Approaches to the Synthesis of Cytochalasans. Part 11. Further transformations and cyclization attempts directed towards proxiphomin

Starting from iodoalcohol 9 , the monoprotected dialdehyde 5 was synthesized (Scheme 2) and converted to 17 by reaction with oxo-phosphonate 15 (Scheme 3). The latter was prepared from 13 . Cyclisation of 17 to the target compound 18 failed. Also the attachment of thiol 22 to lactone 19 was unsatisfactory (Scheme 4). Therefore, the building blocks 28 and 29 were synthesized using diene 33 and diester 30 as starting material for 28 and 9 for 29 (Scheme 5 and 6). Hydroxy acid 28 was converted into formyl-ester 46 (Scheme 7). However, the condensation of its derivatives 48 and 49 with ‘Umpolung’ of the carbonyl reactivity was unsuccessful, probably due to steric hindrance.     

Synthesis of (±)‐Kempa‐6,8‐dien‐3‐ol (=(2aRS,3SR,4aSR,7RS,7aSR,10bSR,10cSR)‐2,2a,3,4,4a,5,6,7,7a,8,10b,10c‐Dodecahydro‐2a,7,10,10c‐tetramethylnaphth[2,18‐cde]azulen‐3‐ol)

The synthesis of kempa‐6,8‐dien‐3β‐ol ( 4a ), as a synthetic leading model of the natural product 4b , was carried out starting from intermediate 12 , the synthetic route of which has been developed previously (Scheme 1). The conversion of 12 to the model compound 4a involved the elaboration of three structure modifications by three processes, Tasks A, B, and C (see Scheme 2). Task A was achieved by epoxy‐ring opening of 41 with Me₃SiCl (Scheme 9), and Task B being performed by oxidation at the 13‐position, followed by hydrogenation, and then epimerization (Schemes 4 and 5). The removal of the 2‐OH group from 12 (Task C) was achieved via 30b according to Scheme 6, whereby 30b was formed exclusively from 30a / 31a 1 : 1 (Scheme 7). In addition, some useful reactions from the synthetic viewpoint were developed during the course of the present experiments.     

Studies Directed Towards the Biosynthesis of the C7N Unit of Rifamycin B: A New Synthesis of Quinic Acid from Shikimic Acid

The synthesis of quinic acid ( 4 ) via epoxide 13 , starting from shikimic acid ( 5 ), is described (Scheme 1). Treatment of 13 with thiophenol yielded not only 17 , but also the γ-lactones 18 and 19 as result of migration of silyl groups within a cis- and trans-diol system. The conversion provides a direct stereoselective epoxidation of a shikimic-acid derivative as well as an alternative pathway for the preparation of 4 . A shorter approach via the disilylated epoxide 22 was unsuccessful because the γ-lactone 25 was obtained in place of the desired α-hydroxy ester 24 (Scheme 2).     

Synthese von 3-(2-Carboxy-4-Pyridyl)- und 3-(6-Carboxy-3-pyridyl)-DL-alanin

Synthesis of 3-(2-Carboxy-4-pyridyl)-and 3-(6-Carboxy-3-pyridyl)-DL-alanine As starting materials for potential photochemical approaches to betalaines C(R = COOH) and to muscaflavine F(R = COOH), β-(2-carboxy-4-pyridyl)- and β-(6(carboxy-3-pyridyl))-DL-alanine ( A and D with R = COOH or 4 and 11 ), respectively, were prepared (Scheme 1). The synthesis of 4 (= A, R = COOH) started with the 2-[(4-pyridyl)methyl]malonate 1 and proceeded via the N-oxide 2 , cyanation and hydrolysis (Scheme 2). Amino acid 11 was obtained from (3-pyridyl)methyl-bromide ( 6 ) via the malonate 7 by an analogous sequence of reactions (Scheme 3).     

A Facile Synthesis of Optically Pure (−)-(S)-Ipsenol Using a Chiral Titanium Complex

A stereocontrolled synthetic route to optically pure (?)-(S)-ipsenol ( 1 ), the pheromone of Pityokteines curvidens and various other bark-beetle species is described. Key step of the synthesis is an enantioselective aldol reaction using a chiral titanium–carbohydrate complex (Scheme 1). The carboxylate function of the optically pure β-hydroxy acid 5 thus obtained in mol quantities is then elaborated to the diene moiety by standard methodology (Scheme 2).     

