Whole-genome sequence assembly of Pediococcus pentosaceus LI05 (CGMCC 7049) from the human gastrointestinal tract and comparative analysis with representative sequences from three food-borne strains

Growth was investigated under different temperature, pH and NaCl conditions. Cell morphologies, motilities and sporulation activities were examined using transmission electron (H-600, Hitachi Ltd., Tokyo, Japan) microscopy. Phenotypic identification was achieved with API CH50 strips and the API CHL medium system according to the manufacturers instructions (BioMrieux SA, Marcy-lEtoile, France). Other physiological and biochemical tests were conducted as described previously [11]. Phylogenetic analysis was conducted using the neighbor-joining method based on the 16S rRNA and housekeeping gene sequences [12].

The genome of P. pentosaceus LI05 was sequenced with the next-generation sequencing platform Illumina HiSeq 2000, and the total number of reads based on a 500-bp library database were 2×11,079,017 (bp). The quality of the sequencing read data was estimated by calculating the quality and GC content of each read. The draft genome sequence was assembled using SOAPdenovo2 [13], and iterative optimization was used to obtain the optimal k-mer value through the use of 31-85 k-mers. The 500-bp libraries were used to build scaffolds, and the SOAPdenovo gap closer software was also used (http://​soap.​genomics.​org.​cn/​soapdenovo.​html). To close the remaining gaps, reference-guided assemblies were carried out with the CLC Genomics Workbench v. 6.05 (CLC bio, Aarbus, Denmark). The combination of de novo assembly and reference-guided assembly was performed manually using the microbial genome-finishing module in the CLC genomics workbench (CLC bio, Aarbus, Denmark). The complete genome sequence of P. pentosaceus ATCC 25745 was used as the reference genome.

P. pentosaceus LI05 genes were identified using Glimmer [14] together with comparative gene prediction by the direct mapping of the ORFs of the P. pentosaceus ATCC reference strain from the NCBI Genome Database. After a round of manual curation, the unannotated predicted coding sequences (CDS) were translated into amino acid sequences for a query using the NCBI non-redundant database as well as the UniProt, Pfam, COG, and InterPro databases to identify the closest existing homology annotations. Transfer RNA (tRNA) genes were detected using tRNAScanSE [15]. Ribosomal RNAs (rRNAs) were identified using a BLASTn [16] search against the ribosomal RNA databases. Signal peptides were predicted using SignalP 4.0 [17], whereas transmembrane helices in proteins were predicted using TMHMM [18]. The Integrated Microbial Genomes (IMG) platform (http://​img.​jgi.​doe.​gov/​) was used to support additional gene prediction analyses and manual functional annotations [19].

A comparative genomic analysis using BRIG [20] was conducted comparing P. pentosaceus LI05 from the human gastrointestinal tract with three food-borne strains with available genomic sequences, including P. pentosaceus ATCC 25745, SL4 and IE-3. The P. pentosaceu s LI05 genome sequences sharing low identities (<50%) with the other strains were designated as the P. pentosaceus LI05-unique regions. The proteins encoded by the genes that only existed in P. pentosaceus LI05 or that possessed sequence similarities of less than 50% with the three food-borne strains were further analyzed by BLASTp.

P. pentosaceus LI05 is a Gram-negative, non-motile, acid-tolerant, non-sporulating, spherical, facultative anaerobe from the human gastrointestinal tract (Additional file 1: Figure S1). It tolerates 6% NaCl in MRS broth. Growth occurs at 15-45°C and at pH 4-8 but optimally at 37°C. The colonies on the MRS agar were white, smooth, shiny, and circular with complete edges. Some carbohydrates, such as L-arabinose, D-ribose, D-xylose, D-galactose, D-glucose, D-fructose, D-mannose, N-acetylglucosamine, amygdalin, arbutin, salicin, D-cellobiose, D-maltose, D-trehalose, gentiobiose, and D-fucose, can be used as the sole carbon sources, whereas glycerol, erythritol, etc. cannot (Additional file 2: Table S1).

