- Research article
- Open Access
Genomic analysis and temperature-dependent transcriptome profiles of the rhizosphere originating strain Pseudomonas aeruginosa M18
BMC Genomicsvolume 12, Article number: 438 (2011)
Our previously published reports have described an effective biocontrol agent named Pseudomonas sp. M18 as its 16S rDNA sequence and several regulator genes share homologous sequences with those of P. aeruginosa, but there are several unusual phenotypic features. This study aims to explore its strain specific genomic features and gene expression patterns at different temperatures.
The complete M18 genome is composed of a single chromosome of 6,327,754 base pairs containing 5684 open reading frames. Seven genomic islands, including two novel prophages and five specific non-phage islands were identified besides the conserved P. aeruginosa core genome. Each prophage contains a putative chitinase coding gene, and the prophage II contains a capB gene encoding a putative cold stress protein. The non-phage genomic islands contain genes responsible for pyoluteorin biosynthesis, environmental substance degradation and type I and III restriction-modification systems. Compared with other P. aeruginosa strains, the fewest number (3) of insertion sequences and the most number (3) of clustered regularly interspaced short palindromic repeats in M18 genome may contribute to the relative genome stability. Although the M18 genome is most closely related to that of P. aeruginosa strain LESB58, the strain M18 is more susceptible to several antimicrobial agents and easier to be erased in a mouse acute lung infection model than the strain LESB58. The whole M18 transcriptomic analysis indicated that 10.6% of the expressed genes are temperature-dependent, with 22 genes up-regulated at 28°C in three non-phage genomic islands and one prophage but none at 37°C.
The P. aeruginosa strain M18 has evolved its specific genomic structures and temperature dependent expression patterns to meet the requirement of its fitness and competitiveness under selective pressures imposed on the strain in rhizosphere niche.
Previously published work from our group has described an effective biocontrol strain M18 that was isolated from sweet melon rhizosphere in Shanghai suburb in 1996 . This strain has been named as the Pseudomonas sp. M18, as its 16S ribosomal DNA sequence shares a high similarity to that of P. aeruginosa, as well as high sequence identities of several global regulatory genes, including gacA, rpoS, qscR, rhlI, lasI and rsmA. However, the strain M18 has developed several unusual features which distinguish it from that of the clinically significant P. aeruginosa strains, and indeed are more similar to soil-inhabiting biocontrol Pseudomonas strains, such as P. fluorescence 2-79 , Pf-5  and P. chlororaphis PCL1391 . Firstly, the predominant phenazine produced by the strain M18 is phenazine-1-carboxylic acid (PCA) rather than pyocyanin (PYO) . PCA is considered to be a major biocontrol compound , while PYO is not necessary for fungal killing . It was determined that more PCA could be produced in the strain M18 at 28°C rather than 37°C compared with that produced by the P. aeruginosa strain PAO1 . Secondly, the strain M18 is the first strain that was reported to produce two antibiotics, PCA and pyoluteorin (Plt) in one single strain . The production of the two antibiotics of the strain M18 can result in synergistic antifungal activities against several phytopathogens. The biosynthesis gene clusters of two antibiotics share 99% nucleotide sequence identity with the clusters of Liverpool Epidemic Strain P. aeruginosa LESB58. However, a 5-base-pair deletion exists in the pltB gene of Plt synthetic cluster resulting in a frameshift mutation in the strain LESB58 . Thirdly, several interesting features with regards to the regulatory mechanism of the production of these two antibiotics were found to be quite different from that in P. aeruginosa. For example, PCA production is negatively regulated and Plt production is positively regulated by a global regulator named GacA . However, the converse relationship occurs in the strain M18 through another global regulator, RsmA . In a more recent study, we demonstrated the negative effect of the QscR regulator on PCA production, but not on Plt in the strain M18 . Studies have also been conducted on the differential regulation of PCA and Plt production by potential quorum sensing (QS) signaling molecule N-acyl homoserine lactones (AHLs) secreted by this strain [5, 6]. The interrelationship between the QS las and rhl system in strain M18 are also different from that in strain P. aeruginosa PAO1 .
Besides the biocontrol strain M18, several P. aeruginosa strains, i.e., 7NSK2 , PNA1 , NJ-15 , PUPa3 , which inhabit the rhizosphere niche, have been found to show excellent plant growth promoting feature and effective biocontrol property against various phytopathogens, and are utilized for biopesticide development. The genetically modified strain M18 has been recently developed as a high yield antifungal antibiotic PCA producer [20–22]. However, little information is available on their specific genome structures, transcriptome profiles and virulence activities compared to their clinical counterparts.
A series of complete genomic sequences of P. aeruginosa strains that originated from patients [14, 23–25] have been published recently, and normal environmental P. aeruginosa isolates were chosen to compare their virulence and genomic structures [26, 27]. To date, none of P. aeruginosa strains that originate from the rhizosphere has been completely sequenced and analyzed in detail. This study aims to delineate the genomic structural differences among various P. aeruginosa strains that originate from nosocomial patients and rhizosphere niches, and to improve our general understanding of what kinds of strategies have been adopted by the P. aeruginosa genome in response to various environment conditions, especially to different temperatures. In this report, the complete genome of the strain M18 was sequenced and compared with those of previously sequenced nosocomial strains. Furthermore, the antimicrobial agent resistance in vitro and the competitive growth ability in a mouse model of acute lung infection in vivo were investigated in the two strains of M18 and LESB58. The temperature dependent genome expression pattern based on the M18 genome was finally analyzed at 28°C and 37°C.
Results and discussion
Strain M18 is an effective biological control agent
The strain M18 was isolated from the sweet melon rhizosphere that grew in Shanghai suburb in 1996, and proved to have effective biocontrol activity to protect plants against various phytopathogens infection . In this report, we first provide the evidence to show that the secondary metabolites produced from the strain M18 have strong effective inhibitory activity against the growth of the Mycosphaerella melonis mycelium plug on a potato dextrose agar (PDA) plate (Figure 1A) and therapeutic anti-fungal effect on the sweat melon plant with full recovery of the plant infected by the M. melonis fungus two days after the culture of 108 CFU/ml of the strain M18 that was sprayed on the infected sweat melon stem (Figure 1B and 1C). We have already found that the strain M18 has a broad spectrum of anti-fungal activities against various phytopathogens, including Rhizoctonia solani, Sclerotina sclerotiorum, Fusarium oxysporum, Colletotrichum gossypii, Mycosphaerella melonis, Phytophthora capsici, Ralstonia solanacearum and others. The strain M18 also shows effective inhibiting activity against both the Gram-negative bacterium Xanthomonas oryzae pv. Oryzae, and the Gram-positive bacterium Bacillus cereus (data not shown). Therefore, the unusual features of the strain M18 against various fungal phytopathogens and bacteria prompted us to sequence the complete genome of this strain to ascertain more about the genomic structures, with specific reference to this expression pattern at different temperatures through analysis of microarray-based transcriptome profiles.
