‘ —-—-

”is; .
Michigan State
University

 

 

This is to certify that the

dissertation entitled

FORMAL TOTAL SYNTHESES OF
(:)— AND (+)-APHIDICOL|N

presented by

Yu-Hwey Chuang

has been accepted towards fulfillment
of the requirements for

Ph.D.
_docia£aJ_de8reein_Chemi_'-d_ny_

\

Steven P. Tanis

 

Major professor

Date March 30, |987

uclr:..-‘a;_-.;...4 - r, u? - . ... 0.12771

 

 

 

MSU

LIBRARIES
-_.

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

 

 

FORMAL TOTAL SYNTHESES OF

(i)- AND (+)-APHIDICOLIN

BY
Yu-hwey Chuang

A DISSERTATION

submitted to
Michigan State University
in partial fulfillments of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1987

ABSTRACT

FORMAL TOTAL SYNTHESES OF
(:)- AND (+)-APHIDICOLIN
by
Yu-hwey Chuang

Formal total syntheses of (1)- and (+)-aphidicolin 1, an
antiviral substance which is used in the study of DNA
replication, have been accomplished (Scheme XVI). The
syntheses employ a furan terminated cationic cyclization of
(i)’ or (+)-107, which were prepared by coupling (LiZCuCl4)
the Grignard reagent derived from 3-chloromethylfuran with
(i)-(79%) or (+)-6,7-epoxy-8-benzyloxygeranyl chloride (75%).
Exposure of compound 107 to BF3-0Et2 yields (i)-108 (66-72%)
and (+)-108 (66-72%). Oxidation and selective reduction of
the 3-one followed by debenzylation provides (i)-85 (78%) and
(+)-85 (74%). Alkylation, furan oxidation and double-bond
reduction affords (i)-18 (67%) and (+)-18 (61%); an advanced

intermediate in McMurry's total synthesis of (:)-1.

ACKNOWLEDGMENTS

The author wishes to thank Dr. Steven P. Tanis for his

patience, support, guidance and friendship throughout this
project.

The author also wishes to acknowledge the members of the

faculty and staff for their assistance and advice throughout
this work.

The author wishes to thank her fellow students for their
advice and companionship.

Special thanks to my family for their love and support
without which this work would not have been possible.

TABLE OF CONTENTS

LIST OF TABLES

LIST OF FIGURES

LIST OF SCHEMES

INTRODUCTION

AN APPROACH TO THE SYNTHESES OF (i)- AND (+)-
APHIDICOLIN 1 KIA FURAN TERMINATED CATIONIC
CYCLIZATION

RESULTS AND DISCUSSION

THE FORMAL TOTAL SYNTHESIS OF (+)-APHIDICOLIN
EXPERIMENTAL

BIBLIOGRAPHY

 

17

20

39

58

91

LIST OF TABLES

 

PAGE
Table 1 Saponification of 104 31
Table 2 Cationic Cyclization of 107 33
Table 3 L-Selectride Reduction of 111 in the presence
of metal salts 37

iii

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure
Figure

Figure

5a

5b

SC

5d

58

8

9

10

LIST OF FIGURES

Furan Equivalencies
Cationic Cyclization Intermediates of 83 and
107

Preparation of 114 and 115

1H-NMR (250 MHz, CDC13) Spectrum of

1H-NMR (250 MHz, coc13) Spectrum of

1H-NMR (250 MHz, CDClB) Spectrum of

115
(i) -85
(i) -85

with First Addition of Eu(hfpc)3

1

H-NMR (250 MHz, CDC13) Spectrum of (:)-85

with Second Addition of Eu(hfpc)3

1H-NMR (250 MHz, CDC13) Spectrum of (+)-85

with Zero to Second Addition of Eu(hfpc)3
lH-NMR (250 MHz, coc13) Spectrum of (+)-85
with Third to Sixth Addition of Eu(hfpc)3
Preparation of 116 and 117

1H-NMR (250 MHz, CDCl3) Spectrum of a mixture
of 118 and 119

Extended Newman Projection of 118 and 119
1H-NMR (250 MHz, coc13) Spectrum of 119

Extended Newman Projection of 120 and 121

iv

PAGE

18

33

41

43

44

45

46

47

48

49

51

52

53

54

Scheme
Scheme
Scheme
Scheme
Scheme
Scheme

Scheme

Scheme
Scheme
Scheme
Scheme
Scheme
Scheme

Scheme

Scheme
Scheme

Scheme

II

III

IV

VI

VII

VIII

IX

XI
XII
XIII

XIV

XV

XVI

XVII

LIST OF SCHEMES

Trost's synthesis of (:)-1

McMurry's synthesis of (i)-1

Corey's synthesis of (:)-l

Ireland's synthesis of (i)-l

van Tamelen's synthesis of (i)-1

Marini Bettolo's synthesis of (i)-1
Proposed Retrosynthetic Scheme to 1 and
its analog 2

Preparation of 84

Preparation of 86

Preparation of 96

Preparation of (i)-18

Preparation of 104

Preparation of 107

Possible Routes to Accomplish Synthesis
of 18 from 108

Preparation of 85

Formal Total Synthesis of (-)-18

Conversion of 18 to 2

12

14

16

19

21

23

26

27

30

32

35

36

41

57

W at l+/-)- MM

 

I. INTRODUCTION

Aphidicolin 1, a diterpenoid tetrol produced by the mold
Cephglgfipgrium aphidiggla Petchl, has provoked wide interest
in recent years owing to its striking biological activity;
more than 100 publications have appeared in the last 7 years
reporting its use in biochemical studies. For example,
aphidicolin shows marked activity against herpes virus, both
in vitro and in the rabbit eye2'3. In addition, aphidicolin
possesses considerable antitumor activity in the C6 mouse

colon and 816 mouse melanosarcoma screens4 and has been shown

to inhibit the growth of leukemic T- and B-lymphocytess.

 

Aphidicolin appears to act as a specific reversible
inhibitor of DNA<1-polymerase6-8. Although the reasons for
this activity are not clear, molecular models reveal the
striking fact that all four of aphidicolin's hydroxyls can

very nearly touch the same flat surface, an observation that

1

2

may well be connected with the biological activity of the
molecule. It appears, however, that not all of the hydroxyl
groups are required for activity. Preliminary structure-
activity studies have indicated that the nonrigidly held
hydroxyl groups at C-17 and C-18 are less important than the
rigidly held hydroxyls at C-3 and C-16. Thus, acetylation of
the C-17 hydroxyl group causes only a 20% diminution of
activity against herpes simplex type 1 virus in human lung
cultures, but oxidation of the C-3 hydroxyl to a ketone
causes a 97% loss of activity.

On the basis of these data and on measurements made from
molecular models, McMiurry9 has hypothesized that
aphidicolin's actvity requires the presence of two hydroxyl
groups rigidly held on a flat carbon framework at a distance
of approximately 6.0 A. Other hydroxyl groups nearby may

also aid in binding but may not be required.

 

Triol 2 appears from molecular models to be an almost
exact duplicate of aphidicolin 1 with respect to the

disposition of its three hydroxyl groups. The intramolecular

3

distance between the hydroxyls at C-3 and C-12 in 2 is
identical with the distance between C-3 and C-16 hydroxyls in
aphidicolin (6.0 A), and the extension of hydroxyls below the
plane of the carbocyclic skeleton is also similar. McMurry9
prepared (i)-2 and (i)-3 and found that triol 2 showed 45% of
the activity of aphidicolin in inhibiting in vitro DNA
synthesis at a concentration of 0.33 pg / ml and 3 was found
to be inactive as expected. Since the material is racemic,
it is possible that the enantiomer of 2 whose absolute
stereochemistry corresponds to aphidicolin would exhibit 90%
of the activity of 1. This point would have to be verified
by synthesis and individual testing of the two enantiomers.

The challenges to synthesis offered by aphidicolin are
several. There is, for example, the stereochemical problem
presented by the A ring. Although the 3a+hydroxy-4a-
hydroxymethyl-4fldmethyl substitution pattern Seems familiar
at first glance, it is in fact unique. No other terpenoid
natural product possesses this stereochemistry. Another and
more severe stereochemical problem is that posed by the spiro
center, C-9. In addition this spiro center, is also adjacent
to another quaternary carbon, c-1o. The presence of these
two adjacent quaternary centers makes this region of
aphidicolin quite crowded and thus a likely source of trouble
during any projected synthesis.

The biological properties mentioned above, together with
the unusual structure of aphidicolin, have occasioned much

activity among synthetic organic chemists. As a result,

4
several total and formal total syntheses of (+/-)-aphidicolin

1 have been reportedlo' 11.

/,O

  

1) Meagfl', NaH
2) KOH. dioxane y

 

 

e . 1
. . 100-0, 16 h . . ( q )
FD“ ; .t . .
~ H 3) 2 N HCI “0 H
// 'h /’
HO 4 RT, overmg t l-D 1

For a synthesis of (1)-1, some simplification in the
target can be accomplished since the ketone 4, which is
obtained by degradation of 1, had previously been converted

into aphidicolin1 (eq. 1). Trost10a

has reported a synthesis
of (i)'4 (Scheme I) from the Wieland-Miescher type ketone 8.
Selective ketalization of 5 (79%) followed by a reductive
formylation gave a single aldol 7 (68%). Reduction of 3-keto
group of the fragile 7 to the 3a alcohol proved troublesome
until Trost discovered that the ate complex generated from t-
BuLi and (i-bu)2AlH gave the desired 8(99%) nearly
quantitatively. Hydrolysis afforded 9 (100%) which led to
acetonide 10 (92%) after exposure to acetone and p-TsOH.
Condensation of the ketone 10 with diphenylsulfonium
cyclopropylide12 under reversible ylide generation conditions
followed by silylation gave 11 (56%). Thermal rearrangement

of 11 gig flash vacuum pyrolysis provided a 2:1 (97%) mixture

of epimers of 12 at C-8. To circumvent this unacceptable

5

Sgheme_1 Trost's synthesis of (i)-1

 

(a) HOCHZCH OH, TsOH, PhH, reflux, Dean-Stark, 79%. (b) 1.
Li, NH , THF, 0.8 eq. of t-BuOH, -78°C, quench with isoprene
and than Et3N, MeBSiC1; 11. CH Li, ether, RT, and then -
78'C, cho, 68%. (c) (iBu)2(tBu)AlHLi, hexane, heptane,
ether, -73'c, 99%. (d) 3N HCl, THF,+RT, 190%. (e) acetone,
TsOH, reflux, 92%. (f) i. c-C3HSS Ph BF ,KOH, Me SO; ii.
PhSeSePh, NaBH4, DME, 60°C: iii. CH cfosiue31=NSiM , Et N,
PhH, 60'C, 56%. (g) Flash vacuum pyro ysis at 610’C, 37%. in)
Pd(OAc) , CH CN, RT, 73%. (1) 1. Li, NHB' THF, 0.8 eq. of t-
BuOH; ii. fie3siCl, 82%. (j) n-BuLi, HMPA, THF; inverse
addition to allyl iodide, 85'C, 35%. (k) 1.
(CH )2CHC(CH3)ZBH , diglyme, O'C, and then NaOH, H20 , 45°C,
57%? ii.pcc, NaOAc, cnzc1 , RT and then 2% KOH, c8303, RT,
54%. (1) 1. one, TsOH, anal , RT; 11. 95% NH NH , KOH,
HO(CH CHZO) H, 140‘-220'C, 91%? iii. 0.5% acetone, T on, RT,
78%; v. seZ (k)ii., 87%.

6

outcome, the mixture was directly oxidized (Pd(OAc)2, 73%) to
the enone 13. Dissolving metal reduction then gave a single
enol silyl ether (i. Li, NH3; ii. Me3SiCl, 82%) 13
corresponding to the minor product of the initial
rearrangement. The correct stereochemistry at C-8 presumably
resulting from protonation of the anion to afford an axial
C-H bond13. Inverse quenching of the generated enolate
solution into hot (85°C) excess allyl iodide gave 15 (29%)
which was converted to 4 (19%) as outlined in Scheme I. This
synthesis demonstrates the successful application of the
cyclopentanone annulation using the cyclopropylide reagent
for creating complex molecular architecture; however the
overall yield of the 20 step sequence is only 1.06% from the
ene-dione 5.

McMurry10b

has reported a concise stereospecific total
synthesis of (1)-aphidicolin from 6 (Scheme II). Alkylation
of 10 with methallyl iodide generated, as expected, 17 (89%)
with the desired stereocenter at C-8. Oxidative cleavage
afforded dione 18 (050

NaIO 86%) and aldol cyclization

4’ 4'
gave McMurry's crucial intermediate 19 (95%). The original
plan for the introduction of the elements of the D-ring
called for a conjugate addition to enone 19. However the
great degree of steric crowding at the enone ‘p-carbon (C-9)
precluded all chances for success. An intramolecular process
was much more likely to succeed and was examined as outlined
in Scheme II. Reduction of 19 (LAH, 95%) and vinyl ether

exchange (CH2=CH0Et, Hg+2, 90%) provided 20. The Claisen

.7

Schemg_11 McMurry's synthesis of (:)-1

 

CH , pyrrolidine, PhH, 75%: ii. HOCH CH OH,
C1,
20, 100%; iii. Li(sec-
C12, 85%. (d) 1.2 eq.
6, 89%. (e) Trace of

(a)i. CH =cncocn

p-TsOH, Benzeze,280 . (b) i. Li, NH , THF and then Mg SI

Et N, 97%; ii. Ch Li, THF and then C

Bu? BH, THF. (c) H coca , p-TsOH, CH
of LICA, THF and than me hallyl iodid
OsO , NaIO , H O, dioxane, 86%. (f) NaH, Trace of t-amyl
alcghol, bénzeng, reflux, 95%. (g) i. LAH, ether, 95%;
ii.CH cnzoca=cnz, Hg(OAc)2, 90%. (h) 0.03% NaO-t-pentyl,
toluefie, 220'c, 60%. (i) 1. LAH, THF, 93%; ii. p-TsCl, pyr.,
95%. (j) Na Fe(CO) , N-methylpiperidone, 30%. (k) i.
CH2=PPh3; ii. 504, p?r., then NaHSO3; iii. H30 , 36%.

8

rearrangement of 20 was carried out initially in the presence
of strong base to yield 21 (60%). Aldehyde 21 was then
further elaborated to tosylate 22 (i. LAH, 93%: ii. p-TsOH,
95%). Tosylate 22 was cyclized by Collman's reagent to
provide a 55% yield of 1:1 mixture of 4 and 23. The McMurry
synthesis concisely constructs the troublesome D-ring, albeit
with some regiochemical problems. The sequence described
above gives 4 in 15 steps and 4.89% overall yield.

Corey10c utilized a polyene cyclization to establish the
A and B rings of aphidicolin with the appropriate
stereochemistry (Scheme III). This strategy very efficiently
affords access to the basic carbocyclic skeleton upon which
the sterically congested spirocyclic D-ring as well as the C-
ring must be appended. Condensation of 8-TBDMSO-geranyl
bromide with the dianion of methyl acetoacetate afforded fl-
keto ester 25 (90%). Formation of enol phosphate 26 (NaH,
(C2H50)2P(O)Cl, no yield reported), followed by mercury II
induced cyclization gave 27 after treatment with aqueous NaCl
(60%). Ring A and B were thus established in a single step
with the proper relative stereochemistry at C-4, C-5, and C-
10. The requisite C-3-oxygen was introduced by free radical

oxygenation (NaBH 02); and the streochemistry was corrected

4,
by oxidation PDC, deprotection (Bu4NF) and L-selectride
reduction, giving acetal 28 (52%) after protection (tBuCHO,
H+). The addition of the spiro D-ring to keto aldehyde 29
was accomplished by the Michael addition to gaseous methyl

vinyl ketone (no yield reported) followed by a Robinson

9

Sgngm§_111 Corey's synthesis of (:)-l

602”“ M6020

(1)2Me
\ o
a O b OP‘OEI)2 C
w —’ 0 fl
\ 0‘ \. \ ;
24 25 26 cmg = 27
Teomso TBDMSO TBDMSO /

o

  

TBDMSO
O

 

(a) i. MsCl, Et3N, -20'C: LiBr; ii. MeLi, Na, methyl
acetoacetate, THF, O'C, 90%. (b) NaH, (CZH O) P(O)Cl, O'C.
(c) Hg(OTFA)2, CH N02, o‘c; aq. NaCl, 60%. (3) i. (HOCH ) ,
p-TsOH, 90%; ii.NgBH , o , DMF; iii. PDC; iv. n-Bu NF v? E-
selectride; vi. (CH CH5, TsOH, 58%. (e) i. LAH; ii. PCC;
iii. acetone-H o-Hald , 90%. (f) THF-tBuOH, DBU, K co ,
MVK(gas). (g) p§rrolidinium acetate. (h) i. (Me SiSCH )2CH3,
ZnI , 88%; ii. MeasiCN, ZnI , 97%; iii. DIBXL, 7;%. (i)
Me SiLi, HMPA, -35°c, 80%. (j? LDA (3 eq.), HMPA; H , 80%.
(k? i. NaZBH , ethanol, THF, -20'c; ii. tBuMe SiCl, DMAP,
Et3N, 90%; iii. 1,3-diiodo-5,5-dimethyl hydantoia, 86%; iv.

