Bacterial strains and plasmids used and constructed in this study are listed in Table 1.

Ralstonia solanacearum was grown and its inocula prepared as described (Stulberg et al., 2015). To isolate R. solanacearum from inoculated plant samples, 0.5-cm plant stem sections were prepared and homogenized as described (Stulberg and Huang, 2015), and dilution plated onto modified semi-selective medium agar plates (Huang and Lakshman, 2010). Escherichia coli strains were cultured at 37°C in Luria-Bertani medium (Miller, 1972). When needed, antibiotics were added at 25 μg/ml for kanamycin and 15 μg/ml for gentamicin. Since R. solanacearum strain UW551 is a select agent pathogen in the United States, manipulation of the strain was conducted in a secured laboratory and virulence assays described below were performed in a secured greenhouse section approved for select agent research by USDA/APHIS using standard operating procedures also approved by APHIS for race 3 biovar 2 strains of R. solanacearum.

Standard molecular biology techniques were used for plasmid isolation, restriction digestion, cloning, and transformation of E. coli strains (Sambrook and Russell, 2001). Total bacterial DNA was extracted using Qiagen’s Blood and Tissue Kit (Qiagen, Chatsworth, CA, United States) following the manufacturer’s instructions.

Primers designed in this study were listed in Table 2. They were designed based on the deposited UW551 draft genome sequence in GenBank (ASM16795v1, GCA_000167955). The regions selected for primer design were entered into the free online A plasmid Editor (ApE) program. Similar design parameters (GC = 45–60%, Tm = 60–64°C, primer length 18–26) were used for primers in each pair. The specificity of each primer pair and amplicon was checked by BLASTn against the UW551 genome, and the nr and WGS databases in GenBank for specificity.

Colony PCR was performed by picking R. solanacearum cells using a sterile toothpick or pipette tip from a single colony grown on a plate and mixing the cells in 100 μl of sterile water. The cell suspension was boiled for 5 min and cooled on ice or stored at -20°C until use. Two to five microliters of the suspension were used for PCR.

PCR to amplify the upstream and downstream prophage regions was conducted in a 20 μl volume containing 1x KAPA HiFi HotStart ReadyMix (Kapa Biosystems, Boston, MA, United States), 5 pmol of each primer, and approximately 20 ng of DNA template. PCR conditions were 1 cycle of 3 min at 95°C, followed by 30 cycles of 20 s at 98°C, 15 s at 58°C, and 30 s at 72°C, with a final extension of 2 min at 72°C. PCR to amplify virulence-related genes was conducted in a 20-μl volume containing 1x GoTaq Green Master Mix (Promega, Madison, WI, United States), 20 ng of template DNA, and 5 pmol of each primer. PCR conditions were 1 cycle of 4 min at 94°C, 30 cycles of 1 min at 94°C, 1 min at 60°C, and 1 min at 72°C, with a final extension of 10 min at 72°C.

To study the role of orf14 of the prophage Rs551 and the prophage itself in R. solanacearum, two mutants of R. solanacearum which had deletions in the prophage were constructed in strain UW551 by homologous double recombination. Mutant UW551ΔϕRs551-orf14, designated as the UW551 orf14 mutant, was generated by replacing a 342-bp prophage region, located in contig 0570 of UW551 (GenBank accession number: AAKL01000012.1), with a 616-bp gentamicin cassette. The 342-bp fragment contained the entire 291-bp prophage region corresponding to ϕRs551’s orf14, coding for a putative type-2 phage repressor, as well as 42-bp upstream and 9-bp downstream of orf14. Mutant UW551ΔϕRs551, designated as the UW551 prophage mutant, was constructed by replacing a 3,321-bp prophage region with a 616-bp gentamicin cassette. The 3,321-bp fragment contained the same 342-bp in the orf14 mutant, as well as an additional 2,979-bp located in the same contig of UW551, corresponding to ϕRs551’s orf1in the replication module, orf2 between the replication and structure modules, and six (orf3 to orf7, and 1,005-bp of the 1,524-bp of orf8) of the seven structural genes in the structural module of the prophage (Figure 1). To make the mutants, a 616-bp gentamicin cassette was first amplified from a colony of TP997, purchased from Addgene (Cambridge, MA, United States), by PCR with primers 5′-CGAATCCATGTGGGAGTTTA-3′ and 5′-TTAGGTGGCGGTACTTGGGT-3′ (Poteete et al., 2006). The cassette was then cloned into the TOPO site of the vector pCR Blunt II TOPO to generate pCR Blunt II-Gm using Invitrogen’s Zero Blunt^® TOPO^® PCR Cloning Kit according to the manufacturer’s instructions. Two regions of DNA, 994-bp in size located upstream and 815-bp downstream of the 342-bp prophage region, were amplified by PCR using primers in Table 2, digested with respective restriction enzymes, and cloned sequentially into the multiple cloning sites before and after the gentamicin cassette in pCR Blunt II-Gm to obtain pCR Blunt II-ϕRs551-orf14 Up-Gm-ϕRs551-orf14 Down. Similarly, the same 994-bp upstream fragment and a 850-bp downstream fragment of the 3,321-bp prophage region was cloned sequentially to obtain pCR Blunt II-ϕRs551 Up-Gm-ϕRs551 Down. The resulting plasmids were electroporated into competent cells of R. solanacearum strain UW551 as described by Ahmad et al. (2017). This was followed by selection on gentamicin-containing triphenyltetrazolium chloride (TZC) plates (Kelman, 1954) for transformants that had undergone homologous double recombination between the Up-Gm-Down region in pCR Blunt II-Up-Gm-Down and the Up and Down prophage sequences in the chromosome of UW551 of R. solanacearum. The knock-out mutants were identified by screening on TZC plates amended with gentamicin, and confirmed by PCR using primer pairs located within the sequences of the mutated regions that showed a lack of any amplified products. To determine if a phage was still produced from the UW551 mutants, an aliquot from the supernatant of the R. solanacearum mutants was subjected to the spot test and plaque-forming assay (Ahmad et al., 2017) using R. solanacearum RUN 302, a strain susceptible to ϕRs551 and which contained no ϕRs551 sequence in its genome before infection. The presence or absence of phage particles in the supernatant of the R. solanacearum mutants was also examined under transmission electron microscope (Ahmad et al., 2017).