Stereoselektive Synthesen von (Z)-(10-Methoxy-4H-benzo[4,5]cyclohepta[1,2-b]thiophen-4-yliden)essigsäure

Stereoselective Syntheses of (Z)-(10-Methoxy-4H-benzo[4,5]cyclohepta[1,2-b]thiophen-4-ylidene)acetic Acid Two stereoselective syntheses for the antiinflammatory compound 1 ((Z)-isomer) are described. In the first approach (Strategy A, Scheme 1) the stereoselective synthesis of 1 was realized via the bicyclic compound 11 under thermodynamic conditions, followed by a thiophene annelation with retention of the double-bond geometry (Schemes 2–4). Optimized conditions were necessary to avoid (E/Z)-isomerization during annelation. In the second approach (Strategy B, Scheme 1), diastereoisomer 17b was obtained selectively from a mixture of the diastereoisomers 17b and 18b by combining thermodynamic epimerization and solubility differences (Scheme 5). Diastereoisomer 17b was converted into the tricyclic compound 23 using a novel thiophene annelation method which we described recently (Scheme 6). In a final step, a stereospecific ‘syn’-elimination transformed the sulfoxide 24 into the target compound 1 (Scheme 7). To avoid (E/Z)-isomerization, it was necessary to trap the sulfenic acid liberated during the reaction. The key reactions of both approaches are highly stereoselective (> 97:3).     

Synthesis of 3-phenyl-2H-1,4-benzoxazin-2-one. Revision of some structural assignments

The reported synthesis of 3-phcnyl-2H-1,4-benzoxazin-2-one (II) via brominution of o-acel-amidophenyl phcnaeyl ether (Scheme) leads in lael to the 7-bromo-3-phenyl-2H-1,4-benzoxazin-2-ol (VIII). The structure of the other synthetic intermediates is also revised and a one-step synthesis of the lactone II is reported hy condensation of methyl phenylglyoxalale and o-amino-phenol.     

Oligosaccharide Analogues of Polysaccharides. Part 1. Concept and synthesis of monosaccharide-derived monomers

It is proposed to study the influence of interresidue H-bonds on the structure and properties of polysaccharides by comparing them to a series of systematically modified oligosaccharide analogues where some or all of the glycosidic O-atoms are replaced by buta-1,3-diyne-1,4-diyl groups. This group is long enough to interrupt the interresidue H-bonds, is chemically versatile, and allows a binomial synthesis. Several approaches to the simplest monomeric unit required to make analogues of cellulose are described. In the first approach, allyl α-D -galactopyranoside ( 1 ) was transformed via 2 and the tribenzyl ether 3 into the triflate 4 (Scheme 2). Substitution by cyanide (→ 5–7 ) followed by reduction with DIBAH led in high yield to the aldehyde 9 , which was transformed into the dibromoalkene 10 and the alkyne 11 following the Corey-Fuchs procedure (Scheme 3). The alkyne was deprotected via 12 or directly to the hemiacetal 13 . Oxidation to the lactone 14 , followed by addition of lithium (trimethylsilyl)acetylide Me₃SiC?CLi/CeCl₃ (→ 15 ) and reductive dehydroxylation afforded the disilylated dialkyne 16 . The large excess of Pd catalyst required for the transformation 11 → 13 was avoided by deallylating the dibromoalkene 10 (→ 17 → 18 ), followed by oxidation to the lactone 19 , addition of Me₃SiC?CLi to the anomeric hemiketals 20 (α-D /β-D 7:2), dehydroxylation to 21 , and elimination to the monosilylated dialkyne 22 (Scheme 3). In an alternative approach, treatment of the epoxide 24 (from 23 ) with Me₃SiC?CLi/Et₂AlCl according to a known procedure gave not only the alkyne 27 but also 25 , resulting from participation of the MeOCH₂O group (Scheme 4). Using Me₃Al instead of Et₂AlCl increased the yield and selectivity. Deprotection of 27 (→ 28 ), dibenzylation (→ 29 ), and acetolysis led to the diacetate 30 which was partially deacetylated (→ 31 ) and oxidized to the lactone 32 . Addition of Me₃SiC?CLi/TiCl₄ afforded the anomeric hemiketals 33 (α-D /β-D 3:2) which were deoxygenated to the dialkyne 34 . This synthesis of target monomers was shortened by treating the hydroxy acetal 36 (from 27 ) with (Me₃SiC?C)₃Al (Scheme 5): formation of the alkyne 37 (70%) by fully retentive alkynylating acetal cleavage is rationalised by postulating a participation of HOC(3). The sequence was further improved by substituting the MeOCH₂O by the (i-Pr)₃SiO group (Scheme 6); the epoxide 38 (from 23 ); yielded 85% of the alkyne 39 which was transformed, on the one hand, via 40 into the dibenzyl ether 29 , and, on the other hand, after C-desilylation (→ 41 ) into the dialkyne 42 . Finally, combined alkynylating opening of the oxirane and the 1,3-dioxolane rings of 38 with excess Et₂Al C?CSiMe₃ led directly to the monomer 43 which is thus available in two steps and 77% yield from 23 (Scheme 6).     