Pd/C; v. Bu NF; vi. p-TsCl, DMAP, Et N. (l) LiN(tBu)2,

H , 3
96%. (m) 1-etho§yethoxymethyllithium.

10
spiroannulation to give 31 (no yield reported). Selective

thioketalization of conjugated ketone ((MeBSiSCH ZnI

14

2-)2I 2]

88%) followed by the conversion of remaining ketone to the

corresponding trimetylsilyl cyanohydrin (Me3SiCN, anz,
97%)15 and reduction of the cyano with diisobutylaluminum
hydride gave aldehyde 32 (75%). Reaction of 32 with
trimethylsilyllithium in HMPA afforded 33 (80%) which was
treated with lithium diisopropylamide and HMPA followed by
quenching with aqueous acid giving 34 (80%). This method
which is a variant of the Peterson olefination technique
might be useful as a one carbon homologation protocol for
those hindered systems plagued by enolization. Reduction of
the aldehyde 34 (NaBH4), tosylation of the primary alcohol,
dithioketalization; and hydrogenation afforded keto-tosylate
35 (70%). Intramolecular alkylation of 35 to form the
aphidicolin ring system completely dependant upon reaction
conditions; in the event the use of kinetically controlled
enolate formation with highly hindered bases at low
temperatures favored the desired alkylation at C-12. Sodium
methoxide in methanol gave exclusively alkylation at C-15.
Reaction of 36 with 1-ethoxyethoxymethyllithium afforded a l
: 1 mixture of (i)- 1 and epimer at C-16 with no yield
reported. The Corey sequence, although effeciently
constructing the basic A-, B-ring system, suffers from
excessive length in the introduction of rings C and D. The
sequence described above gives 1 in 27 steps from 24 in an

overall yield of 6.5%, which does not take into account a

11
number of crucial steps for which yields are not reported.

Ireland10d

reported a stereoselective total synthesis of
(1)-aphidicolin from 37 (Scheme IV). Birch reduction of 37
gave 38 (52%). Methylation of 38 led exclusively to the
desired angularly methylated ketone which was converted to 41
as outlined in Scheme IV. Photooxygenation of 41 gave 42

(sens., AcOEt, O hu, 81%). The hetero-Diels-Alder reaction

2.
of 42 with the methyl(trimethylsilyl)methyl acrylate provided
43 (69%) which was further elaborated to 44 (95%). The
Claisen rearrangement of 44 gave 45 (84%) thus constructing
the 6 membered D-ring. The diazo ketone 46, formed through
the oximino ketone, was photolyzed in the absence of added
nucleophiles, and in an inert solvent medium to afford 48
(51%) as the sole product after cyclobutanone rearrangement
on silica gel. Direct removal of the unwanted C-13 carbonyl
was not successful because of the epimerization of C-8
center. Therefore, carbonyl group was reduced (99%) and
protected by the bulky t-butyldimethylsilyl group (95%) which
also blocked the a-face of the molecule. Osmium tetroxide
hydroxylation of the olefin 49 was now quite efficient and
gave a single diol, and then acetonide 50 (92%), as the only
product. Removal of the C-13 hydroxyl function through
reduction of its derived phosphorodiamidate was not
complicated by overreduction: then ketal exchange provided 51
(82%). Enolization and then trimethylsilyl chloride trapping
led virtually exclusively to the C-3,C-4-silyl enol ether

which was converted to 52 (90%) by procedure of Saegusa and

12

Sgn§m§__ly Ireland's synthesis of (i)'1

 

PhSH2C 5 3

(a) 1. Na, EtOH, NH -Et 0, 52%; ii. Al(OiPr) , CH3COCH ,
toluene. (b) i. KNH , NH -§tzo, Mel; ii. CH Li, E o. (c) i.
oxalic acid, HZO-MeOH, 93%; ii. Li, NH , Et 0, 86%. (d) 1.
Li, NH , Et 0, 66%: 11. (CH CH 0H) , p-TgOH, Benzene, reflux,
85%. ie) hgmatoporphrin diahlgridg, AcOEt, O , hv, 81%. (f)
CH =C(CO CH )CHZSiMe3, lzs'c, sealed tube, hyaroquinone, 69%.
(g? i. BIBXL, EtZO, -78'c: 11. Ph P=CH2, THF, 95%. (h)
150'c, 7.5 h, 84%. (i) i. n-BuLi, l-XmONo, THF, 85%; ii.
NH Cl, THF. (j) hv, EtZO, -73'-O'C. (k) SiO , gel, pet.
etfier/ether, 60%. (1) i. DIBAL, THF, -78'c, 99%? ii. TBSCl,
imidazole, DMF, 95%. (m) i. OsO4, pyr.: NaHSO3; ii.
CH C(OCH ) CH , p-TsOH-H 0; iii. TBAF, THF, 92%. (n) i. n-
BuLi, o /TMEBA, Me Npoci : +Me NH; ii. MeNH , THF, t-BuOH,
Li, NH Cl, 86%; iii. py3.H OT; , acetone, 95 . (o) i. KH,
THF; THSCl, Et N; 11. Pd(OAc) , CH CN, 90%. (p) (CH 0) ,
PhSH, Et3N, EtoH, 78%. (q) Li, NH , Z-Buon-THF; Et N, Tfisci-
THF, 90%. (r) 1. CH Li, THF: HCHB: AcOH; ii. L—silectrlde,

THF, aq. NaOH, 30% H282: 10% HCl-MeOH, 60%.

13

16

Ito . Thus, application of the Petrow reaction17

led to the
phenyl thiomethyl enone 53 (78%), which on lithium / ammonia
reduction and then trimethylsilyl chloride trapping produced
54 (90%). Compound 54 was converted to (1)-1 (60%) as
outlined in Scheme IV. This synthesis .used the
spiroannelation of related 2-methylene ketone through first
hetero-Diels-Alder condensation and then Claisen
rearrangement of the derived allyl vinyl ethers. The central
feature of this synthesis is the rearrangement of the
intermediate ((trimethylsilyl) methyl)cyclobutanone 47 to the
aphidicolin bicyclo[3,2,1] octane ring system. The sequence
described above gives (i)‘1 in 31 steps from 37 and 1.36%
overall yield.

van Tamelenloe

adapted his elegant biomimetic cationic
cyclizations to the synthesis of (1)-1; providing the correct
relative stereochemistry at C-3, -4, -5, and 10 (Scheme V).
Alkylation of the anion of geranyl phenyl thioether with p-
MeO-benzyl chloride provided 55 (86%); reductive cleavage of
the C-S bond (Li / NH3, 47%) and selective functionalization
of the terminal double bond by the van Tamelen procedure (1.
NBS, aq. tBuOH; ii. K2C03) gave epoxide 55 (33%). Epoxide

opening (LiNEt 57%); Sharpless epoxidation (tBuOOH,

2,
VO(acac)2) gave an (1)-erythro glycidol which was protected
as the corresponding benzyl ether 57 (72%). Cyclization
(FeCla, toluene) afforded but 12% of the requisite diol-mono
benzyl ether 58 which was subjected to Birch type conditions

(Li, THF, EtOH), hydrolysis, and acetonide formation to yield

14

figngm§_y van Tamelen's synthesis of (i)'1

 

(a) 1. base, p-methoxybenzyl chloride, 86%; ii. Li, NH3, -
78°C, 47%. (b) 1. N38, THF-HZO: 11. K CO3: MeOH, 33%. (c)
1. LiNEt , 57%; 11. TBHP, vo(acac?2, 80%: 111. NaH,
PhCH 1, 90%. (d) FeCl , toluene, 12%. (e) 1. Li, EtOH-THF, -
78‘c? 11. 0.5M HCl, fiton; 111. CH cocn , TsOH, 55%. (f) 1.
trisylhydrazone, TsOH, 70%: ii. BaLi, TMEDA, hexane, -78'C,
82%. (g) maleic anhydride, benzene, 80‘C, 86%. (h) i. Pd-C,
H O-EtOH, 90%; ii. Pb(OAc) , O -sat'd. pyr., 28%. (i) i.
MEPBA, 100%; ii. Na, PhH, reflux? iii. MsCl. (j) acetone-
HZO, reflux, CaCO3, 60%. (k) CrO3-pyr. 90%.

15
59 (55%).
Construction of the final ring was acomplished by a
Diels-Alder cycloaddition to diene 60 which resulted from the
Bond modification of the Shaprio reaction (1. 2,4,6-

iPrPhSOZNHNH ii. nBuLi, 82%). Addition of maleic anhydride

2.
to 60 gave adduct 61 (86%). Hydrolysis, and hydrogenation of
the diacid (90%) followed by Pb(OAc)4 decarboxylation of the
vicinal diacid yielded olefin 62 (28%). Epoxidation (100%)
of 62 and reduction with Na (20%) yielded an alcohol which
was converted to mesylate 63. Solvolytic rearrangement led
to alcohol 64 (60%) which upon oxidation (Collins, 90%) gave
ketone 4; thus completing a formal total synthesis of (:)-1.

The van Tamelen synthesis is the only one thus far
reported which controls the stereochemistry at C-3.
Unfortunately the extremely low yield obtained in the
cyclization coupled with a number of additional low yield
reactions and the poorly functionalized C-ring results in a
long and very low yield synthesis of ketone 4 (21 steps,
0.007%).

Marini Bettolo10f has also reported employing a
solvolytic rearrangement for a formal total synthesis of
aphidicolin (Scheme VI). The synthesis begins with known
diol 8 prepared by McMurry in 49% yield over 6 steps. Diol 8
was converted to the related bis-protected ketone (90%),
which was formylated (66; NaH, HCOZEt) and treated with
methyl vinyl ketone; the Michael adduct was deformylated

(NaOMe) and cyclized to provide 67 (75%). Photocycloaddition

16

Sghemg VI Marini Bettolo's synthesis of (1)-1

 

(a) see Scheme II, a and b, 6 steps, 58%. (b) i. NaH,
PhChZBr, reflux, 3 h; 6N HCl, 90%. (c) HCO Et, MeONa, PhH; 2N
H SO . (d) 3-buten-2-one, Et3N, 0°C: MeON , 75%. (e) allene,
ha, THF, -78‘c, 86%. (f) 1. p-TsOH, th, ethylene glycol,
reflux, 3 h, 80%. ii. 03, EtOH-CHZCl , -78'C; NaBH , -78'C,
4 h. (g) 1N HCl, THF; 1N NaOH, 76%. (h) 1. NaH; ész, THF,
reflux, 2.5 h; MeI, reflux, 1.5 h, 72%; ii. NaBH ,
EtZO/MeOH. -20'C, 91%; (h) i. o-xylene, reflux, 2.5 h, 90%;
ii. Et N, CH Cl , 0°C; MsCl. (j) acetone-H20, 70'C, 5 h,
97%. (k 1. p05, 31%; 11. Li, NH3, -78'c, 69%.

-17

of 67 with allene at -78'C gave only 68 (86%) which led to
protected aldol adduct 69 (61%) as outlined in Scheme VI.
Ketal hydrolysis and rearrangement provided 70 (76%); which
was converted to the corresponding dithio carbonate (72%);
and the ketone was reduced (NaBH4) to yield 71 (91%). Dithio
carbonate elemination (A, 90%); followed by mesylation gave
72, which suffered smooth solvolytic rearrangement to give 73
(97%). Oxidation (PDC, 81%) followed by Li / NH3 treatment
provided diol 74 (69%) which can be protected as the
corresponding acetonide to give 4 (85%).

The Marini Bettolo sequence borrows from the routes of
Trost, McMurry and van Tamelen to create the sequence. This
formal total synthesis leads to 4 in 5% yield from 2-methyl

1,3-cyclohexane dione.

II. An ApproaCh to the Syntheses of (¢)- and (+)-

Aphidicolin 1 gig Furan Terminated Cation Cyclization

We have demonstrated the use of furans as terminators in
a variety cationic cyclizationsls. As illustrated in eq. 2,
the generation of an electron deficient center (R) in the
side chain of a 3-substituted furan, should lead to 76 after

an electrophilic attack and rearomatization. The products of

(eq. 2)

/\ —-—»/\. ——-—/\
R

'75 ‘76

18

Figure 1 Furan Equivalencies

these cyclization sequences can serve as precursors to a
plethora of functional groups such as those illustrated in
Figure 1. Therefore, as part of a general program in furan
chemistry and an interest in furans as dianion equivalents in
annulation processes19 we were interested in accomplishing
formal total syntheses of (+/-)- and (+)-1 as outlined in
Scheme VII. This study would exploit; (1) a furan-terminated
polyolefin cyclization to rapidly establish the carbocyclic
nucleus of aphidicolin 1, followed by (2) a furan to
butenolide or dione conversion ultimately yielding McMurry's
intermediate dione 18. Cyclization of 79 would afford 78 in
which the A and B rings of the target have been established.
With a furyl moiety integrated into the cyclization product,

an efficient transformation to the McMurry intermediate 18

19
§£D§m§_YII Proposed Retrosynthetic Scheme to Aphidicolin 1

and its analog 2

      

O
1
2 => => .
0‘“ E O“
' H
1 8
o/ +0

McMurry
intermediate

 

OR

should be possible. The advantages of the sequence are rapid
construction of the carbocyclic nucleus with the correct
relative configuration at C-4, C-5, and C-10 resulting
directly from the cyclization. In addition the possibility
of an asymmetric synthesis will be investigated, utilizing
optically pure 79, a product of an asymmetric Sharpless

epoxidationzo.

20

III. Results and Discussion

An analysis of the substituents located at C-3 and C-4
of aphidicolin suggests, as in the Corey synthesisloc, that
an (n-oxygenated geranyl halide, after alkylation and
cyclization would yield the proper relative stereochemistry
at C-4. In addition, should the cyclization substrate
possess a 6,7-epoxide, then the cyclized product would have
an oxygen at C-3, however opposite in stereochemistry to that
desired. We viewed this route as being extremely flexible as
one could readily incorporate an optically pure side chain
into furan 79, utilizing (-)-6,7-epoxy-8-TBDMSQ-geranyl

20

acetate 81 previously prepared by Sharpless as the starting

material.

Our original route to (1)-1821 is described in Scheme
VIII. In the planning stage, we considered the nature of the
protecting group for the C-8-OH incorporated into geranyl
acetate 81. This group must survive conversion to the
chloride 82, coupling22 with 3-ClMgCH2furan, cyclization and
hydroxyl inversion. We anticipated utilizing the 3-one
related to 84 as the vehicle for OH inversion. However, as
has been demonstrated by many other groups exploring the
construction of 1 with the exception of van Tamelen, the C-4-
CHZOR must be deprotected to obtain good selectivity in the
reduction of 3-one. This consideration and the expected

fragility of the derived aldol caused us to employ the TBDMS

group to protect at C-4.

21
W11 Preparation of 84

 

W 1) ACCOCI, pyr. (99%) W

OAc
OH:2) 8e02, TBHP (72%) I 8 0
0H

 

 

1) MCPBA (85%) \ 1) MeOH, K2003 (99%)
2) tBuMeZSiCI 0... ON: 2) nBuLi p -,TsCI LiCl

(94%) OTBDMS 3‘ HMPA (85%)

 

LiZC uCI
MCI Q MgCI \
(79%)

OTBDMS 82
0 83

813-0512 (3 eq.), NEt3 (1.5 eq.);
(25-30%)

 

 

22

In the event, allylic hydroxylation of geranyl acetate

20

using the catalytic SeO protocol of Sharpless , afforded

2
the monoprotected diol 80 (72%). Henbest epoxidation (85%)

and protection (tBuMeZSiCl) afforded silyl ether 81 in 94%

yield. Acetate cleavage (KZCO MeOH, 99%) and chlorination

3!
by the procedure of Stork23 gave the highly oxidized geranyl

chloride 82 in 85% distilled yield. Chloride 82 was smoothly

22,24

coupled with 3-ClMgCH2furan in the presence of Li CuCl4

2
to yield 83 (79%). With 83 in hand, we next examined the

crucial cyclization reaction.

Our first cyclization attempts employed ZnI2 and
Ti(OiPr)3Cl as Lewis acids. These catalysts performed
admirably in our earlier furan terminated epoxide initiated
cyclizations, and in fact these Lewis acids gave 62% and 65%

yields of 3B-OH pallescensin A, the 4a-CH analog of 84, from

3
epoxydendrolasin. In this case, exposure of 83 to ZnI and

2
Ti(OiPr)3Cl provided ‘wildly variable yields (0-49%) of 84.
These results are unacceptable in the racemic route described
and are considered useless for a projected optically pure
synthesis of (+)- 4. After considerable experimentation we

discovered that consistent 25-35% yields of 84 could be

realized after exposure of 83 to BF -OEt2 (3 eq.), Et3N (1.5

3
eq.) in CHZClz-hexane-PhH (1:1:1) at -78°C. Under these
conditions the bulk of the reaction mixture consisted of
acyclic ketones. With tricyclic 84 available we next
examined the inversion of the C-3-Q bond.

As is described in Scheme IX, treatment of alcohol 84

23

Sghgme_1x Preparation of 86

1) PCC (91%)
2) nBu4NF

 

3) L-selectridei
(97%)

   

  

(CH3)ZCO

 

O
‘.

