Draft genome sequences and description of Lactobacillus rhamnosus strains L31, L34, and L35

Lactobacillus rhamnosus is a facultative, lactic acid bacterium in the phylum Firmicutes. Lactobacillus spp. are generally considered beneficial, and specific strains of L. rhamnosus are validated probiotics. We describe the draft genomes of three L. rhamnosus strains (L31, L34, and L35) isolated from the feces of Thai breastfed infants, which exhibit anti-inflammatory properties in vitro. The three genomes range between 2.8 – 2.9 Mb, and contain approximately 2,700 protein coding genes.


Introduction
Lactobacillus is the largest of three genera within the family Lactobacillaceae, and belongs to one of the dominant phyla, Firmicutes, in the human microbiome [1]. Lactobacillus spp. are naturally isolated from fermented foods [2], and are key members of the human microbiota, reviewed in [3]. In humans, they colonize the oral cavity, gastrointestinal and urogenital tracts, and breast milk [4]. As a whole, this genus is beneficial to humans, possesses many probiotic traits, and is rarely associated with disease.
The human-intestinal isolate, L. rhamnosus strain GG, is one of the most studied and applied probiot-ics. Research has shown that L. rhamnosus GG can modulate host immunity in vitro by decreasing inflammatory cytokine production from various eukaryotic cell lines [5,6], induces intestinal mucin gene expression subsequently inhibiting pathogen adherence in vitro [7]; and attenuates in vitro barrier dysfunction induced by inflammatory cytokines [8]. Here we present the draft genomes and classification summary of three potential probiotic L. rhamnosus strains L31, L34, and L35 isolated from the feces of Thai breastfed infants [9]. Genome sequencing and comparisons of L31, L34, and L35 with the species type-strain, L. rhamnosus GG should help researchers identify distinguishing genetic features important for specific probiotic traits.

Classification and features
Within the phylum Firmicutes, the family Lactobacillaceae contains three genera: Lactobacillus,Paralactobacillus, and Pediococcus; Lactobacillus being the largest with latest estimates ranging between 227-230 species (http://www.dsmz.de/bacterialdiversity/prokaryotic-nomenclature-up-todate/prokariotic-nomenclature-up-to-date.html) [10]. Members of Lactobacillus are gram-positive, non-motile, anaerobic, lactic-acid-producing bacilli that are divided into three fermentation groups: A) obligately homofermentative, B) facultatively heterofermentative, and C) obligately heterofermentative [4]. L. rhamnosus resides in fermentation group B and is distinct from the three major Lactobacillus phylogenetic groups based on 16S rRNA gene sequence (L. delbrueckii, L. reuteri, and L. salivarius groups) [4]. L. rhamnosus strains L31, L34, and L35 are phylogenetically similar to L. rhamnosus GG and maintain a distinctive 16S rRNA gene-based phyologeny from the three major Lactobacillus groups ( Figure 1). The basic characteristics of L. rhamnosus L31, L34, and L35 are summarized in Table 1. bulgaricus strain ATCC 11842, NC_008054, L. acidophilus strain NCFM, NC_006814. Full-length 16S rRNA gene sequences were aligned using ClustalW, and phylogenetic inferences were obtained using the maximum-likelihood method within the MEGA 5.2 software [11] with 1,000 bootstraps. B. subtilis strain 6051 HGW (NC_020507) was used as an outgroup. , not directly observed for the living, isolated sample, but based on a generally accepted property for the species, or anecdotal evidence). These evidence codes are from the Gene Ontology project [28].
The colony and Gram stain morphology of L. rhamnosus strains L31, L34, and L35 are each depicted in Figure 2. Supernatants from L. rhamnosus L34 and L35, both isolated from the same 40 day old female, suppress LPS-induced TNF-α production by THP-1 cells [9] and C. difficile-induced IL-8 production by HT-29 cells [29]. Similarly, strain L31, isolated from a 39 day old female, suppresses LPS-induced TNF-α production by THP-1 cells, however does not suppress C. difficile-induced IL-8 production by HT-29 cells [29]. All three strains are resistant to two drugs commonly used to treat C. difficile infection in humans, vancomycin and metronidazole (MIC90 >256µg/mL for each), but are susceptible to low concentrations (MIC90 = 2µg/mL) of the newest antibiotic targeting C. difficile, fidaxomicin. These strain-specific characteristics suggest L. rhamnosus L34 and L35 are potential probiotic candidates for either preventing or treating C. difficile disease.

