Search

A range of analogues of parabactin A 4 and agrobactin A 5 have been synthesised. All these comounds were able to chelate iron, and to promote the growth of iron-starved Paracoccus denitrificans to varying degrees. A degree of flexibility in the siderophore receptor system is thus demonstrated, which may be exploitable for the transport of new siderophore - antibiotic conjugates.

A series of N-propenyl substituted N-benzylcarbamoylcobalt(III) salophens, i.e.11, 20 and 30, have been prepared, and have been shown to be useful precursors in a new approach to β-lactams (viz.12-16, 21 and 31), via 4-exo-trig modes of cyclisation of the corresponding intermediate carbamoyl radicals. A new formal synthesis of (±)-thienamycin 1 from the trans-disubstituted azetidin-2-one 31 produced in one step by heating the carbamoylcobalt salophen 30 in toluene, is also described.

Novel thiazole- and oxazole-containing modified cyclopeptide C3-symmetric cages have been synthesised utilising a trimeric linker to control the assembly and cyclooligomerisation of heterocyclic amino acids.

Macrocyclic and polycyclic ‘cembranoid’ diterpenes are one of the most widespread groups of natural products that are found in the marine milieu. The macrocyclic cembranoids are linked to each other by a network of oxygenation processes, which often climax with the formation of furano- and furanobutenolide-based macrocyclic cembranoids commonly found in gorgonian and soft corals. These macrocycles are then prone to oxidative rearrangements, photochemical ring contraction, and transannular cyclisations amongst others, leading to a plethora of novel and architecturally attractive marine metabolites. Although there is a dearth of knowledge about the enzymes that trigger some of the steps in their biosynthesis, speculations are rife. In this personal perspective we have examined many of the structural relationships within oxycembranoids and their relatives isolated from corals. This has allowed us to speculate on the likely biosynthetic interrelationships between structurally similar metabolites, and then propose some likely key carbon-to-carbon bond forming reactions that are followed in vivo in linking macrocyclic cembranoids to their polycyclic congeners. Biomimetic synthesis studies, which vindicate some of the biosynthetic speculations, are interweaved in the discussion.

Covering: up to July 2010 Polycyclic norcembranoid diterpenes are a rare group of natural products which are found exclusively in soft corals of the genus Sinularia. They co-occur with 14-membered macrocyclic 3(2H)-furanone-based norcembranoids, all of which lack a C18 carbon substituent in comparison with C20-cembranoids. The polycyclic norcembranoids have structural features in common with the better-known C20-cembranoid diterpenes plumarellide, mandapamate and rameswaralide. No synthetic studies or suggestions as to the origins of these novel polycyclic norcembranoids in Nature have been forthcoming. This review summarises the structural relationships that are found between polycyclic norcembranoids and their macrocyclic congeners. It then poses the question: how and when does Nature expel a C18 carbon substituent from a C20-cembranoid precursor during the elaboration of macrocyclic norcembranoids? A number of suggestions and possibilities are then offered, involving both anionic and radical intermediates. Finally, the review speculates on the range of competitive transannular carbon-to-carbon bond-forming reactions that may be used as key steps in the elaboration of polycyclic norcembranoids from their macrocyclic precursors.

Covering: up to July 2010 Polycyclic norcembranoid diterpenes are a rare group of natural products which are found exclusively in soft corals of the genus Sinularia. They co-occur with 14-membered macrocyclic 3(2H)-furanone-based norcembranoids, all of which lack a C18 carbon substituent in comparison with C20-cembranoids. The polycyclic norcembranoids have structural features in common with the better-known C20-cembranoid diterpenes plumarellide, mandapamate and rameswaralide. No synthetic studies or suggestions as to the origins of these novel polycyclic norcembranoids in Nature have been forthcoming. This review summarises the structural relationships that are found between polycyclic norcembranoids and their macrocyclic congeners. It then poses the question: how and when does Nature expel a C18 carbon substituent from a C20-cembranoid precursor during the elaboration of macrocyclic norcembranoids? A number of suggestions and possibilities are then offered, involving both anionic and radical intermediates. Finally, the review speculates on the range of competitive transannular carbon-to-carbon bond-forming reactions that may be used as key steps in the elaboration of polycyclic norcembranoids from their macrocyclic precursors.

Treatment of the acetylene substituted cyclobutanone oxime ether ( 1 ) with (Me 3 Si) 3 SiH leads, in one pot, to the bicycle ( 2 , 70%) by way of a novel double ring expansion-cyclisation process involving aminyl radical intermediates.