((30211)2 (90%) 7 0

+0/ 86

0c provided the 3-one in 91% yield. Deprotection
10b,c

with Pcc1

(Bu4NF) and reduction with L-selectride gave diol 85
(97%). Exposure of 85 to acetone and oxalic acid in CH2C12
afforded the requisite acetonide 86 in excellent yield (90%).

With 86 in hand we examined the conversion of the furyl
moiety to more useful functionality. Initially we attempted
to introduce the necessary methyl at the unsubstituted furyl
a-position by direct metallation and quenching with CH3I.

5

Unfortunately all attempts2 to deprotonate furan 86 and

capture with CH3I or D20 afforded no products of electrophile

capture. Application of more forcing conditions led to
destruction of the acetonide moiety. Changing protecting

OH to MEM, Me

groups for the 3a-OH and 4a-CH Si, or tBuMeZSi

2 3
did not improve the situation. These observations are in good

24
agreement with the similar difficulties in the thiophene
series noted by Heathcock26 during an approach to the
Securinega alkaloids.

A potential solution to this problem might lie in a
selective and careful bromination of the furyl nucleus,
followed by metal-halogen exchange and capture with an
electrophile. In the model compound 87 (eq. 3)27, application
of the mild bromination conditions of Mitchell28 (NBS, DMF)
provided 88 in 84% yield. Repetition of this sequence (eq.
4) with the MEM ether 89 gave bromide 90 (88%) which suffered
smooth metal-halogen exchange (n-BuLi, -78’C) and capture

with Me3SiCl to provide silylfuran 91 (85%).

 

   

 

    

NBS (eq' 3’
DMF
(84%)
R
\ 1)NBS,DMF(88°/o)
2) nBuLi, ~78'C (eq- 4)
3) TMSCI, HMPAV
(85%) MEMO ‘
H 89 R=H
9° R=Br
91 news

Oxidation of 91 (eq. 5) utilizing the procedure reported

by Kuwajma29 (CH3CO3H) led to a mixture of butenolides 92 and

25
93 (4:1, 77%). Enol lactone 92 was readily isomerized to 93
upon attempted purification. Compound 92 is potentially quite
useful in the formal total synthesis of aphidicolin: as the
addition of CH3-M would afford the desired dione directly3o.

Unfortunately, 92 was readily and easily converted to

butenolide 93 rendering this approach useless.

(eq. 5)

 

As a result of the double-bond isomerization of 92, we
examined the introduction of methyl group onto furan in the
alkylation step (Scheme X). In the event, the crude
brominated furan 90 was treated with BuLi and quenched with
MeI (-78‘C) to provide the desired methylated furan 94 in 55%
yield accompanied with 14% of starting material 88. The
mixture of 94 and 89 is not separable at this stage.

However, exposure of 94 to MCPBA31

gave ene-dione 94 in
excellent yield (95%); the unmethylated 89 survived the MCPBA
oxidation quantitatively. These compounds are now readily
separated by silica gel chromatography.

Ene-dione 95 was hydrogenated smoothly to the target
compound of the model system, 96, with the correct

stereochemistry at C-8 after the epimerization on a column of

silica gel. Dione 96 has a strong resemblance to the McMurry

26

Sgngmg_x Preparation of 96

 

     

1) NBS, DMF
2) nBuLi
3) Mel : .
(55%) M300 H 94
MCPBA
(95%)

  
 

H2, Pd/C

A

 

intermediate 18 in his total synthesis of (+/-)-aphidicolin.
Therefore the likelihood of success in the actual system is
high.

Careful bromination (NBS, DMF) of acetonide 86 afforded
the unstable bromide 97. Bromofuran 97 was not purified, but
was immediately subjected to metal-halogen exchange (nBuLi, -

78'C) and alkylation with CH I to provide 68% of the desired

3
methyl furan 98 and 8% of unreacted starting furan 86. The
mixture of 98 and 86 was submitted to the oxidation
conditions in the usual way to give ene-dione 99 in 97% yield
(based upon 98) and recovered 86 (8%). Compound 99 was

hydrogenated to provide the McMurry intermediate 18, which

was identical in all respects to material prepared by

27

§ghgmg_zl Preparation of (1)-18

  

 

 

 

 

1)NBS, DMF :
2) nBuLi, THF =
O. . u
3) CH3| 5 H 97 9:3;
(680/0) 0/ 9 8 RBCHa
o
MCPBA H2. I’d/C _
(97%) (99%)

 

 

McMurrylob; thus accomplishing the formal total synthesis of

(i)-aphidicolin 1.

This sequence was undertaken as a prelude to an
anticipated formal total synthesis of (+)-1 and to
demonstrate the utility of highly oxygenated, complex

substrates as initiators in furan terminated cationic

28
cyclizations. Although the route outlined did indeed provide
(i)-18, thus constituting a formal total synthesis of (i)-1,
the problems encountered in the cyclization affording 84 were
still unresolved. We viewed the 25-35% yield for cyclization
as unacceptable in chiral approach to (+)-1.

We assumed that the size and the fragility of the silyl
ether in 83 were responsible for the low and variable yields
of 84 which were obtained. This assumption was based upon a
poor mass balance and the observation of a sizeable amount of
acyclic ketone in the crude reaction mixture. We reasoned
that the bulk of the t-butyldimethylsilylether was preventing
the starting epoxide from reaching assuming a productive
conformation for cyclization: thus diverting sizable
quantities of material to unproductive epoxide opening. The
reasons for choosing of t-butyldimethylsilyl ether as the
protecting group of choice are (1) silyl groups can be
removed under mild conditions (Bu4NF) without inducing
retroaldol reaction: and (2) t-Butyldimethylsilyl group had
been employed in the cyclization approach to (i)-1 by
Coreyloc. Therefore, an alternate protecting function must
be carefully chosen to allow an inversion of the C-3-QH

without the loss of the C-4-CH OH, 21a retro-aldol of the

2
presumed intermediate 3-one. Of the seven total and formal-

total syntheses reported in the literature only the elegant
erythro-glycidol initiated cationic cyclization approach of

Ge

van Tamelen1 constructed the C-3 and C-4 stereocenters

differently: hence our concern.

29
A benzyl ether is relatively small and stable compared
to an R3Si- group. Also, benzyl ethers are known to be
removed relatively easily by hydrogenolysis which might not
cause the retro-aldol decomposition. Therefore, a benzyl
ether was selected as the protecting group of choice for a

second generation synthesis.

 

o NaH, P1101123; O \ 0A0
0H 1 0 o OCHzPh
25

Benzylation of 6,7-epoxy-8-hydroxygeranyl-acetate 100
with benzylbromide and sodium hydride gave 101 albeit in 21%
yield (eq. 6). However as a result of scrambling of acetyl
group by NaH: and low mass balance we searched for other
alternatives. One obvious solution is the utilization of a
more robust C-l-QH protecting group. Benzoates are usually
stable to such conditions and can be removed without touching
the benzyl protecting group. Therefore, a benzoate was
selected as the protecting group of choice. Benzoylation of

geraniol ( PhCOCl, pyridine, DMAP, CH C12, 99%) followed by

2
application of the Sharpless catalytic allylic oxidation
(3e02, TBHP, 71%) gave 102. Compound 102 was then submitted
to Henbest epoxidation (m-chloroperoxybenzoic acid, CHZClZ)
to afford the 6,7-epoxy-8-hydroxygeranyl-benzoate, 103, in
100% crude yield. Without further purification compound 103
was submitted to standard benzylation conditions (NaH,

PhCHZBr, THF, 7h, room temperature) giving 104 in 22% yield

3O

figh§m§_x11 Preparation of 104

W04 1) PhCOCI’ pyr. 24W
2) $902, TBHP OCOPh

 

 

 

(71%)
MCPBA O NaH, THF(or EtzO)
(100%) 7 1 03 OCOPh PhCHZBr, Bu4N+r
0'" 5 h (88%)

 

 

 

NaH, THF W

PhCH2 Br OCOPh

7 h (22%) OCHzph 04

with 50% recovered starting material. This reaction did
indeed provide the desired 104: however, the reaction rate is
very slow and mass balance is low. We found that the
addition of the phase transfer catalyst (Bu4N+I-, NaH, THF(or
EtZO), 5h, RT) resulted in smooth benzyl ether formation to
give 104 in 88% yield. With 104 in hand, several standard
benzoate saponification methods (entry 1-6 in Table 1)32 were
examined, but none of them gave completely satisfactory

results (Table 1). Although LiOH in THF-H 0 (entry 6) did

2
provide the desired 105 in 91% yield, the extraordinary long
reaction time required rendered these deprotection conditions
less than useful. As we learned from the benzylation of 103,

a phase transfer catalyst can be a very important addend to

the reaction medium. Therefore, 2.2 equivalents of Bu4N+I-

31

Table 1 Saponification of 104

 

 

OCl-IzPh OCHzF’h 105
Conditions Results
1 1% NaOH / MeOH, 5 days no reaction
2 MeONa / MeOH, 2 days 10% conversion by GC
3 NaOH / MeOH, reflux, 4 h decomposed
4 LiOH / THF, reflux, 4 h decomposed
5 MeONa/ MeOH, reflux, 4 h decomposed
L'OH, THF-H 0 5:1
6 I 2 ( ) 91%
14 days
. + -
7 LIOH / THF, Bu4N l(2.2 eq.) no reaction
12 h
H
3 MeONa / M60 (99%)
Bu4N*l'(2.2 eq). 12 h

 

 

was added to the LiOH and NaOMe mixture (entry 7 and 8). As

is shown in Table 1, Bu4N+I- did not significantly accelerate
benzoate hydrolysis in the LiOH-THF system, but it did assist
the saponification of 104 to 105 by MeONa in methanol giving
105 in 99% distilled yield. With a supply of 105 insured we
examined 105 in the chlorination and coupling reactions.
Utilizing Stork's procedure23(l. BuLi, HMPA 2. p-TsCl,

LiCl), 105 was converted to chloride 106 in 85% yield (Scheme

32
XIII). This modified epoxygeranyl chloride 106 was then
coupled with the Grignard reagent produced from 3-furylmethyl

21'22 to afford a 79%

chloride in the usual way (LiZCuCl4)
yield of coupled product 107.

Sghemg_3111 Preparation of 107

o 1) nBuLi, HMPA M
. 5 CI
105 O” 2) TsCl, LICI 106

 

 

(Kit! Ph
Ph (85%) OCH2
MgCl 0
O
LiCuCl4 r
7996
( ) '107
PhCHZO

As was disscussed previously, the benzyl group is more
stable and smaller than the t-Butyl dimethyl silyl protecting
group. Therefore, we expected that the cyclization
intermediate 107 would experience less steric crowding and it
might more readily assume a fruitful cyclization conformation
(Figure 2). When 107 was submitted to the usual cyclization
conditions (Table 2) we obtained 30% of desired cyclized
product 108 and recovered 60% of starting material 107. The
cyclization process was pushed to completion with slight

modification of the Lewis acid-Lewis base solvent brew

33

ol-

   

Hach‘l'la 3 3 1 0 7
“30' “3'0 PhCHZO

I

H3C C’H3

£1gg;e_z Cationic Cyclization Intermediates

of 83 and 107

Table 2 Cationic Cyclization of 107

 
  

 

 

O \ O O
\ \ \
PhH, CHZCIZ, hex; _
o (1:1:1), -78'C ”0 - H O
BnO 1 O7 PhCHZO/ 103 PhCHZO 1 09
starting
Conditions I 108 109 material
BF3-0E12(3 eq.) 30% 0% 60%
Et3N(1.5 eq.)
BFa-OET2(6 eq.) 72% 5% 0%

 

ET3N(3 eq.)

34

(BFBOEt2(6 eq.), Et3N(3 eq.) in hexane, benzene, methylene
chloride (1:1:1), -78'C) to afford 72% of cyclized 108 and 5%
of acyclic ketone 109 (Table 2). This result suggested that
the rationale for the selection of the benzyl protecting
group might be correct and that the nature of the protecting
group plays a major role in the partitioning of 107 between
desired and undesired pathways. We also examined the
cyclization of 107 with Ti(OiPr)3Cl: which led to a mixture
consisting mostly of a monocyclic compound. The vast
improvement in the yield of the crucial cyclization raised
our hopes of successfully completing a formal total synthesis
of (+)-aphidicolin 1 21; a chiral epoxide initiated closure;
however we must demonstrate that the C-3-QH stereochemistry
can be adjusted.

Compound 108 is the 4-CH -O-benzyl equivalent of silyl

2
ether 84: the conversion of the 108 to furan acetonide 86
requires the inversion of configuration at C-3, a
deprotection of the 4-CH2-O-CH2Ph: although not necessarily
in that order: followed by acetonide formation. However as
was mentioned previously we were concerned that a protected
(benzyl ether) aldol intermediate would suffer a retroaldol
reaction when we attempted to deprotect. Therefore we
examined the direct inversion of the C-3-OH gig a Mitsunobu

3 of 108, and nitrite displacement34 of the mesylate

reaction3
of 108. Unfortunately, the highly hindered nature of C-3
hydroxyl of 108 precluded any success with these measures.

We planned then to accomplish the inversion of C-3

35

hydroxyl in a manner identical to that utilized in the

c b

Corey10 and McMurry10

syntheses of (:)-1: that is reduction
of the C-3-one with the C-4-CH2-OH unprotected. A number of
possibilities exist to accomplish this transformation: for
example (Scheme XIV) path A shows a debenzylation of the C-4-
CHZ-O-function to give 110, and in selective silylation of
the produced diol would give silyl ether 84 employed in our
first generation (Scheme VIII) synthesis. A second
possibility (path B) would proceed from ketone 111.
Deprotection of 111 would afford the aldol adduct encountered

in the first generation (Scheme IX) approach.

Sgh§m§_31! Possible Routes to Accomplish Synthesis of 18

from 108

 

36
We examined hydrogenolysis and dissolving metal cleavage
of the benzyl linkage of 108 under a variety of conditions to
no avail. We observed either no cleavage or furan reduction
along with cleavage. Therefore path A (Scheme XIV) could not
be employed and we examined path B. The cyclized 108 was

submitted to the Swern oxidation35

(DMSO, (COC1)2, EtBN, THF)
to afford 111 in 97% yield. Exposure of 108 to PCC provided

wildly variable yields (60-90%) of 111 (Scheme XV).

ggngmg_zy Preparation of 85

DMSO, (cool)2

 

   

112(40 psi), Pd-C
EtOH-HOAc 250:1 ’
( )Ho
12 h

 

 

Attempted hydrogenolysis of the benzyl ether 111 under a
variety of reaction conditions was uniformly unsuccessful. As

shown in Scheme XV, our most successful benzyl cleavage was

37
plagued with concomitant reduction of C-3 ketone. Diol 85 is
useful to us: however 110 is of no direct utility. Milder

nucleophilic thiol cleavage conditions such as those of

6

Fujita3 (BFBOEtZ, CH3CHZSH, CHZClZ) destroyed the starting

benzyl other 111.

Since we were unable to generate the aldol adduct
presumed to be essential for the reported selectivity of L-
selectride reduction, we began to exam the reduction of 111

with L-selectride in the absence and in the presence of a

Table 3 L-Selectride Reduction of 111 in the presence of

metal salts

 

 

 

EntryI Salts 113 108 % Yield
1 no salt 2.8 1 99
2 ZnI2 (1.5 eq.) 1.2 1 93
3 MgBrz-OEt2 (1.5 eq.) 7.0 1 95
4 MgBrz-OEt2 (2 eq.) 8.5 1 95
5 Ti(OiPr)3CI (1.5 eq.) 4.0 1 96
6 Ti(OiPr)4 (1.5 eq.) 5.7 1 93

 

38
variety of metal salts. We assumed that the proper
combination of metal salt and solvent would conspire to
provide a chelated intermediate which might be selectively
reduced37. Pre-complexation of 111 with ZnIZ: MgBrZ-OEtZ:
Ti(OiPr)3Cl or Ti(OiPr)4 followed by reduction with L-
selectride (Table 3) afforded ca. 1.2:1 to 8.5:1 ratios of
3d,/ 3fi alcohols. The best conditions (entry 4) 2 equivalents
MgBrz-OEt2 and L-selectride (CH2C12, -78’C) afforded a

respectable 8.5:1 mixture of 113 and 111 in 95% combined
yield.

 
   

 

 

 

LAH,THF (eq. 7)
12 h (94%):
m.
PhCHZO
Starting
material (
e .8
42%: q )

Our interest in the selective production of mono-
protected diol 113 was based on our assumption that the axial
3a-OH would be able to assist in the cleavage of the
neighboring 4-CH2-O-benzyl ether by analogy to the precedent

of Kutney38. In the event (eq. 7) treatment of 113 with LAH

39
in THF for 12 hours provided exclusively diol 85 in 94%
isolated yield. The mono-protected 3fi-OH diol 108 when
exposed to the Kutney conditions (eq. 8) for 8 days afforded
a ca. 1:1 mixture of diol 110 and starting material. This
result indicates the importance of neighboring group
participation in reductive debenzylation with LiAlH4. Diol
85 was converted to acetonide 86 ((CH3)2CO, (COZH)2, CaSO4,
CH2C12) in a slightly better yield (93%): and the remainder

of the sequence follows our previous described route (Scheme

x1) .