Synthesen des Analgetikums 2-[1-(m-Methoxyphenyl)-2-cyclohexen-1-yl]-N,N -dimethyl-äthylamin

Syntheses of the Analgesic 2-[1-(m-Methoxyphenyl)-2-cyclohexen-1-yl] -N,N-dimethyl-ethylamine Three principal routes to 2-[1-(m-methoxyphenyl)-2-cyclohexen-1-yl]- N,N-dimethyl-ethylamine (13) , a compound with interesting analgesic properties, are described. In the first, derivatives of [1-(m-methoxyphenyl)-2-cyclohexen-1-yl]acetic acid (10) (alternatively the ethyl ester 29 , the dimethylamide 32 or the nitrile 34 ) serve as crucial intermediates. All three can be synthesized from 2-(m-methoxyphenyl)cyclohexanone (1) by sequences comprising successively C-alkylation ( 1→2,4,5; Scheme 1), reduction of the ketone carbonyl group ( 2→6;4→18;5→19; Scheme 1 and 2) and elimination ( 16→29; 18→32; 19→34; Scheme 2). The relative configuration of the cyclohexanols 16, 18, 19 and of a series of related compounds is established by chemical correlation with the lactone 30 the structure of which follows from ¹H-NMR. data (Scheme 2). The second route creates the intermediates 29 and 32 by ester- or amide-enolate-Claisen-type-rearrangement reactions starting from 3-(m-methoxyphenyl)-2-cyclohexen-1-ol ( 39; Scheme 3). Compounds 29, 32 and 34 are transformed into the target molecule 13 by standard reactions. A Hofmann elimination of the quaternary ammonium fluoride 50 (X=F), derived from the known cis-perhydroindoline 48 , is the essential step in the third approach to 13 (Scheme 4).     

β-Cleavage of Potassium Bicyclo[2.2.2]oct-5-en-2-olates. Stereoselective synthesis of (±)-trichodiene

The transformations of 12 bicyclo[2.2.2]oct-5-en-2-ols ( V or VI ) to 3-(cyclohex-3-enyl)-2-alkanones ( III or IV ), via β-cleavage of their potassium alkoxides in HMPA, has been investigated (cf. Table 1). As an illustration of this synthetic methodology, a stereoselective synthesis of (±)-trichodiene ((±)- 1 ) is described which involves the β-cleavage of the tricyclic potassium alkoxides 46a and 47a to cyclopentanone 4 (cf. Scheme 7).     

