Multiple virulence factors regulated by AlgU contribute to the pathogenicity of Pseudomonas savastanoi pv. glycinea in soybean

Bacterial strains, plasmids, and growth conditions

The bacterial strains and plasmids used in this study are listed in Table 1. Pseudomonas strains were routinely cultured on King’s B (KB; King, Ward & Raney, 1954) medium or mannitol-glutamate (MG; Keane, Kerr & New, 1970) medium at 28 °C. Escherichia coli (E. coli) cultures were grown on Luria-Bertani (LB; Sambrook, Fritsch & Maniatis, 1989) medium at 37 °C. The bacterial cell densities at 600 nm (OD₆₀₀) were measured using a Biowave CO8000 Cell Density Meter (Funakoshi, Tokyo, Japan) as described in Sakata et al. (2021).

Bacterial in vitro growth measurements

Wild-type, the algU mutant, and the algU-complemented strain were grown at 28 °C on LB medium. The bacterial suspensions were standardized to an OD₆₀₀ of 0.05 with LB, and bacterial growth was measured at OD₆₀₀ for 6, 9, and 12 h.

Plant material and inoculation procedures

Soybean plants (Glycine max), cultivar “Enrei”, were grown in a growth chamber at 22 °C, with approximately 60% humidity, and a supplementary light intensity of 200–350 μmol/(m² s) for a 14 h photoperiod. All soybean plants used for virulence studies were 3 or 4-week-old.

Psg carrying Tn5 transposon was syringe-infiltrated into soybean leaves at an OD₆₀₀ of 0.1 (5 × 10⁷ CFU/ml) containing 0.02% Silwet L-77 (OSi Specialties Inc., Danbury, CT, USA). The disease symptoms were fully developed at 6 days post inoculation (dpi) (Fig. 1A). The mutants which showed different disease symptoms or virulence reduction in comparison to Psg WT were selected.

For spray inoculation, bacterial suspensions were applied to observe disease symptoms on whole soybean plants as described previously (Uppalapati et al., 2007). Plants were sprayed with a bacterial suspension at an OD₆₀₀ of 0.2 (1 × 10⁸ CFU/ml) in sterile distilled water containing 0.025% Silwet L-77 until runoff. After inoculation, plants were transferred to growth chambers at 28 °C with approximately 90% to 100% humidity for 24 h in the dark before maintaining plants at approximately 70% humidity.

For syringe inoculation, bacteria were suspended at a final concentration at an OD₆₀₀ of 0.1 (5 × 10⁷ CFU/ml), 0.01 (5 × 10⁶ CFU/ml), and 0.001 (5 × 10⁵ CFU/ml), and infiltrated with a one-ml blunt syringe into leaves. The plants were then incubated at 70–80% humidity for the rest of the experimental period. Leaves were removed and photographed at five dpi.

To measure bacterial growth in soybean leaves after spray inoculation, individual second leaf pairs were selected at six dpi, weighed and surface-sterilized in 5% H₂O₂ for 3 min, and then rinsed three times with sterile water. The leaves were then homogenized, and appropriate dilutions were plated on KB medium containing the appropriate antibiotics. The bacterial colony forming units (CFU) were normalized as CFU/g using the total inoculated leaf mass. The population at 0 dpi was estimated using leaves harvested 1 h post inoculation (hpi) without surface-sterilization. For syringe-inoculation, leaf discs were harvested using a 3.5 mm-diameter cork-borer from syringe-infiltrated leaf zone. The leaves were then homogenized, and appropriate dilutions were plated on KB medium containing the appropriate antibiotics. The bacterial colony forming units (CFU) were normalized as CFU/cm² using the leaf square meters. The bacterial populations were evaluated in at least three independent experiments.

Transposon-mediated mutagenesis and identification of mutated genes

Transposon-mediated mutagenesis was carried out, as described previously (Sakata et al., 2019). Briefly, pBSLC1 (Sawada et al., 2018) carrying mini-Tn5 transposon were transferred into Psg to build a mutant library. We developed more than 1,000 individual Psg mutant lines. After the inoculation assay, we identified the mutated genes by rescuing the transposon insertion sites into an E. coli plasmid and sequencing (Sakata et al., 2019).

