Analysis of the Pseudoalteromonas tunicata genome reveals properties of a surface-associated life style in the marine environment

Abstract

Background: Colonisation of sessile eukaryotic host surfaces (e.g. invertebrates and seaweeds) by bacteria is common in the marine environment and is expected to create significant inter-species competition and other interactions. The bacterium Pseudoalteromonas tunicata is a successful competitor on marine surfaces owing primarily to its ability to produce a number of inhibitory molecules. As such P. tunicata has become a model organism for the studies into processes of surface colonisation and eukaryotic host-bacteria interactions.

Methodology/principal findings: To gain a broader understanding into the adaptation to a surface-associated life-style, we have sequenced and analysed the genome of P. tunicata and compared it to the genomes of closely related strains. We found that the P. tunicata genome contains several genes and gene clusters that are involved in the production of inhibitory compounds against surface competitors and secondary colonisers. Features of P. tunicata's oxidative stress response, iron scavenging and nutrient acquisition show that the organism is well adapted to high-density communities on surfaces. Variation of the P. tunicata genome is suggested by several landmarks of genetic rearrangements and mobile genetic elements (e.g. transposons, CRISPRs, phage). Surface attachment is likely to be mediated by curli, novel pili, a number of extracellular polymers and potentially other unexpected cell surface proteins. The P. tunicata genome also shows a utilisation pattern of extracellular polymers that would avoid a degradation of its recognised hosts, while potentially causing detrimental effects on other host types. In addition, the prevalence of recognised virulence genes suggests that P. tunicata has the potential for pathogenic interactions.

Conclusions/significance: The genome analysis has revealed several physiological features that would provide P. tunciata with competitive advantage against other members of the surface-associated community. We have also identified properties that could mediate interactions with surfaces other than its currently recognised hosts. This together with the detection of known virulence genes leads to the hypothesis that P. tunicata maintains a carefully regulated balance between beneficial and detrimental interactions with a range of host surfaces.

Figure 1. Phylogenetic tree based on 16S rRNA gene sequence of Pseudoalteromonas and Alteromonas species compared in the study.

Tree was generated by maximum likelihood analysis of 1276 nucleotide positions. The sequence alignment and phylogenetic calculations were performed and manually checked with the ARB software package [97]. The 16S rRNA gene sequence of Silicibacter pomeroyi DSS-3 was used as an outgroup.

Figure 2. Functional comparison of Pseudoalteromonas and Alteromonas genomes.

Relative abundance compared to all COGs (panel A) and absolute number of categories (panel B) for selected Pseudoalteromonas and Alteromonas species. COGs were extracted from IMG using greater than 30% identity and expectancy values of less than 10⁻⁵ cut-offs and assigned to functional categories.

Figure 3. Phylogeny of pili proteins.

Phylogenetic tree of pili proteins found in the P. tunicata genome. The pili proteins of P. tunicata are highlighted in bold and a specific name is given, if characterised. Other characterised pili proteins are shown in bold. The ten best blast-hits in NCBI's non-redundant database of the putative pili protein of P. tunicata and characterised pili proteins were used to construct the tree. Accession number and taxonomic source are shown. Bootstraps values over 750 are not shown.

Figure 4. Functional properties of predicted P. tunicata secretome.

Major protein families identified in the secretome of P. tunicata according to the Pfam database. Hypothetical proteins (141 entries) did not match to any HMM in the Pfam database; 20 proteins have the TonB-dependent receptor domain (TonB_dep_Rec); the glycolytic hydrolases (‘glyco_hydro’) group contains 11 proteins in total; 36 proteins have HMM domains that are found in different peptidases; ‘a/amido-hydrolases’ refers to alpha- and amido-hydrolases with 6 proteins in total; ‘MipA’ refers to Mlt-interacting protein like sequences with 5 sequences; ‘SBP_bac’ for extracellular solute binding protein includes SBP_bac_1 (1) and SBP_bac_3 (5); ‘Porin-like’ refers to the HMMs for OmpA (2), OmpH (1), OmpW (1), Porin_O_P (1) and Porin_1 (1); ‘MCP_signal’ refers to methyl-accepting chemotaxis protein (5); and ‘Others’ refers to all the other sequences that matched to different protein families in the Pfam database (135 entries).

Figure 5. Peptidases in predicted secretome of P. tunicata.

Comparison of the peptidase profiles in the secretome of the three sequenced Pseudoalteromonas species. The total number of proteins that belong to the peptidase group is 36 in P. tunicata, 12 in P. atlantica and 15 in P. haloplanktis. The colour bars represent the percentage of each subgroup of peptidases. Details of each peptidase family type can be found at http://merops.sanger.ac.uk.