Complete genome sequences of the Serratia plymuthica strains 3Rp8 and 3Re4-18, two rhizosphere bacteria with antagonistic activity towards fungal phytopathogens and plant growth promoting abilities

The Serratia plymuthica strains 3Rp8 and 3Re4-18 are motile, Gram-negative, non-sporulating bacteria. Strain 3Rp8 was isolated from the rhizosphere of Brassica napus L. and strain 3Re4-18 from the endorhiza of Solanum tuberosum L. Studies have shown in vitro activity against the soil-borne fungi Verticillium dahliae Kleb., Rhizoctonia solani Kühn, and Sclerotinia sclerotiorum. Here, we announce and describe the complete genome sequence of S. plymuthica 3Rp8 consisting of a single circular chromosome of 5.5 Mb that encodes 4954 protein-coding and 108 RNA-only encoding genes and of S. plymuthica 3Re4-18 consisting of a single circular chromosome of 5.4 Mb that encodes 4845 protein-coding and 109 RNA-only encoding genes. The whole genome sequences and annotations are available in NCBI under the locus numbers CP012096 and CP012097, respectively. The genome analyses revealed genes putatively responsible for the promising plant growth promoting and biocontrol properties including predicting factors such as secretion systems, iron scavenging siderophores, chitinases, secreted proteases, glucanases and non-ribosomal peptide synthetases, as well as unique genomic islands.


Introduction
Serratia species are well known for their potential as biocontrol agents with broad-spectrum antagonistic activities against common phytopathogens and their plant growth-promoting abilities. Serratia plymuthica 3Rp8 was isolated as an indigenous colonizer of oilseed rape (Brassica napus L.) rhizosphere and is an in vitro antagonist of the soil-borne fungal phytopathogens Verticillium dahliae Kleb., Rhizoctonia solani Kühn and Sclerotinia sclerotiorum [1] which can cause severe yield losses in a large number of different crops. Chitinase and protease activity were demonstrated by plate assays and the production of N-acylhomoserine lactones was detected using bioluminescent sensor plasmid pSB403 [1,2]. Serratia plymuthica 3Re4-18 was isolated from the endorhiza of a potato plant (Solanum tuberosum L.) and was identified as the most effective isolate in an in vitro study screening potato-associated bacterial communities for antagonistic functions against plant pathogenic fungi [3]. Both strains were sequenced to augment current studies targeting novel biotechnological applications for seed and root treatment since the strains represent promising candidates for biological control. In this report, we summarize the complete genome sequences and annotations of S. plymuthica 3Rp8 and 3Re4-18 and describe their genomic properties. Analysis of the genomes of 3Rp8 and 3Re4-18 will provide a framework for further studies of their rhizosphere competence, biocontrol properties, and plant growth promoting activity. 3Rp8 and 3Re4-18 are deposited in the strain collection of antagonistic microorganisms at Graz University of Technology, Institute of Environmental Biotechnology, Austria.

Organism information
Classification and features S. plymuthica 3Rp8 and 3Re4-18 are motile, Gramnegative, non-sporulating Enterobacteriaceae. Colonies appear yellow-beige opaque, domed and moderately mucoid with smooth margins on Luria-Bertani (LB) solid media and form colonies within 24 h at 20°C (Fig. 1a-b). Both strains grow in standard complex media such as LB, potato dextrose agar (PDA), Waksman agar (WA) and nutrient agar (NA) [4] as well as in minimal medium such as Standard Succinate Medium (SSM). The standard growth temperature is at 30°C, but both strains can replicate in liquid LB at 5°C and at 40°C as well. Both strains do not show a production of red pigments on the media mentioned above. The rod-shaped cells are approximately 0.5 μm in width and 2.0 μm in length ( Fig. 1c-d).
3Rp8 was isolated from the roots of oilseed rape cultivar Express grown for a field trial in Braunschweig (Germany) in 1998 [1,5]. 3Re4-18 was isolated from the endorhiza of an early senescent Solanum tuberosum L. cultivar Cilena at the experimental station of the Institute for Plant Diseases, Bonn University in Bonn-Poppelsdorf (Germany) in 2001 [3].
Both bacterial strains are efficient colonizer of oilseed rape and cauliflower [4], lettuce and pumpkin roots (unpublished data) and do not cause any obvious negative effects to those hosts. Priming of oilseed rape and cauliflower seeds with the S. plymuthica 3Rp8 and 3Re4-18 strains had a significant PGP effect on the root weights of the oilseed rape seedlings [4].

