Studies Toward the Total Synthesis of (-)-Zampanolide

Dawn Troast Advisor: John A. Porco, Jr.

1.0 Background and Significance: The unique 20-membered macrolide (-)-zampanolide (1, Figure 1) was isolated in 1996 by Tanaka and Higa from the sponge Fasciospongia rimosa, collected near Okinawa, Japan.1 This structurally interesting molecule has a high degree of unsaturation and an unusual N-acyl hemiaminal side chain.2 In addition to its unique structure, zampanolide displays potent cytotoxic activity (1-5 ng/mL) against P388, A549, HT29, and MEL28 tumor cell lines. Recently, Smith et. al. reported the total synthesis and tentative stereochemical assignment of the antipode (+)-zampanolide as 11R, 15R, 19R, and 20R.3 Our initial studies on zampanolide have focused on the synthesis of the unusual and unstable N-acyl hemiaminal side chain. Preliminary results in the total synthesis have focused on the synthesis of the 2,6-syn-disubstitued exo methylene pyran subunit via an intramolecular silyl-modified Sakurai cyclization (ISMS). Currently, work on C18-C19 bond construction using sp2-sp3 cross-coupling is being investigated. Figure 1: zampanolide

In addition to being a biologically interesting and challenging synthetic target, the macrolactone and the unsaturated side chain of zampanolide are reminiscent of the enamide natural products also being studied in our lab (oximidine II 24 and lobatamide C 3,5 Figure 2). Other natural products similar to zampanolide include the mycalamides (4),2d pederin (5),2e,f the theopederins (6),2g spergualin (7a),2b and 15-deoxyspergualin (7b).2c These compounds contain an N-acyl aminal moiety (4, 5, and 6) or an N-acyl hemiaminal (7a and 7b), which has been proven

Figure 2: Related Natural Products essential for their biological activity.6 1.1 Background of N-Acyl Hemiaminal Synthesis: The most common method used to install the N-acyl aminal in these molecules is reduction of an N-acyl imidate (inset, Figure 2), but this produces a mixture of epimers with moderate to poor selectivity.7 There are relatively few synthetic methods available for the preparation of N-acyl hemiaminals. Smith et. al. installed the N-acyl hemiaminal of (+)-zampanolide using a

O

O

HHMe

O

OHNH

OHO

Me

Me

20

1115

121

1

O O

OHN

O

O

Me

Me

Me

Me

OMe

OH

RO

4: myacalamides A (R = H), B (R = Me)

5: (+)-pederin

O

7a: R = OH: spergualin7b: R = H: (-)15-deoxyspergualin

NH

OO

OMeOMe

MeO

OH

OMe H

OH

Me

Me

MeO OH

O O

OHN

O

O

Me

Me

Me

Me

OMe

MeO OH

OOR

OH

6: theopederins K (R = Me), (L: R = H)

OH

O

O

Me

OH

Me

O

O

HN

ON

MeO

3: lobatamide C

OH

O

O OH

HN

ONOMe

H

2: oximidine II

HN NH

NH

HN

HN

NH2 R O OH

O

NH23 3

R1CO2H

N OO OMe

MeMe

OP

R1

O

N OMe

R2

NH

OO OMe

MeMe

OP

R1 NH

R2

O OMe1. SOCl2, Py, rt

2. HN=C(OMe)R2 Et3N

NaBH4, EtOH

2:7 ratio of epimers

Catecholborane

RhCl[PPh3]3, PhMe, -700C

10:1 ratio of epimers70% overall yield

Inset

- 1 -

stereospecific Curtius rearrangement as a key step.3 Direct condensation of amides and aldehydes has been reported, but is generally limited to very electron-poor aldehydes8,9 or unsubstituted amides10 and typically affords N,N′-alkylidene bisamides via acyl iminium intermediates.11 N-Acyl hemiaminals were also reported as undesired products in an attempted DABCO-mediated Baylis-Hillman reaction of acrylamide and protected amino aldehydes.12 Recently, reduction of an N-acyl imidate13 was used to prepare an N-acyl hemiaminal enroute to a glycosylcarbinolamide.14 1.2 Smith’s Synthesis of (+)-Zampanolide: The recent synthesis of the antipode (+)-zampanolide by Smith et. al.3 (Figure 3) was highlighted by methylenation of dioxanone 8 with Petasis-Tebbe reagent followed by a Petasis-Ferrier rearrangement to construct the cis-pyranone (9) followed Figure 3: Highlights in Smith’s synthesis of (+)-zampanolide

by ketone methylenation to afford the 2,6-syn-disubstituted exo methylene pyran 10. A higher order cuprate derived from vinyl bromide 11 was used to open epoxide 12 to furnish 13. A Curtius rearrangement of 14, followed by thermal rearrangement and trapping of the isocyanate, gave 15, which upon acylation with 16 afforded the PMB protected N-acyl hemiaminal (17). This occurs with complete transfer of the

O

O

MeBr OBPS

O

O

MeBr OBPS

O

H H H H1) Cp2TiMe2, THF, 65oC

19 h (72%, 6:1 at C15)

2) Me2AlCl, CH2Cl2 -78oC-rt separation (59%)

O

MeBr OBPSH H

CH2=PPh3 (98%)

O

MeBrH H

MeTBSO

OTBS 1) tBuLi2) (Th)CuCNLi

3)

DMBO

OPMB

OO

MeH H

MeTBSO

OTBSOH

DMBO

OPMB

(77%)

HO O

O

OPMB

NH

OOPMBO

O

TMS

NH

OOPMBO

Me

1) Hunig's base iBuOCOCl, 0oC2) NaN3, H2O, 0

oC

3) PhCH3, , 15 min. 4) TMSCH2CH2OH , 3h (75%)

1) NaHMDS, THF -78oC, 13 min.

