Metagenome of the Mediterranean deep chlorophyll maximum studied by direct and fosmid library 454 pyrosequencing

Abstract

The deep chlorophyll maximum (DCM) is a zone of maximal photosynthetic activity, generally located toward the base of the photic zone in lakes and oceans. In the tropical waters, this is a permanent feature, but in the Mediterranean and other temperate waters, the DCM is a seasonal phenomenon. The metagenome from a single sample of a mature Mediterranean DCM community has been 454 pyrosequenced both directly and after cloning in fosmids. This study is the first to be carried out at this sequencing depth (ca. 600 Mb combining direct and fosmid sequencing) at any DCM. Our results indicate a microbial community massively dominated by the high-light-adapted Prochlorococcus marinus subsp. pastoris, Synechococcus sp., and the heterotroph Candidatus Pelagibacter. The sequences retrieved were remarkably similar to the existing genome of P. marinus subsp. pastoris with a nucleotide identity over 98%. Besides, we found a large number of cyanophages that could prey on this microbe, although sequence conservation was much lower. The high abundance of phage sequences in the cellular size fraction indicated a remarkably high proportion of cells suffering phage lytic attack. In addition, several fosmids clearly belonging to Group II Euryarchaeota were retrieved and recruited many fragments from the total direct DNA sequencing suggesting that this group might be quite abundant in this habitat. The comparison between the direct and fosmids sequencing revealed a bias in the fosmid libraries against low-GC DNA and specifically against the two most dominant members of the community, Candidatus Pelagibacter and P. marinus subsp. pastoris, thus unexpectedly providing a feasible method to obtain large genomic fragments from other less prevalent members of this community.

Comparison of large-insert, small-insert and pyrosequencing libraries for metagenomic analysis

“…These observations support the previous suggestion of Temperton et al (2009) and Ghai et al (2010), that low-GC taxa in general are less well represented in fosmid libraries, compared with small-insert shotgun clone libraries or 454 sequencing libraries, and that low-GC content itself may be the responsible factor for poor representation in the fosmid libraries. While the GC content does appear to be responsible for much of the observed bias, other factors such as toxic gene products, or chemically modified nucleotide content in the DNA, may also have a role.…”

“…Feingersch and Béjà (2009) on the other hand speculated that the expression of certain genes, in particular ribosomal proteins, may prove toxic to the Escherichia coli host and that BAC libraries are more affected than shotgun libraries, because their larger inserts are more likely to contain at least one such toxic sequence. Subsequently, Ghai et al (2010) reported an apparent GC bias and under representation of Pelagibacter and Prochlorococcus clones in picoplankton libraries from a Mediterranean deepchlorophyll maximum layer (Ghai et al, 2010). To help clarify some of these issues, we report here a more detailed analysis of taxon representation and potential biases associated with multiple sets of deeply sequenced fosmid, shotgun and 454 sequencing libraries, each set prepared from the same sample of open-ocean picoplankton DNA.…”

Complete-fosmid and fosmid-end sequences reveal frequent horizontal gene transfers in marine uncultured planktonic archaea

“…In contrast with Thaumarchaeota, these euryarchaeotal groups lack any cultured representative and their metabolic capabilities remain unknown, except for the indication of a possible phototrophic metabolism in group II archaea from surface waters . Group II Euryarchaeota usually appear to be relatively more abundant in surface waters compared with Thaumarchaeota (Karner et al, 2001;Ghai et al, 2010), but they may be abundant in deep-sea waters as well, reaching in some oceanic regions even higher proportions than Thaumarchaeota (Martín-Cuadrado et al, 2008). The much more enigmatic group III Euryarchaeota have been almost exclusively detected in deep-sea plankton, and are found in general in lower abundance than Thaumarchaeota and group II Euryarchaeota (Fuhrman and Davis, 1997;Martín-Cuadrado et al, 2008).…”

Integrated metatranscriptomic and metagenomic analyses of stratified microbial assemblages in the open ocean

“…Our knowledge of abundance, diversity and gene content of planktonic microbes has been fundamentally advanced over the past three decades, by both model organism-based studies (Giovannoni et al, 2005b;Coleman and Chisholm, 2007), as well as metagenomic surveys of natural microbial communities Rusch et al, 2007;Dinsdale et al, 2008;Ghai et al, 2010). In particular, metagenomic comparisons of distinct microbiomes Martin-Cuadrado et al, 2007;Dinsdale et al, 2008) have revealed habitat-dependent distribution of taxa and gene families, in part shaped by the biogeochemical dynamics characterizing each environment.…”