General genomic features and comparative analysis
The complete genome of the strain M18 was sequenced and assembled as shown in Figure 2. A general comparison of the genomic features of the strain M18 and other sequenced nosocomial P. aeruginosa strains is shown in Table 1. The average GC content of the M18 genome is 66.5%, which is remarkably similar to the previously sequenced P. aeruginosa strains PAO1, PA14, LESB58 and PA7. The genome of the strain M18 is made of a single chromosome of 6,327,754 bp in size, which is slightly larger than the 6,264,404 bp of strain PAO1 , but relatively smaller than the 6,537,648 bp of strain PA14 , 6,601,757 bp of LESB58  and 6,588,339 bp of strain PA7 . The M18 genome was annotated to contain 5,684 open reading frames (ORFs) and 80 RNA genes, which represent 89% of the total genomic DNA. Using the mGenomeSubtractor online software , we found that the genome similarity among these five P. aeruginosa strains is very high; however 93 ORFs in the M18 genome, 109 ORFs in PAO1, 256 ORFs in PA14, 380 ORFs in LESB58 and 929 ORFs in PA7 are strain-specific under the threshold of 0.6 of identity or 0.7 of match length/query length. The strain specific genes contained in the M18 genome are listed in Additional file 1. These results indicated that the available genomes of P. aeruginosa strains originating from both clinical and environmental sources contain a highly conserved core genome in large than 90% of ORFs, with several accessory regions in less than 10% of ORFs remarkably varied depending on their different environmental niches (except taxonomic outlier PA7).
Global alignments were performed by the MUMer software , and the results showed that the large segment inversion of the chromosome by recombination between the rrnA and rrnB genes was found in the M18 genome, as well as in other P. aeruginosa strains LESB58, PA14 and PA7 (Figure 3B) in comparison with that in PAO1 genome sequenced by the Seattle consortium . Interestingly, the inversion also exists in different PAO1 sublines which are studied and stored in German and American laboratories, and the phenotypes among the PAO1 sublines are also different . This suggests that some physiological consequences occur among the PAO1 sublines and other P. aeruginosa strains because of the inversion .
The comparison of the M18 genome and the other available P. aeruginosa genomes with regards to the functional category breakdown of coding sequences (CDSs)  shows a similar distribution in most groups, but the two strains M18 and LESB58 show a greater degree of similarity than others, including a similar number of genes related to DNA replication, recombination and repair, the transport of inorganic ions and amino acids, amino acid metabolism, and the biosynthesis, transport and catabolism of secondary metabolites (Table 1). This suggests that the two strains share a common ancestor as they have the closest relationship in comparison with all other sequenced P. aeruginosa strains. The phylogenetic tree was further generated by maximum parsimony (MP) method, using 58 phylogenetic useful genes in P. aeruginosa. The result confirmed that the strain M18 is the closest relative to strain LESB58 among the sequence available P. aeruginosa strains, as shown in Figure 3A. We also found that nine out of eleven O-antigen loci in the strain M18 share more than 95% similarity to the serotype O6 loci in LESB58 , indicating that M18 may carry a similar serotype O6 O-antigen. Based on these sequence data, the strain M18 is now designated as P. aeruginosa M18, a new isolate of P. aeruginosa.
Genomic islands and prophages
In comparison with other sequenced nosocomial P. aeruginosa genomes, the unique regions of genomic plasticity were predicted in the M18 genome, including five specific genomic islands (GIs) designated MGI-I to V and two novel prophages, I and II. The general features of these GIs and prophages in the strain M18 are summarized in Figure 4A and Additional file 2. However, we have not found homologies of the pathogenic islands against mammals such as PAPI-1 , PAPI-2 , PAGI-5 , LESGI-5 , in the M18 genome.
To the best of our knowledge, the two GIs, MGI-I and MGI-II are initially found in P. aeruginosa M18, and share 38% identity with that in Pseudomonas putida GB-1 and 55% with Nitrosomonas europaea ATCC 19718, respectively. The GC content of MGI-I (51.7%) and MGI-II (55%) are quite lower than the average GC content (66.5%) of the M18 genome. Difference in the percentage of GC content between the GIs and the core genome is considered as an important marker of horizontal gene transfer, suggesting that these two GIs are associated with recently acquired genetic material. MGI-I contains two genes (PAM18_0699 and PAM18_0700) that code for putative integrases, and MGI-II contains a gene (PAM18_2243) that encodes a putative transposase. MGI-I contains another two genes, mod (PAM18_0704) and res (PAM18_0705), which encode two subunits of the type III restriction-modification (RM) system. MGI-II contains another three genes, hsdR (PAM18_2237), hsdM (PAM18_2238) and hsdS (PAM18_2239) that code for all the three subunits in the multifunctional type I RM system which quite differ from the two type I RM systems in the PA7 genome. These two restriction systems contained in a single M18 genome can be predicated to assist this strain in digesting foreign genetic material , which may explain why P. aeruginosa M18 has proven to be recalcitrant to the introduction of foreign DNA and to keep the strain isolated from others within rhizosphere niches.
The GC content of MGI-III (65.6%) is similar to that of the average content in the M18 genome. MGI-III contains a plt synthetic gene cluster with its two flanking regulatory genes and transporter gene cluster (pltMRLABCDEFGZHIJKNO)  and shares 99% nucleotide sequence identity with the LESGI-2 sequence in LESB58, suggesting that the two islands may originate from the same genetic source and remain in the two genomes. However, LESGI-2 contains a frameshift mutation in pltB of the plt gene cluster . The plt synthetic gene cluster is inactivated in strain LESB58, which might the result of the unnecessary Plt production in the nosocomial niche. However, Plt can be produced in the plant-associated pseudomonads such as strain M18  and P. fluorescens Pf-5  and plays a crucial anti-fungal function to suppress a variety of plant diseases in rhizosphere niches.
The GC content of MGI-IV (63.1%) and MGI-V (59.5%) is closer to the average content of the M18 genome than those of MGI-I and MGI-II, indicating that the two islands may be not the recently acquired genetic materials. A homologous sequence of MGI-IV was found in P. aeruginosa PAK . The MGI-IV contains genes encoding acetyltransferase, 3-demethylubiquinone-9,3-methyltransferase, CMP-2-keto-3- deoxyoctulosonic acid synthetase, phenylpropionate dioxygenase, ring-hydroxylating related dioxygenases, and nucleotide sugar transaminase, which are involved in flagellar glycosylation, secondary metabolite synthesis, and redox reaction in the rhizosphere environment. The MGI-V is a novel genomic island which has never been found in other prokaryotic genomes. The gene (PAM18_5672) in this island encodes a putative cytochrome P450, which can catalyze the oxidation of several metabolic intermediates, such as lipids and steroidal hormones, as well as xenobiotic substances, including drugs and other toxic chemicals . The two genomic islands in the M18 genome may assist the strain M18 to utilize various substances and degrade environmental toxic materials and pollutants in the complex rhizosphere environment.
Two novel prophages named prophage I and II were found in the M18 genome. Each of the prophage contains a putative chitinase gene (PAM18_1344 and PAM18_2447 . Chitinase is a specific degradation enzyme that breaks down glycoside bonds in chitin and is essential for the degradation of chitin, a major component of the fungal cell wall. Besides the three copies of capB gene in the M18 genome, an additional copy of capB gene was identified in prophage II. The capB gene is responsible for encoding a cold shock protein B involving in the adaptation to cold stress in the environment . The two chitinase genes and an additional capB gene in the two prophages in the M18 genome suggest that the strain M18 has developed its anti-fungal and cold stress resistant genomic features under the strain-specific environmental selective pressures imposed on it.
Insertion sequence (IS) and clustered regularly interspaced short palindromic repeats (CRISPR)
A striking feature in the M18 genome is the least number and diversity of 3 IS elements named IS222 (1,227 bp), ISPa32 (1,232 bp) and ISPa6 (1,316 bp) as shown in Figure 4B). However, the other completely sequenced P. aeruginosa strains, PAO1, PA14, PA7, and LESB58 carry 12, 10, 15 and 8 IS elements, respectively (Table 1). To our knowledge, the IS elements are postulated to be important drivers of many bacterial genome evolution, such as Xanthomonas. In addition to serving as vectors for lateral gene transfer, IS elements can generate other types of genome modification, including rearrangements, inversions and deletions, any of which can increase the instability of the bacterial genome. Therefore, the least number of IS elements in the genome indicates the relative genome stability of the strain M18 in the rhizosphere niches.