2) -78 -0oC (58%)

Me

O

Cl

8 9

11 12 13

14 1516

17

20

20

1115

10

C20 stereochemistry, however the stereocenter was later epimerized in the final deprotection, illustrating the instability of this functionality as well as the advantage to installing it in the late stages of the synthesis. 2.0 Research Design and Methods 2.1 Retrosynthetic Analysis: In planning a retrosynthesis of zampanolide, a few key issues had to be kept in mind: the instability of the N-acyl hemiaminal, the acid labile exo cyclic methylene

- 2 -

Figure 4: Retrosynthetic Analysis pyran, and the base labile β,γ-unsaturated ketone. The retrosynthetic analysis of zampanolide (Figure 4) relies on a late stage oxidative decarboxylation-hydrolysis protocol for the construction of the sensitive N- acyl hemiaminal. (1) may thus be derived from intermediate 18,in which case the amino acid is

20

1115

1 21O

O

MeO

O

MeNH

O

Me

OHH

H H

20

1115

1 21O

O

MeO

OTBDPS

MeN

H

H H

O

Boc

Me

Me

O

Me

OH

Acetonide/Boc removalAmide Coupling

Oxidation DecarboxylationHydrolysisPG removalDess-Martin [O]

Stork-TakahashiCyanohydrin Alkylation

20

1115

OCHO

O

MeO

MeN

H

H H

O

Boc

Me

Me Cl20

1115

OH

OPMB

O

MeN

H

H H

O

Boc

Me

MeEsterification

MeO

OTBSOH

1115

I

O

MeH HOTBSN

HO

Boc

Me

Me

IWittig

ISMS cyclization

OPMB

OHTMS

I Me

CHOOPMBTMS

MgBr OCu(I) mediated

TMS protection

Zn insertion

ReductionProtection of alcohol

Ph3P CHONi cat. cross-coupling

OPMB

Ph NOtBu

O

Ph

O

MeH H

OPMB

H OOH

OPMB

O

MeH2N

H H

CO2tBu

Asymmetric

Aldol

Inset

epoxide opening

1 1819

20 21

22

24 23

26 25 27 28

29 30 31

PMB removalDess-Martin [O]

20

masked as a 2,2-dimethyloxazolidine, and (Z,E)-sorbic acid 19. Compound 18 may be derived from an intramolecular Stork-Takahashi alkylation15 of the trimethylsilyl cyanohydrin ketone which results from xxx of 20. Due to the instability of the β,γ-unsaturated ketone, reduction and protection as a silyl ether until the final steps of the synthesis may be required. Esterification of 21 with 22 followed by PMB removal, oxidation, and Wittig reaction gives enal 20. The key C9-C20 fragment (21) may be prepared by lithiation then lithium-zinc exchange of vinyl iodide 23 followed by a nickel catalyzed sp2-sp3 cross-coupling with serine-derived alkyl iodide (24).16 Compound (23) will be obtained from allyl silane 25 and iodoenal 26 employing an intramolecular silyl-modified Sakurai cyclization (ISMS) reaction17 to construct the 2,6-syn-disubstituted exo methylene pyran subunit. This route allows flexibility for fragment coupling, as well as for the oxidation state of the N-acyl hemiaminal precursor, without losing the highly convergent nature of the synthesis.

Another approach being considered for the construction of the C19-C20 bond, which is more closely related to a model study done on the side chain (cf. section 2.2), would involve an asymmetric aldol reaction between glycine equivalent 30 and aldehyde 31 (inset, Figure 4). Possible conditions for this reaction will be discussed in section 2.6. Synthetically, only protecting group strategy would change between the two approaches to obtain the C9-C20 fragment. The aldol apprach would result in the presence of a free NH and a t-butyl ester, which would have to both be considerations further in the synthesis. 2.2 Model Studies on the N-Acyl Hemiaminal Side Chain:18 Initial studies focused on the preparation of an N-acyl hemiaminal model system in order to evaluate methods that may be applicable to the synthesis of the natural product. Studies were initiated to determine if N,O-acetals derived from oxidative decarboxylation of N,O-acylamino acids,19 may be hydrolyzed under acidic conditions to afford N-acyl hemiaminals (Scheme 1). L-Threonine was chosen as a model β-hydroxy-α-amino acid, and 2,3-hexadienoic (sorbic) acid as a surrogate for the unsaturated lactone segment of the natural product. Commercially available L-threonine derivative 32 was acylated with sorbic acid to afford hydroxyamide 33 which was further

- 3 -

esterified with sorbic acid using Keck conditions20 to furnish 34. tert-Butyl ester removal was accomplished using TFA/Et3SiH to afford acid 35. Scheme 1: Synthesis of model N-acyl hemiaminals Since we were unable

to achieve direct conversion of the carboxylic acid to the N-acyl hemiaminal,21 we focused on development of conditions to produce the acetate cleanly. N-Acyl-α-amino acid 35

ClH3NOH

tBuO2C

Me

H

Yb(OTf)3

NH

OOHO

Me

O

H

MeH

aq. THFrt, 12 h(88%)

32 33 34

3536

37

NH

OHCO2

tBu

Me H

O

MeHOBT, EDCDIEA, DMF

DMAP-HCl, CH2Cl2

(96%)(97 %)

sorbic acidsorbic acid,DIC, DMAP

NH

OCOOH

Me H

O

MeMe

O

CH2Cl2

NH

OCO2

tBu

Me H

O

MeMe

OTFA, Et3SiH

Pb(OAc)4, Cu(OAc)2pyr., THF

(53%)

(76 %)

NH

OOAc

Me H

O

MeMe

O

Me NH

OOHO

Me

O

H

MeH

38

Me Inset

was subjected to oxidative decarboxylation to afford pure N,O-acetal 36 after extractive workup, with no evidence of N-acyl hemiaminal formation. A number of Lewis acid catalysts were then screened for the solvolysis of 36→37. After considerable experimentation, we found that Yb(OTf)3 (20 mol %, aq. THF), followed by purification through a neutral alumina cartridge,22 gave optimal results to afford N-acyl hemiaminal 37 (88 %). Other acid catalysts either led to no reaction (LiClO4), intolerably slow reactions (Mg(ClO4)2) or destruction of the compound (TMSOTf/CH2Cl2 or BF3-Et2O/aq. CH3CN23), the latter conditions affording considerable amounts of an aldehyde product by 1H NMR. The Z,E sorbamide-based N-acyl hemiaminal (38, inset, Scheme 1) was made through a series of analogous transformations, employing (2Z,4E)-sorbic acid24 in the initial acylation step. In the latter transformation, isomerization of the (Z,E)-diene was a significant concern, but fortunately the (Z)-olefin configuration was maintained throughout the synthesis without difficulty.

