AU2004231239A1 - Hexahydropyrrolo[1,2-C]pyrimidines as antiviral, antifungal and/or antitumor agents - Google Patents

Description

II

AUSTRALIA

Patents Act COMPLETE SPECIFICATION

(ORIGINAL)

Class Int. Class Application Number: Lodged: Complete Specification Lodged: Accepted: Published: Priority Related Art: Name of Applicant: The Regents of The University of California Actual Inventor(s): Larry A Overman, Frank Stappenbeck, Andrew I McDonald Address for Service and Correspondence: PHILLIPS ORMONDE FITZPATRICK Patent and Trade Mark Attorneys 367 Collins Street Melbourne 3000 AUSTRALIA Invention Title: HEXAHYDROPYRROLO[1,2-C]PYRIMIDINES AS ANTIVIRAL, ANTIFUNGAL AND/OR ANTITUMOR AGENTS Our Ref: 733125 POF Code: 463217/193871 The following statement is a full description of this invention, including the best method of performing it known to applicant(s): -1- 6ooeq ^1- HEXAHYDROPYRROLO[1,2-C]PYRIMIDINES AS ANTIVIRAL, ANTIFUNGAL AND/OR ANTITUMOR

AGENTS

O

0 The present application is a divisional application from Australian patent application number 60703/00 the entire disclosure of which is incorporated herein by reference.

Throughout this application various publications are referenced. The disclosures of these publications in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art to which this invention pertains.

FIELD OF THE INVENTION The present invention relates to methods for improved synthesis of guanidinium alkaloids, and more particularly to the total, convergent synthesis of the Crambescidin/Ptilomycalin family of guanidinium alkaloids.

This invention was made with Government Support under Grant No. NIH NHLBIS (HL- 25854), awarded by the National Institutes of Health. The Government may have certain rights in this invention.

BACKGROUND OF THE INVENTION Crambe crambe, a bright red encrusting sponge commonly found at shallow depths along the rocky coast of the Mediterranean is a rich source of structurally novel, bioactive alkaloids (Figure Among the most remarkable marine guanidine natural products are the family of alkaloids depicted in -Figure 1 that have a-rigid pentacyclic guanidine carboxylic acid core linked to an co-hydroxycarboxylic acid, ester or polyamine amide. This family, exemplified by ptilomycalin A (compound the crambescidins (compounds celeromycalin and fromiamycalin (compound 10) are characterized by a structurally unique pentacyclic guanidinium core that has a spermidine or hydroxyspermidine residue tethered by a long chain co-hydrokycarboxylic acid spacer.

The alkaloid, ptilomycalin A, was reported by Kashman, Kakisawa and co-workers from Y'\MhlarNX NO DM.rsI MA17o3-O Div.do.

c1 sponges collected in the Caribbean and Red Sea (Kashman et al., J. Am. Chem. Soc., 1989, 0 111:8925). Ptilomycalin A exhibits cytotoxicity against P388 (IC50 0.1 pg/mL), L1210 z 0.4 g/mL) and KB (IC50 1.3 ig/mL), antifungal activity against Candida albicans (MIC 0.8 4g/mL) as well as considerable antiviral activity against Herpes simplex virus, type 1 (HSV-1) at a concentration of 0.2 .g/mL (Overman, L. et al. supra). Recently, Sptilomycalin A has been shown to inhibit the brain Na K -ATPase and Ca 2 -ATPase Cc, from skeletal sarcoplasmic reticulum with IC50 values of 2pM and 10PM, respectively (Ohtani, et al.. Euro. J. Pharm. 1996, 310,

O

In addition to Ptilomycalin A, numerous other complex marine alkaloids having a hydropyrrolo[1,2-c]pyrimidine-4-carboxylate part structure have been isolated including 13,14,15-isocrambescidin 800, crambescidin 800 and crambescidin 816 from Crambe crambe (Jares-Erijman et al., J. Org. Chem. 1991, 56:5712-5715; Jares-Erigman et al., J. Org. Chem.

1993, 58:4805-4808; Tavares et al., Biochem. Syst. Ecol., 1994, 22:645-646; Berlinck et al., J. Nat. Prod. 1993, 56, 1007-10015.) Ptilomycalin A and several of the crambescidins show substantial antitumor, antiviral and antifungal activities. Crambescidin alkaloids have been described for use in inhibition of calcium channels (Jares-Erijman, et al., J. Org. Chem. 1993, 58:4805); inhibition of Na', K' and Ca 2 -ATPases (Ohizumi et al., Eur. J. Pharmacol., 1996, 310:95). Batzelladine alkaloids, exemplified by batzelladines B andD (Figure 1, Patil et al., J. Org. Chem., 1995, 60:1182; Patil et al., J. Org. Chem., 1997, 62:1814; and Patil et al., J. Nat. Prod., 1997, 60:704), are reported to modulate protein-protein interactions that are important for immunological responses (Patil et al., 1995 and J. Org. Chem., 1997, supra).

As a result of its low abundance, 13,14,15-isocrambescidin 800 has not been extensively screened, although it is reported to be less cytotoxic to L-1210 cells than other crambescidins.

(Jares-Erijman et al., J. Org. Chem., 1993, 58:4805-4808, supra).

The defining structural feature of the crambescidin alkaloids is a pentacyclic guanidine unit Slinked by a straight chain o-hydroxycarboxylic acid tether to a spermidine or hydroxyspermidine unit. Extensive NMR studies demonstrated that the relative stereochemistry of the pentacyclic cores of crambescidin 800, crambescidin 816 and ptilomycalin A is identical (Jares-Erijman et al., supra and Tavares et al., supra), while r n 13,14,15-isocrambescidin 800 is epimeric at C13, C14 and C15 relative to other members of the crambescidin family (Jares-Erijman et al., J. Org. Chem., 1993, supra, and Berlinck et al.,

C

J. Nat. Prod., supra). The absolute configuration of the guanidine moieties of 13,14,15- O isocramescidin 800 and crambescidin 816 was established by oxidative degradation of the S 10 oxepene rings of these alkaloids to yield (S)-2-hydroxybutanoic acid (Jares-Erijman et al., J.

Org. Chem., 1993, supra), while the absolute configuration of the hydroxyspermidine unit of crambescidin 816 was assigned using Mosher's method (Berlinck et al., supra, and Dale et al., J. Am. Chem. Soc., 1973, 95:512-519). Since 'H NMR and 13C NMR chemical shifts in the hydroxyspermidine fragments of 13,14,15-isocrambescidin 800 are nearly identical to those of 2 and 3, it had been assumed that the stereochemistry at C43 is the same for all Crambescidins (Berlinck et al., supra).

Apparent in the alkaloid compounds described (Figure is the occurrence of the hydropyrrolo [1,2c]pyrimidine unit with either the syn or anti relationship of the hydrogens flanking the pyrrolidine nitrogen.

In 1893, Biginelli reported the synthesis of dihydropyrimidines from the condensation of ethyl acetoacetate, aromatic aldehydes and urea. (Biginelli, Gazz. Chem. Ital., 1893, 23:360 (1893). Since Biginelli's disclosure, variations in all three components have led to the synthesis of an array of functionalized dihydropyrimidines and analogues. (Kappe, C. O., Tetrahedron, 49:6937 (1993). In 1993, we reported on the viability of "tethered Biginelli" condensations and verified that the cis orientation of the methine hydrogens was preferentially realized when the dehydrative condensation was promoted under Knoevenagel conditions to form cisi-l-oxohexahydropyrollo[1,2-c]pyrimidine products. (Overman et al., J.

Org. Chem., 1993, 58:3235-3237). These reactions represented the first use of the Biginelli O reactions in stereocontrolled organic synthesis. Tethered Biginelli Cl condensations have already proved to be powerful reactions for the O construction of Crambescidin (Overman et al., J. Am. Chem. Soc., 1995, Z 117:265) and batzelladine alkaloids (Franklin et al., J. Orq. Chem., 1999, 62:6379). Recently it was reported to use acetals in place of alkenes to generate the aldehyde component of a Biginelli cyclization (Cohen et al., tC Organic Letters, 1999, VI N13:2169-2172).

C In 1995, an enantioselective total synthesis of (-)-Ptilomycalin A (Overman et al., J. Am. Chem. Soc., 117:2657 (1995)) was reported, which was the first total cN synthesis of a member of the Crambescidin alkaloid family.

There remains a need for improved methods of total, convergent synthesis of alkaloid compounds having biological activity, such as antifungal, antiviral and/or anti-tumor activity.

The discussion of the background to the invention herein is included to explain the context of the invention. This is not to be taken as an admission that any of the material referred to was published, known or part of the common general knowledge in Australia as at the priority date of any of the claims.

Throughout the description and claims of the specification the word "comprise" and variations of the word, such as "comprising" and "comprises", is not intended to exclude other additives, components, integers or steps.

SUMMARY OF THE INVENTION Accordingly, the present invention provides improved methods for convergent, total enantioselective synthesis of guanidinium alkaloid compounds including compounds having cis- or -trans-1-oxo-and 1-iminohexahydropyrrolo [1,2c]pyrimidine units such as, 13,14,15-Isocrambescidin 800, Crambescidin 800 and Ptilomycalin A, for use as therapeutic agents having antifungal, antiviral and/or antitumor activity.

Y:\clska\nk\species\60703-M)0.doc The compounds of the invention may be represented by the formulae: COMPOUNDS I-V.

1 II H

H

14 N N: H X- HO S4 III IV In which R= H, a carboxylic acid protecting group, an o-alkoxycarboxylic acid or an oalkoxycarboxylic acid ester, and X= any pharmaceutically acceptable counterion.

COMPOUNDS IA-VA H HC2 N 1>4 ;oN kN: OH H 111-A H t H o N N: qOH HO0 V-A IV-A

I

c COMPOUNDS VI-X

O

Hx-H 0 HX-H 0 H QH o O H H O SVI VII H 0H H 4 D/RIE

R

2 1 OR

R

2 N 14 D 1

R

2 14 o 14 0 Hx-H O

H

X-

H H X- H 0 VIII IX X In which, Ri= any alkyl, aryl or substituted alkyl group, R 2 0, OH, OGi, a spermidine moiety or a substituted spermidine moiety, where Gi a carboxylic acid protecting group and X= any pharmaceutically acceptable counterion.

The methods of the invention employ a convergent strategy for obtaining the compounds of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 depicts pentacyclic marine guanidine alkaloids obtained from marine organisms.

Figure 2 depicts a molecular mechanics model of the Ptilomycalin A/Crambescidin core.

Figure 3 depictsa hexahydropyrrolopyrimidine (compound B) having trans stereochemistry prepared by the methods of the invention.

Figure 4 illustrates Biginelli condensations of tethered ureido aldehydes using the methods of 0 the invention, as described in Example I, infra.

SFigure 5 is a synthetic scheme for making compounds 23-24, as described in Example I, infra.

Figure 6 is a synthetic scheme for making compounds 25-28, as described in Example I, N, infra..

O

C 10 Figure 7 presents two hypotheses for tethered Bignelli condensations under Knoevenagel conditions (Y=OH or NR 2 Figure 8 illustrates syntheses for compounds 37 43 as described in Example II, infra.

Figure 9 depicts reactions for synthesis of Ptilomycalin A (compounds 46 and 47) as described in Example II, infra.

Figure 10 depicts syntheses of compounds 49 to 53 as described in Example II, infra.

Figure 11 depicts syntheses of compounds 54 to 56, as described in Example II, infra.

Figure 12 illustrates syntheses of compounds 58 and 54 from compounds 57 and 59, as described in Example II, infra..

Figure 13 illustrates the syntheses of compounds 61 68 and Ptilomycalin A, as described in Example II, infra.

Figure 14 is a model showing expected preference for axial addition in forming the oxepene ring, as described in Example II, infra.

Figure 15 depicts the syntheses of Crambescidin 800 (compound 2) and compounds 71 z as described in Example I, infra.

Figure 16 depicts the syntheses of compounds 76 to 80, as described in Example m, infra.

Figure 17 depicts the syntheses of compounds 81 to 84, as described in Example III, infra.

I Figure 18 depicts the syntheses of compounds 85 to 88, as described in Example m, infra.

N 10 Figure 19 depicts the syntheses of compounds 89 to 93, and compound 2 (Crambescidin 800), as described in Example III, infra.

Figure 20 is a molecular mechanics model of the 13, 14, 15-Isocrambescidin 800 core and the Ptilomycalin A/Crambescidin core, as described in Example IV, infra.

Figure 21 is a retrosynthetic analysis of the Isocrambescidin core, as described in Example IV, infra.

Figure 22 depicts syntheses of compounds 99-103 as described in Example IV, infra.

Figure 23 depicts syntheses of compounds 105a-106 as described in Example IV, infra.

Figure 24 shows the pentacyclic intermediate in the (-)-Ptilomycalin A synthesis as described in Example IV I, infra.

Figure 25 shows the synthesis of compounds 105a and 105b through 108a and 108b, as described in Example IV, infra.

Figure 26 shows the synthesis of Isocrambescidin 800 (compound 2) as described in Example IV, infra.

SFigure 27 depicts the creation of compounds 114-116 as described in Example IV, infra.

Figure 28 depicts the synthesis of compound 117, as described in Example IV, infra.

Figure 29 shows data for Mosher's derivatives of compounds 10 and 117 as described in I Example IV, infra.

C",

Figure 30 is a scheme showing a Biginelli condensation between a tethered guanyl aldehyde and a 0-ketoester afforded 1-iminohexahydropyrrolo[1,2-c]pyrimidine intermediates with the trans relationship about the pyrrolidine ring thus providing a strategy for constructing the Isocrambescidin core.

Figure 31 is three-dimensional models of methyl ester analogs of four pentacyclic guanidine units as described in Example V, infra. Models depicting only heavy atoms are oriented identical to the line drawings; models also showing hydrogen atoms are oriented with the guanidine unit projecting back.

Figure 32 is a scheme showing a Biginelli condensation between a guanyl aldehyde (or aminal) and a P-ketoester afforded 1 -iminohexahydropyrrolo[ 1,2-c]pyrimidine intermediates with the trans relationship about the pyrrolidine ring thus providing a strategy for constructing the Isocrambescidin core.

Figure 33 shows the retrosynthesis of the pentacyclic core ofIsocrambescidin 800 (compound 10) as described in Example V, infra.

Figure 34 depicts the synthesis of compound 134, as described in Example V, infra.

Figure 35 depicts the formation of Pentacycle 135, as described in Example V, infra.

rN Figure 36 depicts the formation of Pentacycle 135 using pyridinium p-toluenesulfonate, as Sdescribed in Example V, infra.

Figure 37 depicts the formation of Pentacycle 135 using HCI, as described in Example V, infra.

Figure 38 depicts models of the methyl ester analog of 139 showing the two chair S conformations of the hydropyran ring. In conformation A, the methyl group is axial and in conformation B it is equatorial.

0 Figure 39 depicts the formation of Penacycle 135b using DCI, as described in Example V, infra.

Figure 40 depicts the formation of compounds 141-143, as described in Example V, infra.

Figure 41 depicts the formation of compounds 141-143 from compound 138, as described in Example V, infra.

Figure 42 depicts the formation of compounds 145-146, as described in Example V, infra.

?0 Figure 43 depicts the formation of compounds 141-147, as described in Example V, infra.

Figure 44 depicts the F-19 NMR data for Mosher derivatives of compounds 10 and 147, as described in Example V, infra.

Figure 45 depicts the relative energy of pentacyclic guanidine isomers, as described in Example V, infra.

Figure 46 is a schematic diagram for a modified enantioselective total synthetic approach.

cl Figure 47 is a schematic diagram of the tethered Biginelli condensation.

O

Z Figure 48 is an improved synthetic approach of compound 152.

Figure 49 is a schematic diagram for making enantiopure Iodide compound 166, as described in Example VI, infra.

Figure 50 is a schematic diagram for coupling of the fragment with the tricyclic intermediate, as described in Example VI, infra.

S Figure 51 is a schematic diagram for making a pentacyclic acid as described in Example VI, infra.

Figure 52 is a schematic diagram for an improved method for making pentacyclic acids, as described in Example VII, infra.

Figure 53 is a diagram showing the synthesis of compounds 180 to 183 as described in Example VII, infra.

Figure 54 is a depiction of the synthesis of compounds 185-189 as described in Example VII, infra.

Figure 55 is a depiction of the synthesis of compound 194 as described in Example VII, infra.

Figure 56 is a mean graph response of Ptilomycalin A as described in Example VIII, infra.

Figure 57 is a mean graph response ofIsocrambescidin 800 trihydrochloride as described in Example VIII, infra.

Figure 58 is a mean graph response of Triacetylcrambescidin 800 chloride as described in 12 Example VIII, infra.

0 Figure 59 is a mean graph response of Crambescidin 657 hydrochloride as described in Example VIII, infra.

Figure 60 is a mean graph response of Crambescidin 800 trihydrochloride as described in Example VIII, infra.

Figure 61 is a mean graph response of Triacetylisocrambescidin 800 chloride as described in Example VI, infra.

Figure 62 is a mean graph response of 13-Epiptilomycalin A as described in Example VIII, infra.

DETAILED DESCRIPTION OF THE INVENTION The present invention provides methods for enantioselective total synthesis of guanidinium alkaloids and congeners in convergent fashion using tethered Biginelli reactions. This invention allows all of the heavy atoms of the pentacyclic core of the Crambescidin/Ptilomycalin A and Isocrambescidin to be assembled in one key step. The compounds produced may be used for pharmacological screening using known methods, to identify compounds having desired biological therapeutic activity, for example as antiviral, antifungal and/or antitumor agents.

For synthesizing the compounds of the invention, a method was developed for controlling stereoselection in tethered Biginelli condensations to synthesize either the cis or trans stereoisomer of 1-oxo and 1 -iminohexahydropyrrolo[ 1,2-c]pyrimidines.

The invention also provides methods for the preparation ofpentacyclic acid compound 7 of Figure 1) and precursor allyl ester compound 8 of Figure 1) intermediates, that allows analogs to be prepared that are not available by degradation of the sponge extracts (Kashman, et al. J. Am. Chem. Soc. 1989, 111, 8925; Ohtani, et al. J. Am. Chem. Soc. 1992, 114, 8472; Jares-Erijman, E. et al. J. Org. Chem. 1991, 56, 5712). It is expected that analogs will show improved pharmacological properties.

The present invention relates to compounds of the general formulae: I II HxH 'C2R 1 14 N N H X- H 6 III IV In which R= H, a carboxylic acid protecting group, an o-alkoxycarboxylic acid or an oalkoxycarboxylic acid ester, and X= any pharmaceutically acceptable counterion.

In one embodiment R= H and X=CI'.

In another embodiment R= allyl and X=Cl".

In another embodiment R= (CH 2 15 C0 2 H and X Cl" The invention includes methods for preparing the compounds. In a method for preparing compound I having the formula: 1 O H N 0 2

R

HX- H 9NN

HXHO

Sin which R= H, a carboxylic acid protecting group, an o-alkoxycarboxylic acid or an coalkoxycarboxylic acid ester, and X= any pharmaceutically acceptable counterion C 5 a compound having the formula:

GO

2 vin which G= a carboxylic acid protecting group, an o-alkoxycarboxylic acid or and c-alkoxycarboxylic acid ester, and Y= alcohol protecting group, is reacted with a compound of the formula:

OP

X

2

NHQ

In which X 2 0 or a ketone protecting group, Z= alkene or carbonyl protecting group, P= alcohol protecting group, and Q= amino carbonyl group, to produce a compound of the formula:

'OP

N0- R

OP

O N

H

in which X 2 O or ketone protecting group, P= alcohol protecting group, and R= carboxylic Sacid protecting group, co-alkoxycarboxylic acid or o-alkoxycarboxylic acid ester which is Z subsequently converted to the pentacyclic compound by deprotection, incorporation of ammonia, and cyclization.

Another embodiment is a method for preparing compound II: Cc, NCI ,,CO 2

R

c x-H0 In which, R= H, a carboxylic acid protecting group, an o-alkoxycarboxylic acid or an oalkoxycarboxylic acid ester, and X= any pharmaceutically acceptable counterion, by epimerizing the stereocenter at carbon-14 of the compound I.

In another embodiment, a method for preparing compounds IV and V: H S14 CO2R N 14

,CO

2

R

H-H

HX-H

IV

V

in which R= H, a carboxylic acid protecting group, an c-alkoxycarboxylic acid or an o- O alkoxycarboxylic acid ester, and X any pharmaceutically acceptable counterion by reacting Z compound

GO

2 C^ Cc in which G= carboxylic acid protecting group, an c-alkoxycarboxylic acid or an co alkoxycarboxylic acid ester, and Y= alcohol protecting group, with compound

OOP

OP

X

2

NHQ

In which X 2 O or ketone protecting group, Z= alkene or carbonyl protecting group P= alcohol protecting group, and Q= amidinyl group to produce a compound of the formula

.OP

,H

O

R

OP

HN N

H

In which X 2 0 or ketone protecting group, P= alcohol protecting group, R= carboxylic acid protecting group, an o-alkoxycarboxylic acid or an coalkoxycarboxylic acid ester which is subsequently converted IV and V by deprotection and cyclization.

cl Another embodiment is a method for preparing compound III: 0 H Z 0 2

R

S In which R= H, a carboxylic acid protecting group, an o-alkoxycarboxylic acid or an oalkoxycarboxylic acid ester, and X= any pharmaceutically acceptable counterion by S 5 epimerizing the stereocenter at carbon-14 and carbon 15 of the compound IV.

Protecting groups' and strategies for synthesis of organic compounds are well known in the art (Protective Groups in Organic Synthesis, 2 nd Ed. T.W. Greene, P.G.M. Wuts, J. Wiley and Sons, Inc. New York, 1991).

Carboxylic acid protecting groups may be chosen from the following groups including, but not limited to, esters and amides.

Alcohol protecting groups may be chosen from the following groups including, but not IS limited to, ether groups, silyl protecting groups, such as TIPS, TBDMS, SEM, THP, TES, 00 TMS, or ester groups, such as acetates, benzoates, and mesitoates.

Carbonyl protecting groups may be chosen from the following groups including, but not limited to, ethers, cyclic or acyclic acetals, ketals, thioketals or thioacetals.

Amine protecting groups may be chosen from the following groups including, but not limited to, N-alkyl, such as benzyl, methyl, N-Silyl groups, N-acyl groups, N-carbamates.

The invention also provides compounds of the general formulas: H

,CO

2 14 OH HO III-A IV-A Another embodiment is a method for preparing compounds I-A to V-A, which can be prepared by following the method of preparing compounds I to V, respectively, and including an additional step of removing the carboxylic acid protecting group or deprotecting the carboxylic acid.

Further, the invention provides compounds having the formula: VI

VII

0R R2 0-V VIII Ix In which, RI any alkyl, aryl or substituted alkyl group,

R

2 O0, OH, OGI, spermidine moiety or substituted spermidine moiety, in which GI =carboxylic acid protecting group and X any pharmaceutically acceptable counterion Methods for preparing compounds

VI-X:

Compound VI is prepared as described for compound I in which R is an coalkoxycarboxylic acid as depicted in the figure below: H H RI OH N 14 In which any alkyl, aryl or substituted alkyl group and including an additional step of reacting the pentacyclic compound of the formula above with a protected spermidine or a protected substituted sperimidine and subsequently deprotecting to produce VI.

Compound VII is prepared as described for compound II in which R is an coalkoxycarboxylic acid acid as depicted in the figure below: In which R 1 any alkyl, aryl or substituted alkyl group and including an additional step of reacting the pentacyclic compound of the formula above with a protected spermidine or a protected substituted sperimidine and subsequently deprotecting to produce VII.

Compound VIII is prepared as described for compound III in which R is an coalkoxycarboxylic acid acid as depicted in the figure below: In which Ri= any alkyl, aryl or substituted alkyl group and including an additional step of reacting the pentacyclic compound of the formula above with a protected spermidine or a protected substituted sperimidine and subsequently deprotecting to produce VIII.

Compound IX is prepared as described for compound IV in which R is an o- 21 alkoxycarboxylic acid acid as depicted in the figure below: 0 0 N N SH X- HO In which Ri= any alkyl, aryl or substituted alkyl group and including an additional step of C reacting the pentacyclic compound of the formula above with a protected spermidine or a 0 protected substituted sperimidine and subsequently deprotecting to produce IX.

Compound X is prepared as described for compound V in which R is an co-alkoxycarboxylic acid acid as depicted in the figure below: 4

,RH

N N- H X- H O In which RI= any alkyl, aryl or substituted alkyl group and including an additional step of reacting the pentacyclic compound of the formula above with a protected spermidine or a protected substituted sperimidine and subsequently deprotecting to produce X.

The compounds of the invention include where applicable, a geometric or optical isomer of the compound or racemic mixture thereof.

Pharmaceutically acceptable counterions may be chosen from the following: acetate, adipate, alginate, aspartate, benzoate, benzenesulfonate, bisulfate, butyrate, citrate, camphorate, camphorsulfonate, cyclopentanepropionate, digluconate, dodecylsulfate, ethanesulfonate, fumarate, glucoheptanoate, glycerophosphate, hemisulfate, heptanoate, hexanoate, hydrochloride, hydrobromide, hydroiodide, 2-hydroxyethanesulfonate, lactate, maleate, c methanesulfonate, 2-naphthalenesulfonate, nicotinate, oxalate, pamoate, pectinate, persulfate, 0 3-phenylpropionate, picrate, pivalate, propionate, succinate, tartrate, thiocyanate, tosylate, and Z undecanoate.

The compounds of the invention may be used in therapy as antiviral, antifungal and/or as c itumor agents. For such uses, the compounds are administered intravenously, intramuscularly, topically, transdermally by means of skin patches, bucally, suppositorally or orally to man or other animals. The compositions can be presented for administration to humans and animals in a variety of dosage forms which include, but are not limited to, liquid solutions or suspensions, tablets, pills, powders, suppositories, polymeric microcapsules or microvesicles, liposomes, and injectable or infusible solutions, granules, sterile parenteral solutions or suspensions, oral solutions or suspensions, oil in water and water in oil emulsions containing suitable quantities of the compound, suppositories and in fluid suspensions or solutions. The preferred form depends upon the mode of administration and the therapeutic application.

For oral administration, either solid or fluid unit dosage forms can be prepared. For preparing solid compositions such as tablets, the compound can be mixed with conventional ingredients e.g. talc, magnesium stearate, dicalcium phosphate, magnesium aluminum silicate, calcium sulfate, starch, lactose, acacia, methylcellulose, and functionally similar materials as pharmaceutical diluents or carriers. Capsules are prepared by mixing the compound with an inert pharmaceutical diluent and filling the mixture into a hard gelatin capsule of suitable size.

Soft gelatin capsules are prepared by machine encapsulation of a slurry of the compound with vegetable oil, light liquid petrolatum or other inert oil.

0 Z Dosage forms for oral administration include syrups, elixirs, and suspensions. The forms can 0> be dissolved in an aqueous vehicle along with sugar, aromatic flavoring agents and preservatives to form a syrup. Suspensions can be prepared with an aqueous vehicle with the aid of a suspending agent for example acacia, tragacanth, methylcellulose and the like.

,For parenteral administration, fluid unit dosage forms can be prepared utilizing the compound and a sterile vehicle. In preparing solutions the compound can be dissolved in the vehicle for injection and filter sterilized before filling into a suitable vial or ampoule and sealing.

C I10 Adjuvants such as a local anesthetic, preservative and buffering agents can be dissolved in the vehicle. The composition can be frozen after filling into a vial and the water removed under vacuum. The dry lyophilized powder can then be sealed in the vial and reconstituted prior to use.

The most effective mode of administration and dosage regimen for the molecules of the present invention depends upon the severity and course of the disease, the subject's health and response to treatment and the judgment of the treating physician. Accordingly, the dosages of the molecules should be titrated to the individual subject.

Adjustments in the dosage regimens and/or modes of administration may be made to optimize the antiviral, antifungal or antitumor efficacy of the compounds of the invention.

Efficacy of the compounds of the invention in therapy may be assessed using known methods. For example, efficacy of the compounds as anti-tumor agents may be assessed by tumor biopsy or non-invasive procedures to determine tumor growth inhibition. Similarly, efficacy of the compounds as anti-viral or anti-fungal agents may be determined using standard protocols such as as assays to detect decreases in numbers of viral particles or fungal cells, or in the numbers of virally or fungally infected cells.

The following examples are presented to illustrate the present invention and to assist one of ordinary skill in making and using the same. The examples are not intended in any way to >otherwise limit the scope of the invention.

EXAMPLE I Synthesis of cis- or trans-l-Oxo-and l-Iminohexahydropoyrrolo[l,2-c]pyrimidines

C-

m This Example describes a method for controlling the stereoselectivity of tethered Biginelli condensations. Modification of the electrophilic reaction component permits access to 0 10 hexahydropyrrolopyrimidines (Compound 10 in Figure 3) having either the cis or trans stereochemistry.

Materials and Methods The strategy for synthesis of compounds 12-18 is depicted in Figure 4, for compounds 21-24 in Figure 5 and for compounds 25-28 in Figure 6. Methods of synthesis were used as previously disclosed and known in the art, e.g. Minor and Overman, J. Org. Chem., 1997, 62:6379, incorporated by reference herein.

Synthesis of (R)-Benzvloxy-7-methyloct-6-en-3.ol (Compound 12). A solution of methyl-3-hydroxy-7-methyl-6-octenoate (Kitamuram et al., Org. Svnth., 1992 71:1) (21.5 g, 0.115 mol) and Et20 (100 mL) was added dropwise to a 0°C suspension of LiAIH 4 (6.8 g, 0.18 mol) and Et20 (0.5 After 1 h, H 2 0 (6.8 mL), 3 M NaOH (6.8 mL), and H 2 0 (20.4 mL) were added sequentially. The resulting mixture was filtered through a pad ofCelite, the filtrate was concentrated, and the resulting oil was purified on silica gel (1:1 hexanes-EtOAc) to provide 13.8 g of (R)-7-methyloct-6-ene-1,3-diol as a colorless oil: 'H NMR (500 MHz, CDC13) 5 5.04-5.08 1H) 3.82 2H) 3.68-3.79 3H) 1.97-2.05 2H) 1.59- 1.67 4H) 1.54-1.60 4H) 1.40-1.48 2H); 3 C NMR (125 MHz, CDC13) 131.8, 123.8, 70.8, 60.7, 38.3, 37.5, 25.5, 24.1, 17.5 ppm; IR (film) 3356 cm'l; [a 23 D [a]23577 [a] 23 5 4 6 [a] 23 42 5 [a] 23 4 0 5 (c 1.2, CHC13). Anal. Calcd for C 9

H,

8 0 2

C,

N 68.31; H, 11.47. Found: C, 68.09; H, 11.54.

0 S A solution of (R)-7-methyloct-6-ene-l,3-diol (7.00 g, 44.3 mmol) and DMF (80 mL) was added dropwise to a -40 0 C suspension of NaH (3.20g, 133 mmol, prewashed with hexanes 3 x 50 mL) and DMF (130 mL). After 15 min, benzyl bromide (5.30 mL, 44.3 mmol) was added, and the reaction was warmed to -10°C over 1 h. The reaction was quenched by pouring into saturated aqueous NH 4 CI (300 mL), and the resulting mixture was extracted with (4 x 150 mL). The combined organic layers were washed with brine (50 mL), dried (MgSO4), and filtered, and the filtrate was concentrated. The crude oil was purified on silica gel (9:1 hexanes-EtOAc to 4:1 hexanes-EtOAc) to provide 7.74 g of 12 as a colorless oil: 'H NMR (500 MHz, CDC13) 5 7.32-7.36 4H) 7.26-7.31 1H) 5.14-5.17 1H) 4.51 2H) 3.78-3.83 1H) 3.66-3.73 1H) 3.62-3.65 1H) 3.04 1H) 2.05-2.16 2H) 1.73-1.77 2H) 1.71 3H) 1.63 3H) 1.44-1.57 2H); 13C NMR (125 MHz,

CDCI

3 137.8, 131.5,128.2, 127.5, 127.4, 124.0, 73.0, 70.4, 68.8, 37.3, 36.3, 25.5, 24.0, 17.5 ppm; IR (film) 3443 cm-l; [a] 23 D 13.0, [a] 2 3 5 77 +13.9, [a] 2 3 546 +15.6, [a] 2 3 435 +26.5, [a]23405 +31.3 (c 1.4, CHCI 3 Anal. Calcd for C1 6

H

2 4 0 2 C, 77.38; H, 9.74. Found: C, 77.25; H, 9.74.

Synthesis of 3 -Amino-l-benzvlox, 7 -methvl-6-octene (compound 13). Diethyl azodicarboxylate (4.12 g, 23.7 mmol) was added dropwise to a solution of 13 (5.05 g, 20.3 mmol), Ph 3 P (6.22 g, 23.7 mmol), HN 3 (12 mL, 2.0 M in toluene), and toluene (75 mL) at 0°C. After 15 min, hexanes (0.2 L) was added, the resulting mixture was filtered through a plug of silica gel (the plug was washed with 30 mL of hexanes), and the eluent was concentrated to yield the crude azide as a slightly yellow oil that was used without further purification.

A solution of this crude azide and Et20 (20 mL) was added dropwise to a stirred 0 C suspension of LiAlH 4 (0.91 g, 24.0 mmol) and Et20 (100 mL), and after 15 min the reaction was warmed to room temperature. After 1 h, the reaction was cooled to 0°C, and H 2 0 (1 mL), 3 M NaOH (1 mL), and H 2 0 (3 mL) were added sequentially. The resulting mixture was filtered through a pad of Celite, and the filtrate was concentrated to provide 4.53 g of amine 13 as a colorless oil that was used without further purification: 'H NNM (400 MLI-z, 0 ~CDCI 3 5 7.35-7.38 (in, 4H) 7.27-7.32 (in, 1H1) 5.1 1-5.14 (mn, IH) 4.52 2H-) 3.56-3.65 (in, Z 2.88-2.95 (in, 1H-) 2.00-2.12 (in, 2H-1) 1.74-1.82 (in, 1IH) 1.70 31-) 1.62 3H) 1.42- 1.60 (in, 2H) 1.21-1.37 (in, 3H); 3 C NMvR (100 IfHz, CDCI 3 138.4, 131.5, 128.3, 127.5, 127.4, 124.1, 72.9, 68.1, 48.8, 38.4, 37.6, 25.6, 24.6, 17.6 ppm; IR (film) 3366 cm-1; [a] 23 D 3.3, [a 2 577 [x 2 546 [ax] 2435 -49 [a 2 405 6 (c 1.0, CHCI 3 Anal. Calcd for C1 6

H

25 N0-HCI: C, 67.71; H, 9.23; N, 4.93. Found: C, 67.68; H, 9.27; N, 5.00.

Synthesis of (S)-I-Benzvloxj- 7 -methyl-3-ureido-6.ocgene (Compound 14h). Trimethylsilyl ri 10 isocyanate (0.90 m.L, 6.7 inmol) was added to a solution of crude 13 (1.15 g, 4.65 inmol) and i-PrOH (7 m.L) at room temperature. After 4 h, the reaction was concentrated, and the resulting oil was purified on silica gel (3:1 hexanes-EtOAc to EtOAc) to provide 873 mng of 14b as a colorless solid: mp 79-81 'H NMR (500 MI-k, CDCI 3 5 7.27-7.36 (in, 514) 5.45 IH) 5.08-5.11 (in, 11-) 5.93 2H) 4.94 2H) 3.53-3.63 (in, 3H-) 2.05 (in, 2141) 1.83-1.90 (in, 1IH) 1.69 3H-) 1.60 4H) 1.

4 2 -1.54 (in, 2H); 3 C NMR (125 MHz, CDCI 3 159.4, 138.0, 131.8, 128.3, 127.6, 127.5, 123.6, 72.9, 67.3, 47.9, 35.7, 35.3, 25.6, 24.4, 17.6 ppm; IR (film) 3340, 1653, 1602 cm-1; [a] 23 D +16.0, [Ca] 23 0 +17.3, [a]1 3 5 46 +19.6, [a] 23 43 +34.5, 23 405 +42.6 (c 1.0, CHCI 3 Anal. Calcd for C1 7

H

26

N

2 0 2 C, 70.3 1; H, 9.02; N, 9. Found: C, 70.39; H, 9.09; N, 9.55.

201 Conversion of Compound M4a to Intermediate Ia-with Ozone. Ozone was bubbled through a solution of urea 14a (120 ing, 0.60 iniol), CH 2

CI

2 (5 and MeOH (I m.L) at -78 0 C until the solution was saturated (blue color appeared and persi sted for 10 min). Nitrogen was then bubbled through the solution to remove excess ozone, Ph3P-Polystyrene (550 ing, 3 inmol Pig resin) was added, and the reaction was allowed to warm to room temperature. After 2 h, the reaction mixture was filtered, inorpholiniuin acetate (140 mng, 0.90 minol) was added to the filtrate, and the resulting solution was concentrated to give a colorless oil that was used without further purification.

ci Representative Procedure for Bi-ainelli Condensation under Knoevena'el Conditions.

Conversion of Comyound la to 17 and 18&. A solution of crude aminal la (0.60 niol), Z benzyl acetoacetate 16 m.L, 0.90 mxnol), morpholinium acetate (140 mg, 0. 90 mmol), and 2,2,2-trifluoroethanol (0.6 m.L) was maintained at 60TC for 2 d. After being cooled to room temperature, the reaction was partitioned between Et 2 O (20 mL) and 50% aqueous NII 4 Cl The layers were separated, the organic layer was dried (MgSO 4 and filtered, and the CK1 filtrate was concentrated. The resulting oil was purified on silica gel (2:1 hexanes-EtOAc to 1: 1 hexanes-EtOAc) to give 126 mg of 17a and 32 mg of 18a.

(4aR. 7S)- 7-(2-Hvdroxvethl)-3-methyi-oxo12.4a,5,6. 7 -hexahydrogyrrolofl.2clnvrimidine-4-carboxylic acid benzyl ester (I 7a): 'H NMR (500 MfiIz, CDGI 3 5 8.67 (s, I1H) 7.29-7.35 (in, 5H) 5.10-5.20 (in, 2H) 4.25 (dd, J= 11.3, 4.7 Hz, I1H) 4.11 (dd, J= 13.8, 8.2 Hz, I H) 3.84 I H) 3.56 (in, 2H-) 2.43-2.48 (in, I1H) 2.22 3H) 2.02-2.08 (in, I1H) 1.81 1.87 (mn, 1H) 1.65-1.74 (in, 3H); 3 C NMR (125 MUIz, CDCl 3 165.6, 154.9, 149.3, 135.9, IS 128.5, 128.3, 128.1, 102.2, 65.9, 59.0, 58.4, 52.2, 39.3, 30.6, 29.8, 18.0 ppm; IR (film) 3356, 1707, 1673, 1627 cm"; [a] 23 -26.5, [a] 2 26.8, 3 54 6 -3 7. 1, [a] 23 43 5 -119, [a] 2 405 -184 (c 1.00, CHC1 3 HRMS (0J) m/z 331.1657 331.1658 calcd for CjgH 23

N

2 0 4 (4aS. 7S)- 7 -(2-HydroxvethvI)-3-methvI..1.oxo..24a,5,6,7-hexahydropyrrolofl,2.

clv yrimidine-4-carboxi'Iic acid benzyl ester (18a): 'H NMvR(500 MiHz, CDC 13) 88.40 11-) 7.30-7.38(in, 51-)5.12-5.22(in, 2H) 4.42(in, 11-)4.35 (dd, J=10.2, 4.5 Hz, 11-)4.33- 4.44 (br s, 3:60(in, 21-) 2 4 O-2.4 5 1H) 2.45 3H) 2.06-2.10 (in,IH) 1.76-1.84(in, IH) 1.39-1.55 (in, 3H); 3 C NMR (125 M&z, CDC1 3 165.8, 153.0, 146.0, 136.1, 128.6, 128.6, 128.1, 99.1, 65.9, 58.9, 57.3, 53.6, 38.3, 34.9, 28.2, 18.3 ppm; IR (film) 3377, 3232, 1713, 1682, 1633 cm-1; [aX] 23 ,D -29.2, [a] 23 577 -29.0, [a] 23 546 -31.0, [a] 23 435 -30.2 (c 1.05, CHCI1 3 HRMS (0I) m/z 331.1629 331.165 8 calcd for for C I H 23

N

2 0 4 Anal. Calcd for C I 8

H

22

N

2 0 4 C, 65.44; H, 6.7 1; N, 8.48.

c-iRepresenative Procedure for Generatin Tethered Bi&ineiPrecursors by, Dih-d-roxylation and 4.2-Diol Cleavage Conversion-of 14b to 15. Osmium tetroxide (0.4 mL, 0. 1 M in t- Z BuON) was added to a solution of 14b (120 mg, 0.41 mmol), N-methylmorpholjne Noxide __(230 mg, 1.96 mmol), pyridine (30 mL, 0.4 mmol), and 10: 1 THF-H 2 0 (8 mL). After 30 min, Florisil (1 NaH-S0 3 (1 and EtOAc (20 m.L) were added, and the resulting mixture was stirred. After 30 min, the reaction mixture was filtered, and the filtrate was concentrated to provide the corresponding I ,2-diol as a colorless oil that was used without further purification.

A solution of this crude diol, Pb(OAc) 4 (0.21 g, 0.48 mmol), and CH 2

CI

2 (8 mL) was maintained for 30 min at room temperature. The reaction mixture was then filtered through a plug of Celite, morpholinium. acetate (92 mg, 0.62 mniol) was added to the filtrate, and this solution was concentrated to provide crude aminal 15 as a slightly yellow oil (Garigipati et al., J. Am. Chem. Soc.. 1985, 107:7790).