Extracellular polysaccharide in the supernatant of R. solanacearum was determined quantitatively, in vitro growth of R. solanacearum strains were measured, and twitching motility examined as described by Ahmad et al. (2017), except that twitching motility was visualized using a Zeiss AxioZoom v16 stereo zoom microscope (Carl Zeiss Microscope GmbH, Germany). Two replicates were used for each strain in the EPS assay and the experiment was repeated three times.

Total bacterial RNA was isolated from 3 ml of R. solanacearum culture at the exponential growth phase (OD₆₀₀ = 0.3) using Qiagen’s RNeasy Protect Bacterial and RNeasy Mini Kits (Qiagen, Inc., CA) according to the manufacturer’s protocol. Ambion^® TURBO DNA-free^TM DNase Treatment and Removal Reagents (Life Technology) were used to remove contaminating DNA from the RNA preparation and to subsequently remove the DNase and divalent cations from the sample. The absence of DNA contaminants was confirmed by PCR using gene-specific primers (Table 2) on the RNA samples. The bacterial genomic DNA of R. solanacearum strain UW551 was used as positive, and sterile water negative controls. The RNA samples were quantified using the Nanodrop ND-1000 spectrophotometer (NanoDrop Tech. Inc.). One microgram of the RNA was reverse transcribed using Quantabio’s qScript^TM cDNA SuperMix (Quantabio, Beverly, MA, United States) according to the manufacturer’s instructions. Quantitative PCR analysis was carried out with gene-specific primers (Table 2) using 1 μl of cDNA as a template in a 20 μl volume containing 10 μl of IQ^TM SYBR^® Green Supermix (Bio-Rad) and 0.5 μM each of the gene primers. Cycling conditions were 95°C for 3 min, followed by 45 cycles of 95°C for 10 s, 60°C for 15 s, and 72°C for 30 s. At the end of the program, melting curve (65–95°C with a heating rate of 0.5°C/min) was analyzed to confirm the specificity of the primer set (Addy et al., 2012a). Relative levels of gene expression were determined using the 2^-ΔΔC_T method (Livak and Schmittgen, 2001), with the 16s rRNA gene as the internal control (Addy et al., 2012a). The experiments were performed three times with two replicates each time.

For virulence assays, tomato plants were seeded, transplanted, inoculated by soil drenching with 50 ml of R. solanacearum (5 × 10⁷ cells/ml) (Ahmad et al., 2017). Inoculated plants were rated using a disease index (DI) ranging from 0 (healthy) to 4 (76–100% leaves wilted) (Roberts et al., 1988). There were 10 plants per treatment and the experiment was repeated three times. To determine if ϕRs551’s orf14 or the phage itself offers any competitive advantage to its carrier strain UW551, a competition assay was similarly performed as described for the virulence assay, except that in addition to inoculation with the ϕRs551-susceptible strain RUN302 alone, tomato plants were also co-inoculated with the following two different R. solanacearum strains, respectively, in a 1:1 (25 ml:25 ml) ratio: (1) wt RUN302 (does not carry and is susceptible to ϕRs551) and the wt UW551 (ϕRs551-carrier), (2) RUN302 and the UW551 orf14 mutant strain, and (3) RUN302 and the UW551 prophage mutant strain. Tomato plants inoculated with water were served as negative controls. As soon as the inoculated plants showed signs of wilting, bacterial colonies were isolated as described above. To estimate the ratio of and to differentiate between the two strains in the stem sections, 50 randomly picked colonies, 10 from each of five different wilted plants per treatment, were subjected to the multiplex PCR developed by Stulberg et al. (2015). The experiment was performed three times.

Data for EPS dry weight and virulence were analyzed by one-way ANOVA using web-based statistical software¹. Means were compared using the Tukey’s Honest Significant Difference test provided by the software. The values in gene expression and competitive fitness were shown as the means of three experiments. Differences were considered statistically significant at P < 0.01.

