Studies of stereocontrolled allylation reactions for the tot

Edited by Lynn Smith-Lovin, Duke University, Durham, NC, and accepted by the Editorial Board April 16, 2014 (received for review July 31, 2013) ArticleFigures SIInfo for instance, on fairness, justice, or welfare. Instead, nonreflective and Contributed by Ira Herskowitz ArticleFigures SIInfo overexpression of ASH1 inhibits mating type switching in mothers (3, 4). Ash1p has 588 amino acid residues and is predicted to contain a zinc-binding domain related to those of the GATA fa

Edited by Kyriacos C. Nicolaou, The Scripps Research Institute, La Jolla, CA (received for review April 9, 2004)

Article Figures & SI Info & Metrics PDF


A highly convergent total synthesis of the potent anticancer agent (+)-phorboxazole A (1) is accomplished. Four components (3–6) are assembled with considerations for control of absolute and relative stereochemistry. Iterative asymmetric allylation methoExecutelogy addresses key stereochemical features in the preparation of the 2,6-cis- and 2,6-trans-tetrahydropyranyl rings of the C3–C19 component (3). The stereocontrolled asymmetric allylation process is also used for development of the C28–C41 fragment (4). Modern Barbier coupling reactions of α-ioExecutemethyl oxazoles and related thiazoles are Characterized with samarium iodide. The convergent assembly of components 4 and 5 features formation of the fully substituted C22–C26 pyran by intramolecular capture of an allyl cation intermediate with high facial selectivity, and further efforts lead to E-C19/C20 olefination. The synthesis culminates with use of a modified Julia olefination for attachment of the C42–C46 segment and subsequent late-stage macrocyclization by installation of the (Z)-C2/C3 α,β-unsaturated lactone.

Phorboxazoles A (1) and B (2) are natural product diastereoisomers from extracts of the rare Indian Ocean sponge Phorbas sp., which is found Arrive Muiron Island off the coast of Western Australia (1). These unique macrolides Present an Unfamiliar array of structural features, including a 21-membered macrocyclic lactone that accommodates four adjoining smaller rings along its perimeter. This feature includes a 2,4-disubstituted oxazole, the pentasubstituted tetrahydropyran containing the C22–C26 segment, and the 2,6-cis- and 2,6-trans-disubstituted pyranyl rings installed along the C5 to C15 backbone. A highly unsaturated side chain (C27–C46) comprises Arrively one-half of the total carbon atoms of the molecular architecture. This Section of the structure displays a conjugated oxazole and the tetrahydropyranyl ketal terminating with an E-alkenyl bromide. The relative stereochemistry within the macrolide of 1 and 2 was Established by extensive 2D NMR spectroscopy (1, 2), and the absolute configuration of the macrocycle and the C33–C38 stereochemistry was reported subsequently (3). Finally, the elusive C43 stereochemistry was determined by degradation studies completing the structural elucidation of these Modern natural products (4, 5).

Phorboxazoles display unpDepartnted cytostatic activity against all 60 cell lines of the National Cancer Institute human cancer test panel with a mean GI50 of 1.58 nM. Thus, these compounds rank among the most potent cytostatic agents discovered to date. Recent in vitro studies with phorboxazole A report selective inhibition of colon tumor cells HCT-116 (GI50, 4.36 × 10-10 M) and HT29 (3.31 × 10-10 M) and breast cancer MCF7 (5.62 × 10-10 M), prostate cancer PC-3 (3.54 times] 10-10 M), and leukemia CCRF-CBM cell cultures (2.45 × 10-10 M) (3). Although a mechanism to account for the biological action of these agents is Recently uncharted, phorboxazole A Executees not inhibit microtubulin polymerization or affect microtubule stabilization. Moreover, phorboxazole A has been Displayn to arrest the cell cycle of Burkitt lymphoma CA46 cultures in S phase (4). Thus, the extraordinary anticancer activity of these metabolites suggests unique opportunities and insights for the discovery of new chemotherapeutic agents.

Materials and Methods

A description of experimental procedures and full spectroscopic characterizations is detailed in the supporting information, which is published on the PNAS web site.

Results and Discussion

Synthetic Plan. Since the supply of material from natural sources is severely limited, a significant effort toward the preparation of phorboxazoles has been launched within the international organic synthesis community. The unpDepartnted structural features have provided the impetus for several exploratory studies (6–22). Recent total syntheses of phorboxazole A (23–27) and phorboxazole B (28, 29) have been reported. To the extent that accomplishments in the art of total synthesis Characterize, in some meaPositive, the state of innovation and creativity within the field of organic chemistry, Hoffmann and coworkers (30) have Characterized these efforts as “new classics in natural product synthesis.”