Conversion of Compound 15 to- 1 7b and 18b under Knoevenaz'el BEkinelli Conditions.

Following the representative procedure for Biginelli condensation under Knoevenagel conditions, crude aminal 15 (0.41 inmol) was condensed with 16, and the crude product was purified on silica gel (2:1 hexanes-EtOAc to 1: 1 hexanes-EtOAc) to provide 140 mg (8 of a 4:1 mixture of 17b and 18b. The isomers were separated by medium-pressure liquid chromatography (MiPLC) on silica gel (2:1 hexanes-EtOAc to 1: 1 hexanes-EtOAc).

(4aR. 7S) 7 -(2-BenzvloxethvI)-3-methvl..l-.oxo1.2.4.5 6. 7-hexahydropyrrolofl.2clpyrimidine-4-carboxiIic acid benzyl ester (I 7b): H NMv.R (5 00 MII-z, CDC 13) 5 8.21 (s, 1H) 7.25-7.38 (in, 10H) 5.11-5.21 (in, 2H) 4.43-4.53 (mn, 2H) 4.28-4.31 (mn, lH) 3.98-4.02 (in, 1H) 3.51-3.55 (in, 2H) 2.43-2.48 (in, 1H) 2.22-2.28 (in, 1H) 2.20 31-) 1.86-1.95 (in, 2H) 1.

7 4 -1.78 IH) 1.

6 1-1.66 (in, I1H); 1 3 C NMR (125 M1z, CDCl 3 165.9, 152.7,148.9, 138.4, 136.1, 128.5, 128.3, 128.3, 128.1, 127.5, 127.4, 101.4, 72.6, 67.8, 65.8, 58.0, 54.4, 33.4, 3 0.6, 2 8.9, 18.2 ppm; IR (film) 16 82, 163 3 cm-n1; [cX23D 18.7, [a] 2 3 57 -20.3, [a] 23 1 46 25.0, [a] 2 3 435 -71.7, [aC] 23 405 -108 (c 1.4, CHC 1 3 Anal. Calcd for C 25

H

28

N

2 0 4 C, 71.4 1; H, 6.71; N, 6.66. Found: C, 71.3 1; H, 6.80; N, 6.69.

(4aS. 7S)- 7 2 -Benzv-loxvethvl)-3-methyil.oxo-..4a, 5.6.7-hexahydirooyrrolo[.2- CI~vrimidine-4-carboxyIic acid benzvl ester 018b): 'H NMvR(500 MI-z, CDC 13) 88.94 (s, 1 H) 7.33-7.40 (in, 9H) 7.26-7.32 (in, I H) 5.14-5.24 (in, 2H) 4.47-4.56 (mn, 2ff) 4.33-4.41 (n 2H) 3.60-3.62 (in, 2H) 2.42-2.47 (in, INH) 2.26 3ff) 2.00-2.12 (mn, 2H) 1. 73-1.79 (in, 11-i) 1.44-1.55 (in, 2H); 1 3 C NMR (125 MHz, CDC1 3 166.0, 151.8, 147.1, 138.4, 136.3, 128.4, 128.2, 128.0, 127.9, 127.5, 127.4, 98.2, 72.8, 67.7, 65.5, 57.2, 54.6, 35.2, 34.8, 28.1, 18.2 ppm; IR (film) 1681, 1640 cm-1; [ca] 23 D -3 7.5, [a] 23 577 -37.0, [C] 2 3 546 -39.7, [a] 23 435 -34.5, [a]23 405 -14.1

CHCI

3 Anal. Calcd for C 25

H

28

N

2 0 4 C, 71.41; H, 6.71; N, 6.66. Found: C, 71.30; H, 6.73; N, 6.59.

Representative Procedure for Birinelli Condensation in the Presence of PPE. Conversion of Compound 14b to 17b and 18b.. Urea 14b (115 mg, 0.400 inmol) was converted to following the general olefln dihydroxylation and 1 ,2-diol cleavage procedure. A solution of the resulting crude aminal 15, benzyl acetoacetate 10 mg, 0. 59 inmol), polyphosphate ester (0.2 rnL), and CH 2

CI

2 (0.2 inL) was maintained at room temperature for 2 d. The reaction was then quenched by adding Et 2 O (20 inL) and 50% aqueous NaHCO 3 (5 inL). The layers were separated, the organic layer was dried (MgSO 4 and filtered, and the filtrate was concentrated. The resulting oil was purified on silica gel (2:1 hexanes-EtOAc to 1: 1 hexanes- EtOAc) to provide a 10 1 mg of a4:1 mixture ofl18b and 17b.

(4aS. 7S)- 7 2 -HvdroxvethvI)-i-imino..3.methvI.1.124a5., 7 -hexahvdro-pvrrolofl.2clpyrimidine-4-.carboxiylic acid benzyl este r hydroformate (Compound 23 ).-.Following the general procedure of Be matowicz, (Bemnatowicz et al., J. Org. Chem. 1992, 57:2497), a solution of 3 -amino- 7-methyl -6-octenolI (Overman et al., J. Am..hem. 1995, 117:2657) (0.95 g, 6.0 inmol), 1 przlelctoxmdn hydrochloride (0.95 g, 6.1 inmol), i-Pr 2 EtN (1.1 InL, 6.3 inmol), and DMNF (2.7 mL) was heated at 60"C. After 4 h, the reaction mixture was concentrated, and the resulting crude 21, a colorless oil, was used without further purification.

Ozone was bubbled through a solution of this sample of crude 21 and MeOH (25 mL) at Z 78*C until the solution was saturated. Nitrogen was then bubbled through the solution to remove excess ozone, Me 2 S (1 mL) was added, and the reaction was allowed to warm to room temperature. After I h, the reaction mixture was dried (MgSO 4 and filtered, and the filtrate was concentrated to give 22 as a yellow oil that was used without further purification.

Following the representative procedure for Biginelli condensation under Knoevenagel conditions, amninal 22 was condensed with compound 16 and the crude product was purified on silica gel (100% CHCI1 3 to 10: 1 CHCJ1 3 -i-PrOH to 10: 1:0. 1 CHC I 3 -i-Pr H-HCO 2 to yield 0.95 g of trans-B iginelli product 23 as a colorless oil: 1H NNM (500 MI-Iz, CDCl1 3 8 10.03 (br s, 2H4) 8.29 2H-) 7.27-7.3 5 (mn, 5N) 5.19 J 12.3 Hz, I H) 5.12 (d, J= 12.3 Hz, IH) 4.28-4.38 (in, 2H) 3.76-3.78 (in, IN) 3.49-3.53 (mn, IN) 2.45-2.50 (in, IH) 2.28 3H) 2.11-2.17 (in, 1.81-1.87 (mn, 1.58-1.67 (in, 2H-) 1.47-1.54 (in, IlH), the OH signal was too broad to observe; 1 3 C NMvR (125 MI~z, CDC1 3 166.6, 164.9, 150.7, 143.8, 135.5, 128.5, 128.2,128.1,101.1, 66.2,57.1, 56.156.0,36.0,34.1, 28.0, l 7 .2ppm; IR (film) 31'80, 1684, 1572 [a] 23 D -30.7, [a]1 3 577 -32.2, [a]13 141 -35.7 (c 3.1, CDCI 3 HRMS (FAB) m/z 330.1820 (MN, 330.1818 calcd for C, 8

H

24 0 3

N

3 (4aS. 7S) -I-(4-Bromobenzoylimino)..

7 -f 2 4 -bromobenzovlox v) ethyll-3-meg-hvi-124a. 5.6.7hexahydropvrro oJ1l.2cyrimidine4-carhoxylic Acid Benz vi Ester 4 -Bromobenzoyl chloride (400 mg, 1.81 inmol) was added at 0 0 C to a solution of 23 (220 mng, 0.60 mxnol), Et 3 N (0.50 m.L, 3.6 mmnol), CH 2 C1 2 (10 and a crystal of 4 -(dimethylamnino)pyridine.

After I h, the reaction was partitioned between Et 2 O (50 mL) and saturated aqueous N1{ 4 C1 (10 mL). The layers were separated, the organic layer was washed with brine (10 mL), dried (MgSO 4 and filtered, and the filtrate was concentrated. The residue was purified on silica gel (4:1 hexanes-EtOAc) to provide 150 mg of 24 as a colorless solid: mp 175-1 761C: 'H NMIR (500 MIHz, CDC1 3 5 7.98 J1= 7.8 Hz, 2H) 7.88 7.8 Hz, 2H).7.56

J=

7.8 Hz, 2H) 7.37-7.40 (in, 5N) 7.31 J1= 7.8 Hz, 2H) 5.15-5.25 (in, 2H) 4.79-4.82 (in, IN) 4.52-4.53 (mn, 21-) 4.4 1-4.45 (in, IN) 2.56-2.61 (in, IH) 2.48-2.53 (in, 1IH) 2.31 3H) 2.13- IN 2.19 1H) 1.92-1.96 1H) 1.56-1.73 2H), the NH signal was too broad to observe; O "C NMR (125 MHz, CDCI 3 176.9, 165.7, 165.4, 152.7, 143.7, 136.8, 135.8, 131.8, 131.0, Z 131.0,130.6, 128.9, 128.6, 128.3, 128.3, 128.2, 126.4, 101.0, 66.1, 62.3, 56.0, 55.9, 34.7, 33.7, 27.4, 18.9 ppm; IR (film) 1716, 1608 cm'; [a] 2 3 D [a] 2 3 7 7 [a] 2 54 6 [a] 2 3 435 +32.5, [a] 23 405 +68.5, (c 1.75, CHC13). Anal. Calcd for C 32

H

29 Br 2 N30 5 C, 55.27; H, S4.20; N, 6.04. Found: C, 55.20; H, 4.16; N, 6.04.

N (S)-N-/(Aminomethvlene)-4-methoxv-2,3,6-trimethylbenzenesulfonamidel-3-amino- 7methyl-6-octenol (25a). A solution of(S)- 3 -amino-7-methyl-6-octenol (Overman et al., J. M.

Chem. Soc. 1995. 117:2657) (19, 1.00 g, 6.36 mmol), S,S,-dimethyl N-(4-methoxy-2,3,6trimethylbenzenesulfonyl)-carbonimidodithioate (1.78 g, 5.34 mmol), and benzene (6 mL) was maintained' at reflux for 2 h. The reaction was quenched by adding Et20 (50 mL) and 0.1 M HCI (5 mL). The layers were separated, the organic layer was dried (MgSO 4 and filtered, and the filtrate was concentrated. The resulting crude oil was purified by MPLC (1:1 hexanes-EtOAc) to provide 1.81 g of the corresponding pseudothiourea as a colorless oil: 'H NMR (500 MHz, CDC13) 5 7.86 J= 9.8 Hz, 1H) 6.40 1H) 5.04-5.06 1H) 3.85 3H) 3.77-3.84 1H) 3.66-3.73 2H) 2.72 3H) 2.64 3H) 2.36 3H) 2.15 3H) 1.96-2.02 2H) 1.84-1.92 2H) 1.69 3H) 1.60-1.68 2H) 1.56 3H), the OH signal was too broad to observe; 3 C NMR (125 MHz, CDC13) 167.4, 158.8, 138.8, 137.0, 132.8, 132.4, 124.9, 122.7, 111.6, 58.7, 55.4, 52.2, 37.7, 35.4, 25.7, 24.1, 24.0, 18.4, 17.6, 14.2, 11.8 ppm; IR (film) 3480, 3290 cm'l; [a] 23 D -15.3, [a] 23 5 7 7 -14.7, [a] 2 3 546 -17.9, [a] 23 43 5 -31.8, [a] 23 405 -39.2 (c 1.9, CHC13). Anal. Calcd for C21H 34

N

2 0 4

S

2 C, 56.98; H, 7.74; N, 6.33. Found: C, 56.90; H, 7.69; N, 6.34.

Silver nitrate (26 mL, 0.2 M in MeCN) was added dropwise to a 0 C solution of a 1.59 g (3.60 mmol) portion of this pseudothiourea and MeCN (75 mL) that had been saturated with

NH

3 (Burgess et al., J. Org Chem.1994, 59:2179). The reaction mixture was allowed to warm to room temperature, and after 18 h, EtOAc (100 mL) was added and the resulting mixture was filtered through a plug of Celite. The eluent was concentrated to provide 1.46 g of 25a as a colorless solid: mp 107-109 0 C: 'H NMR (500 MHz, CDC13) 5 6.51 2H) ri 6.15 18) 4.90 (app s, 11-) 4.36 18) 3.80 (app s, 4H') 3.53-3.66 (in, 3H) 2.64 38) o2.56 3H)2. 10(s, 3M 1.85-1.86 2H) 1.71 (in, 18') 1.56 3H) 1.

3 9-1.32 6H); 3

C

Z NMR (125 Maz, CDC1 3 158.5, 156.8, 138.3, 136.6, 132.9, 131.9,124.8; 123.2, 111.7, 58. 1, 55.3, 47.5, 38.4, 35.3, 25.5, 24.5, 24.1, 18.2, 17.4, 1 1.9 ppm; IR (film) 3442, 3354, 1621 cm-1; [a] 23 D -20.0, [a] 23 577 -20.7, [a]l1 3 56-23.0, [a] 23 45-33.2, [a] 23 405 -35.7 (c 2.4, CHC1 3 Anal.

Calcd for C 2

GH

33

N

3 0 4 S: C, 58.37; H, 8.08; N, 10.21. Found: C, 58.3 1; H, 8.05; N, 10.2 1.

(S)-N-I(Aminomethylene)-4-methoxv-2.3. 6 -trimethvlbenzenesulfonamidel..3.amino-* benzvloxv- 7-methvi-6-octene (25b). Following the procedure described for preparing 25a, 13 I1i (0.807 g, 3.2,62 mmol) was converted in 80% overall yield to 25b a colorless oil; 'H NMIR 500 MHz, DMSO, 80-C) 5 7.25-7.32 (in, 58) 6.65 18) 6.45 18) 6.42 18) 5.01 (in, 18) 4.35 21-1 3.77 38) 3.73 (in, 18) 3.38-3.41 (in, 2H) 3.09 38) 2.63 3H) 2.56 3H-) 1. 88 (in, 2H) 1.69. (mn, 1H)1. 60 (in, 4H)_ 1.49 3H) 1.36-1.42 (in, 28); 13C NMR (125 Mfz, DMSO, 80-C) 157.3, 155.6, 138.2, 137.2, 135.3, 134.8,1 30.5, 127.7, 126.9, 126.8, 123.4, 123.2, 111.6, 71.6, 66.5, 55.1,47.5, 34.3, 34.2, 24.8,23.5, 22.8, 17.4, 16.9, 11.2 ppm; IR (film) 344 5, 3336, 1622, 1538 cm-1; 3 14.6, [a]1 3 57 7 +15.3, [a] 23 546 +18.2, [a] 23 43 +37.4, [a 23 405 +48.9 (c 1.80, CHC1 3 Anal. Calcd for C 27

H

39

N

3 0 4 S: C, 64.64; H, 7.84; N, 8.38. Found: C, 64.77; H, 7.88; N, 8.32.

Conversion of25a to 27c and 28c under Knoevenagel Biginelli Conditions. Following-the represenative olefin dihydroxylation and 1 ,2-diol cleavage procedure, 25a (100 mg, 0.24 mnmol) was converted to 26a. Ainiinal 26a was then condensed with 16 following the representative procedure for Biginelli condensation under Knoevenagel conditions with the exception that the concentration of 26a in 2 2 ,2-trifluoroethanol was 0.5 M. Purification of the crude product on silica gel 1 hexanes-EtOAc) provided 80 mng (61 of a 6:1 mixture of 27a and 28a.

A 1 20 mg (0.22 inmol) sample of a comparable product was esterified with 4-bromobenzoyl chloride (160 mng, 0.72 mmol) following the procedure described for the preparation of 24 to provide a crude residue that was purified on silica gel (3:1 hexanes-EtOAc) to provide 160 N mg (100%) of a 6:1 mixture 27c and 28c. These isomers were separated by HPLC (6:1 O hexanes-EtOAc; 20 mL/min, 300 x 22 mm 10 pm silica Alltech column) to give pure samples 0\ of 27c (tR 62 min) and 28c (tR 53 min).

(4aR, 7S)- 7 2 4 -Bromobenoyloxv)ethyll-l-(4-methoxy-2,3, 6trimethvlbenzenesulfonvlimino)-3-methvl-124a,5,67 -hexahydropyrrolo1,2-cluvrimidiner 4-carboxlic acid bengyzvl ester (27c): 'H NMR (500 MHz, CDC1 3 5 9.33 1H) 7.76 J= 8.4 Hz, 2H) 7.51 J= 8.4 Hz, 2H) 7.32-7.39 5H) 6.48 1H) 5.12-5.21 2H) 4.20- 4.29 2H) 4.13-4.18 1H)4.05-4.09 1H)3.78 3H) 2.66 3H) 2.59 3H) 2.46- 2.55 1H) 2.34 3H) 2.13-2.19 1H) 2.06 3H) 1.93-2.00 1H) 1.75-1.87 (inm, 2H) 1.64-1.71 1H); 13 C NMR (125 MHz, CDC1 3 165.5, 164.9, 158.5, 148.1, 145.6, 138.4, 136.5, 135.6, 132.9, 131.6, 131.0, 128.7, 128.6, 128.4, 128.3, 128.0, 124.7, 111.6, 103.8, 66.3, 62.4, 57.0, 55.3, 54.9, 32.5, 30.0, 28.5, 24.1, 18.5, 18.3, 11.8 ppm; IR (film) 3292, 1716, 1614 [a 1 23 D +55.5, [a] 23 577 +57.7, [a] 2 3 46 +66.5, [a] 23 4 35 +121, 23405 +150 (c 2.1, CHC 13). Anal. Calcd for C 3 5H 38 BrN 3 0 7 S: C, 58.01; H, 5.29; N, 5.80. Found: C, 57.98; H, 5.42; N, 5.52.

(4aS, 7S)- 7 -f 2 4 -Bromobenzovlox,ethyI-l-(4-methoxy-2,3,6trimethvlbenzenesulfonvlimino)-3-methyl-1,2,4a,5,6, 7 -hexahydrovpyrrolofl,2-cpvrimidine- 4-carboxvlic acid benzvl ester (28c): 'H NMR (500 MHz, CDC1 3 6 9.20 1H) 7.76 J 8.4 Hz, 2H) 7.51 J= 8.4 Hz, 2H) 7.33-7.54 (min, 5H) 6.44 1H) 5.12-5.23 2H) 4.36- 4.44 (min, 2H) 4.27-4.29 2H) 3.80 3H) 2.65 3H) 2.56 3H) 2.46-2.51 (min, 1H) 2.29 3H) 2.02-2.10 (min, 4H) 1.75-1.82 1H) 1.48-1.62 (in, 3H); 13 C NMR (125 MHz, CDCl 3 165.6, 165.3, 158.5, 146.7, 143.0, 138.4, 136.9, 135.8, 133.0, 131.6, 131.0, 128.7, 128.6, 128.3, 128.2, 128.0, 124.8, 111.6, 100.6, 66.2, 62.0, 56.2, 56.1, 55.4, 34.5, 33.2, 27.7, 24.0, 18.9, 18.3, 11.8 ppm; IR (film) 3298, 1716, 1614 cm'; [a] 2 3 D -17.7, [a] 2 3 5 7 7 -16.1, [a]23546 -18.3, [a] 23 435 -19.4, [a] 2 3 405 -13.3 (c 0.75, CHC 13). Anal. Calcd for C 3 sH 3 gBrN 3 07S: C, 58.01; H, 5.29; N, 5.80. Found: C, 58.06; H, 5.41; N, 5.55.

CI Conversion of Compound 25b to 2 7b and 28b under Knoevenarel Bi-Vinelli Conditions.

o Following the represenative olefin dihydroxylation and 1 ,2-diol cleavage procedure, Z (100 mg, 0. 20 mmol) was converted to 26b, and this crude material was condensed with 16 following the representative procedure for Biginelli condensation under Knoevenagel conditions with the exception that the concentration of 26b in 2 2 2 -trifluoroethanol was M. Purification of the crude product on silica gel (4:1 hexanes-EtOAc to 2:1 hexanes-EtOAc) provided 106 mg of a 7:1 mixture of 27b and 28b. Characterization data for the major product (4aR,7S)-7-(2-Benzyloxyethyl)- I 4 -methoxy-2,3,6-trixnethylbenzenesulfonylimino 3-methyl-i ,2,4a,5 6 7 -hexahydropyrrolo[

I,

2 -c]pyrixnidine-4-carboxylic acid benzyl ester (27b) as determined from this mixture: 'HNM(500 MHz, CDC 13)89.42 11)7.23-7.42 (in, 1011) 6.52 IH) 5.15-5.25 (in, 2H) 4.28-4.36 (in, 2H) 4.23 J1= 11. 1, 4.0 Hz, I1H) 4.03-4.07 (in, IH) 3.82 (in, 3M1 3.40-3.42 (in, 2H1) 2.70 3H1) 2.62 3H) 2.48-2.50 (in, 1IH) 2.31 311) 2.13 3H) 2.00-2.05 (mn, 1H) 1.93-1.95 (in, 2H) 1.79-1.83 (mn, 11-) 1.47- 1.53 3 C NMvR(125 MHz, CDC1 3 165.1, 158.5, 148.1, 145.6,138.5, 138.2, 136.4, 135.7, 133.2, 128.6, 128.4, 128.2, 128.2, 127.5, 127.4, 124.7, 111.7, 103.9, 72.5, 67.7, 66.3, 57.0, 55.9, 55.3, 33.4, 30.0, 28.8, 24.0, 18.5, 18.3, 11.9 ppm; IR(filrn) 3289, 1704, 1614 cin"'.

Anal. Calcd for C 35

H

4

IN

3 0 6 S: C, 66.54; H, 6.54; N, 6.65. Found: C, 66.66; H, 6.57; N, 6.66.

Conversion of Compound 25b to (4aS. 7 S)-7-(2-Benzy loxvethyl)-1-(4..methoxV-..6.

trimethvlbenzenesulfonylimino)3methy-12,4a 5,6 7 -hxahydropvrrolofl,2.clyyrimtine.

4-carboxvlic Acid Benalv Ester (28b) by Bi-ainelli Condensation in the Presence of PPE.

Following the representative procedure for olefin dihydroxylation and 1 ,2-diol cleavage, (100 mg, 0.20 inmol) was converted to 26b. Crude aminal 26b was then condensed with 16 following the representative procedure for Biginelli condensation in the presence of PPE to give, after purification on silica gel (2:1 hexanes-EtOAc to 1:1 hexanes-EtOAc), 77 mng (61 of 28b, which was contaminated with a trace of 27b 28b: 'H NTM (500 MHz, CDC 13) 589.23 I H) 7.22- 7.42 (in, 1 OH) 6.54 11H) 5.16-5.26 (mn, 2H) 4.36-4.40 (in, 2H) 4.26-4.35 (in, 2H-) 3.84 (in, 311) 3.45-3.48 (in, 211) 2.72 311) 2.65 311) 2.45-2.50 (in, 111) 2.32 3M1)2.15-2.20 (in, 11) 2.14 311)2.00-2.05 (in, 11) 1.62-1.72 (in, 1H) 1. 51 1.60 (in, 2H1); 1 3 C NMVR (125 MIHz, CDCl1 3 8 165.4, 158.5, 146.4, 142.9, 13 8.6, 136.4, 13 5.8, S133.4, 128.6, 128.3, 128.2, 128.2, 127.5, 127.4, 127.4, 124.7, 111.6, 100.5, 72.5, 67.2, 66.1, O 56.4, 56.0, 55.3, 34.5, 34.1,27.7, 24.0, 18.8, 18.3, 11.9 ppm; IR (film) 3290, 1712, 1614 cm'; S[a]23D -65.8, [a] 2 3 5 7 7 -67.5, [a] 23 5 4 6 -76.7, [a]23435 -117, [a] 2 3 4 0 5 -128 (c 1.1, CHC1 3 Anal.

Calcd for C 35

H

4 1

N

3 0 6 S: C, 66.54; H, 6.54; N, 6.65. Found: C, 66.49; H, 6.51; N, 6.56.

Conversion of Compound 28c to Compound 24. A solution of 28c (15 mg, 20 mmol) and STFA (2 mL) was maintained for 1 h at room temperature. The reaction was concentrated, and the resulting crude oil was used without purification. 4-Bromobenzoyl chloride (22 mg, 0.10 mmol) was added to a 0 C solution of this crude guanidine, Et 3 N (0.15 mL, 1.08 mmol),

CH

2

CI

2 (2 mL) and a crystal of 4 -(dimethlyamino)-pyridine. After 1 h, the reaction was quenched to Et20 (10mL) and saturated aqueous NH 4 CI (2 mL). The layers were separated, the organic layer was dried (MgSO 4 and filtered, and the filtrate was concentrated. The residue was purified on silica gel (4:1 hexanes-EtOAc) to provide 4 mg of 24 as a colorless solid.

S,S-Dimethyl N-(4-Methoxy-2,3.6-trimethvlbenenesulfonvl)carbonimidodithioate quanvlatine aeent (Figure Ammonia was bubbled through a solution of 4 -methoxy- 2 3 ,6-trimethylbenzenesulfonyl chloride (Fujino et al., Chem. Pharm. Bull., 1981, 29:2825) (10.3 g, 43.6 mmol) and CH 2 C1 2 (100 mL) at 0°C. After 30 min, acetone (0.5 L) was added, and the reaction mixture was filtered through a plug of silica gel and concentrated. The resulting solid was trituated with Et 2 0 to provide 9.18 g of 4 -methoxy-2,3,6trimethylbenzenesulfonamide as a colorless solid: mp 175-176 0 C; 'H NMR (400 MHz, acetone-d 6 5 6.75 1H) 6.36 2H) 3.86 3H) 2.63 3H) 2.58 3H) 2.05 3H); "C NMR(100 MHz, acetone-d 6 159.7. 139.0,138.0,134.6,125.3, 113.0, 56.2, 24.4, 18.5, 12.3 ppm; IR (KBr) 3385, 3279, 2983, 2942, 1582, 1560, 1486, 1309, 1148, 1113 Anal.

Calcd. for CoHisNO 3 S: C, 52.38; H, 6.59; N, 6.11. Found: C, 52.46; H, 6.55; N, 6.05.

A solution of 4-methoxy-2,3,6-trimethylbenzenesulfonamide (9.15 g, 39.9 mmol) and DMF mL) was added to a mixture of NaH (4.11 g, 98.6 mmol, washed 3x with hexanes) and DMF (20 mL) at 0°C. The reaction was allowed to warm to room temperature and was stirred vigorously for 10 min before CS 2 (6.9 mL, 11 mmol) was added. After another min, Mel (7.85 mL, 126 mmol) was added. After another 15 min, the reaction was poured >into saturated aqueous NH 4 CI (200 mL) and extracted with CHCI 3 (3 x 0.5 The combined z organic layers were dried (MgSO 4 filtered through a plug of silica gel and concentrated.

The crude solid was trituated with MeOH to provide 11.1 g of S,S-dimethyl N-(4methoxy-2,3,6-trimethylbenzenesulfonyl)carbonimidodithioate as a colorless solid: mp 175- 176 0 C; 'H NMR (400 MHz, CDC 3 5 6.56 1H) 3.84 3H) 2.71 3H) 2.57 3H) 2.52 6H) 2.13 3H); 3 C NMR (100 MHz, CDCI 3 182.3, 159.3, 139.2, 138.5, 130.3, 125.0, 111.7, 55.4, 23.9, 18.5, 16.3, 11.9 ppm; IR (film) 2969,2930,1552, 1476,1386,1307, 1146, 997, 925, 804 cm Anal. Calcd. for C1 3

HI

9

NO

3

S

3 C, 46.82; H, 5.74; N, 4.20. Found: C, 46.82; H, 5.73; N, 4.22.

(N

Results Biginelli Condensations of Tethered Ureido Aldehydes To pursue whether the free hydroxyl group in intermediate compound 1 might be influencing stereoselection, Biginelli condensations of this intermediate and benzyl ether derivative were examined (Figure Like compound 1, the benzyl ether congener was accessed from (R)-methyl-3-hydroxy-7-methyl-6-octenoate (11) (Kitamuram et al., Org. Svnth., 1992, 71:1).

Reduction of compound 11 with LiAIH4 and selective monobenzylation of the resulting diol by reaction with excess NaH and benzyl bromide in DMF at -40 to -100C furnished compound 12. Mitsunobu inversion of alcohol 12 with HN 3 (Loibner et al., Helv. Chim.

Acta, 1976, 59:2100) followed by reduction of the resulting azide and reaction of the resulting primary amine with trimethylsilyl isocyanate, provided urea compound 14b in 32% overall yield from compound 11.

In prior studies, the double bond of compound 14a had been cleaved with ozone, using a dimethyl sulfide workup, to generate compound 1 (Overman et al., supra). A more reproducible procedure was to add 1.5 equiv of morpholinium acetate to the crude reaction mixture after reductive workup of the ozonide, but prior to concentration. Replacing N dimethyl sulfide with polymer-bound triphenylphosphine eliminated contamination with O DMSO. Mass spectral data of the product compound la generated in this fashion indicated Z incorporation of morpholine (with loss of H 2 0) and showed the virtual absence of higher molecular weight oligomers.

Alternatively, compound 15 was generated by dihydroxylation of the corresponding alkene "1 precursor, followed by cleavage of the derived 1,2-diol with Pb-(Oac) 4 (Zelle et al., J. rg.

Chem., 1986, 51:5032). Aminals la and 15 were never subjected to an aqueous workup or purification, but rather were used directly following removal of either the phosphine polymer S 10 or lead salts by filtration and concentration of the filtrate after adding morpholinium acetate.

These intermediates are not simply a mixture of stereoisomers, but at least three components as judged by 'H and 3 C NMR data; multiple signals are observed for many carbon atoms in the 1 3 C NMR spectra, while broad peaks are seen in the 'H NMR spectra and no aldehyde signal is apparent.

Biginelli condensations of crude compound 15 or la (generated from 1 equivalent of compound 14a or 14b) were carried out under identical conditions by reaction with equivalents of 1-ketoester 16 and 1.5 equivalents of morpholinium acetate at 60 0 C in 2,2,2trifluoroethanol. These conditions provided the cis- and trans-1-oxohexahydropyrrolo[ 1,2c]pyrimidines 17a and 18a in a 4:1 ratio (80% yield) and the corresponding benzyl ether analogues 17b and 18b in an identical 4:1 ratio (81% yield). The P-oxygen substituent of the side chain clearly plays no significant role. Trifluoroethanol was employed as the reaction solvent since earlier studies with related intermediates had shown that cis stereoselection under Knoevenagel conditions was optimal in this highly polar solvent. For example, stereoselection in the condensation of 1 and 16 was 2:1 when ethanol was employed.

Products 17a and 18a did not interconvert upon resubmission to reaction conditions.

Stereochemical assignments for the hexahydropyrrolo[1,2-c]pyrimidine products followed from diagnostic 'H NMR signals of the angular methine hydrogens H4a and H7: 17a (4.25 and 4.11 ppm) and 17b (4.29 and 4.00 ppm) (Overman and Rabinowitz., J. Org. Chem., 1993, 58:3235).

0 z In a recent investigation, Kappe reported Ore. Chem., 1997, 62:7201) that the mild 0\ dehydrating agent polyphosphate ester (PPE) (Cava et al., J. Org. Chem., 1969,34:2665) was an excellent promoter of the classical three-component Biginelli condensation. Condensation of 15 with P-ketoester 16 at room temperature in a 1:1 mixture of PPE and CH 2 Cl 2 provided Biginelli products 17b and 18b in 60% yield, with the trans isomer 18b now predominating to the extent of 4:1. Identical to what was observed under Knoevenagel conditions, compounds 17b and 18b were recovered unchanged when resubmitted to the PPE reaction conditions for 48 h.

CNI Biginelli Condensations of Tethered Guanyl Aldehydes Although three-component condensations of guanidines, aldehydes, and P-ketoesters are known, this modification of the Biginelli condensation has not been widely explored. (Kappe, Tetrahedron, 1993, 49:6937). To examine the tethered variant, unsaturated guanidinium alcohol 21 was prepared from (S)-amino alcohol 19 (Overman et al., J. Am.Chem. Soc., 1995, 117:2657) by condensation with 1H-pyrazole-l -carboxamidine hydrochloride (20) (Figure Bernatowicz et al., J. Org. Chem., 1992, 57:2497). Ozonolysis of 21 followed by workup with dimethyl sulfide and concentration provided 22, which like its urea counterpart was a mixture of several components. When 22 was concentrated with 1.5 equiv ofmorpholinium acetate, FAB mass spectral data indicated incorporation of morpholine with loss of H 2 0; higher molecular weight oligomers were not observed for either 22 (X OH) or its morpholine adduct. Both intermediates performed identically in Biginelli condensations.

Without purification, 22 was condensed with P-ketoester 16, using Knoevenagel conditions identical to those employed in the urea series, to afford a single Biginelli adduct 23 in 42% overall yield from 19. This product had the trans stereochemistry as rigorously established by single-crystal X-ray analysis of dibenzoyl derivative 24 (Coordinates for compound 24 have been deposited with Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, To pursue the origin of the stereochemical reversal in the urea and guanidine series, Biginelli condensations of tethered N-sulfonylguanidine aldehydes 26 were investigated as depicted in Z Figure 6. Since the pKa ofN-sulfonylguanidinium salts is typically the sulfonylguanidine substituent electronically resembles more closely a urea than a guanidine (Tatlor et al., J.

Chem. Soc. Perkin Trans. 2, 1986, 1765; Yamamoto et al., in Synthesis and Chemistry of Guanidine Derivatives, Yamamoto and Kojima, Ed., Wiley, New York, 1991 (Vol. 2, pp 485- 526). The statistically corrected pK, of a monosubstituted guanidinium salt bearing an

SSO

2

NH

2 substituent has been determined to be 1.83 in water. Using the linear free energy correlation developed by Tatlor et al., supra, the value for the corresponding SO 2 Me- 0 10 substituted guanidinium salt would be 0.2. Treatment of amino alcohol 19, or the corresponding amino ether 13, with S,S-dimethyl N-(4-methoxy-2,3,6trimethylbenzenesulfonyl)-carbonimidodithioate, followed by aminolysis with NH 3 and AgNO 3 afforded the Mtr-protected guanidines 25 in good yield (Burgess et al., J. Org.

Chem., 1994, 59:2179). Dihydroxylation of these intermediates, followed by diol cleavage, provided 26a and 26b. These intermediates were again not simple mixtures ofstereoisomers; multiple signals were observed for many carbon atoms in the 1 3 C spectra, while 'H spectra exhibited broad peaks and showed no apparent aldehyde signal.

Biginelli condensation of crude 26b with P-keto ester 16 under Knoevenagel conditions identical to those employed with the other substrates proceeded in 84% yield to give the cisand trans-1 -iminohexahydropyrrolopyrimidines 27b and 28b in a 7:1 ratio. Nearly identical stereoselectivity was realized in the hydroxyethyl series. In dramatic contrast, when the condensation of 26b and 16 was carried out with PPE, the trans- iminohexahydropyrorolopyrimidine 27b predominated to the extent of 20:1.

Sulfonylguanidine products 27b and 28b were recovered unchanged when resubmitted for 48h to either the Knoevenagel or PPE reaction conditions.

Stereochemical assignments were made by chemical correlation of 28a with 24. Acylation of the crude product mixture produced from Biginelli condensation of 26a and 16 under Knoevenagel conditions with 4-bromobenzoyl chloride followed by separating the isomers by HPLC provided pure samples of 27c and 28c. Exposure of the minor product 28c to TFA at O room temperature removed the Mtr group, and acylation of the resulting free guanidine with Z 4-bromobenzoyl chloride provided 24.

These results demonstrate that stereoselection in tethered Biginelli condensations to form I -oxo- and 1-iminohexahydrophyrrolo[1, 2 -c]pyrimidines varies substantially depending on 1 reaction conditions and the nature of the group X (Figure With substrates having urea and SN-sulfonylguanidine functionality, cis stereoselection is observed when the condensation is accomplished under Knoevenagel conditions, while trans stereoselection (4- O 10 20:1) is observed when the condensation is carried out in the presence ofpolyphosphatester (PPE). Under both conditions, stereoselectivity was highest in the N-sulfonylguanidine series. With a substrate having a basic guanidine unit, the trans product is formed exclusively under Knoevenagel conditions. Since the Knoevenagel conditions are notably mild (morpholinium acetate in CF 3 CH20H at 60 0 this latter guanyl aldehyde route to trans-1iminohexahydropyrrolo[1,2-c]pyrimidines will like be particularly useful for the synthesis of Crambescidin and Batzelladine alkaloids having the anti relationship of the hydrogens flanking the pyrrolidine nitrogen. In the Examples below, the first total synthesis of Isocrambescidin 800 using this approach is described.

The origin of stereoselectivity is unknown. Without wishing to be bound by any theory, the following hypothesis is proposed (Figure Under Knoevenagel conditions, the stereochemistry-determining step in condensations of ureido or N-sulfonylimino aldehyde intermediates 29 could be cyclization of Knoevenagel adduct 31 to give 33. If this reaction has a late transition state, the cis-2,5-disubstituted pyrrolidine should be formed preferentially (molecular mechanics calculations on the model N-acylamino-2,5-disubstituted pyrrolidines 36 show that the cis isomer is 1.9 kcal/mol more stable than the trans isomer. Calculations were done using the MM2 force field and the Monte Carlo search routing ofMacroModel In contrast, in the guanyl aldehyde series, loss of HY from 29 to form the corresponding iminium ion 30 should be particularly favorable, since the nitrogen substituent in 30 is a weakly electron-withdrawing amidine group. If addition of the enol (or enamine) derivative of 16 is controlled primarily by destabilizing interactions with the side chain, trans adduct 32 should be produced preferentially in what could be the stereochemistry- Sdetermining step. Alternatively, the stereochemistry-determining step could be [4 2]cycloaddition of the enol (or enamine) or 30 from the face opposite the side chain, followed by loss of water (or morpholine). In accord with Kappe's recent investigations of the mechanism of the three component Biginelli reaction under classical acidic conditions (Kappe et al., sura), condensations of the ureido or N-sulfonylimino aldehyde intermediates 29 in the presence of polyphosphate ester (PPE) could also proceed by the iminium ion pathway to provide largely trans-l-oxo- and 1-iminohexahydropyrrolo[1,2-c]pyrimidines.

These results demonstrate that stereoselection in the tethered Biginelli condensations depicted in Figure 3 can be tuned to give either the cis or trans product. Under optimum conditions, the trans isomer can be obtained in high stereoselectivity and the cis isomer in moderate selectivity Tethered Biginelli condensations can be extended to include guanyl aldehyde substrates that produce Biginelli products, which should prove particularly useful for preparing complex guanidines.

EXAMPLE II Enantioselective Total Synthesis of Ptilomycalin A, Crambescidin 800 and Selected Congeners Synthesis Plan. A molecular mecfianics model of the methyl ester of the Ptilomycalin A/Crambescidin core is shown in Figure 2. The triazacenaphthalene ring system of these alkaloids is nearly planar with the seven- and six-membered cyclic ethers being oriented on one face. Since the two C-O bonds are axial (cis to the C10 and C13 angular hydrogens), it was surmised that the C8 and C15 spirocenters might assemble with the required stereochemistry if the proper cis stereochemistry of the central triazacenaphthalene unit were in place. Setting the cis stereorelationship of the angular hydrogens at C10 and C13 and relating the chirality of this unit to the C3 and C19 stereogenic centers of the oxepene and hydropyran rings proved to be the critical elements in evolving a stereocontrolled strategy for O preparing this class of guanidine alkaloids.

0 SAs illustrated in Figure 8, disconnection of the C8 aminal and retrosynthetic cleavage of the bond of 36 leads to the 1-oxohexahydropyrrolo[1,2-c]pyrimidine (X and 1 -iminohexahydropyrrolo[ 1,2-c]pyrimidine intermediates (X =NH 2 37. The 4 -alkoxycarbonyl-3,4-dihydropyrimidin-2(lH)-one part structure of 37 suggested that this essential bicyclic intermediate might be prepared by a novel modification of the c three-component Biginelli condensation, in which the urea and aldehyde components would be linked as depicted in 38.

This analysis has the appeal of high convergence, since the left-hand three rings of 36 would derive from acyclic fragment 38, while the right two rings and the ester side chain would be incorporated as the simple 0-ketoester unit 39.

Enantioselective Total Synthesis of Ptilomycalin A. In light of the difficulty experienced during degradation studies in removing the ester side chain of 1, the 1 6 -hydroxyhexadecanoic acid fragment was incorporated from the outset (Figure Alkylation of the dianion of methyl acetoacetate (44)(Huckin and Weiler, J. Am. Chem. Soc., 1974, 96:1082) with enantiopure (R)-siloxy iodide 45 provided 46 in 73% yield. Iodide 45 is conveniently available in high yield from methyl (R)-2-hydroxybutanoate (Kitamura et al., Org. Synth., 1992, 71:1). Selective transesterification (Taber et al., J. Org. Chem., 1985 50:3618) of the P-ketoester functionality with allyl 1 6 -hydroxyhexadecanoate using DMAP 4 -dimethylaminopyridine) as catalyst gave 47 in 64% overall yield from 44.

