.3.

MS (EI): m/z = 274 (35%), 245 (41), 73 (100).

mixture was washed with NaHCO₃solution (10 mL), and H₂O

(10 mL), dried (MgSO₄), filtered, and evaporated to afford 425 mg of

a foamy residue. Purification by flash column chromatography

(CH₂Cl₂/hexanes 10:1) afforded 31 (263 mg, 90%) as a pale yellow

solid; R_f= 0.5 (CH₂Cl₂/hexanes 10:1).

Bis[1,2-dimethylsilyl-2-(propynylbromo)]ethane (28):

To a stirred solution of diol 24 (11.0 g, 43.3 mmol) and CBr₄(27 g,

86.6 mmol) in CH₂Cl₂(200 mL) at 0°C was added dropwise a solu-

tion of PPh₃(34 g, 130 mmol) in CH₂Cl₂(100 mL). The mixture was

stirred overnight and the resulting orange solution was filtered and the

solvent evaporated. Filtration of the residue and trituration with Et₂O

yielded 18 g of crude compound. The oil was purified by flash chro-

matography (hexanes/CH₂Cl₂4:1) to afford 28 (11.1 g, 68%) as a col-

orless oil; R_f= 0.4 (hexanes/CH₂Cl₂4:1).

Found: C, 43.55; H, 5.80. (C₁₂H₂₀ClO₂SSi₂.0.5.H₂O) requires: C,

43.94; H, 6.14 %.

IR (CHCl₃): ν = 2980, 2910, 2190, 1800 cm^–1.

¹H NMR (270 MHz, CDCl₃): δ = 5.74 (dd, J = 12, 2.6, 1H), 4.32 (m,

1H), 4.15 (m, 1H), 0.66 (s, 4H), 0.18 (s, 12H).

¹³C NMR (67.5 MHz, CDCl₃): δ = 99.8, 94.4, 59.7, 42.9, 8.2, –2.8.

MS: m/z = 319 (2), 157 (28), 139 (100).

1,1,8,8-Tetramethyl-1,8-disilacyclodeca-2,6-4-ene (21):

Found: C, 38.00; H, 5.36. (C₁₂H₂₀Br₂Si₂) requires: C, 37.90; H,

5.31%.

To a –78°C solution of MeLi (0.47 mL of a 1.6 M solution in Et₂O,

0.75 mmol) in Et₂O (8 mL), was added sulfone 31 (200 mg,

0.63 mmol) dissolved in Et₂O (15 mL). Immediately after the addi-

tion was complete, sat. aq ammonium chloride was added and the

mixture was diluted with pentane and H₂O. The organic layer was

separated, dried (MgSO₄), filtered, and evaporated to afford 140 mg

of an orange oil. Purification by flash column chromatography (pen-

tane/Et₂O 98:2) afforded 21 (263 mg, 90%) as a colorless semi-solid;

R_f= 0.3 (pentane/Et₂O 98:2).

IR (neat): ν = 2960, 2170, 1245, 1200 cm^–1.

¹H NMR (400 MHz, CDCl₃): δ = 3.92 (s, 4H), 0.58 (s, 4H), 0.16 (s,

12H).

¹³C NMR (67.5 MHz, CDCl₃): δ = 100.5, 91.4, 14.6, 8.1, –2.6.

MS: m/z = 380 (M⁺, 2%), 223 (60), 175 (60), 147 (80).

5-Thia-1,1,8,8-tetramethyl-1,8-disilacyclodeca-2,6-diyne (29):

The dibromide 28 (1.0 g, 2.64 mmol) was dissolved in CH₂Cl₂

(26 mL) and H₂O (35 mL). After complete dissolution tetrabutylam-

monium bromide (0.27 g, 0.789 mmol) and Na₂S.9H₂O (0.76 g,

3.16 mmol) were added under vigorous stirring. After 2 h, the layers

were separated and the organic extract washed with H₂O, dried and

removed under reduced pressure to afford a pale yellow solid. Purifi-

cation by flash column chromatography (hexanes/CH₂Cl₂4:1) afford-

ed 29 (120 mg, 18%) as a white solid; mp 118–121°C; R_f= 0.4

(hexanes/CH₂Cl₂2:1).

IR (neat): ν = 2900, 2110 cm^–1.

UV (CH₃CN): λ _max= 274, 290.

¹H NMR (270 MHz, CDCl₃): δ = 6.0 (s, 2H), 0.66 (s, 4H), 0.16 (s,

12H).

¹³C NMR (67.5 MHz, CDCl₃): δ = 121.8, 103.0, 100.2, 8.7.

1,1,8,8-Tetramethyl-1,8-disilacyclodeca-2,6-diyne-5-propenyl-4-

one (32):

Enediyne 27 (3 mg, 0.01 mmol) was dissolved in benzene-d₆(0.8 mL)

and transferred to a thick-walled NMR tube. Cyclohexa-1,3-diene

(0.08 mL, 0.8 mmol) was added to the mixture, and the sample was

degassed for 30 min with a stream of Ar. After heating the sample at

55°C for 8 h, the starting material remained unchanged. The reaction

temperature was increased to 80°C, and after 60 h the starting mate-

rial was consumed. Evaporation of the solvents yielded 32 (1.3 mg,

43 %) as a yellow oil; R_f= 0.5 (hexanes/EtOAc 10:1).

IR (neat): ν = 2810, 2100, 1725 cm^–1.

Found: C, 55.44; H, 8.07. (C₁₂H₂₀SSi₂.0.5.H₂O) requires: C, 55.11;

H, 8.09 %.

IR (CCl₄): ν = 2910, 2360, 2343, 1186 cm^–1.

¹H NMR (270 MHz, CDCl₃): δ = 3.39 (s, 4H), 0.57 (s, 4H), 0.12 (s,

12H).

¹³C NMR (67.5 MHz, CDCl₃): δ = 101.1, 87.4, 19.9, 8.6, –2.3.

MS: m/z = 252 (M⁺, 1%), 224 (100), 209 (90), 147 (35).

5-Thia-1,1,8,8-tetramethyl-1,8-disilacyclodeca-2,6-diyne 5-Ox-

ide (30):

¹H NMR (270 MHz, CDCl₃): δ = 5.84 (m, 1 H), 5.12 (m, 2H), 3.27

(m, 1H), 2.50 (m, 2H), 0.23 (s, 4H), 0.17 (s, 12 H).

To a precooled (–30°C) solution of the sulfide 29 (1.4 g, 5.56 mmol)

in CH₂Cl₂(85 mL) was added MCPBA (1.6 g, 5.67 mmol). After

30 min, dimethyl sulfide (4 mL) was added and the mixture was

stirred for an additional 15 min. The reaction was concentrated under

reduced pressure, and the residue dissolved in Et₂O. The organic so-

lution was washed with sat. NaHCO₃(2 × 60 mL) and brine (1 ×

30 mL), dried (MgSO₄), filtered, and evaporated to afford 1.3 g of a

light yellow semi-solid residue. Purification by flash column chroma-

tography (Et₂O) afforded 30 (940 mg, 63%) as a white solid; mp

177–178°C; R_f= 0.5 (Et₂O).

¹³C NMR (67.5 MHz, CDCl₃): δ = 185.4, 133.6, 117.9, 106.7, 104.0,

100.1, 91.7, 48.1, 35.4, 7.9, –2.6, –3.3.

(1R,6R)-6-Thexyloxy-2-methylcyclohex-2-enyl-(2-phenylacetate)

(41):

A THF solution (5 mL) of a phosphine reagent [prepared from DEAD

(142 mg, 0.905 mmol) and tributylphosphine (225 mg, 0.905 mmol)]

was added dropwise to a precooled solution (0°C) of alcohol 40

(120.0 mg, 0.444 mmol) and benzoic acid (110.5 mg, 0.905 mmol) in

THF (5 mL) and stirred at r.t. for 20 h. The solvent was removed un-

der reduced pressure, and the residue was purified by column chroma-

tography (EtOAc/hexane 1:9) to yield 41 (113 mg, 68 %) as a viscous,

Found: C, 53.67; H, 7.58. (C₁₂H₂₀OSSi₂) requires: C, 53.66; H, 7.52 %.

IR (CCl₄): ν = 2960, 2910, 2170, 2343, 1186 cm^–1.

¹H NMR (270 MHz, CDCl₃): δ = 3.39 (s, 4H), 0.57 (s, 4H), 0.12 (s,

12H).

25

colorless oil; R_f= 0.25 (EtOAc/hexane 1:9); [α]_D+103.2 (c = 1.18,

CHCl₃).

¹³C NMR (67.5 MHz, CDCl₃): δ = 101.1, 87.4, 19.9, 8.6, –2.3.

MS: m/z = 252 (M⁺, 1%), 224 (100), 209 (90), 147 (35).

¹H NMR (400 MHz, CDCl₃): δ = 8.08 (d, J = 1.2 Hz, 1H), 8.05, (d, J

= 1.4 Hz, 1H), 7.99 (tt, J = 7.8, 1.7 Hz, 2H), 7.66 (br t, J = 1.22 Hz,

1H), 7.55 (tt, J = 7.8, 1.2 Hz, 1H), 5.40 (d, J = 4.9 Hz, 1H), 4.00 (hept,

J = 5.2, 3.1, 2.2 Hz, 1H), 2.20 (m, 1H), 2.05 (m, 1H), 1.81 (m, 1H),

1.74 (m, 1H), 1.68 (br d, 1H), 1.56 (sep, J = 6.5 Hz, 1H), 0.82 (d, 6H),

0.78 (d, 6H), 0.10 (s, 3H), 0.08 (s, 3H).

5-Thia-1,1,8,8-tetramethyl-4-chloro-1,8-disilacyclodeca-2,6-

diyne 5,5-Dioxide (31):

To a –78°C solution of sulfoxide 30 (248 mg, 0.925 mmol) in CH₂Cl₂

(20 mL) was added pyridine (0.26 mL, 3.24 mmol) followed by

SO₂Cl₂(1.9 mL of a 1 M solution in CH₂Cl₂, 1.94 mmol). After 1.5 h,

the solution was quenched with H₂O (4 mL) and warmed to r.t. The

mixture was extracted with sat. NaHCO₃solution (5 mL), H₂O

(5 mL), sat. CuSO₄solution (2 × 5 mL) and brine (10 mL). The organ-

ic solution was dried (MgSO₄), filtered, and evaporated to afford a

residue that was dissolved in CH₂Cl₂, cooled to 0°C and treated with

MCPBA (495 mg, 2.87 mmol). The mixture was stirred overnight and

then two drops of Me₂S were added. After stirring for 15 min, the

¹³C NMR (100 MHz, CDCl₃): δ = 162.3, 132.8, 130.5, 130.3, 129.7,

128.3, 126.8, 75.7, 69.7, 34.2, 27.9, 22.4, 20.2, 18.5, –2.7, –2.9.

(1R,6R)-6-Thexyloxy-2-methylcyclohex-2-en-1-ol (42):

A solution of ester 41 (214 mg, 0.571 mmol) in MeOH/NaOMe

(12 mL, trace) was stirred at r.t. for 46.5 h. The mixture was concen-

trated under reduced pressure to give a residue which was dissolved

in benzene and purified by column chromatography (EtOAc/hexane

SYNTHESIS

Papers

674

3:7) to afford alcohol 42 (143.9 mg, 93%) as a colorless, viscous oil;

R_f= 0.60 (EtOAc/hexane 3:7).

tography (EtOAc/hexane 95:5) to yield the intermediate (3S,4S)-2-(2-

bromoethyl)-3-(1-bromo-2-phenyloxy)-4-(tert-butyldimethylsilyl-

oxy)cyclohexene (132 mg, 44%).

¹H NMR (400 MHz, CDCl₃): δ = 5.41 (br s, 1H), 3.87 (br s, 1H), 3.74

(ddd, J = 10.2, 6.7, 3.5 Hz, 1H), 2.06 (m, 2H), 1.99 (d, J = 4.1 Hz, 1H),

1.76 (m, 4H), 1.63 (m, 2H), 0.88 (br d, J = 6.9 Hz, 6H), 0.84 (s, 6H),

0.14 (s, 3H), 0.12 (s, 3H).

¹H NMR (400 MHz, CDCl₃): δ = 7.53 (dd, J = 7.9, 1.5 Hz, 1H), 7.26

(m, 1H), 7.24 (dt, J = 7.2, 0.9, 1H), 6.83 (dt, J = 7.5, 1.4 Hz, 1H), 5.80

(br s, 1H), 4.63 (br d, J = 4.7 Hz, 1H), 4.04 (m, 1H), 3.41 (dd, J = 7.9,

6.9 Hz, 2H), 2.80 (m, 1H), 2.56 (m, 1H), 2.20 (m, 2H), 1.90 (m, 1H),

1.72 (m, 1H), 0.82 (s, 9H), 0.01 (s, 3H), –0.11 (s, 3H).

¹³C NMR (100 MHz, CDCl₃): δ = 155.3, 133.5, 132.2, 129.5, 128.3,

122.1, 115.0, 112.7, 79.4, 70.0, 37.5, 31.9, 27.8, 25.7, 22.4, 18.9,

–5.0.