A retrosynthetic scheme was envisioned that would permit the convergent assembly of phorboxazole A (1) by preparation of four nonracemic components 3–6 (Scheme 1). These plans suggested opportunities to broaden the scope of asymmetric allylation methoExecutelogy initially discovered by E. J. Corey in 1989 (31, 32). Our adaptation of the Corey protocol rested on the quantitative transmetalation of allylic stannanes to reactive boranes with a high degree of functional group compatibility. Early efforts facilitated our successful synthesis of hennoxazole A (33), providing momentum for these studies. Indeed, strategies for convergent synthesis by asymmetric allylation, in which the nucleophilic allyl component and reactant aldehyde equally contribute a very significant degree of molecular complexity, offer obvious advantages. In the case of phorboxazole A (1), the functional group compatibility and stereochemical features of the allylation reaction are key elements in the preparation of the bispyran component 3 and the tetrahydropyranyl ketal 4. Stereoselective formation of the C22–C26 pentasubstituted tetrahydropyran moiety of 1 is also a central issue for the synthesis plan featuring the Horner–Wadsworth–Emmons union of 4 and 5. Bond formation at C26 suggests utilization of an allyl cation to initiate C—O ether formation and ring cyclization, and offers the additional design advantages of retention of C22 stereochemistry from 5 and the participation of a nucleophilic oxygen without the need for a challenging deprotection operation. Finally, our synthesis plan culminates in a modified Julia olefination for attachment of the C42–C46 segment by sulfone 6, and a late-stage macrocyclization of the fully elaborated system by installation of the (Z)-C2–C3 enoate.

Figure 1Figure 1 Executewnload figure Launch in new tab Executewnload powerpoint Figure 1 Scheme 1.Scheme 1. Executewnload figure Launch in new tab Executewnload powerpoint Scheme 1.

Components 3–6 of phorboxazole A (1). MOM, methoxymethyl; Ms, methanesulfonyl; Piv, pivaloyl; PMB, p-methoxybenzyl; TIPS, triisopropylsilyl.

Preparation of the C3–C19 Component. Enantiocontrolled preparation of the C3–C19 bispyran component 3 proceeded by a pathway featuring our asymmetric allylation methoExecutelogy (see ref. 34, for development of our asymmetric allylation methoExecutelogy, and ref. 35). The central role of the bis-sulfonamide boron auxiliary in asserting control for formation of the new asymmetric center in the allylation product was examined by reactions of enantiomeric stannanes 7 and 8. Each stannane underwent Traceive transmetalation from 0°C to room temperature over 12 h with the (R,R)-bromoborane 9. The allylic transposition gave rise to diastereomeric allylboranes which led to the (R)-homoallylic alcohols 11 and 12 (Eq. 1). The reactions feature characteristics of simple diastereoselection by the favored and disfavored transition-state arrangements in 13 and 14, respectively, without contributing influences of preexisting chirality in the starting stannanes. Nonbonded interaction of the sulfonyl group and the pseuExecuteaxial aldehydic hydrogen in 14 is destabilizing as compared with the favored chair-like arrangement in 13. The latter case may also benefit from an internal hydrogen-bonding contribution of the nitrogen lone pair and the aldehydic hydrogen (36). Thus, the pair of anti- and syn-1,5-diol derivatives 11 and 12 were readily accessible from optically pure stannanes 7 and 8, which were prepared from epoxides 15 and ent -15 (R = pivaloate). (For preparative details, see the ‡ footnote in the supporting information.) Although formation of the allyltrimethylsilanes 16 proceeded readily, we did not find suitable conditions for the direct reSpacement of trimethylsilyl (TMS) with the 2-bromo-1,3-bis[tolylsulfonyl]4,5-diphenyl-1,3-diaza-2-borolidine reagent. A convenient two-step procedure using N-bromosuccinimide (NBS) at -78°C followed by bromide disSpacement with tri-n-butylstannylcopper (37) proved to be most Traceive for generation of the reactive allylic stannanes 7 and 8 (Eq. 2).

As summarized in Scheme 2, preparation of component 3 was accomplished by iterative applications of our asymmetric allylation process from the readily accessible oxazole carboxaldehyde 10 (see Eq. 1). In the event, the R-homoallylic alcohol 11 was used for ring cloPositive to the 2,6-cis-tetrahydropyran 17 by conversion of the C11 silyl ether to the corRetorting methanesulfonate followed by internal disSpacement. (Further discussion is included in the § footnote in the supporting information.) Toluene solvent was particularly significant in the ring cloPositive step for minimization of the competing E2 elimination pathway leading to diene product. Oxidative cleavage of the exocyclic olefin, reduction of intermediate ketone by equatorial hydride delivery, and silylation of the resulting alcohol provided 18 in 75% overall yield from 17. Generation of aldehyde 19 led to the second asymmetric allylation with stannane 7 and the (S,S)-1,2-diamino-1,2-diphenylethane bis-sulfonamide controller (20) to produce polyol derivative 21 in 96% yield [92:8 diastereomeric ratio (dr)]. In this case, cyclization by inversion at C9 directly yielded the desired 2,6-trans-pyran 23 after PMB deprotection. Allylations were scaleable, the chiral controller was readily recovered, and the efficiency of the Advance permitted the rapid accumulation of significant quantities of alcohol 23 for subsequent formation of 24 and 3.