As predicted by the CRISPRFinder , three CRISPR elements were found in the M18 genome (Figure 4B). CRISPR 1 element is 987 bp in length and has 16 spacers each flanked by a 28 bp direct repeat (DR). There is one single similar CRISPR element in the LESB58 genome, in which six spacers are different from the CRISPR 1 element. The CRISPR 2 element consists of 866 bp with 14 spacers each flanked by 28 bp DR, and the CRISPR3 is 687 bp in length with 11 spacers and 28 bp DR. The CRISPR 2 and 3 sequences are similar to those in the PA14 genome, which only has the two homologies. Furthermore, most of spacers in the CRISPR 2 and 3 elements are quite different from those in other P. aeruginosa strains. The strains PAO1 and PA7 only carry some questionable CRISPR elements that have borderline sequence identity. The three CRISPR elements in the M18 genome can be found dispersedly in other sequenced P. aeruginosa genomes, indicating that the M18 genome contains the most number of CRISPR elements among the five available complete P. aeruginosa genomes. The different spacer sequences in the three CRISPR loci of the strain M18 indicate that M18 has developed specific features to combat various phage invasions in the rhizosphere environment, as spacer sequences are known to be critical for bacteria to resist foreign phage invasion [46, 47].
Biocontrol-related gene clusters with their products
We identified six secondary metabolite biosynthetic gene clusters, which are related to biocontrol activities of the strain M18, and responsible for biosynthesis of two siderophores, hydrogen cyanide, Plt and PCA (Additional file 3). Among them, two gene clusters are responsible for the biosynthesis of two siderophores, including pyoverdine (Pvd) and pyochelin (Pch), respectively, which have been demonstrated to suppress target phytopathogens in the rhizosphere through iron competition . The hcn cluster responsible for the production of hydrogen cyanide (HCN) is highly conserved in different P. aeruginosa strains. The Plt biosynthesis gene cluster is located in MGI-III and has been described above. The remaining two gene clusters are responsible for PCA biosynthesis. The two phz gene clusters in the genome of the strain M18 are highly conserved and share 99% homologous sequences to those in the P. aeruginosa PAO1, each of which consists of seven genes (phzA1B1C1D1E1F1G1 and phzA2B2C2D2E2F2G2), named phzA1-G1 and phzA2-G2, respectively . The genes that flank these two clusters are also the same to those of P. aeruginosa PAO1. However, there is a 520 bp intergenic region located directly downstream of the phzA2-G2 gene cluster between the phzG2 gene and the ORF (PAM18_3137) in the strain M18, compared to a shorter 96 bp sequence in the strain PAO1 and a relatively longer 286 bp sequence in the LESB58 genome. All these interval regions are consisted of clustered short repeats, although their function remains unknown.
The predominant phenazine produced in strain M18 is PCA, especially at 28°C, rather than PYO produced in other nosocomial originating P. aeruginosa strains . Both PCA and Plt compounds produced by fluorescent Pseudomonas species are considered to play an important role in microbial competitiveness and combating plant pathogens . To our knowledge, the strain M18 is the first strain that was reported to produce both PCA and Plt simultaneously, and the synergistic antibiotic effect of PCA and Plt can result in strong activity to protect plants from fungal phytopathogen infections. It has been found that the wild-type strain M18 has much greater effective inhibitory activity against various mycelium plug growths than either the single PCA or Plt null mutants on the PDA plate (data not shown). We purified the two compounds of PCA and Plt individually from liquid culture, and measured the antifungal activities of the single compound and the combinations of various weight ratios of the two compounds against Rhizoctonia solani mycelium plug growth on PDA plates, respectively. The 50% value of efficient control (EC50) is 8.38 and 58.13 μg/ml for single PCA and Plt, respectively. However, the EC50 of the two compound combination at a weight ratio of 10 to 1 at pH 7.0 drops to 0.74 μg/ml only, indicating a significant synergistic effect of PCA and Plt combination on fungal inhibition. The combinations of the two antibiotics may play a crucial role for the strain M18 to protect the plant against different phytopathogen infections.
Virulence-related genes against mammals
Besides the virulence related genes located in the accessory genome, the LESB58 genome contains all of the 257 genes described for PAO1 with the exception of PA2399 (pvdD) and PA1392 . Based on the VFDB database , we found that 14 out of 255 virulence genes in the M18 genome are quite different from those of LESB58 after BLASTP searching (Table 2). Firstly, six genes in the M18 genome (i.e., flgL, fliC, flaG, fliD, fliS and fliT) share less than 70% identity with the genes in LESB58, which are involved in the synthesis of flagellum components and related to bacterial motility and biofilm formation. Furthermore, the flagellum of the strain LESB58 belongs to a member of b-type because it lacks flagellar glycosylation, and the flagellum of the strain M18 may belong to the a-type resulted by the flagellar glycosylation function of MGI-IV in the M18 genome. These differences may give rise to different ability of bacterial biofilm formation and induce different mammalian immune system between the two strains. Secondly, five out of 42 pyoverdine synthesis-associated genes, which are involved in the iron uptake and cytotoxic due to its ability to stimulate the production of reactive oxygen species , are quite different between the two strains. The gene pvdS (PALES28681) is completely absent in the M18 genome. Notably, there are three copies of fpvA (encoding the first ferric pyoverdine receptor) and pvdE (encoding an ABC transporter) in the LESB58 genome , but only one copy of fpvA (PAM18_2643) and pvdE (PAM18_2644) in the M18 genome; the two genes have only 24.9% and 57.7% sequence identity, respectively, to those in LESB58. Furthermore, pvdD (PAM18_2640) and pvdJ (PAM18_2641), which are predicted to encode the non-ribosomal peptide synthetase in the pyoverdine synthesis pathway, have only 57.2% and 53.4% sequence identity to those in LESB58, respectively.
In the LESB58 genome, the ppkA gene (PALES-00751) codes a positive regulator of type VI secretion system , and the gene lecA (PALES-27241) codes a factor which plays an important role in biofilm formation and adherence to mammalian cells . However, both of the two homologous ppkA and lecA OFRs (PAM18_2466 and PAM18_0076) in M18 genome are truncated because of a stop codon located in the middle of the two ORFs.
The M18 genome carries a gene cluster responsible for lipopolysaccharide (LPS) O-antigen serotype O6 biosynthesis , which shares larger than 90% identity with that in LESB58. Interestingly, the gmd gene within the LPS biosynthesis gene cluster coding for GDP-mannose 4,6-dehydratase is a pseudogene in the LESB58 genome . However, the gmd gene (PAM18_5576) was found to be a true gene in the M18 genome, although the association with the virulence and O-antigen deficiency remains unclear.
Attenuated competitive activity of the strain M18 in a mouse model of acute lung infection
The different genome structures, especially the different mammalian virulence related genes, prompted us to make a comparison of the infectious activities between the two closely related strains M18 and LESB58 in a mammal model in vivo. Thus, we measured the competitive index (CI) of the paired strains to assess their different ability in growth and maintenance in an acute mouse lung infection model (Figure 5). Equal ratio of the two strains was adjusted in phosphate buffered saline (PBS), and the mixture was inoculated into the mouse lung via the nostril. Bacteria were enumerated in the lung tissue at 24 h and 48 h post-infection. The geometric mean CI values were 0.3909 and 0.1695, respectively. The results showed that the strain M18 was easier to be erased than the strain LESB58 in all of the 10 tested mice, and the growth and maintenance abilities of the M18 strain were more attenuated after a prolonged duration of inoculation than LESB58. The preliminary results indicated that the strain LESB58 is more difficult to be erased than the strain M18 in the acute mouse lung infection mode. Further experiments are now being carried out in this laboratory to elucidate the virulence activities of the strain M18 against humans and animal cell lines, and to identify the relationship between the virulence gene expression profiles of the M18 genome in the mouse lung and the responses of mammalian immune system.
