Independent Research and Publication
Bob began his independent work in 1956 at the London School of Hygiene and Tropical Medicine and over the next 5 years published a series of papers entitled Studies in the Biosynthesis of Fungal Metabolites. This embraced studies of several classes of fungal metabolites among which were phenalenones such as herquinone as well as alternariol and its congeners. The studies involved feeding 14C-labelled precursors such as acetate, mevalonate and simple aromatic phenolic compounds to microorganisms and isolating the labelled metabolites. Chemical degradation then established the distribution of the radiocarbon labels in these compounds. In addition to the biosynthetic work ultimate proof of structure by synthesis was accomplished for several metabolites. The structures of some of the metabolites that he isolated resisted characterisation and were only solved in the 1970s when modern methods such as NMR spectroscopy and mass spectrometry were more generally used. As a result, publication of the structures of mould metabolites continued right up to his retirement, the final ones being published in collaboration with a German research group in 2016/18.
Over the years Bob contributed immensely to biosynthetic theory and his speculative ideas have been proved by experiment. The indole alkaloids are a large and important class of natural products including strychnine, yohimbine and corynanthine and speculation on their biosynthetic origin has abounded, with such luminaries as Sir Robert Robinson, R B Woodward and others making what appeared to be reasonable hypotheses as to their origin. It was however Bob's suggestion in 1961 that a cyclopentanoid monoterpene was involved in the biosynthesis that led to the final correct biosynthesis being proved experimentally in three simultaneous publications by the research groups of Arigoni, Scott and Battersby.
His longstanding interest in polyketides also led him to be the first to suggest that tetracyclic aromatic polyketides may be formed by two possible types of folding, one of which occurs in streptomyces and the other in fungi. These S- and F-foldings can be identified by feeding studies with [1,2-13C]-acetate. His recognition of the subtle difference in the two phyla has permitted some reassignments of natural product structures and no example has yet been found that contradicts the 'Thomas Rule'.
The biosynthesis of penicillins and cephalosporin has excited the interest of many groups over the years, and in 1981, while at Surrey, Bob and his research group were able to make a significant experimental contribution to this work. It had been known that the basic skeleton of these beta-lactam antibiotics was derived from the amino acids L-cysteine and L-valine with inversion of stereochemistry of the latter to D-valine before incorporation into a precursor tripeptide. There were two possible mechanisms for the incorporation of the valine carboxyl unit of the tripeptide into the final antibiotic and Bob used [18O2]-L-valine to show that one of the two isotopically labelled oxygen atoms in this carboxyl group was lost in conversion to penicillin V, suggesting the intermediacy of a nonribosomal thioester.
In 1974, while at Surrey, he was among the first to employ tritium NMR to study biosynthetic problems and this enabled him to demonstrate the locations of the labelled tritons (protons) in biosynthetic products unambiguously. He identified dehydrogenation in ring A of testosterone as being a stereospecific trans elimination and showed that some polyketides incorporated [3H]-acetate to yield label located in the expected positions. This was a useful advance for, in certain examples, it provided information not available from incorporation studies using deuterium-labelled precursors.
In studies using [1,2-13C]- and [1-13C,2H3]-acetate and assay by 13C NMR he demonstrated, respectively, that the tetracyclic skeleton of oxytetracycline was derived from a linear polyketide and that both of the deuterium (hydrogen) atoms in its aromatic A-ring were labelled as would be expected from their polyketide origin. Further, he showed using [1-13C,18O2]-acetate that the expected regiospecific locations of the 18O-oxygen substituents in oxytetracycline corresponded exclusively to those oxygen-bearing carbons that originated biosynthetically from the carboxyl group of acetate.
He published over 60 research papers, completing his last work in his 91st year in collaboration with a German group. This was published posthumously. He wrote several reviews and co-edited three multi-author books from the three UK natural product meetings that he organised.