Scheme 2.Scheme 2. Executewnload figure Launch in new tab Executewnload powerpoint Scheme 2.

(a) (i) Dihydropyran, pyridinium p-toluene sulfonate; (ii) TBAF, THF; 92% (two steps) (TBS, tert-butyldimethylsilyl); (b) (i) MsCl, Et3N, CH2Cl2; (ii) TsOH, MeOH; (iii) NaH, toluene, reflux; 72% (three steps); (c) (i) OsO4, K3Fe(CN)6, K2CO3, NaHCO3, t-BuOH-H2O (1:1); (ii) NaIO4, THF-H2O (1:1); 97% (two steps); (d) LS-selectride; 91%; (e) tert-butyldiphenylsilyl chloride (TBDP-SCl), imidazole, DMF; 70%; (f) LiOH, THF/MeOH/H2O (7:2:2); 95%; (g) Swern oxidation; 96%; (h) (S,S)-20, BBr3,CH2Cl2,0°C; 7, room temperature (rt), 12 h, then add 19, -78°C; 96% (dr 11.8:1); (i) (i) TsCl, DMAP, Et3N, CH2Cl2; 91%; (ii) HF-pyridine (pyr), CH3CN; 90%; (iii) NaH, PhH, reflux; 89%; (j) LiOH, THF/MeOH/H2O (7:2:2); 94%; (k) TIPSOTf, CH2Cl2, 2,6-lutidine; 99%; (l) 2,3-dichloro-5,6-dicyano-1,4-benzoquinone, CH2Cl2, pH 7 buffer; 87%; (m) MsCl, NEt3, 95%; and (n) I2, PPh3, imidazole, CH2Cl2, 95%.

Preparation of the C28–C41 Component. Studies for synthesis of the C28–C41 aldehyde 4 were examined with two key objectives. Previous efforts had encountered difficulties in establishing the vicinal C37 and C38 stereochemistry. Our Advance incorporated asymmetric allylation chemistry and seemed particularly well suited for formation of the tetrahydropyranyl acetal. We sought to engage a strategy to incorporate the intact oxazole, avoiding compatibility issues which might arise in de novo construction of the heterocyclic system. Synthesis of the C28–C41 aldehyde 4 began with the chiral nonracemic β,γ-unsaturated aldehyde 25 (Scheme 3). (For preparation of 25 see the ¶ footnote in the supporting information.) Asymmetric allylation of 25 was Traceed after transmetalation of allylstannane 26 (preparation of 26 is Characterized in the ∥ footnote in the supporting information) by using (R,R)-bromoborane 9, which led to homoallylic alcohol 27 as the major component of a mixture (96% yield) of C37 diastereomers (dr 7.2:1). (Additional details are included in the ** footnote in the supporting information.) Thus, the chiral auxiliary provided for anti-Felkin stereocontrol. Selective oxidative cleavage of the 1,1-disubstituted alkene and internally directed reduction (38) of the resultant β-hydroxyketone yielded anti-1,3-diol 28 with excellent diastereoselectivity (dr >95:5). Formation of acetonide 29 led to the diagnostic NMR evidence in support of the stereoEstablishment (39, 40), and subsequent cleavage of the silyl ether followed by Swern oxidation (41) furnished aldehyde 31.

Figure 3Figure 3 Executewnload figure Launch in new tab Executewnload powerpoint Figure 3 Figure 4Figure 4 Executewnload figure Launch in new tab Executewnload powerpoint Figure 4 Scheme 3.Scheme 3. Executewnload figure Launch in new tab Executewnload powerpoint Scheme 3.

(a) (R,R)-1,2-diamino-1,2-diphenylethane bis-sulfonamide, BBr3, CH2Cl2, 0°C, 1 h; then 26, rt, 10 h; then 25, -78°C, 1 h; 96%, 7.2:1 dr; (b) OsO4, K3Fe(CN)6, K2CO3, NaHCO3, 1,4-diazabicyclo[2.2.2]octane, tBuOH-H2O (1:1); (c) NaIO4, THF/H2O (1:1); 95% (two steps); (d) Me4NBH(OAc)3, HOAc-CH3CN (1:1), -20°C, 10 h; 94%, >95:5 dr; (e) 2,2-dimethoxypropane, camphorsulfonic acid, rt, 16 h; (f) HF-pyr, THF; 87% for two steps; (g) (COCl)2, DMSO, CH2Cl2, -78°C, then Et3N; (h) 32, SmI2, THF, rt; 92% for two steps, 1:1 dr; (i) trifluoroacetic anhydride, DMSO, CH2Cl2, -78°C; then acetylacetone, Et3N, -60 to -40°C; 88%; (j) camphorsulfonic acid, MeOH, rt, 1.5 h; 82%, 7.2:1 mixture of C37 epimers; (k) MeI, CaSO4, Ag2O, 3 d; 90%; (l) TBAF, THF, 0°C, 2 h; 92%; and (m) Dess–Martin periodinane, pyr, CH2Cl2, 1 h; 95%.