IV. THE FORMALITOTAL SYNTHESIS OF (+)fiAPHIDICOLIN

Having successfully overcome the cyclization and
protecting group difficulties in the racemic series we
examined the route to (+)—1 outlined in Scheme XVI. Our sole
concern in this sequence was the transmission of chirality
from the starting epoxy-furan to the product tricyclic
aphidicolin precursor. Given the accepted chair-chair type
folding of acyclic substrates in cationic cyclizations

39 and our

leading to the formation of six-membered rings
construction of the proper relative stereochemistry in the
racemic routes described above we were confident of a
successful outcome provided the C-3-Q bond is not compromised
in the process.

In the event, epoxidation of 8-HO-geranyl benzoate under

20

the conditions of Sharpless afforded the corresponding

40
epoxide in 71% yield (295% ee). The optical purity of (-)-
103 was determined by chiral shift reagent, Eu(HFPC)3, and
HPLC studies of its acetate 114 and Mosher ester4o (MTPA) 115
(Figure 3,4). We have been unable to detect the presence of
the enantiomer of (-)-103. Benzoate cleavage and
chlorination as previously described (Scheme XIII) afforded
the requisite allylic chloride (84%) which was coupled with
3-ClMgCH2-furan (LiZCuC14) to provide the desired cyclization
substrate ([aJD-2.168': c 0.764, CH C12) in 75% yield.

2
Exposure to BF -OEt and Et3N in the benzene, hexane,

3 2
methylene chloride mixture: at -78'C gave (+)-108 in 66-72%
yield ([d]D+46.813°: c 0.455, MeOH).
Swern oxidation led to (+)-111 (95%) which was reduced
(L-selectride, MgBrz-OEtZ) to provide the mono-protected diol

(+)-113 (83%). Treatment of (+)-113 with LiAlH gave (+)-85

4
(94%) which was converted to the desired acetonide as
described in Scheme XVI. Chiral shift reagent studies of 85
(Eu(hfpc)3, Figure 5), demonstrated that, within our limits
of detection, compound 85 was optically pure. However at
this point we cannot yet demonstrate that 85 possesses the
absolute configuration depicted.

Bromination, metallation, methylation and oxidation as
outlined previously yielded a chiral ene-dione which provided
(-)-18 after reduction: thus completing the formal total
synthesis of (+)-aphidicolin 1.

In order to prove that the observed transmission of

chirality was in the desired and anticipated sense, we needed

41

Ac20, pyr.
i .. OCOPh DMAP (92%):O OCOPh
o“ (-)-1 03
OH
MTPACI, pyr.
DMAP (20%) OCOPh

MTPAOO

 

 

 

Eigu;e_3 Preparation of 114 and 115

W Formal Total Synthesis of (-)-18

W“ (RA/Iv
\ \ t s , \ 092
W082 L- diisopropyl tartrate 1‘

O.
HO TBHP HO (+103

(71 %) [0.10 -5.769° (C: 0.364, 011,612)

 

1) NaH, PhCHzBr, .
Bu4Nl (88%) O LIZCuCl4
2) MeONa, MeOH

3) n-BuLi, HMPA 0

OBn (”.1 05 (75 4’)

 

 

4) p-TsCl, LiCI
(85%)

[0110 +2.179° (c: 0.78, CHZCIZ)

BF3-0E12 (6 eq.), NE13 (3 eq.)

 

 

PhH, CH20|2, 06H14’ ‘78°C
(66- 72% )

      

BnO/ (+)-108
[0L 1,, -2.168° (c=0.764, CHZCIZ) 1.110 +46.813° (c: 0.455, MeOH)

(-)-107

42

(COCI)2, omso, TEA

(95%)

 

BnO/ (+)-111

[01 10 +75.2° (c- 0.125, MeOH)

LiAlH,, THF

(94%)

 

BnO / (+)-‘| 13
[01 lo +53.64° (c- 0.11, MeOH)

(CH3)200, (002le

(93%)

 

7L 2
O (+)-86

[01 JD +22.60° (c-0.77,CH2012)

H2, Pd/C

EtOAc (96%)

 

/
; 0 (+99

[01 ID -95.08°(c. 0.061,MeOH)

L-selectride

(83%)

 

.5. H
HO / (+)—85
[01 ID +43.75° (c- 0.16, MeOH)

1) NBS. DMF
2) nBuLi

3) Mel (65%)
4) MCPBA (97%)

 

(+18

[a 10 -44.0°(o- 0.025,MeOH)

43

 

 

 

 

 

 

 

 

 

 

 

 

 

has no souuooom Induce .emz ohms mzzlma e ounces
(“P.Ori ii”.ti lr._..iil ran. «I r r i.b.“> r r rwrflr It .‘PH. . . .MLH. . . .c—p’. r . “Poe. . .LCLIM. r . “PM BRO. WWW. 9P~ Fawn» II I .OPOI I» in
. III. (Illllllll
L E 19.3 .1 s a...) I 4 1‘ \IIII
Qua n ~
_ .I

44

1.1rr

mouxwe mo nonhuman Induce .nmz ohms mxzum

e. w. e on m : erLLem am am we

.. a u m N n 8
rLl h r-rlr-plLl-rI.Flr1 P1thrr L.rrrLlrr,rr h.rrrlr~rurlrlrlrlbllklhl.rlrlbytlelr If rlrrhrlplrlr f- bllrbl? r

H

e ..
Fir-

cm enamfim

mo 80 MO I

Llrblirlrgflrlbvlrrhlrrrrh

 

 

£1? 1

 

 

 

 

45

mxoooceom to commence amuse

cues mmuxue mo souuooom Amaooo .nm: ohms mzznmH on choose

ELL
n. I.~ H.— OH fl.N ..n “.0 I.’ fi.f On H.“ I.U n 0 O... W... I..

Pr>hrrrbrPerhbrhhFrPPPPrPrr—bPPPPPIrrrPPPPblhrIDVLFFPHP—PPrF—rrrrbrbPyrbprPrPFPFrPPPF

 

 

91.3 I I .1

 

 

 

 

 

46

n. O.—

PrrbIPPDPFth

exoouceom no coeuecoc ccooom
cues mouxuv mo saucepan Induce .Nmz ohms mzzlma 0m ounces

Ct‘
fi.— 0” I." 0.0 “.n I,’ “.0 In I.» I.U H 0 IF n... I. n.

PP?»hHPPFPPt—rPhlrthFPrPLIrFPPFPPHPrLP—FIFIPLrNL-IP—yrbrrrk-lPFPbPrPPr—Iprrrhhrrfih

Illli Di

 

 

 

47

exooeceom uo coeuecoc ccooom on once

cues meux+e no souuooom Anaooo .nmz omme mzzumH on ounces
9. M— .N "N Dfi Mn 9' MICKLBW Hm I“ WU Eh Mn 9. fl. 3“
PP r—rbrrbrfsrrhrbPrt—PPr>PrP>L>PthrrPPhPEP~blfrrbrbrPPH pffwrtrihtr kthhr>rhhrkrhrrrr~>r
IL L»

 

P

:‘ 1 open

 

 

 

 

 

 

 

 

Influfll , (A . fl pause -
e _ a - 1 It

 

 

 

UCOUTW

 

 

 

 

 

 

 

 

 

 

 

48

exoouceom mo coeueccc cuxem on cheap

cues mm11+e mo eonuooom Amaooo .um: ohms msz1m on choose

H

CLL
m 9» W. 6N mN On “a 9.. fit an, mm .&0 W3 9.» WP 00 m0 0.

rrPFPrert.PhrPr>hFPLPFNPPFrLerFPrPHrI-rPFPrLPHFPFerPF—blrfibr_Phrrbihtrrble+PPrrrrrhrPhpr

111114 1# ounce

 

 

 

 

 

 

 

 

 

 

 

 

Willi/\I1 _ fimuXfim

113.: 1211 _
111451HgHawHHHWH5$HWHH<11111111312111111 a fi1 _ I1 cause
as;

 

 

 

 

 

 

 

49
to determine the absolute configurations of the cyclized
products. Attempts to prepare bis-bromobenzoate of (+)-110
and (+)-85, in order to utilize the exciton chirality
technique of absolute configuration determination41, were not
successful. After numerous attempts we obtained only mono-
bromobenzoates 116 and 117 (Figure 6). Therefore, we decided
to use the O-methyl mandelate ester for establishment of

absolute configuration42 recently elaborated upon by Trost.

DMAP, pyr.
(85%)

 

   

81431100201
DMAP, pyr. '
(80%)

 

   

Br-Ph002

Eigg;§_§ Preparation of 116 and 117

The technique recently described by Trost for
establishing the absolute stereochemistry of 2'-a1coho1
requires that a racemate and optically pure compound be

available. One converts both the optically pure series and

50
racemate to the corresponding O-methyl mandelate esters and

correlate the 1

H-NMR spectra of the esters utilizing the
following model. One views the ester gig an " extended Newman
projection " in which the intervening ester linkage is
omitted. That substitutent which eclipses the phenyl ring in
such an extended Newman projection is then always shifted
upfield, presumably as a result of the shielding it
experiences by the phenyl ring. These suggestions were
verified by Trost with an extensive series of examples, and
has been confirmed by X-ray.

Therefore (¢)-108 was converted to a mixture of
diasteriomers 118 and 119 with (S)-O-methyl mandelic acid,
DCC and DMAP in CH2C12 (84%) (Figure 7). According to the
Trost model (Figure 8) one would predict that the
diasteriomer corresponding to the coupling of (S)-O-methyl
mandelic acid with the undesired enantiomer of 108 (ie. 118)
would possess a C-4-CH3 which should not experience an
upfield shift: however then in the desired diasteriomer 119
the 4fl-CH3 should be shielded. We observed 2-methyl signals
for the mixture of 118 and 119 5=0.88 and 0.62 ppm
respectively. Therefore, according to the precedent
established by Trost we anticipate observing a methyl signal
for 119 prepared (eq. 9) from (+)-108 at 5=0.62 ppm.
Reaction of (+)-108 with (S)-O-methyl mandelic acid, DCC, and
DMAP in CH2C12 afforded an 83% yield of the corresponding

mandelate ester, which exhibits a C-QO-CH signal at 5=0.62

3
ppm (Figure 9) thus confirming the absolute configuration of

51

GHH 6cm OHH

oo ensures e co eonuooom Amaooo .nmz ohms mzz1m a ounces

H

m. e,_ m- e.~ min 9.1 m.n e... m... e.u min as «I... Q... n...
_

LL—yPkrr_>rrr_rrFPt—krrrbrrrri—.rlLIr—r..r—LIILLIl..r~trer?—Pr> rrrrI—ttrrk_+rrhbrt>l

fl?

1.1L; fi.fi:n§::3..l€i -2: ... 1.- _I . ..:. 44.111 -. ...- _, ...... . , .- 3%..

 

 

 

 

 

 

52

 

W Extended Newman Projection of

118 and 119

53

add no enuuowmm Amauau .umz omm. mzzuma m musmam

I . . . . . .
m a. m. 9 ~ ... ~ 0 n w.n 9+ mé c.m m.m ad m.u 3.5 m.~

_ _ _ p . r
r .rrrr rrrf rrrr rrrr rP+P_.... ..Pr rrrr rkrr Fr*r rprrr rrkrhr

...)...............;..xi.fi.i...l..r.1.,. 2..) .. .. . -

g m .. 1. _

 

 

 

 

 

 

54

 

A: 2-97; 2-68 A: 3.26' 3.16
(AB, J=7.9 Hz, 2H) (AB, J=8.6 Hz, 2H)
8: 3.65 (s, 2H) B: 4.26; 4.19

(A8, J=12.2 Hz, 2H)

W Extended Newman Projection of

120 and 121

55

o
PhVL
H... OH

OCH:

 

   

o
DCC.DMAP Ph
“0 A (83%) H““
a no/(+)-108 (>95%d.e.) 00Ha

the material in hand as that depicted in 119.

Further confirmation of the absolute configuration in
this study was obtained from a similar study employing
coupling (i)- and (+)-113 with (S)-o—methy1 mandelic acid
(DCC, DMAP, CH2C12) gave 120 and 121. Figure 10 illustrates
the Trost prediction for the NMR shifts associated with the
diasteriomers 120 and 121. We expect to observe an upfield
shift of the 4a-CH2-OBn protons of 120 relative to
diasteriomer 121. In this case we observe an AB pattern
assigned to 120 at 5-2.97 (d, J=7.9 Hz); 2.68 ppm (d, J=7.9
Hz) and signal assigned to 121 were observed at 5=3.26 (d,
J=8.6 Hz) and 3.16 ppm (d, J=8.6 Hz). The benzyl ether

signals should also be effected, and we assigned a resonance

0
Ph\/u\
H... 0H 0
00-43 Ph\/Il\
DCC, DMAPfi 0

H
BnO/(+)'113 (32%) OCH:

(eq. 10)

 

   

56
at 5=3.65 (s, 2H) to diasteriomer 120; and signals at 5=4.26
(J-12.2 Hz) and 4.19 ppm (J=12.2 Hz) to diasteriomer 121.
When (+)-113 was coupled to (S)-0-methy1 mandelic acid it
provided diasteriomer 121 as predicted (eq. 10).

We have described formal total syntheses of (i)' and
(+)-aphidicolin 1. The second generation approach affords
(:)- and (-)-18 in 16 steps and 13.9% and 8.4% yields
respectively from geraniol. This represents a considerable
improvement over the first generation synthesis of (1)-18 (16
steps, 5.9%). The completion of the preparation of (-)-18 was
accomplished with 395% ee; demonstrating the utility of the
epoxide initiated-furan terminated cyclization sequence for
the transmission of chirality in a predictable fashion.

An application of this chemistry to the synthesis of
optically pure 2 (Scheme XVII) is under way and will be
reported in due course. McMurry9 has reported that (¢)-2
possessed 45% of the DNA synthesis inhibitory activity of
(+)-1 at a concentration of 0.33gg/m1; should we be able
extrapolate to the correct enantiomer of 2 we would predict
that we would observe 90% of the DNA inhibitory activity of
(+)-1. Such activity in a much simplified structure might

provide routine access to a valuable biochemical tool.

57

W Conversion of 18 to 2

 

   

 

 

   

NaH ‘
t-amyl alcohoT .
(cat) 0"
1) H2, Pd'C
2) P-TsOH, HZO-MeOH
.OH 0
LAH
j 2 / 122

EXPERIHENTAL

General Tetrahydrofuran (THF) was dried by distillation
under nitrogen from sodium benzophenone ketyl: methylene
chloride was dried by distillation under nitrogen from
calcium hydride: N,N-dimethylformamide (DMF) was dried by
distillation at reduced pressure from phosphorous pentoxide;
hexamethylphosphoramide (HMPA) was dried by distillation at
reduced presure from calcium hydride; pyridine was dried by
distillation under nitrogen from calcium hydride:
diisopropylamine was dried by distillation under nitrogen
from calcium hydride. Petroleum ether refers to the 30-60'C
boiling point fraction of petroleum benzin. Diethyl ether
was purchased from Mallinkrodt, Inc., St. Louis, Missouri,
and used as received. n-Butyl lithium in hexane was purchased
from Aldrich Chemical Company, Milwaukee, Wisconsin and
titrated by the method of Watson and Eastham43. N-
bromosuccinimide was crystallized from hot water and dried in
gaggg over P205. All other reagents were used as recieved
unless otherwise stated. All reactions were performed in
oven (150'C) dried glassware under argon with the rigid
exclusion of moisture from all reagents and glassware unless
otherwise mentioned.

Melting points were determined on a Thomas-Hoover
capillary melting point apparatus and are uncorrected.

Infrared spectra were recorded on a Pye-Unicam sp-looo

infrared spectrometer or a Perkin-Elmer Model 167

58

59
spectrometer with polystyrene as standard. Proton magnetic
resonance spectra (la-nun) were recorded on a Varian T-60 at
60 MHz, or a Bruker WM—250 spectrometer at 250 MHz as
indicated, as solutions in deuteriochloroform unless
otherwise indicated. Chemical shifts are reported in parts
per million of the 6 scale relative to a tetramethylsilane
internal standard. Data are reported as follows: Chemical
shift (multiplicity (s = single, d = doublet, t = triplet, q
- quartet, m - multiplet, br - broad), coupling constant

(Hz), integration). 13

C magnetic resonance spectra were
recorded on a Bruker WM-250 spectrometer (68.9 MHz) and are
reported in parts per million from tetramethylsilane on the

scale. Electron impact (EI/MS) and chemical ionization
(CI/MS) mass spectra were recorded on a Finnigan 4000 with an
INCOS 4021 data system. High resolution mass spectra were
performed by the MSU Regional Mass Spectroscopy Facility;
Department of Biochemistry, East Lansing, MI. 48824.
Elemental analyses were performed by Spang Microanalytical
Laboratory, Eagle Harbor, MI. Flash chromatography was
performed according to the procedure of Still, et al44 using
230-400 mesh Merck silica gel and eluted with the solvents
mentioned. Analytical thin-layer chromatography was run on

either Macherey-Nagel polygram SIL G/UV pre-coated plastic

254
sheets or Merck SIL G/UV254 pre-coated glass plates. Spots
were visualized by either dipping into a solution of Vanillin
(1.5 g) in absolute ethanol (100 ml) and concentrated

sulfuric acid (0.5 m1) and heating with a heat gun or

60
spraying with a 5% solution of molybdophosphoric acid in

absolute ethanol and heating to 120'C.

