Dataset for transcriptome analysis of Salmonella enterica subsp. enterica serovar Typhimurium strain 14028S response to starvation

Salmonella enterica is an ubiquitous pathogen throughout the world causing gastroenteritis in humans and animals. Survival of pathogenic bacteria in the external environment may be associated with the ability to overcome the stress caused by starvation. The bacterial response to starvation is well understood in laboratory cultures with a sufficiently high cell density. However, bacterial populations often have a small size when facing this challenge in natural biotopes. The aim of this work was to find out if there are differences in the transcriptomes of S. enterica depending on the factor of cell density during starvation. Here we present transcriptome data of Salmonella enterica subsp. enterica serovar Typhimurium str. 14028S grown in carbon rich or carbon deficient medium with high or low cell density. These data will help identify genes involved in adaptation of low-density bacterial populations to starvation conditions.


a b s t r a c t
Salmonella enterica is an ubiquitous pathogen throughout the world causing gastroenteritis in humans and animals. Survival of pathogenic bacteria in the external environment may be associated with the ability to overcome the stress caused by starvation. The bacterial response to starvation is well understood in laboratory cultures with a sufficiently high cell density. However, bacterial populations often have a small size when facing this challenge in natural biotopes. The aim of this work was to find out if there are differences in the transcriptomes of S. enterica depending on the factor of cell density during starvation. Here we present transcriptome data of Salmonella enterica subsp . enterica serovar Typhimurium str. 14028S grown in carbon rich or carbon deficient medium with high or low cell density. These data will help identify genes involved in adaptation of low-density bacterial populations to starvation conditions. © 2020 The Author(s Value of the data 1. For the first time, transcriptome data were obtained for salmonella starved for carbon and phosphorus at high or low cell density. 2. The transcriptome dataset can be used to identify S. enterica genes differentially expressed under starvation conditions at high or low bacterial population density. 3. These data can help elucidate mechanisms of persistence used by the pathogenic microorganisms for survival and growth in the oligotrophic biotopes.

Data description
The dataset of this article provides information on raw RNA-seq reads obtained from samples of Salmonella enterica serovar Typhimurium 14028S cultures grown in a mineral medium providing carbon and phosphorus starvation, or in the medium supplemented with glucose as a carbon source. The final content of the basic elements in the mineral medium (excluding carbon and nitrogen) is contained in Table 1 . Information on the sampling time and growth rate of the   cultures is presented in Table 2 . This table also provides the NCBI SRA accession numbers of the cleaned FASTQ files for all biological replicates. The transcriptome data obtained are summarized in Table 3 . To identify coding and non-coding transcripts, they were mapped on the reference genome. Up-and down-regulated genes were counted with the Differentially Expressed Genes (DEGs) analysis of transcriptomes of the salmonella cultures starved at high or low cell density, as well as the cultures grown in the glucose rich medium ( Fig. 1 ). Besides, we evaluated the identity of 10 0 0 the most variable genes associated with salmonella transcriptome responses to starvation with the heat map analysis ( Fig. 2 ).

Strains and growth conditions
In this work Salmonella enterica subsp. enterica serovar Typhimurium (ATCC 14,028) strain was used. This strain was cultured on a Luria-Bertani (LB) agar plate (Sigma Aldrich) and a single colony was used to inoculate 10 mL of LB broth. After 12 h aerobic incubation at 37 °C the salmonella cells were collected by centrifugation, washed twice, and transferred to  the mineral carbon and phosphorus deficient AB medium containing 1.0 g/L NH 4 Cl; 0.62 g/L MgSO 4 × 7H 2 O; 0.15 g/L KCl; 0.013 g/L CaCl 2 × 2H 2 O; 0.005 g/L FeSO 4 × 7H 2 O, pH -5.5. Previously, this medium has been used to study the starving cultures of a phytopathogenic bacterium Pectobacterium atrosepticum [ 1 , 2 ]. Similar nutrient limited conditions may be found in some natural water bodies and water courses. The elemental composition of the prepared AB medium was checked using an inductively coupled plasma mass spectrometer Aurora M90 (Bruker Corporation, USA). Only a small amount of phosphorus and sodium as additional components was recorded ( Table 1 ). The cell suspensions were incubated under starvation conditions with initial population density of 10 3 or 10 9 CFU per ml in glass vials without aeration at 28 °C for 4-96 h before sampling ( Table 2 ). The low-density cultures were also incubated in AB medium supplemented with 1% glucose.

