2.1. Bacterial Strains and Media
A2-165 was purchased from DSMZ, Braunschweig, Germany (DSM 17677). It was routinely grown in brain–heart infusion (BHI) medium (Becton Dickinson, Le Pont-de-Claix, France) supplemented with 0.5% yeast extract (Becton Dickinson, Le Pont-de-Claix, France) and 5 mg/L hemin chloride (Calbiochem, VWR International, Libourne, France), with cellobiose (1 g/L; Sigma-Aldrich Co. LLC, St. Louis, MO, USA), maltose (1 g/L; Sigma-Aldrich Co. LLC, St. Louis, MO, USA), and cysteine (0.5 g/L; Sigma-Aldrich Co. LLC, St. Louis, MO, USA), under an atmosphere consisting of 80% N2
. Thirteen lactic acid bacteria/bifidobacteria strains were tested for their impact on growth of F. prausnitzii
). They were maintained in medium M17 lactose (Streptococcus
) or MRS plus 1 g/L l
) or MRS (Lactobacillus
), AES Laboratoire, Combourg, France.
To test the effect of supernatants on the growth of F. prausnitzii, all strains were cultured in YCFA medium optimized for the growth of all bacterial strains. This medium consisted of glucose (2 or 20 g/L, Sigma-Aldrich Co. LLC, St. Louis, MO, USA), tryptone (10 g/L, Becton Dickinson, Le Pont-de-Claix, France), yeast extract (5 g/L, Becton Dickinson, Le Pont-de-Claix, France), sodium acetate (5 g/L, Sigma-Aldrich Co. LLC, St. Louis, MO, USA), monohydrate lactose (5 g/L, Sigma-Aldrich Co. LLC, St. Louis, MO, USA), sodium bicarbonate (4 g/L, Sigma-Aldrich Co. LLC, St. Louis, MO, USA), cellobiose (2 g/L, Sigma-Aldrich Co. LLC, St. Louis, MO, USA), sodium chloride (0.9 g/L, Sigma-Aldrich Co. LLC, St. Louis, MO, USA), ammonium sulfate (0.9 g/L, Sigma-Aldrich Co. LLC, St. Louis, MO, USA), cysteine (0.5 g/L, Sigma-Aldrich Co. LLC, St. Louis, MO, USA), dibasic potassium phosphate (0.45 g/L, Sigma-Aldrich Co. LLC, St. Louis, MO, USA), magnesium sulfate (0.09 g/L, Sigma-Aldrich Co. LLC, St. Louis, MO, USA), calcium chloride (0.09 g/L, Sigma-Aldrich Co. LLC, St. Louis, MO, USA), hemin chloride (0.01 g/L Sigma-Aldrich Co. LLC, St. Louis, MO, USA), and resazurin sodium salt (0.001 g/L, Alfa Aesar, Kandel, Germany). Vitamins and volatile fatty-acid (except acetate) pools were removed to limit F. prausnitzii growth. We also increased the buffering of the medium, to prevent excessive pH variation that might not be tolerated by the commensal bacteria. We supplemented the medium with lactose (0.5%) to support the growth of the tested lactic acid bacteria and bifidobacterial strains. Final pH was 5.7 to mimic proximal gastrointestinal conditions. This modified medium is referred to as YCFAm.
2.4. RNA Isolation and Sequencing (RNA-seq)
We collected 15 mL of F. prausnitzii culture at the end of the exponential growth phase (T22), for each condition and replicate, and added it to 30 mL of RNA Protect bacterial reagent (Qiagen, Hilden, Germany) at pH 6. This mixture was immediately vortexed and incubated for 5 min at room temperature. It was then centrifuged at 5000× g for 10 min and the pellet was recovered and stored at −80 °C until use. RNA was isolated by enzymatic and mechanical lysis followed by use of the RNAeasy Mini Kit (Qiagen, Hilden, Germany). The RNA was concentrated during DNAse treatment with the RNeasy MiniElute Cleanup kit (Qiagen, Hilden, Germany). Quantity and quality of the RNA was assessed with a NanoDrop® photometer (Thermo Scientific, Waltham, MA, USA) and an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA) respectively. We used 1 µg of DNA-free RNA with a RIN value > 8.0 from each sample for ribosomal RNA removal and library construction with Ribo-zero and the Scriptseq v2 RNAseq Library preparation kit (Illumina, San Diego, CA, USA). Quality control was performed on libraries with the Agilent High-Sensitivity DNA Kit (Agilent Technologies, Palo Alto, CA, USA). The 15 total RNA libraries were sequenced in a single lane on a HiSeq 4000 (Illumina, San Diego, CA, USA), in the 50-cycle single-read configuration, according to the manufacturer’s protocol. We used the HiSeq SBS Kit v3 for sequencing, with HiSeq Control Software 3.3.20 and RTA v2.5.2. Reads in bcl format were demultiplexed with the 6 bp Illumina index and CASAVA 1.8, allowing a single base–pair mismatch per library, and were converted to fastq format with bcl2fastq.
2.6. Functional Annotation
Extended functional annotation was performed, comparing F. prausnitzii
A2-165 (GCF_000162015.1) ORFs with the sequences in various databases. The December 2014 update of the COG (Cluster of Orthologous Groups of proteins) database [37
] was used, together with BLAST(p) version 2.7.1 [38
] with a cutoff set at 1 × 10−6
for e-value and 40% for identity and coverage, to assign COG ID, functional classes and COG annotations. COG terms were validated through the CD-Search service from NCBI using RPS-Blast with 1 × 10−6
as e-value cutoff and the best hits were selected in terms of e-value. Proteins were also scanned for carbohydrate-related domains with the dbCAN2 [39
] web server. KEGG Orthology terms and related pathways were identified with the BlastKOALA server [40
]. GO terms for the three ontologies were obtained from the F. prausnitzii
A2-165 page at BioCyc (https://biocyc.org/
). As a result of the re-annotation, 1845 ORFs (67.26% of the total number of 2743 ORFs) were annotated with a known function based on the identification of at least one term from any ontology. F. prausnitzii
A2-165 genome assembly (GCF_000162015.1) was inspected for Lagaffe and Mushu phages [41
] with GenBank accession codes MG711461 and MG711460, respectively. Genome assembly proteins were compared to the 65 Lagaffe and 54 Mushu phage proteins using Blastp v2.7.1 with default parameters. Only hits with e-value lower than 1 × 10−5
, identity higher than 90% and coverage higher than 60% were considered. Hits were then manually curated according to synteny within the phage genome. As a result, 62 and 47 proteins were identified as Lagaffe and Mushu proteins, respectively.