2-Br0Io-4,5,5a,6,7,8,9,9a—octahydro-6,6,9a-trinethyl-
(Saa,10,9afl)-(i)-naphtho[1,2-b]furan-7-ol 88
To a solution of Bfi-hydroxypallescensin A 87 (0.12 g,

0.5 mmole) in dry DMF (5 ml) chilled in an ice-water bath was
added a solution of NBS (0.1 g, 0.5 mmole) in dry DMF (2 ml)
over a period of 5 minutes. Stirring was continued at O'C
for 40 minutes and then quenched with ice-water (50 ml),
ether (50 m1) and pentane (50 ml). The organic phase was
separated; washed with brine (100 ml); dried (MgS04); and
concentrated in 23939 to provide a brown solid. The crude
product was purified by chromatography on a column of silica
gel (230-400 mesh, 30 g, 25 mm o.d., 1:5 ether:pet. ether)
using the flash technique. Fractions 16 to 19 afforded 1.35
g, 84%, of (1)-88 as a white solid. TLC (Merck, UV-vanillin,
etherzhexane 1:5) Rt=0.06.

ln_nu3 (250 MHz, c0c13): 5: 6.02 (s, 1), 3.28 (dd, J=5.1,
11.2 Hz, 1), 2.53-2.25 (m, 2), 2.12 (dt, J=3.0, 12.2 Hz, 1),
1.90-1.25 (m, 7), 1.19 (s, 3), 1.05 (s, 3), 0.86 (s, 3).
zlzus (70 eV): 314, 312 (M+, 12.9, 10.9), 281, 279 (23.2,

24.8), 119, 117 (96.7, 100.0), 91 (42.5).

4,5,5a,6,7,8,9,9a—0ctahydro-7duethoxyethoxylethoxyl-6,6,9a-
trinethyl- (580:, 7,0, 983) - (i) -naphtho[1 , 2-b] furan 89

To a solution of diisopropylethyl amine (0.22 ml, 1.2

61

mmole) in dry methylene chloride (5 ml) chilled in an ice-
water bath was added methoxyethoxymethyl chloride (0.14 ml,
1.2 mmole) followed by a solution of (i)-3E-
hydroxypallescensin A, 87, (0.14 g, 0.6 mmole) in methylene
chloride (10 ml) over 15 minutes. The reaction mixture was
stirred at room temperature for 24 hours: then quenched with
ice-water (100 m1) and cast into ether (200 ml). The organic
phase was separated: washed with saturated aqueous NH4C1 (200
ml), brine (200 m1): dried (M9804) and concentrated in gaggg
to afford crude 89. The crude product was purified by
chromatography on a column of silica gel (230-400 mesh, 30 g,
25 mm o.d., 1:5 ether:pet. ether) using flash technique.
Fractions 15 to 17 afforded 0.183 g, 95%, of 89 as a
colorless viscous liquid. TLC (Merck, UV-vanillin,
etherzhexane 1:1)‘Rf-0.39.

ln_§n3 (250 MHz, c0c13): 5- 7.17 (d, J=2.0 Hz, 1), 6.12 (d,
J=2.0 Hz, 1), 4.88 (d, J=8.3 Hz, 1), 4.73 (d, J=8.3 Hz, 1),
3.88-3.64 (m, 2), 3.57 (t, J-4.2 Hz, 2), 3.40 (s, 3), 3.21
(dd, J=4.2, 12.5 Hz, 1), 2.53-2.24 (m, 2), 2.10 (dt, J=3.0,
12.2 Hz, 1), 1.92-1.26 (m, 6), 1.20 (s, 3), 1.02 (s, 3), 0.88
(S. 3)

31;n§: (33 eV) 322 (M+, 10.2), 231 (19.6), 201 (22.6), 177

(15.9), 89 (58.9), 59 (base)

2-Bro-o-4,5,5a,6,7,8,9,9a-octahydro—7diethoxyethoxynethoxyl-
5:6,9a-trilethyl-(5aa,10,9ap9-(i)-naphtho[1,2-b]furan 90

According to the general procedure outlined for the

62

bromination of 87, on an identical scale, MEM-ether 89 was
brominated to give 90 in 88% yield after purification by
chromatography on silica gel using the flash technique. TLC
(Merck, UV-vanillin, ether:hexane 1:1) Rf=0.42.

ln_un3 (60 MHz, c0c13): 5a 6.00 (s, 1), 4.87 (d, J=7.5 Hz,
1), 4.70 (d, J=7.5 Hz, 1), 3.87-3.40 (m, 4), 3.40 (s, 3),
2.54-2.13 (m, 3), 2.03-1.26 (m, 6), 1.20 (s, 3), 1.02 (s, 3),
0.87 (s, 3)

zlzus: (70 eV) 402, 400 (M+, 0.5, 0.5), 149 (2.5), 131

(2.2), 117 (2.1), 89 (37.0), 59 (base)

4,5,5a,6,7,8,9,9a-Octahydro-7duethoxyethoxylethoxyl-6,6,9a-
trinethyl-Z-trinethylsilyl- ( Sad, 7,0, 911;” - (i) -naphtho [ 1 , 2-b]
furan 91

To a solution of 90 (0.1 g, 0.25 mmole) in dry Et20 (5
ml) cooled in a dry ice-acetone bath was added nBuLi (2.48 M
in hexane, 0.11 ml, 0.27 mmole) and the resulting gelatin-
like solution was stirred for 30 minutes at -78'C. To the
reaction mixture was added trimethylsilyl chloride (0.038 ml,
0.3 mmole), followed by HMPA (0.6 ml). The cooling bath was
removed and the reaction mixture was stirred for 1 hour at
room temperature; then stored in a refrigerator for 10 hours.
The reaction was quenched with water (30 m1) and then cast
into ether (50 ml). The organic phase was separated; washed
with brine (50 ml): dried (M9804); and concentrated in gaggg
to provide a pale yellow liquid. The crude product was

purified by chromatography on a column of silica gel (230-400

63

mesh, 20 g, 20 mm o.d., 1:5 ether:pet. ether) using the flash
technique. Fractions 14 to 17 afforded 0.084 g, 85%, of 91
as white crystals, TLC (Merck, UV-vanillin, ether:hexane 1:1)
RFO.44.

ln_nn3 (250 MHz, c0013): 5- 6.37 (s, 1), 4.86 (d, d=8.1 Hz,
1), 4.72 (d, J=8.1 Hz, 1), 3.87-3.65 (m, 2), 3.56 (t, J=8.2
Hz, 2), 3.39 (s, 3), 3.21 (dd, J=5.4, 10.3 Hz, 1), 2.59-2.17
(m, 3), 2.00-1.26 (m, 6), 1.19 (s, 3), 1.02(s, 3), 0.88(s,
3), 0.21(s, 9)

z;;ua: (70 eV) 394 (M+, 3.7), 303 (5.8), 273 (9.6), 231

(6.0), 89 (70.3), 59 (base)

4,5,5a,6,7,8,9,9a-Octahydro-7-ethoxyethoxynethoxy1-6,6,9a—
trinethyl— ( Sad, 7,0, 9afl) - (1) -naphtho[1 , 2-b] furan-z (33) -one 92
and 4,5,5a,6,7,8,9,9a-octahydro-71-ethoxyethoxylethoxyl-
6,6,9a—trinethyl—(5aa,flfi,9afl)-(i)-naphtho[1,2-b]furan-z(9bfl)-
one 93

To a solution of 91 (0.15 g, 0.38 mmole) in dry
methylene chloride (4 ml) chilled in an ice-water bath was
added NaOAc (0.09 g, 1.1 mmole) in one portion, followed by
peracetic acid (0.4 m1, 6 mmole) over 2 minutes. The
reaction mixture was stirred at 5°C for 3.5 hours then
diluted with ether (50 ml) and washed with water (50 ml),
saturated aqueous NaHCO3 (2x25 ml), saturated aqueous sodium

thiosulfate (2x25 ml) and saturated aqueous NaHCO (25 m1)

3
followed by drying (NaZSOA). Removal of solvent in yagug

afford 0.085 g, of crude product. The crude 1H-NMR showed

64

80% of desired deconjugated butenolide 92. This crude
product was purified by chromatography on a column of silica
gel (230-400 mesh, 30 g, 20 mm o.d., 1:4 ether:pet. ether) to
afford 0.026 g, 20%, of 92, TLC (Merck, UV-iodine,
ether:hexane 1:1) Rf-0.2 and 0.045 g, 35%, 0f 93, Rf=0.02.

W: 11131118 (250 MHz. c0c13): 5= 4.86 (d.
J=8.4 Hz, 1), 4.72 (d, J=8.4 Hz, 1), 3.79-3.70(m, 2), 3.56
(t, J-4.2 Hz, 2), 3.39 (s, 3), 3.18 (dd, J=4.2, 11.2 Hz, 1),
3.10-3.05 (m, 2), 2.15-1.25 (m, 9), 1.67 (s, 3), 1.01 (s, 3),
0.86 (s, 3)

S a o : 1fi_uu3 (250 MHz, D -acetone): 5= 5.69

6
(t, J=1.2 Hz, 1), 4.80 (d, J-6.8 Hz, 1), 4.69 (d, J=6.8 Hz,
1), 4.48 (bds, 1), 3.72-3.66 (m, 2), 3.50 (t, J=4.2 Hz, 2),
3.30 (s, 3), 3.22 (dd, 324.2, 10.4 Hz, 1), 2.92-2.77 (m, 1),
2.49-2.30 (m, 1), 2.10-1.83 (m, 2), 1.68-1.22 (m, 5), 1.08

(s, 3), 0.85 (s, 3), 0.70 (s, 3)

4,5,5a,6,7,8,9,9a—Octahydro-71lethoxyethoxy-ethoxy1-2,6,6,9a—
tetramethyl-(Saa57p,9ap)-(1)-naphtho[1,2-b]furan 94

According to the general procedure for preparation of
98, 89 (0.38 g, 1.18 mmole, in 15 ml DMF) was brominated by
NBS (0.256 g, 1.3 mmole, in 3 ml DMF). The crude bromide 90
was submitted to lithium-halogen exchange by nBuLi (2.5 M in
hexane, 0.57 ml, 1.43 mmole, -78'C) and quenched with MeI
(0.1 ml, 1.61 mmole) to afford 94 in 55% overall yield from
89 with 0.053 g, 14%, of recovered starting material, 89,

after purification by chromatography on silica gel. TLC

65
(Merck, UV-vanillin, ether:hexane 1:1) Rf=0.4.
lgLuua (60 MHz, 001,): 5- 5.52 (s, 1), 4.75 (d, J= 7.0 Hz,
1), 4.55 (d, J= 7.0 Hz, 1), 3.73-3.38 (m, 4), 3.30 (s, 3),
2.53-2.10 (m, 3), 2.17 (8, 3), 2.00-1.26 (m, 6), 1.17 (S, 3),

1.00 (s, 3), 0.83 (s, 3)

Octahydro-S-(methoxyethoxymethoxy)-5,5,8a-trinethy1—(2—
oxopropylene) - ( 4afl, 6,0, Baa) -1-naphthalenone 95
According to the general procedure for preparation of

99, a mixture of 94 (0.22 g, 0.65 mmole) and 89 (0.053 g,
0.16 mmole) was treated with m-chloroperoxybenzoic acid
(0.135 g, 85%, 0.65 mmole) to provide desired 95, 0.219 g, in
95% yield and recovered the unreacted 89, 0.05 9, after
purification by chromatography on silica gel, TLC (Merck, UV-
vanillin, ether:hexane 5:1) Rf-0.21 for 95.

ln_nn3 (250 MHz, c0c13): 5- 5.83 (d, J-1.9 Hz, 1), 4.74 (d,
J=8.4 Hz, 1), 4.62 (d, J=8.4 Hz, 1), 3.69-3.62 (m, 2), 3.48
(t, J=4.2 Hz, 2), 3.30 (s, 3), 3.05 (dd, J=4.2, 12.5 Hz, 1),
2.60-2.20 (m, 2), 2.09 (s, 3), 1.88-1.45 (m, 7), 1.15 (s, 3),
0.94 (s, 3), 0.84 (s, 3)

31;n§: (70 eV) 352 (M+, 0.8), 246 (5.6), 136 (16.5), 121
(5.6), 89 (78.1), 59 (base)

13: (Neat) 2940, 2880, 1700, 1630, 1455, 1365, 1050, 930,

730 cm"1

0ctahydro-6-(methoxyethoxymethoxy)-5,5,8a-trinethy1—(2-

oxopropyl)-(2fl,4afi,6358aa)-1(2B)-naphthalenone 96

66
According to the general procedure for preparation of
18, 95 (43 mg, 0.12 mmole , in 10 ml EtOAc) was hydrogenated
to afford 96, 43 mg, in 99% yield as a colorless viscous
liquid after purification by chromatography on silica gel,
TLC (Merck, UV-vanillin, ether:hexane 5:1) Rf=0.26.

';n_NMB (250 MHz, c0013): 5- 4.79 (d, J=6.5 Hz, 1), 4.66 (d,
J=6.5 Hz, 1), 3.71-3.65 (m, 2), 3.52-3.48 (m, 2), 3.33 (s,
3), 3.09 (dd, J=4.2, 11.2 Hz, 1), 2.81 (dd, J=6.5, 16.4 Hz,
1), 2.15 (s, 3), 2.23-1.98 (m, 5), 1.93-1.42 (m, 6), 1.03 (s,

3), 0.93 (s, 3), 0.85 (s, 3)

8-Bro-o-4a,4b,5,6,9b,10,11,11a-octahydro-2,2,4a,9b-
tetranethyl- (43a, 4bB, 9aot, 11aa) -4H-furo [2 ' , 3 ' : 5 , 6]
naphtha [ 2 , 1-d] -1 , 3-dioxin (i) -97

To a solution of acetonide (1)-86 (80 mg, 0.28 mmole) in
dry DMF (10 ml) was added a solution of NBS (60 mg, 0.3
mmole) in dry DMF (4 m1) over 10 minutes at 0°C. Stirring
was continued for 20 minutes at 0°C and then 50 minutes at
room temperature. The resulting pale yellow solution was
diluted with water (100 ml), and then cast into ether (50 m1)
and pentane (50 ml). The organic phase was separated; washed
with water (100 m1), brine (100 ml); dried (MgSO4): and
concentrated in 29999 to provide 90 mg, 88%, of crude bromide
as a pale yellow solid which was readily transformed into the
methylated adduct without further purification. The crude
product was purified by chromatography on a column of silica

gel (60-230 mesh, 35 g, 9:1 hexane:ether), TLC (Merck, UV-

67

vanillin, ether:hexane 1:9) Rf=0.21, with some decomposition
to provide 40 mg, 40%, of pure bromide (1)-97 as a white
solid for spectral data.

ln_nu3 (250 MHz, c0c13): 5= 6.04 (s, 1), 3.74 (m, 1), 3.70
(d, J=12.8 Hz, 1), 3.34 (d, J=12.8 Hz, 1), 2.49-2.34 (m, 2),
1.98-1.48 (m, 7H ), 1.42 (s, 3), 1.36 (s, 3), 1.22 (s, 3),
0.79 (s, 3)

E;;n9 (70 eV): 370, 368 (M+, 2.0), 297, 295 (11.3, 11.4),

205 (17.9), 149 (99.7), 91 (28.2), 43 (base)

4a,4b,5,6,9b,10,11,11a-Octahydro-2,2,4a,8,9b-pentalethyl-
(4aa, 415,8, 980:, 11nd) -4H-furo[2 ' ,3 ' :5,6]naphtho[2, 1-d]-1, 3-
dioxin (1)-98

To a solution of crude bromide (i)'97 (90 mg, 0.24
mmole) in dry ether (10 ml) cooled in a dry ice-acetone bath
was added n-BuLi (2.5 M in hexane, 0.1 ml, 0.25 mmole) via
syringe and stirred for 30 minutes at -78°C. To the
resulting bright yellow solution was added methyl iodide
(0.02 ml, 0.32 mmole). The mixture was allowed to slowly
warm to room temperature over 1 hour and then stirred for 4
hours at room temperature. Then, this reaction mixture was
diluted with water (50 m1) and ether (50 ml). The organic
phase was separated: washed by brine (50 ml); dried (Na2804)
and concentrated 19 29999 to give a yellow liquid. The crude
product was purified by chromatography on a column of silica
gel (230-400 mesh, 20 g, 20:1 hexane:ether) using flash

technique. Fractions 20 to 24 afforded , TLC (Merck, UV-

68

vanillin, ether:hexane 1:9) Rf=0.26, 57 mg, 68%, of
methylated product (i)-98 and 6 mg, 8%, of starting acetonide
(1)-86 as a mixture. This mixture can be readily separated
at next step.

lH_993 (250 MHz, c0c13): 5- 5.70 (s, 1), 3.72 (t, J=3.0 Hz,
1), 3.70 (d, J=12.3 Hz, 1), 3.32 (d, J=12.3 Hz, 1), 2.50-2.35
(m, 2), 1.97-1.42 (m, 7), 2.21 (s, 3), 1.40 (s, 3), 1.33 (s,
3), 1.18 (s, 3), 0.77 (s, 3)
31:99: (70 eV) 304 (M+, 5.0), 289 (3.8), 231 (30.4), 43

(base).