Synthesis of New Furo[3,4‐b]quinolin‐1(3H)‐one Scaffolds Derived from γ‐Lactone‐Fused Quinolin‐4(1H)‐ones

In the context of our aim of discovering new antitumor drugs among synthetic γ‐lactone‐ and γ‐lactam‐fused 1‐methylquinolin‐4(1H)‐ones, we developed a rapid access to 5‐methyl‐1,3‐dioxolo[4,5‐g]furo[3,4‐b]quinoline‐8,9(5H,6H)‐dione ( 9 ) exploiting the γ‐lactone‐fused chloroquinoline 10 previously synthesized in our laboratory (Scheme 1). We also elaborated efficient synthetic methods allowing for a rapid access to two nonclassical bioisosteres of 9 , i.e., a deoxy and a carba analogue. The deoxy analogue 11 was prepared in two steps from the γ‐lactone‐fused quinoline 13 which was also the synthetic precursor of 10 (Scheme 1). The carba analogue 6,9‐dihydro‐5‐methyl‐9‐methylene‐1,3‐dioxolo[4,5‐g]furo[3,4‐b]quinolin‐8(5H)‐one ( 12 ) was easily prepared by HCl elimination from the 9‐(chloromethyl)dioxolofuroquinoline 15 , which was obtained via a three‐component one‐pot reaction from N‐methyl‐3,4‐(methylenedioxy)aniline (=N‐methyl‐1,3‐benzodioxol‐5‐amine; 16 ), commercially available chloroacetaldehyde, and tetronic acid ( 17 ) (Scheme 2).     

Approaches to the Synthesis of Cytochalasans. Part 3. Synthesis of a substituted tetrahydroisoindolinone moiety possessing the same relative configuration as proxiphomin

The total synthesis of the tetrahydroisoindolinone moiety corresponding to proxiphomin ( 1 ) is described, bearing functional groups for the attachment of the macrocyclic ring. Knoevenagel-Cope condensation of racemic 2-(benzyloxycarbonyl-amino)-3-phenylpropanal ( 2 ) with methyl (4-methyl-2,4-hexadienyl) malonate ( 3 ) yielded a mixture of the (E)- and (Z)-olefins 4a and 4b , which upon heating underwent intramolecular Diels-Alder cyclization (cf. Scheme 1). From the resulting products the tetrahydroisoindoline derivative 6 was isolated. X-ray analysis of 6 [5] revealed the same relative configurations at C(3), C(4), C(5) and C(8) as in 1 , but not at C(9). Hydrolysis of 6 with KOH was accompanied by a change in configuration at C(9) yielding the hydroxy acid 14 which was converted into the hydroxy ester 11 (cf. Scheme 4). The presence of a cis-anellated lactam ring in 11 has been confirmed by X-ray analysis of its O-acetyl derivative 16 [5]. Ring closure of the hydroxy acid 14 gave the lactone 17 , corresponding to the natural product 1 as to the configuration. The presence of the N-benzyloxycarbonyl group in lactone 6 has been shown to be essential for the above-mentioned ‘inversion’ at C(9), because no configurational change occurred with the N-unprotected lactone 8 when treated under the same conditions. The only product obtained was the hydroxy ester 10 possessing the same configuration at C(9) as 8 . Along with stereochemical considerations, mechanistic aspects of the reactions are discussed.     

C-Glykoside der N-Acetylneuraminsäure. Synthese und Untersuchung ihrer Wirkung auf die Vibrio cholerae Sialidase

C-Glycosides of N-Acetylneuraminic Acid The synthesis of the C-glycosides 8 , 15 , and 9 of N-acetylneuraminic acid is described. Hydroxymethylation of the Li-ester enolate, derived from 5 , yielded the protected C-glycosides 7 and 10 (46%; 3:1), which were deprotected to yield 8 (54%) and 15 (51%; Scheme 2). The mesylate 16 was obtained from 7 (73%) and transformed via the azide 17 (75%) into the acid 18 (66%) and the amino acid 9 (Scheme 3). The configuration at C(2) of 17 was proved by transforming 17 into the bicyclic lactam 19 . Both 8 and 15 are very weak inhibitors of Vibrio cholerae sialidase; 9 appears to stimulate this enzyme.     

