Vitamin B12 total synthesis in chemistry describes the total synthesis of the complex biomolecule vitamin B12. A method first reported by the groups of Robert Burns Woodward and Albert Eschenmoser in 1973 [1] is considered a classic in this research field.[2][3][4][5]
The Woodward publications on this topic starting in 1968 in the journal Pure and Applied Chemistry are a transcript of a lecture [6][7] Eschenmoser published an article in 1977 in Science [8] and is also a modification of a lecture. The X-ray crystal structure had already been determined by Dorothy Hodgkin in 1956.[9] The total synthesis was also a scientific breakthrough because in one of its key steps the foundation was laid for what was to become known as the Woodward–Hoffmann rules formulated in 1982.
Contents |
The core of the molecule vitamin B12 (cobalamin) is a corrin structure (depicted in red) with at its center a cobalt ion. Several vitamins exist with different cobalt ligands but the total synthesis concerned the one with a cyano ligand called cyanocobalamin. The corrin rim is lined with methyl groups (8) and lined with amide groups (6) linked through C1 and C2 spacers. A seventh amide group is N-alkylated by a large tail consisting of a isopropanol group, a phosphate group, a ribose group and a dimethylbenzimidazole group. One of the nitrogen atoms on the imidazole is a fifth nitrogen ligand for the cobalt atom. A total of nine carbon atoms on the corrin frame are chiral, adding an additional challenge to the synthesis.
| Vitamin B12 overview |
|---|
In retrosynthesis step 1 was easy. It was already established by Bernhauer in 1960 that the tail can be removed from vitamin B12 by amide hydrolysis to cobyric acid and again replaced.[10] The Woodward/Eschenmoser venture is strictly a formal synthesis because the ultimate target was cobyric acid and tail addition was not included.
| Vitamin B12 retrosynthesis |
|---|
The methyl groups at C5 and C15 were added only after construction of the corrin core. This core was synthesised by joining an AD western part (III) with an BC eastern part (IV). Direct union was not considered feasible due to steric hindrance but both joins were made possible by a sulfur contraction method.
Starting point for the synthesis of ring A was methoxydimethylindol 3 synthesised by condensation of the Schiff base from m-anisidine 1 and acetoin 2. Reaction with the Grignard reagent of propargyl iodide 4 give the propargyl indolenine 5 and ring-closure to 7 was brought about by boron trifluoride and mercuric oxide in methanol through intermediate 6 (electrophilic addition) with the two methyl groups forced into a cis-relationship.
| Vitamin B12 AD ring part A |
|---|
This compound existed as a mixture of two enantiomers (racemic) and chiral resolution using (-)-alpha-phenylethylisocyanate was employed for the isolation of the (+)-enantiomer.
| Vitamin B12 ring A chiral resolution |
|---|
The D ring was synthesized starting from chiral (S)-camphor 8 which was converted to oxime 9 (oxidation / hydroxylamine ), then to amide 10 (hydrolysis), lactam 11 (acid-amide condensation), N-nitroso compound 12, diazo compound 13 and to cyclopentene 14 (carbene, methyl group migration). Reduction (LiAlH4) gave alcohol 15, oxidation (chromic acid) gave aldehyde 16, a Wittig reaction with carbomethoxymethylenetriphenylphosphorane gave trans-alkene 17 and hydrolysis gave carboxylic acid 18.
| Vitamin B12 D ring synthesis |
|---|
Amine 7 and carboxylic acid 18 were condensed through the acid chloride to amide 19. treatment with potassium tert-butoxide in tert-butanol then gave tricycle 20 in a Michael reaction with hydrogen atoms in trans relationship. In anticipation to the partial reduction of the aromatic ring in the later compound protective groups were added: one of the carbonyl groups as the ketal in 21 and the other as an enol ether through the iminium salt 22 (Triethyloxonium tetrafluoroborate) and the orthoamide 23 (sodium methoxide / methanol). Enol ether 24 was obtained by heating in toluene expelling ethanol. Birch reduction provided tetraene 25 and acid treatment gave the dione 26 dubbed pentacyclenone.
| Vitamin B12 AD ring synthesis |
|---|
The second protective group in 26 (acetal, acid hydrolysis) was reconverted to the ketone in 27. The monooxime 28 (at the more hindered ketone group) was synthesised from the dioxime by selective hydrolysis (nitrous acid / acetic acid ). The new nitrogen atom is also the second nitrogen atom required for the AD building block. Both the cyclopentene ring and the cyclohexenone ring were oxidized next in an ozonolysis (ozone) forming triketone 29, an aldol condensation of the 1,5-dicarbonyl unit (pyrrolidine acetate) formed cyclohexene 30 with tosylation of the oxime group, a second oxidation with periodic acid cleaves the cyclohexene ring and diazomethane esterifies the resulting carboxylic acid group in 31. A Beckmann rearrangement (methanol, sodium polystyrene sulfonate 2 hrs, 170°C) took place next to lactam 32 (not isolated) which reacted further to the tetracycle 33 called alpha-corrnorsterone in an amine-carbonyl condensation - aldol condensation cascade. This compound resisted ring-opening of the lactam group due to incorrect stereochemistry of the propionic ester side-chain. The alpha compound was therefore converted its epimer 34 by first equilibrating in excess base followed by reacidifiying and diazomethane treatment. This epimer was then converted to the 35 by the simultaneous action of methanol and thiophenol. This ensured the differentiation of what will become the imidazole tail. The Ozonolysis gave aldehyde 36 with ammonia converting the thioester into an amide group and aldehyde reduction (sodium borohydride) followed by mesylation and bromination (lithium bromide) gave the bromide 37 with conversion of the amide group into a nitrile group as the completed AD section.
