Complete genome sequence of Lactobacillus rhamnosus Pen, a probiotic component of a medicine used in prevention of antibiotic-associated diarrhoea in children

Lactobacillus rhamnosus has been isolated from the human intestinal tract, oral cavity, and vagina. Owing to their beneficial effects on human health, many strains of L. rhamnosus are also used in the dairy and pharmaceutical industries. Examples of such industrially important probiotic strains are Lactobacillus rhamnosus GG and Lactobacillus rhamnosus R0011, as well as Lactobacillus rhamnosus Pen, which is a component of a medicine commonly used to reduce the risk of diarrhoea development during antibiotic therapy [1‐3]. Many characteristics of strain Pen have previously been reported, including carbohydrate utilisation, colony and cell morphology, antibiotic sensitivity, RAPD patterns, and SDS-PAGE and two-dimensional (2D) electrophoretic profiles of surface-associated proteins [4, 5]. Other properties, such as adhesion ability [6], survival rate in acidic pH [7], antiradical activity [8] and production of extracellular ferulic acid esterase [9] have also been analysed. Optimisation of medium composition to enhance growth of L. rhamnosus Pen using response surface methodology was reported by Polak-Berecka et al. [10].

Genomic DNA was isolated and purified using a Genomic Mini AX Bacteria + kit (A&A Biotechnology, Gdynia, Poland); DNA concentration was determined using a NanoDrop spectrophotometer (Thermo Scientific, Waltham, USA). Sequencing was performed at Genomed SA. Briefly, a paired-end library was constructed by using the NEB-Next^® DNA Library Prep Master Mix Set for Illumina (NEB, Ipswich, USA) and subsequently sequenced on an Illumina MiSeq with 2 × 250 paired end sequencing chemistry (Illumina, San Diego, USA). Additionally, a 5–8 kb mate-pair library was constructed according protocol developed in BGI (Shenzhen, China) and sequenced on a HiSeq 4000 with 2 × 100 paired end sequencing chemistry (Illumina, San Diego, USA). A total of 1,270,358,608 bases and 362,759,422 paired reads were yielded. Read trimming and filtering was performed using Cutadapt 1.9.1 [11]. De novo assembly was conducted using SPAdes 3.1.1. [12], which yielded one major contig with 679-fold average coverage. Functional annotation of predicted genes was performed using the NCBI Prokaryotic Genome Annotation Pipeline [13]. The clusters of orthologous groups (COGs) of proteins were determined using eggNOG 4.5 [14]. Ribosomal RNA genes were detected using RNAmer 1.2 [15] and tRNA genes were identified using tRNAscan-SE v. 2.0 [16]. Sequences of proteins which may determine putative probiotic properties of L. rhamnosus Pen were individually search against Conserved Domains Database (NCBI) [17] and InterPro detabase (EMBL-EBI) [18]. Genes potentially involved in the biosynthesis of bacteriocins were identified using BAGEL [19]. The presence of antibiotic resistance genes was tested using ResFinder [20]. Phaster was used to search for prophage sequences [21] and the presence of a CRISPR/Cas system was predicted using CRISPRs finder [22] and the Crispr Recognition Tool [23]. Genome mapping and alignment visualisation were performed using CGView [24] and BRIG [25] respectively.

The complete genome of L. rhamnosus Pen consists of a 2,884,966-nt circular chromosome (GC content of 46.8%) with no plasmid. Among the 2907 identified open reading frames, 2729 contain protein-coding genes. In addition, 59 tRNA genes, 5 rRNA operons, and 101 pseudogenes were identified (Table 1, Additional file 3: Figure S3). Of the identified coding sequences, 2422 (88.7%) were grouped into 20 COG classes. Coding sequences were identified as being involved in carbohydrate transport and metabolism (12%), transcription (7.3%), amino acid transport and metabolism (6.9%), translation, ribosomal structure and biogenesis (5.4%), and replication, recombination and repair of nucleic acids (4.8%) (Table 2, Additional file 3: Figure S3). Comparison of the L. rhamnosus Pen genome with eleven other L. rhamnosus complete genome sequences showed the highest similarity with intestinal isolate L. rhamnosus LOCK900 (symmetric identity 98.76%, gapped identity 99.97; CP005484.1) [26] and substantially lower sequence similarity with the industrially important L. rhamnosus GG (symmetric identity 84.24%, gapped identity 97.50%; AP011548.1) [27] (Fig. 1).

Comparative genomic analysis of L. rhamnosus Pen showed the presence of numerous genes which may determine its putative probiotic properties, supporting use of the strain in prevention of various gastrointestinal disorders. Genetic factors involved in cell surface adherence, biofilm formation, and pathogen inhibition were identified (Additional file 4: Table S1). Such features are known to provide a survival advantage for probiotic strains and are important for effective bacterial colonisation of the human intestine [1, 28‐32]. Additionally, detailed analysis of the genome did not reveal transmissible antibiotic resistance genes in the chromosome of L. rhamnosus Pen. It was previously described that such genetic determinants may constitute a reservoir of antibiotic resistance for food and gut pathogens. On the other hand, presence of intrinsic antibiotic resistance among probiotic strains is valuable factor in restoring the intestinal microbiota after antibiotic treatment [33].

The analysis performed using CRISPRs finder and the Crispr Recognition Tool indicated that the genome contains one regularly interspaced short palindromic repeat locus consisting of a 1092-nt region with 16 spacers (30–31 nt in length) (Fig. 2). The detected CRISPR–Cas system is of type II-A/LsaI1 (four cas genes; cas1, cas2, cas9, csn2, and one CRISPR array), similar to previously described CRISPR loci characteristic of L. rhamnosus strains [34]. BLASTN searches comparing all 16 spacers against the phage and plasmid NCBI databases revealed no sequence identity with known mobile genetic elements of lactobacilli. In a previous report, Douillard et al. [29] observed that many spacer sequences of L. rhamnosus strains fully or partially matched sequenced bacteriophage genomes, such as Lactobacillus rhamnosus phage Lc-Nu and Lrm1, as well as L. casei phages, including φAT3, A2, and PL-1. This phenomenon suggests that CRISPR modules may play an important role in protection against different mobile elements and also provide specific bacteriophage resistance [35]. Interestingly, similar results were not obtained for the CRISPR locus identified for Lactobacillus rhamnosus Pen.

Finally, one intact prophage of ~ 40.7 kb with a GC content of 44.8% was identified. This prophage sequence showed only 94% (query coverage 59%) and 91% (query coverage 21%) similarity with two previously described L. rhamnosus bacteriophages, Lrm1 (EU246945.1) and Lc-Nu (AY131267.2), respectively [36, 37]. However, nearly identical prophage sequences were detected in the genomes of L. rhamnosus CLS17 (NZ_JYCS01000023.1), L. rhamnosus B1 (NZ_NXEU01000011.1), and L. rhamnosus ASCC 3029 (NZ_MLJZ01000021.1). In our previous study, we described the release of phage particles by L. rhamnosus Pen [38]. Although the physiological role of continuous phage particle release in Lactobacillus is not evident, it may be beneficial for the bacterial host. It was previously suggested that such behaviour may enhance biofilm formation and promote horizontal gene transfer. On the other hand, by facilitating binding to human platelets, spontaneous prophage induction may also play an important role in bacterial virulence [39, 40]. Additionally, considering that such bacteriophages may be simultaneously released to the culture medium and that this phenomenon does not lead to complete lysis of the culture, microorganisms containing such phages may have high potential for application as safe food-grade vectors for presenting or producing various biological factors such as antigens, receptors, or virulence proteins [38, 41].