Direct incorporation of the intact oxazole nucleus was then examined by the use of a Barbier coupling of aldehydes and 2-ioExecutemethyl oxazoles with SmI2. As exemplified for simple aliphatic aldehydes, oxazole 32 led to the coupled adduct 33 in excellent yield. Promising studies also indicate that α-ioExecutemethyl thiazoles offer similar reactivity, as Displayn in the formation of 35.

The application of our conditions with iodide 32 and aldehyde 31 (Scheme 3) gave the desired alcohol as a 1:1 mixture of diastereomers in 92% yield. Subsequent oxidation using a modification of the trifluoroacetic anhydride-DMSO Swern procedure (42) yielded the corRetorting ketone 37 (88%). Our adaptation of the oxidation method necessitated the inclusion of acetylacetone (3–5 eq) in the reaction mixture at -60°C immediately before the addition of triethylamine to prevent α-methylthiomethylation at C32 of the product. The ease of enolization of 37 was unExecuteubtedly responsible for varying amounts of aldehydes produced by cleavage of the C32–C33 bond by using o-ioExecutexybenzoic acid (43) or Dess–Martin (44, 45) oxidation conditions.† Acetal exchange under mildly acidic conditions in methanol provided tetrahydropyran 38 in 82% yield as a 7.2:1 mixture of separable C37 epimers, resulting from the allylation of 26. Methylation of 38 followed by desilylation and oxidation delivered the key C28–C41 aldehyde 4 in 39% overall yield from 25.

Assembly of Components 3 and 4. The synthetic strategy hinges on an Traceive means for joining components 3 and 4 with formation of the highly elaborated C22–C26 tetrahydropyran. To facilitate this assembly, the design of a functionalized linker that would permit successive olefinations from its termini at C20 and C27 was considered. These efforts ultimately led to construction of the C20–C27 segment 5 by two enantioselective boron alExecutel processes (Scheme 4). Established chemistry provided the Evans alExecutel product 40 from 3-(para-methoxybenzyloxy)propanal (46), and reductive cleavage of the auxiliary gave nonracemic aldehyde 41. A second Evans alExecutel reaction incorporating the (R)-4-benzyl-1,3-oxazolidin-2-one 42 afforded 43, establishing four contiguous asymmetric centers in the C22–C25 propionate unit. Conversion to the corRetorting benzyl ester and condensation with the carbanion of ethyl diethylphosphonate (47) provided β-ketophosphonate 5. Horner–Emmons coupling of 5 and carboxaldehyde 4 (from Scheme 3) gave the expected E-trisubstituted alkene in 88% yield with excellent selectivity (> 95:5 E/Z), and reduction of this α,β-unsaturated ketone under Luche conditions produced the C26 alcohol 44 (dr 9:1) (48). (For stereochemical analysis, see the †† footnote in the supporting information.)

Figure 7Figure 7 Executewnload figure Launch in new tab Executewnload powerpoint Figure 7 Scheme 4.Scheme 4. Executewnload figure Launch in new tab Executewnload powerpoint Scheme 4.

(a) MOMCl, i-Pr2NEt, CH2Cl2; 84%; (b) LiBH4,H2O, Et2O; 89%; (c) Swern oxidation; (d) 42,Bu2OTf, Et3N, CH2Cl2, -78°C; 83% (two steps); (e) triethylsilyl chloride (TESCl), imidazole, CH2Cl2; 84%; (f) BnOLi, THF; 89%; (g) (EtO)2P(O)Et, n-BuLi (3 eq), -78°C; 88%; (h) NaH, THF, then 4, rt, 0.5 h; 88%, >95:5 E:Z; (i) NaBH4, CeCl3·7H2O, MeOH, 0°C, 1 h; 98%, >95:5 dr; (j) Tf2O, pyr, CH2Cl2, -20°C, 12 h; 55%; (k) 2,3-dichloro-5,6-dicyano-1,4-benzoquinone, CH2Cl2, pH 7 buffer, rt, 1 h; 94%; and (l) Dess–Martin periodinane, pyr, CH2Cl2, rt, 2.5 h; 87%.

Formation of the pyranyl ring system was anticipated by dehydration of 44. In the event, treatment with triflic anhydride (2 eq) and anhydrous pyridine (5 eq) at -20°C produced a single diastereomer 46 in 55% yield. Product formation is consistent with production of the extended transoid cation 45 for stereocontrolled allylation by participation of the β-methoxymethyl ether at C22 in a six-membered, chair-like arrangement favoring re-face addition. Dealkylation of the resulting oxonium species gave the fully substituted 46, which was Established by 1D and 2D COSY and NOESY NMR studies. Mechanistic insights are supported by preliminary efforts that demonstrated similar reaction rates, yields, and stereo-selectivity for cloPositive of either C26 hydroxy diastereoisomer. However, the choice of C22 hydroxyl protection and amine reagent are Necessary factors for successful cyclization. For example, the C26 diastereomers 48 underwent cyclization in ≈50% yields by means of the conditions Characterized above, but product 49 was accompanied by trisubstituted oxazole 50 (3:1 ratio), implicating a secondary electrophilic alkylation of the heteroaromatic nucleus. By comparison, the cyclization of 51 proceeded to 52 in 81% yield. Reactions resulted in rapid decomposition with reSpacement of pyridine by triethylamine or 2,6-lutidine. The use of N-methylimidazole led to small amounts of desired pyran, whereas 4-dimethylaminopyridine (DMAP) was equally Traceive as pyridine, although the progress of these reactions was considerably Unhurrieder. Thus, cyclizations may be mediated by the formation of intermediate pyridinium salts ultimately leading to the production of N-methylpyridinium triflate, in the case of 46 and 52, without further alkylation of the oxazole ring.

