Owing to the extreme salinity (B10 times saltier than the oceans), near toxic magnesium levels (B2.0 M Mg 2 þ ), the dominance of divalent cations, acidic pH (6.0) and high-absorbed radiation flux rates, the Dead Sea represents a unique and harsh ecosystem. Measures of microbial presence (microscopy, pigments and lipids) indicate that during rare bloom events after exceptionally rainy seasons, the microbial communities can reach high densities. However, most of the time, when the Dead Sea level is declining and halite is precipitating from the water column, it is difficult to reliably measure the presence of microorganisms and their activities. Although a number of halophilic Archaea have been previously isolated from the Dead Sea, polar lipid analyses of biomass collected during Dead Sea blooms suggested that these isolates were not the major components of the microbial community of these blooms. In this study, in an effort to characterize the perennial microbial community of the Dead Sea and compare it with bloom assemblages, we performed metagenomic analyses of concentrated biomass from hundreds of liters of brine and of microbial material from the last massive Dead Sea bloom. The difference between the two conditions was reflected in community composition and diversity, in which the bloom was different and less diverse from the residual brine population. The distributional patterns of microbial genes suggested Dead Sea community trends in mono-and divalent cation metabolisms as well as in transposable elements. This may indicate possible mechanisms and pathways enabling these microbes to survive in such a harsh environment.

The aim of this study was to explore the use of large-scale sequencing to better describe the genome content of naturally occurring, uncultured protists. We constructed a metagenomic fosmid library from a picoplanktonic assemblage (0.2-3 mum size cells) collected at the Blanes Bay Microbial Observatory (Western Mediterranean). Seven clones contained a small-subunit ribosomal RNA gene (SSU rDNA) affiliating with prasinophytes and uncultured alveolates. One clone (FBB25; 35 kb in size) was completely sequenced and found to be a tandem repeat array (5.5 times) of the rDNA operon, including three rRNA genes (SSU, large-subunit and 5.8S rDNAs) and three spacer regions (internal transcribed spacers 1, 2 and intergenic spacer). The SSU rDNA of FBB25 affiliated with the marine alveolates group I, cluster 1, and was almost identical to sequences retrieved only in marine surveys from a wide geographic and ecological range. Phylogenetic trees using the different rRNA genes showed FBB25 as an independent branch among the main alveolate groups, but their closest affiliation varied between the SSU tree (dinoflagellates) and the large-subunit and 5.8S trees (perkinsids). The spacer regions of FBB25 were particularly short when compared with other eukaryotes, indicating a possible genome streamlining in this picoeukaryote. Finally, not a single polymorphism was found in the rDNA repeat array, suggesting that the high SSU rDNA variability typically found in molecular surveys derives from organismal and not intragenomic diversity. This first report on the rDNA genomic structure of an uncultured marine alveolate improves their phylogenetic position and helps interpreting data generated during picoeukaryotic molecular surveys.

The Dead Sea is a unique hypersaline ecosystem with near toxic magnesium levels (∼2 M), dominance of divalent cations and a slightly acidic pH. Previously, we reported a haloarchaeon related to Halobacterium salinarum to dominate in a microbial bloom that developed in 1992 in the upper water layers of the lake following massive freshwater runoff. Whether this clade also dominated an earlier bloom in 1980-1982 cannot be ascertained as no samples for cultivation-independent analysis were preserved. The presence of the light-driven proton pump bacteriorhodopsin was reported in the 1980-1982 bloom of prokaryotes that had developed in the Dead Sea. To test the hypothesis that bacteriorhodopsin proton pumping may play a major role in determining what type of haloarchaea may dominate in specific bloom conditions, we compared rhodopsin genes recovered from Dead Sea biomass collected in different periods with genes coding for retinal proteins in isolated haloarchaea. Novel bacteriorhodopsin and sensory rhodopsin genes were found in samples collected in 2007 and 2010. The fact that no rhodopsin genes were recovered from samples collected during the 1992 bloom, which was dominated by a single species, suggests that different clades were present in the 1980-1982 and 1992 blooms, and that bacteriorhodopsin proton pumping did not necessarily play a determinative role in the dominance of specific halophiles in the blooms.