Since the tethered Biginelli condensation had just been verified, in this first generation approach this critical reaction was selected as early as possible in the synthetic sequence. For this reason, a less convergent strategy was pursued in which the electrophilic component of the Biginelli condensation was simplified by deletion of the C1-C7 fragment. The precursor of this intermediate, urea 50, was prepared in three steps from enantiopure methyl 3 -hydroxy-7-methyloct-6-enoate (48)(Kitamura et al., supra) as summarized in Figure iN Mitsunobu displacement of 48 with hydrazoic acid followed by reduction of the crude O P-azido ester with LiAlH 4 gave S amino alcohol 49 in 72% yield and in >98% ee.

Z Entantiometric excess was determined by evaluation of the 9 F NMR spectra of the corresponding and Mosher's amides. Use of other nitrogen nucleophiles such as phthalimide in the Mitsunobu reaction led to significant amounts of the corresponding a, S3P-unsaturated ester.

Condensation of 49 with potassium cyanate and HCI under standard conditions provided O unsaturated urea 50 in 82% yield after recrystallization. Ozonolysis 50 in MeOH at -78 0

C,

,I 10 followed by reduction of the intermediate hydroperoxide with Me 2 S and concentration furnished a viscous yellow oil. Further concentration of this product at 0.1 Torr for 5 days at 0 C to remove residual Me 2 SO lead to a nearly colorless amorphous powder. This intermediate is more complex than formulation 51 implies. Multiple signals were observed for many carbon atoms in the 13C NMR spectra and the 'H NMR spectrum was broad; no aldehyde signal was apparent, and mass spectral data indicated an oligomeric consistency.

All attempts to enhance the purity of 51 by chromatography were unsuccessful.

Biginelli condensation of crude 51 and 47 under the conditions developed during our previous model study (Overman et al., J. Org. Chem., 1993, 58:3235-3237), proceeded in low yield. A number of reaction parameters were surveyed and reaction efficiency was improved in polar solvents. The best result was achieved by heating a mixture of crude 51, 1.5 equivalent of P-ketoester 47, 1 equivalent of morpholinium acetate, a catalytic amount of acetic acid and excess Na 2

SO

4 at 70 0 C in EtOH. Purification of the resulting product on silica gel provided cis adduct 52 in 61% yield and trans adduct 53 in 8% yield. Stereochemical assignments for the hexahydropyrrolo[1 ,2-c]pyrimidine products followed from the similarity of their angular methine hydrogen signals (52: 4.25 and 4.11 ppm and 53: 4.44 and 4.09 ppm) with those of 41 and its trans epimer, the latter of which had earlier been analyzed by single-crystal X-ray analysis (Overman et al., J. Org. Chem., 1993, 58:3235-3237). In a recent detailed examination ofstereoselection in related Biginelli condensations (McDonald and Overman, J.

Org. Chem., 1999, 64:1520-1528) a reproducible procedure for generating the electrophilic reaction component and carrying out the Biginelli condensation was developed; these conditions reliably provide cis adducts in yields of 60-65%.

While 52 could be converted in one step to the spirotricyclic intermediate 54 by exposure to a slight excess ofp-toluenesulfonic acid (p-TsOH), the reaction was more reproducible on a large scale if the TBDMS group was first cleaved with pyridiniump-toluenesulfonate

(PPTS)

in MeOH and the resulting alcohol cyclized at room temperature in CHCI 3 with a catalytic amount ofp-TsOH (Figure 11). This sequence provided a single tricyclic product 54 in near quantitative yield. That this compound was epimeric to Ptilomycalin A at C14 was signaled by the 11.5 Hz diaxial coupling constant of the C14 methine hydrogen. (The Crambescidin numbering system is employed here.) That high diastereoselectivity would be realized in forming the spirohydropyran was established in our earliest model study and can be rationalized as outlined in (Figure 12).

Protonation of the vinylogous carbamate 57 to generate 58 followed by spirocyclization from the convex P-face would produce the spiroaminal having the oxygen axially disposed. Axial protonation of the exocyclic ketene hemiacetal, either prior or subsequent to spirocyclization, would deliver 54.

Although epimerization of 55 to the axial ester might have been possible at this point, this adjustment was deferred to the final stage of the synthesis, hoping to benefit from a presumed thermodynamic.preference for this group to be axial in the natural product. To prepare for the addition of the remaining carbons of the guanidine core, 54 was oxidized with the Swem reagent (Mancuso et al., J. Org. Chem. 1978, 43:2480) to provide 55, whose urea moiety was protected and activated for subsequent guanidine formation by O-methylation. It was critical that this methylation be performed under carefully optimized mild conditions, and that pseudourea 56 be purified rapidly on Et 3N-treated silica gel, or else significant epimerization at C10 resulted.

At this point the remaining C1-C7 carbons of the pentacyclic guanidine unit needed to be

I

appended. This elaboration proved to be extremely challenging. In early studies, we were unsuccessful in efficiently coupling lithium, cerium, titanium, or zirconium reagents derived Z from bromide 61 (Figure 13) to the benzyl ester congener of 56. A complicating issue was the rapid epimerization of 56 at C10 in the presence of Lewis acidic reagents. The Grignard reagent derived from 61 added in acceptable yield to 56 at -78 0 C. Quenching this reaction at low temperature with morpholinium acetate and immediate filtration to remove magnesium I salts provided the corresponding adduct as a mixture of alcohol epimers. Direct oxidation of Sthis intermediate under Swern conditions (Mancuso et al., J. Ore. Chem., 1978, 43:2480) provided 62 in 58% yield from 56. Approximately 5% of a diastereomer, resulting from the 0 10 minor enantiomer of 61, was removed at this point. Bromide 61 was originally prepared in 86% ee by an asymmetric reduction of an ynone precursor (Overman et al., J. Am. Chem.

Soc., 1995, 117:2657-2658). This sequence was extremely sensitive and yields were markedly eroded if magnesium salts resulting from the Grignard step were not removed quickly and thoroughly.

Cleavage of the silyl protecting group of 62 with TBAF furnished alcohol 63 which was then treated with ammonia and ammonium acetate, under conditions similar to those originally reported by Snider (Snider and Shi, J. Am. Chem. Soc., 1994, 116:549-557). After purification of the crude product on silica gel using an eluent containing formic acid, 64 was isolated in 60% as its formate salt NMR 5 8.23, "C NMR 8 165.8). Only a single pentacyclic guanidine was detected, with formation of the spiroaminal again occurring preferentially by axial C-O bond formation. A model of a tetracyclic cation 69 that is the likely direct precursor of the pentacyclic guanidine is shown in Figure 14; the C1-C7 side chain was replaced with a methyl group in generating this molecular mechanics model. A torsional preference for axial addition to the electron-deficient carbon is apparent in this figure and may be responsible for the high selectivity realized.

The total synthesis of Ptilomycalin A was readily completed from 64. The allyl ester of this intermediate was cleanly cleaved (Deziel, Tetrahedron Lett., 1987,28:4371), using palladium catalysis and the resulting acid was coupled with the bis-BOC-protected spermidine (Cohen, et al., Chem. Soc., Chem. Commun., 1992, 298) to generate amide 66 (Figure 13).

0 The ester was then epimerized by heating in MeOH in the presence of excess Et3N, however, z equilibrium for this epimerization favored the P epimer to the extent of 2-3:1. As a result, three recycles were required to obtain a-ester 67 in 50% yield. The equatorial C14 methine hydrogen of 67 showed a diagnostic doublet (J 4.8 Hz) at 5 2.93. Finally, cleavage of the SBOC protecting groups with HCO 2 H, followed by concentration and washing with aqueous NaOH-NaCI provided (-)-Ptilomycalin A trihydrochloride in high yield. Synthetic r compound 1 showed 1H and 3 C NMR spectra consistent with those reported for O (-)-Ptilomycalin A (Kashman et al., J. Am. Chem. Soc., 1989, 111:8925-8926; Ohtani et al., J.

S 10 Am. Chem. Soc., 1992, 114:8472-8479) and was indistinguishable from an authentic sample by TLC comparisons on three adsorbents. Synthetic 1 was converted to derivative compound 68, which also exhibited 'H and "C NMR spectra indistinguishable from those reported (Ohtani et al., supra). Synthetic compound 68 showed [a] 23 D -15.9 (c 0.8, CHCl3), nearly identical to the rotation, [a] 23 D -15.8 (c 0.7, CHCIl), reported for this well-characterized derivative of the natural product (Ohtani et al., supra).

Second Generation Synthesis Plan. A second generation synthesis of the Ptilomycalin A/Crambescidin alkaloids was undertaken with two specific goals in mind; to achieve the high level of convergence originally illustrated in Figure 8, where the entire carbon skeleton ofpentacycle 36 derive from a Biginelli condensation between a fully elaborated electrophilic component (38) and P-keto ester unit 39; and to gain access to either cis or trans 37 from a common precursor, thereby providing a convenient route to both the Crambescidin and Isocrambescidin core from a common intermediate. Details of the total synthesis of 13,14,15-Isocrambescidin 800 are described in the following Example. Critical to both these syntheses is the rapid and stereoselective construction of the common C1-C13 fragment (amine precursor to urea 38). This goal could be accomplished by combining the C1-C7 fragment 56 with the C8-C13 fragment 48 prior to the Biginelli condensation.

SEXAMPLE III 0 Z Synthesis ofCrambescidin 800 (Compound2). The synthesis of the C1-Cl3 fragment began with conversion of 3-butynol (compound 70) to thep-methoxybenzyl (PMB) ether 71 (Figure 15). The alkyne of 71 was deprotonated with n-buthyllithium at -40 0 C and the resulting acetylide treated with anhydrous DMF to provide ynal 72 in 90% yield, after quenching the intermediate a-aminoalkoxide into aqueous phosphate buffer (Journet et al., Tetrahedron SLett., 1988, 39:6427). The C3 stereocenter was introduced by the method of Weber and Seebach (Singh et al., J. Am. Chem. Soc., 987, 109:6187) through condensation of ynal 72 with Et 2 Zn in the presence of(-)-TADDOL (20 mol%) and Ti(Oi-Pr) 4 to give (S)-73 in 94% yield and >98% ee. This asymmetric transformation was reliably performed on a 45 g scale.

Propargylic alcohol 73 was protected as the triisopropylsilyl (TIPS) ether and the alkyne partially hydrogenated with Lindlar's catalyst to provide cis alkene 74. The PMB protecting group was oxidatively removed with DDQ and the resulting alcohol converted to iodide 75 in an overall yield of 89% from 73.

Enantiopure methyl 3 -hydroxy-7-methyloct-6-enoate (compound 48) (Kitamura et al., Org. Synth., 1992,71:1) was converted to amide in 88% yield by reaction with N,O-dimethylhydroxylamine hydrochloride according to the procedure of Weinreb (Garigipati et al, J. Am. Chem. Soc., 1985, 107:7790) followed by protection of the secondary alcohol as the triethylsilyl (TES) ether (Figure 16). Iodide 75 was converted to the corresponding lithium reagent and coupled with 76 to generate dienone 77 in 60-70% yield.

Masking the C8 carbonyl of 77 as the ketal was necessary to prevent dehydration, which occurred under the Mitsunobu conditions employed to install the P-amino functionality.

Ketalization was sluggish, however, when the p-hydroxy group was protected, so optimized reaction conditions were found which cleaved the TES group, did not promote P-hydroxy elimination of the intermediate P-hydroxy ketone and promoted ketalization. The novel ketalization conditions developed involved treatment of 77 with orthoester 78 (Roush and Gillis, J. Org. Chem., 1980, 45:4283-4287; Baganz and Domaschke, Chem. Ber., 1958, 91:650-653) and 1, 3 -propanediol in the presence of Amberlyst-15 to provide ketal 79 in yield. Mitsunobu displacement of the secondary alcohol with azide followed by reduction to 0 the amine provided compound 80 in 77% yield from compound 79. Amine 80, synthesized in 11 steps and in -30% overall yield from commercially available 3-butynol, was to serve as the common C1-C13 fragment for both the Crambescidin and Isocrambescidin syntheses (See Figure 16).

Condensation of amine 80 with TMSNCO yielded urea 81 in 89% yield (Figure 17).

Selective dihydroxylation of the trisubstituted double bond of 81 (Sharpless and Williams, O Tetrahedron Lett., 1975, 3045-3046) followed by cleavage of the vicinal diol with Pb(OAc) 4 S 10 in toluene and addition of morpholinium acetate yielded intermediate 82, which was used without purification. Biginelli condensation of crude 82 with P-ketoester 47 under optimal Knoevenagel conditions (McDonald and Overman, J. Org. Chem., 1999, 64:1520-1528) provided an inseparable 6-7:1 mixture of desired cis adduct 83 and undesired trans adduct 84 in 61% overall yield from urea 81. Stereochemical assignments for the hexahydropyrrolo[l,2-c]pyrimidine products followed from the similarity of their H13 angular methine hydrogen signals (83: 4.22 ppm and 84: 4.44 ppm) with those of 41 and its trans epimer and 52 and 53 (Figure 17).

The silyl protecting groups of 83 were next discharged with TBAF to provide the corresponding urea diol (Figure 18). Brief exposure of this crude diol to p-TsOH induced spirohydropyran formation and ketal deprotection, affording 85 in 71% for the two steps.

After protection of the secondary alcohol of 85 as the chloroacetate the minor trans isomer 12%) was easily separated from the desired cis isomer 86, which was isolated in 86% yield.

It was necessary to protect the C3 alcohol of 85 to prevent methyl ether formation during the methylation of the urea functionality. Exposure of urea 86 to excess MeOTfin the presence of a hindered pyridine base cleanly provided the corresponding methyl pseudourea, which was directly converted to the guanidine without intermediate silica gel purification. It was critical that the methyl pseudourea not be exposed to silica gel chromatography, as decomposition and epimerization at C10 resulted under typical purification conditions. The ability to transform the urea functionality to the guanidine without manipulations of the intermediate pseudourea represents one of the major advantages of the second generation synthesis over S the first. After considerable experimentation, we found that optimal guanylation/cyclization Z conditions, saturated NH3 in allyl alcohol buffered with NH4CI at 60°C for 1 day, cleanly provided pentacycles 87 and 88 in 81% from 86 as a 1.5:1 diastereomeric mixture at C14.

Subjection of pure compound 88 to the reaction conditions established this ratio as the thermodynamic equilibrium (Figure 18).

C",

S These reaction conditions represent a significant improvement from those used in the first generation synthesis, as yield improved dramatically, and both deprotection of the C3 protecting group and the desired epimerization of C14 to its thermodynamic ratio occurred.

The chloride counter ion was also obtained directly, eliminating detrimental aqueous washes.

It should be noted that incomplete deprotection of the C3 alcohol occurred when a simple acetyl protecting group was employed, whereas the chloroacetyl protecting group was quantitatively removed under the guanylation/pentacyclization reaction conditions. Allyl alcohol was employed as solvent to avoid transesterification of the allyl ester that occurred when ethanol or methanol was employed. Furthermore, it was found necessary to saturate the reaction solution with NH 3 at 0°C prior to heating in order to achieve thermodynamic equilibration of the C14 ester side chain. Unfortunately, the thermodynamic ratio of 1.5:1 favored the undesired P-epimer (H14: J 11.5 Hz). Pentacycles 87 and 88 were separated by medium pressure silica gel liquid chromatography, and the 3-epimer twice recycled through the guanylation/cyclization conditions to provide the major a-ester pentacycle 88 in a 52% overall yield from tricyclic urea 86.

The synthesis of Crambescidin 800 (compound 2) was completed as follows (Figure 19).

After removal of the allyl protecting group of 88 with Pd(PPh 3 4 and morpholine (Deziel, supra) acid 89 was coupled with (S)-7-hydroxyspermidine 90 using benzotriazol-1 -yloxytris(dimethylamino)phosphonium hexafluorophosphate (BOP)(Castro et al. Tetrahedron Lett. 1975 1219-1222) to provide the corresponding amide 91 in 82% yield.

Removal of the BOC groups with 3 M HCI in ethyl acetate (Stahl et al., J. Org. Chem., 1978 43:2285-2286) followed by purification of the crude product using reverse-phase

HPLC

yc provided the trihydrochloride salt of Crambescidin 800 in 75% yield. The data for the Strihydrochloride salt of synthetic 2 is in agreement with the 'H and 3 C NMR data reported for Z natural compound 2 (Jares-Erijman et al., J. Ore. Chem., 1991,56:5712-571.5; Berlinck et al., J. Nat. Prod., 1993, 56:1007-1015). Synthetic 2 was also converted to the triacetate derivative, 92. Data for synthetic 92 is also in agreement with 'H and 3 C NMR data reported for 92 prepared from natural 2 The Mosher's derivatives of(43S)- and 4 3R)-crambescidin 800 N (93) were made and compared to the corresponding Mosher's derivative prepared from -150 Stg of natural compound 2. The 9F NMR data is identical for the Mosher's derivative o prepared from natural 2 and synthetic 2, thereby for the first time unambiguously establishing 0 10 that the C43 stereochemistry of Crambescidin 800 is S (Figure 19).

Conclusion. The first total synthesis of Crambescidin 800 was accomplished in a convergent fashion with the longest linear sequence from commercially available starting material being 25 steps and in a 3.0% overall yield. This synthesis demonstrates for the first time that the tethered Biginelli condensation can be accomplished under suitable mild conditions that the aldehyde-urea fragment can contain all the atoms of the three left rings (CI-C13), thus allowing high convergence and efficiency in the synthesis of Crambescidin/Ptilomycalin A alkaloids. These investigations confirm the stereochemical assignment of 2 and rigorously establish that the absolute configuration of its hydroxyspermidine side chain is S.

Experimental Section GeneraL Dry THF, Et20, and CH 2

C

2 from Aldrich were filtered through a column charged with A1 2 0 3 (solvent purification system). Triethylamine (Et 3 pyridine, diisopropylethylamine (i-PrzNEt), diisopropylamine, and acetonitrile were distilled from CaH 2 at atmospheric pressure. Silica gel (0.040-0.063) by Merck was used for flash chromatography. The NMR spectra were recorded on the Bruker instruments (500 MHz and 400 MHz). The IR spectra were measured on Perkin-Elmer Series 1600 FTIR, and optical rotations were measured on Jasco DIP-360 polarimeter. Mass spectra were measured on a MicroMass Analytical 7070E (CI-isobutane) or a MicroMass AutoSpec E (FAB) spectrometer. Infrared spectra were recorded using a Perkin Elmer 1600 FTIR spectrometer.

Z Microanalyses were performed by Atlantic Microlabs, Atlanta, GA: Other general experimental details have been described (Metais et al., J. Org. Chem. 1997, 62:9210, incorporated by reference herein).

N Synthesis of l-( 4 -Methvoxvbenzvloxv)-3-butyne According to established procedures (Takaku et al., Tetrahedron lett., 1983, 24:5363; Nakajima et al., Tetrahedron Lett., 1988, 29:4139, both of which are incorporated by reference herein), TfOH (1.6 mL, 18 mmol) was added dropwise to a 0°C solution of PMBOC(=NH)CCl3 (169.3 g, 0.6 mol), 3-butyn-1-ol (67 g, 0.66 mol) and dry Et20 (600 mL). After 30 min the reaction mixture was quenched by the addition of saturated aqueous NaHCO 3 (100 mL). The phases were separated, and the aqueous phase was extracted with Et20 (50 mL). The combined organic phases were washed with brine (50 mL), dried (MgSO 4 filtered and concentrated. The resulting residue was diluted with hexanes (300 mL), filtered through a plug of silica gel, concentrated and stirred under vacuum (0.1 mm Hg) at 50 0 C for 12 h, yielding 114 g of 71, which was used without further purification: 'H NMR (500 MHz, CDC13) 8 7.28 J= 8.4 Hz, 2 6.89 (d, J 8.4 Hz, 2 4.49 2 3.80 (s,3 3.58 J= 7.0 Hz, 2 2.49 (dt, J= 7.0, 2.7 Hz, 2 2.00 2.6 Hz, 1 3 C NMR (125 MHz, CDCI 3 159.2, 130.0, 129.3,113.7, 81.3, 72.5, 69.2, 67.8, 55.2, 19.8 ppm; IR (film) 3292, 3001, 2936, 2863, 1614, 1514, 823 cm Anal. Calcd for CI 2 Hi 4 0 2 C, 75.76; H, 7.42. Found: C, 75.60; H, 7.49.

Synthesis of5-( 4 -Methoxvbenzvloxv)-2-.entvnal According to established procedures (Journet et al., supra) a hexane solution of n-BuLi (2.5 M, 32 mL) was added dropwise to a -40 0 C solution of 71 (14.45 g, 76.22 mmol) in dry THF (0.2 The reaction temperature did not exceed -35°C. After 10 min anhydrous DMF (11.8 mL, 153 mmol) was added in one portion and the cold bath was removed. After 30 min the reaction mixture was quenched by pouring into a vigorously stirred and cooled (-5 0 C) solution of 10% aqueous KH 2

PO

4 (0.4 L) and methyl tert-butyl ether (MTBE) (0.38 After 20 min the layers were separated and the organic layer was washed with H 2 0 (50 mL). The combined aqueous layers were back extracted with MTBE (100 mL), and the combined organic extracts were washed with brine 0 (50 mL), dried (MgSO 4 filtered and the filtrate concentrated. Purification of the residue on Z silica gel (10:1 hexanes-EtOAc; 6:1 hexanes-EtOAc) provided 14.97 g of 72 as a slightly yellow oil: 'H NMR (500 MHz, CDCI 3 5 9.16 1 7.26 J 8.5 Hz, 2 H), 6.88 (d,J 8.6 Hz, 2 4.48 2 3.79(s, 3 3.61 (t,J 6.7 Hz, 2 2.69 6.7 m Hz, 2 C NMR (125 MHz, CDCI 3 6 177.0, 159.2, 129.6, 129.3, 113.8,95.7, 81.9, 72.7, 66.5, 55.2, 20.6 ppm; IR (film) 3002, 2865, 2205, 1668, 1514, 824 Anal. Calcd for CN CI 3

H,

4 0 3 C, 71.54; H, 6.47. Found: C, 71.42; H, 6.54.

1 0 Synthesis of (5S)-Hvdroxv-l-(4-methoxvbenzvloxv).3-heDtyne According to the general procedure of Seebach (Webber and Seebach, Tetrahedron 1994, 50:7473-7484), incorporated by reference herein, Ti(Oi-Pr) 4 (12.2 mL, 41.0 mmol) was added to a 23°C solution of (4R, 5R)- 2 2 -dimethyl-a,a,a'a'-tetra(naphth-2-yl)- 1, 3 (27.3 g, 41.0 mmol) and dry toluene (340 mL). After 3 h, solvent was removed under reduced pressure (0.1 mm). The resulting residue was dissolved in dry Et 20 (560 mL) and the reaction vessel was cooled to -50 0 C, whereupon Ti(Oi-Pr) 4 (70 mL, 0.24 mmol) and 72 (44.7 g, 0.20 mmol) were added. Diethyl zinc (243 mL, 267 mmol, 1.1 M solution in toluene) was then added slowly over 1 h. The reaction vessel was then warmed to -27 0 C. After 18 h the reaction mixture was quenched with saturated aqueous NH 4 CI (100 mL). The organic phase was dried (MgSO 4 filtered through Celite® and concentrated. The resulting residue was purified on silica gel (20:1 hexanes-EtOAc; 5.6:1 hexanes-EtOAc; 1:1 hexanes-EtOAc) to provide 47.6 g of 73 as a colorless oil: 'H NMR (500 MHz, CDCI 3 5 7.25 J 8.4 Hz, 2 6.86 J= 8.4 Hz, 2 4.46 2 4.26 J= 6.4 Hz, 1 3.78 3 3.53 J= 7.0 Hz, 2 2.58 1 2.49 (dt, J= 7.0, 1.5 Hz, 2 1.66 2 0.97 J= 7.4 Hz, 3 3 C NMR (125 MHz, CDCI 3 5 159.1, 129.9, 129.2, 113.7, 82.3, 81.7, 72.4, 67.9, 63.5, 55.1, 30.9, 19.9, 9 .4 ppm; IR (film) 3418, 2965, 1613, 1514, 1249, 823, 733 cm-'; Anal. Calcd. for CI 5

H

20 0 3 C, 72.55; H, 8.12. Found: C, 72.26; H, 8.14. [a] 25 D [a] 25 57 7 [a] 2 5 5 45 [a] 25 43 5 [a] 25 405 (c 2.35, CHC13).

Following the general procedure of Ward (Ward et al., Tetrahedron Lett., 1991, 32:7165- 53 7166), incorporated by reference herein, 73 (23 mg) was treated wi th 0 (R)-a-methoxy-a-(triflouromethyl)phenylacetic acid chloride [(R)-MTPACI] to give the corre sponding (R)-MTPA ester. Capillary GC analysis [1 50 0 C to 200 0 C/2.0 0 C min-', tR 73-(R)-MTPA =3D 21.13 min, tR ent-50-(R)-MTPA 20.69 min] showed a ratio for 99.7:0.3 of 73-(R)-MTPA and ent-73-(R)-MTPA.

Synthesis of (S)C)](-ehlxbn)loy)5tispovsix--etn (74).

ri Triisopropylsilyl trifluoromethanelsulfonate (19.1 mL, 7 1.1 mmol) was added dropwise over mlin to a 0 0 C solution of 2,6-lutidine (10.3 mL, 88.4 minol), 73 (14.6 g, 58.6 mmol) and 10 dry CH 2

CI

2 (150 mL). After 1 h, the solution was poured into Et 2 O (400 m.L) and washed with IN HC1 (3 x 50 mL) and brine (20 mL). The organic phase was dried (MgSO 4 filtered and the filtrate concentrated. The crude oil was placed under vacuum 1 nun) overnight to provide 24.0 g of (S)-lI-(4-methoxybenzyloxy)-5-tiisopropylsiloxy-3-heptyne as a slightly yellow oil, which was used without further purification: 1 H NAM (400 MI-Iz CDCl 3 5 J 8.6 Hz, 2 6.91 J =8.6 Hz, 2 4.50 2 4.24-4.45 (in, 1 3.83 3 H,3.59 J= 7.2 Hz, 2 2.54 (dt, J= 7.2, 1.9 Hz, 2 1.

6 7 -1.76 2 1.0O1-1. 19(mn, 21 1.02 J= 7.4 Hz, 3 3 C NMR (100 MI-k, CDCI,) 159.1, 130.2, 129.2, 113.7, 82.9, 80.8, 72.5, 68.2, 64.3, 55.1, 32.1, 20.1, 18.0, 12.2, 9.4 ppm; JR (film) 2942, 2866, 1614, 1514,'1464, 1249, 1100 cin'; Anal. Calcd. for C 24

H

40

O

3 Si: C,'71.24; H, 9.86. Found: C, 7.18;H, 1.04;[a] 2 5 D -25.5, [a]15 57 26.3, [a] 255 46 -30.5, [a]11 4 35 -50.8, [a] 2 405 -60.8, (c 1.40, CHCI 3 A mixture of crude (S 4 -methoxybenzyloxy)-5-trisopropylsiloxy3heptyne (24.0 g, 58.6 inmol), freshly distilled quinoline (0.14 inL, 1.18 inmol), Lindlar's catalyst (Pd/CaCO 3 poisoned with PbO, 1.51 g) and dry 3:1 hexanes-EtOAc (360 m.L) was maintained at 23'C under 1 atm H 2 for 17 h. This mixture was then filtered through a plug of Celite, and the eluent was concentrated to yield 24.0 g of 74, which was used without further purification: 1'H NMR (400 MHz, CDCI 3 3 7.30O(d, J1= 8.6 Hz, 2 6.91 J 8.6 Hz, 2 5.47-5.52 (in, I 5.37-5.43 (in, I 4.48-4.52 (in, I 4.48 2 3.83 3 H), 3.46-3.50 (mn, 2 2.35-2.43 (in, 2 1.59-1.68 (in, I 1.47-1.56 (in, I 1.09 (app s, 21 HD, .0.89 J =7.4 Hz, 3 1 3 c Nlvf (100 MHz, CDC1 3 159.1, 135.8, 130.4, 129.2, 0 124.4, 113.7, 72.6, 69.9, 69.4, 55.2, 31.6, 28.7, 18.0, 12.3, 9.3 ppm; IR (film) 2942, 2866, 1613, 1514, 1248, 1097 cm-1; Anal. Calcd. for C 2 4 H.1 42

O

3 Si: C, 70.88; H, 10.41. Found: C, 71.06; H, 10.44; [a] 25 D 18.5, [a] 25 577 +19.7, [a] 25 546 +22.6, [a] 25 43 +41.9, [a] 25 405 +52.0, (c 1.80, CHCl 3 Synthesis of (S)-(Z)-l-Iodo-5friisoproplsiloxv-3-heD tene A solution of crude 74 ri (24.0 g, 58.6 mmol), DDQ (17.3 g, 76.2 mniol) and 20:1 CH 2 C1 2

-H

2 0 (210 mL) was maintained at 23'C for 1 h. The reaction mixture was quenched by pouring into Et 2 O (600 S 10 m.L) and washing with IN NaOH (2 x 200 mL) and brine (200 niL). The organic phase was dried (MgSO 4 filtered and concentrated. Chromnatagraphic separation of p-methoxybenzaldehyde was facilitated by reduction to p-methoxybenzyl alchohol. Towards this end, a solution of the resulting residue, MeOH (200 niL) and NaBH 4 (2.9 g, 77 nimol) .was maintained at 23'C for 1 h. The reaction mixture was quenched by pouring into Et 2

O

(300 mL) and washing with IN HCI (50 mL) and brine (50 niL). The organic phase was dried (MgSO 4 filtered and the filtrate concentrated. The resulting oil was purified on silica gel (20:1 hexanes-EtOAc; 15:1 hexanes-EtOAc; 10: 1 hexanes:EtOAc) to provide 16.0 g of (S)-(Z)-5-triisopropylsiloxy-3-heptenol as a colorless oil: 'H NMR (500 MI-Lz, CDC1 3 8 5-5.51 (in, 1 5.38-5.33 1 4.47 (ddd, J= 13.5, 6.5, 1.5 Hz, 1 3.66 2H), 2.35-2.30 2H), 1.65-1.57 I 1.53-1.46 1 1.41 (br s, 1 1.05 (br s, 21 H), 0. 87 J =7.5 Hz); 1 3 C NNM(125 MIL~z, CDCl 3 137.0, 124.1, 69.9, 62.3, 31.7, 31.6, 18.0, 12.3,9.3; IR (film) 3313,2970,2867;,1485, 1085, 1052 cm-1; Anal. Calcd for C,4{ 34

O

2 Si: C, 67.07; H, 11.96. Found: C, 66.89; H, 11.89; [cz] 2 D +23.2, [a] 25 5 7 7 +25.1, [aL] 2 5 546 +29.2, [a] 25 435 52.9, [a] 25 405 +66.1, (c 1.25, CHCI 3 Following the general procedure of Corey (Singh et al., supra, incorporated by reference herein) iodine (5.03 g, 19.8 mniol) was added in portions over 15 min to a 0 0 C solution of (S)-(Z)-5-triisopropylsiloxy-3-heptenol (5.17 g, 18.0 nimol), PPh 3 (5.19 g, 19.8 mmol), imidazole (1.35 g, 19.8 nimol) and Et 2 O-MeCN 135 mL) and then allowed to warm to, 23'C. After 1 h the solution was partitioned between H 2 0 (150 niL) and Et 2 O (150 mL). The aqueous phase was extracted with Et20 (2 x 150 mL). The combined organic extracts were 0 Sthen washed with Na 2

SO

3 (150 mL) and H 2 0 (150 mL), dried (MgSO 4 and filtered.

Purification of the crude product by flash chromatography (95:5 hexanes-Et 2 0) afforded 6.67 g of iodide 75 as a colorless oil: 'H NMR (500 MHz, CDCl 3 5 5.49-5.53 1 H), 5.28-5.32 1 4.41 (dd, J 7.1, 5.9 Hz, 1 3.10-3.14 2 2.59-2.66 2 H), 1.58-1.62 1 1.48-1.52 1 1.05 21 0.86 J 7.4 Hz, 3 3 C NMR S(125 MHz, CDC13) 136.2, 126.9, 70.0, 32.2, 31.6, 18.1, 12.3, 9.3, 4.4 ppm; IR (film) 3012, 2942, 1464, 1105,883 cm'; Anal. Calcd for C16H 33 0Sil: C, 48.48; H, 8.39. Found: C, 48.63; H, 8.49; [a] 25 D +22.8, [a] 25 77 +24.4, [a] 2 5 546 +23.7, [a] 25 43 +53.1, [a] 2 4 0 5 +65.8, (c 1.2, CHC13).

Synthesis of(R)-Triethvlsiloxv-N-methoxy-N-methvl-7-methvl-6-octenamide(Compound 76). To a 0 0 C solution of the known (Noyori, R. et. al. J. Am. Chem. Soc. 1987, 109, 5868) P-hydroxyester (10.0g, 53.5 mmol) in dry THF (200 mL) was added N, Odimethylhydroxylamine hydrochloride (14 g, 64.2 mmol, 1.2 eq) followed by a 2 M solution oftrimethylaluminum in toluene (60 mL, 2.3 eq) (added dropwise via cannula). The mixture was allowed to warm to room temperature and maintained at this temperature for 3 h. Then the mixture was (carefully) poored into a cold (0 0 C) 2 M solution of tartaric acid (500 mL).

The resulting mixture was stirred for 5 h, after which the layers were separated and the aqueous layer was extracted with EtOAc (3 x 100 mL). The combined organic layers were dried (MgSO 4 and concentrated. Purification of the residue on silica gel yielded 10.2 g ofWeinreb amide. The Weinreb amide (10.2 g, 47.5 mmol) was dissolved in CH 2

CI

2 (150 mL) and treated with Htinig's base (25 mL, 3eq). TESCI (8.6g, 9.7 mL, 1.2eq) was then added dropwise to the mixture. The progress of the reaction was monitored by TLC (hexanes, EtOAc, and, upon completion, the mixture was diluted with water, the layers separated and the aqueous layer extracted with Et20 (3 x 100 mL). The combined organic layers were washed with 0.5 N HCI (2 x 100 mL) and water (2 x 100 mL), dried (MgSO 4 and concentrated. The residue was purified on silica gel (hexane-EtOAc, to give 12.9 g of 76 as a pale yellow oil. 'H NMR (CDCI 3 300 MHz) 8 5.05 1H), 4.20-4.10 (m, 1H), 3.65 3H), 3.14 3H), 2.8-2.6 (dd,J= 17, 2 Hz, 1H), 2.3-2.4 (dd, J= 17, 3 Hz, 1H), 2.1-1.9 (in, 1.65 3H), 1.55 3H), 1.55-1.4 (in, 2H), 1.0-0.9 (in, 9H), 0.6-0.4 (in, z 3 C NMR(CDC 3 75 MIHz)d 172.38, 131.50, 124.10,69.10,61.17,39.61,37.96, 31.90, 25.55, 23.81, 17.56, 6.80, 5.24.

Synthesis of (6R. 1 -1Z' 13S)-2-Methvl-6-triethylsilyloxv-13-trisopropylsiloxvpentadeca-2.11-djen-8-one t-Buli (23.5 mLI 40.0 mmol, 1.7 M) was added to a -78*C solution of iodide 75 (6.67 g, 16.8 mmol) riand Et 2 O0-hexanes 1, 100 mL). The solution was maintained at -7 8*C for 30 min, then a solution amide 76 (6.10 g, 18. 5 nimol) and Et 2 O-hexanes 40 mL) was added. The 71 10 resulting solution was maintained at -78*C for 30 min then allowed to warm to 0 0 C and maintainedfor 2 h. The solution was then added to saturated aqueous NH 4 Cl (150 mL). The phases were separated, and the aqueous phase was extracted with Et 2 O (2 x 150 inL). The combined organic extracts were dried (Mg9SO 4 filtered and concentrated. Purification of the crude product by flash chromatography (98:2 hexanes-Et 2 O) afforded 5.93 g of 77 as a clear oil. The product was ca. 95% pure and was used without further purification. A small sample was further purified by flash chromatography (98:2 hexanes-Et 2 O) to obtain the following data: 'H NMIR (400 MiHz, CDCl 3 8 5.41-5.36 (in, 1 5.29-5.24 (in, 1 5.08 (tt, J= 7.1, 1.3 Hz, I 4.45 (app q,.J 6.7 Hz, 1 4.18 (quintet, J =6.0OHz, 1 2.60 (A of ABX, JAB =313 15.3, JAX =3D) 7.2 Hz, 1 2.48-2.43 (in, 3 2.30-2.24 (in, 2 H), 2.05-1.93 (mn,2 1.68 3H), 164-1.40 4H), 1.59 3H), 1.04 21 0.94 J 7.9 Hz, 9 0. 85 J 7.5 Hz, 3 0. 58 J 7.9 Hz, 6 3 C NMR (100 MI-IZ,

CDCI

3 6 208.9, 135.1, 131.8, 126.7, 123.8, 69.8, 68.7, 50.2, 44.1, 37.9, 31.6, 25.7, 23.8., 21.9,18.1, 18.0, 17.6, 12.3, 9.3, 6.9, 4.9ppm; IR (film) 2958, 2867, 1717, 1463, 1378, 1086, 1014 cm'; MS: HR.MS (FAB) m/z 37.4141 (537.4159 calcd for C 31

H-

61

O

3 Si 2 [a] 25

D

[af 5 577 [cX] 2546 (a] 2 435 +11.0, 2 s 405 14.3 (c 1.6, CHCl 3 Synthesis of liZ, 1 '.3'-dioxan-2-vl)-6-hvdroxy -mhy13 trisopropylsiloxvpentadeca-2,ll-dene A solution of ketone 77 (3.74 g, 6.94 nimol), orthoester 78 10 g, 34.7 rnmol), 1 ,3-propanediol (12.6 inL, 174 nimol), Arnberlyst- 15 resin (278 mng) and CH 3 CN (70 inL) was maintained at rt for 7 h. The mixture was then filtered through Celite and the filtrate was partitioned between Et 2 O (150 mL) and H 2 0 (50 ML). The 0phases were separated, and the organic phase was washed with H 2 0 (250 mL), dried Z (MgSO 4 filtered and concentrated. Purification of the crude product by flash chromatography (85:15 hexanes-Et 2 O) afforded 2.68 g of ketal 79 as clear oil: 'H NMvR (500 MIHz, CDCI 3 565.42-5.29 (in, 2 5.14 (broad t, J= 7.1 Hz, 1 4.45 (app q, J Hz, 1 4.11-4.08 (in, 1 4.02-3.85 (in, 4 3.80 1 2.16-1.96 (in, 6 H), __1.84-1.76 (in, 202 1.68 3 1.

6 5-1.36 (mn,6H), 1.61 3 1.05 21 0.86 (t,J =7.4 Hz, 3H); 3 NMR (125 MlHz, CDC1 3 5 134.6, 131.5, 127.5, 124.3, 101.1, 69.9, 67.0, 59.5, 59.5, 43.7, 37.5, 31.7, 31.3, 25.7, 25.2, 24.1, 22.3, 18.1, 18.0, 17.6, 12.4, 9.3 ppm; IR 19I (film) 3532, 2960, 2866, 1464, 1381, 1246, 1109 cm"; MS: FIRMS (FAB) m/z 505.3683 (505.3691 calcd for C 2 8H5 4

O

4 SiNa). Anal. Calcd for C 2 gH5 4

O

4 Si: C, 69.65; H, 11.27. Found: C, 69.40; H, 11.28; [a]I 2 5 D+1 3.3, [a] 2 5 577 +14.2, 25 546 16.8, 25 435 +30.1, 4 0 5 +3 7.4 (c 1.6, CHCI 3 Synthesis of (6S, 1iZ 13S)-6-amnino-8-(1 '3 '-dioxan-2 '-vl)-2-melhyl-13triso-propvlsiloxvyentadeca-2 11-diene Triphenylphosphine (2.89 g, 11.0 mmol) and hydrazoic acid (5.82 inL, 12.1 minol, 2.08 M in toluene) were added to a 0 0 C solution of alcohol 79 (2.65 g, 5.49 rnmol) and THEF (55 inL), then diethylazodicarboxylate (DEAD) (2.60 rnL, 16.5 inmol) was added dropwise over a period of 15 min. The solution was maintain ed at OTC for 1.5 h, then approximately half of the solvent was removed in vacuo.