Antimicrobial agent resistance spectrum
The minimal inhibitory concentration (MICs) of antimicrobial agents against the two strains M18 and LESB58 were listed in Table 3. The strain M18 is phenotypically resistant to several antimicrobial agents including penicillins, cephalosporins, chloramphenicol, aminoglycosides and macrolides, which is the same as the reported P. aeruginosa strains . The genomic analysis indicated that the strain M18 carries a β-lactamase gene ampC (PAM18_0829) and other eight putative β-lactamase genes (i.e., PAM18_0058, PAM18_1371, PAM18_2047, PAM18_2725, PAM18_3247, PAM18_4437, PAM18_5635, and PAM18_5664), which are responsible for encoding enzymes resistant to β-lactam antibiotics [56, 57]. Furthermore, the M18 genome contains several gene clusters encoding proteins to constitute efflux pumps, such as mexAB-oprM (PAM18_0425-PAM18_0427) , mexCD-oprJ (PAM18_4693-PAM18_4691) , and mexEF-oprN (PAM18_2546-PAM18_2544) , which result in resistance to many antimicrobial agents including chloramphenicol, aminoglycoside, macrolides and rifampicin. Except the aztreonam, the strain M18 is susceptible to most of the clinically applied antimicrobial agents against P. aeruginosa, including third-generation cephalosporins (ceftazidime, ceftriaxone), carbapenems (imipenem, meropenem), monobactam (aztreonam), polymyxin (colistin) and tobramycin. However, comparing with the strain LESB58, the strain M18 is slightly more susceptible to imipenem and meropenem.
Interestingly, compared with P. aeruginosa LESB58, the strain M18 is more susceptible to several antimicrobial agents, such as gentamycin, kanamycin and ofloxacin. Quinolone resistance is usually present in P. aeruginosa strains because of two point mutations in gyrA and parC in the genome, respectively . The lack of the two point mutations in the M18 genome can be postulated to responsible for quinolone susceptibility of the strain M18. Therefore, the results suggest that the rhizosphere isolate M18 has not evolved specific mechanisms including point mutation or acquisition resistant plasmid to resist antimicrobial agents, unlike nosocomial P. aeruginosa strains that face lethal antimicrobial agents stress.
Genomic microarray based transcriptome analysis
Measurement of gene expression levels at two temperatures, 28°C and 37°C, was performed using an oligonucleotide Agilent microarray based on the M18 genome sequence to specify the temperature-dependent expression profiles of this strain. We selected 28°C as a temperature representative for the rhizosphere niches and 37°C for the human body. The results from the temperature-dependent transcriptome analysis were consistent with our previously published data that the expression of phzM, ptsP and lasI gene is up-regulated at 37°C . The comparative analysis of the M18 genome expressional profiles at 28°C and 37°C indicated that the expression levels of a total of 605 genes at 28°C are regulated over two-fold that at 37°C. It shows that a total of 10.6% of the expressed genes in M18 genome are temperature-dependent (Additional file 4).
In comparison with gene expression level at 37°C, the transcriptional expression levels of 277 genes are up-regulated by approximately two-fold at 28°C, including 151 genes with known functions and 126 functions unknown genes. Among these, 77 function known genes are sorted into nine gene families (Table 4). Surprisingly, 24 of these genes are contained in the M18 GIs and prophage, of which eight genes are located in prophage I, two genes are involved in the type I restriction-modification system in MGI-II, seven genes are related to the pyoluteorin transporter in MGI-III, and seven genes in MGI-V are associated with the synthesis of cytochrome P450. We also found that other up-regulated operons are involved in energy production and the conversion of TCA cycle-related oxidative phosphorylation, aromatic amino acid and benzonate degradation, and phosphate uptake. Besides the operons, a potential phenazine modifying enzyme, phzH, a gene that encoded quorum sensing (QS) signal molecule N-butanoyl-L-homoserine lactone synthetase, rhlI, and a gene that encoded the QS repressor, qscR, are promoted at 28°C. The other genes, which are not located in these operons, could be assigned to various functions, including phage-related, FeS cluster assembly, glyoxylate metabolism, benzonate degradation, copper resistance, and pentose phosphate. Hence, the expression levels of these genes involved in antibiotics production, foreign genetic material prevention and nutrient substance degradation are up-regulated at the rhizosphere temperature 28°C, which may give M18 an advantage to resistance the rhizosphere environmental stresses.
In comparison with the gene expression levels at 37°C, a total of 328 gene expression levels are down-regulated by approximately two-fold at 28°C; 53 of these genes have unknown functions, and 189 function known genes are sorted into 19 gene families (Table 4), such as the secretion system, transportation, and ribosomal assembly and etc. These results indicated that the basic protein synthesis, metabolism rate and cell growth rate decrease at the low temperature of 28°C compared to 37°C. Strikingly, none of them is located in M18 GIs or prophages. Interestingly, we found that the 16 down-regulated genes are predicted to encode several mammalian virulence-related factors, including three genes predicated to encode HCN synthetase, three encode proteins involved in type II secretion system, eight genes involved in the type III secretion system, and two genes involved in LasA protease precursor and phospholipase C, respectively.
Transcriptome profile analysis based on the M18 genomic microarray at the two different temperatures implies that the M18 genome has developed its specific features for gene expression to meet the requirement for the competitive fitness and survival in rhizosphere niches except the specific genome structure features.
The strain M18 isolated from the sweet melon rhizosphere in 1996 in Shanghai suburb was named Pseudomonas sp. M18 because of its effective biocontrol ability and several phenotypic features that are similar to Pseudomonas spp. [1, 2, 11]. In this study, the complete genome of the strain M18 was sequenced and shown to share a common core genome that is comprised of more than 90% genome sequence homology to all other sequenced genomes of P. aeruginosa isolates (except taxonomic outlier PA7). Based on the comparative genomic and phylogenetic analysis of the M18 genome with other P. aeruginosa, the strain M18 can now be definitively designated as a new P. aeruginosa strain. The results indicated that the core genome diversity of various P. aeruginosa strains from either nosocomial or rhizosphere niche is quite small in comparison with that of various P. fluorescens strains .
However, the various P. aeruginosa strains originating from different environment niches differ in accessory genome regions, genome expression profiles, virulence activities and antibiotic resistances. The analysis of complete M18 genome sequence and temperature dependent transcriptional profiles indicated several important features distinguished the strain M18 from other nosocomial P. aeruginosa strains. Firstly, the M18 genome contains several specific accessory regions of genomic plasticity which differ significantly from all sequenced nosocomial isolates. This is the first time to sequence completely a rhizosphere originating P. aeruginosa strain M18, and take it as a model strain to analysis and detail the possible determinants related to its biocontrol activity and living ability in rhizosphere niche. We found that the M18 genome contains five specific GIs and two novel prophages, which benefit its survival and biocontrol activity in the rhizosphere niches. Furthermore, all mammalian pathogenicity-related GIs and prophages contained in other sequenced P. aeruginosa strains [14, 33–35] are absent in M18 genome. Secondly, fourteen mammalian virulence-related genes are absent or truncated in M18 genome, and the competitive index analysis of the strains M18 and LESB58 indicated the strain M18 is easier to be erased than that of LESB58 in the acute infection mouse lung model. Comparing with the strain LESB58, the strain M18 is more susceptible to several antibiotics and antimicrobial agents including kanamycin, gentamycin and ofloxacin. Thirdly, the M18 genome based microarray profiles revealed that the strain M18 has developed its specific temperature-dependent gene expression patterns to meet the requirements of survival and thriving in the rhizosphere niche. Especially, we found that none in the GIs expressed higher at 37°C than that at 28°C, but 24 genes in the GIs and prophages are up-regulated at 28°C which is the temperature close to that in rhizosphere niches.