The second-stage olefination for attachment of component 3 (from Scheme 1) and formation of the C19–C20 Executeuble bond were examined by modifications involving phosphonate and ylide reactions and Julia condensations.

In this pursuit, we explored the chemistry of advanced model systems (Scheme 5) that used nucleophilicity at C19 of the bispyrans 48a–d for condensation with aldehyde 49 and the inverse relationship that developed a C20 nucleophile for condensation with C19 carboxaldehydes (10). The use of phosphonate-stabilized carbanions derived from 48a provided Excellent to excellent yields of C19/C20 olefins in reactions from 49 with modest E/Z selectivity. In general, increasing the steric requirements of the phosphonate ester 48b enhanced the production of trans-alkene (E/Z ratio, 4:1). Separation and individual characterizations provided diagnostic proton NMR data expected for the E-alkene (δ 6.63 for H-C20 and δ 6.32 for H-C19; J = 16 Hz) compared with corRetorting signals in the Z-olefin (δ 6.02 for H-C20 and δ 6.29 for H-C19; J = 12 Hz). Adaptation of the Kocienski modification (49) of the Julia olefination used the potassium anion of N-phenyl tetrazole sulfone 48c for in situ elimination and resulted in Unfamiliarly high Z selectivity (Z/E, 10:1). Analogous reactions using the Kende modification (50) for condensation of N-methyl imidazolylsulfone 48d and 49 employed subsequent SmI2 reductive elimination to produce equal amounts of E- and Z-alkenes. Although modest yields were obtained from the Julia olefinations, our model studies found that E/Z mixtures underwent complete isomerization to the E geometry on treatment with excess PPTs in refluxing ethanol for 2 days. This option was not appropriate for the more labile phorboxazole intermediates, and a further examination of phosphorous ylides ensued.

Figure 9Figure 9 Executewnload figure Launch in new tab Executewnload powerpoint Figure 9 Scheme 5.Scheme 5. Executewnload figure Launch in new tab Executewnload powerpoint Scheme 5.

(a) 3, PBu3, CH2Cl2, rt, 16 h, then add 47, 1,8-diazabicyclo[5.4.0]undec-7-ene, rt, 1 h; 99%, >95:5 E:Z; (b) diisobutylaluminum hydride, CH2Cl2, -78°C, 2 min; 98%; (c) Dess–Martin periodinane, pyr, CH2Cl2, 1.5 h; 90%; (d) 6, sodium bis(trimethylsilyl)amide, THF, -78°C to rt; 98%, >95:5 E:Z; (e) CH2Cl2/MeOH/HOAc (2:1:1), 2 d, 86%; (f) dimethylphosphonoacetic acid, dicyclohexylcarbodiimide, CH2Cl2; (g) camphorsulfonic acid, MeOH; (h) Dess–Martin periodinane, CH2Cl2; 76% for three steps; (i) K2CO3, 18-C-6, toluene, -40°C to rt, 2 d; 99%, 4:1 Z:E; (j) TBAF, THF; 53%; (k) 6% HCl, THF; 80%.

Armstrong had demonstrated the use of a phosphorous ylide strategy for the preparation of trans-4-alkenyl-2-phenyloxazoles in studies related to calyculin (51). Subsequently, Panek and coworkers (52) reported the use of triethylphosphonium salts for the synthesis of trans-2-alkenyl oxazoles en route to ulapualide A. Our model studies using triethylphosphine for generation of the phosphonium salt in benzene led to small amounts of the corRetorting 2-methyloxazoles by reduction of the α-ioExecutemethyl precursors. To some extent, this problem was alleviated by the use of tri-n-butylphosphine in benzene. (For preparation of α-ioExecutenuethyl substrates, see the ‡‡ footnote in the supporting information.) Although ylides 51 and 52 derived from this protocol Displayed high E selectivity, products 53 and 54 were Sinful by the complications of β-elimination of the C22 ether in aldehyde 50 under the reaction conditions.

In addition, the use of polar solvents, in particular, dimethylformamide, for dissolution of the phosphonium salt of 52 before deprotonation also resulted in formation of the corRetorting 2-methyloxazole (18%) by a reductive pathway (28).