Complementation of the algU mutant

The algU-complemented strain was constructed as described in Ishiga et al. (2018). Briefly, the pDSKG vector was made from pDSK519 vector (Keen et al., 1988) by replacing kanamycin cassette to gentamycin. The algU and promoter region were transferred into the pDSKG vector to generate pDSKG-algU. The pDSKG-algU construct was introduced into the algU mutant by electrophoresis to generate the complemented strain.

Real-time quantitative RT-PCR

For Psg gene expression profiles, data were collected as previously described in Sakata et al. (2021). Specifically, bacteria were grown in HS medium optimized for COR production (HSC; Palmer & Bender, 1993) for 3 and 48 h. Bacterial RNA was extracted using the ReliaPrep RNA Cell Miniprep System Kit (Promega, WI, USA) according to the manufacture’s protocol. Two micrograms of total RNA were treated with gDNA Remover (TOYOBO, Osaka, Japan) to eliminate genomic DNA, and the DNase-treated RNA was reverse transcribed using the ReverTra Ace qPCR RT Master Mix (TOYOBO). The cDNA (1:10) was then used for RT-qPCR using the primers shown in Table S1 with THUNDERBIRD SYBR qPCR Mix (TOYOBO) on a Thermal Cycler Dice Real Time System (TaKaRa). Psg outer membrane lipoprotein I (oprI) was used to normalize gene expression.

COR quantification by HPLC

Psg WT, the algU mutant, and the algU-complemented strain were cultured in HSC for 7 days. Culture supernatant was obtained by centrifugation (12,000 × g for 5 min). Cell pellets were dried at 65 °C and weighed. The 500 µl of supernatants were extracted twice with 500 µl of ethyl acetate and 25 µl of HCl, and the organic phase was transferred to a new microcentrifuge tube. The sample was dried by centrifugal evaporator at 55 °C, and the dried sample was dissolved with 0.05% trifluoroacetic acid (TFA)/acetonitrile (9:1, v/v). The culture supernatant was analyzed by HPLC with a Shimadzu LC20A system equipped with a Symmetry C8 column (4.6 × 250 mm; Waters Corporation, MA, USA) as described previously (Sakata et al., 2021).

Hypertrophy-inducing activity assay on potato tuber tissue

Potato tubers were cut from the central tuber portion to ensure samples of high uniformity. After washing in tap water for 5 min, each disc was washed with sterile distilled water several times. Potato tuber discs were inoculated using toothpicks by placing the tip in Psg WT, COR-defective mutants (cfa6 and cmaA), the algU mutant, and the algU-complemented strain on a KB medium plate, and then placing the toothpick on the potato tuber disc. The discs were then placed at 23 °C incubator (darkness) for 5 days. Photographs were taken at five dpi.

Biofilm formation assay

Biofilm formation was assayed as described previously (Shao et al., 2019). Briefly, the bacterial strains were incubated overnight in LB broth and resuspended in fresh LB broth to an OD₆₀₀ of 0.1. Bacterial suspensions (120 μl) were put into 96-well plates and incubated at 28 °C for 24, 48, 72, and 96 h. The bacterial solutions were discarded and washed three times with distilled water. The biofilm forming bacteria were treated with 150 μl of 0.1% crystal violet (CV; Fujifilm, Tokyo, Japan) for 20 min without shaking. The dye was discarded and washed twice with distilled water. The plate was dried completely, subsequently the biofilm was eluted with 150 μl of 100% ethanol, and the CV were dissolved completely. Finally, the eluted biofilm sample’s absorbance was measured at OD₅₉₅.

Identification and characterization of reduced virulence mutants

To identify Psg virulence genes, we screened 1,000 transposon insertion mutants for reduced disease symptoms on soybean leaves using the syringe-infiltration method. Disease symptoms caused by Psg WT showed a small water-soaked lesion surrounded by regions of chlorosis (Fig. 1A). A total of 28 mutants showed no or less chlorosis at six dpi (Fig. 1B). Seventeen mutants out of 28 had transposon insertions in genes encoding COR biosynthesis-related genes (Table 1). Soybean leaves inoculated with COR biosynthesis mutants (VTD29, VTE13, VTI15, and VTR4), the algU mutant (VTO41), and an unknown-function mutant (VTF6) showed no chlorosis (Fig. 1B).