In summary, the rhizosphere originated strain M18 has evolved specific genome structures and the temperature dependent expressional patterns to meet the requirement of fitness and survival under the selective pressure imposed on it. Therefore, the specific interesting features found in M18 genome and transcriptome profiles may provide us several cues to design therapeutic strategies against P. aeruginosa infection, and assist us to develop the strain M18 or other rhizosphere originating strains as a safety cell factory for the industrial purposes to produce antifungal compounds or other secondary metabolites through genetically engineered modification.
Bacterial strains, medium, and growth conditions
P. aeruginosa M18 was isolated from sweet melon rhizosphere in Shanghai suburb in 1996 and maintained in our laboratory. P. aeruginosa LESB58 which is a hyper-virulent strain isolated from CF patient , was obtained from Dr. Robert E.W. Hancock. The two strains were typically grown in Luria Bertani (LB) medium at 28°C and 37°C, respectively.
The draft P. aeruginosa M18 genome sequence was obtained using GS 20 system (454 Life Science Corporation). Overall, 547,645 high-quality reads were assembled and had an average length that covered 32.9-fold of the genome, and yielded 227 contigs with the 454 assembly tool. Among these, 172 large contigs (> 500 bp) represented 99.18% of the draft sequence. The complete sequence was assembled and finished according to the method described by Zhang , and the average error rate of the finished genome was lower than 1 bp in 10 kb.
Nucleotide sequence accession number
The annotated genome sequence has been deposited into GenBank under accession no. CP002496.
Genome annotation of the strain M18 was mainly performed as described previously . In brief, putative coding sequences (CDS) were predicted using Glimmer 3 , and the short CDS (< 90 bp) were eliminated. The putative protein function were predicted using BLASTP searches against all the annotated proteins of the P. aeruginosa strains, LESB58, PAO1, UCBPP-PA14 and PA7, and the nr (non-redundant protein sequences) database in NCBI based on the most significant matches. Clusters of orthologous groups (COGs) and subcellular localizations for each CDS were predicted using RPS-BLAST , PSORTb v.2.0  and LipoP1.0 . Transfer RNAs were predicted by TRNASCAN-SE . Ribosomal RNAs and small RNAs were predicted by BLASTN searches against those of P. aeruginosa strains, LESB58, PAO1, UCBPP-PA14 and PA7. ISs were identified by the IS Finder database http://www-is.biotoul.fr/, and CRISPRs were predicted by CRISPRFinder .
The P. aeruginosa M18 genome sequence was aligned to those of other sequenced P. aeruginosa genomes currently at NCBI with the use of MUMmer . The M18 strain-specific CDS were identified with mGenomeSubtractor  with 75% identities, or 70% length ratios of matching regions to questionable CDSs.
Nine completely sequenced P. aeruginosa genomes were selected to assess the phylogenetic status of M18, while P. fluorescens Pf-5 was used as a member of the "out" group. A total of 58 phylogenetically useful genes  which carried more than 10 parsimony informative sites, were selected to construct the phylogenetic tree. Then, the nucleotide sequences of 58 PUG genes in these strains were connected, aligned, and constructed to form a maximum parsimony tree using MEGA 4 software .
Genomic island identification
GIs and prophages located in the 3'-end of the annotated tRNA genes were firstly identified using MobilomeFINDER . GIs located in the other sites of M18 genome were identified by mGenomeSubtractor . The large islands that contained more than five CDS were kept. The direct repeat (DR) elements were identified using BLASTN searches. Prophages were identified by manual inspection of phage-related genes.
Fungal growth inhibition assay
Fungal growth inhibition by P. aeruginosa strain M18 was determined by measuring its ability to inhibit the growth of a plug of Mycosphaerella melonis mycelium on a potato dextrose agar plate (PDA). Briefly, a plug in 5-mm diameter from the leading edge of M. melonis culture grown for 5 days at 28°C on a PDA plate was taken and placed in the center of a fresh PDA plate. The strain M18 cultures grown overnight on a King's medium B (KMB) plate were placed at the position 25 mm away from the mycelium plug on PDA plate. The co-culture plates were incubated at 28°C and scored after 4 or 5 days by measuring the distance between the edges of the bacterial colony and the fungal mycelium plug . Each experiment was repeated twice with three replicates.
Competitive index analysis in mouse
The acute lung infection mouse model has been previously described . Five mice were used for each experiment. According to the guidelines of ethics committee for animal treatment, adult female BALB/c mice that weighted more than 17 g at the age of 6- to 8-weeks-old were used. The animals were anesthetized using 2.5% avertin by intraperitoneal administration. The bacteria were centrifuged and washed twice with PBS, and the final concentration was adjusted to approximately 1 × 108 CFU/ml. Equal ratio of the culture of the two strains M18 and LESB58 was mixed for inoculation. With the mice held in an upright position, 10 μl of the bacterial mixture was inoculated twice (20 μl in total) into mice via application into the nostril with a pipette. Mice were sacrificed at 24 h and 48 h post-inoculation and their lungs were used for CFU counts based on the significant colony size differences of the two strains on LB plates, as the colony diameter of M18 colony and LESB58 is 1-2 mm and 0.2-0.5 mm in size after overnight culture, respectively. The two different strain colonies were further verified by colony PCR using phzG2 interval region primer, and the PCR products of M18 and LESB58 colonies are 1075 bp and 841 bp in length, respectively. The pair of primer sequences is designed as phzG2Iup: ACTGACACTGAGGTGCGAAAGCG; phzG2Idown: ACCGTATGCGCTTCACTTGACC. The M18/LESB58 CIs in vivo were determined according to the method previously described .
Minimum inhibitory concentration (MIC) Determination
The MIC determination method was applied to measure the susceptibility of the two strains of P. aeruginosa M18 and LESB58 to a range of antimicrobial agents, as previously described . For each antimicrobial agent, a serial of two-fold diluted agent was generated using an appropriate starting concentration in Müller-hinton broth media. The MIC corresponds to the smallest agent concentration that inhibits the growth of P. aeruginosa M18 and LESB58.
Oligonucleotide microarray experiments
The P. aeruginosa M18-specific Agilent Oligonucleotide microarray (Gene Expression Omnibus (GEO) Platform: GPL11372) was designed with the eArray software. For each of 5,684 annotated CDS, a 60-mer oligonucleotide was designed and repeated twice on the microarray. The P. aeruginosa strain M18 was cultured overnight in LB medium with 180 rpm at 28°C and 37°C, respectively. The two cultures were inoculated by 100-fold diluted into LB medium for three biological replicates and grown to OD600 ~5.0-6.0. The total RNA was extracted using TRIzol (Invitrogen) and measured with a BioAnalyzer (Agilent Technologies, Palo Alto, CA, USA). Total RNA was eluted in RNase-free water and purified by QIAGEN RNeasy kit. A total of 2 μg RNA was reverse-transcribed to cDNA by a one-step method. The cDNA was transcribed into RNA by T7 RNA polymerase, modified by aa-UTP at 40°C, before labeling with Cy3 fluorescence dye at 25°C, purified by a QIAGEN RNeasy mini kit, and quantified by the BioAnalyzer. After hybridization, the arrays were washed according to the manufacturer instructions and scanned twice by an Agilent scanner at 100% and 10% photo multiplier tube (PMT). The normalized data was analyzed using an R script . Expression levels of the genes were compared using SBC analysis system (http://sas.ebioservice.com/). Average values of each gene were calculated based on the three biological replicates.