¹³C NMR (100 MHz, CDCl₃): δ = 133.6, 123.8, 75.1, 74.0, 34.2, 29.7,

28.5, 23.6, 20.4, 19.4, 18.6, –2.3, –2.8.

3-Bromo-2-{[(1R,6S)-6-thexyloxy]-2-methylcyclohex-2-enyl}oxy-

phenol (43):

To a precooled solution (0°C) of alcohol 42 (75 mg, 0.277 mmol) and

Bu₃P (112 mg, 0.554 mmol) in THF (2 mL) was added dropwise

DEAD (96.6 mg, 0.554 mmol). After stirring for 2 min a solution of

phenol 38 (110 mg, 0.333 mmol) in THF (2 mL) was added dropwise.

The cooling bath was removed and the mixture was stirred for 12.5 h.

The mixture was concentrated under reduced pressure to afford

42.6 mg of crude silyl protected 43. This was dissolved in THF (5

mL), and treated with TBAF (57 mg). After 9 h of stirring at r.t., H₂O

(15 mL) was added, and the THF distilled off under reduced pressure.

The product was extracted with CH₂Cl₂(5 × 15 mL), dried (MgSO₄),

and after filtration, the solvent was removed under reduced pressure

to give crude phenol 43 (~30 mg). The crude material was taken on

directly to the next step.

To a solution of oxazolone 8 (21 mg, 0.247 mmol) in DMSO (5 mL)

was added NaH (10 mg, 0.25 mmol). The mixture was stirred for

15 min to complete the formation of the sodium salt and the interme-

diate bromide (121 mg, 0.247 mmol) in DMSO (2 mL) was added

dropwise via syringe. After stirring at r.t. for 20 h, the reaction was

diluted with Et₂O (50 mL) and quenched with H₂O (30 mL). The or-

ganic phase was separated, washed with brine (2 × 20 mL), dried

(MgSO₄), filtered and concentrated under reduced pressure. The

crude product was purified by column chromatography (EtOAc/hex-

ane 3:7) to afford 46 (68 mg, 56%) as a viscous oil.

¹H NMR (400 MHz, CDCl₃): δ = 7.64 (dd, J = 7.8, 1.5 HZ, 1H), 7.36

(dt, J = 7.3, 1.7 Hz, 1H), 7.27 (dd, J = 8.2 Hz, 1H), 6.94 (br t, J =

7.8 Hz, 1H), 6.80 (d, J = 2.0 Hz, 1H), 6.64 (d, J = 2.0 Hz, 1H), 5.79

(dt, J = 3.8 Hz, 1H), 4.80 (br d, J = 4.1 Hz, 1H), 4.13 (m, 1H), 3.74

(m, 2H), 2.70 (br m, 1H), 2.42 (br m, 1H), 2.28 (m, 1H), 2.16 (m, 1H),

2.00 (m, 1H), 1.80 (m, 1H), 0.91 (s, 9 H), 0.12 (s, 3H), 0.10 (s, 3H).

¹³C NMR (100 MHz, CDCl₃): δ = 155.0, 133.5, 130.3, 130.3, 128.5,

122.2, 116.3, 115.0, 112.5, 79.5, 69.8, 42.7, 33.9, 27.6, 25.7, 22.4,

–4.9, –4.9.

Benzyl (3-Bromo-2-{[(1R,6S)-6-thexyloxy-2-methylcyclohex-2-

enyl]oxy}phenyl Ether (44):

The solution of phenol 43 (34.5 mg, 0.0781 mmol) and benzyl bro-

mide (20.0 mg, 0.117 mmol) in acetone (2.5 mL) was treated with

K₂CO₃(55 mg, 0.390 mmol) and vigorously stirred. After 5 h, the sol-

id was filtered off and the filtrate was concentrated under reduced

pressure. The residue (41 mg) was purified by column chromatogra-

phy (EtOAc/hexanes 1:9) to afford 44 (21.4 mg, 51%) as a viscous,

colorless oil.

10-tert-Butyldimethylsilyloxy-8,16,18-dioxazapentacyclo-

1,9 2,7 14,18

[11.7.0.0 .0 .0

]icosa-2(7),3,5-trien-17-one (47):

To a degassed (Ar sparge) solution of 46 (67 mg, 0.135 mmol) in tol-

uene (10 mL) was added Bu₃SnH (78.9 mg, 73 µL, 0.271 mmol) and

AIBN (cat.). The reaction vessel was submerged into a preheated oil

bath (110°C) and allowed to reflux for 16 h. Two additional spatula

tips of AIBN were added and the reaction was stirred at reflux an ad-

ditional 5 h. The solvent was removed under reduced pressure, and

the crude product was purified by column chromatography (benzene/

Et₂O 9:1). Two compounds with similar R_fwere isolated as a mixture,

which was further purified by preparative TLC (EtOAc/hexane 1:1)

to afford pentacycle 47 (5.5 mg, 10%) as a colorless oil.

¹H NMR (400 MHz, CDCl₃): δ = 7.37 (m, 5H), 7.14 (dd, J = 7.6,

1.8 Hz, 1H), 6.87 (m, 1H), 5.53 (br s, 1H), 5.07 (s, 2H), 4.87 (t, J =

5.4 Hz, 1H), 4.22 (br s, 1H), 2.09 (m, 1H), 1.82 (m, 1H), 1.79 (s, 3H),

1.62 (p, J = 6.9 Hz, 1H), 1.55 (s, 1H), 1.41 (m, 1H), 0.88 (s, 3H), 0.86

(s, 3H), 0.82 (s, 6H), 0.11 (s, 3H), 0.09 (s, 3H).

(5aR,6S,9aS)-9a-Methyl-6-thexyloxy-5a,6,7,8,9,9a-hexahydro-

dibenzo[b,d]furan-4-yl Benzyl Ether (45):

The solution of the bromide 44 (21.0 mg, 0.039 mmol) and Bu₃SnH

(23.0 mg, 0.078 mmol) in refluxing toluene (3.0 mL) was treated with

AIBN. After 24 h of reflux, the reaction was concentrated under re-

duced pressure, and the residue was purified by column chromatogra-

phy (benzene) to yield tricyclic ether 45 (9.8 mg, 55%) as a colorless

oil.

¹H NMR (400 MHz, CDCl₃): δ = 7.26 (m, 1H), 6.97 (br t, J = 7.3 Hz,

1H), 6.86 (m, 2H), 4.48 (t, J = 8.1 Hz, 1H), 4.18 (br s, 1H), 4.10 (dd,

J = 8.7, 5.5 Hz, 1H), 3.83 (ddd, J = 13, 6, 1 Hz, 1H), 3.40 (ddd, J =

10.1, 8, 5.6, 1H), 2.80 (dt, J = 12.8, 3.7 Hz, 1H), 2.65 (dm, J = 13 Hz,

1H), 2.20 (dt, J = 10.1, 5.9 Hz, 1H), 1.50–1.80 (bm, app 6H), 0.81 (s,

9H), –0.05 (s, 3H), –0.12 (s, 3H).

¹H NMR (400 MHz, CDCl₃): δ = 7.44 (m, 2H), 7.35 (m, 2H), 7.29 (m,

1H), 6.75 (m, 2H), 6.70 (dd, J = 6.3, 2.3 Hz, 1H), 5.18 (s, 2H), 4.38

(t, J = 4.9 Hz, 1H), 3.80 (m, 1H), 2.13 (m, 1H), 1.88 (m, 1H), 1.79 (m,

2H), 1.63 (m, 2H), 1.58 (s, 3H), 1.35 (m, 1H), 0.90 (s, 3H), 0.88 (s,

3H), 0.83 (d, J = 6.0 Hz, 6H), 0.08 (s, 3H), 0.05 (s, 3H).

¹³C NMR (100 MHz, CDCl₃): δ = 157.8, 129.5, 123.3, 121.5, 115.3,

66.8, 65.1, 60.1, 41.2, 40.2, 30.6, 29.0, 28.3, 26.8, 25.7, 24.6, 19.7,

17.3, 13.6, –2.4, –2.6.

MS (CI): m/z = 416 (M+H⁺, 28%), 358 (100), 313 (25), 190 (30), 151

(30),

¹³C NMR (100 MHz, CDCl₃): δ = 147.0, 144.0, 140.0, 137.3, 128.5,

127.8, 127.4, 121.1, 114.7, 113.8, 88.8, 71.1, 66.5, 42.5, 34.1, 28.5,

24.8, 22.7, 20.2, 19.6, 18.6, –2.7, –2.9.

HRMS: 416.2276, (C₂₃H₃₃NO₄Si+H) requires, 416.2257.

(3R,4S)-2-(2-Bromoethyl)-4-(tert-butyldimethylsilyloxy)-3-hy-

droxycyclohexene (50):

(3S,4S)-3-(1-Bromo-2-phenyloxy)-4-(tert-butyldimethylsilyloxy)-

2-[2-(3-oxazolone)ethyl]cyclohexene (46):

To a stirred solution of the diol 49 (1.634 g, 7.39 mmol) in anhyd

CH₂Cl₂(50 mL) under Ar, was added diisopropylethylamine (2.70

mL, 15.52 mmol) and the mixture cooled to –78°C, resulting in a

white suspension. After 10 min, tert-butyldimethylsilyl triflate (1.87

mL, 8.13 mmol) was added dropwise, and stirring/cooling continued.

After 5 h, further TBSTf (0.85 mL, 3.70 mmol) was added dropwise,

and the stirred mixture allowed to slowly warm to r.t. overnight. After

20 h, H₂O (80 mL) and saturated aqueous NH₄Cl (30 mL) were add-

ed, and the resulting mixture extracted with further CH₂Cl₂(4 × 80

mL). The combined organic fractions were dried (MgSO₄), filtered

and reduced in vacuo to yield a brown oil. Further purification by

To a solution of alcohol 50 (204.3 mg, 0.609 mmol) and 2-bromophe-

nol (112.8 mg, 0.609 mmol) in THF (10 mL) was added the Mitsuno-

bu reagent [prepared from Bu₃P (296 mg, 1.462 mmol) and DEAD

(255 mg, 1.462 mmol) in THF (10 mL) precooled to 0°C]. The cool-

ing bath was removed and the mixture was stirred for 5 h, at which

time TLC indicated disappearance of starting material. The solvent

was removed under reduced pressure and the crude product was dis-

solved in Et₂O (50 mL). After washing with H₂O (20 mL) and brine

(20 mL), the combined organic phase was dried (MgSO₄), filtered,

and concentrated. The crude product was purified by column chroma-

SYNTHESIS

–1

April 1998

675

flash chromatography (silica ratio 50:1, hexane/EtOAc, gradient elu-

IR (thin film): ν = 2930, 1750, 1720, 1260 cm .

tion, 50:1 to 20:1), yielded silyl ether 50 (1.166 g, 47% from 7a) as a

colorless oil; R_f= 0.30 (hexane/EtOAc 19:1); [α]_D= –42.0 (c = 1.0,

¹H NMR (400 MHz, CDCl₃): δ = 8.07 (m, 2H), 7.59 (ddt, J = 8.0, 6.8,

2 × 1.3 Hz, 1H), 7.46 (m, 2H), 6.70 (d, J = 2.1 Hz, 1H), 6.53 (d, J =

2.1 Hz, 1H), 5.76 (br t, J = 4.0 Hz, 1H), 5.47 (d, J = 4.6 Hz, 1H), 4.05

(ddd, J = 7.4, 4.9, 2.7 Hz, 1H), 3.74 (ddd, J = 13.9, 7.3, 5.0 Hz, 1H),

3.60 (ddd, J = 14.2, 8.4, 7.0 Hz, 1H), 2.38 (m, 1H), 2.25 (m, 2H), 2.12

(m, 1H), 1.85 (m, 1H), 1.75 (m, 1H), 0.84 (s, 9H), 0.07 (s, 3H), 0.04

(s, 3H).

25

CHCl₃).

–1

IR: (thin film): ν = 3552, 2932, 1256, 1087 cm

.

¹H NMR (400 MHz, CDCl₃): δ = 5.65 (m, 1H), 3.90 (t, J = 3.5 Hz,

1H), 3.81 (dt, J = 10.6, 2x 3.7 Hz, 1H), 3.53 (m, 2H), 2.76 (m, 1H),

2.63 (m, 2H), 2.18 (m, 1H), 2.03 (m, 1H), 1.78 (m, 1H), 1.57 (m, 1H),

0.92 (s, 9H), 0.11 (s, 6H).

¹³C NMR (100 MHz, CDCl₃): δ = 166.2 (C), 155.5 (C), 133.1 (CH),

131.0 (CH), 129.8 (C), 129.6 (2 × CH), 129.4 (C), 128.4 (2 × CH),

116.3 (CH), 73.3 (CH), 69.4 (CH), 42.2 (CH₂), 33.1 (CH₂), 27.3

(CH₂), 25.6 (3 × CH₃), 22.1 (CH₂), 17.9 (C), –4.9 (2 × CH₃).