Genome project history
The strains S. plymuthica 3Rp8 and 3Re4-18 were selected for sequencing due to their in vitro activity against V. dahliae and R. solani, their production of hydrolytic enzymes and their root-associated lifestyle on plants [1,3,4]. The sequence data will help to reveal genetic features responsible for their plant growth promoting effects and their ability to protect seeds against fungal threats during germination. The genome project is deposited in the NCBI BioProject database under ID 289082 with the Biosample UIDs 3841799 and 3841798, respectively. The finished genome sequences are deposited in GenBank under the accession numbers CP012096 and CP012097, respectively. A summary of the project information is shown in Table 2.
Growth conditions and genomic DNA preparation 3Rp8 and 3Re4-18 were grown in 50 ml of nutrient broth II (NB II) (Sifin, Berlin, Germany) medium and incubated for 20 h at 30°C. 0.5 ml was then centrifuged at 2500 x g for 5 min at 4°C and genomic DNA was extracted using the MasterPure DNA purification kit (Epicentre, Madison, WI, USA). DNA quality and quantity were checked by agarose gel electrophoresis and spectrophotometry using a UV-Vis spectrophotometer (NanoDrop 2000c, Thermo Fisher Scientific, Waltham, MA USA). Total genomic DNA of 3Rp8 (50.7 μg; 0.8 μg μL -1 ) and of 3Re4-18 (102.8 μg; 1.7 μg μL -1 ) was sent on dry ice to the sequencing service.

Genome sequencing and assembly
PacBio RS libraries with inserts of 8 to 20 kb were constructed and sequenced at GATC Biotech (Konstanz, Germany) using single molecule, real-time (SMRT) sequencing. Assemblies were completed with the Hierarchical Genome Assembly Process v. 2.2.0 (HGAP) algorithm implemented in the PacBio SMRT Analysis software (Pacific Biosciences, Menlo Park, CA, USA). The assembly of the 3Rp8 genome was based on 119,662 quality reads with a mean length of 4581 bp resulting in a single circular chromosome consisting of 5,546,041 bp with 81-fold overall coverage. For assembling the genome of 3Re4-18, 127,834 quality reads with a mean length of 5358 bp were used resulting in a single circular chromosome of 5,439,574 bp with 110fold overall coverage.

Fig. 2
Maximum likelihood 16S rDNA phylogenetic tree indicating the phylogenetic relationship of sequenced isolates. The phylogenetic relationships inferred from the alignment of 1532 bp of 16S rDNA highlighting the positions of S. plymuthica 3Rp8 and 3Re4-18 relative to their closest Serratia strains for which 16S rDNA sequences are publicly available. A representative rhizosphere bacterium from the genera Pseudomonas was used as outgroup. The evolutionary history was inferred by using the Maximum Likelihood method based on the Tamura-Nei model [35]. The percentage of trees in which the associated taxa clustered in the bootstrap test (1000 replicates) is shown next to the branches [36]. Evolutionary analyses were conducted in MEGA7 [37] Fig. 3 Maximum likelihood phylogenetic tree inferred from three housekeeping genes. The phylogenetic relationships inferred from the alignment of 8077 bp of concatenated DNA from three housekeeping genes highlighting the positions of S. plymuthica 3Rp8 and 3Re4-18 relative to their closest Serratia strains for which complete genomes are publicly available. A representative rhizosphere bacterium from the genera Pseudomonas was used as outgroup. For the construction of the tree, the protein-coding house-keeping genes gyrB (2420 bp), rpoP (4146 bp) and nusA (1511 bp) were concatenated and aligned. Then the evolutionary history was inferred by using the Maximum Likelihood method based on the Tamura-Nei model [35]. The percentage of trees in which the associated taxa clustered in the bootstrap test (1000 replicates) is shown next to the branches [36]. Evolutionary analyses were conducted in MEGA7 [37] Among the 5005 predicted genes, 4845 (96.80 %) were identified as protein coding genes, 51 (1.02 %) were designated as pseudo genes, 22 (0.44 %) as rRNAs, 86 (1.72 %) as tRNAs and one (0.02 %) as ncRNA. 19 (0.38 %) genes were frameshifted.
The GC contents of both strains are similar to that of other S. plymuthica strains. The classification of CDSs into functional categories according to the COG database [14,15] is summarized in Table 4 on BASys gene prediction.