A total synthesis of the novel antifungal substance myxothiazol 1 isolated from the myxobacterium Myxococcus fulvus is described. The synthesis is based on elaboration of the S-E,E -diene thioamide 1 5 and the 2 R S , 3 S R amide aldehyde 2 5 as key intermediates, followed by conversion of 1 5 into the bis -thiazole 1 9 and a final Wittig coupling reaction between 2 5 and the salt 2 0 c leading to 7 S ,18 S R , 19 R S myxothiazol.

A total synthesis of the new thiazoline-based siderophore desferrithiocin 1, isolated from Streptomyces antibioticus, is described. Thus, a concise synthesis of (S)-2-methylcysteine hydrochloride 11 is first developed based on a modification of Seebach's “self-reproduction of chirality” protocol using the thiazolidine intermediate 8 derived from (S)-cysteine and pivaldehyde as a key intermediate. When a solution of (S)-2-methylcysteine hydrochloride is heated with 2-cyano-3-hydroxypyridine in the presence of triethylamine, (S)-desferrithiocin is produced in 97% yield. In a similar manner, use of (R)-2-methylcysteine in a cyclocondensation with 2-cyano-3-hydroxypyridine led to (R)-desferrithiocin, in a similar yield.

A concise synthetic route to the tricarbonyl subunit ( 3a ) in the immunosuppressant rapamycin ( 2 ), based on ruthenium-catalysed oxidation of an acetylenic amide precursor ( 17 ) is described. viz ( 17 ) → ( 18 ) → ( 20 ).

Treatment of the N-formyl derivative of the thiazolidine adduct derived from (R-cysteine methyl ester hydrochloride and pivalaldehyde, with LDA-DMPU at −90°C, followed by reaction with iodomethane produces the corresponding methylated thiazolidine containing the methyl and t-butyl groups virtually exclusively anti- to one another. Hydrolysis in the presence of 5M HCl then affords (R-2-methylcysteine hydrochloride in excellent yield and enantiomeric purity. A range of other 2-alkyl substituted cysteines of excellent optical purity are prepared by this modification of Seebach's “self-reproduction of chirality” protocol.

The marine environment is a seemingly inexhaustible treasury of organisms whose secondary metabolites bear witness to the lavishness and inventiveness with which nature is able to manipulate molecular architecture. But to what purpose are these diverse and often grotesque compounds produced? This review is founded on the premise that some of them may be involved in the uptake and transport of metal ions present in the aquatic milieu. Many metabolites produced by terrestrial organisms are known to act as ionophores, but the case for similar behavior by their marine counterparts is far hazier. Notwithstanding the relative abundance of certain metal ions in the oceans, and of metabolite structures possessing features that should facilitate the chelation of metal ions, few attempts to establish a connection between these two phenomena have been reported. We have whittled down the voluminous literature of natural products derived from marine sources to expose a core of observations and speculations germane to our premise. These facts and fantasies are evaluated in this review. A mere handful of metal-containing complexes has actually been isolated; furthermore, attempts to prepare such complexes in vitro are rare, and spectroscopic evidence for metal–metabolite interactions, whether in vivo or in vitro, is not common. Only with the vanadium-sequestering tunichromes does a logical (but by no means complete) picture begin to emerge. In several other cases, the plausibility of metal chelation, though mooted by authors, remains unsupported by experimental evidence. However, continuing efforts to obtain structural, and particularly conformational, information on the metabolites by means of X-ray crystallography, nuclear magnetic resonance spectroscopy, and molecular mechanics calculations would seem to provide the key to a rational approach to this neglected topic. On the basis of recent studies dealing with such structural aspects, we present a selection of candidate compounds, some of which are the targets of our own synthetic attentions, whose potential for binding to metal cations merits further research.

Treament of the bicyclic dienol (8) with iodosylbenzene diacetate-iodine is shown to lead to the 7,5,-tricycle (11, 81%) by way of a cascade ra

A total synthesis of nannochelin A, a novel cinnamoyl-hydroxamate siderophore produced by Nannocystis exedens , is described.

A concise total synthesis of the thiazoline-thiazole containing metabolite didehydromirabazole A 1 is described. The synthesis uses the unusual amino acid (R)-2-methylcysteine 20 in sequential cyclocondensations with imino ethers as key steps, viz22→32 and 37→38.

The acyclic carotenoids phytoene and phytofluene, which are precursors of 9-cis-β-carotene produced by the alga Dunaliella bardawil, have been separated and characterised. Spectroscopic data, and comparison with synthetic model compounds and corresponding carotenoids found in the fruits of the Tangerine tomato Lycopersicon esculentum var., establish that the pigments have the stereochemistries 15-cis-6 and 9-cis-11 respectively. The data suggest that phytoene 6 is the clear branch point for the formation of 9-cis-β-carotene 2 in D. bardawil. In addition, spectroscopic data for β-carotene isolated from D. bardawil showed that it was isomerically pure 9-cis-β-carotene 2.