First Total Synthesis of Prionoid E,A Bioactive Rearranged Secoabietane Diterpene Quinone from Salvia prionitis

The first total synthesis of prionoid E ( 1 ), a rearranged secoabietane diterpene quinone isolated from Salvia prionitis, was achieved efficiently by means of Wacker oxidation (Scheme 5) and aldol condensation (Scheme 7) as the key steps in the synthetic sequence. Thus 1 was prepared in 15 steps in 3.7% yield starting on one hand from anisole (=methoxybenzene) and methylsuccinic anhydride (=dihydro‐3‐methylfuran‐2,5‐dione) via 4 (Scheme 3 and 5), and on the other hand from 2‐hydroxy‐2‐methylpropanoic acid via 5 (Scheme 6).     

A Facile and Efficient Synthesis of Arylsulfonamido‐Substituted 1,5‐Benzodiazepines and N‐[2‐(3‐Benzoylthioureido)aryl]‐3‐oxobutanamide Derivatives

A facile and efficient synthesis of 1,5‐benzodiazepines with an arylsulfonamido substituent at C(3) is described. 1,5‐Benzodiazepine, derived from the condensation of benzene‐1,2‐diamine and diketene, reacts with an arylsulfonyl isocyanate via an enamine intermediate to produce the title compounds of potential synthetic and pharmacological interest in good yields (Scheme 1). In addition, reaction of benzene‐1,2‐diamine and diketene in the presence of benzoyl isothiocyanate leads to N‐[2‐(3‐benzoylthioureido)aryl]‐3‐oxobutanamide derivatives (Scheme 2). This reaction proceeds via an imine intermediate and ring opening of diazepine. The structures were corroborated spectroscopically (IR, ¹H‐ and ¹³C‐NMR, and EI‐MS) and by elemental analyses. A plausible mechanism for this type of cyclization is proposed (Scheme 3).     

A Total Synthesis of Aliskiren

A total synthesis of aliskiren ( 20 ) was accomplished. A key in our synthesis was to use the symmetric trans‐cisoid‐trans‐bis‐lactone 1 as a precursor. It was expediently prepared by three different routes (Scheme 2). Appending the end groups and functional group transformations completed the synthesis (Scheme 3).     

Glycosylidene Carbenes. Part 4. Synthesis of Spirocyclopropanes from Acetamidoglycosylidene-Derived Diazirines

The synthesis of the first glycosylidene-derived 2-acetamido-2-deoxydiazirine 4 from N-acetylglucosamine 6 is described. Thus, 6 was transformed into the 3-O-mesylglucopyranoside 9 by glycosidation with allyl alcohol, benzylidenation, and mesylation (Scheme 2). Solvolysis of 9 gave the allopyranoside 10 which, upon benzylation and glycoside cleavage, yielded the hemiacetals 12 . Using our established method (via the lactone oxime 14 and the diaziridines 16 ), 12 gave the diazirine 4 . Thermolysis of this diazirine in the presence of i-PrOH gave the dihydro-1,3-oxazole 5 (Scheme 1); in the presence of acrylonitrile, the four diastereoisomeric spirocyclopropanes 17–20 and the acetamidoallal 21 were obtained and separated by prep. HPLC (Scheme 3). Assignment of the configuration of 17–20 is based on NOE measurements and on the effect of diamagnetic anisotropy of the CN group. The ratio of the four cyclopropanes, which is in keeping with earlier results, is rationalized.     

Asymmetric Alkylations of a Sultam-Derived Glycine Equivalent: Practical preparation of enantiomerically pure α-amino acids

Alkylation of the chiral glycine derivative 2 with “activated” organohalides under ultrasound-assisted phasetransfer catalysis or with activated and nonactivated organohalides in anhydrous medium provides (mostly crystalline) alkylation products 3 . Acidic hydrolysis of the pure products 3 gives (aminoacyl)sultams 4 which by mild saponification furnish pure α-amino acids 5 in good overall yields from 2 , along with recovered auxiliary 1 (Scheme 1). Pure ω-protected α,ω-diamino acids and α-amino-ω-(hydroxyamino)acids 12–16 are readily accessible from (ω-haloacyl)sultams 3 via reaction with N-nucleophiles followed by acidic and basic hydrolyses (Scheme 2). A reliable determination of the enantiomeric purity of α-amino acids using HPLC analysis of their N-(3,5-dinitrobenzoyl)prolyl derivatives 17 is presented.     