The resulting solution was diluted with hexanes (30 mL) and filtered through a plug of silica gel using 97:3 hexanes-Et 2 O as the eluant. The filtrate was concentrated, and the crude product was purified by flash chromatography (97:3 hexanes-Et2O) affording 2.45 g of the azide as a clear oil: 'H NMvR (500 MiHz, CDCI 3 8 5.41-5.29 (in, 2 5. 10 (broad t, J 7.1 Hz, 1 4.47 (app q, J =7.4 Hz, 1 3.96-3.86 (mn,4H), 3.71-3.66 1 2.12-2.07 (in, 3 2.00-1.72 (in, 6 1.70 3 1.64 3 1.63-1.42 (in, 5 1.05 21 H), 0.87 J 7.4 Hz, 3 1 3 C NMv~R (125 MI-z, CDCI 3 6 134.6, 132.4, 127.7, 123.3, 99. 1, 69.9, 59.6, 59.6, 58.3, 42.2, 36.1, 32.2,31.7,25.7,25.1,24.7,22.2, 18.1, 18.1, 17.6, 12.4,9.4 ppm; IR (filmn)2961,2866,2101, 1463, 1381, 1246, 1145, 1110 cm1; MS:HFRMS (FAB) (M 3D H) ,n/z 506.3776, (506.3781 calcd for C 28

H

52

N

3

O

3 Si). Anal. Calcd for C 28

H

53

N

3

O

3 Si: C, 66.22; H, 10.52. Found: C, 66.27; H, 10.50. [a] 25 D [a] 25 5 77 +10.3, [a] 25 546 +12.1, 0 [(]25435 +24.1, [a] 25 4o0 +31.2 (c 1.6, CHCl 3 A solution of the above azide (2.45 g, 4.82 mmol) and Et 2 0 (18 mL) was added to a 0 C solution of LiAlH 4 (12.1 mL, 12.1 mmol, 1.0 M in Et 2 0) and Et 2 O (18 mL). The ice bath was removed, and the solution was allowed to warm to rt. After 1 h the reaction was quenced by sequential addition of H 2 0 (600 pL), NaOH (600 L, 3 N) and H 2 0 (1.8 mL). The resulting C1 mixture was stirred for 1 h, then MgSO 4 was added. The mixture was filtered through celite O and concentrated to afford a brown oil. Purification of the crude product by flash C 10 chromatography (10:1:0.1 CHC1 3 -MeOH-conc. NH 4 OH) afforded 2.05 g of amine as a light yellow oil: 'H NMR (500 MHz, CDCl 3 8 5.39-5.29 2 5.11 (br t, J 7.1 Hz, 1 4.46 (app q,J 7.4 Hz, 1 3.95-3.84 4 3.15-3.11 1 2.10-1.96 4 1.83-1.69 4 1.68 3 1.63-1.31 6 1.61 3 1.05 21 0.86 (t, J 7.5 Hz, 3 "C NMR (125 MHz, CDC13) 5 134.3, 131.4, 127.9, 124.1, 100.4, 69.8, 59.4,59.2,46.7,43.1,38.8,32.7,31.6,25.6,25.3,24.6,22.1, 18.0, 18.0, 17.6, 12.3,9.3 ppm; IR (film) 3387,3310,2942,2866,1464,1381,1366,1246,1109 MS: HRMS (FAB) (M H m/z 482.4011, (482.4029 calcd for C 28 Hs 6

NO

3 Si). Anal. Calcd for C28Hs 5

NO

3 Si: C, 69.80; H, 11.51. Found: C, 69.85; H, 11.56; [a] 2 5 D+21.2, [a] 2 557 7 +22.7, [a] 25 5 46 +26.1, [a] 25 4 35 +47.2, [a] 25 4 05 +58.1 (c 1.6, CHC1 3 Synthesis of (6S, 11Z. 13S)-8-(1',3 '-Dioxan-2'-vl)-2-methyl-13trisopropvlsiloxv-6-uriedopentadeca-2,11-diene Trimethylsilyl isocyanate (0.55 mL, 4.1 mmol) was added to a rt solution of 80 (1.61 g, 3.35 mmol), CH 2 Cl 2 (6.8 mL) and i-PrOH (0.31 mL). After 15 h, i-PrOH (3 mL) was added and the solution was maintained for 1 h, then concentrated. The resulting oil was purified on silica gel (100% EtOAc) to provide 1.57 g of 81 as a colorless oil: 'H NMR (400 MHz, CDC13) 5 5.24 5.36 2H), 5.03-5.15 4H), 4.41 (dd, J= 13.2, 7.1 Hz, 1H), 3.80-3.91 4H), 3.64 1H), 1.71-2.03 8H), 1.63 3H), 1.55. 3H), 1.36-1.63 6H), 1.00 21H), 0.82 J 7.4 Hz, 3H); 3

C

NMR (100 MHz, CDCl 3 159.3, 134.4, 131.8, 127.5, 123.7, 99.9, 69.7, 59.4, 59.2, 46.7, 36.9, 31.5, 31.1, 25.6, 25.0, 24.5, 22.1, 17.9, 17.8, 17.5, 12.2, 9.2 ppm; IR(film) 3354, 2960, 1660, 59 1600, 1556, 1463,,1381, 1109 cm-; [a] 25 D 5 577 +12.0, 25 s546 +17.3, [aX] 2 5 435 +20.7i 0 [Q] 2 405 +25.4, (c 1.05, CHCI 3 Anal. Calcd for C 29

H

56

N

2

O

4 Si: C, 66.36; H, 10.75; N, 5.34.

ZFound: C, 66.3 1; H, 10. 70; N5.4 1.

Synthesis of (4aR. 7S)-4-J15-(A Ilyloxvcarbonyi)DentadecvloxvcarbonvlI-1 .2.4a,5. 6. 7-hexahYdro-3-I(4S)-g ert-butvldimethvlsiloxvqentvl)I- (7S, '-dioxan-2 7 -triisoroD ylsiloxvnon-5-efl)I-1-oxo-Dyrrolo!1,=2.cI Dynmidine Osmium tetroxide (0.75 mL, 0. 1 M in t-BuOH) was added to a solution of 81 (524 mg, 1D 1.00 minol), NMO (406 mg, 3.46 nunol), and 10: 1 THIF-H 2 0 (25 mL). After 1.5 h, florisil (3 NaHSO 3 (3 and EtOAc (50 mL) were added and the reaction mixture was stirred vigouously. After 30 min, the reaction mixture was filtered, and the filtrate concentrated to provide a colorless oil which was used without further purification.

A solution of this crude diol, Pb(OAc) 4 (532 mg, 1.20 mmol), and toluene (60 mL) was maintained at room temperature. The reaction mixture was filtered through a plug of Celite,® morpholinium acetate (300 mg, 2.0 minol) was added, and the solution concentrated to provide the crude arninal 82 as a slightly yellow oil.

A solution of this crude aniinal, 47 (1.95 g, 3.36 mmol) and 2,2,2-trifluoroethanol (1.0 mL) was maintained at 60 0 C for 2 d. The reaction was quenched by adding Et 2 O (20 mL) and aqueous NH 4 CI (5 mL). The layers were separated, the organic layer was dried (MgSO 4 concentrated, and the resulting oil purified on silica gel (10: 1 hexanes-EtOAc; 7:1 hexanes-EtOAc; 3:1 hexanes-EtOAc) to provide 1.5 g of 24.and 638 mg (6 1 of a -6.5:1 mixture of 83 and 84, which was used without separation. For characterization purposes, a mng sample of this mixure was purified by HPLC (7:1 hexanes-EtOAc; Altima 5 silica). 'H NMR (500 MfHz, CDC1 3 5 6.72 lH), 5.87-5.95 (in, IH), 5.2 1-5.37 (in, 4H)j 4:-56 J =5.7 Hz, 2H), 4. 51 (dd, J= 12.7, 7.1 Hz, I1-H), 4.22 (dd, J= 11, 4.6 Hz, I 4.06-4.13(m 3H), 3.97-3.98 (in, 1H), 3.76-3.88 (in, 4H), 2.47-2.58 (in, 2.39 J 13.6 Hz, 1H), 2.32 J 7.5 Hz, 2H1), 2.26-2.32 (in, IH), 2.15 (dd, J 13.0, 6.0 Hz, lH), 1.99-2.03 (in, (N 1H), 1.50-1.90 13H), 1.41-1.48 3H), 1.11-1.40 23H), 1.10 J= 6. 1 Hz, 3H), O 0.91-1.07 21H), 0.82-0.91 12H), 0.03 3H), 0.02 3H); 1 3 C NMR (125 MHz, SCDC1 3 173.5, 166.0,151.9, 151.2, 134.3,132.2, 128.2, 118.0, 102.1,99.2, 69.9, 68.3, 64.9, 64.2, 59.3, 57.7, 52.7, 39.0, 37.4, 34.5, 34.2, 31.8, 31.3, 30.4, 29.6, 29.57, 29.5, 29.4, 29.3, 29.2, 29.1, 29.0, 26.1, 25.9, 25.3, 24.9, 24.4, 23.6, 21.8, 18.1, 12.3, 9.3, -4.7 ppm; IR (film) 3211,3095, 2927, 2856, 1741, 1682, 1627, 1463, 1435, 1107 cm' [a] 25 D [a] 25 577 [a] 25 5 4 6 [a] 25 43 -15.5, [a] 2 5 4 05 -22.7, (c 0.75, CHC1 3 Anal. Calcd for

C

5 9Hi 0 oN 209 Si 2 C, 67.77; H, 10.41; N, 2.68. Found: C, 67.68; H, 10.27; N 2.65.

(3R.4R,4aR 6'R.7S)-4-f15-(Allvloxvcarbonvl)pentadecyloxvcarbonvll-1,2,4a,5,6, 7-hexahvdro-l-oxo-7-f(7S, 5Z)-7-hydroxv-2-oxo-5-nonenyllpvrrololl.2-clpyrimidine-3-spiro-6'-(2'-methl)-3 '4 '5',6'-tetrahvdro-2H-pyran

A

solution of 83 (1.30 g, 1.24 mmol), TBAF (6.22 mL, 1.0 M solution in Et20), and DMF (31 mL) was maintained at rt for 5 h. The solution was diluted with Et20 (150 mL) and washed with H 2 0 (50 mL) and brine (250 mL). The organic layer was dried (MgSO 4 filtered, and the filtrate concentrated. The resulting residue was used without further purification.

A solution of this crude diol, TsOH'H 2 0 (236 mg, 1.24 mmol), and CHCI 3 (180 mL) was maintained at 60 0 C for 15 min. The reaction was quenched by adding saturated aqueous NaHCO 3 (20 mL). The layers were separated and the organic layer was washed with brine mL), then the organic layer was dried (MgSO 4 concentrated, and the resulting oil purified on silica gel (1:3 hexanes-EtOAc; 100% EtOAc) to provide 630 mg of a -6.5:1 mixture isomers. 62: 'H NMR (500 MHz, CDCI 3 5 5.87-5.95 1H), 5.56 1H), 5.34-5.43 2H), 5.31 (dd, J 17.2, 1.5 Hz, 1H), 5.22 (dd, J 10.6, 1.3 Hz, 1H), 4.57 (dd, J= 4.3, 1.3 Hz, 2H), 4.38 (dd, J 14.5, 6.8 Hz, 1H), 4.29-4.31 1H), 4.08-4.18 2H), 4.02 (dt, J =11.1, 4.8 Hz, 1H), 3.77-3.80 1H), 3.37 J 16.8 Hz, 1H), 2.52-2.60 (m, 2H), 2.43-2.50 1H), 2.32 J 7.5 Hz, 2H), 2.22-2.27 2H), 2.04-2.20 4H), 1.69-1.76 4H), 1.56-1.65 7H), 1.42-1.48 3H), 1.24-1.28 21H), 1.06-1.09 (m, 1H), 1.05 J 6.0 Hz, 3H), 0.89 J 7.5 Hz, 3H); 3 C NMR (125 MHz, CDC1 3 209.0, 173.5, 168.9, 153.0, 134.1, 132.3, 129.8, 118.1,82.2,68.4,66.2,65.1,64.9,55.0,54.0,53.2, 46.2, 42.7, 34.3, 32.2, 32.1, 30.3, 30.0, 29.6, 29.57, 29.5, 29.4, 29.3, 29.2, 29.1, 28.7, 26.0,

O

24.9,22.0,21.7, 18.8 ppm; IR (film) 3450,3231, 3081, 2927,2855, 1732,1715, 1659, 1651, 1470, 1373, 1262, 1013 cm'; [a]25D +42.2, 2 5 5 77 +42.7, [a] 25 546 +49.8, [a] 2 5 435 +91.0, [a] 25 4 0 5 +114, (c 0.60, CHCI 3 Anal. Calcd for C 41

H

68

N

2 0 8 C, 68.68; H, 9.56; N, 3.91.

Found: C, 68.71;H, 9.51; N 3.84.

Synthesis of (3R,4R,4aR. 6'R, 7S)-4-fl5-(Allvloxcarbonyl)pentadecyloxvcarbonvll-1,2,4a 5.6, 7-hexahyd ro-l-oxo-7-(7S. 5Z)- 7 -chloroacetox-2-oxo--nonenylpyrrolofl2-clpyrimidine-3-spiro-6'-(2'-methyl)-3',4',5',6'-tetrahydro-2H-pyran (86).

Chloroacetyl chloride (0.34 mL, 0.46 mmol) was added dropwise to a 0 0 C solution of (0.63 g, 0.88 mmol), pyridine (1.42 mL, 17.6 mmol), and CH 2 C1 2 (50 mL). The solution was immediately allowed to warm to rt. After 1 h, the solution was quenched by adding (200 mL) and washed with IN NaOH (25 mL), CuSO 4 (225 mL), and brine (25 mL). The organic layer was dried (MgSO 4 filtered, and the filtrate concentrated. The resulting residue was purified on silica gel (2:1 hexanes-EtOAc; 1:1 hexanes-EtOAc; 1:2 hexanes-EtOAc) to yield 600 mg of the desired cis isomer 86 as a colorless oil, and -85 mg of an undesired trans isomer which was derived from 84. 86: 'H NMR (500 MHz, CDCI 3 6 6.34 (s, 1H), 5.87-5.94 1H), 5.48-5.56 2H), 5.27-5.32 2H), 5.22 J 10.4 Hz, 1H), 4.56 J 5.7 Hz, 2H), 4.31-4.33 1H), 4.09-4.19 2H), 4.03 2H), 4.00-4.06 (m, IH), 3.77-3.81 1H), 3.34 J 16.6 Hz, 1H), 2.40-2.48 3H), 2.25-2.38 2.05-2.17 3H), 1.69-1.74 4H), 1.55-1.62 7H), 1.42-1.50 1H), 1.24-1.31 (m, 22H), 1.06-1.15 1H), 1.05 J= 6.0 Hz, 3H), 0.89 J= 7.5 Hz, 3H); 'C NMR (125 MHz, CDCl 3 207. 9, 173.4, 168.8, 166.6, 153.4, 133.0, 132.3, 127.8, 118.0, 82.1,73.7,66.1, 64.9, 64.8, 54.9, 53.9, 53,1, 46.3, 42.3, 41.1, 34.2, 32.2, 32.0, 29.5, 29.49, 29,4, 29.3, 29.2, 29.1, 29.07, 29.0, 28.6, 27.4, 25.9, 21.8, 21.6, 18.5, 9.3 ppm; IR (film) 3296, 2928, 2855, 1732, 1652, 1466, 1303, 1174, 1013 [a 25 D +42.7, [a] 2 5 7 7 +47.0, [a] 25 546 +52.6, [a] 25 43 +96.1, [a] 25 405 +120, (c 1.00, CHCl 3 Anal. Calcd for C 4 3

H

69

N

2 0 9 C1: C, 65.09; H, 8.77; N, 3.53. Found: C, 65.16; H, 8.79; N 3.57.

ri Synthesis ofPentacvles 87 and 88. A solution of 86 (327 mg, 0.412 mmol), MeOTf(1.29 mL, 8.21 mmol), 2,6-di-t-butylpyridine (0.46 mL, 2.1 mmol), and CH 2 C1 2 (20 mL) was Z maintained at room temperature for 8 h. The solution was then poured into Et20 (100 mL) S and washed with 1 N NaOH (2 x 10 mL) and brine (10 mL). The organic layer was dried (MgSO 4 filtered, and the filtrate concentrated. The resulting residue was used without C further purification.

C Ammonia was bubbled through a room temperature solution of the above crude pseudourea,

N

4 HCI (50 mg, 0.93 mmol), and allyl alcohol (5 mL) for 20 min (saturated solution). The 0 10 reaction vessel was sealed and heated to 60 0 C for 28 h. The reaction mixture was then cooled rt, concentrated, and the resulting oil purified by silica gel MPLC (100:0.6 CHCl 3 -i-PrOH) to provide a 147 mg of 87 and 98 mg of 88. 87 was twice recycled through the above guanylation reaction conditions to yield an additional 60 mg of 88 (52% combined yield). 87: 'H NMR (500 MHz, CDCl 3 8 8.68 1H), 8.56 1H), 5.88-5.95 1H), 5.64-5.67 (m, 1H), 5.48 J= 10.9 Hz, 1H), 5.33 (dd, J= 17.2, 1.5 Hz, 1H), 5.25 (dd, J= 10.4, 1.2 Hz, 1H), 4.57 J 5.7 Hz, 2H), 4.48 J= 10.3 Hz, 1H), 4.32-4.38 1H), 4.10-4.24 (m, 3H), 3.78-3.81 1H), 2.56-2.61 2H), 2.45 J 11.6 Hz, 1H), 2.32 J 7.6 Hz, 2H), 2.26-2.36 3H), 2.15-2.18 2H), 2.00 (dt, J= 13.8,4.7 Hz, 1H), 1.87 (dd,J= 14.6, 5.4 Hz, 1H), 1.61-1.78 10H), 1.53-1.58 1H), 1.42-1.49 1H), 1.23-1.35 22H), 1.05-1.15 1H), 1.05 J= 6.1 Hz, 3H), 0.81 J= 7.2 Hz, 3H); 3 C NMR (125 MHz,

CDCI

3 173. 5, 167.6, 147.3, 133.4, 132.3, 129.7, 118.0, 83.9, 81.7, 70.9, 67.6, 65.7, 64.9, 53.4, 53.3, 36.8, 36.2, 34.2, 31.8, 30.6, 29.7, 29.6, 29.5, 29.46, 29.4, 29.2, 29.1, 29.08, 29.0, 28.5, 25.9, 24.9, 23.6, 21.3, 17.9, 10.1 ppm; IR (film) 3268, 3147, 3020, 2927, 2855, 1732, 1660, 1608, 1465, 1283, 1164, 1029 cm'; [c] 2 5 D +12.2, [a] 2 5 57 7 +13.1, [a] 25 54 6 +14.1, (c 2.00, CHC13). HRMS (FAB) m/z 698.5108 cald for C 4 1

H

68

N

3 0 6 found 698.5096 88: 'H NMR (500 MHz, CDCI 3 5 8.54 1H), 8.43 1H), 5.88-5.95 1H), 5.64-5.67 1H), 5.48 (d,J 10.9 Hz, 1H), 5.31 (dd, J= 17.2, 1.5 Hz, 1H), 5.22 (dd,J 10.4, 1.2 Hz, 1H), 4.57 (dd, J= 5.7, 1.2 Hz, 2H), 4.48 J= 9.7 Hz, 1H), 4.29-4.33 1H), 4.08 J 6.8 Hz, 2H), 3.99-4.05 1H), 3.84-3.87 1H), 2.93 J= 4.8 Hz, 1H), 2.55-2.63 2H), 2.32 J 7.6 Hz, 2H), 2.26-2.36 2H), 2.13-2.24 3H), 1.98 (dd, J 14.7, 5.3 Hz, 1H), 1.78-1.84(in, 1H), 1.5 1-1.76(in, I1OH), 1.3871.48(in, 2H), 1.21-1.30 22H), 1.07-1.20(in, 1 1.05 J 6.1 Hz, 3H), 0. 81 J 7.2 Hz, 3H); 1 3 C NMvR (125 MI-Iz, CDCI 3 173.5, 168.3, 148.4, 133.5, 132.3, 129.7, 118.0, 84.0, 80.8, 70.9, 67.2, 65.5, 64.9; 54.1, 52.2, 49.9, 36.7, 36.1, 34.4, 31.8, 31.7, 30.5, 29.6, 29.56, 29.5, 29.45, 29.4, 29.2, 29.1,29.06, 28.4, 26.7, 25.8, 24.9, 23.6, 21.5, 17.7, 10.0 ppm; IR (film) 3263, 3154, 3025, 2928, 2854, 1732, 1652, ci1614, 1516, 1464, 1298 cin'; [a] 25 D [a] 2 5 5 7 7 -10.5, [a]"s 546 [aX] 25 435 416.5, [a] 2 40 -17.2, (c 0.75, CHC1 3 HRMS (FAB) m/z 698.5108 cal'd for C 4

IH

68

N

3 0 6 Found .698.5106 cI 10 Synthesis of Carboxylic Acid 89: A solution of 88 (27 ing, 37 i.±iol), Pd(PPh 3 4 (21 mg, 18 pmol), morpholine (13 giL, 0. 15 innol), and MeCN (1.0 m.L) was maintained at rt for 5 h.

The solution was diluted with Et 2 O (30 and washed with 0. 1 N HC1 (25 inL) and brine The organic layer was dried (MgSO 4 filtered, and the filtrate concentrated. The resulting residue was purified on silica gel (100: 1 CHCl 3 -MeOH; 33:1 CHC1 3 -MeOR) to yield 24 ing of the desired product 89 as a colorless oil: 'H NMR (500 MiHz, CDCl 3 8 5.63-5.66 (in, I1H), 5.46-5.49 (in, I 4.48 (app d, J 10.2 Hz, IRH), 4.27-4.31 (in, I1H), 4.04-4.12 (in, 2H), 3.96-4.03 (in, 11-1), 3.85-3.88 (in, I1H), 2.92 J 4.9 Hz, INH), 2.62 J -13.8 Hz, I1H), 2.55 (dd, J 12.7, 4.7 Hz, IH)J, 2.12-2.32 (in, 7H), 1. 86 (dd, J =14.8, 5.3 Hz, 1H), 1.77-1.81 (in, lH), 1.60-1.73 (in, 9H), 1.51-1.59 (in, 1.37-1.45 (in, 2H), 1.20-1.30(in, 22H), 1.16-1.20(in, 1W, 1.04(d,J= 6.1 Hz, 3W, 0.80(t,J= 7.2 Hz,3N), the NH- and OH signals are not observable; 3 C NMR (125 MI-z, CDCl 3 179.1, 168.4, 148.7, 133.6, 129.8, 83.9, 80.8, 70.8, 67.0, 65.4, 54.0, 52.0, 50.0, 36.7, 36.0, 31.9, 31.8, 30.5, 29.4, 29.3, 29.29, 29.26, 29.1, 29.0, 28.4, 26.7, 25.8, 25.5, 23.7, 21.5, 17.8, 10.0 ppm; hR (film) 3261, 3138, 2919, 2849, 1728, 1658, 1606, 1465, 1284, 1154, 1031 cin'; [a] 2 5 D -17.7, [a] 25 5 7 7 -17.0, [a]25546 -18.7, [a] 2 435 -28.5, (c 1.10, CHC 3 1-IMS (FAB) m/z 658.4795 cal'd for CHH6-L N 3 0 6 found 65 8.4791 Synthesis of 41.45-bis-t-Butoxvcarbonv Crambescidin 800 (91).

Benzotriazol- 1-yloxytris(diinethylamnino)phosphonium hexafluorophosphate (22 mg, 39 pmnol) was added to a rt solution of carboxylic acid 89 (23 mg, 33 pmol), amine 90 (18 mng, unol), Et 2 N (0.15 mL, 1.1 mmol), and CH 2 C1 2 (5 mL). After 4 h, the reaction was diluted O with Et20 (20 mL), and washed with saturated aqueous NH 4 CI (5 mL) and brine (5mL). The organic layer was dried (MgSO 4 filtered, and the filtrate was concentrated. The resulting residue was purified on silica gel (50:1 CHCl 3 -MeOH) to yield 28 mg of the desired product 91 as a colorless oil: 'H NMR (500 MHz, d4-MeOD) 5 5.70-5.73 1H), 5.47-5.52 1H), 4.40 (br d, J 10.3 Hz, 1H), 4.33-4.37 1H), 4.10-4.16 2H), 4.02-4.09 (m, 1H), 3.75-3.85 2H), 3.34-3.59 2H), 3.23-3.29 2H), 3.12-3.20 2H), 3.07 J CN 4.8 Hz, 1H), 2.94-3.06 2H), 2.64 (dd, J 13.0, 4.7 Hz, 1H), 2.26-2.46 6H), O 2.10-2.20 1H), 2.00 (dd, J= 13.9, 5.8 Hz, 1H), 1.79-1.85 3H), 1.50-1.77 11H), c 10 1.36-1.47 20H), 1.22-1.35 25H), 1.09 J 6.1 Hz, 3H), 0.85 J 6.1 Hz, 3H); 3 C NMR (125 MHz, d4-MeOD) 176.6/176.2, 170.2, 158.5, 150.2, 134.3, 131.3, 85.1, 82.2, 80.0, 79.96, 72.3, 69.1, 68.4, 68.37, 66.5, 55.6, 55.0, 54.2, 53.5, 50.7, 45.1, 38.9, 38.7, 38.3, 38.1, 37.9, 36.2, 34.3, 34.1, 33.0, 32.6, 31.5, 30.8, 30.7, 30.6, 30.5, 30.3, 30.2, 29.6, 28.8, 28.7, 27.6, 27.0, 26.7, 26.6, 24.4, 21.8, 19.5, 10.8 ppm; IR (film) 3356, 2934, 2858, 1732, 1706, 1657, 1613, 1509, 1459, 1251, 1170 cm'l; [a] 2 D [a] 2 5 577 [a] 25 5 46 -2.8, [a] 25 435 [a] 2 5 4 05 (c 0.75, CHC13). HRMS (FAB) m/z 1001.7 cald for C 55

H

9 7

N

6 01 0 found 1001.7 Synthesis ofCrambescidin 800 Trihydrochloride A solution of 91 (13 mg, 13 pmol) and 1.3 mL of a 3.0 M solution of HCI in EtOAc was maintained at rt for 20 mins and then concentrated. Purification of the residue by reverse phase HPLC (4:1 MeOH-0.1 M NaC1, Altima C 18, 5 column) gave -11.8 mg of crambescidin 800 as its trihydrochloride salt (a light yellow oil). Data for this sample were consistent with data published for natural 2.

Data for synthetic 2: 'H NMR (500 MHz, d4-MeOD) 5 5.71 1H), 5.50 (app d, J 10.9 Hz, 1H), 4.41 1H), 4.33 1H), 4.13 1H), 4.05 1H), 3.96 1H), 3.85 1H), 3.65 2H), 3.55 1H), 3.44 2H), 3.11 2H), 3.07 J 4.8 Hz, 1H), 2.97 (m, 2.88 1.5H), 2.64 (dd, J 12.8, 4.7 Hz, 1H), 2.23-2.51 7H), 2.17 1H), 1.50-2.10 15H), 1.42 12.2 Hz, 1H), 1.20-1.40(m, 25H), 1.09(d, J= 6. 1 Hz, 3H), 0.85 J 7.2 Hz, 3H); 'H NMR (500 MHz, CDCl 3 6 9.74 1H), 9.50 1H), 8.00 (br s, 6H), 5.67 (app s, 1H), 5.47 (app d, J 10.4 Hz, 1H), 4.49 1H), 4.28 1H), 4.07 (m, 2H), 3.97 2H), 3.45-3.66 3H), 3.29 2H), 3.11 2H), 2.95 2H), 2.55 (m, 1 IH), 2.10-2.50 7H), 2.05 1H), 1.95 1H), 1.50-1.70 15H), 1.40-1.50 2H), Z 1.20-1.40 25H), 1.05 (d,J 5.4 Hz, 3H), 0.83 (t,J 6.6 Hz, 3H); 1 3 C NMR(125 MHz, Sd4-MeOD) 177.5, 170.2, 150.2, 134.3, 131.3, 85.1, 82.1, 72.3, 69.4, 68.4, 66.5, 55.6, 54.8, 54.1, 50.7, 43.8, 38.5, 38.3, 38.2, 37.9, 33.0, 32.9, 32.6, 31.5, 30.8, 30.7, 30.67, 30.6, 30.3, 30.26, 29.6, 27.6, 27.0, 26.6,26.55,24.4,21.8, 19.4, 10.8 ppm; 3 C NMR (125 MHz, CDCI 3 175.5/175.0, 168.3, 148.7, 133.6, 129.8, 83.5, 80.6, 70.9, 67.1, 65.4, 54.4, 53.9, 51.8, 49.5, 44.0,37.9, 37.6, 37.0, 36.9, 33.5, 33.2, 32.0, 31.8, 30.8, 30.6, 29.7, 29.6, 29.59, 29.55, 29.5, 29.1, 29.0, 28.4, 26.9, 25.8, 25.5, 25.4, 23.4, 21.4, 18.3, 10.1 ppm; IR (film) 3382, 3231, 2923,2852,1732,1659,1614,1469,1167,1086,1015 [a]255 7 7 [a]25 5 46 [a] 2 5 4 35 [a] 25 405 (c 0.70, CHCI 3 HRMS (FAB) m/z 801.6217 cald for

C

4 5

H

8 siN 6 0 6 found 801.6222 Synthesis of PeracetvIlcrambescidin 800 A solution of crambescidin 800 (5.0 mg, 5.5 pig), Ac 2 0O (0.5 mL), and pyridine (1 mL) was maintained at rt for 23 then concentrated in vacuo (0.9 mm Hg, 23°C), diluted with CHCI 3 (20 mL) and washed with 0.1 M HC1 (5 mL), and brine (5 mL). The organic layer was dried (MgSO 4 filtered, and the filtrate was concentrated. The resulting residue was purified on silica gel (20:1 CHCI 3 -MeOH; 10:1

CHCI

3 -MeOH) to yield 2 mg ofperacetylcrambescidin 800 (92) as a white wax. Data for this sample were consistent with data published for naturally derived peracetylcrambescidin 800. Data for synthetic 92: 1H NMIR (500 MHz, CDC13) 5 9.88 1H), 9.64 1H), 6.75 (br s, 0.7H), 6.38 (br s, 0.7H), 6.18 (br s, 0.7H), 5.67 (app t, J 10.5 Hz, 1iH), 5.49 (app d, J 11.0 Hz, 1H), 5.10 1H), 4.51 1H), 4.28 (app dt,J 9.8,4.9 Hz, 1H), 4.12 1H), 4.06 IH), 4.04 2H), 3.46-3.64 2H), 3.20-3.39 4H), 2.99-3.16 2H), 2.94 J 4.6 Hz, 1H), 2.55 (dd, J 12.6, 4.6 Hz, 1H), 2.49 1H), 2.17-2.38 7H), 2.05 3H), 2.00 3H), 1.98 3H), 1.92-196 1H), 1.52-1.82 (m, 14H), 1.43 J 12.2 Hz, 1H), 1.20-1.40 25H), 1.05 J 6.1 Hz, 3H), 0.83 J 7.2 Hz, 3H); 'H NMR (500 MHz, d4-MeOH) 5 5.72 (app t, J 10.9 Hz, 1H), 5.51 (app d, J 11.0 Hz, 1H), 5.15 1H), 4.32-4.39 2H), 4.13 (dt, J 6.6, 1.8 Hz, 2H), 4.07 1H), 3.83 1H), 3.39-3.63 4H), 3.14-3.25 4H), 3.08 J 4.9 Hz, 1H), 2.64 (dd, J 13.0,4.8 Hz, 1H), 2.29-2.47 7H), 2.17 1H), 2.02 3H), 2.01 1H), 1.94 0 1.93 1.5H), 1.92 1.5H), 1.91 1.5H), 1.53-1.86 14H), 1.42 J= 12.6 Hz, 1H), 1.20-1.40 25H), 1.09 J= 6.2 Hz, 3H), 0.85 J 7.2 Hz, 3H); 1 3 C NMR (125 MHz,

CDC

3 174.5, 171.0, 170.8, 170.5, 170.4, 168.3, 148.8, 133.7, 129.8, 83.6, 80.7, 71.0, 70.6, 67.3, 65.4, 53.9, 51.8, 50.5, 49.7, 42.6, 37.0, 36.96, 35.9, 35.3, 33.2, 32.4, 32.0, 30.6, 29.7, mc, 29.6, 29.5, 29.4,29.1,28.5,27.1,26.8, 25.8,25.5,23.5,21.4, 18.4, 10.1 ppm; IR (film) 3457, 3240, 2923, 1739, 1732, 1660, 1643, 1614, 1463, 1372, 1238, 1015 cm'; [a] 25 D -37 (c 0.2, C CHCI 3 HRMS (FAB) m/z 927.6534 cald for C 51

H

87

N

6 0 9 found 927.6547 C 10 These results demonstrate the first total synthesis of Crambescidin 800 (compound 2) in a convergent fashion with the longest linear sequence from commercially available starting material in 25 steps and in a 3.0% overall yield. These experiments confirm the stereochemical assignment of Crambescidin 800 (compound 2) and rigorously establish that the absolute configuration of it hydroxysperimidine side chain is S.

EXAMPLE IV Initial Method for enantioselective total synthesis of 13,14,15-Isocrambescidin 800 Synthesis Plan. Molecular mechanics models of the methyl esters of the 13,14,15- Isocrambescidin 800 and Ptilomycalin A/Crambescidin are shown in Figure 20. These models clearly illustrate the structural differences and similarities between the two structures. For instance, the C 10 and C 13 angular hydrogens are trans in the Isocrambescidin core and cis in the Ptilomycalin A/Crambescidin core, but the relationship between the substituents at C13, C 14 and C 15 is the same in both structures. Also, the C-O bonds in both structures are axial.

Thus, as in the Ptilomycalin A synthesis, we surmised that the C8 and C 15 spirocenters could be constructed with the desired stereochemistry if the required trans stereochemistry of the central triazacenaphthalene ring system was in place. This strategy would require setting the trans relationship of the C10 and C 13 angular hydrogens and relating this chirality to the C3 and C 19 stereocenters of the oxopene and hydropyran rings.

O The initial synthetic challenge was constructing the triazacenaphthalene ring system with the desired trans stereochemistry. In the (-)-Ptilomycalin A synthesis the required cis stereochemistry was established via a "tethered Biginelli" condensation of a ureido aldehyde and a P-ketoester. (See Overman, L. Rabinowitz, M. H. J. Org. Chem. 1993, 58, 3235m 3237; Overman, L. Rabinowitz, M. Renhowe P. A. J. Am. Chem. Soc. 1995, 117, S2657-2658 and Kappe, C. O. Tetrahedron 1993, 49, 6937-6963).

"1 O Further studies in our laboratories revealed that a Biginelli condensation between a tethered C 1 10 guanyl aldehyde and a p-ketoester afforded 1-iminohexahydropyrrolo[1,2-c]pyrimidine intermediates with the trans relationship about the pyrrolidine ring (Figure 30) (McDonald, A. Overman, L. E. J. Org. Chem. 1999, 64, 1520-1528) thus providing a strategy for constructing the Isocrambescidin core.

A retrosynthetic analysis of the Isocrambescidin core is shown in Figure 21. Disconnection of the C8 and C15 aminals of 94 leads to a 1-iminohexahydropyrrolo[1,2-c]pyrimidine intermediate such as 95. With the guanidine unit in place, we envisaged formation of the final three rings of the pentacyclic core directly from an intermediate such as 95. We further envisaged 95 being formed via a Biginelli condensation of guanyl aldehyde 96 and a Pketoester such as 97. As in the (-)-Ptilomycalin A synthesis, this strategy is very attractive since it is highly convergent. In order for this synthesis to be completed in this convergent fashion, however, the guanidine unit"ould have to be introduced very early in the synthesis.

Early installation of the guanidine is attractive in that further manipulations to install the guanidine at a later stage can be avoided, but since we did not want to protect the guanidine, we were forced to deal with this highly polar functionality for several steps of the synthesis (Figure 21).

Results and Discussion 0 Synthesis of 13,14,15-Isocrambescidin 800. The synthesis began with amine 98, which had also been utilized in the second generation synthesis of (-)-Ptilomycalin A and the synthesis of Crambescidin 800 (See Figure 22). Treatment of 98 with 1-H-pyrazole-1 -carboxamidine Shydrochloride (Bernatowicz et al., J. Org. Chem. 1992, 57:2497-2502) and Hunig's base at 0 C afforded guanidine 99 (ca 99% yield) which was used without purification (Figure 22).

C Before examining the key Biginelli condensation, the trisubstituted olefin had to be 8 selectively oxidized. In the event, treatment of 99 with catalytic osmium tetroxide (OsO4) C1 10 resulted in selective dihydroxylation of the trisubstituted double bond. (Sharpless and Williams, Tetrahedron Lett. The corresponding diol was cleaved with Pb(OAc) 4 in the presence of morpholinium acetate affording aminal 100. Following cleavage of the diol, we found it was optimal if 100 was filtered, and used immediately in the Biginelli condensation, and not exposed to an aqueous workup. Furthermore, 100 is actually a mixture of several components as judged by 'H and "C NMR. Multiple signals are observed for many carbon atoms in the 3 C NMR spectra, while broad peaks are seen in the 'H NMR spectra and no aldehyde signal is apparent (Figure 22).

Biginelli condensation of 100 and P-ketoester 101 (Overman et al., J. Am. Chem. Soc., 1995, 117:2657-2658) in EtOH proceeded with modest selectivity (3:1 trans:cis). Fortunately, we found that heating 100 (1 equiv) and 101 (1.5 equiv) in 2,2,2-trifluoroethanol at 60 0 C for 20 h improved the diastereoselectivity to 7:1 trans:cis. After purification on silica gel deactivated with pH 7.0 buffer. The deactivated silica was prepared by taking Merck silica gel (0.040- 0.063) adding 10% (by weight) pH 7.0 buffer and mixing until homogeneous. The desired trans adduct 102 was isolated in 49% yield and cis adduct 103 in ca. 5% yield. The cis adduct 103 was slower moving on silica get than trans adduct 102. Thus, it was somewhat difficult to isolate pure 103 since some of trans adduct 102 would trail off of the column. The stereochemistry of trans adduct 102 was initially assigned based on results from previous model studies (McDonald et al. J. Org. Chem 1999, 64, 1520-1528). This assignment was more rigorously established using pentacyclic intermediates (see 105a and 105b) produced

O

N later in the synthesis. Deprotection of 102 with TBAF in DMF for 36 h afforded diol 104 in yield (Figure 23). Often, this reaction did not go to completion, and partially deprotected intermediates

(R

2 TBS, R 3 H) were isolated in 10-15% yield. Heating the reaction mixture at 60°C consistently afforded fully deprotected 104, but other products were formed and the isolated yield of 104 was lower.

c N A. Initial Stereochemical Assignment of pentacycle 105a Diol 104 was now properly functionalized for conversion to the isocrambescidin pentacyclic S 10 core. In the event, 104 was exposed to p-toluenesulfonic acid (p-TsOH) (3 equiv) in CHCl 3 for 24 h. The reaction mixture was then washed with aqueous

HCO

2 Na affording an approximate 1:1 mixture of the desired pentacycle 105a and isomerized pentacycle 106 (pentacycle 20 is an approximate 1:1 mixture of diastereomers, and the structure was assigned based on 'H NMR COSY studies) in ca 50% yield (Figure 23). Since the tosylate couterion interconverted slowly, several washings were required to obtain the formate salts exclusively, and we were concerned that the multiple washings may result in lower yields for this transformation. The formate salts were prepared to allow direct comparisons to pentacycle 107 (Figure 24), an intermediate in the first generation synthesis of (-)-Ptilomycalin

A

(Overman et al. J. Am. Chem. Soc. 1995, 117, 2657-2658). 'HNMR NOE studies confirm that 105a is epimeric at C13, C14 and C15 with 107. Thus, trans diaxial addition of the side chain alcohol to the double bond during the conversion of 10 4 -105a orients the hydropyran in the Isocrambescidin core as in the Ptilomycalin A/Crambescidin core. Furthermore, that 105a is epimeric with 13 ,1 4 ,15-Isocrambescidin 800 at C14 was signaled by the 11.8 Hz coupling constant of the C14 methine hydrogen (McDonald et al. J. Org. Chem 1999, 64, 1520-1528) (See Figures 23 and 24 [Scheme 3, Figure Further studies revealed that the ratio of compounds 105a and 106 could be controlled by varying the reaction times and equivalents ofp-TsOH. Larger amounts ofp-TsOH and longer reaction times favored the formation of 106, and we found that treating pure 105a with p- TsOH also afforded 106. Conditions were found (2 equivp-TsOH,

CHCI

3 7 h) that gave a 5:1 mixture of 105a and 106. Unfortunately, we could not find conditions usingp-TsOH that gave 105a exclusively, and since 105a and 106 were somewhat difficult to separate, the Z isolated yields of 105a were never greater than 45-50%. Thus, a better method for the preparation of 105a was required.

Treatment of 104 with pyridiniump-tolunesulfonate (PPTS) (2 equiv) at 60 0 C in CHCI 3 for 7 h followed by a HCO 2 Na wash afforded a 1:5 mixture of the desired 105a and tetracycle 108a (Figure 25). Slight modifications of the reaction conditions (2 equiv PPTS, CHCI 3 90°C in a sealed tube, 24 however, afforded a 2:1 mixture of 105a and 108a (Figure Unfortunately, we could not find conditions using PPTS to convert 104 completely to 105a, but 105a and 108a could be separated by column chromatography. After separation, resubjection of 108a to the reaction conditions afforded, again, a 2:1 mixture of 105a and 108a. The desired 105a could be isolated in 75% combined yield after one recycle. Initially, for comparative purposes, 105a and 108a were converted to the formate salts before chromatography, and they were separated using 95:5:0.1 EtOAc-isopropanol-formic acid. We later found, however, that the hydrochloride salts, 105b and 108b, were easier to separate than the corresponding formate salts (105a and 108a). The hydrochloride salts were prepared by washing the reaction mixture with 0.1 N HCI or saturated aqueous sodium chloride and separated on silica gel using 99:1 CHCl 3 -MeOH--95:5 CHCI 3 -MeOH. As in previous cases, several washings were required to completely convert the tosylate salt to the hydrochloride salt. Since the formate and hydrochloride salts could be obtained in virtually identical yields, the ease of separation made use of the hydrochloride salts optimal (Figure 25). Reagents used were PPTS, HCI 3 90 0 C 24 h; HCO 2 Na wash or 0.1 N HCI wash While pentacycles 105a and 105b could be prepared in synthetically useful yields, the sequence was somewhat cumbersome. Ideally, we needed to find conditions that did not promote the formation of isomerization products such as 106 but would afford complete conversion to 105a or 105b. Toward this end, we found that treatment of 104 with HCI (3 equiv) in EtOAc afforded 105b exclusively in 78% yield. Since the hydrochloride salt was formed during the reaction, the somewhat troublesome counterion exchange was avoided Nc (Figure 26).