Arylmethyl- and allyl-carbonylcobalt salophen complexes, e.g.(7), (8), (9), and (22), readily undergo carbon-to-cobalt-bond homolysis and in situ decarbonylation, producing new alkyl radical centres which can be intercepted with oxygen-, nitrogen-, halogen-, sulphur-, and selenium-containing radical-trapping agents leading to functionalised noralkanes. The sequence constitutes a useful, and in some cases more flexible, variant of the classical Hunsdiecker reaction, and amounts to a cobalt equivalent of the Barton radical decarboxylation reaction of carboxylic acids via their corresponding thiohydroxamic esters. In a similar manner, irradiation of the oxy-substituted acylcobalt salen reagents (25) and (26) in the presence of tetramethylpiperidine oxide produces the products (14) and (27), respectively, resulting from homolysis–decarboxylation and alkyl-radical trapping. In the absence of radical-trapping agents, irradiation of (7c) produces (18), and irradiation of (29) leads to the but-2-enolide (32).

Hydrocobaltation of myrcene (3) by pyridinatocobaloxime leads to a 2:1 mixture of the E- and Z-allylcobaloxime (4a) and (4b) in good yield. When a solution of the cobaloximes (4a) and (4b) in toluene is heated in the presence of tetramethylpiperidine oxide (TEMPO) the hydroxylamines (7a) and (7b) result, which can be converted into geraniol (8a) and nerol (8b) by reduction using zinc in acetic acid. By contrast, in the presence of molecular oxygen, the allylcobaloxime (4) is converted into the allylperoxycobaloxime (15) which on reduction produces (±)-linalool (16). In addition, when the cobaloxime (15) is heated with TEMPO it undergoes cyclisation to the tetrahydrofuran (17), precursor to (+)-linalool oxide (18). Whereas the 1,3-dienes (20)–(24) all failed to undergo hydrocobaltation reactions, both cobaloximes (27) and (28) were easily obtained from 2-methyl-(25) and 2,3-dimethyl-buta-1,3-diene (26), respectively. Using similar chemistry to that described for compound (4), the allylcobaloxime (27) was smoothly converted into the hydroxylamine (29) and into the epoxy-TEMPO derivative (32).

A concise synthesis of optically active (−)-zearalenone (1) which uses a novel 14- endo trig macrocyclisation from an allylic radical intermediate (Scheme 1) as a key feature, is described.

Alkyl radicals generated by photolytic homolysis of cobalt salophen reagents, e.g.(6) and (15), are shown to add to activated carbon-to-carbon double bonds, i.e. ethyl acrylate, methyl vinyl ketone, acrylonitrile, and styrene, producing preparatively useful yields of new alkene products, viz.(9), (10), (12) and (14)[from (6)], and (16)[from (15)]. The reactions proceed via radical (Michael) additions, followed by ‘dehydrocobaltation’ from the presumed organocobalt intermediates (2)(Scheme 4). By use of this chemistry, in combination with the Schrauzer ‘hydrocobaltation’ reaction of alkenes, a new method for the cross-coupling reactions between two alkenes, leading to new functionalised alkenes, e.g.(24), (25), (26), (27), (28), and (29), is developed [Equation (2)].

Irradiations of the alkylcobalt salophen complexes (14), (21), (22), and (23) in the presence of radical-trapping agents, e.g. molecular oxygen, tetramethylpiperidine oxide, nitrogen monoxide, diphenyl disulphide, diphenyl diselenide, methanesulphonyl chloride, bromotrichloromethane, or iodine, leads to oxgen-[e.g., (18), (19), and (25)], nitrogen-[e.g., (27) and (28)], sulphur/selenium-[e.g., (30) and (31)] or halogen-(34) functionalised products. When these radical-trapping methodologies are combined with cobalt-mediated radical cyclisation reactions (Scheme 2) a powerful synthetic procedure, i.e. radical carbon-to-carbon bond formation with simultaneous functionalisation of the product radical centre, becomes available.

Acylcobalt salophen reagents (6) are conveniently synthesised from carboxylic acid chlorides following treatment with the sodium derivative (5) produced from reduction of cobalt(II) salophen with sodium amalgam in tetrahydrofuran at room temperature. The acylcobalt salophens (6) undergo homolytic cleavage in the presence of light from a conventional 300 W sunlamp to give acyl radicals (24) which then undergo additions to activated carbon-to-carbon double bonds, leading to enones, viz.(26)[following dehydrocobaltation from presumed organocobalt intermediates (25)], or saturated ketones, viz.(34)[following H-quenching of intermediates (25)]. Intramolecular cyclisations of the acylcobalt salophens (42a) and (42b) lead to the ylidenecyclopentanones (43) and (45), respectively.