Substrates lacking either one or both of the sorbic side chains were also evaluated in the hydrolysis protocol, but these gave poor results. The surprising stability of model compounds such 37 and 38 in comparison to other N-acyl hemiaminals may be due to stabilization resulting from a hydrogen bonding network as well as the electron-withdrawing effects of the unsaturated ester in 37 and 38, which would stabilize the tetrahedral N-acyl hemiaminal and discourage formation of the transient iminium ion species. For model compound 37, the stabilization through a hydrogen bonding network has been shown by 1H NMR experiments. Evaluation of coupling constants, H-D exchange, and chemical shift of the amide proton all support the formation of a hydrogen bond network. These studies indicate that a hydrogen bond network could also stabilize and provide structural rigidity of the N-acyl hemiaminal side chain of zampanolide. Similar 1H NMR studies will be performed on the natural product to determine if a hydrogen bonding network exists and understand its importance in the potent cytotoxicity of zampanolide. 2.3 Synthesis of the C9-C17 Fragment: Preliminary studies have established methodology for preparation of the 2,6-syn-disubstituted exo-methylene pyran portion (23) of zampanolide using the ISMS reaction. Such reactions have been found to afford 2,6 syn-disubstituted exo methylene pyrans with high levels of stereocontrol.17 We have prepared the allyl silane via two different routes,

- 4 -

Scheme 3: Syntheses of allyl silane 25 as summarized in Scheme 3. Route 1 required the preparation of a β-keto ester that was difficult to make cleanly and therefore made the route unreliable. Route 2 involved an achiral

PMBOCHO

PMBO OMe

OO

TMS

Me

CeCl3, THF1) Me3SiCH2MgCl

1) Me2AlCl, CH2Cl2 (36%)

1) NaBH4, MeOH (38%)2) TMSCl, Et3N, THF (79%)

2) TMSNHCONHTMS, CH2Cl2 (91%)

TMSMgBr

OPMB

O CuI

PMBOTMS

PMBOTMS

OH

PMBO OMe

O

PMBOTMS

2) SiO2, CH2Cl2, (63%)3) TMSCl, Et3N, THF (75%)

THF

Route 1

Route 2

Route 3

TMSO TMSO

TMSO

27 28 25

ene reaction, which occurred with very low yields.25 This reaction initially offered an opportunity to develop an asymmetric ene reaction, but poor initial results and literature precedent26 led us to explore a third route to prepare 25. The current approach (route 3) begins with (+)-epoxide 28, which is prepared from an m-CPBA epoxidation followed by a hydrolytic kinetic resolution.27 Scheme 4: ISMS cyclization reaction Use of Grignard reagent 27 in a

copper(I)-mediated epoxide opening28 of 28 should give secondary alcohol 25. Preliminary results did provide some product in the opening of the racemic epoxide (% yield), but generation of the Grignard reagent on small scale was problematic. Due to the cost of the

O

MeI

OPMBH H

OPMBTMSI Me

CHO

OTMS TMSOTf, CH2Cl22,6-di-t-butyl-4-methylpyridine

(60%, unoptimized)

J (Hz)J15-16J14a-15J14b-115

zampanolide 7.6 11.3 2.7

X 7.611.22.8

Inset26 39 23

vinyl bromide precursor of 27 ($65.20/g) efforts to synthesize it from 2,3-dibromopropene on larger (10–20g) scale are underway.29 The ISMS reaction of iodoenal 2630 with 41 was accomplished using TMSOTf (10 mol %) and 2,6-di-t-butyl-4-methylpyridine at –78 ºC to afford pyran 23 as a single diastereomer (60%, unoptimized). The cis pyran stereochemistry of 23 was confirmed by evaluation of coupling constants (see Scheme 4, inset), which indicate that the C15 methine is in an axial position. Once conditions to produce allyl silane 41 on large scale are worked out, it can be used to optimize the ISMS reaction. 2.4 Synthesis of C9-C20 Fragment 21: Fragment 40 was derived from the serine-derived Garner aldehyde31 by stereoselective epoxidation using dimethylsulfonium methylide.32 The coupling of 40 with model compound 41 was studied extensively, focusing mainly on the use of Lipshutz’s higher order cuprate, (2-th)Cu(CN)Li (Table 1). There was some difficulty in generation of the reactive cuprate in this reaction due to its extreme air sensitivity.ref-smith In these reactions, mostly starting material was recovered, along with a complex mixture of products. There are examples where a Lewis acid is used to activate the epoxide,33 but even with a large excess of Lewis acid (3.0 equiv. BF3·Et2O) no product was formed. In some cases, it appeared that some epoxide had been opened, by a halogen to form a halohydrin side-product. Table 1: Cuprate Opening of an Epoxide

20

1115

OH

OPMB

O

MeN

H

H H

O

Boc

Me

Me

21

O

N OBoc

MeMe

I Me

OTBS

O

N OH

Boc

Me

MeMe

OTBS

1) conditions

40

41 42

2) additive3)

- 5 -

Entry Conditions Additive Results 1 (2-th)Cu(CN)Li, tBuLi ---- Starting material 2 (2-th)Cu(CN)Li, tBuLi BF3·Et2O Starting material and complex mixt. 3 CuCn, tBuLi ---- Starting material and protonated 41 4 tBuLi BF3·Et2O Starting material and protonated 41 5 iPrnBu2MgLi ---- Starting material 6 tBuLi, Me3Al BF3·Et2O Starting material and complex mixt. 7 (2-th)Cu(CN)Li, tBuLi (purchased) BF3·Et2O Starting material

Precedent for opening this serine-derived epoxide with an acetylide anion,34 led us to try this transformation, which was successful, but the subsequent hydrozirconation, tranmetallation to aluminum, and quenching with paraformaldehyde35 produced mostly starting material and the protonate alkene. The epoxide has a lot of steric bulk, which could be preventing attack on the epoxide, so ring opening of an epoxide without the acetonide (43, Figure 5) was also attempted, Figure 5: hoping that this would make the epoxide more accessible to the nucleophile.

Attempted copper(I)-mediated ring opening with isopropenyl magnesium bromide gave a small amount of product, but mostly starting material was recovered. Without any promising results for this type of epoxide opening, we have looked for alternate routes to make this fragment.