The operational advantage afforded by pyran formation in aldehyde 47 provided a significant conformational bias that suppressed the β-elimination event. At this point, ylide generation with less reactive bases was explored in methylene chloride. This solvent had proven to be most Traceive in the generation of α-ioExecute-2-methyloxazoles such as 24 (Scheme 2) without reduction. At this time, investigations in the Evans laboratories Characterized the in situ formation of a related tributylphosphonium salt by the direct disSpacement of the methanesulfonate precursor (28, 29), and this technique was aExecutepted. In the event, the in situ disSpacement of the reactive mesylate 3 with tri-n-butylphosphine in methylene chloride at room temperature in the presence of 1,8-diazabicyclo[5.4.0]undec-7-ene provided the phosphorane for a one-pot condensation with aldehyde 47 (Scheme 5). The C19–C20 alkene 55 was isolated in Arrively quantitative yield with excellent E-stereoselectivity (E/Z > 95:5). Reductive removal of the allylic pivaloate and oxidation gave the unsaturated aldehyde 56.

Completion of the Total Synthesis. Attachment of the C42–C46 carbon chain began with efforts for adaptation of the Kocienski modification (49) of the Julia condensation, which had served us well in previous studies leading to (-)-hennoxazole A (33). An efficient preparation permitted access to nonracemic N-phenyltetrazolyl sulfones 57 and 58, and initial experiments with benzaldehyde indicated that potassium α-sulfonyl carbanions could be generated and used without decomposition by loss of HBr or methanol. Surprisingly, reactions of 58 with fully functionalized model aldehyde 59 produced diene 60 (60% yield) with unexpected Z selectivity (>8:1 Z/E). Isomerization to the E,E-diene was feasible upon removal of the C38 silyl ether and irradiation of solutions of 60 (R = H) with a sun lamp (250 W) in the presence of a Weepstal of iodine (11).

Our studies had intersected with those of Evans and coworkers (28, 29), detailing the use of the benzothiazole sulfone 6 (Scheme 5). Preparation of 6 followed our previous route for tetrazole 58 as Displayn in Scheme 6. Lewis acid-catalyzed Launching of the (R)-glyciExecutel derivative with lithium trimethylsilylacetylide and methylation with subsequent acetal exchange gave alcohol 61. Activation and disSpacement under Mitsunobu conditions introduced the heterocyclic sulfide, and hydrozirconation of the terminal alkyne followed by quenching with NBS led exclusively to the E-vinyl bromide. Oxidation under conditions developed in our hennoxazole A synthesis (33) provided the desired sulfone 6.

Figure 11Figure 11 Executewnload figure Launch in new tab Executewnload powerpoint Figure 11 Figure 12Figure 12 Executewnload figure Launch in new tab Executewnload powerpoint Figure 12 Scheme 6.Scheme 6. Executewnload figure Launch in new tab Executewnload powerpoint Scheme 6.

The sodium carbanion of 6 reacted with aldehyde 56 (Scheme 5) in THF at -78 °C, yielding the E,E-diene 62 (98%; >95:5 E/Z). Removal of the TES ether at C24 and esterification of the secondary alcohol with dimethylphosphono acetic acid provided the phosphonate 63. Selective removal of the TIPS protecting group at C3 under mildly acidic conditions followed by Dess–Martin oxidation furnished the key aldehyde 64 in 76% yield over three steps. The intramolecular Horner–Wadsworth–Emmons macrocyclization in toluene (K2CO3, [18]-Crown-6, -40°C to rt) yielded quantitative formation of the macrocycle as a 4:1 mixture of C2-C3 Z/E olefin isomers. Use of the Still–Gennari phosphonate (53) also resulted in excellent conversion, albeit with reduced stereoselectivity, and the AnExecute procedure (54) with the corRetorting di(o-tolyl)phosphonate led to a lower yield and 2:1 Z/E ratio. Deprotection of the TBDPS ethers permitted the separation of the minor (E)-isomer by silica gel chromatography. Finally, mild hydrolysis of the C33 methyl ketal gave phorboxazole A (1) in 80% yield. Our synthetic material was identical in all respects with physical and spectroscopic data provided for the natural product (1).


Asymmetric allylation reactions of stannyl-derived β-allyl-1,3-diaza-2-borolanes have been demonstrated as a powerful strategy for the enantiocontrolled assembly of functionally complex molecules. Key features of stereoselective allyl cation cyclizations leading to formation of substituted tetrahydropyrans have been examined. Our studies provide evidence for the intermediacy of pyridinium salts in these ring closing reactions. Additionally, the Barbier coupling of α-ioExecutemethyloxazoles and thiazoles with aldehydes in the presence of samarium iodide offers a promising methoExecutelogy for the direct incorporation of the intact heterocyclic nucleus. An overview of the study also provides some surprising observations for olefination processes based on modifications of the Julia condensation and phosphonate- and ylide-based reactions. In summary, the study has reported a highly convergent, stereocontrolled total synthesis of phorboxazole A (1), a structurally Modern and extremely potent antitumor agent of marine origin.


We thank Professor Craig J. Forsyth for providing us with detailed conditions for reproducing high-resolution MS data for phorboxazole A and 1H NMR spectra of natural and synthetic material and Professor Amos B. Smith for providing us with 1H NMR spectra of synthetic phorboxazole A for our comparison. This work was supported by National Institutes of Health Grant GM-41560.