The total synthesis of (±)-citreoviral, which co-occurs with citreoviridin, citreoviridinol, and citreodiol in Penicillium citreoviride, is described. The synthesis is based on a biogenetic model, and has as a key feature the elaboration of the dihydroxytetrahydrofuranyl ring portion 10 by an acidcatalysed cyclisation of the central epoxy alcohol intermediate 9 which is derived in ten steps from methyl tiglate.

A total synthesis of the all-cis cyclopentanetetraol moiety 2,3,4,5-tetrahydroxycyclopentane-1-acetonitrile, present in the antiviral substance funiculosin produced by Penicillium funiculosum, is described. Conversion of the known cyclopent-4-ene-1,2,3-triol into the 1-[1-(t-butyldimethylsiloxy)vinyl] 2,3-methylene trisether, followed by Claisen–lreland rearrangement in xylene at 190 °C, first provided the key intermediate all-cis-4,5-(methylenedioxy)cyclopent-2-ene-1-acetic acid. Iodolactonisation of this acid then led to the iodo lactone 4β-iodohexahydrofuro[2′,3′:4,3]cyclopenta [d][1,3]dioxolan-6-one, which on direct nucleophilic displacement using potassium superoxide in the presence of 18-crown-6 produced the corresponding 4β-hydroxy lactone by an SN1 mechanism probably via the C-4 radical intermediate. The structure and stereochemistry of the hydroxy lactone were determined by X-ray analysis of its t-butyldimethylsilyl ether.The above iodo lactone was converted into the iodohydrin-acetonitrile 2α-hydroxy-3β-iodo-4α,5α-(methylenedioxy)cyclopentane-1α-acetonitrile in three steps, which was then treated with methyl (chloroformyl)acetate to give the corresponding mixed malonate diester. Treatment of this diester with sodium hydride resulted in smooth, intramolecular oxygen nucleophile displacement leading to the all-cis ketene acetal methyl [7-(cyanomethyl)-3b,6a,7,7a-tetrahydro-3aH-cyclopenta [1,2-d;3,4-d′]-di [1,3]dioxol-2-ylidene]acetate in 92% yield, hydrolysis of which produced a mixture of positional isomers of all-cis-2-(cyanomethyl)-5-hydroxy-3,4-(methylenedioxy)cyclopentyl methyl malonate, which was then saponified to give the all-cis cyclopentane-substituted 2,3-dihydroxy-4,5-(methylenedioxy) cyclopentane-1-acetonitrile. The full structure and stereochemistry of this acetonitrile followed from X-ray analysis. The synthesis of (±)-all-cis-2,3,4,5-tetrahydroxycyclopentane-1-acetonitrile was then completed following treatment of the 4,5-methylenedioxy compound with ozone, leading to the 4-formate (and its 5-isomer), and saponification.

A number of complementary biomimetic approaches to the tetrahydrofuranyl portion of the dienal 13, which was used in a synthesis of preaurovertin, are described. Epoxidation of the trienoate 31 produces largely the β-epoxide 32, which on treatment with aqueous trifluoroacetic acid undergoes deacetalisation and concomitant cyclisation leading to compound 34. After conversion of 34 into dienal 13, a Wittig reaction with the ylide derived from the phosphonium salt 37 produces preaurovertin.Preaurovertin is a putative biogenetic precursor of aurovertin found in Calcarisporium arbuscula. Citreoviridin produced by Penicillium citreoviride relates to preaurovertin, and the citreoviridinols are related structurally to aurovertin. In addition, aurovertin is related biogenetically to asteltoxin found in Aspergillus stellatus. These structural and biosynthetic interrelationships are considered alongside the synthetic work.

Reactions between substituted maleic anhydrides, viz. 1, and sodium dimethyl phosphite in refluxing benzene are shown to lead to the corresponding furan-2(5H)-one phosphonates 2 and 8. Wadsworth–Emmons olefination reactions involving 2 and heteroarylaldehydes and arylbenzoyl formates 14 then provides a new synthesis of heterocyclic analogues of pulvinones, i.e.10–13, and of permethylated pulvinic acids, e.g.16–17 which are found in higher fungi.