NH

PivO

OBoc

43

2.5 sp2-sp3 Coupling for C9-C20 Fragment 21: The next plan for the synthesis of fragment 21 involves metal catalyzed sp2-sp3 cross-coupling. The most common method for this type of coupling is palladium-mediated coupling using an alkyl boron and a vinyl halide.36 There are some examples where yields of sp2-sp3 coupling were improved using an alkyl zinc reagent due to its higher reactivity in the transmetallation step.37 Initially, zinc insertion into a model alkyl iodide containing a β-hydroxy group was attempted. The zinc inserted compound was prepared using zinc dust,38 zinc/copper couple,39 and lithium-zinc exchange.40 The zinc insertion on its own was a difficult task. The lithium-zinc exchange seemed to work best, but complete consumption of the starting material was not seen in any attempt. A concern with a substrate such as 24 was the β-elimination once the zinc inserted. Experiments on model compound 4441 Table 2: Zn insertion on model compound 44

Entry PG Zn source Results (44:45:46)

1 TIPS Zn0/TMSCl 12:trace:1 2 THP Zn0/TMSCl 1.3:0:1 3 TIPS Zn/Cu couple only 44 4 TIPS Li-Zn exchange 1:trace:1.2

showed that this would likely be a problem (Table 2). The hydroxyl protecting group was initially TIPS, hoping that the bulkiness of this group would prevent it from eliminating, but this did not show promise. A THP group was also employed thinking that the pyran oxygen could chelate to the zinc and stabilize the intermediate, but this

OPMBI

OPGZinc source

Me OPMB

OPG

OPMB

OPG

44 45 46

also produced a significant amount of eliminated product (and starting material). The use of an alkyl boron reagent was predicted to give similar difficulties and was not attempted. Considering these negative results, attention was turned to the cross coupling with a reversal in the reactivity. The vinyl iodide 43 was lithiated42 then a lithium-zinc exchange afforded the vinyl zinc, which

- 6 -

Scheme 5:

Entry Conditions 1 Ni(acac)2, THF:NMP (2:1), ligand 48 2 Ni(acac)2, THF:NMP (2:1), ligand 48 3 Pd(PPh3)4, DMF 4 CuCN ·2LiCl, THF

was directly added to a solution of alkyl iodide (44) and nickel(II) catalyst. In the absence of the styrene ligand 48, no product formed (Scheme 5). Knochel has reported that styrene ligand 48 can be used to facilitate the reductive elimination of the product (entry 1).19,43 Bu4NI has also been used as an additive in sp3-sp3 coupling of alkyl zinc reagents and alkyl halides. It is thought that the I- could coordinate to the alkyl zinc complex to form a more reactive zincate species.44 The palladium(0) system,

I Me

OTBS

1) tBuLi, Et2O, -78OC

2) ZnBr2, THF

3) conditionsO

N OTBS

IBoc

MeMe

F3C

O

N OTBS

Boc

Me

Me

Me

OTBS

4841

24

47

entry 3, has been used in the alkylation of dihydropranyl acetates with vinyl zinc compounds.45 The copper zinc reagent (entry 4) is another possible condition, whih has been used to couple alkyl zinc reagents with a variety of electrophiles.43-check These conditions will all be used on a model system with alkyl iodide 44 and vinyl iodide 41. The promising conditions will then be applied to the real system in order to optimize the conditions. 2.6 Asymmetric Aldol to Construct the C19-C20 Bond: An alternative route in the synthesis of fragment 29, would involve an asymmetric aldol reaction between glycine equivalent 30 and aldehyde 31. β-Hydroxy-α-amino acids are important building blocks in organic synthesis,46 but the asymmetric synthesis of these compounds is still limited.47 There are fewer examples of the use of aldol reactions to make β-hydroxy-α-amino acids.48 These methods tend to give moderate selectivity or depend greatly on the substrates. Glycine equivalent 30 is well known in literature, mainly involving alkylation reactions. Asymmetric alkylations with this substrate have shown very high selectivity in many cases,49c,49 but this selectivity has yet to be carried over to aldol reactions. Cinchonidine-derived catalyst have been employed for aldol reactions,50,49c(check) which give moderate selectivity with certain substrates, but it drops quickly with other substrates, limiting the utility of this system. Titanium enolates of N-alkylideneglycinates have been used in direct aldol reactions with simple aliphatic aldehydes to give anti-selectivity, but poor selectivity was seen with vinylic or aromatic aldehydes.51 A Ni(II) complex which contains homochiral Schiff base ligands has also been used in asymmetric aldol reactions to yield β-hydroxy-α-amino acids in high enantiomeric excess.52 Unfortunately, this procedure requires one equivalent of catalyst and two equivalents of the aldehyde, which makes this approach not as attractive as a catalytic reaction. Another approach uses acylation of t-butyl-N,N-dibenzylglycinate with an acid chloride followed by NaBH4 reduction to produce β-hydroxy-α-amino acids with high selectivity in two steps, but only moderate selectivity was reported for the direct aldol reaction using lithium enolates.53 From these examples, it is apparent that there is still a need for a general, highly selective method for the synthesis of β-hydroxy-α-amino acids using catalytic asymmetric methods.

Lanthanides have become very important reagents in organic synthesis and are especially effective in stereo-controlled reactions.54 There are many examples in literature for use of lanthanides in asymmetric aldol reaction,55 but none for the synthesis of β-hydroxy-α-amino acids. We would like to employ a lanthanide metal with a chiral ligand to catalyze this reaction. The first step in developing this reaction will be to screen ligands which have been reported in lanthanide catalyzed aldol reactions. Some of the more common chiral ligands used in reactions

- 7 -

with lanthanides, include the pybox class of ligands56 and chiral crown ethers.60a,57 As a model system, ethyl ester 49 will be used as the glycine equivalent and phenylacetaldehyde (50) will be used as a model β,γ-unsaturated aldehyde. Preliminary work on an asymmetric aldol will involve screening of chiral ligands with various lanthanides to look for reactivity. Once a few sets of good reaction conditions are found, a screen for enantiomeric excess will be done based on these results. Catalysts will be tested in an arrayed catalyst evaluation protocol58 performed in 96 well plates or in block synthesizers such as Radley’s Greenhouse (24 reactors).59 These are multi-variable reactions, therefore the outcome will depend on; lanthanide metal, chiral ligand, solvent, Figure 8: Screen for Asymmetric Aldol Reaction additives, and counter ion (Figure

8). Ideally all possible combinations will be tried, but initial tests for reactivity will only involve a lanthanide triflate and a chiral ligand in CH2Cl2. The progress of the reactions will be monitored by TLC and the enantiomeric excess will be determined using chiral HPLC. The second phase of screening will look at other variables such as solvent and counter ion effects to optimize the reaction conditions and examine the selectivity. The exact conditions used will probably be changed

