Page 96 - Read Online

Page 4 of 19 Ambros et al. Microbiome Res Rep 2023;2:34 https://dx.doi.org/10.20517/mrr.2023.18

incubation at 37 °C for 1.5 h. Additionally, cell disruption via glass beads was carried out using the FastPrep ®
-24 (20 s, 4 m/s, 24/2) from MP Biomedicals.

Bacterial genome sequencing, assembly, and annotation
Extracted DNA was sequenced with the Illumina HiSeq technology by Eurofins (Germany). Unicycler
[29]
version 0.4.8 on usegalaxy.eu was used for contig assembly. Default parameters were used, and the
FASTA file contig length cut-off was set to 1,000. The NCBI Prokaryotic Genome Annotation Pipeline
(PGAP) [30-32] was used for the annotation of the bacterial genomes.

Viral DNA extraction and sequencing
For virion concentration, residual cell debris was removed from 10 mL virion-containing, post-induced
(UV light) lysate by centrifugation (6,000 × g, 5 min, 20 °C) and sterile filtration (induction and filtration as
described before). Virions were precipitated at 4 °C using 0.5 M NaCl and 10% (w/v) PEG8000 (final
concentrations) until the next day. After centrifugation (16,000 × g, 30 min, 4 °C), precipitated virions were
harvested by resuspending the phage-containing pellet in 500 µL SM buffer .
[33]

Prior to viral capsid digestion, bacterial gDNA and RNA were removed by adding 1.25 µL DNase I (Qiagen)
and 2.5 µL RNase A (10 mg/mL; Carl Roth) and incubating the samples at 37 °C for 1 h. Viral capsids were
digested at 60 °C for 1 h after the addition of 1.25 µL Proteinase K (20 mg/mL; Omega Bio-tek), 25 µL 10%
(w/v) sodium dodecyl sulfate (SDS; Serva) stock, and 20 µL 0.5 M EDTA (pH 8.0; VWR™). Samples were
briefly cooled down to ambient temperature (app. 20 °C), then DNA was extracted by phenol-chloroform
extraction and precipitated using ethanol (protocol after Center for Phage Technology ). Afterwards, DNA
[34]
was dissolved in 30 µL Elution Buffer (E.Z.N.A. Bacterial DNA Kit; Omega Bio-Tek).
®

Sequencing of viral genomes was performed using Eurofins (Germany) INVIEW Virus Sequencing for
dsDNA viruses. Assembly was performed as previously described for bacterial genomes, with the exception
of using a 100 bp contig cut-off for the generation of FASTA files.

Genome selection for PHASTER analysis
The genomes of 56 strains, including genomes at all assembly stages (“complete”, “genome”, “scaffold”,
“contig”), were downloaded from the NCBI website (Last download: 10th August 2022). The first analysis
with the JSpeciesWS Webtool (Ribocon GmbH; version 3.9.5) allowed the exclusion of genomes with
[35]
high similarity (99.00 ANIb over an alignment percentage cut-off of 95%), ensuring a diverse genome set for
subsequent analyses. The 16 strains excluded are listed in the supplemental material
[Supplementary Table 1]. The 38 genomes remaining in the analyses are listed in the “Availability of data
and materials” section. Furthermore, 7 strains from the in-house strain collection were also included in this
study, and are now provided as well at the NCBI website. A total of 45 genomes remained for further
analysis (“Availability of data and materials”), including the genomes provided within this study. Of note
are two strains: Strain TMW 1.706, which shares high sequence similarities (100.00% ANIb over 99.95% of
aligned nucleotides) with L. curvatus type strain DSM 20019 and strain TMW 1.1447, which shares 99.52%
T
ANIb over 95.39% of the aligned nucleotides with strain DRD-164. While we did not have access to the
L. curvatus type strain for the induction experiments, strain TMW 1.706 serves as a closely related
replacement. Strain DRD-164 was included as it seems to differ in its prophages, despite the overall high
genome similarities to strain TMW 1.1447. Notably, L. curvatus strain VRA_2sq_f was included within this
study, albeit its genome was found to have a low average nucleotide identity in relation to the genomes of all
other included L. curvatus strains (e.g., 88.24% ANIb over 66.84% of aligned nucleotides with strain DSM
20019 ). While this strain is listed as L. curvatus at the time of creating this study, and seems more closely
T
related to this species than to other species within the Latilactobacillus genus (data not shown), this might