Oligonucleotide microarray data accession number
The microarray data has been deposited in to GEO under accession no. GSE26518.
Hu H, Xu Y, Chen F, Zhang X, Hur BK: Isolation and characterization of a new fluorescent Pseudomonas strain that produces both phenazine 1-carboxylic acid and pyoluteorin. J Microbiol Biotechnol. 2005, 15: 86-90.
Ge Y, Huang X, Wang S, Zhang X, Xu Y: Phenazine-1-carboxylic acid is negatively regulated and pyoluteorin positively regulated by gacA in Pseudomonas sp. M18. FEMS Microbiol Lett. 2004, 237: 41-47. 10.1111/j.1574-6968.2004.tb09676.x.
Ge Y, Pei D, Feng P, Huang X, Xu Y: Autoinduction of RpoS biosynthesis in the biocontrol strain Pseudomonas sp. M18. Curr Microbiol. 2007, 54: 97-101. 10.1007/s00284-006-0072-z.
Wang Y, Huang X, Hu H, Zhang X, Xu Y: QscR acts as an intermediate in gacA-dependent regulation of PCA biosynthesis in Pseudomonas sp. M18. Curr Microbiol. 2008, 56: 339-345. 10.1007/s00284-007-9087-3.
Yan A, Huang X, Liu H, Dong D, Zhang D, Zhang X, Xu Y: An rhl-like quorum-sensing system negatively regulates pyoluteorin production in Pseudomonas sp. M18. Microbiology. 2007, 153: 16-28. 10.1099/mic.0.29211-0.
Chen Y, Wang X, Huang X, Zhang X, Xu Y: Las-like quorum-sensing system negatively regulates both pyoluteorin and phenazine-1-carboxylic acid production in Pseudomonas sp. M18. Sci China C Life Sci. 2008, 51: 174-181. 10.1007/s11427-008-0026-8.
Zhang X, Wang S, Geng H, Ge Y, Huang X, Hu H, Xu Y: Differential regulation of rsmA gene on biosynthesis of pyoluteorin and phenazine-1-carboxylic acid in Pseudomonas sp. M18. World J Microbiol Biotechnol. 2005, 21: 883-889. 10.1007/s11274-004-6358-z.
Gurusiddaiah S, Weller DM, Sarkar A, Cook RJ: Characterization of an antibiotic produced by a strain of Pseudomonas fluorescens inhibitory to Gaeumannomyces graminis var. tritici and Pythium spp. Antimicrob Agents Chemother. 1986, 29: 488-495.
Howell CR, Stipanovic RD: Suppression of Pythium ultimum-induced damping-off of cotton seedlings by Pseudomonas fluorescens and its antibiotic, pyoluteorin. Phytopathology. 1980, 70: 712-715. 10.1094/Phyto-70-712.
Chin-A-Woeng TFC, Bloemberg GV, van der Bij AJ, van der Drift K, Schripsema J, Kroon B, Scheffer RJ, Keel C, Bakker P, Tichy HV, et al: Biocontrol by phenazine-1-carboxamide-producing Pseudomonas chlororaphis PCL1391 of tomato root rot caused by Fusarium oxysporum f. sp. radicis-lycopersici. Molecular Plant-Microbe Interactions. 1998, 11: 1069-1077. 10.1094/MPMI.19188.8.131.529.
Huang J, Xu Y, Zhang H, Li Y, Huang X, Ren B, Zhang X: Temperature-dependent expression of phzM and its regulatory genes lasI and ptsP in rhizosphere isolate Pseudomonas sp. strain M18. Appl Environ Microbiol. 2009, 75: 6568-6580. 10.1128/AEM.01148-09.
Mavrodi DV, Blankenfeldt W, Thomashow LS: Phenazine compounds in fluorescent Pseudomonas spp. biosynthesis and regulation. Annu Rev Phytopathol. 2006, 44: 417-445. 10.1146/annurev.phyto.44.013106.145710.
Gibson J, Sood A, Hogan DA: Pseudomonas aeruginosa-Candida albicans interactions: localization and fungal toxicity of a phenazine derivative. Appl Environ Microbiol. 2009, 75: 504-513. 10.1128/AEM.01037-08.
Winstanley C, Langille MG, Fothergill JL, Kukavica-Ibrulj I, Paradis-Bleau C, Sanschagrin F, Thomson NR, Winsor GL, Quail MA, Lennard N, et al: Newly introduced genomic prophage islands are critical determinants of in vivo competitiveness in the Liverpool Epidemic Strain of Pseudomonas aeruginosa. Genome Res. 2009, 19: 12-23.
Lu J, Huang X, Zhang M, Li S, Jiang H, Xu Y: The distinct quorum sensing hierarchy of las and rhl in Pseudomonas sp. M18. Curr Microbiol. 2009, 59: 621-627. 10.1007/s00284-009-9483-y.
Buysens S, Heungens K, Poppe J, Hofte M: Involvement of pyochelin and pyoverdin in suppression of Pythium-Induced damping-off of tomato by Pseudomonas aeruginosa 7NSK2. Appl Environ Microbiol. 1996, 62: 865-871.
Anjaiah V, Cornelis P, Koedam N: Effect of genotype and root colonization in biological control of fusarium wilts in pigeonpea and chickpea by Pseudomonas aeruginosa PNA1. Can J Microbiol. 2003, 49: 85-91. 10.1139/w03-011.
Bano N, Musarrat J: Characterization of a new Pseudomonas aeruginosa strain NJ-15 as a potential biocontrol agent. Curr Microbiol. 2003, 46: 324-328. 10.1007/s00284-002-3857-8.
Kumar RS, Ayyadurai N, Pandiaraja P, Reddy AV, Venkateswarlu Y, Prakash O, Sakthivel N: Characterization of antifungal metabolite produced by a new strain Pseudomonas aeruginosa PUPa3 that exhibits broad-spectrum antifungal activity and biofertilizing traits. J Appl Microbiol. 2005, 98: 145-154. 10.1111/j.1365-2672.2004.02435.x.
Li Y, Jiang H, Du X, Huang X, Zhang X, Xu Y: Enhancement of phenazine-1-carboxylic acid production using batch and fed-batch culture of gacA inactivated Pseudomonas sp. M18G. Bioresour Technol. 2010, 101: 3649-3656. 10.1016/j.biortech.2009.12.120.
Su J, Zhou Q, Zhang H, Li Y, Huang X, Xu Y: Medium optimization for phenazine-1-carboxylic acid production by a gacA qscR double mutant of Pseudomonas sp. M18 using response surface methodology. Bioresour Technol. 2010, 101: 4089-4095. 10.1016/j.biortech.2009.12.143.
Zhou Q, Su J, Jiang H, Huang X, Xu Y: Optimization of phenazine-1-carboxylic acid production by a gacA/qscR-inactivated Pseudomonas sp. M18GQ harboring pME6032Phz using response surface methodology. Appl Microbiol Biotechnol. 2010, 86: 1761-1773. 10.1007/s00253-010-2464-z.
Stover CK, Pham XQ, Erwin AL, Mizoguchi SD, Warrener P, Hickey MJ, Brinkman FS, Hufnagle WO, Kowalik DJ, Lagrou M, et al: Complete genome sequence of Pseudomonas aeruginosa PAO1, an opportunistic pathogen. Nature. 2000, 406: 959-964. 10.1038/35023079.