Insights from the genome sequence
Both strains share a collection of genes that may be important contributors to biological control with other S. plymuthica strains already published, like genes annotated as secretion systems, iron scavenging siderophores (locus tags ADP72_19185, ADP73_16995), chitinases (e.g. locus tags ADP72_04805, ADP73_00825), secreted proteases (e.g. locus tags ADP72_11930, ADP73_24375), glucanases (e.g. locus tags ADP72_10355, ADP73_00890) and non-ribosomal peptide synthetases (e.g. locus tags ADP72_05100, ADP73_05800). Additionally, genes predicting plant growth promotion, like spermidine synthases (e.g. locus tags ADP72_15170, ADP73_11985), indole-3pyruvate decarboxylases (locus tags ADP72_18190, ADP73_17980) or diacetyl-reductase (locus tags ADP72_1 9475, ADP73_16745) were detected. Unique genomic islands were identified in both strains with IslandViewer 3 software [16][17][18]. In 3Rp8 coding regions containing high similarities on DNA-level with a region in Photorhabdus luminescens TT01 [19] as well as a region annotated as type IV/VI secretion system were found. In 3Re4-18 unique coding regions for proteins related to type VI secretion systems as well as other islands with putatively phage origin were detected.

Conclusions
Here, we announce the complete genome sequences of Serratia plymuthica 3Rp8 and 3Re4-18, two enterobacteria that were originally isolated in Germany from oilseed rape rhizosphere and from endorhiza of potato, respectively. Both strains were selected for sequencing based on their ability to control soil-borne plantpathogenic fungi. Such properties likely have origins in a repertoire of genes probably involved in fungal cell wall degradation expressed by chitinases, proteases or non-ribosomal peptide synthetases. They also share a Fig. 4 Phylogenomic overview using ANI data calculated from whole genome sequences. The heat-plot was compiled in Gegenees [6] and is based on a fragmented alignment using BLASTN made with settings 200/100 (accurate calculation). The cutoff threshold for non-conserved material was set to 30 %   . The outer scale is marked every 10 kb. Circles range from 1 (outer circle) to 7 (inner circle). Circle 1 and 2, ORFs encoded by leading and lagging strand respectively, with color code for functions: salmon, translation, ribosomal structure and biogenesis; aquamarine, RNA processing and modification; light blue, transcription; cyan, DNA replication, recombination and repair; tan, chromatin structure and dynamics; turquoise, cell division; dark orange, defense mechanisms; deep pink, post-translational modification, protein turnover and chaperones; dark olive green, cell envelope biogenesis; purple, cell motility and secretion; lavender, intracellular trafficking, secretion, and vesicular transport; forest green, inorganic ion transport and metabolism; pink, signal transduction; red, energy production; sienna, carbohydrate transport and metabolism; yellow, amino acid transport; orange, nucleotide transport and metabolism; gold, co-enzyme transport and metabolism; cornflower blue, lipid metabolism; blue, secondary metabolites, transport and catabolism; gray, general function prediction only; yellow green, unknown function; black, function unclassified or unknown. Circle 3 and 4, distributions of tRNA genes and rrn operons respectively. Circle 5, distribution of pseudogenes. Circle 6 and 7, G + C content and GC skew (G-C/G + C) respectively collection of genes known to be responsible for specific PGP features and both carry unique genomic islands with interesting genes for agricultural applications. Further functional studies and comparative genomics with related isolates will greatly enhance the understanding of biocontrol and PGP features. The percentage is based on the total number of protein coding genes in the genome based on BASys gene prediction [7][8][9]