OO

O

OOO

Ph

Ph

NOO

OON

Me

Me Me

Me

NN

OO

NR R

A BC: R = PhD: R = tBu

NOTMS

Ph

PhOEt

OHC

1) LnX3 chiral ligand

solvent, -78oC OH

CO2Et

H2N

2) H+49 50 51

Sc La Yb Zn Er HfLn(OTf)3

Chi

ral L

iga n

d AB

CD

Sc La Yb Zn Er Hf Sc La Yb Zn Er Hf

Solvent

LnCl3 Ln(OTf)3

CH2Cl2 EtOH/H2O

LnCl3

Sc La Yb Zn Er Hf

CH2Cl2EtOH/H2O

**

based on the results of the first screen. The lanthanide metals tested will be based on performance in the first screen and on availability (an x-ed out box indicates that the catalyst is unavailable). The Z and E-enolate can both be accessed for this gylcine equivalent,60 which will provide access to another variable to examine in the later stages of development. Using this method, we expect to uncover a reactive catalyst which may be used to prepare β-hydroxy-α-amino acids with high selectivity. It is expected that particular catalyst systems may be uncovered which afford either high syn or anti selectivity. Screening reactions conditions in parallel should lead to an effective catalyst-ligand system for asymmetric aldol reactions in a very quick and efficient manner. 2.7 Completion of the Synthesis: The initial approach to the completion of the macrolide involves esterification of 21 with E,E-dienoic acid 52 to afford 53 (Scheme 6). 52 will be derived from a Stille coupling between cis-2-methyl-3-trimethylstannayl-2-propen-1-ol61 and trans-3-iodo-acrylic acid ethyl ester.62 Removal of the PMB protecting group and subsequent oxidation followed by a wittig reaction will afford the precursor for the Stork-Takahashi cyanohydrin alkylation 20. Due to the instability of the β,γ-unsaturated ketone, reduction and protection of the hydroxyl group may be necessary for the remaining steps in the synthesis. Initial experiments will be done with the β,γ-unsaturated ketone, but we will be looking for migration of the double bond. A benefit of this approach is that if there is a problem with the intramolecular Stork-Takahashi cyanohydrin alkylation reaction, this can also be performed intermolecularly followed by a more common intramolecular esterification to close the macrolactone.

- 8 -

Scheme 6: Final Transformations in Synthesis of Zampanolide Final transformations to complete the total synthesis of zampanolide are outlined in Scheme 6. Concomitant hydrolysis of the acetonide and the Boc protecting groups of 18 will be effected by treatment with HCl/EtOAc or TFA/H2O according to well established protocols.63 One problem that could occur with this deprotection is translactonization. In an earlier model study, removal

O

Me

OH

O

N

Me

MeBoc OPMB

H HO

Me

O

O

N

Me

MeBoc OPMB

H H

O Me

OTBS

O

Me

O

O

N

Me

MeBoc

H H

OMe

ClCHO

O

Me

O

O

N

Me

MeBoc

H H

O Me

OTBDPS

O

Me

O

HO

H3N

H H

O Me

OTBDPS

O

Me

O

HO

NH

H H

O Me

OTBDPSO

Me

O

Me

ONH

H H

O MeO

Me

O

Me

OOH

NH

H H

O M

OO

Me

HO

O

MeOTBS

1) DDQ, CH2Cl2 (aq.)2) Dess-Martin, CH2Cl23) Ph3PCH2CHO, KHMDS, THF

4) TBAF, THF, 0oCl5) MsCl, 2,6-lutidine, LiCl DMF, 0oC

1) (TMS)CN, cat. KCN/18-c-6 complex2) LiHMDS, THF, -78oC3) AcOH, THF-H2O; 1% NaOH aq.

4) BH3-THF, 5% CBS reagent, THF, -10oC

5) TBDPSCl, imid, DMF, rt.

HCl, EtOAc

Me

O

OHEDC, HOBTDIEA, DMF

1) Pb(OAc)4, Cu(OAc)2 pyr., THF, 0oC-rt

1) TBAF, THF, 0oC

2) Yb(OTf)3, THF aq.

21

52

53 20

18 54

19

55 56 (±)12) MnO2, CH2Cl2

HO

EDC, HOBTDIEA, DMF

orTFA/H2O

O

X

20 20

15 11

e

of a trityl group with p-TsOH resulted in transesterification64 of the sorbic chain (Scheme 7). Hopefully, the constraints from the macrolide will prevent this, but it will be a concern. The resulting amino alcohol hydrochloride 54 will be acylated with Z,E-sorbic acid 19 to afford N- Scheme 7: Transesterification acylamino alcohol 54. It should be noted

that at this stage alternate acids may be coupled to prepare analogues or probe reagents derived from the zampanolide core. Removal of the C7 hydroxyl and oxidation

TrOO

Me

O Me

p-TsOHMeOH, 2h, rt HO

O

Me

O Me

Me O

O

Me

OH

3:1

of the allylic alcohol will yield the β,γ-unsaturated ketone 56. This represents a C20 methylene- inserted zampanolide analogue which will also be tested for biological activity. The identification of stable and bioactive analogues of 1 which lack the N-acyl hemiaminal would be extremely advantageous from the point of view of shelf-stability and further mechanism of action studies. Along these lines, methylene-inserted analogues of the α-hydroxyglycine of spergualin (cf. 7a, Figure 2) have been shown to retain the antitumor activity of the parent compound, demonstrating the possibility that the N-acyl hemiaminal of 1 could conceivably be replaced.65 Final oxidation of 56 with Pd(OAc)4 or the derived carboxylic acid (PDC/DMF) should form N,O-acetal intermediates which will be hydrolyzed with aqueous Yb(OTf)3 to form zampanolide 1 and its C20 epimer. It has been shown that these compounds can be separated to produce pure(-)-zampanolide 1.ref-smith Scheme 8: Final Transformations for Aldol Pathway The final transformation if the

aldol pathway is used are shown in Scheme 8. Although this pathway is more closely related to the model study (cf. section 2.2), it does not provide access to the methylene inserted compound 56. The aldol adduct 29, after protection of the amine, will undergo the same transformations as 21 (Scheme 6) to yield the macrolactone 58. Removal of the amine protecting group will afford

O

Me

OHCO2

tBu

NH OPMB

H H

O

Me

O

tBuO2C

NH

H H

O Me

OTBDPSO

Me

O

Me

ONH

H H

O MeO

Me

O

Me

OOH

NH

H H

O Me

OO

Me

Me

O

OHEDC, HOBTDIEA, DMF

1) Pb(OAc)4, Cu(OAc)2 pyr., THF, 0oC-rt

1) TBAF, THF, 0oC

2) Yb(OTf)3, THF aq.