Towards the Synthesis of Azoacetylenes

The synthesis of azoacetylenes (=dialkynyldiazenes) 1 and 2 has been investigated. They represent a still elusive class of chromophores with potentially very interesting applications as novel bistable photochemical molecular switches or as antitumor agents (Fig. 1). Our synthetic efforts have led us alongside three different approaches (Scheme 1). In a first route, it was envisioned to generate the azo (=diazene) bond by photolysis of N,N′‐dialkynylated 1,3,4‐thiadiazolidine‐2,5‐diones that are themselves challenging targets (Scheme 2). Attempts are described to obtain the latter by alkynylation of the parent heterocycle with substituted alkynyliodonium salts. In a conceptually similar approach, the no‐less‐challenging dialkynylated 9,10‐dihydro‐9,10‐diazanoanthracene ( 29 ) was to be generated by alkynylation of the unsubstituted hydrazine 28 (Scheme 6). In a second route, the generation of the N?N bond from Br‐substituted divinylidenehydrazines (ketene‐azines) 35 was attempted in a synthetic scheme involving an aza‐Wittig reaction between azinobis(phosphorane) 36 and (triisopropylsilyl)ketene 37 (Scheme 7). Finally, a third approach, based on the formation of the central azo bond as the key step, was explored. This route involved the extrapolation of a newly discovered condensation reaction of N,N‐disilylated anilines with nitroso compounds (Scheme 11, and Tables 1 and 2) to the transformation of N,N‐disilylated ynamine 55 and nitroso‐alkyne 54 (Scheme 13).     

Synthesis of the 6-C-Methyl and 6-C-(Hydroxymethyl) Analogues of N-Acetylneuraminic Acid and of N-Acetyl-2,3-didehydro-2-deoxyneuraminic Acid

The synthesis of 6-C-methyl-Neu2en5Ac ( 4 ), 6-C-(hydroxymethyl)-Neu2en5Ac ( 5 ), and 6-C-methyl-Neu5Ac ( 6 ) is described. The 4-methylumbellyferyl glycosides 8 and 9 were also prepared but proved unstable. Protection of the previously reported nitro ether 10 (→ 11 ) followed by a Kornblum reaction gave the branched-chain derivative 13 which was transformed into aldehyde 14 and hence via 16 into the-protected 6-C-hydroxymethylated 20 and into the 6-C-methyl-substituted 18 (Scheme 1). Debenzylidenation of 20 and 18 afforded the diols 21 and 19 , respectively. Selective oxydation of 19 followed by esterification (→ 22 ), acetylation (→ 23 ), and elimination led to the protected 6-C-methyl-Neu2en5Ac derivative 24 (Scheme 2). Bromomethoxylation yielded mainly 25 and some 26 , which were reductively debrominated to 27 and 28 , respectively. Attempted deprotection of 27 did not lead to the corresponding acid, but to the 2,7- and 2,8-anhydro compounds 29 and 30 which were characterised as their peracetylated esters 31 and 32 (Scheme 3). The structure of 32 was established by X-ray analysis. Oxydation of 19 and 21 , followed by deprotection, esterification, and acetylation gave 37 and 38 , respectively (Scheme 4). The branched-chain Neu2en5Ac derivatives 4 and 5 were obtained by β-elimination (→ 39 and 40 ) and deprotection. Omission of the esterification after oxydation of 33 and 34 gave the lactones 35 and 36 which were transformed into 37 and 38 , respectively. Bromoacetoxylation of 39 gave 41-43 which were reductively debrominated to 44 (from 41 and 42 ) and 45 (Scheme 5). Bromoacetoxylation of 40 yielded 46 which was debrominated to 47. Glycosidation of the glycosyl chlorides obtained from 44 and 47 led to the α -D-glycosides 48 and 49 and to the elimination products 39 and 40 , respectively (Scheme 6). Transesterification of 48 , followed by saponification gave the unstable glycoside 8 and hence 6-C-methyl-Neu5Ac ( 6 ). The unstable glycoside 9 was obtained by similar treatment of 49 but yielded 50 under acidic conditions. The branched-chain 4 and 5 were weak inhibitors of Vibrio cholera sialidase, and 8 and 9 were very poor substrates.