Comparative genomics and metabolic profiling of the genus Lysobacter

Abstract

Background: Lysobacter species are Gram-negative bacteria widely distributed in soil, plant and freshwater habitats. Lysobacter owes its name to the lytic effects on other microorganisms. To better understand their ecology and interactions with other (micro)organisms, five Lysobacter strains representing the four species L. enzymogenes, L. capsici, L. gummosus and L. antibioticus were subjected to genomics and metabolomics analyses.

Results: Comparative genomics revealed a diverse genome content among the Lysobacter species with a core genome of 2,891 and a pangenome of 10,028 coding sequences. Genes encoding type I, II, III, IV, V secretion systems and type IV pili were highly conserved in all five genomes, whereas type VI secretion systems were only found in L. enzymogenes and L. gummosus. Genes encoding components of the flagellar apparatus were absent in the two sequenced L. antibioticus strains. The genomes contained a large number of genes encoding extracellular enzymes including chitinases, glucanases and peptidases. Various nonribosomal peptide synthase (NRPS) and polyketide synthase (PKS) gene clusters encoding putative bioactive metabolites were identified but only few of these clusters were shared between the different species. Metabolic profiling by imaging mass spectrometry complemented, in part, the in silico genome analyses and allowed visualisation of the spatial distribution patterns of several secondary metabolites produced by or induced in Lysobacter species during interactions with the soil-borne fungus Rhizoctonia solani.

Conclusions: Our work shows that mining the genomes of Lysobacter species in combination with metabolic profiling provides novel insights into the genomic and metabolic potential of this widely distributed but understudied and versatile bacterial genus.

Fig. 1

Whole genome phylogeny of Lysobacter species. The tree is based on the total nucleotide genome sequences using the Gegenees software. Asterisks indicate the previously published and publically available Lysobacter genome sequences. Genome sequences of Stenotrophomonas maltophilia and Xanthomonas were used as outgroups. S. mal: Stenotrophomonas maltophilia; X. alb: Xanthomonas albilineans; X. axo_cit_F1: Xanthomonas axonoponis citrumelo F1; X. axo_cit 306: X. axonoponis citri 306; X. cam: Xanthomonas campestris; X. cam_rap: X. campestris raphani; X.ory: Xanthomonas oryzae pv. oryzae; X.ory_ory: X. oryzae pv. oryzicola

Fig. 2

Genomic diversity of Lysobacter. The number of unique coding sequences (CDSs) shared by the Lysobacter strains representing the core genome is shown in the centre. The total number of CDSs in the analysis was 32019. Overlapping regions show the number of CDSs conserved only within the specified genomes. Numbers in non-overlapping portions of each oval show the number of CDSs unique to each strain. The total number of protein CDSs within each genome is listed below the strain name

Fig. 3

Antimicrobial and extracellular enzyme activity of Lysobacter. Antimicrobial activity against a) Rhizoctonia solani on R2A medium, and b) Xanthomonas campestris pv campestris (Xcc) on 1/5th strength Potato Dextrose Agar. The Lysobacter strains were spot inoculated on medium containing Xcc cells. c Chitinase activity on R2A medium supplemented with 0.2 % colloidal chitin. d siderophore production on CAS medium. When iron is chelated, the medium turns from blue to orange. e lipase activity on R2A medium supplemented with 0.01 % CaCl2 and 1 % Tween 80. Lipase activity is visible as a precipitation of calcium crystals surrounding the colony. 1: L. antibioticus ATCC29479; 2: L. antibioticus 76; 3: L. capsici 55; 4: L. gummosus 3.2.11; 5: L. enzymogenes C3; 6: L. enzymogenes DCA is a mutant of L. enzymogenes strain C3 with a Tn5 insertion in the gene encoding the global regulator Clp

Fig. 4

Lysobacter genes encoding the diffusible signal factor (DSF)-dependent system or the CRP-like protein Clp. The gene clusters were identified by BLASTp analysis using the reference gene cluster from Stenotrophomonas maltophilia. Black arrows indicate the CDSs with an identity >60 %; the grey arrows indicate the CDSs with an identity >40-60 %

Fig. 5

Genetic organization of Type III, Type IV and Type VI secretion pathways in the Lysobacter genomes. Arrows represent relative position and transcriptional direction for each gene. Gene/COG calls are indicated above the arrows. Similar coloured boxes between strains represent homologues and numbers within arrows represent GeneIDs. Gene sizes are not drawn to scale. Coloured boxes represent known core type III secretion homologues; closed boxes represent conserved hypothetical proteins; dashed boxes represent non-homologous hypothetical proteins (between strains). a Type III secretion. b Type IV secretion; virB5a and virB5b represent duplicate genes. Arrows below virB6 and virB5b represent duplicate gene pairs located throughout the genomes. The dashed line between virD4 and virB6 indicates that genes are distally located. c Type VI secretion with each cluster represented as a contiguous gene set as indicated by gene identification number. Two gene clusters are identified in L. enzymogenes clusters and one in L. gummosus 3.2.11. Stippled boxes represent type VI secretion genes unique to each cluster

Fig. 6

MALDI imaging mass spectrometry (IMS) of Lysobacter species. At the top right, the positions of the Lysobacter strains grown on R2A medium on the MALDI slide are shown. Lysobacter enzymogenes DCA is a mutant of L. enzymogenes strain C3 with a Tn5 insertion in the gene encoding the global regulator Clp. The mass spectra are shown on the left and the images at specific m/z values are shown on the right. The colour gradient bar shows the % intensity based on absorbance units after normalisation to the total ion count

Fig. 7

MALDI imaging mass spectrometry of Lysobacter-fungus interactions. At the top, the mass spectra of the Lysobacter strains grown on R2A medium in absence and presence of the fungus Rhizoctonia solani are shown. At the bottom the images of metabolites with specific m/z values are shown. The colour gradient bar shows the % intensity based on aborbance units after normalisation to the total ion count. The dashed circle depicts the point where the fungus R. solani was inoculated