57

19

60 61

(±)1

2) MnO2, CH2Cl2 or Dess-Martin, CH2Cl23) HF/CH3CN

tBuO2C

O

Alloc

O

Me

O

tBuO2C

NH

H H

O Me

OTBDPSAlloc

1) Pd(OAc)2, Et3N, Et3SiH

2) H3O+ O

Me

O

tBuO2C

Cl-H3N

H H

O Me

OTBDPS

58 59

20

15 11

- 9 -

59, which will be acylated with 19 to give 54. TBAF deprotection, oxidation of the allylic alcohol (again, only if necessary to reduce the β,γ-unsaturated ketone), and removal of the tert-butyl group will afford N-acyl-α-amino acid 61. Finally, oxidative decarboxylation followed by hydrolysis will yield (±)-zampanolide 1. 2.8 Pyran Ring as a Scaffold for Diversity Oriented Synthesis: Diversity oriented synthesis66 is a new and useful strategy to synthesize large numbers of complex molecules based on natural products. Natural product-like compounds can be used as scaffolds to create new, highly functionalized structures. The ISMS reaction provides an efficient, stereoselective route to 2,6-syn-disubstituted exo methylene pyran rings. Using an allyl silane such as 39 and aromatic aldehydes such as p-bromobenzaldehyde 62, this reaction can provide a template such as 63 for metal- catalyzed, cross-coupling reactions (Scheme 7). The right side chain alcohol could be used to attach a solid support, or it can be oxidized to an aldehyde and further functionalized. Scheme 7: Pyran Ring as a Scaffold The aromatic portion can be functionalized

using Stille,67 Suzuki,36 or Sonogarshira68 couplings. Finally, the exo methylene provides a position for intermolecular Pauson-Khand reactions.69 The stereoselectivityat the emerging spiro stereocenter will also be investigated and determined by the appropriate NMR techniques such as NOE.70 These reactions will provide densely functionalized spiro compounds with two functionalized arms

O

BrH H

TMSOPMB

OTMS

CHO

BrTMSOTf, CH2Cl22,6-di-t-butyl-4-methylpyridine

OPMB

O

BrH H

OPMB

R1Co2(CO)8

O R1

*

O

R2H H

OPMB

O R1

*

Suzuki, Stille, Sonogarshira

R2 = aryl, alkenyl, alkynyl

39

62

63

6465

on the pyran ring. This could create an interesting library of complex compounds that could be screened for biological activity. 3.0 Summary: This proposal discusses the planned total synthesis of antitumor macrolide (-)-zampanolide. This synthesis will include a nickel(II) catalyzed sp2-sp3 coupling reaction to make fragment X, an ISMS cyclization reaction to create the pyran ring moiety Y, and a intramolecular Stork-Takahashi cyanohydrin alkylation to close the macrolactone. A model study on the side chain of zampanolide has established an oxidative-decarboxylation/hydrolysis protocol to install the N-acyl hemiaminal. This study also showed evidence for a hydrogen-bonding network in the side chain model, which could have some effect on the biological activity of zampanolide. Plans to create a library of highly functionalized spiro-ketal compounds base on the pyran moiety of zampanolide has also been discussed as an extension this project. …….

- 10 -

References Cited: (1) (a) Tanaka, J-i; Higa, T. Tetrahedron Lett. 1996, 37, 5535-5538. (b) Higa, T.; Tanaka, J-i; Garcia Gravalos., D. PCT Int. Appl. 1997, 25 pp. WO 9710242 A1. For a related natural product (dactylolide) lacking the N-acyl hemiaminal side chain, see: Cutignano, A.; Bruno, I.; Bifulco, G.; Casapullo, A.; Debitus, C.; Gomez-Paloma, L.; Riccio, R. Eur. J. Org. Chem. 2001, 775-778. (2) For examples of N-acyl hemiaminal-containing natural products, see: (a) echinocandin B: Benz, F.; Knuesel, F.; Nuesch, J.; Treichler, H.; Voser, W.; Nyfeler, R.; Keller-Schierlein, W. Helv. Chim. Acta 1974, 57, 2459-2477. spergualin: (b) Umezawa, H.; Kondo, S.; Iinuma, H.; Kunimoto, S.; Ikeda, Y.; Iwasawa, H.; Ikeda, D.; Takeuchi, T. J. Antibiot. 1981, 34, 1622-1624. 15-deoxyspergualin: (c) Groth, C. G. Ann. N.Y. Acad. Sci. 1993, 685, 193-xxx. For N-acyl aminal natural products, see: mycalamides: (d) Perry, N. B.; Blunt, J. W.; Munro, M. H. G.; Pannell,