¹³C NMR (100 MHz, CDCl₃): δ = 134.7 (C), 127.7 (CH), 70.7 (CH),

68.6 (CH), 38.4 (CH₂), 31.8 (CH₂), 25.7 (3 × CH₃), 25.4 (CH₂), 23.9

(CH₂), 18.0 (C), –4.5 (CH₃), –4.9 (CH₃).

+

MS (EI, 70 eV): m/z = 335 (M+H⁺, 10%).

MS (EI, 70 eV): m/z = 444 (M+H⁺, 20%), 428 (M-CH₃, 42).

HRMS: 335.1042, (C₁₄H₂₈O₂⁷⁹BrSi) requires, 335.1042.

Also isolated, in varying amount, was the bis silylated compound; R_f

HRMS: 444.2204, (C₂₄H₃₄NO₅Si) requires, 444.2206.

Also isolated was the elimination product, (3R)-Benzoyloxy-(4S)-

tert-butyldimethylsilyloxy-2-vinylcyclohexene (53): (0.459 g,

11%) as a colorless oil; R_f= 0.49 (hexane/EtOAc 19:1); [α]_D²⁵+ 246.4

25

= 0.59 (hexane/EtOAc 19:1); [α]_D–42.2 (c = 1.0, CHCl₃).

–1

IR (thin film): ν = 2995, 1470, 1255, 1090 cm

.

¹H NMR (400 MHz, CDCl₃): δ = 5.65 (br t, J = 3.3 Hz, 1H), 3.96 (br

s, 1H), 3.72 (dt, J = 10.5, 2x 2.8 Hz, 1H), 3.42 (m, 2H), 2.57 (t, J =

7.9 Hz, 2H), 2.16 (m, 1H), 2.01 (m, 1H), 1.90 (m, 1H), 1.51 (m, 1H),

0.90 (s, 9H), 0.89 (s, 9H), 0.12 (s, 3H), 0.10 (s, 3H), 0.08 (s, 3H), 0.07

(s, 3H).

(c = 1.0, CHCl₃).

IR (thin film): ν = 2930, 2860, 1720, 1260, 1090 cm

–1

.

¹H NMR (400 MHz, CDCl₃): δ = 8.05 (m, 2H), 7.55 (m, 1H), 7.43 (m,

2H), 6.31 (dd, J = 17.6, 11.1 Hz, 1H), 6.11 (dd, J = 5.0, 3.0 Hz, 1H),

5.62 (d, J = 3.2 Hz, 1H), 5.11 (d, J = 17.7 Hz, 1H), 4.90 (d, J = 11.0

Hz, 1H), 4.10 (t, J = 4.9 Hz, 1H), 2.44 (m, 1H), 2.17 (dm, J = 18.9 Hz,

1H), 1.85 (dddd, J = 13.5, 10.7, 5.5, 2.0 Hz, 1H), 1.75 (m, 1H), 0.89

(s, 9H), 0.15 (s, 3H), 0.11 (s, 3H).

¹³C NMR (100 MHz, CDCl₃): δ = 136.4 (C), 125.9 (CH), 72.2 (CH),

71.2 (CH), 37.6 (CH₂), 32.0 (CH₂), 26.1 (6x CH₃), 25.4 (CH₂), 24.7

(CH₂), 18.4 (2x C), -3.6 (CH₃), –4.3 (CH₃), –4.4 (CH₃), –4.6 (CH₃).

¹³C NMR (100 MHz, CDCl₃): δ = 165.5 (C), 137.5 (CH), 134.1 (CH),

132.9 (CH), 131.7 (C), 130.3 (C), 129.7 (2 × CH), 128.3 (2 × CH),

111.2 (CH₂), 68.7 (CH), 67.5 (CH), 25.7 (3 × CH₃), 21.8 (CH₂), 21.3

(CH₂), 18.1 (C), –4.8 (CH₃), –5.0 (CH₃).

(3S,4S)-3-Benzoyloxy-4-tert-butyldimethylsilyoxy-2-(2-bromo-

ethyl)cyclohexene (51):

To stirred solution of alcohol 50 (4.609 g, 13.74 mmol) and benzoic

acid (1.846 g, 15.12 mmol) in anhyd THF (12 mL) at 0°C, was added

a solution of the Mitsunobu reagent [previously prepared by addition

of DEAD (4.786 g, 27.48 mmol) to a stirred solution of tributylphos-

phine (5.559 g, 27.48 mmol) in anhyd THF (10 mL) at 0°C], and the

mixture allowed to warm to r.t. with stirring. After 16 h, solvent was

removed in vacuo to yield a brown oil, which was prepurified by pas-

sage through a silica plug with benzene. Further purification by flash

chromatography (silica 300 g, benzene) yielded benzoate 51 (5.117 g,

84%) as a highly viscous, colorless oil, which solidified on prolonged

standing; mp 65–66°C (benzene); R_f= 0.20 (hexane/EtOAc 19:1);

+

MS (CI/methane): m/z = 359 (M+H⁺, 1%), 237 (M-PhCO₂, 52).

HRMS: 359.2030, (C₂₁H₃₁O₃Si) requires, 359.2042.

(3R,4S)-3-[2-(1-Bromo-3-methoxy)phenyloxy]-4-(tert-butyldi-

methylsilyloxy)-2-[2-(3-oxazolone)ethyl]cyclohexene (55):

To a stirred solution of benzoate 52 (3.683 g, 8.30 mmol) in MeOH

(50 mL) at r.t., was added 1 M aq NaOH (20 mL), resulting in mild

emulsification. After 6 h, the MeOH was removed in vacuo, brine

(100 mL) was added, and the resulting mixture extracted with EtOAc

(3 × 150 mL). The combined organic fractions were dried (MgSO₄),

filtered and reduced in vacuo to yield the intermediate alcohol 54 as a

highly viscous, colorless oil, which was used directly without further

purification.

¹H NMR (400 MHz, CDCl₃): δ = 6.77 (d, J = 2.1 Hz, 1H), 6.54 (d, J

= 2.1 Hz, 1H), 5.39 (br t, J = 3.0 Hz, 1H), 3.94 (m, 1H), 3.87 (ddd, J

= 6.4, 7.6, 16.0 Hz, 1H), 3.74 (ddd, J = 3.2, 5.9, 9.1 Hz, 1H), 3.64 (dt,

J = 6.4 x 2, 13.9 Hz, 1H), 2.75 (d, J = 5.6 Hz, 1H), 2.52 (m, 1H), 2.37

(m, 1H), 2.10 (m, 1H), 1.98 (m, 1H), 1.77 (m, 1H), 1.60 (m, 1H), 0.9

(s, 3H).

¹³C NMR (100 MHz, CDCl₃): δ = 156.1 (C), 132.7 (C), 127.8 (2 × C),

116.0 (CH), 73.62 (CH), 73.58 (CH), 43.13 (CH₂), 34.2 (CH₂), 27.5

(CH₂), 25.8 (3 × CH₃), 23.1 (CH₂), 18.0 (C), –4.5 (CH₃), –4.7 (CH₃).

To a stirred solution of the alcohol 54 (2.54 g, 7.48 mmol) and phenol

33 (1.52 g, 7.48 mmol) in anhyd THF (10 mL) under Ar at 0°C, was

added dropwise a solution of the Mitsunobu reagent [previously pre-

pared by addition of DEAD (2.356 mL, 14.96 mmol) to a stirred so-

lution of Bu₃P (3.727 mL, 14.96 mmol) in anhyd THF (30 mL) at

0°C], and the mixture was allowed to warm to r.t. with stirring. After

6 h, the solvent was removed in vacuo to yield a brown oil, which was

prepurified by passage through a silica plug (hexane/EtOAc 3:2). Fur-

ther purification by flash chromatography (silica ratio 200:1, hexane/

EtOAc 3:2) yielded aryl ether 55, (1.095 g, 28%) as a highly viscous,

colorless oil; R_f= 0.22 (hexane/EtOAc 2:1); [α]_D²⁵–179.5 (c = 1.0,

CHCl₃).

25

[α]_D+102.0 (c = 1.0, CHCl₃).

–1

IR: (KBr disc): ν = 2930, 1710, 1260, 1110, 840 cm

.

¹H NMR (400 MHz, CDCl₃): δ = 8.05 (dm, J = 8.5 Hz, 2H), 7.57 (ddt,

J = 7.9, 6.9, 1.4 Hz, 1H), 7.45 (tm, J = 7.6 Hz, 2H), 5.84 (br t, J =

3.6 Hz, 1H), 5.40 (d, J = 4.4 Hz, 1H), 4.02 (ddd, J = 7.0, 4.7, 2.9 Hz,

1H), 3.45 (m, 2H), 2.60 (m, 1H), 2.50 (m, 1H), 2.30 (m, 1H), 2.11 (m,

1H), 1.82 (m, 1H), 1.74 (m, 1H), 0.84 (s, 9H), 0.50 (s, 3H), 0.30 (s,

3H).

¹³C NMR (100 MHz, CDCl₃): δ = 166.1 (C), 133.0 (CH), 130.8 (C),

130.3 (CH), 130.1 (C), 129.7 (2 × CH), 128.4 (2 × CH), 73.4 (CH),

69.2 (CH), 37.5 (CH₂), 31.1 (CH₂), 27.2 (CH₂), 25.6 (3 × CH₃), 22.1

(CH₂), 17.8 (C), –4.86 (CH₃), –4.88 (CH₃).

+

MS (EI, 70 eV): m/z 439 (M , 10%), 359 (MH , 100).

HRMS: 439.1291, (C₂₁H₃₂O₃⁷⁹BrSi) requires, 439.1304.

(3S,4S)-3-Benzoyloxy-4-tert-butyldimethylsilyl-2-[2-(oxazol-2-

on-1-yl)ethyl]cyclohexene (52):

To a stirred suspension of NaH (0.3353 g, 1.40 mmol) in anhyd

DMSO (20 mL) was added, in one portion, the oxazolone 8 (1.1885 g,

13.97 mmol) and the resulting mixture stirred at r.t. for 20 min. After

this time, the mixture was cooled slightly (~10°C) and a solution of

bromide 51 (5.1168 g, 11.64 mmol) in DMSO (40 mL) was added

dropwise. The mixture was allowed to warm to r.t. with stirring. After

18 h, the reaction was cooled to 0°C and quenched with brine

(200 mL) and H₂O (200 mL). The resulting mixture was extracted

with EtOAc (1 × 200 mL, 3 × 100 mL), the combined organic frac-

tions were washed with brine (1 × 50 mL), dried (Na₂SO₄) and the

solvent was removed in vacuo to yield a pale yellow oil. Further pu-

rification by flash chromatography (silica 200 g, EtOAc/hexane 1:1)

yielded pure ester 52 (3.683 g, 71%) as a highly viscous, colorless oil;

R_f= 0.14 (hexane/EtOAc 4:1); [α]_D²⁵+ 83.7 (c = 1.0, CHCl₃).

–1

IR (KBr disc): ν = 3150, 2960, 1750, 1470, 1260, 1120, 840 cm

.

¹H NMR (400 MHz, CDCl₃): δ = 7.07 (dd, J = 7.2, 2.4 Hz, 1H), 6.79

(m, 2H), 6.70 (d, J = 1.8 Hz, 1H), 6.61 (d, J = 1.8 Hz, 1H), 5.58 (m,

1H), 5.13 (d, J = 2.9 Hz, 1H), 3.88 (s, 3H), 3.87 (m, 1H), 3.74 (m, 2H),

2.69 (m, J = ~7Hz, 1H), 2.55 (m, 1H), 2.44 (m, 1H), 2.21 (m, 1H),

2.04 (m, 1H), 1.58 (m, 1H), 0.60 (s, 9H), –0.10 (s, 3H), –0.28 (s, 3H).

SYNTHESIS

Papers

676

¹³C NMR (100 MHz, CDCl₃): δ = 155.9 (C), 151.8 (C), 146.6 (C),

131.6 (C), 131.2 (2 × CH), 125.6 (CH), 122.8 (CH), 117.5 (C), 116.8

(CH), 111.3 (CH), 78.2 (CH), 73.0 (CH), 55.6 (CH₃), 43.1 (CH₂),

35.2 (CH₂), 25.7 (CH₂), 25.6 (3 × CH₃), 24.9 (CH₂), 17.9 (C), –5.0

(CH₃), –5.4 (CH₃).

MS: m/z = 342 (M⁺², 4%), 340 (M⁺, 10), 259 (31), 183 (62), 55 (100).

HRMS: 337.9515, (C₁₁H₁₆O₂⁷⁹Br₂) requires, 337.9517.

