Abstract
The fraction of dissolved dimethylsulfoniopropionate (DMSPd) converted by marine bacterioplankton into the climate-active gas dimethylsulfide (DMS) varies widely in the ocean, with the factors that determine this value still largely unknown. One current hypothesis is that the ratio of DMS formation: DMSP demethylation is determined by DMSP availability, with 'availability' in both an absolute sense (i.e. concentration in seawater) and in a relative sense (i.e. proportionally to other labile organic S compounds) proposed as the critical factor. We investigated these models during an experimentally induced phytoplankton bloom using a taxon-specific microarray targeting DMSP-related gene transcription in members of the Roseobacter clade, a group hypothesized to play an important role in the surface ocean sulfur cycle and well represented by genome sequences. The array consisted of 1578 probes to 431 genes and was designed to target diverse Roseobacter communities in natural seawater by using hierarchical probe design based on 13 genome sequences. The prevailing pattern of Roseobacter gene transcription showed relative depletion of DMSP-related transcripts during the peak of the bloom, despite increasing absolute concentrations and flux of DMSP-related compounds. DMSPd thus appeared to be assimilated by Roseobacter populations in proportion to its relative abundance in the organic matter pool (the 'relative sense' hypothesis), rather than assimilated in preference to other labile organic sulfur or carbon compounds produced during the bloom. The relative investment of the Roseobacter community in DMSP demethylation was not useful for predicting the formation of DMS, however, suggesting a complex regulatory process that may involve multiple taxa and alternative fates of DMSPd.
