A method is described for simultaneously determining retinol, alpha-tocopherol, beta-carotene, alpha-carotene, lycopene, and beta-cryptoxanthin in 0.25 mL of plasma. Plasma mixed with sodium dodecyl sulfate is deproteinized with ethanol containing tocopherol acetate, then extracted with heptane. The evaporated organic layer is reconstituted with mobile phase (methanol/acetonitrile/chloroform, 47/47/6 by vol) and injected onto a 100 x 4.6 mm 3-micron column of Spherisorb ODS-2 (LKB) at 1.5 mL/min. The alpha- and beta-carotenes are well resolved during the 6.5-min run. Retinol is monitored at 325 nm, the tocopherols at 292 nm, and the carotenoids at 450 nm. Extraction of concentrations as great as 135 mumol/L is complete. Intrabatch CVs were 1.7%, 2.3%, 4.1%, 10.4%, 6.4%, and 3.6% for retinol, alpha-tocopherol, beta-carotene, alpha-carotene, lycopene, and beta-cryptoxanthin, respectively. Interbatch CVs for measurements on 30 occasions over 11 weeks were about 10% for all components except alpha-tocopherol (5.3%). Results agree well with those for retinol, alpha-tocopherol, and beta-carotene in quality-control samples.

The vitamin B,, coenzyme has been partially synthesised from hydroxocobalamin (vitamin Blzb) by reduction followed by reaction with a 5'-tosyladenosine. Many analogues, including simple alkyl and acyl derivatives, have been obtained by similar methods. The reactive intermediate is postulated as a cobalt hydride on the basis of its addition to a variety of ethylenic and acetylenic compounds, yielding further analogues of the coenzyme. Reaction of the cobalt hydride with diazomethane gives the cobalt-methyl derivative. Certain reactions of the products are described.THE vitamin B,, coenzymes were first isolated by Barker and his colleagues,l and the chemical structure (I) of one of them, the 5,6-dimethylbenzimidazolylcobamide coenzyme was established by X-ray crystallographic and chemical analysis., The other coenzymes of the group differ only in the nature of the base contained in the nucleotide fragment. The novel cobalt-carbon bond linking the metal to the 5'-deoxyadenosine group of the coenzyme (I ; R = 5'-deoxyadenosyl) is easily broken by a variety of chemical reagents as well as by light, and many transformation products of the coenzyme have been characterised (see, e.g., refs. 3 and 4). Apart from the possible addition of hydrogen at C,-C,, in the corrin nucleus * the main ambiguity of structure (I; R = 5'-deoxyadenosyl) concerns the cobalt atom, which is probably tervalent on the basis of the observed chemical reactions, although there have been claims that the coenzyme is paramagnetic, which would suggest the presence of bivalent cobalt.* Until this ambiguity is resolved, the " cobalamin " nomenclature for the coenzymes and their analogues (I), e.g., " methylcobalamin " for (I; R = Me), has been avoided:

Structures for vitamin B,, (I) and its hexacarboxylic acid degradation product (11) containing six rather than five conjugated double bonds are adopted and reactions bearing on this point are considered. With mild alkali in the presence of oxygen the acetamide side-chain of ring B of the vitamin undergoes cyclisation, to give dehydrovitamin BIB, a 7-lactam, whereas with an equivalent of a halogenating agent the same side-chain is converted into a y-lactone ; mechanisms for these cyclisations are considered. The reaction of vitamin B,, with more than one equivalent of halogenating agent has a marked effect on the chromophore ; under analogous conditions, dehydrovitamin B,, gives a crystalline monohalogeno-derivative.

The hydrolysis of vitamin BIZ under a wide variety of experimental conditions has been studied. Electrophoresis on paper has proved most useful for analysing the complex mixtures of basic, neutral, and acidic products so formed. Vitamin BIZ contains three primary amide groups, which can be hydrolysed stepwise, under comparatively mild conditions, to the corresponding acids. The acids so obtained, three mono-, three di-, and one tri-basic, have all been reconverted into the parent vitamin. Conditions for the hydrolytic removal of the 5 : 6-dimethylbenziminazole nucleotide, both before and after hydrolysis of the amide groups, have been defined and the crystalline free nucleotide has been isolated. 1-Aminopropan-2-01 is believed to be attached by ester linkage to the nucleotide and by amide linkage to the rest of the molecule. The moderately stable cobalt-containing end products from vigorous acid hydrolysis contain five, six, and seven acidic groupings and, from alkaline hydrolysis, five and six acidic groupings.THE combined results of the degradative work so f a r reported on vitamin B,, the antipernicious anaemia factor, have led to the partial formula ( I ) for the vitamin. Its precise molecular formula is not known with certainty, but C,,_,H84-92013-14NlpPCo covers the * A feature of the structure ( I ) is the 5 : 6-dimethyl-l-(a-D-ribofuranosyl)benzhinazoIe-2'(or 3') phosphate residue, which must be fully esterified as the vitamin itself contains no free acidic groups. The benziminazole nucleotide has been isolated from the products of acid hydrolysis of vitamin B,,, both as the barium salt (Buchanan, Johnson, Mills, and Todd, J., 1950, 2845) and as the free acid (Kaczka, Heyl, Jones, and Folkers, J . Amer. Chem. SOC., 1952, 74, 5549; see also Experimental section), and a synthesis has been reported (Kaczka et al., Zoc. cit.), although the point of attachment of the phosphate to the sugar side-chain (i.e,, whether 2' or 3') is still uncertain. The corresponding nucleoside has also been isolated from acid hydrolysates (Brink and Folkers, J . Amer. Chem. SOC., 1952, 74, 2856) and its structure, 5 : 6-dimethyl-l-(~~-~-~bofuranosyl)benz~~nazole, has been confirmed by synthesis (Holly, Shunk, Peel, Cahill, Lavigne, and Folkers, ibid., p. 4521) ; 5 : 6-dimethylbenzirninazole itself has been isolated from the products of vigorous acid hydrolysis of the vitamin (Brink and Folkers, ibid., 1950, 72, 4442; Beaven, Holiday, Johnson, Ellis, Mamalis, Petrow, and Sturgeon, J . Pharm. Pharmacol., 1949, 1, 957).Ammonia and D,-1-aminopropan-2-01 (Wolf, Jones, Valiant, and Folkers, J . Amer. Chem. SOC., 1950, 72, 2820) are also produced on acid hydrolysis, and it has been claimed by Chargaff, Levine, Green, and Kream (Exferientia, 1950, 6, 229) that two moles of the alkanolamine and four-five moles of ammonia are formed from each mole of the vitamin after hydrolysis with G~-hydrochloric acid for 6 hr. at 100" in a sealed tube. On the other hand Cooley, Davies, Ellis, Petrow, and Sturgeon ( J . Pharm. Pharmacol., 1953, 5, ...

No abstract