3-[2-(6-Bromo-2,2-dimethyl-(3aS,7aR)-4,5-dihydroben-

+

zo[d][1,3]dioxol-7-yl)]ethyl-2,3-dihydro[1,3]oxazol-2-one (67):

A flame dried, Ar filled flask was charged with oxazol-2-one 8

(1.085 g, 12.75 mmol) and NaH (460 mg, 11.5 mmol, 60% susp., Ald-

rich) and cooled to 0°C. Anhyd DMSO (5 mL) was added via syringe

and the cooling bath was removed. After stirring for 15 min at r.t. (to

complete the salt formation), the mixture was placed in a ice bath and

allowed to partially solidify. The solution of the acetonide (2.168 g,

6.375 mmol) in DMSO (3 mL) was added dropwise. An additional

2 mL of DMSO was used to rinse the flask containing the acetonide and

was subsequently added to the mixture. The temperature was allowed

to rise gradually to r.t. overnight. The mixture was diluted with Et₂O

(10 mL) and quenched with H₂O/brine mixture (1:1, 10 mL). After fur-

ther extraction with Et₂O (3 × 30 mL) and drying (MgSO₄), the com-

bined organic extracts were filtered through a plug of Celite and

evaporated to dryness to give 2.4 g of a crude oil. Purification by flash

chromatography (hexane/EtOAc 7:3), yielded N-alkylated compound

67 (843 mg, 38%) and somewhat impure 71 (981.2 mg, 59%). The lat-

ter was used without purification in the hydroboration/oxidation se-

quence described below.

MS (EI, 70 eV): m/z = 524 (M , 1%), 468 (M-tert-Bu , 40).

HRMS: 524.1467, (C₂₄H₃₅NO₅⁷⁹BrSi) requires, 524.1468.

10-tert-Butyldimethylsilyloxy-6-methoxy-(1S,10S,13S,14R)-

8,16,18-dioxazapentacyclo[11.7.0.0^1,9.0^2,7.0^14,18]icosa-2(7),3,5-

trien-17-one (56a):

A large, heavy walled, resealable glass tube was fitted with a septum,

subjected to vacuum flame drying, Ar flooding three times, and al-

lowed to cool under Ar. Bromide 55 (215 mg, 0.410 mmol) was taken

up in anhyd benzene (3 × 14 mL) and added to the tube. The resulting

solution (~0.01 M) was rigorously degassed under a stream of Ar for

30 min. After that time, tris(trimethylsilyl)silane (190 µL,

0.615 mmol) was added, followed by a catalytic quantity of AIBN,

while taking precautions to exclude oxygen. The tube was sealed, and

the solution heated to 140°C in an oil bath. Further AIBN was added

over regular intervals (~5 h), and after 20 h, additional TTMSS (63

µL, 0.205 mmol). After 24 h total, heat was removed, the tube al-

lowed to cool, and opened. Solvent was removed in vacuo, and the

residue subjected to purification by a combination of flash chroma-

tography (silica ratio 100:1, hexane/EtOAc 3:2) and preparative

HPLC (Supelco C₁₈reverse phase column, MeOH/H₂O 90:10, 15

mL/min). Further purification of the enriched fractions yielded penta-

cycle 56a (15.4 mg, 8%) as a white solid; R_f= 0.50 (hexane/EtOAc

¹H NMR (400 MHz, CDCl₃): δ = 6.82 (dd, J = 17.7, 11.3 Hz, 1Η),

5.60 (d, J = 16.6 Ηz, 1Η), 5.29 (d, J = 11.3 Ηz, 1Η), 4.81 (d, J =

5.3 Hz, 1H), 4.39 (m, 1H), 2.54 (dt, J = 18, 5.3, Hz 1H), 2.02 (m, 1H),

1.86 (m, 1H), 1.41 (s, 3H), 1.39 (s, 3H).

25

In a similar procedure, starting with 73 the desired product 67 was ob-

tained in 60 % yield.

1:1); [α]_D–11.5 (c = 0.2, CHCl₃).

¹H NMR (400 MHz, CDCl₃): δ = 6.82 (t, J = 7.8 Hz, 1H), 6.74 (dd, J

= 8.1, 1.1 Hz, 1H), 6.60 (dd, J = 7.4, 1.1 Hz, 1H), 4.57 (d, J = 4.0 Hz,

1H), 4.53 (t, J = 8.3 Hz, 1H), 4.21 (td, J = 6.2, 6.2, 4.1 Hz, 1H), 3.99

(dd, J = 8.6, 5.7 Hz, 1H), 3.87 (s, 3H), 3.83 (ddd, J = 13.7, 5.0, 2.9 Hz,

1H), 3.74 (ddd, J = 11.0, 7.8, 5.8 Hz, 1H), 2.99 (ddd, J = 13.7, 12.1,

3.3 Hz, 1H), 2.01 (ddd, J = 10.9, 5.4, 2.7 Hz, 1H), 1.81 (m, 1H), 1.70

(ddd, J = 14.3, 12.3, 5.1 Hz, 1H), 1.65 (dt, J = 8.1, 6.0, 6.0 Hz, 1H),

1.23 (ddt, J = 14.3, 2 × 5.7, 2.7 Hz, 1H), 0.76 (s, 9H), 0.06 (s, 6H).

¹³C NMR (100 MHz, CDCl₃): δ = 156.5 (C), 144.1 (C), 135.1 (C),

121.3 (CH), 113.6 (CH), 112.3 (CH), 110.4 (C), 87.1 (CH), 68.3

(CH₂), 68.2 (CH), 55.9 (CH₃), 55.5 (CH), 47.8 (CH₂), 40.3 (CH), 37.6

(CH₂), 36.4 (CH₂), 25.7 (3 × CH₃), 24.2 (CH₂), 18.4 (C), –4.8 (CH₃),

–5.0 (CH₃).

67: R_f= 0.43 (hexane/EtOAc, 1:1); [α]_D²⁷+42.4 (c = 0.83, CHCl₃).

Found: C, 48.78; H, 5.27. (C₁₄H₁₈O₄NBr) requires: C, 48.84; H,

5.28%.

IR (neat): ν = 3490, 3200, 1520, 1350 cm^–1.

¹H NMR (400 MHz, CDCl₃): δ = 6.77 (d, J = 2.1 Hz, 1H), 6.57 (d, J

= 2.1 Hz, 1H), 4.54 (d, J = 5.5 Hz, 1H), 4.36 ( ddd, J = 5.3, 5.3, 3.0

Hz, 1H), 3.87 (m, 1H), 3.63 (m, 1H), 2.69 (m, 3H), 2.38 (ddd, J =

17.2, 4.6, 4.6 Hz, 3H), 1.97 (m, 1H), 1.86 (m, 1H).

¹³C NMR (100 MHz, CDCl₃): δ = 155.6 (C), 130.4 (C), 127.4 (CH),

136.6 (CH), 115.8 (CH), 109.0 (C), 75.2 (CH), 72.5 (CH), 41.4 (CH₂),

33.4 (CH₂), 31.9 (CH₂), 27.7 (CH₃), 26.4 (CH₂), 26.3 (CH₃).

+

MS (EI): m/z = 344 (M+H , 56%), 286 (100), 264 (48), 206 (62).

HRMS: 344.0520, (C₁₄H₁₉O₄⁷⁹BrN) requires, 344.0498.

Also isolated were various amounts of tetracyclic enol ether 58:

¹H NMR (400 MHz, CDCl₃): δ = 6.92 (m, 2H), 6.82 (m, 2H), 4.46 (t,

J = 8.5 Hz, 1H), 4.25 (br s, 1H), 4.08 (dd, J = 8.9, 5.5 Hz, 1H), 3.89

(s, 3H), 3.80 (ddd, J = 13.0, 6.1, 1.4 Hz, 1H), 3.40 (ddd, J = 10.1, 7.9,

5.5 Hz, 1H), 2.75 (dt, J = 12.8, 3.7 Hz, 1H), 2.64 (br d, J = 14.2 Hz,

1H), 2.15 (m, 1H), 1.82 (m, 2H), 1.67 (m, 3H), 0.96 (s, 9H), –0.05 (s,

3H), –0.18 (s, 3H).

Procedure for regeneration of 67:

2-(6-Bromo-2,2-dimethyl-(3aS,7aR)-4,5-dihydrobenzo[d][1,3]di-

oxol-7-yl)ethanol (72):

The crude (3aR,7aS)-5-bromo-2-2-dimethyl-4-vinyl-3a,6,7,7a-tet-

rahydro-1,3-benzodioxole 71 (2.9 g, ~11.19 mmol) was dissolved in

anhyd THF (20 mL) and a solution of 9-BBN (54 mL, 27 mmol, 0.5 M

sol. in THF, Aldrich) was added via syringe. After stirring at r.t. for

20 h, the mixture was cooled to 0°C, and H₂O (2 mL), an aqueous so-

lution of NaOH (6 mL, 3M), EtOH (10 mL) and H₂O₂(10 mL, 30%)

were added. The oxidation was complete after 30 min, and the mixture

was concentrated under reduced pressure. The residue was diluted with

H₂O/brine mixture (1:1, 60 mL) and extracted with EtOAc (5 x 50 mL).

The combined organic phases were dried (MgSO₄), filtered through a

plug of Celite, and evaporated to dryness. Flash column chromatogra-

phy (silica gel, 87 g, EtOAc/hexanes 1:1) afforded 72 (1.392 g, 45%);

R_f= 0.50 (hexane/EtOAc 7:3); [α]_D²⁶+87.4 (c = 1.00, CHCl₃).

Found: C, 47.50; H, 6.30. (C₁₁H₁₇O₃Br) requires: C, 47.67; H, 6.18%.

IR (neat): ν = 3400, 2950, 2930 cm^–1.

¹³C NMR (100 MHz, CDCl₃): δ = 156.7, 148.7, 147.7, 147.1, 123.1,

121.9, 120.5, 114.4, 112.0, 66.8, 65.4, 60.1, 55.8, 41.4, 40.1, 30.7,

25.7, 24.6, 19.7, 17.9, –5.2, –5.3.

6-Bromo-7-(2-bromoethyl)-2,2-dimethyl-(3aS,7aR)-4,5-dihy-

drobenzo[d][1,3]dioxole (66):

The PAD reduced diol 65 (776 mg, 2.59 mmol) was dissolved in ac-

etone (10 mL). 2,2-Dimethoxypropane (6 mL) and a catalytic amount

of p-TsOH were added. After stirring the mixture at r.t. for 3 h the

contents were concentrated to afford 843 mg of a brown oil. Purifica-

tion by flash chromatography (10% deactivated silica gel, hexane/

EtOAc 9:1), yielded 66 (790 mg, 90%) as a pale yellow oil; R_f= 0.21

(hexane/EtOAc, 20:1); [α]_D²⁵+93.7 (c = 1.2, CHCl₃).

¹H NMR (400 MHz, CDCl₃): δ = 4.52 (d, J = 5.6 Hz, 1H), 4.38 (m,

1H), 3.77 (m, 2H), 2.56 (m, 2H), 2.41 (m, 3H), 2.0 (m, 1H), 1.90 (m,

1H), 1.38 (s, 3H), 1.28 (s, 3 H).

Found: C, 39.18; H, 4.42. (C₁₁H₁₆O₂Br₂) requires: C, 38.85; H, 4.74%.

IR (neat): ν = 3100, 2900, 1100 cm^-1.

¹H NMR (400 MHz, CDCl₃): δ = 4.52 (d, J = 4.4 Hz, 1H), 4.38 (1m,

1H), 3.54 (m, 1H), 2.83 (m, 3H), 2.42 (ddd, J = 17.5, 4.3, 3.4 Hz, 1H),

2.03 (m, 1H), 1.87 (m, 1H), 1.37 (s, 3H), 1.35 (s, 3H).

¹³C NMR (100 MHz, CDCl₃): δ = 132.0 (C), 126.2 (C), 109.1 (C),

75.7 (CH), 72.4 (CH), 37.2 (CH₂), 31.7 (CH₂), 27.8 (CH₃), 26.5

(CH₂), 26.2 (CH₂).

¹³C NMR (100 MHz, CDCl₃): δ = 132.0 (C), 126.2 (C), 109.0 (C),

76.1 (CH), 72.6 (CH), 60.9 (CH₂), 38.1 (CH₂), 32.0 (CH₂), 27.6

(CH₃), 26.8 (CH₂), 26.2 (CH₃).

+

MS (EI): m/z = 277 (M , 80%), 261 (38), 200 (82), 139 (100).

HRMS: 277.0404, (C₁₁H₁₈⁷⁹BrO₃) requires, 277.0439.

SYNTHESIS

April 1998

677

5-Bromo-4-(2-methanesulfonyloxyethyl)-2,2-dimethyl-

(3aS,7aR)-6,7-dihydrobenzo[d][1,3]dioxole (73):

of silica gel and evaporated to dryness. Purification by flash chroma-

tography (silica gel, EtOAc/EtOH 8:2) afforded pure diol 74 (923.2

30

A solution of (3aR,7aS)-4-(2-hydroxyethyl)-3a,6,7,7a-tetrahydro-

1,3-benzodioxole (1.297 g, 4.68 mmol) and diisopropylethylamine

(1.63 mL, 9.36 mmol) in anhyd CH₂Cl₂(20 mL) was cooled to 0°C,

and mesyl chloride (0.707 mL, 5.614 mmol) was added via syringe.

