The strain AUS0004 of E. faecium was used for this study [27]. It has VanB-type resistance to vancomycin and was isolated in 1998, at Austin Hospital (Melbourne, Australia) from the blood of a patient (NCBI – GenBank: CP003351.1). BHI (Brain Heart Infusion) broth was used to cultivate E. faecium at 37° C without agitation. Plate counts were performed on TS (tryptone soya) agar plates incubated 48 h at 37 °C and MH medium (Mueller-Hinton) was used for antibiograms and cultures with antibiotics.

100 µL of the overnight culture of E. faecium were washed with MH medium and used to inoculate 9.9 mL of MH broth and incubated at 37 °C until an OD600 of 0.3 was reached. At this time point, ciprofloxacin was added to obtain 20 µg/mL corresponding to 10-fold the minimum inhibitory concentration (MIC). Samples were taken at 24 and 48 h to determine the number of CFUs (Colony Forming Units) by plate counts after serial dilutions.

A bacterial suspension of density equivalent to a MacFarland of 0.5 was used to inoculate an MH agar plate using a sterile swab. An Etest® (Biomérieux, Marcy-l’Etoile, France) containing ciprofloxacin was loaded onto the agar and incubated overnight at 37 °C before MIC determination. In parallel, an antibiogram of Enterococci was performed by disk diffusion method using the following molecules (Bio-Rad, Hercules, CA, USA): ampicillin (2 µg), imipenem (10 µg), norfloxacin (10 µg), rifampicin (5 µg), erythromycin (10 µg), clindamycin (2 µg), quinuspristin-dalfopristin (15 µg), tigecycline (15 µg), linezolid (10 µg), levofloxacin (5 µg), gentamicin (30 µg), streptomycin (300 µg), vancomycin (5 µg), teicoplanin (30 µg), nitrofurantoin (100 µg), fosfomycin (200 µg). The interpretations were carried out according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines (https://www.eucast.org/clinical_breakpoints, 2022).

To perform whole genome sequencing by the Illumina technique, an overnight culture was prepared with wild type strain E. faecium AUS0004, as well as with a persister strain harvested after 48 h in presence of ciprofloxacin (10 x MIC). From these cultures, DNA extractions were carried out using “NucleoBond® AXG 100 Columns” and “NucleoBond® Buffer Set III” buffers, according to the supplier’s recommendations (Macherey-Nagel, Hoerdt, France). A mechanical lysis step was added by transferring cell pellets into screw-capped tubes containing 500 µg of glass beads and disrupted using the Fast Prep instrument (MP Biomedical LLC, Santa Ana, CA, USA) for 3 min at 6.5 m/s. At the end of the protocol, in order to precipitate the DNAs, the solutions were left overnight at -20 °C after addition of isopropanol. The tubes were then centrifuged at 10,000 rpm for 25 min at 4 °C. The DNA pellets were collected and then washed with 70% ethanol. DNAs were dried, resuspended in 500 µL of water and quantified with Nanodrop (Thermo Fisher Scientific, Hampton, NH, USA). Finally, libraries were prepared using the DNA prep kit and sequenced with the Illumina Nextseq 500 sequencer by the Platform for Shared Microbiology (P2M) of the Pasteur International Bioresources Network (PIBNet; Institut Pasteur, Paris, France). After quality controls using fastqc (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/), genomes were assembled using SPAdes v3.12 with recommended parameters for Illumina sequencing. We then compared the sequence of our original strain with that of the persistent strain in order to detect putative mutations using the snippy pipeline (https://github.com/tseemann/snippy).

Cell pellets were obtained from 20 mL of cultures harvested at OD600 of 0.3, after 48 h of stationary phase and from persisters (48 h in presence of ciprofloxacin). In order to avoid extracting proteins from dead cells, we first used the protocol published by Sulaiman [16]. Briefly, the cells from 48-hour cultures or peristers were resuspended with 100 µL of saline buffer. Then, 100 µL of a suspension of magnetic beads (100 µg, 50 nm in diameter) (MACSflex MicroBead Kit, Miltenyi Biotec, Paris, France) coupled to propidium iodide (50 µg/mL) (Sigma-Aldrich, Saint-Louis, Missouri, USA) were added. The mixture was incubated for 15 min at room temperature, in the dark and then placed in a magnetic rack (MagAttract® Magnetic Rack, Qiagen, Hilden, Germany). The propidium iodide, linked to the beads, will interact with the DNA of cells whose membrane was altered, including dead cells. These magnetic beads were retained by the magnet, and thus only intact cells can be recovered. This step was repeated three times to remove as many beads as possible. The tubes containing the living cells were centrifuged at 6000 rpm, 15 min at 4 °C, and the pellets were resuspended with 500 µL of the recovery buffer (Tris HCl 50 mM, Na2SO4 50 mM, glycerol 15%) and stored at -80 °C. For protein extractions, cells were disrupted by adding 500 µg of glass beads and using the Fast Prep instrument as previously described for DNA extractions. Protein quantifications were performed with the “Pierce BCA protein assay kit” (Thermo Fisher Scientific) according to the supplier’s recommendations.

For our global proteomic analysis by Mass Spectrometry (MS), 5 µg of each protein extract were digested with trypsin/Lys-C overnight at 37 °C. MS/MS spectra were obtained with a NanoElute ultra high-pressure Nano flow system (Bruker Daltonics, Billerica, MA, USA) as previously described [28]. For each condition, three independent experiments were carried out and sample tests were performed using Student’s t-test with a permutation-based FDR (False Discovery Rate) of 0.05.