Decahydro-3,3,6a,10b-tetra-ethyl-8-(2-oxoprqpy1ene)-(4ad,6aa,
83,103,10ba)-GH-naphtho[2,1-d][1,3]dioxin-7-one (i)-99

To a solution of the mixture (114 mg, 0.38 mmole, of
methylated acetonide (1)-98 and 12 mg, 0.04 mmole, of
acetonide (i)-86) in dry methylene chloride (6 ml) chilled in
an ice-water bath was added m-chloroperoxy benzoic acid (76
mg, 85%, 0.38 mmole) in one portion. Stirring was continued
for 1 hour at 0°C and then 7 hours at room temperature. The
reaction mixture was then diluted with ether (60 m1), aqueous
NaHCO3 (60 ml), water (50 m1), brine (50 ml), dried (Na2804)
and concentrated ig 29999 to give a yellow solid. The crude
product was purified by chromatography on a column of silica
gel (230-400 mesh, 30 9, 1:1 hexane:ether) using flash
technique. Fractions 13 to 14 afforded , TLC (Merck, UV-
vanillin, ether:hexane 5:1) Rf=0.2, 116 mg, 97%, of (i)'99 as

a white solid and recoverd 12 mg of starting acetonide (i)-

69
86.

lH_gug (250 MHz, coc13): 5- 5.99 (d, J=1.9 Hz, 1), 3.67 (t,
J=2.9 Hz, 1), 3.60 (d, J=12.4 Hz, 1), 3.31 (d, J=12.4 Hz, 1),
2.65-2.15 (m, 3), 1.91-1.56 (m, 6), 2.18 (s, 3), 1.39 (s, 3),
1.35 (s, 3), 1.23 (s, 3), 0.80 (s, 3)

91:99 (70 eV): 320 (M+, 0.3), 204 (30.0), 149 (base), 134
(54.2)

13 (Neat): 2940, 2850, 1690, 1625, 1455, 1385, 1195, 1095,

A1000, 915, 855, 730 cm"1

(')-99

According to the procedure outlined for bromination of
(i)-86, (+)-86 (150 mg, 0.52 mmole) was brominated with NBS
(112 mg, 0.57 mmole) in 3 ml of DMF to give crude chiral
bromide which was utilized without further purification. The
chiral bromide 97 was metallated by n-BuLi (0.3 ml, 2.43 M in
hexane, 0.73 mmole) and quenched with MeI (0.13 ml, 1.5
mmole) in 20 m1 of dry ether at -78°C to provide 102 mg, 65%,
of chiral 98 and 13.5 mg, 9%, of (+)-86 as a mixture which
was oxidized with MCPBA (70 mg, 85%, 0.34 mmole) in 30 ml of
CH2C12 to give 104 mg, 97%, of (-)-99.

[a]D= -95.08 (c: 0.061, MeOH)

Decahydro-3,3,6a,10b-tetranethyl-8-(2-oxopropyl)-(4aa,6aa,
83,103,10ba3-7H-naphtho[2,1-d][1,3]dioxin-7-one (i)-18
A solution of conjugated dione (1)-99 (50 mg, 0.16 mmole)

and Pd-C (10%, 0.01 g) in ethyl acetate (20 ml) was

7O

hydrogenated under H (1 atm) for 7 hours at room

2
temperature. The catalyst was then filtered off and rinsed
with ether (10 ml). The solution was concentrated 13 Egggg
and MeOH (20 ml) was added followed by K2C03 (0.5 g). The
resulting suspension was stirred for 5 hours at room
temperature. The solution was filtered through a pad of
silica gel and concentrated to give a pale yellow solid. The
crude product was purified by a column of silica gel (230-400
mesh, 10 9, 4:1 hexane:ether), TLC (Merck, UV-vanillin,
ether:hexane 1:1) Rf=0.13, to provide 50 mg, 99%, of (i)-18
as a white solid. m.p.=104-105°C.

ln_nng (250 MHz, c0c13): 5-r3.63 (t, J=2.7 Hz, 1), 3.58 (d,
J=12.5 Hz, 1),3.26 (d, J-12.5 Hz, 1), 2.93 (dd, J-8.4, 17.5
Hz, 1), 2.23-1.94 (m, 5), 2.18 (s, 3), 1.37 (s, 3), 1.32 (s,
3), 1.88-1.52 (m, 6), 1.19 (s, 3), 0.79 (s, 3)

1;9_uug (250 MHz, c0c13): 5==214.36, 207.74, 98.22, 72.65,
67,96, 47.92, 43.49, 41.73, 41.41, 35.88, 32.84, 30.57,
29.39, 26.04, 23.25, 20.23, 18.93 (2), 17.54

91:99: (70 eV) 323 (M++1), 307 (8.2), 171 (9.3), 158 (5.3),
119 (6.0), 107 (8.1), 105 (6.1), 93 (10.0), 86 (19.8), 84
(33.8), 43 (base) :3: (neat) 2930, 2870, 1705, 1455, 1375,

1255, 1235, 1195, 1160, 1090, 1000, 855 cm-1.

(-)-18

According to the procedure for the hydrogenation of (1)-
99, (-)-99 (30 mg, 0.1 mmole) was hydrogenated (1 atm. H2,
Pd-C) in 10 ml of ethylacetate to provide 29 mg, 96%, of (-)-

71
18.

[a]D= -44.0 (c: 0.025: MeOH).

6,7-Epoxy-8-benzyloxygerany1 alcohol 101

To a solution of 100 (4.3 g, 18.9 mmole) and benzyl
bromide (3.87 g, 22.6 mmole) in dry THF (100 ml) chilled in
an ice-water bath was added NaH (1 g, 20.8 mmole, 50% in oil,
washed with dry hexane twice) in one portion. The resulting
suspension was stirred at 0°C for 1 hour and room temperature
for 2 hours, then quenched with saturated aqueous NH4Cl (100
ml) and cast into ether (200 ml). The organic phase was
separated: washed with water (100 ml), brine (100 m1); dried
(Na2804): and concentrated 19 99999 to afford yellow liquid.
The crude mixture was separated by chromatography on a column
of silica gel (230-400 mesh, 30 9, 1:1 hexane:ether) to
provide a mixture of 101 1.405 g, 21%, TLC (Merck, UV-

vanillin, ether:hexane 3:1) R =0.18.

f
lfi_gug of 101 (60 MHz, coc13): 5- 7.30 (s, 5), 5.42 (t,
J=7.0 Hz, 1), 4.53 (s, 2), 4.10 (d, J=7.0 Hz, 2), 3.43 (s,
2), 2.83 (t, J=6.0 Hz, 1), 2.45-2.00 (m, 2), l.88-1.66 (m,

3), 1.66 (s, 3), 1.33 (s, 3)

8-Hydroxygeranyl-benzoate 102

To selenium dioxide (0.30 g, 2.7 mmole) and salicylic
acid (1.88 g, 13.6 mmole) in dry methylene chloride (50 ml),
chilled in an ice-water bath under argon was added t-butyl

hydroperoxide (90%, 54.6 ml, 0.49 mole) in one portion. To

72
this mixture was added a solution of geranyl benzoate (32.85
g, 0.13 mole) in methylene chloride (50 ml) over 30 minutes.
The resulting colorless solution was stirred for 24 hours at
room temperature, then was diluted with benzene (70 ml) and
concentrated in 99999. Ether (150 ml) was added to the
residue; the organic phase was separated; washed with 10% KOH
(4x50 ml) and concentrated in 99999 to provide a yellow
liquid. The reaction mixture was dissoved in cold (ice-
water) acetic acid (30 ml) and dimethyl sulfide (33 ml) was
added slowly at 0°C and the resulting solution was stirred
for 5 hours at room temperature. The solution was then
chilled in an ice-water bath and neutralized with 20% aqueous
K2C03 and was then cast into ether (300 ml). The combined
organic phase was washed with water (300 ml), brine (300 ml):
dried (Nazso4); and concentrated in 99999 to provide a yellow
liquid. The crude product was dissoved in EtOH (100
m1), cooled in an ice-water bath and NaBH4 (5 g, 0.13 mole)
was added slowly. Stirring was continued for 10 minutes and
reaction was quenched by carefully addition of 1N aqueous HCl
(250 ml).The mixture was cast into water (300 ml), extracted
with Et 0 (3x250 ml), and the combined organic extracts were

2

washed with brine (800 ml), dried (Na $04) and concentrated

2
in 99999 to give a colorless liquid. The crude product was
purified by chromatography on a column of silica gel (230-400
mesh, 120 g, 50mm o.d., 4:1 hexane:ether) using flash
technique. Fractions 13 to 16 afforded 3.74 g, 11%, of

starting material, TLC (Merck, UV-vanillin, ether:hexane 1:1)

73
Rf=0.67 and fractions 18 to 29 afforded 22.34 g, 64% (71%
based on the recovery of starting material) of 8-hydroxy
geranyl benzoate 102, Rf=0.2, as a colorless liquid.

19_999 (250 MHz, c0013): 5- 8.05-8.00 (m, 2), 7.58-7.36 (m,
3), 5.45 (t, J=6.5 Hz, 1), 5.38 (t, J=6.5 Hz, 1), 4.82 (d,
J=6.5 Hz, 2), 3.95 (s, 2), 3.95 (s, 2), 2.22-2.04 (m, 5),
1.75 (s, 3), 1.64 (s, 3).

91:99 (7 eV): 275 (M++l, 6.7), 274 (9+, 0.5), 257 (35.5),
153 (32.8), 135 (base), 107 (17.7)

13 (Neat): 3430 (br.), 3030, 2930, 1715, 1600, 1585, 1450,

1385, 1315, 1275, 1105, 1070, 1025, 715 cm”1
Anal. Calcd. for C171122 3: C, 74.4: H, 8.08. Found: C,

74.41; H, 8.27.

(i)-6,7-Epoxy-8-hydroxygerany1 benzoate (1)-103

To a solution of 8-hydroxy geranyl benzoate 102 (9.65 g,
35.19 mmole) in dry methylene chloride (130 ml) chilled in an
ice-water bath was added a solution of m-chloroperoxybenzoic
acid (7.5 g, 85% purity, 36.95 mmole) in methylene chloride
(110 ml) over a period of 1 hour. Stirring was continued for
another 0.5 hour at 0°C and then 2 hours at room temperature.
The reaction mixture was then cast into ether (300 ml) and
saturated aqueous NaHCO3 (300 ml). The organic phase was
separated; washed with aqueous NaHCO3 (300 ml), water (300
ml), brine (300 ml): dried (MgSO4); and concentrated in 99999
to give 10.2 g, 100% crude yield, of 6,7-Epoxy-8-

hydroxygeranyl benzoate 103 as a colorless liquid which was

74
utilized without further purification.

199193 (250 MHz, c0c13): 5- 8.05-7.97 (m, 2), 7.54-7.36 (m,
3), 5.48 (t, J=7.3 Hz, 1), 4.81 (d, J=7.3 Hz, 2), 3.65-3.52
(m, 2), 2.99 (t, J=6.4 Hz, 1), 2.28-2.10 (m, 2), 1.75 (s, 3),
1.76-1.62 (m, 3), 1.25 (s, 3)

91:99: (7 eV) 291 (n+1, 0.4), 205 (11.9), 169 (46.5), 151
(48.1), 111 (base)
IR (neat): 3440 (br.), 3030, 2930, 1715, 1450, 1385, 1315,

1275, 1105, 1070, 1025, 715 cm'1

(-)-6,7-Epoxy-8-hydroxygeranyl benzoate (-)-103

To a solution of L-diisopropyltartrate (5.87 g, 25 mmole)
in dry methylene chloride (230 ml) cooled in a dry ice-CCl4
bath was added Ti(OiPr)4 (7.82 ml, 26.3 mmole). The solution
was allowed to stir for 5 minutes then 6.87 g (25 mmole) of
8-hydroxygeranyl benzoate in 40 ml of CH2C12 was added over a
period of 30 minutes followed immediately by the addition of
TBHP (3.5 M in toluene, 14.5 ml, 50 mmole). The resulting
yellow solution was stirred for 2 hours at -23°C and then was
stored in a freezer overnight. To the chilled (dry ice-CC14)
reaction mixture was added 4.5 ml of Mezs and the solution
was stirred for 40 minutes at -20°C; then the cold reaction
mixture was added slowly to a vigorously stirred saturated
aqueous NaF solution (550 ml) at room temperature. The
solution was stirred for 10 minutes: the aqueous layer was

saturated with NaCl: and the gel-like precitipated inorganic

fluorides were removed by filtration through a pad of celite.

75

The organic phase were separated: the aqueous layer was
extracted with CH2C12 (3x200 ml) and the combined organic
extracts were washed with water, brine (600 ml each): and
dried (Na2S04). The crude product was purified by
chromatography on a column of silica gel (60-240 mesh, 200 g,
1:1 hexane:ethyl acetate) using flash technique. Fractions
30 to 45 afforded 5.16 g, 71%, of (-)-103 as a viscous
colorless liquid.

[aun- -5.769 (c: 0.364, CH2C12)

(i)-6,7-Epoxy-8-benzyloxygerany1 benzoate (1)-104

To a solution of crude (i)-6,7-epoxy-8-hydroxygeranyl
benzoate 103 (10.2 g, 35.19 mmole) in dry THF (200 ml) was
added NaH (1.86 g, 38.66 mmole, 50% in oil, washed with dry
hexane (2K)) in one portion. The mixture was stirred for 20
minutes at room temperature under argon. To the resulting
dark brown solution was added benzyl bromide (4.6 ml, 38.66
mmole) in one portion, followed immediately by tetrabutyl
ammonium iodide (5.19 g, 14.05 mmole): the resulting mixture
was allowed to stir for 5 hours at room temperature. Solvent
was removed 19 99999 and the reaction was quenched by
carefully addition of saturated aqueous NH4C1 (300 ml). The
solution was cast into ether (200 ml); the organic phase was
separated and the aqueous layer was extracted with ether
(2X300 ml). The combined organic extracts were washed with
water, brine (500 ml each); and dried (Na2S04); and

concentrated 19 V9C99 to provide a yellow liquid. The crude

76
product was purified by chromatography on a column of silica
gel (230-400 mesh, 120 g, 50 mm o.d., 9:1 hexane:ethyl
acetate) using flash technique. Fractions 20 to 38 afforded
11.89 g, 89%, of 104 as a pale yellow liquid, TLC (Merck, UV-

vanillin, ether:hexane 1:1) R =0.41.

f

19_999 (250 MHz, 00013): 5- 8.01-7.93 (m, 2), 7.51-7.33 (m,
3), 7.31-7.22 (m, 5), 5.50 (t, J=8.0 Hz, 1), 4.82 (d, J=8.0
Hz, 2), 4.54 (d, J=12.2 Hz, 1), 4.48 (d, J=12.2 Hz, 1), 3.51
(d, J=10.3 Hz, 1), 3.44 (d, J-1o.3 Hz, 1), 2.88 ( t, J=6.2
Hz, 1), 2.35-2.17 (m, 2), 1.88 (s, 3), 1.88-1.66 (m, 2), 1.36

(S. 3)

Chiral 104

According to the procedure of preparation of (1)-104, (-
)-103 (4 g, 13.8 mmole) was benzylated by benzylbromide (1.8
ml, 15.1 mmole), NaH (0.7 g, 14.5 mmole, 50% in oil) and
Bu4NF (2 g, 5.4 mmole) in 200 ml of THF to provide 4.61 g,

88%, of chiral 104.

(i)—6,7-Epoxy-8-benzyloxygeranyl alcohol (1)-105

A mixture of 104 (7.51 g, 19.76 mmole), sodium methoxide
(32 g, 0.59 mole) and tetrabutyl ammonium iodide (8.76 g,
23.71 mmole) in MeOH (475 ml) was stirred for 12 hours at
room temperature. Then, MeOH was removed 19 99999 and water
(80 ml) was added. The solution was extracted with ether
(2x200 ml), water (400 ml): dried (Na2804); and concentrated

19 99999 to provide a yellow liquid. The crude product was

 

77
purified by chromatography on a column of silica gel (60-230
mesh, 100 g, 1:1 hexane:ether) using flash technique.
Fractions 16 to 32 afforded 5.4 g, 99%. of 105 as a pale
colorless liquid. TLC (Merck, UV-vanillin, ether:hexane 1:1)
Rf=0.09.