| Vitamin B12 AD ring synthesis II |
|---|
The starting material for the synthesis of ring C was chiral (+)-camphorquinone 38 [11] which can be converted to the acetoxy trimethylcyclohexene carboxylic acid ester 39 by addition of trifluoroborane in acetic anhydride, a reaction pioneered by Manasse & Samuel in 1902.[12] Ester hydrolysis to carboxylic acid 40 and amidation to amide 41 was followed by ozonolysis to peroxide 42 that was reduced to succinimide 43 by zinc and methanol. treatment with methanolic hydrochloric acid gave lactam 44 and pyrolysis gave the complete C ring 45.
| Vitamin B12 C ring synthesis |
|---|
The starting material for ring B was 3-methyl-4-oxo-2-pentenoic acid 46 which was reacted with butadiene in a Diels-Alder reaction (stannic chloride) to racemic cyclohexene 47. This reaction is stereospecific with the methyl group and the carboxylic acid group ending up in a cis relationship (with hindsight a disrotatory electrocyclic reaction). Chiral resolution using alpha-phenylethylamine gave optically active 47. Oxidation of the double bond with chromic acid gave the triacid 48 as an intermediate which gave dilactone 49 in two intramolecular esterifications. An Arndt–Eistert reaction elongated the carboxylic acid chain in 50, reaction with ammonia gave lactam 51 and reaction with phosphorus pentasulfide give thiolactam 52.
| Vitamin B12 D ring synthesis |
|---|
Ring B (52) and ring C (45) were joined with benzoyl peroxide/HCl to sulfur bridged 53, the sulfur atom was extruded with triethylphosphite to enamine-imine 54 in a first of two sulfur contractions and the lactam group converted to the thiolactam 55 (trimethyloxonium fluoroborate / hydrogen sulfide)
| Vitamin B12 BC ring system synthesis |
|---|
The eastern part of the molecule BC (cyanobromide 37) and the western part AD (thiodextolin 55, with the propionic acid ester group racemized) were then joined using potassium t-butoxide to thioether 56 (through a sulfide ion intermediate). A second sulfur contraction (cyanoethyl phosphine/ trifluoroacetic acid/sulfolane) yielded cyanocorrigenolide 57 with the propionic acid ester group of ring C also racemised. Due to the steric bulk of both reactants this contraction was the only successful method. The oxygen atoms in the lactam and the lactone group were replaced by sulfur (phosphorus pentasulfide/4-methylpyridine) in dithiocyanocorrigenolide 58 and the S-methyl derivate 59 was formed by reaction with trimethyloxonium fluoroborate. Dimethylamine addition opened the thiolactone ring with an exocyclic alkene group in 60 by elimination of the sulfide anion from the methyl group. In an early example of template-directed synthesis this compound was isolated as the cobalt adduct.
| Vitamin B12 BCAD construction |
|---|
The final cyclisation reaction of 60 to 61 was facilitated by the central cobalt ion (placing the ends in close proximity) and consisted of another type of sulfur contraction employing basic conditions (diazabicyclononane / dimethylacetamide). This reaction takes place with racemisation of the propionic acid ester group of ring C. Oxidation (iodine/acetic acid) formed lactone 62 and restored the correct stereochemistry at the ring B propionic acid ester tail.
The final efforts were directed at placing methyl groups at position 5 and 15. With position 10 sufficiently shielded reaction with chloromethyl benzyl ether produced the di(chloromethyl) adduct which was further converted to the dithiophenyl compound 63 using thiophenol, isolation of which required plate chromatography. Desulfurisation took place with Raney nickel and the reduction reaction also opened the lactone ring to the carboxylic acid which was converted to the ester 64 by reaction with diazomethane. At this stage the number of isomers in the mixture was reduced by HPLC to just two with the racemic propionic acid ester group at C13 (ring C) remaining in 65. Reaction with sulfuric acid converted the cyano group to the amide group in 66, again destroying stereochemistry at C13. The correct isomer 67 (minor yield) was isolated again by HPLC.
| Vitamin B12 BCAD construction part II |
|---|
The amide group was converted to the carboxylic acid group in 68 by action of the cyclohexylnitrone derived from chloroacetaldehyde in combination with silver tetrafluoroborate and in the final step 6 ester groups were converted to the amide groups in cobyric acid 69 by reaction with ammonia and ammonium chloride.
| Vitamin B12 BCAD construction part III |
|---|