The mixture was allowed to warm up to r.t. over a period of 30 min

and was quenched with brine (30 mL). Extraction with CH₂Cl₂

(3 × 50 mL), drying (MgSO₄), filtration and evaporation of the sol-

vent afforded 1.78 g of crude product. Further purification by flash

chromatography (silica gel, 80 g, EtOAc/hexanes 1:1) gave the de-

sired mesylate 73 (1.595 g, 95%), which was immediately used in the

next step.

mg, 94%); mp 150–153°C; R_f= 0.43 (EtOAc/MeOH 4:1); [α]_D

–22.8 (c = 0.67, MeOH).

Found: C, 58.41; H, 6.75; N, 6.16. (C₁₁H₁₆NO₄) requires: C, 58.66;

H, 6.71; N, 6.22%.

IR (KBr): ν = 2870, 1740 cm^–1.

¹H NMR (400 MHz, CDCl₃): δ = 4.47 (t, J = 8.7 Hz, 1H), 4.25 (t, J =

6.7 Hz, 1H), 3.97 (m, 2H), 3.89 (s, 1H), 3.75 (s, 1H), 3.06 (ddd, J =

13, 11, 4.7 Hz, 1H), 2.61 (d, J = 4.7 Hz, 1H), 2.40 (m, 1H), 2.25 (m,

1H), 1.97 (s, 2H), 1.80 (m, 2H), 1.70 (s, 1H).

¹³C NMR (100 MHz, CDCl₃): δ = 130.4 (C), 130.1 (C), 124.9 (C),

69.0 (CH), 68.3 (CH), 66.6 (CH₂), 55.3(CH), 38.1 (CH₂), 25.3 (CH₂),

24.9 (CH₂), 24.3 (CH₂).

MS (CI): m/z = 226 (M⁺, 100%), 207 (18), 190 (29).

HRMS: 266.1064, (C₁₁H₁₆NO₄) requires, 226.1072.

2,2-Dimethyl-(3aR,9aR,11aS)-4,5,10,11-tetrahydro[1,3]dioxo-

lo[4,5-f]oxazolo[4,3-a]isoquinolin-7-one (68a): and 2,2-Dimethyl-

(3aR,9aS,11aS)-4,5,10,11-tetrahydro[1,3]dioxolo[4,5-f]oxazo-

lo[4,3-a]isoquinolin-7-one (68b):

7-Hydroxy-8-tert-butyldimethylsilyloxy-(7R,8S,10bS)-

A flame dried round bottom flask was set under static Ar and charged

with the solution of 67 (2.75 g, 8.0 mmol) in freshly distilled benzene

(500 mL). After degassing, (stream of Ar, 30 min.) Bu₃SnH (2.15 mL,

16.0 mmol) was added via glass pipette followed by AIBN (132 mg,

0.80 mmol). The mixture was heated to reflux for 1 h. The solvent was

evaporated and the crude product was disproportionated between

MeCN and hexane. The MeCN phase was evaporated to dryness, and

the residue was purified by flash chromatography (silica gel, 6:4 hex-

ane/EtOAc) to give 68a (1.193 g, 56%) as the major isomer and 68b

(0.633 g,31 %) as the minor isomer.

5,6,7,8,9,10-hexahydrooxazolo[4,3-a]isoquinolin-3-one (75):

The solution of the diol 74 (882.7 mg, 3.912 mmol) and diisopropyl-

ethylamine (1.363 mL, 7.824 mmol) in CH₂Cl₂(200 mL) was stirred

and cooled to –78°C. A solution of tert-butyldimethylsilyltrifluo-

romethanesulfonate (1.078 mL, 4.694 mmol) in CH₂Cl₂(50 mL) was

added dropwise during a period of 30 min and the stirring at –78°C

was continued for 4 h. The reaction was quenched at –78°C with H₂O

(60 mL) and allowed to warm up to r.t. The CH₂Cl₂layer was sepa-

rated, and the aqueous phase was extracted with CH₂Cl₂(2 × 50 mL).

The combined organic extracts were dried (MgSO₄), filtered through

Celite and evaporated to dryness. The crude product was dissolved in

a mixture of EtOAc/hexane (7:3, 50 mL) and passed through a plug

of silica gel. The filtrate was evaporated to dryness and purified by

flash chromatography (silica gel, EtOAc/hexane 7:3) to yield the de-

sired product 75 (1.135 g, 85.5%) and bis-protected 7,8-di(tert-bu-

tyldimethylsilyloxy)-(7R,8S,10bS)-5,6,7,8,9,10-hexahydro[1,3]oxa-

zolo[4,3-a]isoquinolin-3-one (173 mg, 10%). The latter was cleaved

using tetrabutylammonium fluoride to regenerate the starting diol 74.

(68a) (Major Isomer):

R_f= 0.16 (hexane/EtOAc 7:3); [α]_D²⁵–100.8 (c = 0.66, CHCl₃).

Found: C, 62.98; H, 7.25; N, 5.08. (C₁₄H₁₉NO₄) requires: C, 63.34;

H, 7.22; N, 5.28%.

IR (neat): ν = 2980, 1760 cm^–1.

¹H NMR (400 MHz, CDCl₃): δ = 4.49 (t, J = 8.8 Hz, 1H), 4.24 (m,

3H), 3.97 (m, 2H), 3.01 (ddd, J = 13.6, 11.6, 4.8 Hz, 1H), 2.32 (d, J =

12.4 Hz, 1H), 2.21 (m, 1H), 2.01 (m, 1H), 1.83 (m, 3H), 1.37 (s, 6H).

¹³C NMR (100 MHz, CDCl₃): δ = 157.4 (C), 130.2 (C), 128.4 (C),

108.5 (C), 73.6 (CH), 72.7 (CH), 66.7 (CH₂), 54.9 (CH), 38.3 (CH₂),

28.0 (CH₃), 26.2 (CH₃), 25.4 (CH₂), 24.1 (CH₂), 21.0 (CH₂).

MS (EI): m/z = 265 (M⁺, 8%), 190 (100), 151 (39), 105 (48).

HRMS: 266.1392, (C₁₄H₂₀NO₄) requires, 266.1393

(75):

mp 200–201°C; R_f= 0.40 (EtOAc/hexane 7:3); [α]_D²⁴–180.9 (c = 1.1,

CHCl₃).

Found: C, 59.97; H, 8.63; N, 3.91. (C₁₇H₂₉NO₄Si) requires: C, 60.14;

H, 8.61; N, 4.13%.

(68b) (Minor Isomer):

IR (KBr): ν = 3510, 2920, 2895, 2840, 1745, 1405, 1370, 1305, 1245,

mp 102–105°C; R_f= 0.09 (hexane/EtOAc, 7:3); [α]_D²⁶+201.2 (c =

0.89, CHCl₃).

1173, 1100, 1050, 973, 920, 820 cm^–1.

¹H NMR (400 MHz, CDCl₃): δ = 4.44 (t, J = 8.4 Hz, 1H), 4.21 (br t,

J = 6.9 Hz, 1H), 3.97 (d , J = 6.7 Hz, 1H), 3.94 (dd, J = 6.7 , 4.4 Hz,

1H), 3.73 (m, 2H), 3.06 (ddd, J = 13.6, 11.6, 4.88 Hz, 1H), 2.73 (br s,

1H), 2.37 (br dd, J = 17.1, 2.1 Hz, 1H), 2.24 (m, 1H), 1.87 (m, 3H),

1.60 (m, 1H), 0.9 (s, 9H), 0.1 (s, 6 H).

Found: C, 63.31; H, 7.19; N, 5.27. (C₁₄H₁₉NO₄) requires C, 63.34; H,

7.22; N, 5.28%.

IR (neat): ν = 2980, 1720 cm^–1.

¹H NMR (400 MHz, CDCl₃): δ = 4.47 (t, J = 8.5 Hz, 1H ), 4.35 (m,

1H), 4.26 (m, 1H), 3.98 (q, J = 6.6 Hz, 1H), 3.91 (t, J = 7.9 Hz, 1H),

3.01 (ddd, J = 13.3, 11.9, 4.7 Hz, 1H), 2.55 (m, 1H), 2.12 (m, 2H),

1.87 (d, J = 16.0 Hz, 1H), 1.71 (m, 1H), 1.36 (s, 3H), 1.31 (s, 3H).

¹³C NMR (100 MHz, CDCl₃): δ = 129 (C), 128.3 (C), 108.9 (C), 75.7

(CH), 72.4 (CH), 66.8 (CH₂), 55.0 (CH), 37.8 (CH₂), 27.9 (CH₃), 26.5

(CH₃), 25.5 (CH₂), 24.3 (CH₂), 19.7 (CH₂).

¹³C NMR (100 MHz, CDCl₃): δ = 157.3 (C), 131.1 (C), 130.0 (C),

70.3 (CH), 68.5 (CH), 55.4 (CH), 38.0 (CH₂), 25.7 (CH₃), 24.9 (CH₂),

24.4 (CH₂), 18.0 (C), –4.5 (CH₃), –4.9 (CH₃).

+

MS (CI): m/z = 340 (M+H , 100%), 322 (10), 282 (12), 208 (1), 191

(6), 190 (47).

HRMS: 340.1958, (C₁₇H₃₀NO₄Si) requires, 340.1944.

MS (EI): m/s = 265 (M⁺, 2%), 250 (28), 190 (51), 105 (100).

HRMS: 266.1394, (C₁₄H₂₀NO₄) requires, 266.1393.

7-(2-Bromo-6-methoxyphenyloxy)-8-tert-butyldimethylsilyloxy-

(7S,8S,10bR)-5,6,7,8,9,10-hexahydrooxazolo[4,3-a]isoquinolin-3-

one (76):

5-Bromo-4-(2-methanesulfonyloxyethyl)-2,2-dimethyl-

(3aS,7aR)-6,7-dihydrobenzo[d][1,3]dioxole (73):

A solution of 75 (551 mg, 1.623 mmol) and 6-bromo-2-methoxyphe-

nol (346 mg, 1.704 mmol) in THF (10 mL) was cooled to 0°C with

stirring. To this solution, a reagent, prepared from Bu₃P (657 mg,

3.246 mmol) and DEAD (565 mg, 3.246 mmol) in THF (10 mL), was

added dropwise and the mixture was stirred at 0°C for 90 min. The

reaction was quenched with MeOH (3 mL), and the solvent was evap-

orated at reduced pressure. The crude product (1.003 g) was further

purified by flash chromatography (silica gel, 86 g, EtOAc: hexane

1:1) to yield pure 76 (803 mg, 94%) as a viscous, colorless oil; R_f=

0.23 (CH₂Cl₂/acetone 97:3); T_r= 8.80 min., (Prodigy 5 µ, 80%

MeCN/20% H₂O, 5 mM Et₃N.HOAc, 1.0 mL/min., UV λ_max236 nm);

[α]_D²⁷–11.5 (c = 1.1, CHCl₃).

¹H NMR (400 MHz, CDCl₃): δ = 4.48 (br d, J = 5.3 Hz, 1H), 4.41 (m,

1H), 4.35 (m, 2H), 3.02 (s, 3H), 2.75 (m, 3H), 2.41 (dt, J = 17.0, 5.0

Hz, 1H), 2.02 (m 1H), 1.88 (m, 1H), 1.36 (s, 3H), 1.34 (s, 3H).

¹³C NMR (100 MHz, CDCl₃): δ = 129.7, 126.8, 109.1, 75.8, 72.4,

67.3, 37.5, 33.6, 31.7, 27.8, 26.4, 26.2.

7,8-Dihydroxy-(7R,8S,10bR)-5,6,7,8,9,10-hexahydrooxazolo[4,3-

a]isoquinolin-3-one (74):

A suspension of the acetonide 68a (1.16 g, 4.37 mmol) and Dowex

50X8-100 strongly acidic resin (625 mg) in aq MeOH (50 mL, 90%)

was stirred at r.t. overnight. The mixture was filtered through a plug

SYNTHESIS

Papers

678

+

Found: C, 54.59; H, 6.41; N, 2.26. (C₂₄H₃₄BrNO₅Si) requires: C,

54.96; H, 6.53; N, 2.67%.

MS (FAB) m/z = 446 (M+H , 100%), 388 (25), 312 (3), 270 (2), 209 (3).

HRMS: 446.2340, (C₂₄H₃₆NO₅Si) requires, 446.2363.

IR (KBr): ν = 3440, 2930, 2850, 1760, 1570, 1470, 1450, 1440, 1260,

1120, 1080, 1030, 840, 750 cm^–1.

10-Hydroxy-6-methoxy-(1R,9S,10S,13S,14R)-8,16,18-dioxaza-

¹H NMR (500 MHz, CDCl₃): δ = 7.12 (dd, J = 1.5, 7.9 Hz, 1H), 6.92

(t, J = 2x 8.1 Hz, 1H), 6.85 (dd, J = 8.2, 1.5 Hz, 1H), 4.53 (d, J = 8.9,

8.1 Hz, 1H), 4.39 (br s 1H), 4.31 (br t, J = 8.9 Hz, 1H), 4.07 (m, 2H),

3.93 (dd, J = 13.1, 6.1 Hz, 1H), 3.84 (s, 3H), 3.01 (ddd, J = 13.1, 11.9,

4.4 Hz, 1H), 2.53 (m, 1H), 2.33 (m, 1H), 2.16 (m, 1H), 1.99 (br d, J =

16.9 Hz, 1H), 1.7 (m, 2H), 0.78 (s, 9H), –0.08 (s, 3H), –0.12 (s, 3H).

¹³C NMR (125 MHz, CDCl₃): δ = 157.1 (C). 153.4(C), 144.2 (C),

132.4 (C), 125.9 (C), 125.1 (CH), 124.9 (CH), 118.1 (CH), 111.5

(CH), 81.6 (CH), 67.5 (CH₂), 66.9 (CH), 55.7 (CH₃), 54.5 (CH), 38.1

(CH₂), 27.9 (CH₂), 25.6 (CH₃), 25.1 (CH₂), 20.5 (CH₂), 18.0 (C), –5.1

(CH₃), –5.2 (CH₃).