- 11 -

L.K. J. Am. Chem. Soc. 1988, 110, 4850-4851. pederin: (e) Cardani, C.; Ghiringhelli, D.; Mondelli, R.; Quilico, A. Tetrahedron Lett. 1965, 57, 2537. (f) Matsumoto, T.; Yanagiya, M.; Maeno, S.; Yasuda, S. Tetrahedron Lett. 1968, 60, 6297-6300. theopederins: (g) Fusetani, N.; Sugawara, T.; Matsunaga, S. J. Org. Chem. 1992, 57, 3828-3838. tallysomycins (glycosylcarbinolamide): (h) Konishi, M.; Saito, K.; Numata, K.; Tsuno, T.; Asama, K.; Tsukiura, H. Naito, T.; Kawaguchi, H. J. Antibiot. 1977, 30, 789-805. (3) Smith, A. B., III; Safonov, I. G.; Corbett, R. M. J. Am. Chem. Soc. 2001, 123, 12426-12427. (4)Kim, J. W.; Shin-ya, K.; Furihata, K.; Hayakawa, Y.; Seto, H. J. Org. Chem. 1999, 64, 153-155. (5) McKee, T. C.; Galinis, D. L.; Pannell, L. K.; Caredellina, J. H., II; Laasko, J.; Ireland, C. M.; Murray, L.; Capon, R. J.; Boyd, M. R. J. Org. Chem. 1998, 63, 7805-7810. (6) Abell, A. D.; Blunt, J. W.; Foulds, G. J.; Munro, M. H. G. J. Chem. Soc., Perkin Tans. 1 1997, 11, 16471654. (7) (a) Matsudea, F.; Tomiyoshi, N.; Yanagiya, M.; Matsumoto, T. Tetrahedron 1988, 44, 7063-7080. (b) Kocrenski, P.; Jarowicki, K.; Marczak, S. Synthesis 1991, 1191-1200. (8) Glyoxylic acid: Schouteeten, A.; Christidis, Y.; Mattioda, G. Bull. Soc. Chim. Fr. 1978, (5-6 Pt. 2), 248-254. (9) Perhaloaldehydes; Ingrassia, L.; Mulliez, M. Synthesis 1999, 1731-1738. (10) Johnson, A. P.; Luke, R. W. A.; Steele, R. W.; Boa, A. N. J. Chem. Soc., Perkin Tans. 1 1996, 883-893. (11) (a) Fernandez, A. H.; Alvarez, R. M.; Abajo, T. M. Synthesis 1996, 1299-1301. (b) For a recent example, see: Labrecque, D.; Charron, S.; Rej, R.; Blais, C.; Lamothe, S. Tetrahedron Lett. 2001, 42, 2645-2648. (12) Bussolari, J. C.; Beers, K.; Lalan, P.; Murray, W. V.; Gauthier, K.; McDonnell, P. Chem. Lett. 1998, 787-788. (13) Matsuda, F.; Tomiyoshi, N.; Yanagiya, M.; Matsumoto, T. Tetrahedron 1988, 44, 7063-7080. (14) Sznaidman, M. L.; Hecht, S. M. Org. Lett. 2001, 3, 2811-2814. 15 Takahashi, T.; Nemoto, H.; Tsuji, J. Tetrahedron Lett. 1983, 24, 3485. (16) Giovannini, R.; Knochel, P. J. Am. Chem. Soc. 1998, 120, 11186-11187. (17)(a) Marko, I. E.; Bayston, D. J. Tetrahedron Lett. 1993, 34, 6595-6598. (b) Sung, M.; Kwak, W. Y.; Kang, K. T. Bull. Korean Chem. Soc. 1998, 19, 862-868. (c) Leroy, B.; Marko, I. E. Tetrahedron Lett. 2001, 42, 8685-8688 (18) Troast, D. M.; Porco, J. A., Jr. Org. Lett. 2002, 4, in press. (see appendix) (19) Wang, X.; Porco, J. A., Jr. J. Org. Chem. 2001, 66, 8215-8221. (20) Boden, E. P.; Keck, G. E. J. Org. Chem. 1985, 50, 2394-2395. (21) Oxidative decarboxylation with Pb(OAc)4, Cu(OAc)2 with iPr2EtN in THF afforded a mixture of N-acyl hemiaminal and acetate products, in low (10-20 % ) yields. (22) A Waters Sep-Pak® neutral alumina cartridge (12 cc, 2 g) was utilized. Attempted purification of N-acyl hemiaminal products such as 7 using silica gel chromatography led to low recoveries of product. (23) Askin, D.; Angst, C.; Danishefsky , S. J. Org . Chem. 1987, 52 , 62 -35 . (24) Prepared by Stille coupling of tributyl-(1E)-1-propenyl-stannane with (Z)-3-iodoacrylic acid cf. Abarbri, M.; Parain, J.-L.; Cintrat, J.-C.; Duchene, A. Synthesis 1996, 82-86. (25) Snider, B. B. Acc. Chem. Res. 1980, 13, 426-432. (26) Ref for ene rxn w/ electron poor aldehydes (27) Schaus, S. E.; Brandes, B. D.; Larrow, J. F.; Tokunaga, M.; Hansen, K. B.; Gould, A. E.; Furrow, M. E.; Jacobsen, E. N. J. Am. Chem. Soc. 2002, 124, 1307-1315. (28) Nishiyama, H.; Yokayama, H.; Narimatsu, S.; Itoh, K. Tetrahedron Lett. 1982, 23, 1267-1270. (29) Smith, J. G.; Drozda, S. E.; Petraglia, S. P.; Quinn, N. R.; Rice, E. M.; Taylor, B. S.; Viswanathan, M. J. Org. Chem. 1984, 49, 4112-4120. (30) Takeuchi, R.; Tanabe, K.; Tanaka, S. J. Org. Chem. 2000, 65, 1558-1561. (31) Campbell, A. D.; Raynham, T. M.; Taylor, R. J. K. Synthesis, 1998, 1707-1709. (32) Moore, W. J.; Luzzio, F. A. Tetrahedron Lett. 1995, 36, 6599-6602. and references within. (33) Alexakis, A.; Hanafzi, J.; Jachiet, D.; Normant, J. F. Tetrahedron Lett. 1990, 31, 1271-1274. (34) Koviach, J. L.; Chappell, M. D.; Halcomb, R. L. J. Org. Chem, 2001, 66, 2318-2326. (35) Okukado, N.; Negishi, E-i. Tetrahderon Lett. 1978, 27, 2357-2360. (36) Suzuki, A. Acc. Chem. Res. 1982, 15, 178-184. (37) Negishi, E-i.; Takahashi, T.; Baba, S.; Van Horn, D. E.; Okukado, N. J. Am. Chem. Soc. 1987, 109, 2393-2401. (38) Dudda, R.; Eckardt, M.; Furlong, M.; Knoess. H. P.; Berger, S.; Knochel, P. Tetrahedron Lett. 1994, 35, 2415-2432. (39) Jackson, R. F. W.; Wishart, N.; Wook, A.; James, K.; Wythes, M. J. J. Org. Chem. 1992, 57, 3397-3404. (40) Knochel, P.; Singer, R. D. Chem. Rev. 1993, 93, 2117-2188. (41) Jackson ref on elimination (42) Earlier experiments showed that this compound could be lithiated cleanly, tBuLi (2 equiv.) in Et2O.