Lee DG, Urbach JM, Wu G, Liberati NT, Feinbaum RL, Miyata S, Diggins LT, He J, Saucier M, Deziel E, et al: Genomic analysis reveals that Pseudomonas aeruginosa virulence is combinatorial. Genome Biol. 2006, 7: R90-10.1186/gb-2006-7-10-r90.
Roy PH, Tetu SG, Larouche A, Elbourne L, Tremblay S, Ren Q, Dodson R, Harkins D, Shay R, Watkins K, et al: Complete genome sequence of the multiresistant taxonomic outlier Pseudomonas aeruginosa PA7. PLoS ONE. 2010, 5: e8842-10.1371/journal.pone.0008842.
Wolfgang MC, Kulasekara BR, Liang X, Boyd D, Wu K, Yang Q, Miyada CG, Lory S: Conservation of genome content and virulence determinants among clinical and environmental isolates of Pseudomonas aeruginosa. Proc Natl Acad Sci USA. 2003, 100: 8484-8489. 10.1073/pnas.0832438100.
Mathee K, Narasimhan G, Valdes C, Qiu X, Matewish JM, Koehrsen M, Rokas A, Yandava CN, Engels R, Zeng E, et al: Dynamics of Pseudomonas aeruginosa genome evolution. Proc Natl Acad Sci USA. 2008, 105: 3100-3105. 10.1073/pnas.0711982105.
Shao Y, He X, Harrison EM, Tai C, Ou HY, Rajakumar K, Deng Z: mGenomeSubtractor: a web-based tool for parallel in silico subtractive hybridization analysis of multiple bacterial genomes. Nucleic Acids Res. 2010, 38 (Suppl): W194-200.
Delcher AL, Phillippy A, Carlton J, Salzberg SL: Fast algorithms for large-scale genome alignment and comparison. Nucleic Acids Res. 2002, 30: 2478-2483. 10.1093/nar/30.11.2478.
Klockgether J, Munder A, Neugebauer J, Davenport CF, Stanke F, Larbig KD, Heeb S, Schock U, Pohl TM, Wiehlmann L, Tummler B: Genome diversity of Pseudomonas aeruginosa PAO1 laboratory strains. J Bacteriol. 2010, 192: 1113-1121. 10.1128/JB.01515-09.
Winsor GL, Van Rossum T, Lo R, Khaira B, Whiteside MD, Hancock RE, Brinkman FS: Pseudomonas Genome Database: facilitating user-friendly, comprehensive comparisons of microbial genomes. Nucleic Acids Res. 2009, 37: D483-488. 10.1093/nar/gkn861.
Raymond CK, Sims EH, Kas A, Spencer DH, Kutyavin TV, Ivey RG, Zhou Y, Kaul R, Clendenning JB, Olson MV: Genetic variation at the O-antigen biosynthetic locus in Pseudomonas aeruginosa. J Bacteriol. 2002, 184: 3614-3622. 10.1128/JB.184.13.3614-3622.2002.
He J, Baldini RL, Deziel E, Saucier M, Zhang Q, Liberati NT, Lee D, Urbach J, Goodman HM, Rahme LG: The broad host range pathogen Pseudomonas aeruginosa strain PA14 carries two pathogenicity islands harboring plant and animal virulence genes. Proc Natl Acad Sci USA. 2004, 101: 2530-2535. 10.1073/pnas.0304622101.
Harrison EM, Carter ME, Luck S, Ou HY, He X, Deng Z, O'Callaghan C, Kadioglu A, Rajakumar K: Pathogenicity islands PAPI-1 and PAPI-2 contribute individually and synergistically to the virulence of Pseudomonas aeruginosa strain PA14. Infect Immun. 2010, 78: 1437-1446. 10.1128/IAI.00621-09.
Battle SE, Meyer F, Rello J, Kung VL, Hauser AR: Hybrid pathogenicity island PAGI-5 contributes to the highly virulent phenotype of a Pseudomonas aeruginosa isolate in mammals. J Bacteriol. 2008, 190: 7130-7140. 10.1128/JB.00785-08.
Kobayashi I: Behavior of restriction-modification systems as selfish mobile elements and their impact on genome evolution. Nucleic Acids Res. 2001, 29: 3742-3756. 10.1093/nar/29.18.3742.
Huang X, Yan A, Zhang X, Xu Y: Identification and characterization of a putative ABC transporter PltHIJKN required for pyoluteorin production in Pseudomonas sp. M18. Gene. 2006, 376: 68-78. 10.1016/j.gene.2006.02.009.
Huang X, Zhu D, Ge Y, Hu H, Zhang X, Xu Y: Identification and characterization of pltZ, a gene involved in the repression of pyoluteorin biosynthesis in Pseudomonas sp. M18. FEMS Microbiol Lett. 2004, 232: 197-202. 10.1016/S0378-1097(04)00074-6.
Nowak-Thompson B, Chaney N, Wing JS, Gould SJ, Loper JE: Characterization of the pyoluteorin biosynthetic gene cluster of Pseudomonas fluorescens Pf-5. J Bacteriol. 1999, 181: 2166-2174.
Arora SK, Bangera M, Lory S, Ramphal R: A genomic island in Pseudomonas aeruginosa carries the determinants of flagellin glycosylation. Proc Natl Acad Sci USA. 2001, 98: 9342-9347. 10.1073/pnas.161249198.
Werck-Reichhart D, Feyereisen R: Cytochromes P450: a success story. Genome Biol. 2000, 1: REVIEWS3003-
Bhattacharya D, Nagpure A, Gupta RK: Bacterial chitinases: properties and potential. Crit Rev Biotechnol. 2007, 27: 21-28. 10.1080/07388550601168223.
Schindelin H, Jiang W, Inouye M, Heinemann U: Crystal structure of CspA, the major cold shock protein of Escherichia coli. Proc Natl Acad Sci USA. 1994, 91: 5119-5123. 10.1073/pnas.91.11.5119.
Monteiro-Vitorello CB, de Oliveira MC, Zerillo MM, Varani AM, Civerolo E, Van Sluys MA: Xylella and Xanthomonas Mobil'omics. OMICS. 2005, 9: 146-159. 10.1089/omi.2005.9.146.
Grissa I, Vergnaud G, Pourcel C: CRISPRFinder: a web tool to identify clustered regularly interspaced short palindromic repeats. Nucleic Acids Res. 2007, 35: W52-57. 10.1093/nar/gkm360.
Brouns SJ, Jore MM, Lundgren M, Westra ER, Slijkhuis RJ, Snijders AP, Dickman MJ, Makarova KS, Koonin EV, van der Oost J: Small CRISPR RNAs guide antiviral defense in prokaryotes. Science. 2008, 321: 960-964. 10.1126/science.1159689.
Marraffini LA, Sontheimer EJ: CRISPR interference limits horizontal gene transfer in staphylococci by targeting DNA. Science. 2008, 322: 1843-1845. 10.1126/science.1165771.
Briat JF: Iron assimilation and storage in prokaryotes. J Gen Microbiol. 1992, 138: 2475-2483.
Li Y, Du X, Lu ZJ, Wu D, Zhao Y, Ren B, Huang J, Huang X, Xu Y: Regulatory feedback loop of two phz gene clusters through 5'-untranslated regions in Pseudomonas sp. M18. PLoS ONE. 2011, 6: e19413-10.1371/journal.pone.0019413.
Miller SH, Mark GL, Franks A, O'Gara F: Pseudomonas-Plant Interactions. 2008, Wiley-VCH Verlag GmbH & Co. KGaA
Yang J, Chen L, Sun L, Yu J, Jin Q: VFDB 2008 release: an enhanced web-based resource for comparative pathogenomics. Nucleic Acids Res. 2008, 36: D539-542.