A neighbor-joining tree (Figure 1A) based on the 16S rRNA gene sequence of the strain LI05 shows the phylogenetic relationships between the species of the genus Pediococcus. This organism formed a distinct branch with P. pentosaceus, which was separate from those formed by other members of the genus Pediococcus. Sequence analyses of the dnaA, dnaJ, dnaK, pheS, pryH, recA, recH, tuF, gryB and rplB housekeeping genes were carried out for the definitive identifications of P. pentosaceus LI05, P. pentosaceus ATCC 25745, P. pentosaceus SL4 and P. pentosaceus IE-3. As shown in Figure 1B, the combination of the above housekeeping genes provided good phylogenetic resolution of the four strains. The P. pentosaceus strain IE-3 was the closest evolutionary relative of strain LI05.

The genome of P. pentosaceus LI05 was sequenced by the Illumina method (see Methods). A total of 11.05 million 100-bp paired-end reads were generated, which provided over 500-fold coverage of the reference genome. High-quality reads with Q?>?30 were assembled using de novo methods to obtain a draft genome of 1.75 Mbp with 8 contigs (the N50 of the assembled contigs was 34.3 Kb; the max length was 318 Kb). The G?+?C content of P. pentosaceus LI05 was 37.29%. For the main chromosome, 1,638 genes were predicted, 1,555 of which were protein-coding genes. A total of 1,321 protein-coding genes were assigned to putative functions, and the remainder were classified as hypothetical proteins. This genome contained 50 tRNAs and a complete 5S-23S-16S rRNA gene family. The properties and statistics of the genome are shown in Table 1 and Figure 2. As shown in Figure 3, the genome sequence of P. pentosaceus LI05 was highly conserved compared with those of P. pentosaceus ATCC 25745, P. pentosaceus SL4 and P. pentosaceus IE-3.

In a previous study, we have observed that P. pentosaceus LI05 is resistant to gastric acidity and bile compounds [7]. This was supported by the genomic data from this study, in which a gene encoding cholylglycine hydrolase, which is related to bile salt resistance, and genes encoding F0F1 ATP synthases, which are important for acid tolerance, were detected (Table 2). Additionally, six universal stress proteins (UspA), the chaperone protein DnaJ, the cofactor GrpE, which participates in the hyperosmotic and heat shock responses, the chaperone GroEL, which protects against environmental stresses, an enterocin A immunity family protein, an arsenate reductase, and methionine sulfoxide reductase A, which protects against oxidative stresses, were annotated. These representative stress resistance genes were highly conserved between P. pentosaceus LI05 and P. pentosaceus ATCC 25745, P. pentosaceus IE-3 or P. pentosaceus sL4, but most of them showed significant divergences from other species.

The ability to adhere to gastrointestinal mucosa is an important property of most probiotics [21],[22]. Several proteins encoded by P. pentosaceus LI05 genes had predicted adhesive potentials (Table 3). For example, sortase attaches surface proteins, including enzymes, pilins and adhesion-mediating large surface glycoproteins, to cell walls. Other proteins included a pilin-like competence protein ComGC, elongation factor Tu (EF-Tu), an enolase capable of binding to host extracellular fibronectin and the pilus biosynthesis protein HicB. Abundant adhesion proteins encoded by the genomic regions were consistent with the strong adhesion properties of P. pentosaceus LI05. However, these proteins have also been predicted in the other tested strains. These findings may represent a possible reason for the extensive colonization of P. pentosaceus in the gut. However, the examinations of many more genes or proteins may be required to evaluate the adhesive abilities of probiotics.