- 12 -

(43) Giovannini, R.; Studemann, T.; Dussin, G.; Knochel, P. Angew. Chem. Int. Ed. Engl. 1998, 37, 2387-2390. (44) (a) Jensen, A. E.; Knochel, P. J. Org. Chem. 2002, 67, 79-85. (b) Piber, M.; Jensen, A. E.; Rottlander, M.; Knochel, P. Org. Lett. 1999, 1, 1323-1326. (45) Dunkerton, L. V.; Euske, J. M.; Serino, A. J. Carbohydr. Res. 1987, 171, 89-107. (46) For a few examples of natural products contains b-hydroxy-a-amino acids, see : (a) echinocandins: Benz, v. F.; Knusel, F.; Nuesch, J.; Treichler, H.; Voser, W.; Nyfelder, R.; Keller-Schierlein, W. Helv. Chim. Acta. 1974, 57, 2459-2477. (b) vancomycin: Williams, D. H. Acc. Chem. Res. 1984, 17, 364-369. (c) cyclosporin: Aebi, J. D.; Deyo, D. T.; Sun, C. Q.; Guillaume, D.; Dunlap, B.; Rich, D. H. J. Med. Chem. 1990, 33, 999-1009. (d) bouvardin: Jolad, S. D.; Hoffmann, J. J.; Torance, S. J. Wiedhopf, R. M. Cole, J. R.; Arora, S. K.; Bates, R. B.; Garguilo, R. L. Kriek, G. R. J. Am. Chem. Soc. 1977, 99, 8040-8044. (47) (a) Zandbergen, P.: Brussee, J.; van der Gen, A. Tetrahderon: Asymmetry 1992, 3, 769-774. (b) Corey, E. J. Tetrahedron Lett. 1991, 2857. (c) Kurokawa, N. Ofune, Y. E. J. Am. Chem. Soc. 1986, 108, 6041-6043. (d) Masse, C. E. Morgan, A. J. Panek, J. S. Org. Lett. 1999, 2, 2571-2573. (e) Han, B. h.; Kim, Y. C.; Park, M. K.; Park, J. H.; Go, H. J.; Yang, H. O.; Suh, D. Y.; Kang, Y. H. Heterocycles, 1995, 41, 1909-1914. (f) Moore, W. J.; Luzzio, F. A. Tetrahedron Lett. 1995, 36, 6599-6602. (48) (a) Van der Werf, A. W.; Kellogg, R. M. J. Chem. Soc., Chem. Commun. 1991, 10, 682-683. (b) Evans, D. A.; Janey, J. M.; Magomedov, N.; Tedrow, J. S. Angew. Chem. Int. Ed. 2001, 40, 1884-1888. (c) Corey, E. J.; Xu, F.; Noe, M. C. J. Am. Chem. Soc. 1997, 119, 12414-12415. (49) Gmouh, S. Jamal-Eddine, J.; Valnot, J. Y. Tetrahedron 2000, 56, 8361-8366. c) Lygo, B.; Crosby, J.; Lowdon, T. R.; Wainwright, P. G. Tetrahedron 2001, 57, 2391-2402. D) Ooi, T.; Kameda, M.; Maruoka, K. J. Am. Chem. Soc. 1999, 121, 6519-6520. (50) (a) Horikawa, M.; Busch-Petersen, J.; Corey, E. J. Tetrahedron Lett. 1999, 40, 3843-3846. b) Gasparski, C. M.; Miller, M. J. Tetrahedron 1991, 47, 5367-5378. (51) Kanemasa, S.; Mori, T.; Wada, E.; Tatsukawa, A. Tetrahedron Lett. 1993, 34, 677-680. (52) Belokon, Y. N.; Kochetkov, K. A.; Ikonnikov, N. S.; Strelkova, T. V.; Harutyunyan, S. R.; Saghiyan, A. S. Tetrahedron: Asymmetry 2001, 12, 481-485. (53) Guanti, G.; Banfi, L.; Narisano, E.; Scolastico, C. Tetrahedron Lett. 1984, 25, 4693-4696. (54) Lanthanides: Chemistry and Use in Organic Synthesis; Kobayashi, S., Ed.; Topics in Orgnometallic Chemistry; Springer: Heiderberg, 1999. (55) (a) Yoshikawa, N.; Yamada, Y. M. A.; Das, J.; Sasai, H.; Shibasaki, M. J. Am. Chem. Soc. 1999, 121, 4168-4178. (b) Kobayashi, S.; Hamada, T.; Nagayama, S.; Manabe, K. Org. Lett. 2001, 3, 165-167. (c) Gong, L.; Streitwieser, A. J. Org. Chem. 1990, 55, 6235-6236. (d) Vougioukas, A. E.; Kagan, H. B. Tetrahedron Lett. 1987, 28, 5513-5516. (e) Kobayashi, S. Chem. Lett. 1991, x, 2187-2190. (f) Kobayashi, S. Synlett 1994, 689-701. (56) (a) Evans, D. A.; Sweeney, Z. K.; Rovis, T.; Tedrow, J. S. J. Am. Chem. Soc. 2001, 123, 12095-12096. (b) Fukazawa, S.-i.; Matsuzawa, H.; Metoki, K. Synlett 2001, 5, 701-711. (c) Qian, C.; Wang, L. Tetrahedron Lett. 2000, 41, 2203-2206. (d) Schaus, S. E.; Jacobsen, E. N. Org. Lett. 2000, 2, 1001-1004. (e) Evans, D. A.; Fitch, D. M. Ang. Chem. Int. Ed. Engl. 2000, 39, 2536-2540. (f) Aspinall, H. C.; Greeves, N.; Smith, P. M. Tetrahedron Lett. 1999, 40, 1763-1766. (57) Nagayama, S.; Kobayashi, S. J. Am. Chem. Soc. 2000, 122, 11531-11532. (58) Taylor, S. J.; Morken, J. P. J. Am. Chem. Soc. 1999, 121, 12202-12203. (59) greenhouse (60) (a) Guanti, G.; Banfi, L.; Narisano, E.; Scolastico, C. Tetrahedron 1988, 44, 3671-3684. (b) Georg, G. I.; Akgun, E. Tetrahedron Lett. 1991, 32, 5521-5524. (61) syn of vinyl stannane (62) Takeuchi, R.; Tanabe, K.; Tanaka, S. J. Org. Chem. 2000, 65, 1558-1561.(same as 30) (63) For an example, see: Roush, W. R.; Hunt, J. A. J. Org. Chem. 1995, 60,798-806. (64) Coery, E. J.; Brunelle, D. J.; Nicolaou, K. C. J. Am. Chem. Soc. 1977, 99, 7359-7360. (65) Nishizwa, R.; Takei, Y.; Yoshida, M.; Tomiyoshi, T.; Saino, T.; Nishikawa, K.; Nemoto, K.; Takahashi, K.; Fujii, A.; Nakamura, T.; Takita, T.; Takeuchi, T. J. Antibiot. 1988, 41, 1629-1643. (66) Schreiber, S. L. Science, 2000, 287, 1964-1969. (67) Stille, J. K. Angew. Chem. Int. Ed. Engl. 1986, 25, 508-xxx. (68) Sonogashira, K. Tohda, Y. Hagihara, N. Tetrahedron Lett. 1975, x, 4467-xxx. (69) Ishizaki, M.; Kasama, Y.; Zyo, M.; Niimi, Y.; Hoshino, O. Heterocycles, 2001, 55, 1439-1442. (70) Use of NOE to determine stereochemistry of the pauson khand product: