Chemical mediation of bacterial surface colonisation by secondary metabolites from the red alga Delisea pulchra

We investigated the effects of halogenated furanones from the red alga Delisea pulchra on colonisation of surfaces by marine bacteria. Bacterial abundance on the surface of D. pulchra, assessed using scanning electron microscopy (SEM), was significantly lower than on the surfaces of 3 co-occurring algal species, all of which lack furanones. There was also a strong inverse correlation between bacterial abundance and furanone content (previously determined) for different sections of the thallus of D. pulchra, consistent with inhibition of bacteria by furanones. Based on these observations we experimentally investigated inhibition of marine bacteria by furanones, initially testing the effects of crude extract of D. pulchra (about 50 % of which is furanones) on the growth of 144 strains of bacteria isolated from the surfaces of D. pulchra, nearby rocks, or a co-occurring alga (Sasgassum vestitum). This crude extract did not strongly inhibit growth of these bacteria; 79% of the strains grew at 50 pg ml(-1) of crude extract, and 63 % grew at 500 mu g ml(-1). Inhibition of growth that did occur was strongly source dependent, with bacteria isolated from rocks the least affected, and strains from D. pulchra the most. As inhibition of growth did not provide an adequate explanation for the inverse relationship between levels of furanones and bacteria abundance on D. pulchra, we proceeded to investigate the effects of these metabolites on other bacterial characteristics relevant to colonisation - attachment, swarming, and swimming. Individual furanones or crude extract at natural concentrations strongly inhibited bacterial attachment in the laboratory and in the field. In laboratory assays, attachment of 3 strains isolated from rocks was much more strongly affected than that of 3 isolates from D. pulchra, in contrast to the pattern for growth inhibition. We also tested individual furanones against swimming and swarming of the same 6 bacterial isolates (3 from rocks, 3 from D. pulchra) used in the attachment assays. At least some furanones inhibited swarming or swimming at non-growth-inhibitory concentrations for all isolates, again indicating specific effects against bacterial characteristics. As for attachment, there were significant differences in the responses of different isolates to furanones. We also found that the ability to swarm was widespread among these surface associated marine bacteria, suggesting that swarming may be ecologically important in these systems. Overall, we found that the effects of furanones on bacteria varied among (1) furanones, (2) bacterial phenotypes, (3) different isolates and (4) different sources of isolation (e.g. rocks or algae). This differential inhibition of different bacterial isolates or phenotypes by furanones, as well as affecting overall bacterial abundance on the alga, should have strong effects on the species composition of the bacterial community on the alga's surface. The effects of furanones on specific bacterial colonisation traits are discussed in the light of recent evidence demonstrating that furanones interfere with bacterial acylated homoserine lactone regulatory systems.