Meyer JM, Neely A, Stintzi A, Georges C, Holder IA: Pyoverdin is essential for virulence of Pseudomonas aeruginosa. Infect Immun. 1996, 64: 518-523.
Mougous JD, Gifford CA, Ramsdell TL, Mekalanos JJ: Threonine phosphorylation post-translationally regulates protein secretion in Pseudomonas aeruginosa. Nat Cell Biol. 2007, 9: 797-803. 10.1038/ncb1605.
Diggle SP, Stacey RE, Dodd C, Camara M, Williams P, Winzer K: The galactophilic lectin, LecA, contributes to biofilm development in Pseudomonas aeruginosa. Environ Microbiol. 2006, 8: 1095-1104. 10.1111/j.1462-2920.2006.001001.x.
Zavascki AP, Carvalhaes CG, Picao RC, Gales AC: Multidrug-resistant Pseudomonas aeruginosa and Acinetobacter baumannii: resistance mechanisms and implications for therapy. Expert Rev Anti Infect Ther. 2010, 8: 71-93. 10.1586/eri.09.108.
Jacoby GA: AmpC beta-lactamases. Clin Microbiol Rev. 2009, 22: 161-182, Table of Contents. 10.1128/CMR.00036-08.
Livermore DM, Woodford N: The beta-lactamase threat in Enterobacteriaceae, Pseudomonas and Acinetobacter. Trends Microbiol. 2006, 14: 413-420. 10.1016/j.tim.2006.07.008.
Li XZ, Nikaido H, Poole K: Role of mexA-mexB-oprM in antibiotic efflux in Pseudomonas aeruginosa. Antimicrob Agents Chemother. 1995, 39: 1948-1953.
Poole K, Gotoh N, Tsujimoto H, Zhao Q, Wada A, Yamasaki T, Neshat S, Yamagishi J, Li XZ, Nishino T: Overexpression of the mexC-mexD-oprJ efflux operon in nfxB-type multidrug-resistant strains of Pseudomonas aeruginosa. Mol Microbiol. 1996, 21: 713-724. 10.1046/j.1365-2958.1996.281397.x.
Kohler T, Michea-Hamzehpour M, Henze U, Gotoh N, Curty LK, Pechere JC: Characterization of MexE-MexF-OprN, a positively regulated multidrug efflux system of Pseudomonas aeruginosa. Mol Microbiol. 1997, 23: 345-354. 10.1046/j.1365-2958.1997.2281594.x.
Jalal S, Wretlind B: Mechanisms of quinolone resistance in clinical strains of Pseudomonas aeruginosa. Microb Drug Resist. 1998, 4: 257-261. 10.1089/mdr.1998.4.257.
Silby MW, Cerdeno-Tarraga AM, Vernikos GS, Giddens SR, Jackson RW, Preston GM, Zhang XX, Moon CD, Gehrig SM, Godfrey SA, et al: Genomic and genetic analyses of diversity and plant interactions of Pseudomonas fluorescens. Genome Biol. 2009, 10: R51-10.1186/gb-2009-10-5-r51.
Cheng K, Smyth RL, Govan JR, Doherty C, Winstanley C, Denning N, Heaf DP, van Saene H, Hart CA: Spread of beta-lactam-resistant Pseudomonas aeruginosa in a cystic fibrosis clinic. Lancet. 1996, 348: 639-642. 10.1016/S0140-6736(96)05169-0.
Zhang ZY, Liu C, Zhu YZ, Zhong Y, Zhu YQ, Zheng HJ, Zhao GP, Wang SY, Guo XK: Complete genome sequence of Lactobacillus plantarum JDM1. J Bacteriol. 2009, 191: 5020-5021. 10.1128/JB.00587-09.
Delcher AL, Harmon D, Kasif S, White O, Salzberg SL: Improved microbial gene identification with GLIMMER. Nucleic Acids Res. 1999, 27: 4636-4641. 10.1093/nar/27.23.4636.
Marchler-Bauer A, Panchenko AR, Shoemaker BA, Thiessen PA, Geer LY, Bryant SH: CDD: a database of conserved domain alignments with links to domain three-dimensional structure. Nucleic Acids Res. 2002, 30: 281-283. 10.1093/nar/30.1.281.
Gardy JL, Laird MR, Chen F, Rey S, Walsh CJ, Ester M, Brinkman FS: PSORTb v.2.0: expanded prediction of bacterial protein subcellular localization and insights gained from comparative proteome analysis. Bioinformatics. 2005, 21: 617-623. 10.1093/bioinformatics/bti057.
Juncker AS, Willenbrock H, Von Heijne G, Brunak S, Nielsen H, Krogh A: Prediction of lipoprotein signal peptides in Gram-negative bacteria. Protein Sci. 2003, 12: 1652-1662. 10.1110/ps.0303703.
Lowe TM, Eddy SR: tRNAscan-SE: a program for improved detection of transfer RNA genes in genomic sequence. Nucleic Acids Res. 1997, 25: 955-964. 10.1093/nar/25.5.955.
Tamura K, Dudley J, Nei M, Kumar S: MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol Biol Evol. 2007, 24: 1596-1599. 10.1093/molbev/msm092.
Ou HY, He X, Harrison EM, Kulasekara BR, Thani AB, Kadioglu A, Lory S, Hinton JC, Barer MR, Deng Z, Rajakumar K: MobilomeFINDER: web-based tools for in silico and experimental discovery of bacterial genomic islands. Nucleic Acids Res. 2007, 35: W97-W104. 10.1093/nar/gkm380.
Thomashow LS, Weller DM: Role of a phenazine antibiotic from Pseudomonas fluorescens in biological control of Gaeumannomyces graminis var. tritici. J Bacteriol. 1988, 170: 3499-3508.
Kukavica-Ibrulj I, Bragonzi A, Paroni M, Winstanley C, Sanschagrin F, O'Toole GA, Levesque RC: In vivo growth of Pseudomonas aeruginosa strains PAO1 and PA14 and the hypervirulent strain LESB58 in a rat model of chronic lung infection. J Bacteriol. 2008, 190: 2804-2813. 10.1128/JB.01572-07.
Ihaka R, Gentleman R: R: a language for data analysis and graphics. J Comput Graph Stat. 1996, 5: 299-314. 10.2307/1390807.
Ghai R, Hain T, Chakraborty T: GenomeViz: visualizing microbial genomes. BMC Bioinformatics. 2004, 5: 198-10.1186/1471-2105-5-198.
Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, Valentin F, Wallace IM, Wilm A, Lopez R, et al: Clustal W and Clustal × version 2.0. Bioinformatics. 2007, 23: 2947-2948. 10.1093/bioinformatics/btm404.
We are grateful to the sequencing team at Chinese National Human Genome Center at Shanghai for their assistance of sequencing and annotation; ShanghaiBio Corporation for assistance of microarray experiment and analysis. This work was funded by the National Key Basic Research Program (973 Program, No. 2009CB118906), the National High Technology Research and Development Program of China (863 Program, No. 2006AA10A209 and 2007AA02Z215), the Shanghai Science and Technology Program (No. 08391911900), and the Shanghai Leading Academic Discipline Project (No. B203).
YQX, DQW and HYO conceived and designed the experiments. DQW and YWH HYO were responsible for sequencing, finishing and annotating. DQW, YQX, YWH, JY and XW performed experiments and data analysis. YQX, HYO and XQH contributed reagents/materials/analysis tools. YQX, DQW, YWH and HYO wrote the paper. All authors read and approved the final manuscript.