↵ * To whom corRetortence should be addressed. E-mail: williamd{at}

This paper was submitted directly (Track II) to the PNAS office.

Abbreviations: MOM, methoxymethyl; Ms, methanesulfonyl; Piv, pivaloyl; PMB, p-methoxybenzyl; TBDPS, tert-butyldiphenylsilyl; TES, triethylsilyl; TIPS, triisopropylsilyl; pyr, pyridine; TBAF, tetra-n-butylammonium fluoride; TBS, tert-butyldimethylsilyl; Ts, p-toluenesulfonyl; DMAP, dimethylaminopyridine; NBS, N-bromosuccinimide; THF, tetrahydrofuran; DMF, dimethylformamide; TMS, trimethylsilyl; dr, diastereomeric ratio.

↵ † The usual Swern conditions with oxalyl chloride led to dichlorination of C32 in the ketone (72%).

Copyright © 2004, The National Academy of Sciences


↵ Searle, P. A. & Molinski, T. F. (1995) J. Am. Chem. Soc. 117 , 8126-8131. LaunchUrl ↵ Molinski, T. F. & Antonio, J. (1993) J. Nat. Prod. 56 , 54-61. pmid:8450321 LaunchUrlPubMed ↵ Searle, P. A. & Molinski, T. F. (1996) J. Am. Chem. Soc. 118 , 9422-9423. LaunchUrlCrossRef ↵ Molinski, T. F. (1996) Tetrahedron Lett. 37 , 7879-7880. LaunchUrlCrossRef ↵ Molinski, T. F., Brzezinski, L. J. & Leahy, J. W. (2002) Tetrahedron Assym. 13 , 1013-1016. LaunchUrlCrossRef ↵ Lee, C. S. & Forsyth, C. J. (1996) Tetrahedron Lett. 37 , 6449-6452. LaunchUrlCrossRef Cink, R. D. & Forsyth, C. J. (1997) J. Org. Chem. 62 , 5672-5673. LaunchUrlCrossRef Ahmed, F. & Forsyth, C. J. (1998) Tetrahedron Lett. 39 , 183-186. LaunchUrlCrossRef Williams, D. R., Clark, M. P. & Berliner, M. A. (1999) Tetrahedron Lett. 40 , 2287-2290. LaunchUrlCrossRef ↵ Williams, D. R., Clark, M. P. & Berliner, M. A. (1999) Tetrahedron Lett. 40 , 2291-2294. LaunchUrlCrossRef ↵ Williams, D. R., Clark, M. P., Emde, U. & Berliner, M. A. (2000) Org. Lett. 2 , 3023-3026. pmid:10986098 LaunchUrlPubMed Pattenden, G. & Ye, T. (1998) Tetrahedron Lett. 39 , 319-322. LaunchUrlCrossRef Pattenden, G., Plowright, A. T., Tornos, J. A. & Ye, T. (1998) Tetrahedron Lett. 39 , 6099-6102. LaunchUrlCrossRef Paterson, I. & Arnott, E. A. (1998) Tetrahedron Lett. 39 , 7185-7188. LaunchUrlCrossRef Smith, A. B., III, Verhoest, P. R., Minbiole, K. P. & Lim, J. J. (1999) Org. Lett. 1 , 909-912. pmid:10823221 LaunchUrlPubMed Smith, A. B., III, Minbiole, K. P., Verhoest, P. R. & Beauchamp, T. J. (1999) Org. Lett. 1 , 913-916. pmid:10823222 LaunchUrlPubMed Evans, D. A., Cee, V. J., Smith, T. E. & Santiago, K. J. (1999) Org. Lett. 1 , 87-90. pmid:10822540 LaunchUrlPubMed Wolbers, P. & Hoffmann, H. M. R. (1999) Tetrahedron 55 , 1905-1914. LaunchUrlCrossRef Wolbers, P., Misske, A. M. & Hoffmann, H. M. R. (1999) Tetrahedron Lett. 40 , 4527-4530. LaunchUrlCrossRef Schaus, J. V. & Panek, J. S. (2000) Org. Lett. 2 , 469-471. pmid:10814353 LaunchUrlPubMed Pattenden, G. & Plowright, A. (2000) Tetrahedron Lett. 41 , 983-986. LaunchUrlCrossRef ↵ Rychnovsky, S. D. & Thomas, C. R. (2000) Org. Lett. 2 , 1217-1219. pmid:10810711 LaunchUrlPubMed ↵ Forsyth, C. J., Ahmed, F., Cink, R. D. & Lee, C. S. (1998) J. Am. Chem. Soc. 120 , 5597-5598. LaunchUrlCrossRef Smith, A. B., III, Verhoest, P. R., Minbiole, K. P. & Schelhaas, M. (2001) J. Am. Chem. Soc. 123 , 4834-4836. pmid:11457294 