0 Z Epimerization to the axial ester at C 14 was best accomplished after removal of the allyl group Sof the terminal ester. To this end, the allyl group in 105b was removed with (Ph 3

P)

4 Pd and morpholine (Figure 26). The C14 ester was then epimerized with Et 3 N in MeOH at affording a 2-3:1 mixture of the desired P-epimer, 109, and a mixture of the starting a-epimer and a product analogous to 108a resulting from elimination of the oxygen at C15. After S purification by flash chromatography, 109 was isolated in 50-60% yield for the two steps. A diagnostic 3.0 Hz coupling constant is observed for the C14 methine hydrogen of 109. In contrast to precursors of(-)-Ptilomycalin A and Crambescidin 800, the axial ester is favored at equilibrium in the Isocrambescidin series.

The synthesis of 13,14,15-Isocrambescidin 800 was readily completed from 109 (Figure 26).

The (S)-7-hydroxyspermidine fragment 110 was prepared from (R)-epichlorohydrin and coupled with pentacycle 109 using benzotriazol-l-yloxytris(dimethylamino)phosphonium hexafluorophosphate (BOP) (Castro et al. Tetrahedron Lett. 1975, 1219-22) to provide 111 in 71% yield. Removal of the BOC groups with 2 M HCI in ethyl acetate (Stahl et al. J. Org.

Chem. 1978, 43, 2285-2286) and purification of the crude product by reverse-phase HPLC provided the trihydrochloride salt of 13,14,15-Isocrambescidin 800 [a]D 23 -67.7 (c 0.7 MeOH), in 70% yield. The data for the trihydrochloride salt of synthetic 10 is in agreement with 'H and "C NMR data reported for natural 10. The 'H NMR spectrum (500 MHz, of synthetic dl trihydrochloride is identical to the spectrum of natural 10 published in the Supplementary Materials of references 3b (Spectrum Spectrum S-3 is apparently also of the tricationic salt. There are errors in the tabulated data in Table 1 of reference 3b, since there are discrepancies between the tabulated data and Spectrum S-3. The 3 C NMR spectrum of synthetic 10 trihydrochloride is also identical to the spectrum of natural published in the Supplementary Material of reference 3b (Spectrum S-8a). Again, there are errors in the tabulated data in Table 1 of reference 3b, since there are discrepancies between the tabulated data and Spectrum S-8a. Synthetic 10 was also converted to the triacetate derivative, 112. Data for synthetic 112 is also in agreement with 'H and 3 C NMR data reported for 112 prepared from natural 10 (Jares-Erijman, et al J. Org. Chem. 1993 58: 0 4805).

The trihydrochloride salt of 10 was believed to be obtained, since a basic workup was not performed after the removal of the BOC groups (111 10). Natural 10, however, was c reported with the amines as the free bases, but the 'H and "C NMR spectra of synthetic and natural 10 were indistinguishable. Treatment of synthetic 10 with 0.1 N NaOH saturated with NaCI resulted in downfield shifts of the C41 and C45 protons. To investigate this O further, 114 was prepared from acid 113 to model the hydroxyspermidine region of 13,14,15- C 10 Isocrambescidin 800 (Figure 27). Deprotection of 114 with 2.0 M HCI in EtOAc afforded 115 in 95% yield. The chemical shifts of the C37-C45 protons were identical in 115 and [C41, 2.99-2.84 ppm C45, 3.14-3.08 ppm (Table but the absence of the guanidine unit made the spectral analysis of 115 somewhat easier. Treatment of 115 with 0.1 N NaOH afforded 116 as the free base. There were significant upfield shifts of the C41 and C45 protons in 116 [C41, 2.66-2.58 ppm C45, 2.86-2.78 ppm This upfield shift is consistent with conversion of a hydrochloride salt to a free base. Therefore, this supports the conclusion that natural 13 ,1 4 ,15-Isocrambescidin 800 was actually isolated as the trihydrochloride salt (Figure 27).

TABLE 1 Comparison of the Chemical Shifts of the C41 and C45 Protons of Compounds 115 and 116 'H NMR (CDOD, 500 MHz), (5 ppm),mult Position 115 116 41 2.99-2.84, m 2.6-2.58, m 3.14-3.08, m 2.86-2.78, m Assignment of the C43 Stereocenter ofl3,14,15-Isocrambescidin 800. The C43 stereocenter z in 13,14,15-Isocrambescidin 800 was assigned as S based on analogy to crambescidin 816 S(compound Our completed total synthesis of compound 10 seemed to confirm this assignment, but, since the C43 stereocenter is far removed the other stereocenters, we were not certain that (43S)-13,14,15-Isocrambescidin 800 could be spectroscopically distinguished m from (43R)-13,14,15-Isocrambescidin 800. To this end (43R)-13,14,15-Isocrambescidin 800 S(117) was prepared from 109 and ent-110. Ent 110 was prepared from (S)-epichlorohydrin S(Figure 28). As anticipated, 117 was indistinguishable from synthetic 10 and natural 10 by 0 'H and "C NMR and HPLC.

N ID To unambiguously make this stereochemical assignment, we anticipated having to compare a derivative of natural 10 to derivatives of synthetic 10 and 117. The preparation of Mosher's derivatives of synthetic 10, 117 and natural 10 and subsequent analysis by 9 F NMR was an appealing option. Mosher's derivatives 118 (43S) and 118 (43R) were prepared according to the method developed by Ward (Figure 29). The corresponding Mosher's derivative of natural 10 was then prepared and compared to 118 and 119. The 9 F NMR data is identical for the Mosher's derivatives prepared from natural 10 and synthetic 10. Due to rotamers about the C38 amide, there are 6 peaks in the 9 F NMR spectra (See Figure 27). This unambiguously establishes that the stereochemistry at C43 in 1 3 ,1 4 ,15-Isocrambescidin 800 is S.

The first total synthesis of 13,14,-15-Isocrambescidin 800 (10) was accomplished in convergent fashion with the longest linear sequence from commercially available material being 21 steps. These investigations confirm the stereochemical assignment of 10 and rigorously establish that the absolute configuration of its hydroxyspermidine side chain is S.

In its present form, this synthesis can provide substantial quantities of 10 and congeners for pharmacological evaluation. This enantioselective total synthesis demonstrates for the first time that: the tethered Biginelli strategy can be extended to guandidine intermediates, (b) the key Biginelli condensation can be accomplished under sufficiently mild conditions that fragments containing the full functionality of the Crambescidin core can be employed, and (c) c that the spiroaminal units in the Isocrambescidin series assemble with high stereochemical fidelity.

Experimental Section GeneraL Dry THF, Et20, and CH 2 C1 2 from Aldrich were filtered through a column charged I with A1 2 0 3 (solvent purification system). Triethylamine (Et 3 pyridine, Sdiisopropylethylamine, diisopropylamine, and acetonitrile were distilled from CaH 2 at atmospheric pressure. Silica gel (0.040-0.063) by Merck was used for flash chromatography.

S 10 Reverse phase -HPLC separations were performed using an HPLC system composed of a Waters 590 pump and a Waters 486 UV detector. NMR spectra were recorded on Bruker instruments (500 MHz and 400 MHz). IR spectra were measured on Perkin-Elmer Series 1600 FTIR, and optical rotations were measured on Jasco DIP-360 polarimeter. Mass spectra were measured on a MicroMass Analytical 7070E (CI-isobutane) or a MicroMass AutoSpec E (FAB) spectrometer. Microanalyses were performed by Atlantic Microlabs, Atlanta, GA.

Other general experimental details have been described (Metais, E. et al. J. Org. Chem. 1997, 62, 9210-9216).

(6S, liZ, 13 S)-6-amino-N-carboxamidine-8-( '-dioxan-2 '-l)-2-methyl-13trisopropvlsiloxpentadeca-2,11-diene A solution of amine 98 (2.95 g, 6.12 mmol), 1- H-pyrazole-1-carboxamidine hydrochloride (2.70 g, 18.4 mmol), Hunig's base (4.37 mL, 24.5 mmol) and DMF (6.0 mL) was maintained at rt for 16 h then warmed to 60 0 C and maintained for 4 h. The solution was cooled and partitioned between CHCl 3 (300 mL) and 0.1 N HCI mL). The organic phase was washed with 0.1 N HCI (75 mL) and H 2 0 (75 mL), dried (Na 2

SO

4 filtered and concentrated affording a 2:1 mixture of guanidine 99 and amine 98.

This mixture was dissolved in DMF (6.0 mL) and treated with 1-H-pyrazole-1 -carboxamidine hydrochloride (1.35 g, 9.2 mmol) and Hunig's base (2.19 mL, 12.3 mmol). The solution was maintained at room temperature (rt) for 16 h then warmed to 60 0 C and maintained for 4 h.

The reaction was worked up as previously described and placed on a vacuum pump (0.1 mm Hg) to remove the residual DMF. This afforded 3.20 g of guanidine 99 as a light yellow oil which was used without further purification: 'H NMv~R (500 MI-z, CDCI 3 5 7.82 z (app d, J= 6.7 H-z, I 7.24 (br s, 1 5.43-5.39 (in, I 5.29-5.24 (mn, 1 5.09 (br t, J Hz, 1 4.46 (app q, J= 7.3 Hz, I 3.98-3.76 (in, 4 3.60 (in, I 2.20-2.13 (in, 2 2.02-1.74 (overlapping mn, 2 1.69 3 1.64-1 .58 (overlapping m, 2 1.62 3 1. 51-1.3 8 (in, 2 1. 05 (in, 21 0. 87 J =7.4 H-z, 3 3 C NMv~R (12 5 MHz,

CDGI

3 8157.6, 135.0, 132.7, 126.9, 123..1, 100.5, 69.8, 59.8, 59.3, 46.6, 45.0, 36.5, 31.7, 3 0.5, 2 5.7, 2 5.0, 24.8, 22.2, 18.1, 18.0, 17.6, 12.3, 9.3 ppm; IIR(filmn) 2961, 2865, 16 51, 1463, 1383, 1246, 1109 cm"; MS: HRMS (FAB) (M Cl) m/z 524.4225, (524.4250 calcd for

C

27 H58N 3

O

3 Si); [a] 25 D+1.7, [a] 25 5 77 [a]1 5 546 [a] 25 435 [a] 25 405 +9.3 (c L.3, ID CHCI 3 (4aS. 7 S)-4-ll5-(Allvloxycarbonv)DventadecyloxvcarbonyII..3-ff4S)-4-.

butvldimethylsioxvpentvll- '-dioxan-2 7 1-1,2,4a, 5,6, 7 -hexahydro-1-imino-gyrrolofl,2.c..pvrjmidine hydrochloride (102). 4- Methylinorpholine-N-oxide (2.16 g, 18.4 rnmol) and 0S0 4 (3.1 mEL, 0.24 ininol, 2% in tbutanol) were added to a solution of guanidine 99 (3.2 g, -6.1 nol), TI-F (105 m.L) and

H

2 0 (15 inL). The mixture was stirred for 8 h, Florisil (1.5 g) and NaHSO 3 (1.5 g) were added and the resulting mixture was stirred for an additional 10 h. Celite and MgSO 4 were added and the mixture was filtered and the eluent was concentrated to give a brown oil. This oil was dissolved in toluene (120 inL), then inorpholinium acetate (3.6g, 24.5 inmol) and Pb(OAc) 4 (3.3 g, 7.3 innol) were added. The solution was maintained for 45 min, then Celite was added. The mixture was filtered through a plug of Celite, the eluent was diluted with toluene (200 mE) and this solution was concentrated to give a brown oil. The oil was azeotroped to dryness with toluene (200 mE) and the residue was combined with P-ketoester 15 (5.3 g, 9.2 minol) and 2,2,2-trifluoroethanol (9.0 This solution was maintained at 60 0 C for 20 h and then partitioned between CHC1 3 (250 mE) and 0. 1 N HCI (50 mE). The organic phase was wa shed with 0. 1 N HCI (50 m.L) and brine (50 dried (Na 2 SOA) filtered and concentrated. 'H NMR analysis revealed a 7:1 ratio of trans:cis Biginelli adducts. Purification of the crude mixture by flash chromatography (CHCl 3 99:1 CHCI 3 -MeOH 98:2 CH.Cl MeOH) using silica gel deactivated with pH 7.0 buffer afforded 3.22 g of the desired trans adduct, 102, as a light brown oil and 331 mng of cis adduct 103. Data for 102: 'H NMvR (500 M4Hz, CDCI 3 89.06 I 7.33 11-1), 5.95-5.88 (in, 1 5.43 (app t, 1= 9.8 ZHz, 1 5.31 (app dq, J= 17.2, 1.5 Hz, 1 5.27-5.25 (in, 1 5.23 (appdq, J= 10.4, 1.3 I 4.5 7 (br d, J 5.7, 2 4.46-4. 41 (in, 2 4.27-4.24 (in, 1 4.17-4.07 (in, 2 4.01-3.95 (in, 2 3.91-3.78 (in 3 2.77-2.71 (mn, 2 2.65-2.59 (in, 1.14), 2.45-2.40 (in, I 2.32 J =7.6 Hz, 2 2.07-1188 6 1.79-1.55 (mn, 11I 1.53-1.43 (m 4 1.31-1.25 (in, 21 1. 13 J= 6.1 Hz, 3 1.05 21 0.87 J= 7.4 Hz, 3 H), 0.86 9 0.037 3 0.032 3 3 C NMv~R (100 MI-Iz, CDCI 3 6173.4, 165.0, 149.9, 147.3, 135.3, 132.2, 126.4, 117.9, 100.9, 100.3, 69.8, 68.3, 64.8, 64.7, 59.9, 59.4, 57.5, S 10 54.1, 46.1,39.0, 34.8, 34.2, 33.3,31.6,30.9,30.3,29.6,29.52,29.48, 29.42,29.3, 29.1,29.0, 28.5, 26.0, 25.8, 24.83, 24.76, 24.4, 23.6, 22.1, 18.01, 17.98, 12.3, 9.2, -4.7 ppm; 1k (film) 2926, 2856, 1738, 1713, 1681, 1538, 1462,1382,1256, 1086 cm"; HvMS(FAB)v(M- CI) "n/z 1044.6, (1044.8 calcd for C 59 H, joNASi 2 [a] 25 D-21 .2 [aj 25 s5 77 -21.3, [a] 5 546 -23.3, [a] 2 5 435 -28.8, [a] 25 405 -25.1 (c 1.9, CHCl 3 (4aS. 7

S)-

4 -lS-(Alloxvcarbonv1)Dentadecyloxvcarbonvi; 7-(SZ, 7S)-2-(1,3 '-dioxan-2 vi)- 7-h vdroxv-5-nonen vlI-1 .2.4a, 5.6. 7-he~xahvdro-3-f(4 -yroxvenvII-J-imino- Pvrrololl,2,cI-fyrimidine hydrochloride (104). A solu tion of 102 (2.80 g, 2.59 inmol), TBAF (13.0 mL, 13.0 iniol, 1.0 M) and DMF (26 m.L) was maintained at rt for 24 h, then more TBAF (6.0 m.L, 6.0 inmol, 1.0 M) was added. The solution was maintained for 6 h then partitioned between CHC1 3 (200 mL) and 0. 1 N HCI (75 mL). The organic phase was washed with saturated aqueous NaHCO 2 (Wx5 inL), dried (Na2SO 4 filtered and concentrated. The crude product was purified by flash chromatography (95:5:0.1 EtOAc-isopropanol formic acid 90:10:0.1 EtOAc-isopropan6irinic acid 85:15:0.1 EtOAicopropanol: formic acid) using silica gel deactivated with pH 7.0 buffer to afford the formnate salt of the diol 1 .68g as a light brown oil. The formate salt was easier to purify, but the chloride salt was more stable. Therefore, after purification, the forinate salt was converted quantitatively to chloride salt 104 by partitioning the formate salt between CHC1 3 (150 mQL and 0. 1 N HCI mEL) and washing with 0. 1 N HCI (25 mE) and brine (25 mL). The organic phase was dried (Na 2

SO

4 filtered and concetrated to afford diol 104: 'H NMR (500 MIL~z, CDCI 3 5 8.63 I N, 7.43 1 5.95-5.87 1 5.51-5.42 2 5.31 (ddd,J= 17.2, 3.0, 1.5 Hz, 1 5.22 (ddd, J= 9.2, 3.0, 1.3 Hz, 1 4.57 (dt, J= 5.7, 1.3 Hz, 2 4.43 (dd, J= 9.9,4.3 Z Hz, 1 4.32 (app q, J =7.1 Hz, 1 4.28-4.25 1 4.17-4.08 2 4.05-3.92 (m, 3 3.89-3.82 2 2.91-2.86 1 2.62-2.58 1 2.52 (td,J= 11.8,4.6 Hz, 1 2.42-2.39 1 2.32 J= 7.6 Hz, 2 2.16-1.96 6 1.86-1.44 14 H), 1.30-1.24 22 1.19(d, J= 6.2 Hz, 3 0.91 J= 7.4 Hz, 3 3 C NMR(125 MHz, c CDCI 3 8 173.5, 165.0, 149.7, 147.5, 133.5, 132.3, 130.4, 118.0, 101.0, 100.5, 68.7, 65.4, S 64.85, 64.76, 60.1, 59.6, 57.6, 54.2, 45.8, 38.1, 34.7, 34.2, 33.1, 30.4, 30.2, 29.6, 29.51, 29.46, 29.37, 29.2, 29.1, 28.6, 26.0, 24.9, 24.7, 24.0, 23.5, 22.2, 9.7 ppm; IR (film) 3344, 2925, 2854, 1736, 1685, 1542,1462,1384,1259,1170,1084, 1001 cm'; MS: HRMS (FAB) (M Cl) m/z 774.5615, (774.5632 calcd for C44H 7 6

N

3 Og); [a] 2 5 D-39.4, [a] 25 5 77 -40.2, [a] 2 5 46 44.8, [a] 2 5 4 35 -66.0, [a]2 405 -70.0 (c 1.2, CHCI 3 Pentacycle 105b. Acetyl chloride (320 jL, 4.5 mmol) was added to a 0°C solution of MeOH (200 pL, 5.0 mmol) and EtOAc (30 mL) to give a 0.15 M solution ofHCl in EtOAc. Diol 104 (1.10 g, 1.36 mmol) was then dissolved in 27 mL of this solution (4.1 mmol of HCl). The solution was maintained at rt for 6 h, then partitioned between CHCl 3 (250 mL) and brine mL). The organic phase was dried (Na 2

SO

4 filtered and concentrated. Purification of the residue by flash chromatography

(CHCI

3 99:1 CH MeOH 98:2 CHCI 3 -MeOH) gave 780 mg of pentacycle 105b as a light yellow oil. In some instances, pentacycle 105b was contamintated with ca. 5% of an unidentified impurity. This impurity could be removed by further purification by reverse phase HPLC, but the recovery of the desired pentacycle, 105b, was low. Therefore, pentacycle 105b was not purified further, and the unknown impurity was removed after the next transformation.

Data for pure 105b: 'H NMR (500 MHz, CDC 3 8 10.45 (br s, 1 8.89 (br s, 1 5.95- 5.87 1 5.68-5.64 1 5.48 (broad d, J= 11.0 Hz, 1 5.31 (app dd, J= 17.2, Hz, 1 5.23 (app dd, J= 10.4,1.3 Hz, 1 4.57 (brd,J= 5.7 Hz, 2 H),4.51 (br d,J= 7.7 Hz, 1 4.25-4.21 2 4.12-4.07 1 3.98-3.95 1 3.77-3.72 1 H), 2.91 J= 11.8 Hz, 1 2.61-2.56 1 2.55 (dd, J= 12.5, 2.9 Hz, 1 2.32 (t,J= Hz, 2 2.30-2.28 3H), 2.21-2.17 2H), 1.91 (dd,J= 14.6, 5.3 Hz, 1 1.85 (br 0 d, J= 12.9 Hz, I 1.78-1.36 13 1.32-1.20 21 1.12 J= 6.0 Hz, 3 H), z 1.12-1.10 1 0.86 7.3 Hz, 3 1' 3 C NMR (125 MHz, CDCI 3 8 173.5, 169.0, S150.8, 133.3, 132.3, 129.8, 118.0, 85.5, 84.8, 70.8, 68.7, 65.5, 64.8, 58.7, 55.1, 52.1, 37.2, 37.1, 34.2, 32.9, 32.1, 30.9, 30.0, 29.6, 29.51, 29.45, 29.4, 29.2, 29.11, 29.07, 28.4, 25.9, mr 24.9,23.8,22.0, 17.9, 10.2 ppm;IR (film) 2926; 2853, 1732,1659,1615,1462,1349,1202, r^1022 cm'; HRMS (FAB) (M-CI) m/z 698.5117, (698.5108 calcd for C 41

H

68

N

3 0 6 [a] 25

D

(N 54.6, [a] 25 77 -55.6, [a]25546 -64.2, [a] 2 5 435 -114.8, [a] 2 5 40 5 -141.3 (c 1.25, CHCI 3 S 10 Carboxylic Acid 109. A solution ofpentacycle 105b (50 mg, 0.068 mmol), morpholine (24 0.27 mmol), (Ph 3

P)

4 Pd (16 mg, 0.014 mmol) and CH 3 CN (5 mL) was maintained for 2 h.

More morpholine (12 p.L, 0.13 mmol) and (Ph 3

P)

4 Pd (8 mg, 0.007 mmol) were added and the solution was maintained for 2 h. The solution was then partitioned between CHCI 3 (50 mL) and 0.1 N HCI (10 mL). The organic phase was washed with 0.1 N HCI (10 mL), dried (Na 2

SO

4 filtered and concentrated to give a brown oil. The brown oil was filtered through a plug of silica gel (99:1 CHC1 3 :MeOH 98:2 CHC1eOH), concentrated and dissolved in Et 3 N (95 gL, 0.68 mmol) and MeOH (7 mL). The resulting solution was maintained at for 36 h then partitioned between CHCI 3 (50 mL) and 0.1 N HCI (8 mL). The organic phase was washed with 0.1 N HCI (8 mL), dried (Na 2

SO

4 filtered and concentrated. Purification of the crude product by flash chromatography (99:1 CHCI 3 :MeOH 98:2 CHCI 3 -MeOH 95:5 CHCl 3 :MeOH) afforded 28 mg of 109 as a light yellow oil: 'H NMR (500 MH-z,

CDCI

3 810.07(brs, 1 9.28 (br s, 1 5.64 (app t, J= 8.1 Hz, 1 5.50(d, J=10.6Hz, 1 4.58 (br s, 1 4.17-4.12 1 4.02-3.97 2 3.92-3.88 1 3.71- 3.68(m, 1 3.46 3.0 Hz, 1 2.63-2.55 1 2.52 (d,J=l 1.0 Hz, 1 2.30 (t, J= 7.4 Hz, 2 2.29-2.26 1 2 2 2 -2.16(m, 3 1.85-1.80 4H), 1.73-1.42 11 1.40-1.24 23 1.18 J= 5.9 Hz, 3 0.95 J= 7.2 Hz, 3 3 C NMR (125 MHz, CDCI 3 5178.6, 167.8, 149.4, 133.7, 129.6, 85.0, 83.0, 70.7, 69.0, 65.2, 52.9, 52.1, 41.7, 37.9, 37.4, 35.1,32.5,31.5, 30.2, 29.7, 29.4, 29.33, 29.29, 29.19,29.0,28.4,27.9,25.7, 25.3, 24.1,22.2, 19.7, 10.2 ppm; IR (film) 3200, 2924, 2852, 1732, 1660, 1621, 1189, 1167, CK11027 cm-1; HRMS (FAB) (M-CI) m/z 658.4789, (658.4795 calcd for C 3 sI-14N 3

O

6 [a] 25 o47.3, [a] 25 577 -49.5, [a] 2 '1 46 -55.9, [a] 25 435 -99.8, [a] 25 405 -121.5 (c 1.2, Cl-Id 3 41,45-bis-t-Butoxcarbonvl-13 14.1 5-Isocrambescidin 800 (111). Benzotriazol. 1yloxytris(dimethylamino)phosphonium hexafluorophosphate (28 mg, 0.064 mniol) was added to a solution of carboxylic acid 109 (30 mg, 0.043 mmol), amine 110 (23 mg, 0.064 mmol), Et3N (29 p1, 0.22 mmol) and CH 2 C1 2 (2.0 mL). The solution was maintained for 1 h and then was partitioned between Et 2 O (40 mL) and 0.1 N HCI (10 mL). The organic phase was washed with brine (2x 10 dried (MgSOA) filtered and concentrated to afford a crude oil.

Purification of this residue by flash chromatography (99:1 CHCI 3 -MeOH 97:3 CHC 13- MeOH) gave 32 mg (7 of Il1 as a colorless foam: 'H NMvR(500NMH,

CD

3 OD)585.70 (br t, J 8.8 Hz, I 5.51 J =11. 1 Hz, I 4.45 (br s, 1 4 .1 9 4 .06 3 3.92- 3.78 (in, 3 3.84 (dJ= 3.4 Hz, 1 3.59-3.23 (in, 3 3.19-3.12 (mn, 3 3.06-2.97 (in, 2 2.5 8 (dd, J1 12.8, 2.3 Hz, 1 2.45-2.25 (mn, 6 2.18-2.12 (in, I 1.96 (dd,J 13.1, 6.1 H~z, 1 M, 1.81-1.44 (in, 18 1.43 18 1.38-1.17 (mn,23 1. 16 J= Hz, 3 0.95 J= 7.3 Hz, 3 1 3 C NMvR (125 M4Hz, CD3OD)(the C3 8 amide exists as an approximate 1: 1 mixture of rotamers). Some of the signals of carbons in close proximity to C38, including the carbons of the hydroxyperinidine unit, are doubled. These signals are listed in parentheses: 5(176.6/176.2), 169.8, 158.6, 158.4, 150.2, 134.1, 131.3, 86.7, 84.6, 80.02, 79.96, 72.0, 70.1, (69.0/68.3), 66.2, (55.0/53.4), 54.8, 54.3, 45.0, 42.6, 39.1, (38.9/38.7), 38.1, 36.2, 34.3,34.1, 33.7, 32.9, 31.0, 30.78, 30.75, 30.67, 30.64, 30.57, 30.54, 30.50, 30.24, 30.16, 29.7, 28.9, 28.8, 28.7, 27.0, (26.7/26.6), 25.0, 22.4, 21.0, 10.8 ppm; IR (film) 3385, 2927, 2854, 1731, 1668 1614, 1449, 1366, 1253, 1167, 1028 H-RMS (FAB) m/z 1001.7 (1001.7 calcd for C 55

H

97

N

6 01 0 [a] 2 h,-68.7, [a] 22 577 -72.9, [a] 2 1 2 5 46 -83.3, [aX] 22 435 -147.7 (c 0.6, CHC1).

l3,J4,15-Isocrambescidin 800 Trihydrochloride A solution of 111 (30 mg, 0.029 minol) and 2.9 mL of a 2.0 M solution of HCl in EtOAc was maintained at rt for 30 min and then concentrated. Purification of the residue by reverse phase HPLC (3.5:1 MeOH-0. I M NaC, Altiina C 18, 5 column) gave 18 mg of l 3 ,l 4 ,IS-isocrambescidin 800 as its trihydrochloride salt (a light yellow oil). Data for this sample were consistent with data >published for natural 0 Data for synthetic 10: 'H NMR (500 MHz, CD 3 OD) 6 5.70 (br t, J= 9.1 Hz, 1 5.51 (br d, J= 11.2, 1 4.46 (br s, 1 4 .18- 4 .06 3 3.97-3.86 3 3.84 J= 3.3 Hz, 1 3.70-3.38 3 3.31-3.07 3 2.99-2.84 2 2.57 (dd, J= 12.9, 2.4 Hz, 1 C 2.56-2.36 5 2.31-2.24 3 2.18-1.43 18 1.28 (app s, 22 1.16-1.15 (overlapping m, 1 1.16 J= 6.0 Hz, 3 0.95 J= 7.4 Hz, 3 "C NMR (125 S MH-z, CD30D) (the C38 amide exists as an approximate 3:1 mixture ofrotamers. Only the carbons in close proximity to C38, including some carbons of the hydroxyspermidine unit, exhibit signals due to the minor rotamer. The carbon signals of the rotamers are listed in parentheses with the major rotamer listed first: 5(177.5/176.4), 169.8, 150.2, 134.1, 131.3, 86.8, 84.6, 72.0, 70.2, (68.6/69.4), 66.2, (54.9/53.2), 54.8, 54.3, (43.9/47.9), 42.7, 39.1, 38.6 (two peaks), (38.27,38.33), 38.1, (34.2/34.0). 33.7, 32.9 (2 peaks), 31.0, 30.84, 30.80, 30.75, 30.67, 30.59, 30.54, 30.4, 30.3, 29.7, 28.9, 27.0, (26.63/26.57), 25.0, 22.5, 21.0, 10.8 ppm; MS: HRMS (FAB) m/z 801.6223 (M Cl), (801.6217 calcd for C 45

H

81

N

6 0 6 [act 22 -67.7, [a]22577 -70.9, [a] 2 2 5 4 6 -80.6, [a] 22 435 -147.7 (c 0.73, MeOH).

Peracetvl-13,14,15-Isocrambescidin 800 Hydrochloride (112). A solution of 13,14,15isocrambescidin 800 acetic anhydride (1.2 mL) and pyridine (2.4 mL) was maintained at rt for 20 then concentrated using a vacuum pump. The resulting residue was dissolved in

CHCI

3 (40 mL).and washed sequentially with brine (10 mL), 0.1 N HCI (10 mL) and brine mL). The solution was dried (Na 2

SO

4 filtered and concentrated. Purification of the residue by flash chromatography (95:5 CHCI 3 -MeOH) gave 8 mg of Peracetylisocrambescidin 800 (112). 'H NMR and 13C NMR data for synthetic 112 was in agreement with data reported for Peracetylioscrambescidin 800 prepared from natural 1 3 ,1 4 ,15-Isocrambescidin 800.

Data for synthetic 112: 'H NMR (500 MHz, CDCI 3 810.0 1 9.97 1 H of minor rotamer), 9.23 1 9.19 I H of minor rotamer), 6.86 1 6.70 1 H of minor rotamer), 6.57 1 H of minor rotamer), 6.40 1 5.66 (br t, J= 8.7 Hz, 1 5.50 (br d, 0 J= 11.0 Hz, 1 5.13-5. 07 1 4.55 (br s, 1 4.19-4.13 1 4.02-3.97 2 z 3.91-3.88 1 3.72-3.69 1 3.62-3.37 4 3.46 J= 2.8 Hz, 1 H), 3.32-3.12 3 3.05-2.98 1 2.55-2.52 2 2.37-2.18 7 [2.05, 2.04, 2.01, 2.00, 1.99 (singlets of the acetate methyl groups, 9 2.00-1.37 18 1.28-1.18 Cn 23 1.18 J= 6.0 Hz, 3 0.96 J= 7.3 Hz, 3 'C NMR (125 MHz, CDC1 3 8 174.4, 173.8 (minor rotamer), 170.9, 170.8, 170.7, 167.7, 149.3, 133.6, 129.7, 85.0, 82.9, N 70.8, 70.5 (minor rotamer), 65.3, 52.9, 52.1, 50.5, 48.4 (minor rotamer), 46.4 (minor 0 rotamer), 42.6, 41.73, 41.68 (minor rotamer), 38.2, 37.4, 37.0, 36.1, 35.6, 35.4, 33.2, 33.1, 32.9, 32.3, 31.4, 30.2, 29.6, 29.5, 29.4, 29.1, 29.0, 28.5, 27.9, 27.0, 25.8, 25.5, 25.4, 24.0, 23.2,22.1,21.2, 20.9, 19.6, 10.2 ppm; MS: LRMS (ES) m/z 927.70 (M Cl), (927.6534 calcd for C 5 IHs 7

N

6 0 9 [c] 25 D-56.1 (c 0.3, CHCI 3 EXAMPLE V Synthesis Plan. The structural differences and similarities between the two Crambescidin families are apparent in molecular mechanics models of the methyl esters of the 13,14,15- Isocrambescidin and Crambescidin/Ptilomycalin A pentacyclic guanidine moieties (Figure The lowest energy conformation found from Monte Carlo searches using Macromodel version 5.5 and the OPLS force field is depicted (Chang, Guida, W. Still, W. C. J. Am.

Chem. Soc. 1989, 111, 4379-4386). Ten thousand starting conformations were examined; in all cases, several conformations that differ only in the spatial orientation of the methyl ester fragment were within a few kcals of the global minimum. For instance, the C 10 and C 13 angular hydrogens are trans in the isocrambescidin core and cis in the corresponding Crambescidin/Ptilomycalin A unit, while the stereochemical relationship between the substituents at C13, C14 and C15 is the same in both structures. For both alkaloid families, the C-O bonds of the hydropyran and oxepene units are axial. Thus, as in the Crambescidin/Ptilomycalin A series (Snider, B. Shi, Z. Tetrahedron Lett. 1993,34, 2099- 2102; Snider, B. Shi, Z. J. Am. Chem. Soc. 1994, 116, 549-557; Overman, L. E.; Rabinowitz, M. Renhowe P. A. J. Am. Chem. Soc. 1995, 117, 2657-2658;), it was anticipated that the C8 and C15 spirocenters of the Isocrambescidins would evolve with the desired stereochemistry if the central triazaacenaphthalene ring system was constructed with

O

z the proper trans stereochemistry.

An intramolecular variant of the venerable Biginelli condensation that was introduced several years ago (Overman, L. Rabinowitz, M. H. J. Org. Chem. 1993, 58, 3235-3237) has proven to be highly useful in the design of concise strategies for synthesizing complex guanidine alkaloids. As detailed in Example I, tethered Biginelli condensation of a ureido aldehyde and a P-ketoester can be employed to combine all the carbons of the Crambescidin/Ptilomycalin A pentacyclic core and set the pivotal cis relationship of the and HI13 hydrogens (Kappe, C. O. Tetrahedron 1993, 49, 6937-6963). Recent exploratory studies of stereoselection in tethered Biginelli condensations were critical in the planning on how to synthesize the Isocrambescidin alkaloids (McDonald, A. Overman, L. E. J. Org.

Chem. 1999, 64, 1520-1528). These investigations revealed that the stereochemical outcome of tethered Biginelli condensations could be reversed if the urea component was replaced with a basic guanidine. Thus, Biginelli condensation of guanidine aldehyde (or aminal) 122 with benzyl acetoacetate provided trans-1 -iminohexahydropyrrolo[ 1,2-c]pyrimidine 123 with high selectivity (Figure 32) (McDonald, A. Overman, L. E. J. Org. Chem. 1999, 64, 1520- 1528).

Based on these exploratory studies and the experience in the Crambescidin/Ptilomycalin

A

series, a convergent plan for preparing 1 3 ,14,15-Isocrambescidin 800 (10) readily emerged (Figure 33). Tethered Biginelli condensation of guanidine aldehyde 126 and P-ketoester 127 would be employed to set the critical trans C10-C13 stereorelationship and unite all the heavy atoms of the pentacyclic guanidine moiety. It was hoped that acid promoted dehydration of 125 would then generate the remaining three heterocyclic rings of 124 in a single step. Mindful from the outset of one challenge posed by this strategy: the guanidine functional group would be introduced early in the synthesis. Unless protection and deprotection steps are added, this highly polar functionality would be forced to carry through several stages of the synthesis.

>Results and Discussion 0 Synthesis oftrans-l-Iminohexahvdropvrrolofl,2-clvrimidine 134. The total syntheses of 10 and lOa began with diene amine 128, which was also utilized in the synthesis of Crambescidin 800 (Figure 34). Treatment of 128 with l-H-pyrazole-l-carboxamidine Shydrochloride (Bematowicz, M. Wu, Matsueda, G. R. J. Org. Chem. 1992, 57, 2497- 2502) and diisopropylethylamine at 60 0 C generated guanidine 129, which was utilized directly without purification. The trisubstituted double bond of this intermediate next had to be cleaved to liberate the electrophilic component of the Biginelli condensation. Fortunately, the oxidation strategy that was employed to realize this degradation in the related urea series was compatible with the guanidine functionality. Thus, selective dihydroxylation of the trisubstituted double bond of 129 with catalytic osmium tetroxide (Os0 4 and Nmethylmorpholine-N-oxide (NMO) (Sharpless, K. Williams, D. R. Tetrahedron Lett.

1975, 3045-3046), followed by cleavage of the resulting diol with Pb(OAc) 4 in the presence of morpholinium acetate provided 130. This intermediate was purified only by filtration to remove PbO 2 and was a mixture of several components as judged by 'H and 3 C NMR analysis. Multiple signals were observed for many carbon atoms in 3 C NMR spectra of 130 and 'H NMR spectra showed several broad peaks; no aldehyde signal was apparent.

Biginelli condensation of crude 130 and P-ketoester 131 in EtOH at 60 0 C proceeded with modest trans selectivity Fortunately, it was found that heating 130 with 1.5 equiv of 131 in 2,2,2-trifluoroethanol at 60 0 C for 20 h improved diastereoselection to 7:1. After purification of the crude products on silica gel deactivated with pH 7.0 buffer (deactivated silica was prepared by adding 10% (by weight) pH 7.0 phosphate buffer to Merck silica gel (0.040-0.063 A) and mixing until homogeneous), the desired trans adduct 132 was isolated in 48% yield and cis adduct 133 in ca. 5% yield (since the cis adduct 133 was slower moving on silica gel than 132, it was difficult to isolate pure 133). The stereochemistry of 132 was provisionally assigned based on the earlier exploratory studies (McDonald, A. Overman, L.

E. J. Org. Chem. 1999, 64, 1520-1528). As seen shortly, this assignment could be confirmed 0 C rigorously at a later stage. Deprotection of 132 with tetra-n-butylammonium fluoride (TBAF) O in N,N-dimethylformamide (DMF) at room temperature for 36 h gave rise to diol 134 in Z yield. In some runs, this reaction did not go to completion and intermediates in which only 0\ the TIPS group had been removed were isolated in 10-15% yield. Heating the reaction mixture at 60 0 C avoided this complication, however, other unidentified products were formed and the isolated yield of 134 was not improved.

r Cyclization to Form Pentacvcle 135. Initially guanidine diol 134 was exposed at room temperature to 3 equiv ofp-toluenesulfonic acid monohydrate (p-TsOH-H 2 0) in CHCl 3 for 24 h (Figure 35). After washing the reaction mixture with aqueous HCO 2 Na, a 1:1 mixture of a pentacyclic product, subsequently shown to be 135a, and tetrahydrofuryl isomer 136a were isolated in ca. 50% yield (exchange of the tosylate couter ion for formate required several washings with aqueous sodium formate, which led to some erosion in yield).

The constitution of these pentacyclic products was ascertained as follows. The gross structure of pentacycle 136a, a -1:1 mixture of stereoisomers at the center carrying the 1-butenyl side chain, was secured by 'H NMR COSY and "C NMR studies. The stereochemistry of 136a at (the crambescidin numbering system is employed in the discussion of synthetic intermediates; correct IUPAC names and numbering can be found in the Experimental Section) followed from the chemical shift of the C14 methine hydrogen (6 2.88) (The C14 methine hydrogen of 135 is observed at 5 2.91, while this hydrogen of 139 is occurs at 6 2.30.

The C 15 stereochemistry of these products was rigorously determined (vide infra)), while the stereochemistry at C8 was not determined and is assigned on the basis of analogy only.

Pentacyclic guanidines 135a and 136a were isolated as their formate salts to allow direct comparisons with pentacycle 137, an intermediate in the original synthesis of Ptilomycalin A (Example I and Overman, L. Rabinowitz, M. Renhowe P. A. J. Am.

Chem. Soc. 1995, 117, 2657-2658). That 135a was epimeric to 137 at C13 was signaled by the absence of an 'H NMR NOE between H10 and H13 in the former, while the 11.7 Hz coupling constant of the C 14 methine hydrogen of 1353a showed that the ester side chain was equatorial.

Since none of the pentacyclic guanidine intermediates or products prepared during the Z investigations were crystalline, 'H NMR NOE studies proved indispensable in assigning stereochemistry. A molecular mechanics model of the guanidine moiety of 135a (an intermediate having the 13-Epicrambescidin core), which helped in analysis of critical NOE enhancements, is provided in Figure 31 (the lowest energy conformation found from Monte C, Carlo searches using Macromodel version 5.5 and the OPLS force field is depicted. Ten c thousand starting conformations were examined; in all cases, several conformations that differ only in the spatial orientation of the methyl ester fragment were within a few kcals of the global minimum. As discussed later in the text, the conformation of the 13,15- Isocrambescidin core depicted in Figure 31 is undoubtedly not the lowest energy one). Also provided in Figure 31 are models of the two additional guanidine pentacycles (13,14,15isocrambescidin and 13,15-epicrambescidin ring systems) and, for reference, a model of the Crambescidin/Ptilomycalin A pentacyclic guanidine moiety.