19_999 (250 MHz, coc13): 5- 7.34-7.20 (m, 5), 5.42 (t,
J=8.0 Hz, 1), 4.55 (d, J=12.1 Hz, 1), 4.49 (d, J=12.1 Hz, 2),
4.13 (d, J=8.0 Hz, 2), 3.47 (s, 2), 2.87 (t, J=6.5 Hz, 1),
2.27-2.16 (m, 2), 1.79-1.67 (m, 3), 1.72 (s, 3), 1.35 (s, 3)

91:99 (7 eV): 277 (M++1, 1.2), 259 (31.0), 151 (19.0), 111
(base), 91 (72.2)
13 (Neat): 3420 (br.), 3030, 2920, 2860, 1670, 1450, 1380,

1195, 1000, 740, 700 cm'1

(-)-6,7-Epoxy-8-benzyloxygeranyl alcohol (-)-105
According to the procedure of preparation of (1)-105,
the chiral 104 (2.45 g, 6.4 mmole) was treated with MeONa
(10.5 g, 0.19 mole) and Bu4NF (2.9 g, 7.8 mmole) in 120 ml of
MeOH to provide 1.76 g, 99% of (-)-105.
[“10“ -6.132 (c: 0.424, MeOH)
M+-OH cald. for C17H2302 259.1698, M+-OH found 259.1694
(i)-6,7-Epoxy-8-benzyloxygeranyl chloride (1)-106
To a solution of (1)-105 (2.1 g, 7.61 mmole) in dry ether
(18 ml) and HMPA (6 ml) chilled in an ice-water bath was
added n-BuLi (2.4 M in hexane, 3.2 ml, 7.61 mmole) over 10

minutes. The resulting mixture was stirred for 10 minutes at

78

0°C: then p-toluenesulfonyl chloride (1.45 g, 7.61 mmole) in
dry ether (5 ml) was added in one portion followed by
anhydrous lithium chloride (0.64 g, 15.2 mmole). The
resulting mixture was allowed to stirr for 1 hour at 0°C and
2 hours at room temperature: then was cast into saturated
aqueous NaHCO3 (200 ml) and ether (300 ml). The organic
phase was washed with water (2x200 ml), brine (200 ml): dried
(Na2804); and concentrated in 99999 to give a yellow liquid.
The crude product was purified by chromatography on a column
of silica gel (60-230 mesh, 50 g, 1:1 hexane:ether), TLC
(Merck, UV-vanillin, ether:hexane 1:1) Rf=0.73, to provide
1.9 g, 85%, of 106 as a colorless liquid.

l9__999 (250 MHz, c0013): 5- 7.42-7.24 (m, 5), 5.47 (t,
J=6.7 Hz, 1), 4.56 (d, J=12.2 Hz, 1), 4.50 (d, J=12.2 Hz, 1),
4.06 (d, J=7.9 Hz, 2), 3.50 (d, J=11.2 Hz, 1), 3.42 (d,
J=11.2 Hz, 1), 2.83 (t, J=6.4 Hz, 1), 2.28-2.09 (m, 2), 1.72
(s, 3), 1.73-1.61 (m, 2), 1.31 (s, 3)

91:99 (7 eV): 295 (9+, 0.7), 169 (11.4), 151 (17.9), 111
(base), 91 (65.2)

19 (Neat): 3020, 2920, 1665, 1455, 1385, 1255, 1095, 740,

700 cm-1

(+)-6,7-Epoxy-8-benzyloxygeranyl chloride (+)-106

According to the procedure for chlorination of (1)-105.
(-)-105 (1.22 g, 4.4 mmole) was treated with nBuLi (2 ml,
2.43 M in hexane, 4.86 mmole) and HMPA (3.5 ml) in 60 ml of

ether; followed by p-TsCl (1 g, 5.2 mmole) and LiCl (0.37 g,

79
8.8 mmole) to give 1.11 g, 85%, of (+)-106.

[a]D- +2.179 (c: 0.78, CHZClz)

(i)-1-(3-Fury1)-7,8-epoxy—9-benzyloxy non-3-ene (1)-107

In the usual procedurezz, 3-chloromethylfuran (0.83 g,
7.1 mmole) in 10 ml of THF was converted to the corresponding
Grignard reagent and coupled with (1)-6,7-epoxy-8-
benzyloxygeranyl chloride 106 (1.9 g, 6.5 mmole).
Purification by chromatgraphy on a column of silica gel (240-
400 mesh, 100 g, 9:1 hexane:ether) using flash technique
afforded 1.73 g, 79%, of (1)-107 as a colorless viscous

liquid, TLC (Merck, UV-vanillin, ether:hexane 1:4) R 80.34.

f
19_999 (250 MHz, c0013): 5- 7.41-7.24 (m, 6), 7.18 (m, 1),
6.25 (m, 1), 5.19 (t, J-7.5 Hz, 1), 4.57 (d, J-12.1 Hz, 1),
4.51 (d, J=12.l Hz, 1), 3.48 (d, J=ll.0 Hz, 1), 3.40 (d,
J=ll.0 Hz, 1), 2.82 (t, J=6.4 Hz, 1), 2.47-2.38 (m, 2), 2.29-
2.04 (m, 4), 1.69-1.56 (m, 2), 1.59 (s, 3), 1.32 (s, 3)
91:99 (70 eV): 340 (9+, 0.3), 173 (15.0), 91 (base), 81
(30.0)
13 (Neat): 3020, 2920, 1500, 1450, 1380, 1105, 1025, 875,

735, 700 cm‘1

(-)-1-(3-Fury1)-7,8-epoxy-9-benzyloxy non-3-ene (-)-107
According to the procedure for the preparation of (i)-

107, (+)-106 (0.42 g, 1.43 mmole) was coupled with 3-

furfurylMgCl (0.2 g, 1.7 mmole, of furfurylchloride and 0.08

g of Mg) and LiZCuCl4 in 20 ml of THF to give 0.364 g, 75%,

80
of (-)-107.
[an= -2.168 (c: 0.764, CH2c12)

9+ cald. for czszgo3 340.20386, 9+ found 340.2033
6-Benzyloxylethylene-4,5,5a,6,7,8,9,9a-octahydro-6,9a-
dinethyl- (Saar, 6a, 7,8, 9afl) - (3;) -naphtho[1 , 2-b] furan-7-ol
(i) ’108

To a solution of (1)-107 (0.11 g, 0.32 mmole) in dry
methylene chloride (3 ml), benzene (3 ml) and hexane (3 ml)
was added triethylamine (0.14 ml, 1 mmole). The solution was
cooled in a dry ice-acetone bath and a solution of boron
trifluoride etherate (0.24 ml, 1.95 mmole) in methylene
chloride (7 ml) was added very slowly over a period of 2
hours. The resulting yellow solution was stirred for 45
minutes at -78°C. The reaction mixture was cast into ether
(150 ml) and aqueous NaHCO3 (100 ml). The organic phase was
separated: washed with 0.1 N HCl (100 ml), brine (100 ml);
dried (Na2504): and concentrated 19 99999. The crude product
was purified by chromatography on a column of silica gel
(240-400 mesh, 50 9, 4:1 hexane:ether) using the flash
technique. Fractions 13 to 16 afforded 5.5 mg, 5%, TLC
(Merck, UV-vanillin, ether:hexane 1:1) Rf=0.45, of uncyclized
ketone 109 and 79 mg, 72%, of cyclized product (1)-108 as a

White solid, R =0.22. m.p.=92-93°C.

f
19_999 (250 MHz, coc13): 5a 7.26 (m, 5), 7.14 (d, J=2.0 Hz,
1), 6.08 (d, J=2.0 Hz, 1), 4.52 (s, 2), 3.74 (dd, J=5.0, 10.0

Hz, 1), 3.61 (d, J=8.8 Hz, 1), 3.30 (d, J=8.8 Hz, 1), 2.55-

81
2.13 (m, 2), 1.90-1.49 (m, 8), 1.27 (s, 3), 1.06 (s, 3).
51:95 (70 eV): 340 (M+, 18.0), 201 (21.1), 173 (24.9), 159
(22.7), 91 (base)
13 (Neat): 3440 (br.), 3030, 2930, 1505, 1455, 1375, 1075,

1025, 735, 695 cm’1

(+)-108
According to the procedure for the preparation of (i)-
108, (-)-107 (0.26 g, 0.76 mmole) was cyclized with BF3-OEt2

(0.57 ml, 4.56 mmole, in 17 ml CH2C12) and Et N (0.32 ml,

3

2.28 mmole) in hexane, benzene, CH2C12 (7 ml each) to afford
0036 g! 72%, Of (+)-1080 pomo=84-85.Co

[a]D= +46.813 (c: 0.455, MeOH)

Anal. Calcd. for C22H28 3: C, 77.61: H, 8.28. Found: C,

77.51: H, 8.31.

Benzyloxylethylene-4,5,5a,6,9,9a-hexahydro-6-6,9a-dimethyl-
(5aa,6a,9qfl)-(i)-naphtho[1,2-b]furan-7(8H)-one (i)-111

To a solution of oxalyl chloride (51 ul, 0.58 mmole) in
dry THF (2 ml) cooled in a dry ice-acetone bath was added
DMSO (43 ul, 0.6 mmole). The solution was warmed to -35°C
for 3 minutes. The resulting greenish yellow solution was
cooled in a dry ice-acetone bath and then a solution of ,fl-OH
benzylether 108 (0.19 g, 0.56 mmole) in THF (2 ml) was added
over 5 minutes. The solution was warmed to -35°C for 15
minutes, cooled back to -50°C and triethylamine (0.4 ml) was

added. The resulting cloudy solution was warmed to room

82
temperature and stirred for 1 hour. The reaction mixture was
diluted with ether (50 ml) and water (30 ml). Aqueous layer
was furthered extracted with ether (2x50 ml). The combined
organic extracts were washed with 0.1 N HCl (50 ml), brine
(50 ml); and dried (Nazso4). Removal of solvent followed by
chromatography on a column of silica gel (240-400 mesh, 30 g,
4:1 hexane:ether) using the flash technique afforded 0.183 g,
97%, of keto benzylether (1)-111 as a white solid, TLC

(Merck, UV-vanillin, ether:hexane 1:1) R =0.5.

f
l9__999 (250 MHz, c0c13): 5s 7.34-7.19 (m, 6), 6.14 (d,

J-1.7 Hz, 1), 4.50 (d, J-12.4 Hz, 1), 4.36 (d, J-12.4 Hz, 1),

3.48 (d, J=8.4 Hz, 1), 3.29 (d, J=8.4 Hz, 1), 2.68-2.22 (m,

6), 2.03-1.89 (m, 1), 1.66-1.53 (m, 2), 1.19 (s, 3), 0.98 (s,

3)

91:99 (70 eV): 338(M+, 4.0), 232 (8.6), 217 (10.5), 149

(8.9), 105 (8.5), 91 (base)

19 (Neat): 3030, 2950, 2850, 1700, 1500, 1450, 1375, 1230,

1105, 1030, 735, 700 cm'1

(+)-111
According to the procedure of preparation of (:)-111,
(+)-108 (70 mg, 20.6 mmole) was oxidized with oxalyl chloride

(0.05 ml), DMSO (0.05 ml) and Et N (0.4 ml) in 5 m1 of THF to

3
provide 66 mg, 95%, of (+)-111.

[d]D= +75.2 (c: 0.125, MeOH)

General Procedure for the Reduction by L-selectride and Salt:

83
Preparation of benzyloxynethylene—4,5,5a,6,7,8,9,9a-
octahydro-6-6 , 9a—dinethyl- (Sad, 60!, 7 a, 9ap) -(1-_) -naphtho[1 , 2-b]
furan-7-ol (1)-113
To a solution of keto benzyl ether (:)-111 (100 mg, 0.3
mmole) in dry methylene chloride (25 ml) chilled to -10°C

(ice-salt) was added MgBrz-Et 0 (153 mg, 0.6 mmole) in one

2
portion and stirred for 1 minute at -10°C. The result
suspension was then cooled in a dry ice-acetone bath and
added L-selectride (1 M in THF, 0.9 ml, 0.9 mmole) over a
period of 5 minutes. The reaction mixture was stirred for
0.5 hour at -78°C and warmed to room temperature slowly over
1 hour. Stirring was continued at room temperature for 20
minutes. The reaction was quenched by carefully adding
methanol (4 ml), followed by 20% aqueous NaOH (6 ml) and 30%
aqueous H202 (12 ml). The solution was stirred at room
temperature for 1.5 hours and then diluted with ether (100

ml), washed with water (100 ml), saturated aqueous NH Cl (100

4
ml) and saturated aqueous NaHCO3 (100 ml). The organic phase
was dried (Na2804) and removal of solvent 19 99999 followed
with purification by chromatography on a column of silica gel
(240-400 mesh, 50 g, 9:1 hexane:ether) using flash technique.
Fractions 30 to 48 afforded 85 mg, 85%, of desiredc»0H benzyl
ether (1)—113, TLC (Merck, UV-vanillin, ether:hexane 1:1)

R I=0.31 and fractions 52 to 56 afforded 10 mg, 10%, of fl-OH

f
benzyl ether (1)-108 which was resubmitted to Swern
oxidation, Rf=0.22.

199199 (250 MHz, cuc13): 5- 7.25 (m, 5), 7.14 (d, J=2.0 Hz,

84
1), 6.07 (d, J=2.0 Hz, 1), 4.57 (d, J=12.7 Hz, 1), 4.41 (d,
J=12.7 Hz, 1), 3.77 (m, 1), 3.54 (d, J=8.5 Hz, 1), 3.31 (d,
J=8.5 Hz, 1), 2.49-2.29 (m, 2), 2.01-1.51 (m, 8), 1.25 (s,
3), 0.87 (s, 3) .

91:99 (70 eV): 340 (9+, 1.4), 217 (17.0), 149 (36.3), 122
(13.8), 91 (base)

19 (Neat): 3480(br.), 3030, 2930, 1505, 1455, 1065, 730, 695

cm-1

(+)-113

According to the procedure of preparation of (1)-113,
(+)-111 (530 mg, 1.57 mmole) was reduced with L-selectride
(6.4 ml, 1 M in THF, 6.4 mmole) and MgBrz-OEt2 (810 mg, 3.13
mmole) in 130 ml of CH2C12 to give 496 mg, 93%, of a mixture
of (+)-108 and (+)-113 (8.5:1).

[a]D=+53.64 (c: 0.11, MeOH)

4,5,5a,6,7,8,9,9a-Octahydro-6-hydroxymethylene-6,9a-dinethy1—
(Sad, 6a, 73, 9ap) - (i) -naphtho[1 , 2-b] furan—7-ol (:L') -110
According to the general procedure outlined for the
preparation of (1)-85, (i)-108 was converted to (1)-110 in
50% yield with 42% recovered starting material after
purification by chromatography on silica gel.
199199 of 110 (60 MHz, 00013): 6= 7.12 (d, J=2.0 Hz, 1),
6.05 (d, J=2.0 Hz, 1), 3.90-3.20 (m, 3), 2.60-2.10 (m, 2),

2.00-1.40 (m, 9), 1.20 (s, 3), 0.95 (s, 3)

85

General Procedure for the Reductive Debenzylation:
Preparation of 4,5,5a,6,7,8,9,9a—octahydro-6-hydroxy
methylene-6,9a—dinethyl-(5aa,6d,7a;9ap)-(1)-naphtho[1,2-b]
furan—7-ol (i)-85

To a solution of lithium aluminum hydride (25 mg, 0.66
mmole) in dry THF (5 ml) was added a solution of benzyl ether
(1)-113 (15 mg, 0.04 mmole) in THF (5 ml) over 10 minutes and
stirred for 11 hours at room temperature. The reaction
mixture was carefully quenched with 20% aqueous NaOH (30 ml)
and cast into ether (50 ml). The organic phase was
separated: washed with saturated aqueous NH4C1 (50 ml), water
(50 m1), brine (50 m1); dried (Na2804) and concentrated in
99999 to provide a yellow solid. The crude product was
purified by chromatography on a column of silica gel (230-400
mesh, 8 g, 1:2 hexane:ether), TLC (Merck, UV-vanillin,
ether:hexane 5:1) Rf-0.18, to provide 10.4 mg, 94%, of (i)-85
as a white solid. m.p.=119-121°C.

19_999 (250 MHz, cnc13): 6= 7.13 (d, J=2.0 Hz, 1), 6.09 (d,
J=2.0 Hz, 1), 3.79 (t, J=3.0 Hz, 1), 3.59 (d, J=12.5 Hz, 1),
3.48 (d, J=12.5 Hz, 1), 2.54-2.27 (m, 2), 1.96-1.54 (m, 9),
1.26 (s, 3), 0.83 (s, 3).

91:99 (70 eV): 250 (9*, 31.6), 235 (21.47), 217 (base), 199
(19.3), 159 (22.3), 91 (36.4)
13 (neat): 3350, 2950, 1505, 1480, 1455, 1385, 1270, 1210,

1165, 1135, 1055, 1000, 890, 745 cm'1

(+)-85

86
According to the procedure of preparation of (1)-85,
(+)-113 (400 mg, 1.18 mmole) was debenzylated by LAH (0.45 g)
in 100 ml of THF to give 276 mg, 94%, of (+)-85. m.p.=138-
140°C

[a]D= +43.75 (c: 0.16, MeOH).

4a,4b,5,6,9b,10,11,11a-0ctahydro-2,2,4a,9b—tetranethy1—
(4aa, 4b,0, 9ba, 11aa)-4H-furo[2 ' , 3 ' : 5 , 6]naphtho[2 , 1-d] -1 , 3-
dioxin (1)-86

To) a solution of diol (1)-85 (250 mg, 1.0 mmole) in dry
methylene chloride (30 ml) and dry acetone (2 ml) was added

oxalic acid (few crystals) and anhydrous CaSO (1 g, 4.77

4
mmole). The mixture was stirred overnight at room
temperature. The solution was diluted with ether (30 ml),

washed with aqueous NaHCO (2x30 ml) and dried (Na2804).