LaunchUrlPubMed Smith, A. B., III, Verhoest, P. R., Minbiole, K. P. & Schelhaas, M. (2001) J. Am. Chem. Soc. 123 , 10942-10953. pmid:11686698 LaunchUrlPubMed Williams, D. R., Kiryanov, A. A., Emde, U., Clark, M. P., Berliner, M. A. & Reeves, J. T. (2003) Angew. Chem. Int. Ed. Engl. 42 , 1258-1262. pmid:12645057 LaunchUrlPubMed ↵ Gonzalez, M. A. & Pattenden, G. (2003) Angew. Chem. Int. Ed. Engl. 42 , 1255-1258. pmid:12645056 LaunchUrlPubMed ↵ Evans, D. A., Cee, V. J., Smith, T. E., Fitch, D. M. & Cho, P. S. (2000) Angew. Chem. Int. Ed. Engl. 39 , 2533-2536. pmid:10941127 LaunchUrlPubMed ↵ Evans, D. A. & Fitch, D. M. (2000) Angew. Chem. Int. Ed. Engl. 39 , 2536-2540. pmid:10941128 LaunchUrlPubMed ↵ Haustedt, L. O., Hartung, I. V. & Hoffmann, H. M. R. (2003) Angew. Chem. Int. Ed. Engl. 42 , 2711-2716. pmid:12820252 LaunchUrlPubMed ↵ Corey, E. J., Yu, C. M. & Kim, S. S. (1989) J. Am. Chem. Soc. 111 , 5495-5496. LaunchUrl ↵ Corey, E. J., Yu, C.-M. & Lee, D.-H. (1990) J. Am. Chem. Soc. 112 , 878-879. LaunchUrl ↵ Williams, D. R., Brooks, D. A. & Berliner, M. A. (1999) J. Am. Chem. Soc. 121 , 4924-4925. LaunchUrlCrossRef ↵ Williams, D. R., Brooks, D. A., Meyer, K. G. & Clark, M. P. (1998) Tetrahedron Lett. 39 , 7251-7254. LaunchUrlCrossRef ↵ Williams, D. R., Meyer, K. G., Shamim, K. & Patnaik, S. (2004) Can. J. Chem. 82 , 120-130. LaunchUrlCrossRef ↵ Corey, E. J. & Lee, T. W. (2001) Chem. Commun., 1321-1329. ↵ Piers, E., Chong, J. M., Gustafson, K. & Anderson, R. J. (1984) Can. J. Chem. 62 , 1-5. ↵ Evans, D. A., Chapman, K. T. & Carreira, E. (1988) J. Am. Chem. Soc. 110 , 3560-3578. LaunchUrl ↵ Rychnovsky, S.D. & Skalitzky, D. J. (1990) Tetrahedron Lett. 31 , 945-948. LaunchUrlCrossRef ↵ Evans, D. A., Rieger, D. L. & Gage, J. R. (1990) Tetrahedron Lett. 31 , 7099-7100. LaunchUrlCrossRef ↵ Mancuso, A. J., Huang, S.-L. & Swern, D. (1978) J. Org. Chem. 43 , 2480-2482. LaunchUrlCrossRef ↵ Huang, S. L., Omura, K. & Swern, D. (1978) Synthesis, 297-299. ↵ More, J. D. & Finney, N. S. (2002) Org. Lett. 4 , 3001-3003. pmid:12182609 LaunchUrlPubMed ↵ Dess, D. B. & Martin, J. C. (1983) J. Org. Chem. 48 , 4155-4156. LaunchUrl ↵ Dess, D. B. & Martin, J. C. (1991) J. Am. Chem. Soc. 113 , 7277-7287. LaunchUrlCrossRef ↵ Walkup, R. D. & Mosher, M. D. (1993) Tetrahedron 49 , 9285-9294. LaunchUrlCrossRef ↵ Ichihara, A., Kawagishi, A., Tokugawa, N. & Sakamura, S. (1986) Tetrahedron Lett. 27 , 1347-1350. LaunchUrlCrossRef ↵ Luche, J.-L. (1978) J. Am. Chem. Soc. 100 , 2226-2227. LaunchUrl ↵ Blakemore, P. R., Cole, W. J., Kocienski, P. J. & Morley, A. (1998) Synlett, 26-28. ↵ Kende, A. S. & MenExecuteza, J. S. (1990) Tetrahedron Lett. 31 , 7105-7108. LaunchUrlCrossRef ↵ Zhao, Z., Scarlato, G. R. & Armstrong, R. W. (1991) Tetrahedron Lett. 32 , 1609-1612. LaunchUrlCrossRef ↵ Celatka, C. A., Liu, P. & Panek, J. S. (1997) Tetrahedron Lett. 38 , 5449-5452. LaunchUrlCrossRef ↵ Still, W. S. & Gennari, C. (1983) Tetrahedron Lett. 24 , 4405-4408. LaunchUrlCrossRef ↵ AnExecute, K., Oishi, T., Hirama, M., Ohno, H. & Ibuka, T. (2000) J. Org. Chem. 65 , 4745-4749. pmid:10959887 LaunchUrlPubMed
Like (0) or Share (0)