1,9 2,7 14,18

pentacyclo[11.7.0.0 .0 .0

]icosa-2(7),3,5-trien-17-one (79):

The solution of TBS-pentacycle 78 (116 mg, 0.260 mmol) and tet-

rabutylammonium fluoride trihydrate (100 mg) in THF (5 mL) was

stirred at r.t. for 4 h. The mixture was filtered through Celite, washed

with MeOH and evaporated to dryness. The crude product was puri-

fied by flash chromatography (silica gel, 86 g, CH₂Cl₂/acetone 7:3) to

yield the desired alcohol 79 (86 mg, quant.).

Cyclization from 77: To a degassed (Ar sparge) solution of 77

(44.4 mg, 0.108 mmol) in benzene (25 mL) was added Bu₃SnH (63 mg,

58 µl, 0.216 mmol) followed by AIBN (cat.). The reaction vessel was

submerged into a preheated oil bath and refluxed was maintained until

HPLC indicated the disappearance of starting material, approx. 80 min.

The solvent was evaporated under reduced pressure and the oily residue

was disproportionated between hexane (60 mL) and MeCN (30 mL).

The MeCN layer was extracted with hexane two additional times

(2 × 30 mL), and the MeCN was evaporated under reduced pressure.

The crude residue was purified by column chromatography (EtOAc/

EtOH/NH₄OH 90:5:5) to afford alcohol 79 (10.3 mg, 29%), spectral

characteristics of which were identical to those obtained by the TBAF

mediated deprotection; mp 118–120°C; R_f= 0.45 (benzene/dioxane/aq

NH₄OH/H₂O 50:40:5:5); T_r= 6.88 min, (Prodigy 5µ ODS2, 30%

MeCN/70% H₂O, 5 mM triethylammonium acetate, UV λ_max236 nm,

1.0 mL/min.); [α]_D²⁵+31.3 (c = 0.38, CHCl₃).

MS (FAB): m/z = 524 (M⁺, 7%), 322 (65), 190 (100).

HRMS: 524.1450, (C₂₄H₃₅NO₅⁷⁹BrSi) requires, 524.1468.

7-(2-Bromo-6-methoxyphenyloxy)-8-hydroxy-(7S,8S,10bR)-

5,6,7,8,9,10-hexahydrooxazolo[4,3-a]isoquinolin-3-one (77):

To a solution of silyl ether 76 (39 mg, 0.076 mmol) in THF (2 mL)

was added TBAF/silica (30 mg) and TBAF/H₂O (30 mg). The mix-

ture was stirred at r.t. for 3.5 h. The crude mixture was concentrated

under reduced pressure and purified by column chromatography (sil-

ica gel, EtOAc) to afford alcohol 77 (31.2 mg, quant.) as an oil; R_f=

0.30 (CH₂Cl₂/Et₂O 8:2); T_r= 6.86 min., (Prodigy 5 µ, 35% MeCN/

65% H₂O, 5 mM Et₃N.HOAc, 1.0 mL/min., UV λ_max254 nm).

¹H NMR (400 MHz, CDCl₃): δ = 7.13 (dd, J = 8.1, 1.7 Hz, 1H), 6.91

(t, J = 8.2 Hz, 1H), 6.85 (dd, J = 8.2, 1.5, 1H), 4.66 (br s, 1H), 4.50 (t,

J = 8.7 Hz, 1H), 4.30 (br t, J = 6.9 Hz, 1H), 4.18 (m, 1H), 4.04 (t, J =

8.1 Hz, 1H), 3.95 (dd, J = 13.1, 6.6 Hz, 1H), 3.84 (s, 3H), 3.02 (ddd,

J = 13.3 11.9, 4.4 Hz, 1H), 2.60 (m, 1H), 2.20 (m, 1H), 2.00 (m, 3H),

1.80 (m, 1H).

Found: C, 50.49; H, 4.94; N, 3.07. (C₁₈H₂₁NO₅+CHCl₃) 450.744

(Solvate with CHCl₃, four analyses) requires: C, 50.63; H, 4.90; N,

3.10%.

–1

IR (CHCl₃): 2936, 1749, 1614, 1492, 1455, 1283, 1215, 908 cm

.

¹H NMR (500 MHz, CDCl₃): δ = 6.92 (t, J = 7.8 Hz, 1H), 6.80 (dd, J

= 8.7, 1.0 Hz, 1H), 6.65 (dd, J = 7.8, 1.0 Hz, 1H), 4.53 (d, J = 7.3 Hz,

1H), 4.52 (t, J = 8,8 Hz, 1H), 4.05 (dd, J = 8.8, 5.9 Hz, 1H), 3.89 (ddd,

J = 11.2, 8.1, 5.9 Hz, 1H), 3.88 (s, 3H), 3.80 (ddd, J = 13.9, 5.1,

1.7 Hz, 1H), 3.63 (ddd, J = 11.7, 7.3, 5.9 Hz, 1H), 3.07 (ddd, J = 13.4,

13.4, 3.8 Hz, 1H), 2.19 (ddd, J = 11.2, 4.6, 2.2 Hz, 1H), 1.81 (ddd, J

= 14.2, 2.4, 2.4 Hz, 1H), 1.75 (m, 1H), 1.65 (ddd, J = 13.7, 3.4, 3.4 Hz,

1H), 1.53 (ddd, J = 13.2, 11.7, 2.9 Hz, 1H), 1.46 (ddd, J = 14.7, 6.4,

3.4 Hz, 1H), 1.37 (ddd, J = 13.3, 13.2, 4.9 Hz, 1H).

¹³C NMR (100 MHz, CDCl₃): δ = 157.0, 153.0, 144.8, 131.1, 127.7,

125.2, 124.8, 117.5, 111.7, 83.4, 69.4, 55.7, 54.8, 37.9, 27.3, 26.2,

22.1.

10-tert-Butyldimethylsilyloxy-6-methoxy-(1R,9S,10S,13S,14R)-

8,16,18-dioxazapentacyclo[11.7.0.0 .0 .0

trien-17-one (78):

1,9 2,7 14,18

]icosa-2(7),3,5-

¹³C NMR (125 MHz, CDCl₃): δ = 156.7 (C), 145.9 (C), 145.1(C),

135.5 (C), 122.4 (CH), 113.2 (CH), 112.0 (CH), 89.6 (CH), 71.7

(CH), 67.6 (CH₂), 55.9 (CH₃), 53.4 (CH), 49.0 (CH₂), 38.6 (CH), 37.6

(C), 34.6 (CH₂), 25.1 (CH₂₎, 21.1 (CH₂).

A solution of 76 (747 mg, 1.425 mmol) in benzene (500 mL) was de-

gassed at r.t. with a stream of Ar (30 min.). Bu₃SnH (1.659 g,

5.7 mmol, glass pipette), followed by AIBN (90 mg, 0.548 mmol)

was added and the reaction flask was submerged into a preheated

(100°C) oil bath. HPLC (Prodigy 5µ ODS2, 80% MeCN/20% H₂O,

5 mM Et₃N.AcOH buffer, UV λ_max210 nm, 1.0 mL/min.) indicated

complete reaction after 50 min. The solvent was removed in vacuo

and the remaining oil was distributed between MeCN (100 mL) and

hexane (200 mL). The MeCN phase was extracted with hexane

(2 × 100 mL), and the combined hexanes back-washed with MeCN

(100 mL). The combined MeCN extracts were evaporated in vacuo to

yield 833 mg of crude product. Column chromatography (silica gel,

86 g, CH₂Cl₂/acetone 97:3) gave the pure pentacycle 78 (297.1 mg,

47%); R_f= 0.53 (EtOAc/hexanes 1:1); T_r= 14.4 min., (Prodigy 5µ

ODS2, 80% MeCN/20%H₂O, 5 mM triethylammonium acetate,

MS (FAB): m/z = 332 (M⁺, 100%), 242 (9), 155 (36), 118 (57).

HRMS: 332.1500, (C₁₈H₂₁NO₅+H) requires, 332.1498.

5-Hydroxymethyl-13-methoxy-4-methyl-(1R,5R,6S,9S,10S)-11,4-

1,6 12,17

oxazatetracyclo[8.7.0 .0

(80):

]heptadeca-12(17)13,15-trien-9-ol

A flame dried round-bottom flask was set under static Ar, and was

charged with the solution of the pentacyclic alcohol 79 (86 mg, 0.259

mmol) in CH₂Cl₂(20 mL). The stirred solution was cooled to 0°C and

solution of DIBAL-H in THF (2.6 mL, 1M, Aldrich) was added, drop-

wise. After 30 min., HPLC (Primesphere 5µ, C18HC, 30% MeCN/

70% H₂O, 5 mM ammonium carbonate, 1.0 mL/min., UV λ_max

284 nm) indicated complete conversion and the reaction was

quenched at 0°C with H₂O (1.8 mL), followed by MeOH (1.8 mL)

and sat. aq NaHCO₃(3.6 mL). The mixture was allowed to warm to

r.t. and the solid was filtered off and washed with CH₂Cl₂(4 ×

40 mL). The combined filtrates were dried (Na₂SO₄) and the solvent

was removed under reduced pressure. The crude product was purified

by flash chromatography (silica gel, 85 g, EtOAc/EtOH/aq NH₄OH

75:20:5) to yield the pure product 80 (72 mg, 87 %) as a viscous, color-

less oil; R_f= 0.4 (EtOAc/ EtOH/aq NH₄OH 75:20:5); T_r= 2.99 min.

(Primesphere 5µ, C18HC, 30% MeCN/70% H₂O, 5mM ammonium car-

28

UV λ_max210 nm, 1.0 mL/min.); [α]_D+ 23.3 (c = 0.89, CHCl₃).

Found: C, 64.26; H, 7.99; N, 3.20. (C₂₄H₃₆NO₅Si) requires: C, 64.69;

H, 7.92; N, 3.14%.

IR (CHCl₃): ν = 3010, 2930, 1750, 1620, 1450, 1270, 1120, 910,

840 cm^–1.

¹H NMR (500 MHz, CDCl₃): δ = 6.87 (dd, J = 8.1, 7.3 Hz, 1H), 6.79

(dd, J = 8.1, 1.1 Hz, 1H), 6.63 (dd, J = 7.5, 1.1 Hz, 1H), 4.49 (d, J =

5.6 Hz, 1H), 4.46 (t, J = 8.1 Hz, 1H), 4.08 (ddd, J = 11.1, 7.9, 5.5 Hz,

1H), 4.02 (dd, J = 8.4, 5.7 Hz, 1H), 3.86 (s, 3H), 3.79 (ddd, J = 13.9,

5.3, 2.0 Hz, 1H), 3.68 (ddd, J = 8.9, 5.5, 5.5 Hz, 1H), 3.09 (ddd, J =

13.3, 13.3, 2.9 Hz, 1H), 2.13 (ddd, J = 11.0, 5.3, 2.1 Hz, 1H), 1.78 (m,

1H), 1.67 (m, 1H), 1.55 (m, 2H), 1.47 (m, 1H), 1.35 (m, 1H), 0.9 (s,

9H), 0.14 (s, 3H), 0.02 (s, 3H).

26

bonate, 1.0 mL/min., UV λ_max284 nm); [α]_D+34.7 (c = 1.51, CHCl₃).

IR (CHCl₃): ν = 2960, 1380, 1130, 1220, 770 cm^–1.

1H NMR (500 MHz, CDCl₃): δ = 6.87 (t, J = 7.8 Hz, 1H), 6.75 (d, J

= 8.3 Hz, 1H), 6.73 (d, J = 7.8 Hz, 1H), 4.49 (d, J = 7.3 Hz, 1H), 3.89

¹³C NMR (125 MHz, CDCl₃): δ = 156.8, 145.8, 145.7, 135.6, 122.1,

113.5, 112.3, 89.4, 71.8, 67.6, 56.0, 54.0, 48.6, 38.5, 37.7, 35.4, 26.4,

25.8, 20.2, 18.0.

SYNTHESIS

April 1998

679

(dd, J = 11.7, 2.9 Hz, 1H), 3.85 (s, 3H), 3.64 (dd, J = 11.7, 1.0 Hz,

1H), 3.58 (ddd, J = 11.7, 6.8, 5.8 Hz, 1H), 2.78 (ddd, J = 12.2, 3.4, 3.4

Hz, 1H), 2.57 (bdd, J = 10.7, 3.4, ~1.5 Hz, 1H), 2.54 (ddd, J = 12.2,

12.2, 2.0 Hz, 1H), 2.35 (s, 3H), 2.21 (br d, J = 10.7 Hz, 1H), 1.73 (m,

3H), 1.65 (m, 1H), 1.46 (m, 3H).