The 7,929-nucleotide genome sequence of ϕRs551 corresponds to nucleotides 73,039–80,967 in contig 0570 of the deposited genome sequence of R. solanacearum strain UW551 (Figure 1). R. solanacearum mutants were confirmed by their ability to grow on TZC plates containing gentamicin, and by PCR for the absence of the 342-bp prophage region in the UW551 orf14 mutant and the 3,321-bp region in the UW551 prophage mutant (data not shown). The UW551 orf14 mutant was found to produce phage particles spontaneously in its supernatant at a rate similar to the wt strain UW551. This was shown when the supernatant of the overnight culture of the UW551 orf14 mutant strain was subjected to the spot test and plaque-forming assay using ϕRs551-suceptable strain R. solanacearum RUN302, similar plaque formation was observed, and a similar number of plaques was obtained as with the supernatant of the wt UW551 (data not shown). On the contrary, no plaques were formed when the supernatant of the prophage mutant strain was subjected to the same phage susceptibility assay, and no phage particles were observed under transmission electron microscope.

To characterize the UW551 orf14 and prophage mutant strains, we first compared the in vitro growth of the mutants with their wt strain UW551, and found all three strains grew at a similar rate (data not shown). When the three strains grew on regular TZC medium plates, however, the colonies of the mutant strains appeared more fluidal and irregular than those of the wt UW551, suggesting a high production of EPS. This observation was confirmed by an EPS quantitative assay that showed both mutant strains produced significantly higher amounts of EPS (73.6 ± 4.5 mg/10 ml for UW551 prophage mutant, and 59.6 ± 9.5 mg/10 ml for UW551 orf14 mutant) than the wt strain (44.3 ± 7.2 mg/10 ml). The difference in EPS production between the two mutant strains was not significant. The two mutant strains also displayed distinctly different twitching motility when compared with the wt strain UW551 (Figure 2). For the wt strain, we observed twitching motility under a microscope as indicated by the formation of corrugated trajectories with smooth edge around the margin of its colonies (Figure 2, left). The size of the trajectories, however, was larger with irregular edges in R. solanacearum mutant strains, especially in the orf14 mutant (Figure 2, middle).

To study the effect of deletion of the targeted prophage regions in the virulence of R. solanacearum, we compared the virulence of the wt to that of the mutant strains of R. solanacearum (Figure 3). The wt strain UW551 did not cause any disease symptoms until 8 days after soil drenching inoculation (DI > 0), and reached a DI of 3.1 at day 21 (Figure 3). The virulence level caused by the two mutant strains of R. solanacearum, however, was significantly higher (Figure 3). The mutants started to cause disease symptoms 5 days after inoculation and completely wilted all inoculated plants (DI = 4) by day 17 (Figure 3). DIs caused by the UW551 prophage mutant were statistically similar to the ones caused by the UW551 orf14 mutant, except at days 7 and 8 (Figure 3).

To identify the factors contributing to increased virulence of the R. solanacearum mutant strains, expression of five genes (pilT, egl, pehC, hrpB, and phcA), all known virulence factors of R. solanacearum, was compared between the mutant and the wt strains. The expression of all five genes was increased in the two mutant strains, with the pilT gene showing the greatest increase: 8-fold in the orf14 mutant and 21-fold in the prophage mutant (Figure 4). The level of expression of the egl, pehC, hrpB, and phcA genes was increased between 2.0- and 4.5-fold in the two mutants (Figure 4).

The effect of UW551 and its mutant strains on plant colonization by RUN302, a strain of R. solanacearum lacking the prophage Rs551, was studied. RUN302 was co-inoculated with one of the UW551 strains for infection of tomato plants. The ratio of the two mixed strains was determined in stem sections of infected tomato plants. When the wt ϕRs551-carrier strain UW551 was co-inoculated with the wt ϕRs551-lacking strain RNN302, 30 out of 50 randomly picked bacterial colonies isolated from the tomato stems belonged to UW551 (Figure 5), significantly more than the number of RUN302 colonies. On the contrary, when each of the two UW551 mutant strains was co-inoculated with RUN302, only 10 and 19 out of 50 were colonies of the prophage mutant strain and the orf14 mutant strain, respectively, significantly less than the number of RUN302 colonies in the stems (Figure 5).

To study the effect of UW551 and its mutant strains on the virulence of R. solanacearum strain RUN302, tomato plants were inoculated by soil drenching with RUN302 alone, or together (1:1) with the wt UW551, the UW551 orf14 mutant or the UW551 prophage mutant, respectively (Figure 6). When RUN302 was co-inoculated with UW551, the co-inoculation caused a delayed and significantly lower DI 4 days after inoculation than inoculation with RUN302 alone (Figure 6). On the contrary, when RUN302 was co-inoculated with the UW551 prophage mutant, its overall virulence was similar to that caused by RUN302 alone (Figure 6). DIs caused by RUN302 co-inoculated with the UW551 orf14 mutant strain were lower than the ones caused by RUN302 alone or co-inoculation with RUN302 and the UW551 prophage mutant, but the difference was not significant 9 days after inoculation and all inoculated plants were completely wilted 6 days after that (Figure 6).