Additional investigation revealed that formation of tetrahydrofuran isomer 136a from 134 could be controlled by varying reaction time and equivalents of p-TsOH*H 2 0. Larger amounts of acid and longer reaction times favored the formation of 136a. Exposing 135a top- TsOH*H 2 0 at room temperature for extended periods also led to 136a. The best conditions found for generating 135a involved exposing 134 to 2 equiv ofp-TsOH'H 2 0 in CHCI 3 for 7 h at room temperature; a 5:1 mixture of 135a and 136a was produced. Since these isomers were difficult to separate, the isolated yield of 135a produced in this way was never greater than Pyridinium p-toluenesulfonate (PPTS) was examined to cleave the 1,3-dioxane protecting group of 134 and promote cyclization of the resulting keto guanidine diol. With this weaker acid, higher reaction temperatures were required and mixtures of 135a, tetracyclic vinylogous carbamate 138a and several unidentified minor byproducts were produced (Figure 36). When 134 was heated with 2 equiv of PPTS at 60 0 C in CHCI 3 for 5 h and the crude product was washed with aqueous HCO 2 Na, 135a and 138a were generated in a 1:5 ratio. Increasing the c, reaction temperature to 90 0 C (sealed tube) for 24 h provided 135a and 138a in a 2:1 ratio (a small amount, -10% relative to 138a, of the formate analog of 139 was also produced. When Z 138a was heated with PPTS at 90 0 C, 135a and 138a were formed also in a -2:1 ratio).

Separation of these products on silica gel, followed by resubjection of 138a to PPTS at 90 0

C

gave 135a in 75% combined yield.

N Initially, 135a and 138a were converted to their formate salts prior to chromatography and C were eluted from deactivated silica gel using 95:5:0.1 EtOAc-isopropanol-formic acid. It was later found that the hydrochloride salts, 135b and 138b, were easier to separate on silica gel.

O 10 These salts were prepared by washing the reaction mixture with 0.1 M HCI or saturated aqueous sodium chloride; several washings were required to completely exchange the tosylate counter ion.

Since both NH hydrogens were readily apparent in 'H NMR spectra of 135b, extensive NMR studies COSY, HMQC, HMBC and NOESY) eventually revealed that 135b had thel3- Epicrambescidin stereochemistry the spiro hydropyran and ester side chain are both epimeric to those of 10 and 10a). Key findings were diagnostic 'H NMR NOEs observed between N2H and H19, N2H and H17(axial), and H13 and H16(axial); see the modelof the 13-Epicrambescidin core in Figure 31. It is shown in the following discussion, the stereochemistry of both the spiro hydropyran and ester side chain can be readily inverted, allowing 135b to be a viable intermediate for accessing Isocrambescidins 10 and Although the procedures just described provided pentacyclic guanidine salts 135 in synthetically useful yields, these sequences were cumbersome. Ideally, it is needed to find acidic conditions for cyclizing 134 that would not promote allylic rearrangement of the C3 alcohol, yet would irreversibly transform tetracyclic vinylogous carbamate intermediate 138 to a pentacyclic guanidine isomer. It was eventually discovered that treatment of 134 with 3 equiv ofHCl in EtOAc at room temperature delivered 135b in 78% yield (Figure 37). Careful purification of the crude cyclization product by reverse phase HPLC (9:1 MeOH-0.1 M NaCI) afforded, in addition to 135b, 5-7% of pentacyclic guanidine 139.

That 139 was epimeric to the Isocrambescidins only at C14 (ester side chain) was apparent Z from 'H NMR COSY, HMQC, HMBC and NOESY experiments. The stereochemistry at S C15 followed directly from diagnostic 'H NMR NOEs observed between N2H and the H17(axial) and N2H and H20, and the lack of NOE between N2H and H19. This NOE data is consistent with the hydropyran ring of 139 preferentially adopting a chair conformation having the methyl substituent axial (Figure 38, conformation This conformational Spreference undoubtedly derives from two factors: In the alternate hydropyran chair conformer, destabilizing syn pentane interactions would exist between C 17 and C 19 of the hydropyran ring and the carbonyl carbon of the ester group; for two views of this conformation, see Figure 38, conformation B and the model of the 13,15-Epicrambescidin core in Figure 31. Conformer A would be stabilized by an anomeric interaction between the hydropyran oxygen and the C15-N2 bond (Kirby, A. J. Stereoelectronic Effects; Oxford University Press: Oxford, 1995; pp 3-24; Kirby, A. J. The Anomeric Effect and Related Stereoelectronic Effects at Oxygen; Springer: Berlin, 1983; Deslongchamps,

P.

Stereoelectronic Effects in Organic Chemistry; Pergamon: Oxford, 1983.).

To gain more insight into the mechanism of hydropyran formation, pure 135b was resubjected to the cyclization conditions (3 equiv HCI in EtOAc at room temperature) to yield an approximate 8-9:1 mixture of 135b and 139 (Figure 37). That this represents an equilibrium ratio of the C15 epimers under these conditions was established by: (a) demonstrating that the 8-9:1 mixture.of 135b and 139 was unchanged when resubjected to the reaction conditions for an additional 24 h, and showing that pure 139 gives an identical ratio of epimers when exposed for 24 h to 3 equiv HCI in EtOAc. Since no intermediates or by-products having the ester side chain on the P face were detected in HCIpromoted cyclization of 134, or HCl-promoted equilibrations of the spiro hydropyran epimers, it was surmised that the equilibration of 135b and 139 did not involve tetracyclic intermediates such as 138. Consistent with this hypothesis, exposure of 139 to DCI in EtOAc gave an approximate 8-9:1 mixture of 135b and 139 without incorporation of deuterium into 135b (Figure 39). Iminium cation 140 is the likely intermediate in the equilibration of the

O

C, spiro hydropyran epimers (although not rigorously precluded, the alternate possibility that epimerization at C15 occurs by cleavage of the N2-C15 bond to form a six-membered Z oxocarbenium ion intermediate to be less likely). It was concluded from these studies that formation of 135b as the major product from HCl-promoted cyclization of 134 arises from kinetically-controlled axial protonation of the vinylogous carbamate unit of 134 to generate the protio equivalent of 140, which undergoes thermodynamically-controlled spirocyclization to generate 135b preferentially (in our syntheses of Ptilomycalin A and Crambescidin 800, Sthe only spirohydrofuran products formed from acid-promoted cyclization of related vinylogous carbamates have the oxygen axial. In those cases, equilibration of hydropyran epimers by a pathway related to that depicted in Figure 39 would occur at slower rates since less stable N-acyliminium cations would be involved).

Epimerization of 135b at C14 and C15 to Give Pentacyclic Guanidine Acid 141 and Total Synthesis ofl3,14,15-isocrambescidin 657 Not long after 135a was first prepared, it was established that exposure of this intermediate to Et 3 N in hot methanol provided a pentacyclic guanidine whose stereochemistry was identical to that of 13,14,15- Isocrambescidin 800 Although this fact was not initially appreciated, epimerization at C14 and C15 is a coupled event. This reorganization was best accomplished after removal of the allyl'group of the hexadecanoate ester. To this end, the 8-9:1 mixture of 135b and 139 resulting from HCl-promoted cyclization of 134 was deprotected with (Ph 3

P)

4 Pd and morpholine (Figure 40). The resulting mixture of acids was then epimerized by heating in MeOH at 60 0 C in the presence of 10 equiv of Et 3 N. Acidification of this product with 0.1 M HCI yielded a mixture of pentacyclic guanidine acids 141 and 142 and tetracyclic guanidine 143 in an approximate ratio of 10-14:1:1 (the ratio of 141 to (142 and 143) was determined from the crude product mixture by 'H NMR analysis at 500 MHz. Due to the complexity of this spectrum and the presence of minor impurities, it was estimated that this ratio is only accurate to The ratio of 142:143 was more difficult to ascertain, although these products appeared to be formed in similar amounts. Attempts to resolve this mixture by HPLC were unsuccessful). The pentacyclic guanidine acid resulting from deallylation of 139 was not detected. After purification by flash chromatography on silica gel, 141, which exhibits a diagnostic 3.3 Hz coupling constant for the equatorial C 14 methine hydrogen, was 0 isolated in 50-60% yield for the two steps. A similar mixture of products was obtained when z pure samples of 135b or 139 were individually deallylated and heated with Et 3 N in MeOH.

In contrast to precursors of (-)-Ptilomycalin A and Crambescidin 800 (Overman, L.

Rabinowitz, M. Renhowe P. A. J. Am. Chem. Soc. 1995, 117, 2657-2658), the axial ester is highly favored in the Isocrambescidin series.

The structure of 141 was secured by extensive 'H NMR COSY, HMQC, HMBC and NOESY experiments. The stereochemistry of 141 at C15 followed from diagnostic 'H NMR NOEs observed between H19 and H14 and between H19 and H13 (weaker), and the absence of NOEs between N2H and H19 (see the 3-dimensional model of the 13, 14 core in Figure Carboxylic acid 141 was quantitatively converted to the corresponding inner salt by washing with dilute NaOH. This product showed 'H and 'C NMR data fully consistent with those reported (Kashman, Hirsh, McConnell, O. Ohtani, Kusumi, T; Kakisawa, H. J. Am. Chem. Soc. 1989, 111, 8 925-8926) for 13 ,1 4 ,15-Isocrambescidin 657 The specific rotation of synthetic 10a was [a] 23 D -35.4 (c 0.8 MeOH), which agrees well with the specific rotation, [a] 23 D -32.7 (c 0.3 MeOH), reported (Kashman, Hirsh, S.; McConnell, O. Ohtani, Kusumi, T; Kakisawa, H. J. Am. Chem. Soc. 1989, 111, 8925- 8926) for natural 13,14,15-Isocrambescidin 657 (10a). Complete assignments of the 'H and 13C chemical shifts of 10a and 141 are provided.

Since a pure sample of 138b was available from our earlier studies of the cyclization of 134 with PPTS, this tetracyclic guanidine was deallylated to form 143 (Figure 41). Exposure of 143 to Et 3 N and MeOH at 60 0 C provided a product mixture containing 141, 142 and 143 in an approximate 12:1:1 ratio. As in the related conversions of 135b and 139, the acid congener of 139 was not detected. The experiment summarized in Figure 41 provides permissive evidence for the intermediacy of 143 in the epimerization of 135b at C14 and Cl 5 to provide 141.

Total Synthesis of 13,14,15-isocrambescidin 800 The 7 -hydroxyspermidine Sfragment 144, which is available from (R)-epichlorohydrin (Coffey, D. McDonald, A. I.; Overman, L. E. J. Org. Chem. 1999, 64, 8741-8742), was coupled to pentacyclic acid 141 using benzotriazol-l-yloxytris(dimethylamino)phosphonium hexafluorophosphate

(BOP)

(Castro, Dormoy, J. Evin, Selve, C. Tetrahedron Lett. 1975, 1219-1222) to provide 145 in 71% yield. Removal of the BOC protecting groups with 2 M HCI in ethyl Sacetate (Stahl, G. Walter, Smith, C. W. J. Org. Chem. 1978, 43, 2285-2286) and purification of the crude product by reverse-phase HPLC gave the trihydrochloride salt of 0 1 3 ,14,15-isocrambescidin 800 [a]o 23 -67.7 (c 0.7 MeOH), in 70% yield. A specific N 10 rotation of [a] 23 D -48 (c 0.5 MeOH) is reported for natural 1 3 ,14,15-Isocrambescidin 800 (Jares-Erijman, E. Ingrum, A. Carney, J. Rinehart, K. Sakai, R. J. Org.

Chem. 1993, 58, 4805-4808). Since the counter ion of natural 10 was not described, the significance, if any, of this discrepancy in rotation magnitude is unknown. NMR data for the trihydrochloride salt of synthetic 10 were in good agreement with those reported for natural 10, (Jares-Erijman, E. Ingrum, A. Carney, J. Rinehart, K. Sakai, R. J. Org.

Chem. 1993, 58, 4805-4808). The trihydrochloride salt of 10 was obtained, since a basic workup was not performed after the removal of the BOC.groups. However, natural 10 has been depicted with the spermidine nitrogens in the free base form (Berlinck, R. G. S.; Braekman, J. Daloze, Bruno, Riccio, Ferri, Spampinato, Speroni, E. J Nat. Prod. 1993, 56, 1007-1015; Jares-Erijman, E. Ingrum, A. Carney, J. Rinehart, K. Sakai, R. J. Org. Chem. 1993, 58, 4805-4808), yet the 'H and 1 3 C NMR spectra of synthetic 10 and natural 10 were indistinguishable. Treatment of synthetic 10 with 0.1 M NaOH saturated with NaCI resulted in downfield shifts of the C41 and C45 hydrogens. To investigate this issue further, i was prepared to model the hydroxyspermidine unit of 1 3 ,14,15-Isocrambescidin 800. Chemical shifts of the hydrogens of the hydroxyspermidine units of i and synthetic 10 were nearly identical; the absence of the guanidine unit made assignments for i straightforward. Treatment ofi with 0.1 M NaOH gave ii as the free base.

As summarized in the Table below, there were significant upfield shifts of the C41 and hydrogens in ii upon deprotonation. This study and related experiments with synthetic provide confidence that natural 13,14,15-Isocrambescidin 800 (10) was isolated as the c1 trihydrochloride salt. Synthetic 10 was indistinguishable from a natural sample of 10 by o HPLC comparisons using three eluents.

41 SNAcO NHR

,NHR

OH

SiR H*HCI II R H 1H NMR shifts of the C41 and C45 hydrogens.a 8 (ppm), mult position i ii 41 2.99-2.84, m 2.66-2.60, m 3.14-3.08, m 2.86-2.78, m aln CD 3 OD at 500 MHz To provide one additional point of comparison, synthetic 10 was converted to triacetylated derivative 146. Data for this product agreed perfectly with 'H and "C NMR data reported for this derivative of natural 10 (Berlinck, R. G. Braekman, J. Daloze, Bruno, I.; Riccio, Ferri, Spampinato, Speroni, E. J. Nat. Prod. 1993, 56, 1007-1015).

Proofthat the C43 Stereocenter of 13,14,15-Isocrambescidin 800 (10) is S. As noted earlier, the S configuration of the C43 stereocenter of 13,14,15-Isocrambescidin 800 (10) had been proposed solely by analogy with Crambescidin 816 (Berlinck, R. G. Braekman, J. C.; Daloze, Bruno, Riccio, Ferri, Spampinato, Speroni, E. J. Nat. Prod. 1993, 56, 1007-1015; Jares-Erijman, E. Ingrum, A. Carney, J. Rinehart, K. Sakai, R.

J. Org. Chem. 1993, 58, 48054808). On the surface, our total synthesis of 10 appeared to confirm this assignment. However, since the C43 stereocenter is far removed from stereocenters of the pentacyclic guanidine moiety, it was not confident that epimers at this sterogenic center would be readily distinguished. To pursue this issue further, (43R)- 13,14,15-Isocrambescidin 800 (147) was prepared from 141 and ent-144 (Figure 43) (Hydroxyspermidine derivative ent-144 was prepared from (S)-epichlorohydrin (Coffey, D.

92 McDonald, A. Overman, L. E. J. Org. Chem. 1999, 64, 8741-8742)). 147 was O indistinguishable from synthetic 10 and natural 10 by 'H and 3 C NMR comparisons as well as by HPLC analysis.

To unambiguously differentiate the C43 epimers of 13,14,15-Isocrambescidin 800, a common Cc derivative of natural 10, synthetic 10 and 147 were prepared. Since only 200 pg of natural was available, it was chosen to employ Mosher derivatives and do the analysis by 9 F NMR C1 spectroscopy (Dale, J. Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512-519). The tris O Mosher derivatives 148 (43S) and 149 (43R) were prepared from 1 10 methoxy-a-(trifluoromethyl)phenylacetic acid (MTPA), synthetic 10 and 147 according to the method developed by Ward (Ward, D. Rhee, C. K. Tetrahedron Lett. 1991, 32, 7165- 7166) and their 19F NMR spectra were recorded. Since these products were mixtures of two rotamers on the NMR time scale, six 19F signals were observed. Several of the signals were substantially different in diastereomers 148 and 149 (Figure 44). The (S)-MPTA derivatives of natural and synthetic 10 were identical, thus unambiguously establishing that the stereochemistry of 13,14,15-isocrambescidin 800 (10) at C43 is S.

Relative Energies ofPentacyclic Guanidine Stereoisomers. In contrast to the studies with Ptilomycalin A/Crambescidin compounds, the investigations with the Isocrambescidin compounds provided access to several pentacyclic guanidine stereoisomers. The relative energies of the 13,15-Epicrambescidin and 13-Epicrambescidin pentacyclic guanidine moieties are readily discerned, since 139 and 135 equilibrate at room temperature in the presence of HCI (Figure 37). No similarly clean equilibration allows us to precisely specify the relative energy of the 13,14,15-Isocrambescidin ring system. Nonetheless, that the 13,14,15-Isocrambescidin ring system is considerably more stable than the 13- Epicrambescidin guanidine moiety was signaled early in the studies when it was observed that the 13-Epicrambescidin ester 135a was converted in good yield to the allyl ester analog of the 13,14,15-Isocrambescidin acid 141 upon treatment with Et 3 N in hot methanol.

Moreover, exposure of 142, 143, or the acid derived from 139 to methanolic Et 3 N at 60 0

C

provided the 13,1 4 ,15-Isocrambescidin acid 141 and the 13-Epicrambescidin acid 142 in an 0C approximate ratio of 12:1 (Figure 40). Although the complexity of this reaction mixture, the Sinability to isolate 142 in pure form, and analytical difficulties prevents unambiguous Z specification that this ratio of 141 and 142 accurately represents thermodynamic equilibrium at 60 0 C, this ratio is a reasonable estimate. Using this estimate, the energetic ordering of the 13-Epicrambescidin, 13,15-Epicrambescidin and 13,14,15-Isocrambescidin pentacyclic 0 guanidine ring systems depicted in Figure 45 is obtained.

That epimerization of the 1 3 ,15-Epicrambescidin guanidine moiety at C14 would be highly favored is apparent in the molecular models shown in Figures 31 and 38. In one hydropyran S 10 chair conformer of the 13,15-Epicrambescidin ring system the ester substituent is thrust over the hydropyran ring (conformer B of Figure 38 and alternate views shown in Figure 31) and in the other chair conformer, which relieves this interaction, the methyl group is axial (conformer A of Figure 38). No such destabilizing interactions exist in the 13,14,15- Isocrambescidin ring system.

Conclusion. The first total syntheses of 1 3 ,1 4 ,15-Isocrambescidin 800 (10) and 13,14,15- Isocrambescidin 657 (10a) were accomplished in convergent fashion. The synthesis of was achieved in 11% overall yield from amine 128 by a sequence involving 5 isolated intermediates. As detailed previously, 128 can be accessed from commercially available 3butyn- 1-ol in 30% overall yield by way of nine isolated and purified intermediates. Thus, the approach to the Isocrambescidins recorded herein is capable of providing these guanidine alkaloids on meaningful scales.

The total syntheses detailed herein confirm the stereochemical assignments of 10 and 10a and rigorously establish that the absolute configuration of the hydroxyspermidine side chain of is S. Moreover, this Example demonstrated for the first time that the tethered Biginelli strategy for preparing crambescidin alkaloids can be extended to guanidine intermediates and that the key Biginelli condensation can be accomplished under sufficiently mild conditions that fragments containing the full functionality of the Crambescidin core can be employed.

Experimental Section (Experimental details are the same as those described in the o preceding example) (6S. JZ13S) -6-A mino-N-carboxamidine-8.(I '-dioxan-2 '-0l-2-methvl-13riiopropylsiloxvgentadeca-2l1..-iene (129). A solution of arnine 128 (2.95 g, 6.12 mmol), 1-H-pyrazole-1-carboxamidine hydrochloride (2.70 g, 18.4 mmol), i-P r 2 EtN (4.4 mL, 24 C1 mrnol) and D~vf (6 mnL) was maintained at rt for 16 h and then at 601C for 4 h. The solution was cooled to rt and partitioned between CHC1 3 (300 m.L) and 0.1 M HCI (75 mL). The organic phase was washed with 0. 1 M HCI (75 mL) and, H 2 0 (75 dried (Na2SO 4 filtered and concentrated to give a 2:1 mixture of guanidine 129 and amine 128. This mixture was dissolved in DMF (6 mL) and again allowed to react (rt for 16 h and 60*C for 4 h) with l-H-pyrazole-1-carboxarnidine hydrochloride (1.35 g, 9.2 mmol) and i-Pr 2 EtN base (2.2 mL, 12 mmol). The reaction was worked up as previously described, residual DMF was removed by evacuation for several hours at 0. 1 mm to provide 3.20 g of crude guanidine 129 as a light yellow oil. This intermediate was used without further purification: 'H NMR (500 M2z, CDCI 3 5 7.82 (app d, J 6.7 Hz, IlH), 7.24 (br s, 1I-H), 5.43-5.39 (in, I 5.29-5.24 (in, INH), 5.09 (br t, J =7.0 Hz, IlH), 4.45 (app q, J =7.3 Hz, INH), 3.98-3.76 (in, 4H), 3.62- 3.59 (in, INH), 2.2 0-2.13 (in, 2H), 2.02-1.74 (overlapping in, 6H), 1. 74-1.67 (in, 2H), 1. 69 (s, 3H), 1. 64-1.5 8 (overlapping m, 2H), 1.62 3H), 1. 51-1.3 8 (in, 2H), 1. 05 (in, 2 1H), 0. 87 (t, J 7.4 Hz, 3H), 3 C NMIR (125 MIHz, CDCI 3 8 157.6, 135.0,' 132.7, 126.9, 123.1, 100.5, 69.8, 59.8, 59.3,46.6,45.0,36.5,31.7,30.5,25.7,25.0,24.8,22.2, 18.1, 18.0, 17.6, 12.3, 9.3 ppm; ]A (film) 2961,2865, 1651, 14637,1383, 1246, 1109 cin'; HIRMS (FAB) m/lz 524.4225 (524.4250 calcd for C 27

H

58

N

3

O

3 Si, M-CI); [a 25 DE0+1.7, [a] 25 577 [a] 25 546 [a]l 2 5 43 [a] 2 405 +9.3 (c 1.3, CHCI 3 (4a.S 1-Al~-vabnl~etdclxcroyl3 4Sbutvldimethylsioxvpenvyll7-1(SZ. 7S-2-(1 '-dioxan-2 7 nonenyll-1,2,4a..6. 7 -hexahvdro-J -imino-pvrrolofJ 2 -c-pyrimidine Hydrochloride (132).

N-Methylmorpholine-N-oxide (2.16 g, 18.4 inmol) and 0504 (3.1 inL, 0.24 mmol, 2% in tert- 3D butanol) were added to a solution of guanidine 129 (3.2 g, -6.1 inmol), 'VHF (105 mL) and

H

2 0 (15 mL). The mixture was stirred at rt for 8 h, Florisil (1.5 g) and NaHSO 3 (1.5 g) were 0 added, and the resulting mixture was stirred for an additional 10 h. Celite and MgSO 4 then were added, the mixture was filtered and the eluent was concentrated to give the corresponding crude diol as a brown oil. This oil was dissolved in toluene (120 mL) and morpholinium acetate (3.6 g, 24 mmol) and Pb(OAc) 4 (3.3 g, 7.3 mmol) were added. The resulting mixture was maintained at rt for 45 min and Celite was added. This mixture was S filtered through a plug of Celite, the eluent was diluted with toluene (200 mL) and the Ssolution was concentrated to give a brown oil. This oil was azeotroped to dryness with O toluene (200 mL) and the residue was combined with P-ketoester 131 (5.3 g, 9.2 mmol) and 10 2,2,2-trifluoroethanol (9 mL). The resulting solution was maintained at 60 0 C for 20 h and then partitioned between CHC13 (250 mL) and 0.1 M HC1 (50 mL). The organic phase was washed with 0.1 M HCI (50 mL) and brine (50 mL), dried (Na 2

SO

4 filtered and Sconcentrated. Analysis by 'H NMR revealed a 7:1 ratio of trans:cis Biginelli adducts.

Purification of the crude mixture by flash chromatography

(CHCI

3 99:1 CHCl 3 -MeOH 98:2 CHC13-MeOH) on silica gel deactivated with pH 7.0 buffer (McDonald, A. Overman, L. E. J. Org. Chem. 1999, 64, 1520-1528) provided 3.22 g (48% from 128) of the desired anti adduct 132 as a light brown oil and 331 mg from 128) of syn adduct 133. Data for 132: 'H NMR (500 MHz, CDCI 3 5 9.06 1H), 7.33 1H), 5.95-5.88 1H), 5.43 (app t, J= 9.8 Hz, 1H), 5.31 (app dq, J= 17.2, 1.5 Hz, 1H), 5.

27 -5.25 1H), 5.23 (app dq, J= 10.4, 1.3 Hz, 1H), 4.57 (br d, J= 5.7, 2H), 4.46-4.41 2H), 4.27-4.24 1H), 4.17-4.07 (m, 2H), 4.01-3.95 2H), 3.91-3.78 3H), 2.77-2.71 2H), 2.65-2.59 1H), 2.45- 2.40 1H), 2.32 7.6 Hz, 2H), 2.07-1.88 6H), 1.79-1.55 11H), 1.53-1.43 (m, 4H), 1.31-1.25 21H), 1.13 J= 6.1 Hz, 3H), 1.05 21H), 0.87 J= 7.4 Hz, 3H), 0.86 9H), 0.037 3H), 0.032 3H); 13C NMR (100 MHz, CDCl 3 5 173.4, 165.0, 149.9, 147.3, 135.3, 132.2, 126.4, 117.9, 100.9, 100.3, 69.8, 68.3, 64.8, 64.7, 59.9, 59.4, 57.5, 54.1, 46.1,39.0, 34.8,34.2, 33.3, 31.6, 30.9, 30.3, 29.6,29.52, 29.48, 29.42, 29.3, 29.1, 29.0, 28.5, 26.0, 25.8, 24.83, 24.76, 24.4, 23.6, 22.1, 18.01, 17.98, 12.3, 9.2, -4.7 ppm; IR (film) 2926, 2856, 1738, 1713,1681,1538,1462,1382,1256, 1086 cm'; HRMS (FAB) m/z 1044.6 (1044.8 calcd for C 59 H, IoN 3 08Si 2 M-CI); []25D -21.2, [a] 25 57 7 -21.3, [a] 25 546 -23.3, [a] 25 43 -28.8, [a] 25 405 -25.1 (c 1.9, CHCI 3 0 0 0 (4aS, 7 S) -4-fl5-(Allloxc pe dearbonyl)D dec loxcarbonyl/-7-(5Z, 7S)-2-(1 '.3'_-dioxan-2'-p Z 7-hdroxy-5-nonenvll-1,2.4a,5,6,7-hexahydro-3-f4S)-4-hydroxvpentyvl-l iminopvrrolofl,2-clDvrimidine Hvdrochloride (134). A solution of 132 (2.80 g, 2.59 mmol), tetrabutylammonium fluoride (TBAF, 13 mL, 13 mmol, 1.0 M) and DMF (26 mL) was maintained at rt for 24 h, then more TBAF (6 mL, 6 mmol, 1.0 M) was added. The solution r was maintained at for 6 h then partitioned between CHC13 (200 mL) and 0.1 M HC1 (75 mL).

The organic phase was washed with saturated aqueous HCO 2 Na (2 x 50 mL), dried (Na 2

SO

4 S filtered and the filtrate was concentrated. The crude product was purified by flash 0 10 chromatography (95:5:0.1 EtOAc-isopropanol-formic acid 90:10:0.1 EtOAcisopropanol-formic acid 85:15:0.1 EtOAc-isopropanol-formic acid) on silica gel deactivated with pH 7.0 buffer to give the formate salt of the diol 1.68 g as a light brown oil.

The formate salt was easier to purify, but the chloride salt was more stable. Therefore, after purification, the formate salt was converted quantitatively to chloride salt 134 by partitioning the formate salt between CHCI 3 (150 mL) and 0.1 M HCI (25 mL) and washing the organic layer with 0.1 M HCI (25 mL) and brine (25 mL). The organic phase was dried (Na 2

SO

4 filtered and concentrated to give diol 134: 'H NMR (500 MHz, CDCl 3 5 8.63 1H), 7.43 1H), 5.95-5.87 1H), 5.51-5.42 2H), 5.31 (ddd, J= 17.2, 3.0, 1.5 Hz, 1H), 5.22 (ddd, J= 9.2, 3.0, 1.3 Hz, 1H), 4.57 (dt, J= 5.7, 1.3 Hz, 2H), 4.43 (dd, J= 9.9, 4.3 Hz, 1H), 4.32 (app q, J= 7.1 Hz, 1H), 4.28-4.25 1H), 4.17-4.08 2H), 4.05-3.92 3H), 3.89-3.82 2H), 2.91-2.86 1H), 2.62-2.58 1H), 2.52 (dt, J= 11.8, 4.6 Hz, 1H), 2.42-2.39 1H), 2.32 J= 7 .6 Hz, 2H), 2.16-1.96 6H), 1.86-1.72 3H), 1.70-1.44 11H), 1.30-1.24 22H), 1.19 J= 6.2 Hz, 3H), 0.91 J= 7.4 Hz, 3H); 3 C NMR (125 MHz, CDC1 3 8 173.5, 165.0, 149.7, 147.5, 133.5, 132.3, 130.4, 118.0, 101.0, 100.5, 68.7, 65.4, 64.85, 64.76, 60.1, 59.6,.6, 57.6, 54.2, 45.8, 38.1,34.7, 34.2, 33.1, 30.4, 30.2, 29.6, 29.51, 29.46, 29.37, 29.2, 29.1, 28.6, 26.0, 24.9, 24.7, 24.0, 23.5, 22.2, 9.7 ppm; IR (film) 3344, 2925, 2854, 1736, 1685, 1542, 1462, 1384, 1259, 1170, 1084, 1001 cm'; MS: HRMS (FAB) m/z 774.5615 (774.5632 calcd for C 44

H

76

N

3 0 8 M-CI); [a] 25 D -39.4, [a] 25 577 -40.2, N [a]25546 -44.8, [a] 25 43 -66.0, [a]2405 -70.0 (c 1.2, CHCI 3 0 Z Formation ofPentacycle 19b from 18 by Reaction with Methanolic HCL Acetyl chloride (320 pL, 4.5 mmol) was added to a 0°C solution of MeOH (200 pL, 5.0 mmol) and EtOAc (30 mL) to give a 0.15 M solution of HCI in EtOAc. Diol 134 (1.10 g, 1.36 mmol) was then dissolved in 27 mL of this solution. This solution (containing 4.1 mmol of HCI) was I maintained at rt for 6 h, then partitioned between CHC1 3 (250 mL) and brine (50 mL). The organic phase was dried (Na 2

SO

4 filtered and concentrated. Purification of the residue by flash chromatography

(CHCI

3 99:1 CHC13-MeOH 98:2 CHC13-MeOH) gave 780 mg of an approximate 8-9:1 mixture ofpentacycles 135b and 139 as a light yellow oil (it was difficult to measure accurately the ratio of 135b and 139, since many peaks in the 'H NMR spectra overlapped). This mixture was used without further purification in the next step.

For characterization purposes, a sample of this mixture was purified by reverse phase HPLC (9:1 MeOH-0.1 M NaCI). To insure that.the counterions of 135b and 139 were uniquely chloride, pure samples of 135b and 139 were dissolved in CHCI 3 (50 mL), washed with 0.1 M HCI (10 mL), and the organic phases were dried (Na 2

SO

4 filtered and concentrated (there are small differences in the 'H NMR and 3 C NMR spectra of 135b, 139 and 10a before and after washing with 0.1 M HCI. In Example IV, 135b and 10a were not washed with 0.1 M HCI after purification.

Data for 135b: 'H NMR (500 MHz, CDC3) 5 10.37 1H), 9.81 1H), 5.95-5.87 1H), 6 9 -5.65 1H), 5.48 (br d, J= 10.9 Hz, 1H), 5.31 (dq, J= 17.2, 1.5 Hz, 1H), 5.22 (dq, J= 10.4, 1.3 Hz, 1H), 4.57 (dt, J= 5.7, 1.4 Hz, 2H), 4.50 (br d, J= 8.1 Hz, 1H), 4.31-4.27 (m, 1H), 4.26-4.21 1H), 4.12-4.07 1H), 3.98-3.95 1H), 3.77-3.72 1H), 2.91 (d,J 11.7 Hz, 1H), 2.58-2.53 2H), 2.32 J= 7.6 Hz, 2H), 2.31-2.28 3H), 2.21-2.17 2H), 1.93 (dd, J= 14.5, 5.3 Hz, 1H), 1.86-1.72 3H), 1.69-1.60 7H), 1.57-1.36 6H), 1.32-1.20 19H), 1.1 7 -1.12(m, 1H), 1.13 J=6.0 Hz, 3H), 0.87 J= 7.3 Hz, 3H); 3 C NMR (125 MHz, CDCI 3 8 173.4, 169.0, 150.9, 133.3, 132.3, 129.8, 118.0, 85.6, 84.7, 70.8, 68.8, 65.5, 64.8, 58.5, 55.1, 52.2, 37.5, 37.2, 34.2, 33.0, 32.1, 30.9, 30.0, 29.56, 29.53, 29.46, 29.38, 29.2, 29.11,29.09,28.5, 25.9, 24.9,23.8, 22.0, 18.0, 10.2 ppm; IR (film) Z 2926, 2853, 1732, 1659, 1615, 1462, 1349, 1202, 1022 HRMS (FAB) m/z 698.5117 (698.5108 calcd for C 4 1

H

68

N

3 0 6 M-C1); [a]25D -54.6, [a] 25 577 -55.6, [a] 25 5 46 -64.2, [a] 25 4 3 -115, [a] 25 405 -141 (c 1.25, CHCI 3 Data for minor pentacycle 139: 'H NMR (500 MHz, CDC1 3 8 10.23 1H), 9.59 1H), 5.96-5.88 lH), 5.68-5.64 1H), 5.48 (br d,J= 11.0 Hz, 1H), 5.31 (dq, J= 17.2, Hz, 1H), 5.23 (dq, J= 10.4, 1.3 Hz, iH), 4.57 (dt, J= 5.7, 1.3 Hz, 2H), 4.56 (br s, IH), 4.16 S D1 J= 6.7 Hz, 2H), 4.08 (dt, J= 11.0, 5.4 Hz, 1H), 3.97-3.92 IH), 3.91-3.88 1H), 2.57-2.52 2H), 2.46-2.43 2H), 2.33 J= 7.5 Hz, 2H), 2.30 J= 11.1 Hz, 1H), 2.30-2.26 1H), 2.25-2.17 2H), 1.92 (dd, J= 14.2, 5.8 Hz, 1H), 1.77-1.42 16H), 1.36 J= 12.3 Hz, IH), 1.33 J= 6.7 Hz, 3H), 1.32-1.24 19H), 0.85 J= 7.3 Hz, 3H); 3 C NMR (125 MHz, CDC 3 5 173.5, 167.9, 148.9, 133.3, 132.3, 129.8, 118.1, 84.7, 82.9, 70.7, 70.1, 65.6, 64.9, 54.6, 53.0, 52.5, 37.8, 36.8, 34.2, 31.1, 30.33, 30.31, 29.61, 26.56, 29.49, 29.42, 29.23,29.15,29.11,28.6, 28.4,25.9,24.9,23.9,21.8, 14.1, 10.3 ppm; IR (film) 2926, 2853, 1732, 1662, 1620 cm'; LRMS (FAB) m/z 698.51 (698.5108 calcd for

C

4

,H

6 8

N

3 0 6 M-CI); [a] 2 5 D -73.2, [a] 25 5 77 -67.3, [a] 25 5 46 -81.5, [a] 2 5 43 5 -149, [a] 25 405 -184 (c 0.3, CHCI 3 Carboxylic Acid 25 and 13,14,15-Isocrambescidin 657 A solution of the 8-9:1 mixture of 135b and 139 (50 mg, 0.068 mmol), morpholine (24 L, 0.27 mmol), (Ph 3

P)

4 Pd (16 mg, 0.014 mmol) and MeCN (5 mL) was maintained at rt for 2 h. Additional morpholine (12 pL, 0.13 mmol) and (Ph 3

P)

4 Pd (8 mg, 0.007 mmol) were added and the solution was maintained at rt for an additional 2 h. The solution was then partitioned between CHCI 3 mL) and 0.1 M HCI (10 mL). The organic phase was washed with 0.1 M HCI (10 mL), dried (Na 2

SO

4 filtered and concentrated to give a brown oil. The brown oil was filtered through a plug of silica gel (99:1 CHCI 3 -MeOH 98:2 CHCI 3 -MeOH), concentrated and the residue was dissolved in Et 3 N (95 pL, 0.68 mmol) and MeOH (7 mL). The resulting solution was maintained at 60 0 C for 36 h and then partitioned between CHCI 3 (50 mL) and 0.1 M HCI (8 99 ML). The organic phase was washed with 0. 1 M HCI (8 mL), dried (Na 2 SOA) filtered and z concentrated. Purification of the residue by flash chromatography (99:1 CHCI 3 -MeOH 98:2 CHCI 3 -MeOH 95:5 CHCl 3 -MeOH) provided 28 mg of 141 as a light yellow oil. To insure that the counterion was uniquely chloride, 141 was dissolved in CHC1 3 (50 m.l) and washed with 0. 1 M HCI (10 The organic phase was dried (Na2SO 4 filtered and concentrated. Data for 141: 'H NMR (500 MHz, CDCI 3 5 10.00 lH), 9.23 IHM, 5.64 (app t, J=8.1 Hz, INH), 5.50 (br d, J= 11.0 Hz, I 4.57 (br s, 4.16-4.11 (in, I1H), 4.03- 3.99 (in, IlH), 4.00-3.97 (mn, INH), 3.92-3.88 (in, INH), 3.72-3.68 (in, I 3.45 J= 3.3 Hz, INH), 2.59-2.51 (in, 2H), 2.33 J =7.5 Hz, 211), 2.29-2.24 (in, iN), 2.24-2.17 (mn, 3H), N- 10 1.89-1.8.0 (in, 4H), 1.

7 5 -1.45 I OH), 1.39 J= 12.3 Hz, IN), 1.

3 0 1.24 (in, 23H), 1. 18 J 6.0 Hz, 3M), 0.95 J1= 7.3 Hz, 3Hf) 3 C NMR (125 MHz, CDCI,) 5 178.4, 167.7, 149.3, 133.6, 129.6, 85.0, 82.9, 70.8, 69.1, 65.3, 52.8, 52.0,41.7,38.1, 37.4, 33.9, 32.7, 31.4, 30.2, 29.5, 29.43, 29.37, 29.35, 29.2, 29.1, 29.0,28.5, 27.9, 25.8,24.7, 24.0, 22.1, 20.0, 10.2 ppm; JR (film) 3200, 2924, 2852, 1732, 1660, 1621, 1189, 1167, 1027 cm-1; HRMS (FAB) m/lz 658.4789, (658.4795 calcd for C 3 &H,6N 3

O

6 M-Cl); [a] 2 5 D0-47.3, 7 -49.5, [a] 2 5 4 6 55.9, [a 2 5 435 -99.8, [cc] 25 405 -122 (c 1.2, CHCI 3 Carboxylic acid 141 was quantitatively converted to the carboxylate inner salt by washing a CHC1 3 (5 m.L) solution of the acid (5 mng) with 1 M NaOH (1 mL) and brine (1 The organic layer was dried (Na 2

SO

4 and then concentrated to provide 10a as a colorless oil:.

[fI 25 D -3 5.4 (c 0. 8, MeOI{). Spectroscopic and mass spectromnetric data for this sample were consistent with data published for natural 4 l.

4 5-Di-:ert-buoxvcarbonv131415ocrambescidin 800(045). A solution of carboxylic acid 141 (30 ing, 0.043 inmol), benzotriazol1-yloxyis(dimethylamino)phosphniu hexafluorophosphate (28 ing, 0.064 mmol), (S)-hydroxyspermidine derivative 144 (23 ing, 0.064 mol), Et 3 N (29 .iL, 0.22 inmol) and CH 2

CI

2 (2.0 mL) was maintained at rt for 1 h and then partitioned between Et 2 O (40 mL) and 0. 1 M HCl (10 inL). The organic phase was washed with brine (2 x 10 dried (MgSO 4 filtered and concentrated. Purification of this residue by flash chromatography (99:1 CNCI 3 -MeOH 97:3 CHCI 3 -MeOH) gave 32 mg 100 of 145 as a colorless foam: 'H NMR (500 MHz, CD 3 0D) S 5.70 (br t, J= 8.8 Hz, 1H), 0 5.51 J= 11.1 Hz, 1H), 4.45 (brs, 1H), 4.19-4.06 3H), 3.92-3.78 3H), 3.84 (dJ= Z 3.4 Hz, 1H), 3.59-3.23 3H), 3.19-3.12 3H), 3.06-2.97 2H), 2.58 (dd, J= 12.8, S2.3 Hz, 1H), 2.45-2.32 4H), 2.31-2.24 2H), 2.18-2.12 1H), 1.96 (dd,J= 13.1, 6.1 Hz, 1H), 1.

8 1 -1.

4 4 18H). 1.43(s, 18H), 1.