¹H NMR (500 MHz, benzene-d₆): δ = 6.76 (t, J = 7.8 Hz, 1H), 6.59

(d, J = 3.2 Hz), 6.57 (d, J = 4.7 Hz, 1H), 4.47 (d, J = 7.1 Hz, 1H), 3.63

(dd, J = 11.7, 3.1 Hz, 1H), 3.47 (s, 3H), 3.46 (m, 1H), 3.43 (br d, J =

10.9 Hz, 1H), 2.58 (br d, J = 10.7 Hz, 1H), 2.26 (dt, J = 12.1, 3.6 Hz,

1H), 2.03 (td, J = 12.7, 2.2 Hz, 1H), 1.90 (s, 3H), 1.77 (br d, J =

10.9 Hz, 1H), 1.60 (dt, J = 13.8, 1.9 Hz, 1H), 1.35 (br m, 5H).

¹³C NMR (125 MHz, CDCl₃): δ = 145.7, 145.2, 137.3, 122.1, 113.5,

111.5, 90.0, 72.9, 63.4, 58.4, 55.9, 52.0, 49.2, 42.8, 35.6, 33.3, 25.8,

22.6.

¹H NMR (500 MHz, CDCl₃): δ = 6.87 (t, J = 7.81 Hz, 1H), 6.77 (d, J

= 7.6 Hz, 1H), 6.74 (d, J = 7.3 Hz, 1H), 4.46 (d, J = 6.6 Hz, 1H), 3.86

(s, 3H), 3.20 (ABq, J = 12.7 Hz, 2H), 3.60 (m, 1H), 3.05 (dd, J = 14.2,

7.1 Hz, 1H), 2.83 (ddd, J = 12.0, 3.9, 3.9 Hz, 1H), 2.53 (m, 3H), 2.41

(s, 3H), 2.30 (dddd, J = 11.2, 11.2, 3.7, 3.7 Hz, 1H), 1.95 (br d, J =

10.3 Hz, 1H), 1.6 (m, 3H), 0.90 (s, 3H), 0.14 (s, 3H), 0.01 (s, 3H).

13-Methoxy-4-methyl-9-oxo-(1R,5R,6S,10S)-11,4-oxazatetracy-

1,6 12,17

clo[8.7.0 .0

(83):

]heptadeca-12,(17),13,15-triene-5-carbaldehyde

A flame dried round-bottom flask was set under static Ar atmosphere,

charged with CH₂Cl₂(4 mL) and oxalyl chloride (86 µL, 2 M solution

in CH₂Cl₂, Aldrich, 0.172 mmol,). The mixture was cooled to –78°C

and a CH₂Cl₂solution of DMSO (344 µl, 1M, 0.344 mmol) was in-

troduced via syringe. After stirring for 10 min. a solution of the tetra-

cyclic diol 80 (5.5 mg, 0.0172 mmol) in CH₂Cl₂(1.0 mL) was added

and the temperature was allowed to warm up to 0°C during 4 h. A

CH₂Cl₂solution of Et₃N (412 µL, 0.412 mmol, 1M) was added, the

cooling bath was removed and stirring was continued for an addition-

al 30 min. The reaction was quenched with sat. NaHCO₃(2.0 mL,

1:10 solution), stirred for 5 min and the organic solvent was separat-

ed. The aqueous layer was further extracted with CH₂Cl₂(2 × 4 mL),

the combined organic phases were dried (Na₂SO₄) and the solvent

was removed under reduced pressure. Column chromatography (sili-

ca gel, 5.4 g, EtOAc/ethanol/aq NH₄OH 90:5:5) gave the pure product

(3.6 mg, 66%) as a viscous oil, which was used immediately in the

next step; R_f= 0.6 (EtOAc/EtOH/aq NH₄OH 90:5:5); T_r= 12.8 min.

(Primesphere 5 µ, C18HC, 30 % MeCN / 70 % H₂O, 5 mM ammoni-

um carbonate, UV λ_max283 nm, 1.0 mL/min.).

MS (FAB): m/z = 320 (M+H⁺, 100%), 288 (38), 219 (3), 213 (3), 169

(5), 154 (10), 137 (15), 109 (21), 94 (44).

HRMS (FAB): 320.1877, (C₁₈H₂₆NO₄+H) requires, 320.1862.

5-Hydroxymethyl-9-tert-butyldimethylsilyoxy-13-methoxy-4-

methyl-(1R,5R,6S,9S,10S)-11,4-oxazatetracyc-

1,6 12,17

lo[8.7.0.0 .0

]hep-tadeca-12(17),13,15-triene (81):

A solution of the pentacyclic substrate 78 (25.0 mg, 0.056 mmol) in

freshly distilled CH₂Cl₂(6 mL) was set under static Ar atmosphere,

and cooled with stirring to 0°C. A solution of DIBAL-H (281 µL, 1M

in CH₂Cl₂, Aldrich) was added via syringe. Stirring at 0°C was con-

tinued for 1 h, and the mixture was allowed to warm up to r.t. After

additional 2 h, the reaction was quenched with H₂O (0.8 mL), fol-

lowed by MeOH (0.8 mL) and sat. aq NaHCO₃(1.6 mL) and stirring

continued for 30 min. The solid was filtered off, washed with CH₂Cl₂

(6 × 6 mL) and H₂O (6 × 6 mL). The organic phase was separated,

dried (MgSO₄), and the solvent was evaporated under reduced pres-

sure to yield the crude product (27 mg). HPLC analysis (Primesphere

5m C18HC, 70% MeCN/30% H₂O, 5 mM triethylammonium acetate,

UV λ_max210 nm, 1.0 mL/min.) indicated 79% purity. Column chro-

matography (silica gel, 15 g, EtOAc/EtOH/NH₄OH 70:25:5) gave

pure 81 (19.3 mg, 79%); R_f= 0.53 (EtOAc/EtOH/NH₄OH 70:25:5);

T_r= 7.94 min. (Primesphere 5µ C18HC, 70% MeCN/30% H₂O,

5 mM triethylammonium acetate, UV λ_max210 nm, 1.0 mL/min.).

¹H NMR (500 MHz, CDCl₃): δ = 6.87 (dd, J = 8.1, 7.5 Hz, 1H), 6.77

(dd, J = 8.2, 1.1 Hz, 1H), 6.74 (dd, J = 7.3, 1.1 Hz, 1H), 4.45 (d, J =

6.6 Hz, 1H), 3.90 (dd, J = 11.9, 3.2 Hz, 1H), 3.86 (s, 3H), 3.64 (dd, J

= 11.6, 1.0 Hz, 1H), 3.57 (m, 1H), 2.77 (ddd, J = 12.1, 3.8, 3.8 Hz,

1H), 2.63 (ddd, J = 12.7, 12.7, 2.4 Hz, 1H), 2.57 (br d, J = 11.2 Hz,

1H), 2.38 (s, 3H), 2.33 (br d, J = 10.7 Hz, 1H), 1.76 (ddd, J = 14.2,

2.9, 2.9 Hz, 1H), 1.50 (m, 4H), 0.90 (s, 9H), 0.13 (s, 3H), 0.02 (s, 3H).

¹³C NMR (125 MHz, CDCl₃): δ = 145.7, 145.6, 137.6, 121.6, 113.5,

111.9, 90.7, 73.3, 63.5, 58.4, 55.9, 51.9, 49.1, 42.7, 35.8, 33.2, 27.6,

25.8, 22.6, 22.2, 18.1.

¹H NMR (500 MHz, CDCl₃): δ = 9.45 (s, 1H), 6.93 (t, J = 7.8 Hz, 1H),

6.79 (d, J = 7.3 Hz, 1H), 6.70 (t, J = 7.8 Hz, 1H), 4.60 (s, 1H), 3.88

(s, 3H), 3.87 (m, 1H), 2.90 (br d, J = 12.2 Hz, 1H), 2.50 (m, 5H), 2.27

(s, 3H), 2.20 (m, 2H), 1.88 (ddd, J = 13.9, 2.4, 2.4 Hz, 1H), 1.77

(dddd, J = 14.9, 14.9, 7.6, 3.4 Hz, 1H).

10-Hydroxy-14-epi-ent-dihydrocodeinone (84):

The crude oxo aldehyde 83 (12.0 mg, 0.038 mmol) was dissolved in

trifluoromethanesulfonic acid (neat, 400 µL) at r.t. A deep red colored

solution resulted. The reaction progress was monitored by HPLC

(Prodigy 5µ, ODS2, MeCN/H₂O 30:70, 5 mM ammonium acetate

buffer, 1.0 mL/min., UV λ_max210 nm), which, after 10 min indicated

complete disappearance of the starting material (T_{r, SM}= 11.10 mins,

T_r, product 3.97 min).

The mixture was diluted with CHCl₃(8 mL), cooled to –5°C, and

carefully quenched with 5 g of ice. The aqueous layer was basified

(1.0 M KOH), and extracted with CHCl₃(6 × 6 mL). The combined

organic extracts were dried (Na₂SO₄), and the solvent was removed

under reduced pressure to yield 9.3 mg of the crude base, as an vis-

cous oil. Column chromatography (silica gel, 7.0 g, CHCl₃/MeOH/

NH₄OH 90:9:1) gave 6.9 mg (58 %) of the pure morphinan; R_f= 0.30

(CHCl₃/MeOH/NH₄OH 90:9:1); λ_max(MeCN): 286, 248, 236 and

214 nm.

¹H NMR (500 MHz, benzene-d₆): δ = 6.84 (dd, J = 8.3, 0.7 Hz, 1H),

6.67 (d, J = 8.3 Hz, 1H), 4.80 (s, 1H), 4.18 (s, 1H), 3.65 (s, 3H), 2.77

(br d, J = 2.1 Hz, 1H), 2.28 (m, 1H), 2.20 (s, 3H), 2.10 (m, 1H), 2.08

(dd, J = 10.2, 6.2 Hz, 1H), 1.99 (dt, J = 12.2, 3.6 Hz, 1H), 1.86 (ddd,

J = 16.0, 10.6, 2.1 Hz, 1H), 1.76 (dt, J = 12.6, 5.7 Hz, 1H), 1.31 (m,

1H).

5-Chloromethyl-9-tert-butyldimethylsilyoxy-13-methoxy-4-

1,6 12,17

methyl-(1R,5R,6S,9S,10S)-11,4-oxazatetracyclo[8.7.0.0 .0

]

heptadeca-12(17),13,15-triene (82):

A flame dried round-bottom flask was set under static Ar atmosphere,

and was charged with LiCl (30 mg, 0.708 mmol). A solution of sub-

strate 81 (12.0 mg, 0.028 mmol) in CH₂Cl₂(3 mL) was introduced via

syringe, followed by Et₃N (5.6 mg, 0.055 mmol, in 100 µL of CH₂Cl₂,

stock solution). The mixture was cooled with stirring to 0°C and a so-

lution of MsCl (2.57 µL, 0.033 mmol) in CH₂Cl₂(100 µL, stock solu-

tion) was added. The reaction was monitored via HPLC (Prodigy 5 µ,

80% MeCN/20% H₂O, 5 mM triethylammonium acetate, UV λ_max

236 nm, 1.0 mL/min.). The starting material (T_r= 5.16 min) disap-

peared within 1 h, and the mixture contained 72% of the (presumed)

mesylate (T_r= 6.03 min.), and 9% of the desired chloride (T_r= 12.5

min.). After a total of 6 h, HPLC indicated less than 3% of the mesylate

and solvent was removed under reduced pressure. Flash chromatogra-

phy (silica gel, 8.6 g, EtOAc, saturated with NH₄OH) yielded pure 82

(10.9 mg, 87%) as a viscous oil; R_f= 0.55 (EtOAc/EtOH/NH₄OH

70:25:5); T_r= 12.51 min. (Prodigy 5µ, 80 % MeCN/ 20 % H₂O, 5 mM

triethylammonium acetate, UV λ_max236 nm, 1.0 mL/min.).

¹H NMR (500 MHz, MeOH-d₄): δ = 6.95 (dd, J = 8.3, 0.7 Hz, 1H),

6.84 (d, J = 8.3 Hz, 1H), 4.98 (s, 1H), 4.97 (s, 1H), 3.88 (s, 3H), 2.99

(br d, J = 2.1 Hz, 1H), 2.77 (ddd, J = 13.1, 10.4, 6.2 Hz, 1H), 2.72 (dt,

J = 12.3, 3.5 Hz, 1H), 2.57 (dd, J = 11.8, 4.9 Hz, 1H), 2.53 (s, 3H),

2.47 (dt, J = 12.8, 5.3 Hz, 1H), 2.38 (m, 1H), 2.29 (dt, J = 12.2, 3.5

Hz, 1H), 1.89 (br dt, J = 13.0, 3.2 Hz, 1H), 1.75 (br d, J = 13.3 Hz,

1H), 1.60 (br d, J = 10.3 Hz, 1H).

