A search for intermediates in the bacterial biosynthesis of PQQ

Studies on the biosynthesis of pyrroloquinoline quinone (PQQ) were performed with Acinetobacter calcoaceticus PQQ- -mutants belonging to genetically different complementation groups. All mutants were unable to grow on L-arabinose, the conversion of this substrate by the organism only occurring via membrane-bound quinoprotein (PQQ-containing) glucose dehydrogenase. In general, the same observation and conclusion applied to shikimate and quinate, requiring active quinoprotein quinate dehydrogenase (EC 1.1.99.--), although some mutants appeared to be leaky with respect to PQQ biosynthesis under this condition. A number of mutants were unable to grow on anthranilate and accumulated this compound when the growth medium was supplemented with L-kynurenine. Combined with other observations, it strongly suggests that these are deletion mutants, missing a gene for synthesis of anthranilate hydroxylase (EC 1.14.12.1) as well as nearby located genes for the biosynthesis of PQQ. Supplementation of the growth media with amino acids did not result in stimulation of PQQ biosynthesis. Also cross-feeding experiments, using normal and permeabilized cells with extensive variation in combination and conditions, resulted in neither stimulation nor reconstitution of PQQ synthesis. Under conditions optimal for PQQ production in the wild-type strain, as well as under stress conditions using a limiting amount of added cofactor, excretion of intermediates by PQQ- -mutants could not be detected. Similar results were obtained with PQQ- -mutants from Methylobacterium organophilum and Pseudomonas aureofaciens. A tentative explanation, accounting for the absence of detectable intermediates in the biosynthetic route, is given.