BUSCADOR RECOLECTA

A) RNA sampling conditions from environmental (soil and mineral water) and previously obtained samples (rich B medium reference and three in planta conditions). B) Transcriptomic data analysis pipeline. First, raw RNA-seq data quality was evaluated with FastQC (v.0.11.5), trimmed with trimGalore (v.0.6.1) and potential rRNA contaminants were filtered out with the SortMeRNA software (v.4.2.0). Reads were mapped with Bowtie2 (v. 2.4.4) and alignments quantified with FADU v. 1.8. The R. solanacearum UY031 genome GCF_001299555.1_ASM129955v1 was used. DEG analyses were performed with Deseq2 (v. 1.34.0). Genes with |log2(fold-change)|>1.5 and adjusted p-value <0.01 were considered as differentially expressed (DEG) when compared to the reference medium. The UpsetR package (v. 1.4.0) (90) was used to detect unique DEG and intersections among the in the different conditions. Deseq2 transformed counts normalized for sample size were used for principal component analysis. C) DEGs in the different conditions. Bars show the total number of up- (yellow) and downregulated (blue) genes for each condition compared to the reference Rich B medium. The line graph and the right Y axis indicate the percentage of DEGs. Drawings created with BioRender., Peer reviewed

Dot plots of the KEGG (left) and GO (right) enrichment analyses of differentially expressed genes (DEGs) from soil (brown) and water (blue) conditions. Dot sizes represent the number of genes associated with each term and dot colour indicates the p. adjusted value. The gene ratio represented in the X axis is the proportion of associated genes to a term from the total gene set. The DEGs were extracted with DEseq2 using the thresholds: p-adj.value > 0.01 and log 2 FC ± 1.5 and ClusterProfiler was used to calculate the enrichment., Peer reviewed

R. solanacearum cultures grown overnight in rich B medium were washed and diluted to OD600 = 0.1 in water and luminescence and OD600 values were measured over a 24h period. Relative luminescence units (RLU) were normalised by bacterial concentration measured as OD600. RLU values were divided by 1000 to facilitate visualisation., Peer reviewed

Time-course expression of prhG, hrpB and hrpY reporter strains at native basic (A to F) or neutral pH (G) and after pH neutralisation with HCl or alkalinisation with KOH. For each time point, luminescence was measured (RLU) and normalised by OD600. All values were divided by 1000 to facilitate visualisation. Letters indicate the water sources detailed in the methods section., Peer reviewed

A) Representation of the main components of the nitrogen metabolism and their expression in different conditions (log2 fold change with respect to rich B medium). Genes depicted in A correspond to genes highlighted in bold in B. B) Heatmap representation of the log2 fold change in expression with respect to growth in rich B medium for all the genes classified in the nitrogen metabolism group. The colour palette ranges from blue (downregulated) to yellow (upregulated genes) as indicated in the key. Locus names are presented without the preceding letters RSUY_., Peer reviewed

Heatmap representation of gene log2 fold change with respect to the rich B medium in the different conditions for A) All stress response genes and B) T3SS and T3E gene categories according to the curated classification (see M&Ms). The colour palette ranges from blue (downregulated) to yellow (upregulated genes) as indicated in the key. Locus names are presented without the preceding letters RSUY_., Peer reviewed

Real time q-PCR was performed to validate gene expression in soil and water at 3 dpi. The genes selected were ahpC (RSUY_35830), katE (RSUY_45660), katG (RSUY_09930) and oxyR (RSUY_07740). Relative expression of target genes was calculated with the DeltaCt method using the geometric mean of the housekeeping gene serC (RSUY_11010) as reference. Error bars represent standard error across three technical replicates., Peer reviewed

A) Representative confocal laser scanning microscope images of bacterial cells extracted from soil microcosms at 28 dpi. Cells were stained using the BacLight LIVE/DEAD bacterial viability kit. Green cells represent viable cells and red represents dead cells. Scale bar represents 50 μm. B) Quantification of dead cells at different time points of soil microcosms based in BacLight LIVE/DEAD staining. Cells were quantified using Fiji ImageJ. Detailed methods:0.5 g from soil microcosms experiments was resuspended with 1 mL MilliQ water, samples were vortexed and soil particles allowed to sediment. Supernatants were recovered, centrifuged at 1000 rpm for 5 min to eliminate soil and supernatant was transferred to a fresh tube and cells were pelleted by centrifugation at 10000 rpm, 5 min and finally, cells were resuspended in 300 μL 0.85% NaCl. Then, 50 μL of cells were stained with 0.2 μL of BacLight LIVE/DEAD bacterial viability kit (ThermoFisher), incubated for 5 min in the dark and visualized promptly with a confocal microscope. Green SYTO9 fluorescence was measured with excitation at 488 nm and emission at 500–520 nm and red PI fluorescence was measured by excitation at 561 nm and emission at 610–630 nm. The alive control was prepared by washing an overnight culture with 0.85% NaCl and the dead control was prepared by killing cells from the overnight culture with 70% isopropanol for 1 h, followed by centrifugation of cells and resuspension in 0.85% NaCl. The number of alive and dead cells was quantified using ImageJ by setting a “threshold method = Moments” and particles size = 0.5-Infinity., Peer reviewed

Bacterial pathogens exhibit a remarkable ability to persist and thrive in diverse ecological niches. Understanding the mechanisms enabling their transition between environments is crucial to control dissemination and potential disease outbreaks. Here, we use Ralstonia solanacearum, the causing agent of the bacterial wilt disease, as a model to investigate the bacterial transcriptomic adaptation to two environments, water and soil, and compare it with the well-studied in planta gene expression. We identified extensive transcriptional reprogramming in response to each environment. Strikingly, gene expression in water resembled expression during late xylem colonization, with an intriguing induction of the type 3 secretion system (T3SS). We reveal that basic pH and also to nutrient scarcity - conditions also encountered in planta during late infection stages - trigger the T3SS induction. In the soil environment, R. solanacearum upregulated stress-responses and alternate carbon source metabolism genes, mostly phenylacetate metabolism and glyoxylate cycle and downregulated virulence-associated genes. Furthermore, we proved through gain- and loss-of-function that genes encoding proteins associated with oxidative stress, such as the oxidative stress regulator OxyR and the catalase KatG, are key for survival of the bacterium in soil. This work identifies essential factors necessary for R. solanacearum to complete its life cycle and provides valuable insights into the biological processes driving pathogen adaptation to unexplored environments out of their hosts.