Experiment design
In order to study the dynamics of transcriptome response during salmonella adaptation to starvation, sampling was conducted at two time points corresponded to the periods of maximum growth and transition to the stationary phase, respectively.
In our experiments the number of CFU increased after inoculation in all salmonella cultures. In starving high-density cultures, an eight-fold increase in CFU number observed over four hours was replaced by a gradual decrease, which stabilized by the fourth day of incubation at a final value of about 0.06% of the inoculation titer. Such dynamics of the salmonella culture is in a good agreement with the "altruistic" model of the bacterial stress response associated with programmed death [3] . Based on this dynamics of CFU, the starving high-density cultures were sampled after 4 h and 4 days of incubation ("High_4h" and "High_4d", Table 2 ).
Low-density cultures entered the stationary phase just after the period of exponential growth was over. The control low-density cultures in carbon-supplemented AB medium ("Glucose_" in Table 2 ) were sampled after 6 and 24 h of incubation that corresponded to periods of the maximum growth and the transition to the stationary phase, respectively. However, the growth of experimental low-density cultures incubated in the carbon-deficient medium was significantly impaired. Thus, sampling of these cultures delayed up to 24 h and 3 days ("Low_24h" and "Low_3d", Table 2 ).
Total RNA was isolated and cDNA libraries were prepared for RNA-sequencing. Directional libraries were sequenced on Illumina Hiseq 2500 in single reads. The RNA-seq raw reads were stored in FASTQ files, and further analyzed to get the clean reads.

Library construction and sequencing
Bacterial cells were harvested by filtration through nitrocellulose membranes with diameter of pores 0.22 μm (Millipore, USA). The membranes were placed into tubes containing 1 ml of Extract reagent (Evrogen, Russia) and heated at 55 °C 10 min. Then total RNA was extracted according to the manufacturer's protocol. DNA contaminants were removed using RNase-free DNase I kit (Ambion, USA). RNA integrity was checked with an Agilent 2100 bioanalyzer (Agilent,USA). rRNA depletion was performed using Ribo-Zero rRNA Removal Kit for Gram-Negative Bacteria (Illumina, USA). NEBNext Ultra Directional RNA Library Prep Kit for Illumina was used to prepare RNA-seq libraries. The resulting average size of the cDNA libraries was approximately 300 bp. The libraries were sequenced in a single lane of a flow cell on the HiSeq 2500 (Illumina) platform.

Sequence QC and filtering
A total 173,505,388 raw single-end reads 60 bp long were obtained ( Table 3 ). Quality of the raw reads was controlled with FastQC software v. 0.11.3 [4] . Reads belonged to rRNA were removed using SortMeRNA [5] . Trimming of reads and adapters removal were performed with Trimmomatic v.0.35 [6] .

Reads alignment to the reference genome and data analysis
Filtered high-quality reads were mapped onto the reference genome sequence of Salmonella enterica subsp. enterica serovar Typhimurium 14028S assembly GCA_0 0 0 022165.1 [7] with TopHat 2 [8] . Before mapping the reference genome was indexed with Bowtie2 [9] . Coverage estimates and statistics of the reads mapping are presented in Table 3 . Gene level count tables were obtained using the prepDE.py Python script supplemented to Stringtie tool [10] . Differential expression of genes was calculated via DESeq2 [11] . DEGs analysis was carried out for those genes with fold change (FC) 4.0 and fold discovery rate (FDR) adjusted p-value threshold of 0. All ethical requirements were observed in the preparation of the publication. The work was not related to the use of human objects and did not include experiments with animals.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.