Genome sequencing information
Genome project history L. rhamnosus strains L31, L34, and L35 were selected for sequencing based on the properties described above. The draft genome sequence for each strain was finished in October 2012. The Whole Genome Shotgun projects for L. rhamnosus L31, L34, and L35 have been deposited at DDBJ/EMBL/GenBank under the accession numbers AYTQ00000000, AYTR00000000, and AYTP0000000, respectively. The versions de-scribed in this paper are AYTQ01000000, AYTR01000000, and AYTP0100000, respectively. The genome projects for L31, L34, and L35 are listed in the Genome OnLine Database (GOLD) [30] as projects Gi0036900, Gi0036903, and Gi0036905, respectively. Genome sequencing and assembly was completed at Baylor College of Medicine's Human Genome Sequencing Center (BCM-HGSC). Automatic annotation was performed using the DOE-JGI Microbial Annotation Pipeline (DOE-JGI MAP). Table 2 shows the project information and its association with MIGS version 2.0 compliance [31], .

Genome annotation
Open Reading Frames (ORFs) were predicted using Prodigal [33,34] as part of the Oak Ridge National Laboratory genome annotation pipeline, followed by a round of manual curation using the JGI GenePRIMP pipeline [35]. The predicted protein coding sequences (CDSs) were translated and searched against the National Center for Biotechnology Information (NCBI) non-redundant database, UniProt, TIGRFam, Pfam, PRIAM, KEGG, COG, and InterPro databases [35]. These data sources were combined to assert a product description for each predicted protein. Additional gene prediction analysis and manual functional annotation was performed with the Integrated Microbial Genomes Expert Review (IMG-ER) platform [36]. Noncoding genes and miscellaneous features were predicted using tRNAscan-SE [37], RNAMMer [38], Rfam [39], TMHMM [40], and signalP [41].

Genome properties
The properties and statistics for the three L. rhamnosus genomes are summarized in Table 3. The distribution of genes into COG functional categories for each genome is detailed in

Comparison with Lactobacillus rhamnosus strain GG
The beneficial effects of human-intestinal derived L. rhamnosus GG have been studied for two decades [42][43][44][45] and its complete genome is available in NCBI [46]. We have compared the draft genome sequences of the potential probiotic L. rhamnosus strains L31, L34, and L35 to L. rhamnosus GG. The L. rhamnosus GG genome (3,010,111 bp, 46.69% GC content) is slightly larger than the new genomes presented here, and has approximately the same GC content (Table 3). In a recent compara-tive genomics study of 100 L. rhamnosus strains, Douillard, et al. [47] delineated seventeen variable chromosomal regions of L. rhamnosus strain GG (annotated in Figure 3), and the majority of these regions are absent or incomplete in the genomes of strains L31, L34, and L35 (Figure 3), notably the spaCBA pili gene cluster required for mucus adhesion [46]. The galactitol PTS region important for dulcitol utilization, a trait that typically belongs to L. rhamnosus isolates adapted to the intestinal tract [47], is conserved in L31, L34, and L35. Similar to L. rhamnosus GG, L31, L34, and L35 each contain genes annotated as L-lactate dehydrogenase (ldhL) and D-lactate dehydrogenase (ldhD) important for synthesizing L-lactate and D-lactate from pyruvate, respectively [49]. L. rhamnosus GG is unable to metabolize either maltose due to an inserted gene between the maltose-specific transport genes and hydrolase, or lactose because of a 38 bp N-terminal truncation in lacT and a disrupted lacG [47,50]. Strains L31, L34, and L35 all have an intact maltose locus and carry nonmutated copies of lacT and lacG (locations indicated on Figure 3), and therefore are predicted to utilize both maltose and lactose. a The total is based on either the size of the genome in base pairs or the total number of protein coding genes in the annotated genome  In line with the anti-inflammatory phenotypic differences already noted [9,29], differences in genomic features between L. rhamnosus L31 and the two isolates, L34 and L35, can also be made relative to strain GG. The taurine transport system deemed important for bile resistance as well as the fucU, fucI, fcsR, and α-L-fucosidase genes required for metabolizing fucosylated compounds present in gastrointestinal environments are found in L34 and L35 genomes, but not in L31. L. rhamnosus GG, despite belonging to a species known for rhamnose utilization, possesses an altered rhamnose locus and cannot utilize rhamnose [46]. L. rhamnosus L31 contains an intact rhamnose locus, while this locus in strains L34 and L35 looks similarly disrupted to that of strain GG. It is also noteworthy that L. rhamnosus L31 contains an iron-transport and a general secretion system not present in strains L34, L35, or GG.

Conclusion
Here we have presented the draft genomes of three potential probiotic strains of L. rhamnosus: L31, L34, and L35. Brief genome comparisons indicate that strains L34 and L35 are most similar to L. rhamnosus GG, while L31 contains marked differences suggesting it may have originated from a slightly different ecological niche [47]. L. rhamnosus L34 and L35 were isolated from the same host based on initial distinguishing colony morphology [9], however current colony morphology for these strains is not unique (Figure 2) and comparison of the draft genomes suggests the two genomes are nearly identical and similarly distinct from L31. It is possible that L34 and L35 may represent isolates of the same strain. Future studies will combine functional data with genomics, which is a powerful method for not only validating probiotic features of beneficial mi-crobes, but also for learning about the environmental adaptations that have favored their mutual relationship with human hosts.