Reduced disease symptoms and bacterial growth in soybean

We identified the 28 mutants which showed reduced disease symptoms compared to Psg WT by syringe-inoculation (Fig. 1B). We further investigated whether these mutants also showed reduced virulence via spray-inoculation, and selected two mutants related to COR biosynthesis (cmaA and cfa6), and others. Soybean leaves inoculated with all mutants showed significantly reduced bacterial populations than those of Psg WT (Fig. 2A). Among all mutants, the algU mutant (VTO41) showed dramatically reduced bacterial populations and disease symptoms at six dpi (Figs. 2A and 2B). To confirm whether the altered algU mutant phenotype originates from a corresponding mutation, an algU-complemented strain was generated. Psg WT and the algU-complemented strain showed the same bacterial population levels as well as disease symptom development in soybean (Figs. S1A and S1B). We also confirmed that algU is apparently dispensable for Psg growth in rich LB medium, since no growth difference was observed among WT, algU mutant, and algU-complemented strain (Fig. S2).

To further investigate the algU contribution to Psg virulence, we conducted syringe infiltration with WT, algU mutant, and algU-complemented strain. As a result, the algU mutant showed reduced symptoms and bacterial populations at all inoculum levels we tested (Figs. 3A–3F). Taken together, these results indicate that AlgU contributes to growth both on leaf surface and in apoplast, and to causing disease.

AlgU regulates the expression of Psg virulence genes in HSC medium

Pst DC3000 AlgU positively regulates virulence gene transcription (Markel et al., 2016; Ishiga et al., 2018). To investigate whether Psg AlgU also regulates virulence genes, we analyzed virulence gene expression profiles in HSC medium. COR biosynthesis-related genes including cmaA and corR, in the algU mutant showed reduced expression at 48 h after incubation (Figs. 4A and 4C). Moreover, hrpL, encoding HrpL (an alternative sigma factor recognizing the hrp box in the promoter of T3SS genes), also showed significantly less expression in the algU mutant at both 3 and 48 h after incubation compared to Psg WT (Fig. 4D). These results indicate that AlgU positively regulates COR biosynthesis-related genes and hrpL.

To investigate whether AlgU can coordinate gene expression involved in Psg motility, we determined the expression profile of fliC (encoding flagellin, relating to flagellar mobility). At 3 h after incubation, there was no difference in the flagellar-encoding gene expression between the algU mutant and Psg WT. However, after 48 h incubation, relative fliC expression was greater in the algU mutant compared to Psg WT (Fig. 4E). Additionally, algD expression was down regulated in the algU mutant (Fig. 4F), indicating that AlgU positively regulates alginate biosynthesis-related genes.

AlgU contributes to COR biosynthesis and biofilm formation in Psg

We demonstrated that COR biosynthesis-related genes in the algU mutant showed reduced expression in HSC medium (Figs. 4A–4D). To investigate whether AlgU contributes to COR production, we first conducted a hypertrophy-inducting activity test on potato tuber tissues for COR detection (Sakai et al., 1979; Völksch, Bublitz & Fritsche, 1989). Potato tuber tissues inoculated with Psg WT showed hypertrophy response, but those inoculated with COR-defective mutants (cmaA and cfa6) showed no response (Fig. 5A). The algU mutant-inoculated potato tuber tissues showed less hypertrophy response compared to those inoculated with Psg WT and the reduction was restored in the algU-complemented strain (Fig. 5A). Furthermore, we also quantified COR production by using HPLC. Psg WT produced around 90 ng/g of COR in HSC medium (Fig. 5B). However, the algU mutant produced only around one third as much COR compared to Psg WT (Fig. 5B). We also confirmed that the algU-complemented strain recovered COR production more than WT (Fig. 5B). Taken together, these results suggest that AlgU contributes to COR biosynthesis in Psg.

Since we also demonstrated that alginate biosynthesis-related genes algD showed reduced expression in algU mutant (Fig. 4F), we next investigated the biofilm formation ability in Psg WT and the algU mutant. The algU mutant showed a reduction in biofilm formation at 24, 48, and 72 hpi (Fig. 6). These results suggest that AlgU also contributes to biofilm formation in Psg.