3 8 -1.17(m,23H), 1.16(d,J=6.0 Hz,3H), 0.95 J= 7.3 Hz, 3H); 3 C NMR (125 MHz, CD 3 OD) (The C38 amide exists on the NMR time scale as an approximate 1:1 mixture ofrotamers. Some of the signals of carbons in close proximity to C38, including the carbons of the hydroxyspermidine unit, are doubled. In cases Swhere the rotamers can be distinguished, these signals are listed in parentheses) C 110 5 (176.6/176.2), 169.8, 158.6, 158.4, 150.2, 134.1,131.3,86.7,84.6,80.02,79.95,72.0,70.1, (69.0/68.3), 66.2, (55.0/53.4), 54.8, 54.3, 45.0, 42.6, 39.1, (38.9/38.7), 38.1,36.2, 34.3, 34.1, 33.7, 32.9, 31.0, 30.78, 30.75, 30.67, 30.64, 30.57, 30.54, 30.50, 30.24, 30.16, 29.7, 28.9, 28.8, 28.7, 27.0, (26.7/26.6), 25.0, 22.4, 21.0, 10.8 ppm; IR (film) 3385, 2927, 2854, 1731, 1668 1614,1449,1366,1253, 1167, 1028 HRMS (FAB) m/z 1001.

7 (1001.7 calcd for Cs 55

H

7

N

6 Olo, [a] 22 DO-68.7, 22 577 -72.9, [a] 22 546 -83.3, [a] 2 2 435 -148 (c 0.6,

CHCI

3 1 3 14 ,15-Isocrambescidin 800 Trihydrochloride A solution of 145 (30 mg, 0.029 mmol) and 2.9 mL of a 2.0 M solution of HCI in EtOAc was maintained at rt for 30 min and then concentrated. Purification of the residue by reverse phase HPLC (3.5:1 MeOH-0.1

M

NaCI, 5 p Altima C18 column) gave 18 mg of 13,1 4 ,15-Isocrambescidin 800 a light yellow oil, as its trihydrochloride salt: [af22D -67.7, [a]22 77 -70.9, [a] 22 546 -80.6 (c 0.73, MeOH). NMR data for this sample were consistent with data published for natural 10 and synthetic 10 was indistinguishable from a natural sample of 10 by HPLC comparisons using three eluents.

Preparation of Peracetvl-13,14,15isocrambescidin 800 Hydrochloride (146). A solution of 1 3 ,1 4 ,15-isocrambescidin 800 acetic anhydride (1.2 mL) and pyridine (2.4 mL) was maintained at rt for 20 h then concentrated using a vacuum pump. The residue was dissolved in CHC13 (40 mL) and washed sequentially with brine (10 mL), 0.1 M HCI (10 mL) and brine (71 (10 mL). The solution was dried (Na 2 SO4), filtered and concentrated. Purification of the oresidue by flash chromatography (95:5 CHC1 3 -MeOH) gave 8 mg (70 of Z peracetylisocrambescidin 800 (146). 'H NMR and 3 C NMIR data for synthetic 146 were in perfect agreement with data reported for this derivative of natural 13,14,1 800 (Berlinck, R. G. Braekman, J. Daloze, Bruno, Riccio, Ferri, S.; Spampinato, Speroni, E. J1 Nat. Prod 1993, 56, 1007-1015).

(4aR, 7S)-4-fJ5-(Allyloxvcarbonyl)Dentadecyloxvcarbonvyl-.34(4S)-4..

butvldimethvlsi~oentvll- 7-ISZ, 7S)-2-(1 '-dioxan-2 nonen 1 -1 2 4a 5 6 7-hexah vdro-l-imino yrrolo 1 2 c rimidine Hydrochloride (133).

'H NMvR (500 MlHz, CDCl 3 8 9.16 1H), 6.99 iN), 5.94-5.87 (in, IN), 5.42 (br t, J1= 9.8 Hz, 1H), 5.30 (dq, J =17.2, 1.5 Hz, 1H), 5.27-5.24 (in, IlH), 5.22 (dq,1J 10.4, 1.3 Hz, IH), 4.56 (dt, J= 5.6, 1.4 Hz, 2H), 4.46-4.41 (in, 2H), 4.24-4.21 (in, iN), 4.18-4.08 (in, 2H), 4.04-3.89 (in, 5H), 2.82-2.77 (in, 1H), 2.66-2.57 (in, 2H), 2.32 7.6 Hz, 2H), 2.27-2.19 (in, I 2.03-1.5 5 (in, 17H), 1.31-1.24 (mn, 25H), 1. 12 J1=6.0 Hz 3H), 1. 04 2 1H), 0.87 J 7.6 Hz, 3H), 0.85 9H), 0.028 0.024 3H); 3 C NMvR (125 MfHz,

CDCI

3 5 173.4, 164.8, 150.7, 149.6, 135.2, 132.2, 126.5, 117.9, 102.8, 100.0, 69.8, 69.7, 68.2, 64.8, 60.0, 59.4, 57.8, 52.2, 44.4, 39.0, 34.2, 33.5, 31.6, 30.8, 30.1, 30.0, 29.54, 29.52, 29.48, 29.41, 29.3, 29.1, 29.0, 28.5, 26.0, 25.8, 24.8, 24.4, 23.5, 22.1, 18.01, 17.99, 12.3, 9.2, 29 -4.7 ppm; MS (FAB) m/z 1044.3 (1044.8 calcd for C 59

HI

10

N

3

O

8 Si 2 M-Ci).

Tetracyclic Guanidine 138b. 'H NMR (500 MI-z, CDCl 3 8 10.46, 5.94 -5.87 (mn, I1H), 5.67-5.64 (mn, INH), 5.48 (br d, J= 10.9 Hz, 1H), 5.30 (dq,1J 17.2, 1.5 Hz, INH), 5.22 (dq, J1= 10.7, 1.3 Hz, I1H), 4.56 (dt, J= 5.7, 1.3 Hz, 2H), 4.56 (br s, iN), 4.20-4.08 (mn, 3H), 4.05-3.99 (mn, 3.94-3.91 (in, 1H), 3.68 (br s, 1Ff), 2.99-2.94 (in, 1H), 2.70-2.58 (in, 2.52-2.45 (in, 2.39-2.30 (in, 2.32 J= 7.6 Hz, 2H), 2.24-2.19 (in, 1Ff), 2.04-1.99 (in 1.91 (dd, J 14.8, 5.2 Hz, 1H), 1.87-1.71 (in, 4H), 1.68-1.58 (mn, 1.57-1.40 1.38-1.23 (mn, 20H), 1.21 J= 6.2 Hz, 3H), 0.84 7.2 Hz, 3H); 1 3 c NMR (125 M2Hz, CDCI 3 8 173.5, 164.4, 151.4, 149.4, 133.2, 132.3, 129.7, 118.0, 104.0, 85.1, 71.3, 65.9, 65.0, 64.9, 55.6, 51.8, 37.7, 37.0, 36.4, 34.2, 31.7, 31.3 30.2, 29.6, 29.54,

I

29.48, 29.40, 29.2, 29.1, 28.6, 26.1, 24.9, 24.4, 24.2, 23.6, 10.4 ppm; IR (film) 3372, 2925, 17 3 7, 1689, 1 6 5 1, 1613, 1547, 1455, 1341 MS (FAB) m/z 698.5106 (698.5108 calcd for Z C 4 1

H

68

N

3 0 6

M-CI).

N-Acylated Hydroxyspermidine Hydrochloride Salt i. 'H NMR (500 MHz, CD 3 0D) 5 4.04 J= 6.7 Hz, 2H), 3.97-3.95 1H), 3.69-3.38 3H), 3.32-3.21 1H), 3.14-3.08 (m, 2H), 2.99-2.84 2H), 2.54-2.39 2H), 2.01 3H), 2.00-1.80 3H), 1.74-1.71 (m, S 1H), 1.63-1.58 4H), 1.33-1.29 (m 22H); "C NMR (125 MHz, CD30D) (The amide t exists as an approximate 3:1 mixture of rotamers on the NMR time scale. Carbons in close proximity to the amide, including some carbons of the hydroxyspermidine unit, exhibit two signals. In cases where two rotamers were observed, carbon signals of the rotamers are listed in parentheses with the major rotamer listed first) 5 (177.5, 176.4), 173.0, (68.6, 69.4), 65.7, (54.8, 53.2), (43.9, 47.8), 38.5, (38.2, 38.3), 34.2, 34.0, (32.9, 33.0), 30.75, 30.72, 30.67, 30.61, 30.5, 30.3, 29.7, 27.0, (26.61, 27.8), (26.54, 26.59), 20.8 ppm.

N-Acvlated Hvdroxvspermidine Free Base ii. 'H NMR (500 MHz, CD30D) 6 4.04 J= 6.7 Hz, 2H), 3.91-3.85 1H), 3.65-3.32 3H), 3.27-3.13 1H), 2.86-2.78 2H), 2.66- 2.60 2H), 2.45-2.37 2H), 2.00 3H), 1.77-1.68 2H), 1.63-1.51 6H), 1.32- 1.24 22H); 3 C NMR (125 MHz, CD 3 OD)(the amide exists as an approximate 1:1 mixture of rotamers) 8 176.6, 176.3, 173.1, 70.0, 68.9, 65.7, 55.0, 53.7, 44.5, 40.0, 39.5, 39.4, 38.1, 37.9, 34.3, 34.0, 32.8, 30.8, 30.73, 30.67, 30.63, 30.5, 30.4, 29.7, 27.0, 26.7, 20.8 ppm.

EXAMPLE VI This example describes methods for preparing novel pentacyclic intermediates for the preparation of the Crambescidin/Ptilomycalin family of guanidinium alkaloids and congeners.

This example further relates to improved chemical synthesis of pentacyclic intermediates for the preparation of the Crambescidin/Ptilomycalin family of guanidinium alkaloids and congeners.

SSynthesis Z Tethered Biginelli Condensation. The allyl ester was chosen to protect the C(22) carboxylic C0 acid, since this protecting group can be removed in the presence of a guanidinium salt (Overman, L. et al. J. Am. Chem. Soc. 1995, 117, 2657). Biginelli condensation between compounds 151 (Overman, L. et al. J. Am. Chem. Soc. 1995, 117, 2657) and 152 S(Overman, L. et al. 1995, supra), using the previous conditions (Overman, L. et al.

S1995, supra), gave only 30-40% of product compound 153 with poor diastereoselectivity Attention was turned to the integrity of compound 152 (Figure 48). Urea 155 (Figure 48) was synthesized in an improved yield by reaction of precursor amine 154 (Overman, L. et al.

1995, supra) with trimethylsilyl isothiocyanate (Vishnyakova, T. et al. Russ. Chem. Rev.

1985, 54, 249) (Figure 48). When the ozonolysis of compound 155 was quenched with H2 and 10% Pd/C, followed by filtration and concentration, a solid product was obtained after 1 h under reduced pressure (0.1 mm) at 23 0 C. This material, gave superior yields in the Biginelli condensation Diastereoselectivity, however, was still poor (ds Extensive optimization of reaction conditions showed that in the non-typical solvent trifluoroethanol, the Biginelli condensation proceeded with good diastereoselectivity (ds=6.5:1 0.5 M, ds 4:1 1.7 The use of this solvent to improve efficiency and stereoselectivity in Biginelli condensations was recently reported (McDonald and Overman, J. Org. Chemisiry, 1999, 64:1520-1528).

Morpholinium acetate was selected as a catalyst for the Biginelli reaction (Renhowe, P. A.

Ph.D. Thesis, University of.California, Irvine. 1995). An important discovery regarding the use of morpholinium acetate was made during optimization of the Biginelli reaction. After reductive hydrogenation of the ozonolysis product of compound 152, but prior to filtration and concentration, morpholinium acetate was added to the methanolic solution of compound 152. The solution was then concentrated to give a viscous oil compound 156 that was characterized by HRMS analysis. Subjection of this oil to Biginelli condensation provided compound 153 in a much improved yield of 80%. Moreover, this modification resulted in 104 halving the reaction time to 1.5 days.

0 0 Synthesis of Enantiopure Iodide Compound 166. Previous synthesis formed iodide 166, the fragment, in only moderate enantiomeric purity (86% ee) by enantioselective reduction of an ynone precursor (Overman, L. et al. J. Am. Chem. Soc. 1995, 117, 2657; c Renhowe, P. A. Ph.D. Thesis, University of California, Irvine. 1995). A shorter synthesis that provides this intermediate in enantiomeric purity is summarized in Scheme VIII (Figure 34).

Diethylzinc addition to aldehydes 159 or 160, which were synthesized from compounds 157 O and 158 respectively, in the presence of a TADDOLate catalyst (Weber, Seebach, D.

NC 10 Tetrahedron, 1994, 50, 7473-7484) gave chiral alcohols 161 and 162 in good yield and >99% ee as determined by GLC analysis of the derived Mosher esters (Seebach, et al. Helv.

Chim. Acta 1987, 70, 954; Seebach, et al. Chimia 1991, 238; Seebach, et al. Helv.

Chim. Acta 1992, 75, 438; Seebach, et al. Helv. Chim. Acta 1992, 75,2171; Seebach, D.; et al. Tetrahedron 1994, 50, 4363; Weber, Seebach, D. Tetrahedron 1994, 50, 7473).

Alcohol 161 was found identical, except for optical rotation, to the intermediate employed in our original synthesis (Overman, L. et al. J. Am. Chem. Soc. 1995, 117, 2657).

Enantiopure 163 was converted to (S)-(Z)-l-iodo-5-triisopropylsiloxy-3-heptene (Overman, L. et al. J. Am. Chem. Soc. 1995, 117, 2657). At this point, the TIPS protecting group was deemed unnecessarily robust and was replaced with TBDMS. PMB protection of the primary alcohol allowed for TBDMS protection of the secondary alcohol, thus delivering primary iodide 166 as summarized in Figure 49. Organolithium 167 was generated from 166 by lithium-iodide exchange at -78 0 C (Dale, J. et al. J. Org. Chem. 1969, 34, 2543; Dale, J.

Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512; Ward, D. Rhee, C. K. Tettrahedron Lett. 1991, 32, 7165).

Coupling of the Fragment with the Tricvclic Intermediate. This stage of the synthesis was the least satisfactory of the earlier synthesis due to the lability of the aldehyde related to 169 (Overman, L. et al. J. Am. Chem. Soc. 1995, 117, 2657; Renhowe, P. A.

Ph.D. Thesis, University of California, Irvine. 1995). In the present allyl series, when compound 168 was oxidized with the Swem reagent substantial epimerization at C(8) occurred. No epimerization occurred with Dess-Martin periodinane oxidation (Figure The resulting aldehyde 169 was O-methylated according to established protocol (Overman, L.

SE.; et al. J. Am. Chem. Soc. 1995, 117, 2657; Renhowe, P. A. Ph.D. Thesis, University of Z California, Irvine. 1995). However, addition of compound 167 to aldehyde 170, followed by oxidation of the crude epimeric alcohols, provided ketone 171 in low yield When pure compound 170 or 171 was exposed to either commercial or deactivated silica gel for 1 h, significant lost of mass was observed. This observation partially explains of the low yields for these two steps. An alternative sequence of events was developed to overcome this difficulty (Figure 51).

Zt- ID 0 SAddition of 2.2 equivalent of compound 167 to aldehyde 169, and subsequent oxidation, yielded ketone 172 in an unoptimized yield of 30-40% (46-51% based on consumed compound 169). Ketone 172 was O-methylated, guanylated, deprotected and cyclized to pentacyclic allyl ester 8 (without intermediate purification) in an unoptimized 25-30% overall yield. This sequence should be optimizable and minimizes the loss of material upon silica gel chromatography.

Synthesis of Pentacvclic Acid 7. The allyl ester was successfully removed under standard conditions with Pd(0)/dimedone to furnish pentacyclic acid 7 (Figure 51).

The following examples are presented to illustrate the present invention and to assist one of ordinary skill in making and using the same. The examples are not intended in any way to otherwise limit the scope of the invention.

General experimental details: All reactions were carried out under an atmosphere of Ar or N 2 and concentrations were performed under reduced pressure with a BUchi rotary evaporator. Tetrahydrofuran (THF), EtzO and CHCI, were degassed with Ar then passed through two 4 x 36 inch columns of anhydrous neutral A-2 alumina (8 x 14 mesh; LaRoche Chemicals; activated under a flow of Ar at 350 0 C for 3h) to remove water.

Toluene was degassed with Ar then passed through one 4 x 36 inch column of Q-5 reactant (Englehard; activated under a flow of 5% H2/N, at 250 0 C for 3 h) to remove 02 then through one 4 x 36 inch column of anhydrous neutral A-2 alumina (8 x 14 mesh; LaRoche Chemicals; activated under a flow ofAr at 350 0 C for 3h) to remove water. Triethylamine (Et 3 pyridine, diisopropylethylamine (i-Pr2NEt), diisopropylamine, and acetonitrile were distilled from CaH2 at atmospheric pressure. Indicated molarities oforganolithium reagents were c established by titration with menthol/fluorene (Posner, G. Lentz, C. M. J. Am. Chem. Soc.

O 1979, 101, 934). Instrumentation and Chromatography: 300 MHz 'H and 75 MHz 3

C

Z spectra were obtained on a BrUker QE 300 FT NMR; 500 MHz 'H and 125 MHz 3 C NMR spectra were obtained on a BrUker GN 500 FT NMR or BrUker Q 500 FT NMR. 'H NMR chemical shifts are reported as 6 values in ppm. Coupling constants are reported in Hz and Cr\ refer to apparent multiplicities.

CMultiplicity is indicated as follows: s (singlet); d (doublet); t (triplet); m (multiplet); app t (apparent dd (doublet of doublets) etc. Mass spectra were measured on a MicroMass Analytical 7070E (CI-isobutane) or a MicroMass AutoSpec E (FAB) spectrometer. Infrared spectra were recorded using a Perkin Elmer 1600 FTIR spectrometer. Microanalyses were performed by Atlantic Microlabs, Atlanta, GA. Optical rotations were measured using a JASCO DIP-360 digital polarimeter. TLC and column chromatography were performed using E. Merck silica gel (43-60 gm) with a loading of approximately 30:1 SiO 2 :substrate.

0 0 QTBDMS AllylO C17H320 4 Si FW 328. 52 (R)-AllI- 7-(t-butvldimethvlsiloxv)-3-oxooctanoate (Compound 151). Freshly distilled allyl acetoacetate (5.0 mL, 37 mmol) was added dropwise to a 0°C mixture ofhexane-washed NaH (1.73 g, 43 mmol) and dry THF (50 mL). After 10 min, n-butyllithium (14.9 mL of a 2.7 M solution in hexanes) was then added dropwise and the resulting red solution was maintained for an additional 10 min at 0°C. A solution of compound 150 (4.53 g, 14.4 mmol) and dry THF (20 mL) was then added dropwise. After 20 min at 0°C the reaction mixture was quenched with saturated aqueous NH 4 Cl (20 mL). The layers were separated, and the H 2 0 layer was extracted with Et20 (2 x 15mL), and the combined organic layers were washed with brine (15 mL), dried (MgSO 4 and concentrated. Purification of the residue on silica gel (20:1 hexanes-EtOAc) provided 2.84 g of compound 151 as a colorless oil (9:1 z mixture of keto and enol forms by IH NMR analysis): IH NMR (500 MHz CDCl3) 6 5.86- 5.90 (m 11) 5.31 (d J1 17.0 Hz 111) 5.23 (d J 10.5 Hz 1H) 4.60 (d J= 5.7 Hz 2H) 3.76 (dd J= 11.9 6.0 Hz 11) 3.44 (s 2H) 2.52 (t J= 7.3 Hz 2H) 1.63-1.66 (m 1H) 1.55-1.58 (m 11) 1.35-1.40 (m 2H) 1.09 (d J= 6.0 Hz 31) 0.85 (s 91) 0.02 (s 61); 013 C NMR(125 MHz CDC 3 202.5 166.9 131.5 118.8 68.2 65.9 49.1 43.1 38.8 25.9 23.7 19.7 18.1 -4.4 -4.7ppm;IR(film)2956 2857 1748 1716 1255 1149 836 775cm- 1; [aI: -12.8 577: -12.9' [a]s25: -14.1" 35: 23.8* 05: 28.1P (c 1.05 CHC13). Anal. Calcd for C 17

H

3 2

O

4 Si: C 62.15: H 9.82. Found: C 62.42; H 9.93.

H

OH

TBDMSO HN N C0AIIyI

C

24

H

42

N

2 0 5 Si FW 466.70 (4aR. 7S)-4-(A Ilyloxvcarbonvl)-.2,4a, 5,6, 7-hexahydro- 7-(2-hvdroxvethyl)-3-!(4S)-4-(t.

butvldimethylsioxvpentvl) I-1oxopyrroloJl,2-clpvrimidine (153a) and aSM7S)-4- (Allyloxvcarbonyl)-1.2,4a,5.6, 7-he-xahvdro- 7-(2-hvdroxvethyl)-3-!(4S)-4-(tbutvldimethylsloxvpentvI)I-]1 xopyrroloI12ajpyrimidjln (Compound 153b). A solution of crude (S)-156 (2.3 g 9 mmol) 151 (2.2 g 6.7 mmol) and trifluoroethanol (4 mL) was heated at 60 0 C for 2 d. The reaction mixture was quenched by pouring into Et 2 O (50 mL) washing with saturated aqueous NH 4 CI (2 x 10 mL) and brine (10 mL). The organic layer was dried (MgSO 4 concentrated and purified on silica gel (1:1 hexanes-EtOAc) to yield 2.01 g of the desired cis-Biginelli product 153a and 0.51 g of the trans-Biginelli product 153b.

Compound53a: 1 HNMR(500 MHz CDCI 3 )5 8.26 11) 5.

8 9 -5.97 (m 1H) 5.30(dd J c-i 16.7 1.2 Hz IH) 3.22 (dd J= 10.4 1.0 Hz 111) 4.60 (ddd J= 22.6 13.1 5.9 Hz 2H-) 4.26 (dd J =11. 3 4.9 Hz I1H) 4.10-4.14 (m I1H) 3.74-3.80 (m 1IH) 3.40-3.67 (m 2H) Z2.58 (t J= 7.5 Hz 2H) 2.47-2.52 (m LH) 2.02-2.11 (m 1H) 1.84 (m IH) 1.60-1.77 (m 4H) 1.36-1.53 (m 3H) 1. 10 (d J= 6.0 Hz 3H) 0.85 (s 9H) 0.02 (s 3H) 0.01 (s 3H); 13

C

NMR (125M1-lz CDC13) 8 165.2 155.0 152.9 132.3 118.5 102.2 68.6 64.8 59.0 58.3 52.2 39.4 38.9 31.1 30.6 29.8 25.8 .25.0 23.7 18.1 -4.4 -4.7 ppm; LR (film) 3450 3225 [cC]i 25 3095 2954 1682 1626 1431 1111 835 776 cm- 1 D -28.6" [aL] 5 77 546: c-i -36.80 435:-~97.7' 405: -138' (c.=2.20 CHCl 3 Anal. Calcd for C 24

H

42

N

2

O

5 Si: C 61.77;H 9.07;N 6.00. Found: C 61.66;H 9.15;N 5.92.

Compound 153b: IH NMR (500MN1-z CDCl 3 88.52 (s 1H) 5.87-5.94 (m 1H) 5.28 (d J= 17.5 Hz 1H) 5.21 (d J= 10.4 Hz 1H) 4.60 (ddd J= 13.4 7.5 6.0 Hz 2H) 4.39-4.45 (m 1H) 4.34 (dd J10.5 4.6 Hz 1H) 3.77 (dd J= 11.5 5.7 Hz 1H) 3.56-3.66 (m 2H) 2.66- 2.71 (m IH) 2.49-2.54 (m 1H) 2.42-2.46 (m 1H) 2.08 (dd J= 20.7 8.7 Hz 1H) 1.76-1.81 (m I1H) 1.62-1.67 (m I H) 1.37-1.54 (m 6H) 1. 10 (d J1=6.04 Hz 3H) 0. 8.(s 9H) 0.02 (s 3H) 0.0 1 (s 3H); 3 CNMR (125 MHz CDC1 3 165.3 153.2 150.1 1 32.4 118.2 98.7 68.5 64.7 58.9 57.3 53.7 38.9 38.3 35.1 31.3 28.2 25.9 24.5 23.6 18.1 -4.5 -4.7 ppm; IR (film) 3442 3256 3100 2930 2897 1708 1668 1634 1463 1236 1082 736 cm1; D: -26.3* 25577: -26.8 25546: -29.3* 25 45-54.70 25 405 -122' (c=2.30 CHC1 3 Anal. -Calcd for C 24

H

42

N

2

O

5 Si: C. 61.77; H 9.07; N 6.00. Found: C 61.75; H 9. 10; N 5.96.

OH

2n

C

11

H

21

N

3 0 3 FW 243.31 0 Compound (156). Ozone was bubbled through a solution of compound 155 (1.74 g 9 mmol) and MeOH (50 mL) at -78 0 C until the solution was saturated (blue color appeared and persisted for -10 min). Nitrogen was then bubbled through the solution to dissipate the c excess ozone. 10% ,d/C (0.6 g) was added to the colorless solution and the reaction mixture was maintained at -78 0 C under 1 atm ofH 2 After 30 min the cooling bath was removed morpholinium acetate (2.0 g 13 mmol) was added and the reaction mixture was allowed to O warm to 23 0 C. After 4 h the reaction mixture was dried (MgSO 4 filtered and the filtrate was CI 10 concentrated. The resulting residue was diluted with trifluoroethanol (30 mL) and concentrated to give a yellow oil which was used without further purification: MS (CI) m/e cald for CH 21

N

3 0 3 243.1583 found 243.1588

PMBO

C

12

H

1 402 FW 190.24 1-(4-Methvoxvbenvloxy)-3-butyne (Compound 158). According to established procedures (Takaku H. et al.; Tetrahedron Lett. 1983 24 5363; Nakajima et al. Tetrahedron Lett.

1988 29 4139) TfOH (1.6 mL 18 mmol) was added dropwise to a 0 C solution of ,MBOC(=NH)CCl3 (169.3 g 0.6 mol) 3-butyn-l-ol (67 g 0.66 mol) dry Et 2 0 (600 mL).

After 30 min the reaction mixture was quenched by the addition of saturated aqueous NaHCO 3 (100 mL) the layers were separated the aqueous layer was extracted with Et20 mL) and the combined organic layers are washed with brine (50 mL) dried (MgSO 4 and concentrated. The resulting residue was diluted with hexanes (300 mL) filtered through a plug of silica gel concentrated and stirred under vacuum (0.1 mm Hg) at 50 0 C for 12 h yielding 158 which is used without further purification. 1 H NMR (500 MHz CDCl 3 5 7.28(d J= 8.4 Hz 2H) 6.89 (d J= 8.4 Hz 2H) 4.49 (s 2H) 3.80 (s 3H) 3.58 (t J= 7.0 Hz 2H) 2.49(dt J= 7.0 2.7 Hz 2H) 2.00(t J= 2.6 Hz 1H); C NMR (125 MHz 110 CDC1 3 5 159.2 130.0 129.3 113.7 81.3 72.5 69.2 67.8 55.2 19.8 ppm; IR (film) 3292 3001 2936 2863 1614 1514 823 cm- 1 Anal. Calcd for C 12

H

14 0 2 C 75.76; H 7.42.

Found: C 75.60; H 7.49.

0 TBDMSO"" c-I

C

11

H

20

O

2 Si FW 212.12 S-(t-Butvldimethylsiloxv)-2-DenlnaI (Compound 159). A hexane solution of n-BuLi (2.5 M 4.8 mnL) was added to a-78 0 C solution of compound 157 (2.0 g 10.9 mmol) in dry THF mL). After 10 min the reaction mixture was placed into an ice bath and dry DMIF (5 mL) in THE (20 mL) was added. After 30 min at OTC the reaction mixture was quenched by pouring into a vigorously stirred solution of 5% I 2 S0 4 (20 mL). After 1 h the layers were separated the H2 layer was extracted with Et20 (3 x 15 mL) and the combined organic layers were washed with saturated aqueous NaHCO 3 (1 x 15 mL) and brine (1 x 15 m.L) dried (MgSO 4 and concentrated. ,urification of the residue on silica gel (4:1 hexanes-EtOAc) provided 0.921 g (5 of compound 159 as a slightly yellow oil: I.H.NMR (500 MiHz CDCI 3 589.17 (s I1-H) 3.79 (t J1= 6.7 Hz 2H) 2.62 (t J1= 6.7 Hz 2H) 0.9 (s 9H) 0. 1 (s 6H); 3 C NMR (125 MI-Iz CDCl 3 5 177.0 96.2 82.3 60.6 25.8 23.5 18.3 -5.3 -5.4 ppm; IR (film) 2930 2205 1671 1111 cm- 1 FIRMS (CI isobutane) ie calcd for C 11

H

20

O

2 Si 212.1232 found 197.0998 (M CH 3 0

PMBO"

C

13

HI

4 0 3 FW =218.25 5-(4-Methoxvbenzvloxv)-2-Dentvnal (compound 160). A hexane solution of n-BuLi (2.5 M 9.34 mL mL) was added to a -78 0 C solution of compound 158 (4.04 g 21.2 mmol) in dry THF (100 mL). After 10 min the reaction mixture was placed into an ice bath and dry DMF (10 mL) in THF (100 mL) was added. After 30 min at 0 C the reaction mixture was quenched by pouring into a vigorously stirred solution of 5% aqueous H 2

SO

4 (100 mL).

"1 After 1 h the layers were separated the H 2 0 layer was extracted with Et 2 O (3 x 30 mL) and the combined organic layers were washed with saturated aqueous NaHCO 3 (1 x 30 mL) and brine (1 x 30mL) dried (MgSO 4 and concentrated. ,urification of the residue on silica gel N 10 (4:1 hexanes-EtOAc) provided 2.55 g of compound 160 as a slightly yellow oil: IH NMR (500 MHz CDCI) 5 9.16 (s 1H) 7.26 (d J= 8.5 Hz 2H) 6.88 (d J= 8.6 Hz 2H) 4.48 (s 2H) 3.79 (s 3H) 3.61 (t J= 6.7 Hz 2H) 2.69 (t J= 6.7 Hz 2H); C NMR (125 MHz CDC13)5 177.0 159.2 129.6 129.3 113.8 95.7 81.9 72.7 66.5 55.2 20.6ppm;IR (film) 3002 2865 2205 1668 1514 824 cm- 1 Anal. Calcd for C, 3

H,

4 0 3 C 71.54; H 6.47. Found: C 71.42; H 6.54.

OH

PMBO

C

1 5H 20 0 3 FW 248.33 (5S)-Hvdroxv-l-(4-methoxvben7vloxv)-3-heptvne (compound 162). According to the general procedure of Seebach Ti(Oi-,r)4 (0.50 mL 1.68 mmol) was added to a 23°C solution of (4R 5R)-2 2 -Dimethyl-a,a,ac,a3-etra(naphth-2-yl)-1 3-dioxolan-4 dimethanol (1.12 g 1.67 mmol) and dry toluene (15 mL). After 3 h solvent was removed under reduced pressure (0.1 mm). The resulting residue was dissolved in dry Et 2 O (33 mL) and the reaction vessel was cooled to -26 0 C whereupon Ti(Oi-,r) 4 (3.0 mL 10 mmol) compound 160 (1.83 g 8.37 mmol) and Et 2 Zn (9.1 mL of a 1.1 Msolution in toluene)were (71 added. After 18 h. at -26*C the reaction mixture was quenched with saturated aqueous

NH-

4 C1 (lmiL) dried (MgSO 4 filtered through Celite®, concentrated and the resulting residue was puified on silica gel (4:1 hexanes-EtOAc) to provide 1.833 g (88%)Oof compound 162 as a colorless oil: IH NMR (500 MfHz CDCI 3 5 7.2 5 (d J 8.4 Hz 2H) 6.86 (d J 8.4 Hz 2H-) 4.46 (s 2H) 4.26 (t J= 6.4 Hz 11-) 3.78 (s 3H1) 3.53 (t J= 7.0 Hz 2H) 2.58 (s I1H) 2.49 (dt J =7.0 1.5 Hz 211) 1.66 (m 2H) 0.97 (t J =7.4 Hz 3H1); 13 C NMR (125 MiHz CDCl3) 5 159.1 129.9 129.2 113.7 82.3 81.7 72.4 67.9 63.5 55.1 30.9 19.9 9.4 1; [a]25 ppmn; IR (film) 3418 2965 1613 1514 1249 823 733 cm' 5[D] 3 2 77 36 25 c~KI[a] 545: 43: 405: (c 2.35 CHC 3 Anal. Calcd for ID C 15

H

20 0 3 C 72.55; H 8.12. Found: C 72.26; H 8.14.

Following the general procedure of Ward (Ward D. Rhee C. K. Tetrahedron Let. 1991 32 7165) compound 162 (23 mg) was treated with (R)-ax-methoxy-a-(triflouromethyl)phenylacetic acid chloride [(R)-MT,ACI] to give the corresponding (R)-MT,A ester. Capillary GC analysis 150'C to 200'C/2.0*C min tR 162- (R)-MT,A 21.13 min tR ent -162-(R)-MT,A 20.69 min] showed a ratio for 99.7:0.3 of 162-(R)-MT,A and ent -162-(R)-MT,A.

OTBDMS

PMBO'^

C21 H343Si FW 362.59 (SS)-(t-Butidimethvlsiloxl)-14(4-methoxvbenzvIoxy)-3heptyne. TBSCI (1.08 g 7.2 mrnol) was added in portions over 15 min to a 23'C solution of imidazole (0.53 g 7.8 mmol) compound 162 (1.48 g 6 mmol) and dry DMIF (5 After standing at 23'C for 2 h the solution was poured into 20 mL 1120 and extracted with Et 2 O (4 x 20 mL). The combined organic layers were washed with brine (20 m.L) dried (MgSO 4 and concentrated. The crude 113 oil was placed under vacuum 1 mm) overnight to provide 2.16 g (100%) of the desired 0 product as a colorless oil which was used without fuirther purification: IHMNNM. 500MHz, z

CDCI

3 5 7.28 (d J= 8.5 Hz 21H) 6.89 (d 1= 8.5 Hz 21-H) 4.48 (s 2H) 4.28 (dt J= 6.2 1.7 Hz I H) 3.80 (d J1 1.6 Hz 311) 3.56 (dt J 7.2 1.52 Hz 2H) 2.51 (dt J 7.2 1.7 Hz 211) 1.66 (appt J= 7.0OHz 211) 0.96 (dt J= 7.3 1.3 Hz 3H) 0.91 (d J=l1.4 Hz 911) 0. 13 13 (d ~J =1.4 Hz 3H-) 0. 11 (d J =1.4 Hz 3H); C NNM (125 M4Hz CDCl 3 5 159.2 130.2 129.2 113.73 82.8 80.9 72.6 68.3 64.4 55.2 31.9 25.8 20.1 18.3 9.2 -4.5 (film) 2930 1614 1514 1249 1099 837 cm- 1 -34.50 25 -3S.0* 2546 25 [a 2 5 40.9' 4.35: -69.5, a 405: -83.5* 5.35 CHCI 3 Anal. CalcdforC 2

,H

34

O

3 Si: C 69.56; H 9.45. Found: C 69.49; H 9.50.

OTBDMS

HO,"

C

13

H

26

O

2 Si FW =242.44 (S)-(5)-(t-Butfdimehvsoxy)..3.heptpnol A solution of (5S)-(Q-butyldimethylsiloxy). 1 methoxybenzyloxy)-3-heptyne (0.17 g 0.46 mmol) DDQ (0.16 g 0.68 mmol) and 20:1

CH

2 Cl 2

-H

2 0 (3 mL) was maintained at 23TC for 2 h. The reaction mixture was quenched by pouring into Et 2 O (25 mL) and washing with saturated aqueous NaHCO 3 (2 x 5 mL) followed by brine (5 mL). The organic layer was dried (MgSO 4 concentrated and the resulting residue was purified on silica gel (4:1 hexanes-EtOAc) to provide 88 mg (80%)flof the desired product as a colorless oil: I H NMR (500 MIHz CDCI 3 6 4.26 (t 1= 6.3 Hz 111) 3.69 (t J= 6.6 Hz 2H) 2.47 (dt J= 6.3 1.6 Hz 2M) 1.98 (s IH) 1.

6 1-1- 6 7 (m 2H-) 0.94 (t 1= 7.4 13 Hz 3 H) 0.-9 (s 9H) 0. 11 (s 3 H) 0. 10 (s 31-1); C NMIR (75 MI-Iz CDCI 3 6 84.0 80.6 64.4 61.1 31.9 25.8 23.1 18.3 9.7 -4.6 -5.0 ppm; IR (film) 3388 2958 2858 1472 1256 1059 cm HERMS (EI-GCMS) m/e calcd for C, 2

H

26 0 2 Si 242.1701 found 242.1655 []5):4.1a"77-48.0 LaJ 546* 54.50 25 43:93.2' 25405: Z 1.4 CHCl 3 PMBO,9- O

C

1 5 H2O 3 FW 250.34 (3Z5)I(-ehlxbnvoy--ein5o (compound 163). A mixture of compound 162 (13.8 g 55.6 mmol) freshly distilled quinoline (138 mL 1.20 mmol) Lindlar's catalyst (,d/CaCO 3 poisoned with ,bO 1.29 g)anddry3:l hexanes -EtOAc (138 mL) was maintained at 23'C under 1 atm H 2 for 3 d. This mixture was then filtered through a plug of Celite the plug was washed with 3:1 hexanes-EtOAc (400 mL) and the elutent was concentrated to yield 13.9 g (100%) of compound 163 which was used without fuirther purification: 1

H

NlvMR (300 M1Hz CDCI 3 5 7.24 (d J= 8.4 2H) 6.86 (d J= 8.5 2W) 5.5 1-5.55 (mn 2H) 4.44 (s 214 4.30 (appq J= 7.0 114 3.79 (s 3M 3.34-3.53 (mn 2H) 2.25- 2.55 (mn 2H) 2.2 1-2.24 (mn I1W 1.41-1.64 (mn 2H) 0. 86 (t J =7.4 3H) 1 C NMv~R (75 M4Hz CDCI 3 159.1 135.2 129.8 129.3 128.6 113.7 72.7 68.61 68.0 55.2 29.7 28.3 9.7 ppm; IR (film) 3421 3007 2961 2860 1613 1514 1249 1094 821 cm- 1 [a]25D 1 7 9 0 [a]2 57 17.8' 25 5 46 20.60 [JX25 45= -35.0' [a] 2 405 42.6' (c =2.2 1CDC1 3 Anal. Calcd for C 15

H

22 0 3 C 71.97; H 8.86. Found: C 71.97; H 8.90.

OTBS

PM8QJ, C21 H36O 3 Si FW =364.61 0 (3Z5S)-4--Butldimethylsiloxv)-l-(4-methoxvbenvlox hetene (compound 164 O TBSCI (0.51 g 3.4 mmol) was added in portions over 15 min to a solution ofimidazole (0.48 g 7.0 mmol) compound 163 (0.7 g 2.8 mmol) and dry DMF (1.4 mL) at 23 0 C. After standing at 23 0 C for 2 h the solution was poured into 20 mL H 2 0 and extracted with Et 2 0 (4 x 20 mL). The combined organic layers were washed with brine (20 mL) dried (MgSO 4 and concentrated. The crude oil was placed under vacuum (0.1 mm Hg) overnight to provide 1.02 g (100%) of 164 as a colorless oil: 1 H NMR (300 MHz CDCI 3 6 7.26 (d J= 8.3 Hz 2H) 6.88 (d J= 8.5 Hz 2H) 5.

34 -5.46 (m 2H) 4.45 (s 2H) 4 3 8 (appq 1H) 3.80 (s 3H) 0 3.45 (t J= 7.0 Hz 2H) 2 .35 (m 2H) 1.

3 8 -1.56(m 2H) 0.

84 -0.88 (m 12H) 0.05 (s 3H) 0.02 (s 3H) C NMR (75 MHz CDC1) 159.1 135.8 130.4 129.2 124.6 113.7 72.6 70.2 69.4 55.2 31.3 28.6 25.8 18.2 9.8 -4.4 4 .8ppm;IR(film)2967 2856 1616 1514 1464 1250 1098 836 cm [a]25 14.40 [a]57 15.7 o [a]2546 18.4° [a]25 43 33.5 [a] 25 405 42.40 (c 1.98 CHC1 3 Anal. Calcd for C 2 1

H

36 03Si: C 69.18; H 9.95.

Found: C 69.30; H 10.03.

OTBDMS

HO,

C13H2802Si FW 244.45 (S)-(Z)-5-(t-Butldimethvlsiloxv)-3-heptenol (compound 165). A solution of compound 164 (0.17 g 0.47 mmol) DDQ (0.16 g 0.68 mmol) and 20:1 CH 2

C

2

-H

2 0 (3 mL) was maintained at 23 0 C for 2 h. The reaction mixture was quenched by pouring into Et 2 O mL) and washing with saturated aqueous NaHCO 3 (2 x 5 mL) followed by brine (5 mL). The organic layer was dried (MgSO 4 concentrated and the resulting residue was purified on silica gel (4:1 hexanes-EtOAc) to provide 92 mg (80%)Dof the desired product as a colorless oil: 1 HNMR (500 MHz CDCI 3 )S 5.

4 6 -5.50(m 1H) 5.

3 0-5.36(m 1H) 4.31 (dd J= 14.6 6.7 Hz 1H) 3.63 (dt J= 6.6 2.1 Hz 2H) 2.29-2.34 (m 2H) 1.94 (s 1H) 1.50-1.60 (m 1H) 1.

3 7 -1.24(m 1H) 0.