3
Removal of solvent followed with purification by
chromatography on a column of silica gel (60-230 mesh, 8 g,
8:1 hexane:ether) using flash technique. Fractions 6 to 9
afforded 270 mg, 93%, of acetonide (:)-86 as a white solid,

TLC (Merck, UV-vanillin, ether:hexane 1:9) R =0.27. m.p.=137-

f
139'c.

l9_999 (250 MHz, d6-acetone): 5: 7.26 (d, J=2.0 Hz, 1),
6.15 (d, J=2.0 Hz, 1), 3.78 (t, J=3.0 Hz, 1), 3.70 (d, J=12.0
Hz, 1), 3.37 ( d, J=12.0 Hz, 1), 2.55-2.30 (m, 2), 2.11-1.50
(m, 7), 1.43 (s, 3), 1.28 (s, 3), 1.21 (s, 3), 0.82 (s, 3).

91:99 (70 eV): 290 (9*, 41.5), 275 (25.8), 232 (4.5), 217

87

(base), 149 (47.6)
13 (KBr): 3010, 2970, 2890, 1505, 1480, 1380, 1205, 1165,

1095, 1010, 860, 765 cm"1

(+)-86
According to the procedure of preparation of (1)-86,
(+)-85 (250 mg, 1 mmole) was treated with acetone (2 ml),

oxalic acid (few crystals) and CaSO (1 g) in 30 ml of CH2C12

4
to give 270 mg, 93%, of (+)-86.

[“30' +22.60 (c: 0.77, CH2C12)

6,7-Epoxy-8-acetyloxygerany1 benzoate 114

To a solution of (-)-103 (50 mg, 0.17 mmole) in a mixture
of dry methylene chloride (20 m1) and pyridine (0.5 ml, 6.2
mmole) chilled in an ice-water bath was added a solution of
acetic anhydride (20 mg, 0.19 mmole) in methylene chloride
(15 ml) slowly. The mixture was stirred overnight at room
temperature. The resulting solution was cast into water (40
ml) and ether (50 ml). The organic phase was separated;

washed with 5% HCl (3x40 ml), aqueous NaHCO (40 ml), water

3
(40 ml), brine (40 ml): dried (MgSO4) and concentrated 19
99999 to provide a yellow liquid. The crude product was
purified by chromatography on a column of silica gel (60-240
mesh, 20 g, 4:1 hexane:ether) using flash technique.

Fractions 6 to 8 afforded 114 ,53 mg, (93%). TLC (Merck, UV-

vanillin, ether:hexane 5:1) Rf=0.55.

I“

‘fim‘ L' .n ':

88

;n_993 (250 MHz, 00013): 5- 7.98-7.92 (m, 2), 7.50-7.30 (m,
3), 5.48 (t, J=7.4 Hz, 1), 4.82 (d, J=8.5 Hz, 2), 4.13 (d,
J=11.7 Hz, 1), 3.92 (d, J=11.7 Hz, 1), 2.89 (t, J=6.4 Hz, 1),
2.36-2.17 (m, 2), 2.10 (s, 3), 1.83-1.63 (m, 2), 1.82 (s, 3),

1.35 (s, 3).

HTPA ester 115

A solution of (-)-O¢-Methoxy-01-(trifluoromethyl)phenyl
acetic acid (60 mg, 0.26 mmole) and SOC12 (20 ml) was
refluxed for 45 hours. The excess SOCl2 was removed 19 99999
and diluted with dry methylene chloride (5 ml). To this crude
MTPACl solution was added a solution of (-)-103 (50 mg, 0.17
mmole) in methylene chloride (10 ml), followed by pyridine (5
ml) and DMAP (5 mg). The reaction mixture was refluxed for
1.5 hours and quenched with 1 N HCl (50 ml) and cast into
ether (50 ml). The organic phase was separated: washed with

saturated aqueous NaHCO (50 ml), water (50 ml), brine (50

3
ml): dried (MgSO4) and concentrated 19 99999 to provide a
yellow liquid. The crude product was purified by
chromatography on a column of silica gel (230-400 mesh, 10 9,
4:1 hexane:ether) using flash technique. Fractions 9 to 11
afforded 115, 18 mg, in 20% yield, TLC (Merck, UV-vanillin,
ether:hexane 1:1) Rf=0.35 and 20 mg of starting material was
recovered.

19_993 (250 MHz, c0c13): 5- 8.02-7.74 (m, 2), 7.56-7.35 (m,
3), 5.49 (t, J=7.4 Hz, 1), 4.84 (d, J=8.5 Hz, 2), 4.39 (d,

J=11.7 Hz, 1), 4.25 (d, J=11.7 Hz, 1), 3.59 (q, J=1.4 Hz, 3),

89
2.93 (t, J=6.4 Hz, 1), 2.36-2.10 (m, 2), 1.87-1.47 (m, 2),

1.82 (s, 3), 1.36 (s, 3)

Mbnobronobenzoate 116

According to the general procedure outlined for the
preparation of MTPA ester 115, (+)-110 was converted to 116
in 85% yield after purification by chromatography on silica

gel, TLC (Merck, UV-vanillin, ether:hexane 1:1) R =0.55.

f

199199 (250 MHz, c0c13): 5-I7.90-7.77 (m, 2), 7.67-7.44(m,
2), 7.13 (d, J-2.0 Hz, 1), 6.09 (d, J-2.0 Hz, 1), 4.40 (d,
J=10.0 Hz, 1), 4.25 (d, J-10.0 Hz, 1), 3.81 (m, 1), 2.63-2.30

(m, 2), 2.10-1.53 (m, 8), 1.31 (s, 3), 1.13 (s, 3)

General Procedure for the Preparation of (S)-O-
Hethylnandelate ester: Preparation of 119

A few crystals of DMAP was added all at once to a
colorless solution of (+)-108 (10 mg, 0.03 mmole), (S)-0-
methylmandelic acid (5 mg, 0.03 mmole) and 1,3 dicyclohexyl
carbodiimide (6 mg, 0.03 mmole) in 25 ml of methylene
chloride. After 24 hours, the reaction mixture was diluted
with ether (25 ml) and washed with 1 N aqueous HCl (20 ml),
saturated aqueous NaHCO3 (20 ml), brine (20 ml), dried
(MgSO4) and concentrated 19, 99999 to provide a viscous
liquid. The crude product was purified by chromatography on
a column of silica gel (230-400 mesh, 5 g, 4:1 hexane:ether)
to afford 12 mg, 83% (295% de), of 119.

¥9_999 of 118 (250 MHz, c0013): 5- 7.48-7.22 (m, 10), 7.16

90
(d, J=1.8 Hz, 1), 6.08 (d, J=1.8 Hz, 1), 5.09 (dd, J=5.0,
11.7 Hz, 1), 4.73 (s, 1), 4.10 (d, J=12.2 Hz, 1), 3.85 (d,
J=12.2 Hz, 1), 3.41 (s, 3), 2.86 (d, J=9.8 Hz, 1), 2.39-2.30
(m, 2), 2.15 (d, J-9.8 Hz, 1), 2.10 (dt, J-3.2, 12.5 Hz, 1),
1.97-1.40 (m, 6), 1.21 (s, 3), 0.88 (s, 3)

19_999 of 119 (250 MHz, 00013): 5- 7.48-7.26 (m, 10), 7.16
(d, J=1.8 Hz, 1), 6.08 (d, J=1.8 Hz, 1), 5.09 (dd, J=5.0,
11.7 Hz, 1), 4.73 (s, 1), 4.11 (d, J=12.2 Hz, 1), 3.85 (d,
J-12.2 Hz, 1), 3.41 (s, 3), 2.86 (d, J=9.8 Hz, 1), 2.40-2.30
(m, 2), 2.16 (d, J=9.8 Hz, 1), 2.11 (dt, 323.2, 12.5 Hz, 1),
2.01-1.39 (m, 6), 1.18 (s, 3), 0.62 (s, 3)

19_999 of 120 (250 MHz, 00013): 5= 7.49-7.20 (m, 10), 7.07
(m, 1), 6.16 (d, J=1.8 Hz, 1), 5.06 (t, J=1.5 Hz, 1), 4.69
(s, 1), 3.65 (s, 2), 3.34 (s, 3), 2.97 (d, J=7.9 Hz, 1), 2.68
(d, J=7.9 Hz, 1), 2.60-2.36 (m, 2), 2.08-1.57 (m, 6), 1.21
(s, 3), 1.00 (s, 3)

19_999 of 121 (250 MHz, c0013): 5- 7.49-7.20 (m, 10), 7.10
(d, J=1.8 Hz, 1), 6.14 (d, J=1.8 Hz, 1), 5.02 (t, J=1.5 Hz,
1), 4.58 (s, 1), 4.26 (d, J=12.2 Hz, 1), 4.19 (d, J=12.2 Hz,
1), 3.30 (s, 3), 3.26 (d, J=8.6 Hz, 1), 3.16 (d, J=8.6 Hz,
1), 2.60-2.36 (m, 2), 2.08-1.57 (m, 6), 1.15 (s, 3), 1.01 (s,

3)

10.

11.

REFERENCES

(a) Brundret, K. M.: Dalziel, W.; Hesp, 8.: Jarvis, J.
A. J.; Neidle, S. C
1972, 1027. (b) Dalziel, W., Hesp, B.; Stevenson, K.

M-: Jarvis, J- A. J. MWIJQQJ 1973.
2841.

Bucknall, R. A.: Moores, J.; Simms, R.; Hesp, B.
AnLimisrgh._Agean_§hemgsher. 1973. 4. 294.

Ikegami, 8.; Taguchi, T.: Ohashi, M.: 0guro, M.:
Nagano, H.: Mano, Y. 991919 (London) 1978, 279, 458.

Douros, J: Suffness, M. "New Anticancer Drugs";
Carter, 8. K.: Sakurai, Y.; Eds; Springer-Verlag:
Berlin, 1980, 29.

Pedrali-Noy, G.: Belvedere, M,: Crepaldi, T.: Focher,
F.: Spadari, S. 999991_3991 1982, 52, 3810.

Ohashi, M.: Taguchi, T.: Ikegami, 8. .91999999
Bigphx§._Be§._ngmnn. 1973, £2.1084-

Review: Huberman, J.A. 9911 1981,29, 647.

Review: Spadari, S: Sals, F.: Pedrali-Noy, G. 199999
Eigshem._§921 1982. 1. 29.

McMurry, J. E.: Webb, T. R. 1. 9e9, 9999, 1984, 91,
1367.

For Previous total and formal-total synthesis of
aphidicolin, see: (a) Trost, B. M.; Nishimura, Y.:
Yamamoto, K. 1. Am, 99999_§999 1979, 191, 1328. (b)
McMurry, J. E.: Andrus, A.: Ksander, G. M.; Musser,
G. H.; Johnson, M. A. g, 99, 99e9. 99c, 1979, 191,
1330 ; ibid, 19119999199 1981, 91, Suppl. 1, 319.
(c) Corey, E. J.: Tius, M. A.: Das, J. J, 99. C999.
5991 1980, 192, 1742. (d) Ireland, R. E.; Godfrey,

J. D.; Thaisrivongs, S. 9. 99. C99m, Soc. 1981,
191, 2446: Ireland, R. E.: Dow, W. C. : Godfrey, J.

d.: Thaisrivongs, S. 1, 919, 99999 1984, :9, 1001.
(e) van Tamelen, E. E.: Zawacky, S. R.; Russell, R.
K.: Carlson. J. G. 1._Am._9hsm.__§es. 1983. 125. 142:

van Tamelen, E. E.: Zawacky, S. R. 19919999999_99991
1985, 29, 2833. (f) Bettolo, R. A.: Tagliatesta P.:

Lupi, A.; Bravetti, D. 991v, 9919, 9ct9 1983, 99,
1922.

For approaches to 1 and related compounds, see:

(a) Cargell, R. L.: Bushey, D. F.: Dalton, J. R.:
Prasad, R. 8.; Dyer, R. D. 11_9199_99999 1981, 59,

91

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.
23.

24.

92

3389.

(b) Nicolaou, K. C.; zipkin, R. E. ew C e
Ed.__En911 1981: 29. 785-

(c) Kametani, T.: Honda, T.: Shiratori, y.;

MatsumotO. H-: FukumotO. K. l.__§hem.__§99.__zerkin
1 1981, 1386.

(d) Pearson, A. J.: Heywood, G. C.; Chandler, M. 99

thm.__§921__29231n__1 1982. 2631-
(e) Koyama, H.: Okawara, H.: Kobayashi, S.; 0hno, M.

Tesrahedr9n_Le§t. 1986. 25. 2685-

Trost, B. M.: Bogdanowicz, M. J. 1, 99, 9999, 999,
1973, 99, 5321.

stark. 6.: d'AngeIO. J. 1._Am.__§hem.__§99. 1974. 29.
7114.

Evans, D. A.: Truesadale, L. K.: Grimm, K. G.:
Nesbitt, S. L. 19_991_99999__9999 1977, 29, 5009.

(a) Parham, W. E.: Roosevelt, C. S. 19919999199_L9111
1971, 923.

(b) Evans, D. A.: Truesdale, L. K.: Carroll, G. L. 11

Qnem._§9§.._chem1_§9mmun1 1973. 55-
(c) Evans, D. A.: Truesdale, L. K.: Carroll, K. L. {9

W 197‘: 3.2! 914-

Ito, Y: Hirao, T.: Saegusa, T. 19_9199_99991 1978 51,
1011.

Kirk, D. N.; Petrow, V. 11_99999_§999 1962, 1091.
For some recent reports, see:

Tanis, S. P.: Herrinton, P. M. Q, 999, 9999. 1985,
99, 3988: Tanis, S. P.: Herrinton, P. M.: Dixon, L.

A. 19919999199_L9919, 1985, 29, 5347: and references
cited there in.

Tanis, S. P.: Herrinton, P. M. ;9_9;99_99999, 1983,
19, 4572.

Katsuki, T.: Sharpless, K. B. J. Am. Ch99, Soc.
1980, 192, 5974.

Tanis, S. P.: Chuang, Y. H.; Head, D. 8. 19999999199
LQEE. 1985. 29. 5147-

TaniS. 8- P- Tetrahedrgn_he§§. 1982. .2. 5509-

Stork, G. 9191_§991 1978, 91, 68.
Tamura, M.: Kochi, J. 599199919 1971, 303.

25.

26.

27.

28.

29.

30.

31.

32.
33.

34.

35.

36.

37.

38.

39.

40.

41.

93
For previous results about the approach to the
synthesis of 1, see:
Head, D. B. Ph.D. Dissertation, Part III, 1985.

Heathcock, C. H.;,Jennings, R. A.; von Geldern, T. W.

1L_QIQL_§DEEL 1933. AB. 3428-

Tanis, S. P.: Herrinton, P. M. g9_Q;g9_§99m9 1983,
59, 4572.

Mitchell, R. H.; Lai, Y. H.; Williams, R. V. J. 999.
Qhfimi 1979. 11. 4733-

Kuwajima. 1.; Urabe. H. Istrangdrgn_hgttl 1981. 22.
5191.

Williams, D. R.; Barner, B. A. I99;9999;99_L99§9
1983, 21, 427.

Williams, P. D.; LeGoff, E. I99;999Q;99_L9999 1985,
29, 1367.

Mashimo, K.;,Sato, Y. $9999999999, 1970, 29, 803.

Mitsunobu, 0. 929999919 1981, 1.

Brooks, D. W.; Grothaus, P. G.; Palmer, J. T.
1982, 2;, 4187; Raduchel, B.

Istrahsdr9n__Lettl
synthesis. 1980, 292.
Mancuso, A: Swern, D. 19_QIQL_§9999 1978, 9;, 2480.

Fuji, K.: Ichikawa, K; Node, M.; Fujita, E. . 0 .
$9999 1979, AA, 1661.

Brunet, J. J.; Mordenti, L.; Caubere, P. J. 0:9.
QBQEL 1973. $1. 4808.

Kutney, J. P.: Abdurahman, N.; Gletsos, C.;
LeQuesne, P.: Piers, E.; Vlattas, I. q9__999__g9999
$999 1970, 22, 1727.

For reviews of polyene cyclization, see: Johnson, W.
S. 9999__Q9999__3999 1968, l, 1: van Tamelen, E. E.
1919 1975, 9, 152; Johnson, W. S. ' o . e
1975. i. 513 Johnson, W. S. An§§!1__§h§m1__lnti_fifii
Engi 1975. 15. 9-

Dale, J. A.; Mosher, H. S. g, 99, Q9e9. §oc. 1973,
29, 512.

Dillon, J.; Nakanishi, K 19_Am9_§99m9__§999 1975, 91,
5409.

94
42. Trost, B. M.; Belletire, J. L.; Godleski, S.;

McDougal, P. G. Balkovec, J. M. g9_ggg9_g9999 1986,
91, 2370.

43. Watson, 8. L.: Eastham, J. F. . a t. C em.
1967, 9, 165.

44. Still, W. C.: Kahn, M.: Mitra, A. ;9_ng9_§9999 1978,
$1, 2923. '

“WI

966

1293 03046 3

3

IIHIIIWIW