Comparative Genomics of Pseudomonas sp. Strain SI-3 Associated With Macroalga Ulva prolifera, the Causative Species for Green Tide in the Yellow Sea

Abstract

Algae-bacteria associations occurred widely in marine habitats, however, contributions of bacteria to macroalgal blooming were almost unknown. In this study, a potential endophytic strain SI-3 was isolated from Ulva prolifera, the causative species for the world's largest green tide in the Yellow Sea, following a strict bleaching treatment to eliminate epiphytes. The genomic sequence of SI-3 was determined in size of 4.8 Mb and SI-3 was found to be mostly closed to Pseudomonas stutzeri. To evaluate the characteristics of SI-3 as a potential endophyte, the genomes of SI-3 and other 20 P. stutzeri strains were compared. We found that SI-3 had more strain-specific genes than most of the 20 P. stutzeri strains. Clusters of Orthologous Groups (COGs) analysis revealed that SI-3 had a higher proportion of genes assigned to transcriptional regulation and signal transduction compared with the 20 P. stutzeri strains, including four rhizosphere bacteria, indicating a complicated interaction network between SI-3 and its host. P. stutzeri is renowned for its metabolic versatility in aromatic compounds degradation. However, significant gene loss was observed in several aromatic compounds degradation pathways in SI-3, which may be an evolutional adaptation that developed upon association with its host. KEGG analysis revealed that dissimilatory nitrate reduction to ammonium (DNRA) and denitrification, two competing dissimilatory nitrate reduction pathways, co-occurred in the genome of SI-3, like most of the other 20 P. stutzeri strains. We speculated that DNRA of SI-3 may contribute a competitive advantage in nitrogen acquisition of U. prolifera by conserving nitrogen in ${\text{NH}}_4^{+}$ form, as in the case of microalgae bloom. Collectively, these data suggest that Pseudomonas sp. strain SI-3 was a suitable candidate for investigation of the algae-bacteria interaction with U. prolifera and the ecological impacts on algal blooming.

Keywords: Comparative genomics; Pseudomonas; Ulva prolifera; algae-associated bacteria; genome sequence.

Figure 1

The epiphytes removal effect of ethanol plus bleach sterilization using SEM. (A) Untreated U. prolifera. (B) Ehtanol plus bleach treated U. prolfiera.

Figure 2

Genome features of Pseudomonas sp. strain SI-3. (A) Graphical map of chromosome of Pseudomonas sp. strain SI-3. The outer scale is marked in 0.1 Mb. From the outside to the center of each circle: Circle 1 and 2, genes encoded on forward and reverse strands, respectively. Coding sequences are colored by COG categories. Circle 3, RNA genes. Circle 4, GC content. Circle 5, GC skew (G–C/G+C). (B) COG function classification of SI-3. The ordinate axis indicates the gene numbers in each COG functional category. (C) Genomic islands predicted by two methods. Blue lines present the genomic islands predicted by IslandPATH-DIMOB method. Yellow lines present the genomic islands predicted by SIGI-HMM method. Red lines present the integrated genomic islands predicted by two methods. The second circle indicates the GC content.

Figure 3

Phylogenetic trees of Pseudomonas sp. strain SI-3 and 20 other P. stutzeri strains. The Maximum Likelihood (ML) phylogeny trees were constructed based on amino acid sequences of housekeeping gene rpoD (A) and concatenated amino acid sequences of single copy orthologous genes (B). The topology of the tree was tested with 1,000 bootstrap replications and bootstrap values were shown near the horizontal branches of the trees.

Figure 4

Venn diagram showing the pan- and core-genome of Pseudomonas sp. strain SI-3 and 20 P. stutzeri strains. The center red circle represented the core-genome. The strain name and total gene number of each ellipse presented were marked outside. The specific genes number of each strain was marked in corresponding ellipse.

Figure 5

Enhanced transcription and signal transduction network in Pseudomonas sp. strain SI-3. The proportions of genes assigned into COG K (A) and COG T (B) categories were normalized to the total gene number of all COG categories in each strain.

Figure 6

Relative contents (RCs) of gene numbers of KEGG metabolic pathways in Pseudomonas sp. strain SI-3. The average gene numbers for each metabolic pathway in strain SI-3 and 20 P. stutzeri strains were set to 1, after which the gene numbers of corresponding KEGG metabolic pathways of SI-3 were normalized. All pathways of SI-3 were ranked based on the RC of genes. The pathways with RCs between 0.8 and 1.2 were omitted to make the results clear. The embedded panel in the upper right was the entire profile of all pathways of SI-3.

Figure 7

Loss of genes involved in aromatic compounds degradation pathways in Pseudomonas sp. strain SI-3. The gene numbers in each strain were normalized to the average gene numbers of SI-3 and 20 P. stutzeri strains in xylene degradation (A), toluene degradation (B), fluorobenzoate degradation (C), and degradation aromatic compounds (D) pathways.