86 (apps 12H) 0.03(s 3H) 0.01 (s 3H); C NMR(125 MHz 116 CAD) 5 136.6 125.2 70.4 62.1 31.8 31.7 26.1 18.4 10.0 -4.0 -4.1 PPM; IR (filnm)3354 295 z47.40 o]45 59.70 (c =2.30 CDCI 3 Anal. Calcd for C 13

H

28

O

2 Si: C 63.88; H 11.55.

Found: C 63.82; H 11.53.

CI

OTBS

C

13

H

27 OISi FW 354.35 (compound 166). Following the general procedure of Corey (Singh S. et al. J Am. Chem. Soc. 1987 109 6187; Garegg Samuelsson B. J1 Chem.. Soc., Perkin Trans. 1.1980 2866) iodine (2.09 g 8.24 mmol) was added in portions over 15 min to a 0 0 C solution of compound 165 (1.83 g 7.49 mmol) ,,h3 (2.03 g 9.0Ommol)- imidaz ole (0.61 g 8 9 9 mmol) and Et 2 O-vjeCN (3:1 40 m1)and then allowed to warm to 23 0 C. After 1.5 h the solution was diluted with 1: 1 hexanes-EtOAc' (200 mL) then filtered through basic alumina (activity-IV) and concentrated. The resulting mixture was flushed though a plug of silica gel (9:1 hexane-Et 2 O) to yield 2.5 g of the desired product as a colorless oil which was used without any further purification: 111 NMR (300 MII-z CDCI 3 5 5.46-5.52 (mn 11-) 5.23-5.32 (m 11) 4.25 (dd J= 14.4 6.6 Hz 111) 13 3 .11-3.16 (m 2H) 2 6 0- 2 .68 (m 2H1) 1.

3 7 -1.60 (m 2H1) 0.

84 -0.89 (m 121H); C NM MiHz CDCI 3 136.2 127.0 70.2 32.0 31.3 25.8 18.2 9.8 4.6 -4.3 -4.7 ppm; JR (film) 3612 29 57 2530 2857 1699 1650 1252 cmi; 25 D= 21.9' [a] 25 77 =22.6' [a2554 26.20 [a 25 435 49.5' [a 25 405 62.20 (c =2.00 CHCI 3 Anal. Calcd. for C 13

H

27 OIS1: C 44.07; H 7.68. Found: C 44.24; H- 7.64.

0

OTIPS

C

16

H

33 OISi FW 396.43 (S)-(Z)-J-Iodo-5-(risopro~v~silox)3heptene- Following the general procedure of Corey (Singh S. et al. J Am. Chem. Soc. 1987 109 6187; Garegg Samuelsson B. 1.

Chem. Soc., Perkin Trans. 11980 2866) iodine (0.80 g 3.5 mmol) was added in portions IDover 15 min to *a 0 0 C solution of (.S)-(Z)-5-(triisopropylsiloxy)..3.heptenol (0.900 g 3.14 mmol) 3 (0.78 g'3.5 mmol) iniidazole (0.24 g 3 .5 mmol) and Et 2 O-MeCN (3:1 5 mL) and then allowed to warm to 23'C. After 1.5 h the solution was diluted with 1: 1 hexanes- EtOAc (50 mL) then filtered through basic alumina (activity-I-V) and concentrated. The resulting mixture was flushed though a plug of silica gel (9:1 hexane-Et 2 O) to yield 1.29 g of the desired product as a colorless oil which was used without any further purification: I H NNvI(500M-lz CDCl 3 65.49-5.53 (m IlH) 5.

2 8 -5.32 (m 1ff) 4.41 (dd 1=7.1 5.9Hz IH) 3 .l0- 3 .14(m 2H-) 2 .5 9 2 .66(mr 211) 1.58-1.62(m 1ff) 1.

4 8 -1.52(m 1ff) 1.05(s 21$) 0.86(t 1=7.4Hz 13 CNMR(I25MIHz CDCI3) 136.2 126.9 70.0 32.2 311.6 18.1 12.3 9.3 4 .4 ppm; IR (filmn) 3012 2942 1464 1105 883 cm-; 22.80 [01 25577:. 24.40 [cc] 25546: 23.70 2545 53.10 25405: 65.80 (c 1.2 CHCI 3 Anal. Calcd for C 6

H

33 OSi1: C 48.48; H 8.39. Found: C 48.63; H 8.49.

TBDMSO HN'% zCHO CO2AIyI

C

24

H

40

N

2 0 5 Si fFW 464.68 (aR. 7S)-4-(A Ilvloxvcarbonyl)-.2.4a, 5.6, 7-hexahvdro-3-f4S)-4-(t- 0 b utvldimethvlsiloxvpentvl)l-1-oxo- 7 2 -oxvethvl)Dvrrololl.2-clpvrimidine. Dess-Martin periodinane (Dess D. Martin J. C. J Org. Chem., 1983 48 4155) (0.50 g 1.2 mnol) was added to a 23 0 C solution of compound 153a (0.46 g 1 mmol) and CH 2 Cl 2 (10 mL). After 1 h the reaction mixture was poured into Et 2 O (50 mL) and washed with saturated aqueous Na 2

S

2

O

3 (2 x10 mL) 1 N NaOH (2 x 10 mL) and brine (10 mL). The organic layer was dried (MgSO) concentrated and purified on silica gel (1:1 hexanes-EtOAc) to yield 0.404 g of desired product as a colorless oil: 'HNMR (500 MHz CDC 3 89.73 (s 11) 8.21 (s IH) 5.88-5.95 (m 111) 5.30 (dd J= 16.7 1.2 Hz 11) 5.22 (d J= 10.4 Hz 111) 4.60 I0 (ddd J= 22.6 13.1 5.9 Hz 2H) 4 .29-4.35(m 21) 3.75-3.78(m LR) 3.15 (dd J= 16.7 3.8 Hz 1I) 2.57-2.62(m 1H) 2.52-2.56(m 21) 2.44-2.51 (m 111) 2 .11- 2 .14(m IH) 1.62-1.73 (m 1.57-1.61 (m 1H) 1.53-1.56 (m 1H) 1.39-1.45 (m 2H) 1.09 (d J= Hz 3H) 0.85 (s 91) 0.02 (s 31) 0.01 (s 31); 13C NMR (75 MHz CDC1 3 5 200.0 165.3 152.6 152.1 132.3 118.5 101.0 68.4 64.8 58.2 51.0 48.4 39.1 31.1 30.6 29.6 25.8 24.6 23.7 18.1 -4.5 -4.7 ppm; IR (film) 3218 3096 2955 2856 2730. 1722. 1679 1630 1439 1253 836 775 734 cm- 1 [a]25D: -35.40 [a]5 577: -35.5O 2546: -44.61 45: -61.6' 405: 19.8' (c 1.85 CHCI3. Anal. Calcd for C 2 4

H

40

N

2

O

5 Si: C 62.04; H 8.68; 6.03. Found: C 61.75; H 8.68; N 6.00.

211 HN J N

H

CO cly, C1 8

H

2 8

N

2 0 FW 352.43 (3R.4R.4aR 6 7S)-4-(A llyloxvcarbonyl) -1,2,4a,5, 6, 7-hexahydro- 7-(2-hydroxvethyl)-J- ,oxorroloJJ.2-clprimidine-3-spiro-6'-(2 '-methvl)-3 '4'.56'-1etrahvdro-2H- yran (compound 168). A solution of compound 153a (0.486 g, 1.04 mmol), PPTS (0.262 g, 1 .04 mmol), and MeOH (2OmnL) was heated at 50TC for 5 h. The resulting solution was concentrated, flushed through a plug of silica gel (20:1 EtOAc-MeOH), and concentrated.

Z The resulting residue was dissolved in a solution of CHI-C 3 andp-TsOI- (45 mg, 0.24 mmol), C7* 5 which was maintained at 23TC for 1 h, then poured into Et 2 O (60 mL). The solution was washed with saturated aqueous NaHCO 3 (2 x 10 brine (10 dried (MgSO 4 and concentrated to yield 0.345 g 168 as a slightly yellow oil, which was used without (Ni further purification: 'H NMR(500 MJ-z, CDCI 3 566.2 6 I 5.84-5.92 (in, 1 5.3 2 (d, J =J 17.4 Hz, I 5.22 J =10.4 Hz, I1H), 4.62 (ddd, J= 21.1, 12.5, 6.2 Hz, 2H), 4.3 3 (s, 1lH), 4.13-4.33 (in, I1H), 4.02 (dt, J 11. 1, 5.0 Hz, I1H), 3.77-3.80 (in, I1H) 3.53-3.58 (mn, ri2H), 2.32 J =11. 1 Hz, I 2 .1 3 2 2H4), 1.98-2.03 (in, I1H), 1.52-1.74 (in, 8H), 13 1.05-1.09 (in, I1H), 1.02 J =6.1 Hz, 3H); C NMIR (125 MI-Iz, CDCI 3 6 168.4, 154.6, 131.7, 118.5, 82.5, 66.2, 65.4, 59.2, 55.3, 54453639.7,32.4,32.2, 30.3, 29.4, 21.7, 18.6 ppm; IR (film) 3297, 3084, 2934, 1731, 1659, 1633, 1480, 1012, 733 cm- 1 [cCI 25 D: 1390, [ax]2 577: 145', 25546: 1660, 25 4 3 5: 2850, 25405: 3450, (c 2.25, CHCI 3 Anal. Calcd for C 8

H

28

N

2 C, 61.34;1 H, 8.00; N, 7.95. Found: C, 61.08; H, 8.08; N, 7.78.

0 '..CHO HN N 200

H

CO2AIyI

C

18

H

26

N

2 0 FW 350.42 (3R. 4R ,4aR. 6 R. 7S) 4 -(Allloxvcarbonyi)l42,4a 56. 7-hexahvdro-1 -oxo- 7-(2 -oxvegh vi)- Dyrrolofl2-cpvrimidine3sDiro-6.(2 '-methyl)-3 6 t -tetrahydro- 2H-Pyran (compound 169). Dess-Martin periodinane Dess, D. Martin, J. C. J Org. Chem. 1983, 48, 4155) (0.72 g, 1 .7 mmol) was added to a 23 0 C solution of compound 168 (0.500 g, 1 .42 inmol) and CH,C1 2 (35 mL). After I h the reaction mixture was poured into Et 2 O (100 inL) and washed with saturated aqueous Na 2

S

2

O

3 (2 xl10 1 N NaOH (2 x 20 mL), brine (20 mL). The organic layer was dried (MgSO 4 concentrated, and purified on silica gel (EtOAc; 20:1 EtOAc-MeOH) to yield 0.404 g (8 of compound 169 as a colorless-oil: 'H NMR (500 z MHz, CDCI 3 6 9.66 I 6.84 I1H), 5.79-5.87(in, I1H), 5.28 (d,J=.17.1 Hz, 1H), 5.16 S 5 J= 10.5 Hz, IH), 4.59-4.63 (mn, 1H) 4.51-4.55 (mn, 1H) 4.32 J= 12.5, 7.9 Hz, lH), 4.00 (dt, J= 11.2, 4.7 Hz, 11-1), 3.76 (dd,J= 11.1, 5.9 Hz, 1H), 3.09 (dd, J= 16.7, 4.1 Hz, IH), 2.33 (dd, 16.7, 7.9 Hz, 2.28 11.2 Hz, 1H), 2.09-2.13 (in, 1.96-2.07 (in, 1.82 (dd, J= 25 12.2 Hz, I 1.39-1.64 (in, 5H), 1.00-1.04 (mn, I 0.97 J= 13 6.1 Hz, 3H); C NNM(125 Mflz, CDCI 3 200.2,168.3, 153.3, 131.6, 118.2,82.1,66.0,65.2, ~t 10 54.7, 53.8, 51.8, 48.6, 32.1, 31.9, 29.5, 29.3, 21.5, 18.2 ppm; JR (film) 3229, 3079, 2932, 2730, 1732, 1660, 1651, 1470, 1014, 733 cm- 1 210 5 [x57: 11 P5, [a]2 5 4 6 1320, 435 238', 25405: 299', (c =2.50, CHCI 3 Anal. Cacid for C 18

H

26

N

2 0 5 C,61.70; H, 7.48; N, 7.99. Found: C, 61.80; H, 7.53; N, 8.06.

OMe CHO N

N

H

C2AIyI

C

1 9

H

28

N

2 FW 364.45 (3R,4R.4aR, 6'R. 7 S)-4-(Allyloxvcarbonyl)..34,4a,S, 6,7-h exah vdro-J -methoxy- 7-(2oxehl-yrlo 'cprmdie3sio6-(2 '-methsyl)-3'.4',5 '6 '-tetrahydro-2H-yyvran (compound 170). A solution of compound 169 (0.285 g, 0.8 13 minol), MeOTf (0.368 mL, 3.26 mmol), 2 6 -di-t-butyl-4-inethylpyridine (0.25 g, 1.22 inunol), and dry CH1 2

CI

2 (5 m.L) was maintained at 23'C for 5 h. The solution was then poured into Et 2 0 (40 m.L) and washed with 1 N NaOH (2 x 10 m.L) and brine (10 dried (Na 2 S0 4 filtered, concentrated, and the resulting residue was purified on 10% pH 7 phosphate buffered silica gel (4:1 hexanes- EtOAc; 3:1 hexanes-EtOAc) to yield 2 00 mg of compound 170 as a colorless oil: I H 121 NMR (500 MII-z, CDC1 3 )5 9.68 I1H) 5.87-5.95 (in, I1H) 5.35 J 17.4Hz, 1H) 5.21 (d, J= 10.5, 1H)4.64-4.68(m, I1H)4.58-4.61 (mn, I1H)4.29-4.33(m, IlH)4.03-4.07(m, I1H)3.87 (dtJ= 11. 1, 5.3 Hz, I H) 3.66 3H) 2.76-2.80 (in, I1H) 2.34-2.39 (in, I1H) 2.14-2.24 (in, 0 2H)2.02-2.1O(m,2H) l.96(dt,J= 12.8,3.9Hz, I1H), l.66(dt,J= 12.8,6.5 Hz, 1H) 1.51- 1.55 (in, 2H) 1.

3 9 -1.46 (mn, 1H) 1.34 J= 12.6 Hz, lH) 1.05 (ddd, J= 13 4 ,11.6, 4.0 Hz, I H) 0.97 J= 6.3 Hz, 3H); 3C NMR (125 MIHz, CDCl1 3 200.5, 170.6, 15 0.3, 13 2.2, 117.8, 84.9, 65.6, 64.9, 56.6, 54.3, 52.5, 51.8, 50.0, 35.0, 33.6, 29.9, 29.2, 22.2, 19.4 ppm; JR (film) 2932, 2725, 1727, 1636, 1455, 1393, 1017, 754 cm- 1 [a]'12 D: 177', 25 577 1850, 2546 23,[a] 4 35: 387', (c =2.00, CHCI 3 Anal. Calcd for C,9H 28

O

5 2: C, 62.62; H, 7.74; N, 7.69. Found: C, 62.36; H, 7.77; N, 7.52.

0 OTBOMS H -N

H

CO

2 AIly

C

31

H

52

N

2 0 6 Si FW 576.86 (3R,4R,4aR. 6 R. 7S)-4 -(AllI voxycarbonyl) -1,2.4a,5, 6.7-hexahvdro-J -oxo- -2oxo- 7 t-butpdimet hyIsiloxv) -5-non en vil-pyrrolo[i 2 -clvrimidine-3spiro6.'(2 '-methyl)- 3 49 .5'W6-Ietrahydro-2H-Pyran ompound 172). t-BuLi (1.83 m.L, 1.44 M inhexanes) was added to a -78'C solution of compound 166 (439 mg, 1.24 minol), Et 2 O (5 mL), and hexanes mL). After 20 min, the solution is cannulated into a -78'C solution of compound 169 (0.20 g, 0.57 mmol) and TI-F (10 inL). After 5 min, the reaction mixture is quenched with saturated aqueous NIH 4 C1(10 ml). The layers were separated, and the aqueous layer extracted with EtO (10 The combined organic layers were washed with brine (5 dried (MgSO 4 and concentrated to yield a slightly yellow oil, which was used without further purification.

The crude oil, Dess-Martin periodinane (48 mg, 0. 11 mmol), and CH 2 C1 2 (10 m.L) were maintained at 23 0 C for 45 min. The reaction mixture was quenched with saturated aqueous z Na 2

S

2

O

3 (10 mL), saturated aqueous NaHCO 3 (10 mL) and Et 2 O (30 mL). The layers were separated and the organic layer was washed with brine (5 mL), dried (MgSO 4 and concentrated to yield a slightly yellow oil, which was purified on silica gel (1:1 hexanes- EtOAc) to obtain 99 mg of the desired product as a colorless oil: 1 NMiR(500HMiz, Cl 3 6 6.31 INH), 5.84-5.90 (in, INH), 5.29-5.34 (mn, 2H), 5.17-5.22 (in, 2H), 4.64-4.68 (in,IN), 4.56-4.59 lH), 4.25-4.30 2H), 4.02 (dt, J=11.2, 5.0 Hz, IN), 3.77-3.81(in, 1D 3.3 8 (dd, J =16.6, 2.1 Hz, INH), 2.33-2.45 (in, 2H), 2.20-2.27 (mn, 4H), 2.00-2.26 (in, 3H), 1.22-1.77'(in, 8H), 1.06-1.09 (in, INH), 1.02 J= 6.1 Hz, 3H), 0.79-0.83 (mn, 12H), 0.00 3H), -0.03 3H); 13C NMR (75 MfHz, CDCI 3 208.2, 168.5, 153.1, 135.0, 131.6, 126.7, 118.4, 52.1, 69.8, 66.0, 65.3, 54.9, 53.8, 53.0, 46.4, 42.6, 32.2, 32.1, 29.4, 28.9, 25.7, 21.7, 21.6, 18.4, 18.1, 9.7, -4.9 ppm; JR (film) 3226, 3079, 2931, 1736, 1717, 1652, 1472, [a25 25 25 81,[]25 1375, 1255, 1084, 1015 crn'; [D 64.6', 77 67.90, I(a] 546 43 781,[a 1440, [U] 25 40 5 1770, (c 2.251, CHCl 3 Anal. Caicd. for C 3 H5 2

O

6

N

2 Si: C, 64.5 5; H, 9.06; N, 4.87. Found: C, 64.39; H, 8.98; N, 4.77.

0 HC0 2

HN

I ~H 0

H

CO2AIyI C26H 39

N

3 0 6 FW =489.62 AIMy ester Compound 8. A solution of compound 1 7 2 (110img, 0.19 rnmol), MeOTf (0.37 piL, 3.3 inmol), 2 6 -di-t-butyl-4-inethylpyridine (10 mg, 0. 05 mrnol), and dry CH 2

CI

2 (8 mnL) 123 rN was maintained at 23 0 C for 12 h. The solution was then poured into Et 2 O (30 mL) and O washed with 1 N NaOH (2 x 5 mL) and brine (5 mL) dried (Na 2 SO4) filtered concentrated Sand the restilting residue was used without further purification.

Anhydrous NH3 was bubbled through a 0 C solution of the crude residue and MeOH (25 mL) in a resealable tube. After 15 min the tube was sealed and heated to 50 0 C. After 2 d the solution was concentrated and the crude residue was used without further purification.

The crude residue TsOH (95 mg 0.50 mmol) and CHCI3 (10 mL) was maintained at 23 0

C.

C1 10 After 8 h the reaction mixture was quenched with saturated aqueous NaHCO 3 (2 mL). The layers were separated and the aqueous was extracted with Et 2 0 (2 x 5 mL). The combined organic layers were dried (MgSO 4 and concentrated to yield a slightly yellow oil which was purified on silica gel (10:1:0.1 CHCl 3 :i-,rOH:HCO 2

H

to obtain 23 mg of the desired product as a slightly yellow oil.

HCO,- HN OHN N 62 H

CO

2

H

C

2 3

H

3 5

N

3 0 4 FW 449.55 PentacyclicAcid Compound 7. A solution of compound 8 (23 mg 0.05 mmol) 3 (4 mg 3 pmol) dimedone (35 mg 0.25 mmol) and THF (1 mL) was maintained at 23 0 C. After min the reaction mixture was concentrated and purified on silica gel (10:1:0.1 CHCI 3 :i- ,rOH:HCO 2 H -4:1 CHCI 3

:HCO

2 H) to obtain 3 mg of the desired product as a slightly yellow oil: HRMS (FAB) m/z 404.2549 calcd for C 22

H

34 0 4

N

3 found 404.2541.

organic layers were dried (MgSO 4 and concentrated to yield a slightly yellow oil, which was purified on silica gel (10:1:0.1 CHCl 3 :i-PrOH:HCO 2 H) to obtain 23 mg of the desired product as a slightly yellow oil.

0 HC02- HN OHN

N

H

C02H 62 C23H 35

N

3 0 4 FW 449.55 PentacvclicAcid Compound 7. A solution of compound 8 (23 mg, 0.05 mmol), Pd(PPh 3 (4 mg, 3 pmol), dimedone (35 mg, 0.25 mmol), and THF (1 mL) was maintained at 23 0 C. After min, the reaction mixture was concentrated and purified on silica gel (10:1:0.1 CHC13:i- PrOH:HCO 2 H 4:1 CHCI 3 :HCOH) to obtain 3 mg of the desired product as a slightly yellow oil: HRMS (FAB) m/z 404.2549 calcd for C 22 H3 4 0 4 N, found 404.2541.

EXAMPLE VII An Improved Synthesis of Pentacyclic Acid This Example provides an improved method for synthesizing pentacyclic acid compounds.

Chemical synthesis procedures are as described above for Example VI. A convergent synthetic strategy for the guanidinium alkaloids is shown in Figure 46.

Figure 52 depicts the synthesis strategy for a new method of preparing pentacyclic acid compounds and the compounds produced, e.g. compounds 176 and 177, using compound 173 as starting material. Compound 61 is an urea compound obtained as shown in Figures 53, 54 N and 55 as follows: 3-butynol (compound 178) is converted to the p-methoxybenzyl

(PMB)

O ether 179 (Figure 53). The alkyne of compound 179 was deprotonated with n-buthyl lithium Z at -40°C and the resulting acetylide treated with anhydrous DMF to provide ynal 180 in yield, after quenching the intermediate -aminoalkoxide into aqueous phosphate buffer (Joumet et al., Tetrahedron Lett., 1988, 39:6427). The C3 stereocenter was introduced by the method of Weber and Seebach (Singh et al., J. Am. Chem. Soc., 1987, 109:6187) through condensation of ynal 180 with Et 2 Zn in the presence of(-)-TADDOL (20 mol%) and Ti(Oi- S Pr) 4 to give (S)-181 in 94% yield and >98% ee. This asymmetric transformation was reliably performed on a 45 g scale. Propargylic alcohol 181 was protected as the triisopropylsilyl (TIPS) ether and the alkyne partially hydrogenated with Lindlar's catalyst to provide cis alkene 182. The PMB-protecting group was oxidatively removed with DDZ and the resulting alcohol was cdnverted to iodide 183 (Kitamura et al., Org. Synth., 1992, 71:1) in an overall yield of 89% from 181.

Enantiopure methyl R-3-hydroxy-7-methyloct-6-enoate (Kitamura et al., Org. Synth., 1992, 71:1) was converted to amide 185 in 88% yield by reaction with N,Odimethylhydroxylarminde hydrochloride according to the procedure ofWeinreb (Garigipati et al., J. Am. Chem. Soc., 1985, 107:7790) followed by protection of the secondary alcohol as the triethylsilyl (TES) ether (Figure 54). Iodide 183 was converted to the corresponding lithium reagent and coupled with 185 to generate dienone 186 in 60-70% yield. Masking the C8 carbonyl of 186 as the ketal was necessary to prevent a B-hydroxy elimination, which occurred under the Mitsunobu conditions employed to install the B-amino functionality.

Ketalization was sluggish, however, when the B-hydroxy group was protected, so optimized reaction conditions were found which cleaved the TES group, did not promote the B-hydroxy elimination of the intermediate B- hydroxy ketone and promoted ketalization. Optimized ketalization conditions involved treatment of 186 with orthoester 187 (Roush and Gillis, J Org. Chem., 1980,45:4283-4287; Baganz and Domascke, Chem. Ber., 1958,91:650-653) and 1,3-propanediol in the presence of Amberlyst-15 to provide ketal 188 in 80% yield.

Mitsunobu displacement of the secondary alcohol with azide followed by reduction to the amine provided compound 189 in 77% yield from compound 188.

SCondensation of amine 189 with TMSNCO yielded urea 190 in 89% yield (Figure Z Amine 189 is used to prepare pentacyclic compound 177 as shown in Figure 52.

These results demonstrate a method that permits preparation of both pentacyclic compounds 7 and 177. Having access to both types of compounds, allows side chains to be attached either before or after epimerization of the ester of the pentacyclic compound.

EXAMPLE VIII This example describes the in vitro screening of 60 tumor cell lines against the compounds of the invention: ptilomycalin A, isocrambescidin 800 trihydrochloride, triacetylcrambescidin 800 chloride, crambescidin 657 hydrochloride, crambescidin 800 trihydrochloride, triacetylisocrambescidin 800 chloride, and 13-epiptilomycalin A to determine anti-tumor activity.

The screening methods utilized the National Cancer Institute (NCI) DTP Human Tumor Cell Line Screen protocol as described by Monks et al., J. Nat'l. Cancer Inst. 83:757-766 (1991); and Boyd In "Cancer Drug Discovery and Development, Vol. 2; Drug Development; Preclinical Screening, Clinical Trial and Approval, Humana Press, 1997, pp 23-43. The origins and processing of the cell lines used are described in Alley et al., Cancer Res., 1988, 48:589-601; Shoemaker et al., Prog. Clin. Biol. Res., 1988, 276:265-286; and Stinson et al., Proc. Am. Assoc. Cancer Res., 1989, 30.613.

In brief, in the screen protocol, cell suspensions were diluted depending on cell type and the expected target cell density (approximately 5000-40,000 cells per well) into a 96 well microtiter plate. Inoculates were preincubated for 24h at 37 OC for stabilization. Dilutions at twice the intended test concentrations were added at time zero in 100 pL aliquots to the microtiter plate wells. Test compounds were evaluated at five 10-fold dilutions. Routine test concentrations have the highest well concentration at 1 0E-4M, but for the standard agents, the highest well concentration used depended on the agent used. Incubations lasted 48 hours in 127

CO

2 atmosphere and 100% humidity. The cells were assayed by Sulforhodomine B assay 0 as described by Rubenstein et al., JNCI, 1990, 82:1113-1118 and Skehan et al., JNCI, 1990, Z 82: 1107-1112. Optical densities were read with a plate reader and the data processed using a microcomputer into special concentration parameters.

c NCI renamed the IC50 value, the concentration that causes 50% growth inhibition the value to emphasize the correction for the cell count at time zero; thus GI50 is the concentration of test drug where 100 X 50 (Boyd et al., In Cytotoxic 0 Anticancer Drugs: Models and Concepts for Drug Discovery and Development, Vleriote et S 10 al., Eds., Kluwer Academic, Hingham, MA, 1992, pp 11-34; and Monks et al., JNCI. 1991, 83, 757-766. The optical density of the test well after a 48 hr period of exposure to the test compound is the optical density at time zero is TO and the control optical density is The "50" is called the GI50PRCNT, a T/C-like parameter that can have values from +100 to -100. The GI50 also measures the growth inhibitory power of the test compound. The TGI is the concentration of test drug where 100 X Thus, the TGI signifies a cytostatic effect. The LC50 which signifies a cytotoxic effect, is the concentration of the test compound where 100 X (T-TO)/TO=-50. The control optical density is not used in the calculation of These concentration parameters are interpolated values. The concentrations giving G150PRCNT values above and below the reference values 50 for G150) are used to make interpolations on the concentration axis. Currently, about 45% of the G150 records in the database are "approximated." In 42% of the records, the GI50PRCNT for a given cell line-does not go to 50 or below. For mean graph purposes, the value assumed for the Gl 50 in such a case is the highest concentration tested (HICONC). Similar approximations are made when the G150 cannot be calculated because the G150PRCNT does not go as high as 50 or above of total). In this case, the lowest concentration tested is used for the G150.

Corresponding approximations are made for the TGI and for the The results of the tumor cell screening are shown in the mean graphs of 56-62.

N 0 Nc The mean graphs are a presentation of the in vitro tumor cell screen results developed by the SNCI to emphasize differential effects of test compounds on various human tumor cell lines Z (Boyd et al., In Cancer:Principles and Practice of Oncology, DeVita et al., Eds., Lippincott, Philadelphia, PA, 1989, Vol. 3, pp. 1-12; Paull et al., JNCI, 1989, 81:1088-1092; and Paull et al., Proc. Am. Assoc. Cancer Res., 1988, 29:488. The mean graph bar graphs depict patterns 0 created by plotting positive and negative values generated from a set of G1 50, TGI or CN values. The positive and negative values are plotted against a vertical line that represents the Smean response of all the cell lines in the panel to the test compound. Positive values project to the right of the vertical line and represent cellular sensitivities to the test agent that exceed

O

the mean. Negative values project to the left and represent cell sensitivities to the test compound that are less than the average value. The positive and negative values, called "deltas", are generated from the G150 data (or TGI or LC40 data) by a three-step calculation.

The G150 value for each cell line tested against a test compound is converted to its logl0 G150 value. The logio G150 values are averaged. Each loglo G150 value is subtracted from the average to create the delta. Thus, a bar projecting 3 units to the right denotes that the G150 (orTGI or LC50) for that cell line occurs at a concentration 1000 times less than the average concentration required for all the cell lines used in the experiment. Thus the cell line is usually sensitive to that compound. If for a particular compound and cell line it was not possible to determine the desired response parameter by interpolation, the bar length shown in either the highest concentration tested (and the listed loglo of the response parameter will be preceded by a or the lowest concentration tested (and the listed loglo will be preceded by a The, values at either limit or are also calculated in the mean used for the meangraph. Therefore, the mean used in the meangraph may not be the actual mean of the G150 for instance. For this reason, this value is referred to as the MgMID (meangraph midpoint).

These results demonstrate that certain cancer cell lines are sensitive to ptilomycalin A, triacetylcrambescidin 800 chloride, crambescidin 657 hydrochloride, crambescidin 800 trihydrochloride, and 13-epiptilomycalin A. Isocrambescidin 800 trihydrochloride and triacetylisocrambescidin 800 chloride display a lesser effect on the cell lines tested.

Claims (19)

1. A compound of the formula: Wherein, R= H, a carboxylic acid protecting group, an o-alkoxycarboxylic acid or an o- alkoxycarboxylic acid ester, and X= any pharmaceutically acceptable counterion.

2. A compound of the formula: Wherein, R= H, a carboxylic acid protecting group, an co-alkoxycarboxylic acid or an alkoxycarboxylic acid ester, and X= any pharmaceutically acceptable counterion.

3. A compound of the formula: S Wherein, R= a carboxylic acid protecting group, an (o-alkoxycarboxylic acid or an co- alkoxycarboxylic acid ester, and X= any pharmaceutically acceptable counterion.

4. A compound of the formula: Wherein, R= H, a carboxylic acid protecting group, an o-allcoxycarboxylic acid or an w- alkoxycarboxylic acid ester, and X= any pharmaceutically acceptable counterion. A compound of the formula: Wherein, R= H, a carboxylic acid protecting group, an o-alkoxycarboxylic acid or an o- alkoxycarboxylic acid ester, and X= any pharmaceutically acceptable counterion.

6. A compound of the formula:

7. A compound of the formula:

8. A compound of the formula:

9. A compound of the formula: A compound of the formula:

11. A method for synthesizing a pentacyclic compound of the formula: O O Wherein, R= H, a carboxylic acid protecting group, an co-alkoxycarboxylic acid or an o- alkoxycarboxylic acid ester, and SX= any pharmaceutically acceptable counterion O 1 10 which method comprises reacting a compound of the formula: GO 2 wherein G= a carboxylic acid protecting group, an o-alkoxycarboxylic acid or and c-alkoxycarboxylic acid ester, and Y= alcohol protecting group with a compound of the formula: OP X 2 NHQ wherein X 2 0 or ketone protecting group Z= alkene or carbonyl protecting group P- alcohol protecting group and Q= amino carbonyl group to produce a compound of the formula: O Swherein X 2 0 or ketone protecting group P= alcohol protecting group, and R carboxylic acid protecting group, co-alkoxycarboxylic acid C^'1 5 M, or o-alkoxycarboxylic acid ester which compound is subsequently converted to the pentacyclic compound by 0 deprotection, incorporation of ammonia, and cyclization.

12. The method of claim 11, wherein when R a carboxylic acid protecting group, the method further comprises the step of deprotecting the pentacycle compound of claim 11.

13. A method for synthesizing a pentacyclic compound of the formula Wherein, R= H, a carboxylic acid protecting group, an oD-alkoxycarboxylic acid or an co- alkoxycarboxylic acid ester, and X= any pharmaceutically acceptable counterion, which comprises epimerizing the stereocenter at carbon-14 of the compound of the formula: O O

14. The method of claim 13, wherein when R= a carboxylic acid protecting group, the S method further comprises the step of deprotecting the pentacycle compound of claim 13. C- A method for synthesizing pentacyclic compounds B and C of the formulae: H H 1 sCO 2 R B C Wherein, R= H, a carboxylic acid protecting group, an o-alkoxycarboxylic acid or an o- alkoxycarbQxylic acid ester, and X= any pharmaceutically acceptable counterion, which comprises reacting a compound of the formula: wherein G= a carboxylic acid protecting group, an co-alkoxycarboxylic acid or an co-alkoxycarboxylic acid ester, and Y= an alcohol protecting group O _with a compound of the formula: OP 2 NHQ C wherein X 2 0 or a ketone protecting group Z= an alkene or carbonyl protecting group P= an alcohol protecting group, and Q= an amidinyl group To produce a compound of the formula: .PP H H O' OP HN N H wherein X 2 O or a ketone protecting group P= an alcohol protecting group and R= a carboxylic acid protecting group, an o-alkoxycarboxylic acid or an co-alkoxycarboxylic acid ester which is subsequently converted to the pentacyclic compound by deprotection and cyclization.

16. The method of claim 15, wherein when R= a carboxylic acid protecting group, the method further comprises the step of deprotecting the pentacycle compound B of claim

17. The method of claim 15, wherein when R= a carboxylic acid protecting group, the method further comprises the step of deprotecting the pentacycle compound C of claim

18. A method for synthesizing a pentacyclic compound of the formula: R= H, a carboxylic acid protecting group, an o-alkoxycarboxylic acid or an o- alkoxycarboxylic acid ester, and X= any pharmaceutically acceptable counterion. which comprises epimerizing the stereocenter at carbon-14 and carbon 15 of the compound of the formula:

19. The method of claim 18, wherein when R= a carboxylic acid protecting group, the method further comprises the step of deprotecting the pentacycle compound of claim 18. The compound of claim 1, 2, 3, 4, or 5 wherein R= allyl and X= Cl

21. The compound of claim 1, 2, 3, 4, or 5 wherein R=H, and X= C1-.

22. The compound of claim 1, 2, 3, 4, or 5 wherein R= (CH 2 15 C0 2 G, Wherein G=H, a counterion of a carboxylate salt, or a carboxylic acid protecting group, and X= CI-

523. The compound of claim 1, wherein Rz= (CH 2 15 C0 2 H and X ClF. 24. The compound of claim 2, wherein R= (CH 2 5 C0 2 H and X =Cl-. The compound of claim 3, wherein, R= (CH 2 )15CO 2 H and X =cL. LI0 26. The compound of claim 4, wherein R= (CH 2 15 C0 2 H and X =CU-. 27. The compound of claim 5, wherein R= (CH 2 15 C0 2 H and X =ClF. 28. A compound of the formula: wherein RI= any ailkyl, aryl or substituted alkyl group R 2 OH, 0GI, a spermidine moiety or a substituted spermidine moiety wherein G, a carboxylic acid protecting group and X= any pharmaceutically acceptable counterion. 29. A compound of the formula: wherein any alkyl, aryl or substituted alkyl group R 2 OH, 0G 1 a spermidine moiety or a substituted sperniidine moiety wherein G 1 a carboxylic acid protecting group and X= any pharmaceutically acceptable counterion. 30. A compound of the formula: wherein Rj= any alkyl, aryl or substituted alkyl group R 2 OH, 0GI, a spermidine, moiety or a substituted sperniidine moiety wherein G, a carboxylic acid protecting group and X= any pharmaceutically acceptable counterion. O O 00 31. A compound of the formula: wherein RI= any alkyl, aryl or substituted alkyl group R 2 O, OH, OGI, a spermidine moiety or a substituted spermidine moiety wherein G =carboxylic acid protecting group, and X= any pharmaceutically acceptable counterion. 32. A compound of the formula: wherein Rj= any alkyl, aryl or substituted alkyl group R2 OH, OGI, a spermidine moiety or a substituted spermnnidine moiety 1 wherein G, =carboxylic acid protecting group and X= any pharmaceutically acceptable counterion. 33. The method of claim 11, wherein when R is an w-alkoxycarboxylic acid, the method further comprises the step of reacting the pentacyclic compound of the formula: wherein, Ri= any alkyl, aryl or substituted alkyl group with a protected spermidine or a protected substituted sperimidine and subsequently deprotecting to produce the compound of the formula: wherein RI any alkyl, aryl or substituted alkyl group R 2 a spermidine moiety or a substituted spermidine moiety and X= any pharmaceutically acceptable counterion. i f 34. The method of claim 13, wherein when R is an o-alkoxycarboxylic acid the method further comprises the step of reacting the pentacyclic compound of the formula: with a protected spermidine or a protected substituted sperimidine and subsequently deprotecting to produce the compound of the formula: wherein RI= any alkyl, aryl or substituted alkyl group R 2 a spermidine moiety or a substituted spermidine moiety, and X= any pharmaceutically acceptable counterion. The method of claim 15, wherein when R is an co-alkoxycarboxylic acid the method further comprises the step of reacting the pentacyclic compound of the formula: wherein, Ri= any alkyl, aryl or substituted alkyl group with a protected spermidine or a protected substituted sperimidine and subsequently deprotecting to produce the compound of the formula: wherein Ri= any alkyl, aryl or substituted alkyl group R 2 a spermidine moiety or a substituted spermidine moiety and X= any pharmaceutically acceptable counterion. 36. The method of claim 15, wherein when R is an o-alkoxycarboxylic acid the method further comprises the step of reacting the pentacyclic compound of the formula: Y7ilKX-IkIRO OH wherein, Ri= any alkyl, aryl or substituted alkyl group with a protected spermidine or a protected substituted sperimidine and subsequently deprotecting to produce the compound of the formula: wherein R 1 any alkyl, aryl or substituted alkyl group R 2 a spermidine moiety or a substituted spermidine moiety and X= any pharmaceutically acceptable counterion. 0 37. The method of claim 18, wherein when R is an o-alkoxycarboxylic acid the method further comprises the step of reacting the pentacyclic compound of the formula: Ca 1 4 7P -H X- N wherein, Ri= any alkyl, aryl or substituted alkyl group with a protected spermidine or a protected substituted sperimidine and subsequently deprotecting to produce the compound of the formula: OR R2 wherein RI= any alkyl, aryl or substituted alkyl group R 2 a spermidine moiety or a substituted spermidine moiety and X= any pharmaceutically acceptable counterion. 38. A method for synthesizing Ptilomycalin of the formula: ptilomycalin A which comprises reacting the pentacyclic compound of claim 22 with the compound of the formula: H S NHR2 HI NHR 2 wherein R 2 an amine protecting group to produce a compound of the formula: which is subsequently deprotected to produce Ptilomycalin A. 39. A method for synthesizing Crarnbescidin 800 of the formula: N N H, crambescidin 800 which comprises reacting the pentacyclic compound of claim 22 with the compound of the formula: H NR NHR 2 OH wherein R 2 an amine protecting group to produce a compound of the formula: H C FH OH which is subsequently deprotected to produce Cranibescidin 800. O O 0 0 NH2 0-6 A method for synthesizing 13, 14, 15-Isocrambescidin 800 of the formula: ^N tN NH SX-O OH 0 13,14,15-isocrambescidin 800 which comprises reacting the pentacyclic compound of claim 24 with the compound of the formula: H N' NHR2 ,I^^NHR 2 OH wherein R2 an amine protecting group to produce a compound of the formula: H ,IH 'ii N NHR 2 1 4 NHR 2 C OdH which is subsequently deprotected to produce 13, 14, 15-Isocrambescidin 800. 41. An antitumor composition comprising a compound of claim 1, 2, 3, 4, 5, 6, 7, 8, 9 or in admixture with a pharmaceutically acceptable carrier. 42. An antiviral composition comprising a compound of claim 1,2,3, 4, 5, 6, 7, 8,9, or 10 in admixture with a pharmaceutically acceptable carrier. 149 0 O 43. An antifungal composition comprising a compound of claim 1, 2, 3, 4, 5, 6, 7, 8, 9 or in admixture with a pharmaceutically acceptable carrier. 44. A method for treating tumors comprising administering to a subject in need of said treatment, an effective amount of compound of claim 1, 2, 3, 4, 5, 6, 7, 8, 9 or i 45. A method for treating viral infections comprising administering to a subject in need of Ssaid treatment, an effective amount of compound of claim 1, 2, 3, 4, 5, 6, 7, 8, 9 or 46. A method for treating fungal infections comprising administering to a subject in need of said treatment, an effective amount of compound of claim 1, 2, 3, 4, 5, 6, 7, 8, 9 or DATED: 18 November 2004 PHILLIPS ORMONDE FITZPATRICK Attorneys for: The Regents of The University of California ,9 j s.