MS (FAB-Glycerine): m/z = 316.1561 (52%), 298.1383 (100),

225.1987 (62), 207.0850 (10), 115.0453 (6), 98.0988 (5).

MS (FAB-NBA): m/z = 316.1546 (62%), 225.1802 (17), 107.0620

(21).

SYNTHESIS

Papers

680

MS (EI, 70 eV): m/z = 315.1511 (62%), 258.1210 (17), 224.1931

(18), 143.1195 (38), 98.0914 (56), 70.0646 (100).

(8) a) Gates, M.; Tschudi, G. J. Am. Chem. Soc. 1952, 74, 1109.

b) Elad, D.; Ginsburg, D. J. Am. Chem. Soc. 1954, 76, 312.

c) Barton, D.H.R.; Kirby, G.W.; Steglich, W.; Thomas, G.M.

Proc. Chem. Soc. 1963, 203.

HRMS (FAB-Glycerine): 316.1561 (1.2 mmu), (C₁₈H₂₂NO₄+H) re-

+

quires, 316.1549. For (M+H) – H₂O: 298.1383 (6.0 mmu),

(C₁₈H₂₂NO₄+H) – H₂O requires 298.1443.

FAB-Nitrobenzyl alcohol: 316.1546 (0.3 mmu), (C₁₈H₂₂NO₄+H) re-

quires 316.1549.

d) Grewe, R.; Friedrichsen, W. Chem. Ber. 1967, 100, 1550.

e) Morrison, G.C.; Waite, R.O.; Shavel, J., Jr. Tetrahedron Lett.

1967, 4055.

EI (70 eV): 315.1511 (-4 mmu), (C₁₈H₂₁NO₄) requires, 315.1471.

f) Kametani, T.; Ihara, M.; Fukumoto, K.; Yagi, H. J. Chem.

g) Schwartz, M.A.; Pham, P.T.K. J. Org. Chem. 1988, 53, 2318.

h) Lie, T.S.; Maat, L.; Beyerman, H.C. Recl. Trav. Chim. Pays-

Bas. 1979, 98, 419.

The authors are grateful to Mallinckrodt Specialty Chemicals, TDC

Research, Inc., NSF (CHE-9315684, CHE-9521489, and CHE-

9615112) and the University of Florida for ﬁnancial support of this

work. Special thanks are extended to Khalil A. Abboud (University of

Florida) for determination of the absolute stereochemistry of diol 74

and Matthew R. Ellis (Summer Undergraduate Research Participant)

for his assistance in preparation of some starting materials.

i) Rice, K.C. J. Org. Chem. 1980, 45, 3135.

j) Evans, D.A.; Mitch, C.H. Tetrahedron Lett. 1982, 23, 285.

k) Moos, W.H.; Gless, R.D.; Rapoport, H. J. Org. Chem. 1983,

48, 227.

l) White, J.D.; Caravatti, G.; Kline, T.B.; Edstrom, E.; Rice,

K.C.; Brossi, A. Tetrahedron 1983, 39, 2393.

m) Ludwig, W.; Schäfer, H.J. Angew., Chem. 1986, 98, 1032;

Angew. Chem., Int. Ed. Engl. 1986, 25, 1025.

n) Toth, J.E.; Fuchs, P.L. J. Org. Chem. 1987, 52, 473.

o) Tius, M.A.; Kerr, M.A. J. Am. Chem. Soc. 1992, 114, 5959.

p) Parker, K.A.; Fokas, D. J. Am. Chem. Soc. 1992, 114, 9688.

r) Hong, C.Y.; Kado, N.; Overman, L.E. J. Am. Chem. Soc.

1993, 115, 11028.

(1) For a preliminary account on this work see: Butora, G.; Hudlic-

ky, T.; Fearnley, S.P.; Gum, A.G;. Stabile, M.R; Abboud K.A.

Tetrahedron Lett. 1996, 8155.

(2) White, P.T.; Raymer, S. “The Poppy,” National Geographic

Feb. 1985, vol. 167, pp 142–188.

See also Rice, K.C. in The Chemistry and Biology of Isoquino-

line Alkaloids; Phillipson, et al, Eds.; Springer: Berlin, 1985;

pp 191–203.

Terry, C.E; Pellens, M. The Opium Problem; Ch. 2, Bur. Soc.

Hyg. New York, 1928.

U.S. Drug Enforcement Agency, “Controlled Substance Aggre-

gate Production Quota History”, Federal Register, July 7, 1994.

National Narcotics Intelligence Consumers Committee Report

1993, “The Supplies of Illicit Drugs to the United States ”,

DEA-94066, August 1994.

s) Mulzer, J.; Durner, G.; Trauner, D. Angew. Chem. 1996, 108,

3046; Angew. Chem., Int. Ed. Engl. 1996, 35, 2830.

(9) a) Hudlicky, T.; Butora, G.; Fearnley, S.P.; Gum, A.G.; Stabile,

M.R. Studies in Natural Product Chemistry 1996, 18, 43.

b) Maier, M. Synthesis of Morphine in Organic Synthesis High-

lights II Waldman, H.; Ed.; VCH: Weinheim, 1995, pp

357–369.

(10) See ref. 9a for the list of approaches and dissertations dealing

with morphine synthesis.

Pollan, M. Harper’s Magazine 1997 35, 294.

(11) Hudlicky, T.; Boros, C.H.; Boros, E.E. Synthesis 1992, 174.

(12) Butora, G.; Gum, A.G.; Hudlicky, T.; Abboud, K.A. Synthesis

1998, 275.

(13) For reviews on Cascade or “Domino” reactions see Tietze, L.F.;

Beifuss, U. Angew. Chem.. 1993, 105, 137; Angew. Chem., Int.

Ed. Engl. 1993, 32, 131.

(3) a) Morphine, named after the Greek god of dreams, Morpheus,

is one of the oldest drugs on record. The first documented use of

opium goes back to 1500 BC and the power of the drug as an

analgesic is mentioned in sources as old as Homer’s Odyssey

and the New Testament.

b) The primary medical use of morphine is for the relief of very

strong pain, such as that derived from severe wounds, some tu-

mors or major surgery procedures. Because morphine acts on

respiratory function in higher doses while maintaining no influ-

ence on cardiac activity it is an ideal anesthetic for heart surge-

ry. Derivatives of morphine such as codeine are used as cough

suppressants, while antagonists such as naloxone or naltrexone

are used in the treatment of accidental overdose. Further chemi-

cal manipulation of morphine and other natural opium alkaloids

provides a wide family of agonists and antagonists that have be-

come essential in modern medical procedures.

(14) a) Stabile, M.R.; Hudlicky, T.; Meisels, M.L. Tetrahedron:

Asymmetry 1995, 6, 537.

b) Stabile, M.R.; Hudlicky, T.; Meisels, M.L.; Butora, G.; Gum,

A.G.; Fearnley, S.P.; Thorpe, A.J.; Ellis, M.R. Chirality 1995,

7, 556.

(15) Gibson, D.T. Science 1968, 161, 1093.

Gibson, D.T.; Hensley, M.; Yoshioka, H.; Mabry, T. J.

Biochemistry 1970, 9, 1626.

(16) Zylstra, G.J.; Gibson, D.T. J. Biol. Chem. 1989, 264, 14940.

(17) Hudlicky, T.; Endoma, M.A.E.; Butora, G. J. Chem. Soc., Per-

kin 1 1996, 2187.

(4) Wilson, M.I.; Coscia, C.J. J. Am. Chem. Soc. 1995, 97, 431.

(5) Kutney, J.P. Acc. Chem. Res. 1996, 26, 559.

Tencheva, Z.S.; Belcheva, M.M.; Coscia, C.J. Mol. Brain Res.

1997, 48, 156.

(18) For comprehensive reviews of arene cis-diol chemistry see:

Brown, S.M.; Hudlicky, T. In Organic Synthesis: Theory and

Applications; Vol 2; Hudlicky, T., Ed.; JAI Press: Greenwich,

CT, 1993; p. 113.

Scheneider, B.; Zenk, M.H. Phytochemistry 1993, 33, 1431.

Long, M.T.; Hailes, A.M.; Bruce, N.C. Appl. Environ. Microbi-

ol. 1995, 61, 3645.

French, C.E.; Hailes, A.M.; Rathbone, D.A.; Long, M.T.; Wil-

ley, D.L.; Bruce, N.C. Biotechnology, 1995, 13, 674.

Hailes, A.M.; Bruce, N.L. Appl. Environ. Microbiol., 1993, 59,

2166.

Widdowson, D.A.; Ribbons, D.W.; Thomas, S.D. Janssen Chi-

mica Acta 1990, 8, 3.

Carless, H.A.J. Tetrahedron: Asymmetry 1992, 3, 795.

Hudlicky, T.; Reed, J. In Advances in Asymmetric Synthesis;

Hassner, A., Ed.; JAI Press: Greenwich, CT, 1995; p. 271.

Grund, A.D. SIM News 1995, 45, 59.

Hudlicky, T. In Green Chemistry: Designing Chemistry for the

Environment, ACS Symposium Series; Vol. 626; Anastas, P.T.;

Williamson, T., Eds.; Washington, DC, 1996; p. 180.

Hudlicky, T.; Thorpe, A.J. J. Chem. Soc., Chem Commun. 1996,

1993.

Bruce, N.J; French, C.E.; Hailes, A.M; Long, M.T.; Rathbone,

D.A. Trends Biotech. 1995, 13, 200.

(6) Lenz, R.; Zenk, M.H. Tetrahedron Lett. 1995, 36, 2449.

(7) a) Muhtadi, F.J. “Analytical Profile of Morphine” in Analytical

Profiles of Drug Substances 1988 17, 259.

Hudlicky, T. Chem. Rev., 1996, 96, 3.

b) Muhtadi, F.J.; Hassan, M.A. “Codeine Phosphate” in Analy-

tical Profiles of Drug Substances 1981 10, 93.

Hudlicky, T.; Entwistle, D.A.; Pitzer, K.K.; Thorpe, A.J. Chem.

Rev. 1996, 96, 1195.

c) Wyatt, D.K.; Grady, L.T. “Heroin” in Analytical Profiles of

Drug Substances 1981 10, 357.

(19) Nicolaou, K.; Sorensen, E.; Hwang, C.-K.; Discordia, R.; Berg-

man, R.; Minto, R. U. S. Patent 5183942.

SYNTHESIS

April 1998

681

(20) Sakai, Y.; Nishiwaki, E.; Shishido, K.; Shibuya, M. Tetrahe-

dron Lett. 1991, 32, 4363.

(21) Nicolaou, K.; Zuccarello, G.; Ogawa, Y.; Schweiger, E.; Ku-

maa, T. J. Am. Chem. Soc. 1992, 110, 4866.

(22) Jones, R.; Bergman, R. J. Am. Chem. Soc. 1972, 94, 660.

(23) Nicolaou, K.; Zuccarello, G.; Riemer, C.; Estevez, V.; Dai, W.-

M. J. Am. Chem. Soc. 1992, 114, 7361.

(30) David, S.; Hanessian, S. Tetrahedron 1985, 41, 643.

(31) Hudlicky, T.; Stabile, M.R.; Gibson, D.T.; Whited, G.M. Org.

Syn. 1998, 76, in press.

(32) Ishizaki, M.; Ozaki, K.; Kanematsu, A.; Isoda, T.; Hoshino, O.

J. Org. Chem. 1993, 58, 3877.

(33) Mitsunobu, O. Synthesis 1981, 1.

Huges D.L., Org. React. 1992, 42, 335.

(24) Semmelhack, M.; Gallagher, J. Tetrahedron Lett. 1993, 34,

4121.

Mitsunobo, O.; Jenkins, I.D. in Encyclopedia of Reagents for

Organic Synthesis, Paquette, L.A., Ed.; Wiley: New York,

1995; Vol. 8, pp. 5379.

(25) Vollhardt , P.; Winn, L. Tetrahedron Lett. 1985, 26, 709.

(26) Namy, J.; Souppe, J.; Kagan, J. Tetrahedron Lett. 1983, 24, 765.

(27) Vollhardt, P.; Berries, B. Tetrahedron 1982, 19, 2911.

(28) For a review see: Block, E. Org. React. 1984, 30, 457.

(29) For a review see: Paquette, L. Org. React. 1977, 25, 1.

(34) Hudlicky, T.; Boros, C.H. Tetrahedron Lett. 1993, 34, 2557.

(35) Hudlicky, T.; Boros, C.H.; Boros, E.E. Tetrahedron: Asymme-

try 1993, 4, 1365.

(36) Kraus, A.G.; Gottschalk, P. J. Org. Chem. 1983, 48, 2111.

Article Doi

Toward a practical synthesis of morphine. The first several generations of a radical cyclization approach

DOI: 10.1055/s-1998-5931

Source and publish data:

Authors:

Article abstract of DOI:10.1055/s-1998-5931

Full text of DOI:10.1055/s-1998-5931

Products guided by the article

R&D Labs maybe for 207788-80-5

Relevant to this article

Hot Product