The P. pentosaceus LI05 genes also encoded three antimicrobials, which is consistent with the excellent antimicrobial ability of this strain. As shown in Table 3, genes encoding prebacteriocin were annotated in the genomes of both P. pentosaceus LI05 and P. pentosaceus ATCC 25745. Alternatively, the pedA gene (PCPN_1274) encoding pediocin PA-1 was detected in P. pentosaceus IE-3, but it was not identical to the prebacteriocin gene of P. pentosaceus LI05. Furthermore, genes encoding colicin V, which is a peptide antibiotic that kills sensitive cells by disrupting their membrane potentials [23], were found in these four P. pentosaceus strains. However, the colicin V discovered in strain L105 was different from that of the other spicies. Additionally, genes encoding lysin were detected in P. pentosaceus LI05 and P. pentosaceus ATCC 25745. As an antimicrobial agent, lysin is potentially immunogenic [24]. Therefore, P. pentosaceus LI05 can achieve “competitive exclusion” not only by limiting the surface area available but also by secreting antimicrobial substances.

In the genome of P. pentosaceus LI05, we also detected potentially beneficial properties that were not experimentally confirmed. This strain contained genes involved in the biosynthesis of not only important vitamins, such as riboflavin, folate, thiamine and biotin but also of functional factors, such as gamma-aminobutyrate (Table 3) [25]. In Gram-positive bacteria, peptidoglycan is one of the most important host immune regulators [26]. Although the genes and coding proteins related to the peptidoglycan pathway were conserved in the four strains of P. pentosaceus, they were not significantly similar to those of the other species. These findings will contribute to the elucidation of the mechanisms of immune regulation in P. pentosaceus LI05.

Fifty-three proteins encoded by P. pentosaceus LI05 genes were not detected or had sequence similarities of less than 50% in the comparative analysis with the three known food-borne strains, P. pentosaceus ATCC 25745, SL4 and IE-3. Among these proteins, 21 hypothetical proteins with no clear functions were not further analyzed; the other 32 proteins are listed in Table 4, demonstrated in Figure 3, and further discussed below.

Five putative exopolysaccharide biosynthesis proteins were detected only in P. pentosaceus LI05, including an epimerase, a capsular polysaccharide biosynthesis protein, two glycosyltransferases (key enzymes for the biosyntheses of the exopolysaccharide repeating units) and a polysaccharide biosynthesis protein. Four of these enzymes need to be examined in further detail because they are not only potentially novel but also probably induce variations in the structures of their encoded polysaccharides that may have influenced adherence, biofilm formation and the nature of the immune response [27].

P. pentosaceus LI05 was characterized by three extra-environmental stress tolerance proteins, including a putative ferritin-like DNA-binding protein, which maintains a steady state of iron ions and responds to stresses, such as those involving temperature, humidity, and ionizing and redox processes [28], a putative PadR family transcriptional regulator, which functions against phenolic acid stress, and a putative ThiJ/PfpI family protein, which is involved in cellular protection against environmental stresses [29].

Fourteen proteins related to the intrusion of exogenous DNA were identified in P. pentosaceus LI05. One group was comprised of twelve prophage-related proteins, including a phage integrase family site-specific recombinase, two integrases, a putative prophage repressor, two phage proteins, a replisome organizer, a terminase, two minor capsid proteins, a capsid protein and a tail protein. It is not rare for bacteria to contain multiple prophages in their chromosomes, which then constitute a sizable proportion of their total chromosomal material [30]. Pathogenic, commensal, and symbiotic bacteria have been observed to play roles in a variety of bacterial adaptations in hosts [31]. Phage-related proteins were encoded by genes in each of the three food-borne strains. The other genes detected in the P. pentosaceus LI05 included two encoding bacterial DNA type I restriction endonucleases, which are involved in prokaryotic DNA restriction-modification mechanisms that protect the bacteria against invading foreign DNA [32].

Two putative doxorubicin-daunorubicin resistance proteins existed in P. pentosaceus LI05. One ORF encoded DrrA, which is part of the ABC transporter complex DrrAB. The other ORF encoded DrrC, which is part of the ABC transporter permease protein. This finding partially reflects the complex interactions between drugs and gut-associated microbes [33]. Both daunorubicin and doxorubicin are antitumor drugs and are thus not suitable for antibacterial applications. Therefore, these two genes will not affect the control of P. pentosaceus LI05.