The 2,6-dioxabicyclo[3.2.1]octane ring systems, viz. systems 24 and 34, present in citreoviridinol and the aurovertins, have been produced in a stereoselective manner by treatment of the corresponding 4-hydroxytetrahydrofuran epoxides 22 and 33, respectively, with toluene-p-sulphonic acid. A number of related epoxy alcohol cyclisations leading to isomeric dioxabicyclooctanes are reported. These studies have led to a greater understanding of the biosynthetic pathway to the citreoviridinols and the aurovertins, and a greater appreciation of the role of epoxide intermediates in the biosynthesis apparatus.

A total synthesis of all-E-citreomontanin, produced by Pencillium pedemontanum and believed to be the putative precursor of citreoviridin and citreoviridinol, involving successive epoxidations of its terminal triene portion, is described. Elaboration of the pyrone 9a to the phosphonium salt 13b was smoothly accomplished via the key pyrone aldehyde intermediates 10 and 12. The all-E-trienal 7 required for Wittig coupling with compound 13b was readily produced from the aldehyde 14 by use of two successive stereoselective Wittig olefination reactions involving the stabilised phosphorane 15; i.e., 14+15→16, and 17+15→18. A Wittig coupling reaction between the ylide 8 derived from 13b and the trienal 7 then produced the all-E-polyene 1, as an orange solid, which was identical with citreomontanin isolated from P. pedemontanum.

A concise total synthesis of (±)-citreoviral which illustrates a new stereoselective approach to hydroxy-substituted tetrahydrofurans is described. Key features in the synthesis are: (i) elaboration of the Z-unsaturated epoxy ester 10, (ii) acid-catalysed cyclisation of ester 10 to the 4-hydroxyethyl-substituted but-2-enolide 14, (iii) stereoselective vicinal hydroxylation of the butenolide 14 to the diol15, (iv) cyclic ether formation from diol 15 to the bicyclic lactone 8via the benzylidene acetal 9, (v) transesterification of bicycle 8 to the tetrahydrofuran 21 containing three of the four centres in citreoviral in the correct relative configuration. Inversion of the secondary 3-OH group in compound 21, and elaboration of the C-2 side-chain, then completed the synthesis of (±)-citreoviral.

A synthesis of the grevillin group of pyrandione pigments e.g.3, 23 and 24 present in fungi is described. The synthesis, which is based on a biogenetic model, uses bis-benzylacyloins 9 and their corresponding oxalate derivatives as key intermediates (Scheme 3). Treatment of the grevillins 25–c with sodium ethoxide in ethanol effects their quantitative isomerisation into the corresponding terphenylquinone pigments 4a–c. Perkin-type condensations between the terphenylquinones 4 and arylacetic acids in the presence of sodium acetate–acetic anhydride then produces the xylerythrin pigments 29a–e, whereas rearrangements of 4 in the presence of dimethyl sulphoxide leads to pulvinic acid derivatives, e.g.31, 32 and 5. These synthetic studies interrelate the biosynthetic origins of the pigment types 3, 4, 5 and 8 together with the related pulvinone 6 and furanone 7 fungal pigments.

This presentation summarises the progress maded in the constructing models for the biological activities of the ulapualides,lissoclinamides and leinamycin

A total synthesis of the new thiazoline-based siderophore desferrithiocin 1, isolated from Streptomyces antibioticus, is described. Thus, a concise synthesis of (S)-2-methylcysteine hydrochloride 11 is first developed based on a modification of Seebach's “self-reproduction of chirality” protocol using the thiazolidine intermediate 8 derived from (S)-cysteine and pivaldehyde as a key intermediate. When a solution of (S)-2-methylcysteine hydrochloride is heated with 2-cyano-3-hydroxypyridine in the presence of triethylamine, (S)-desferrithiocin is produced in 97% yield. In a similar manner, use of (R)-2-methylcysteine in a cyclocondensation with 2-cyano-3-hydroxypyridine led to (R)-desferrithiocin, in a similar yield.

A concise synthetic route to the tricarbonyl subunit ( 3a ) in the immunosuppressant rapamycin ( 2 ), based on ruthenium-catalysed oxidation of an acetylenic amide precursor ( 17 ) is described. viz ( 17 ) → ( 18 ) → ( 20 ).

Treatment of the N-formyl derivative of the thiazolidine adduct derived from (R-cysteine methyl ester hydrochloride and pivalaldehyde, with LDA-DMPU at −90°C, followed by reaction with iodomethane produces the corresponding methylated thiazolidine containing the methyl and t-butyl groups virtually exclusively anti- to one another. Hydrolysis in the presence of 5M HCl then affords (R-2-methylcysteine hydrochloride in excellent yield and enantiomeric purity. A range of other 2-alkyl substituted cysteines of excellent optical purity are prepared by this modification of Seebach's “self-reproduction of chirality” protocol.