Introduction

Xanthomonas oryzae pv. oryzae (Xoo) causes bacterial leaf blight, a prevalent rice disease observed worldwide. The pathogen typically invades rice vascular tissues through wounds or natural openings1. Particularly, indica rice IR24 demonstrates high susceptibility to Xoo. The pathogen employs the type III secretion system (T3SS) to release diverse effector proteins during interactions with both host rice and non-hosts, such as tobacco, leading to disease manifestation and triggering hypersensitive response (HR)2,3. Xoo's pathogenic arsenal includes exoenzyme, HR and pathogenicity, and effector genes, which contribute to its virulence4,5. Additionally, extracellular polysaccharides (EPSs) play a crucial role in Xoo pathogenicity. Efficient colonization of rice leaves by Xoo relies on EPSs, which facilitate attachment to the leaf surface. Upon infection, Xoo produces substantial EPSs, leading to vascular bundle occlusion and disrupted water transport, resulting in wilting and eventual death of the host plant6.

Polyhydroxyalkanoates (PHAs) are biological polyesters synthesized by microorganisms under carbon excess and nitrogen- or phosphate-limited conditions7. Among the various PHAs, polyhydroxybutyrate (PHB) stands out as the most prevalent monomer, characterized by different side-chain groups on C-3. Notably, most microorganisms synthesize a single type of PHA8. Bacteria utilize three primary biosynthetic pathways to produce PHA: acetyl-CoA9, fatty acid β-oxidation10, and fatty acid synthesis pathways11. Xoo, akin to other PHA-producing bacteria, harbors genes, such as phaC and phaE, responsible for PHB biosynthesis, along with phaR for regulatory functions and phaZ for PHB degradation12. In conditions of carbon excess and nutrient constraints, bacteria utilize PHA polymerase (PhaC, also known as PHA synthase) to store surplus carbon as PHAs. Conversely, under starvation conditions, PHA depolymerase (PhaZ) catalyzes the degradation of PHAs, releasing R-hydroxyalkanoic acid, which can be subsequently utilized as a carbon and energy source.

PHA synthase, the key enzyme in PHA biosynthesis, catalyzes the polymerization of 3-hydroxyacyl-CoA (3HACoA) into PHA while releasing CoA13,14. Based on their substrate specificities and subunits, PhaC enzymes are classified into four groups. Class I synthases utilize short-chain hydroxyacyl-coenzyme A as a substrate, while class II synthases can utilize both short- or long-chain hydroxyacyl-CoA. Class III synthases consist of PhaC and PhaE subunits, whereas class IV synthases additionally contain a PhaR subunit15,16. Although all four classes of PHA synthases possess PhaC subunits, only Class I and Class II PHA synthases consist of a single PhaC subunit17,18. Notably, the phaC gene plays a pivotal role in the PHA synthesis pathway and is considered the most critical and indispensable gene for PHA biosynthesis12.

While previous studies have elucidated the impact of the phaR gene on various bacterial characteristics of Xoo19, the relationship between PHA synthesis and Xoo virulence remains unclear. Notably, phaR is downstream of phaC15; however, it is only present in class IV synthases. Conversely, the absence of phaC alone is sufficient to completely block PHA synthesis.

PHA depolymerase, which degrades PHA into R-hydroxyalkanoic acids that can be utilized as carbon and energy sources under starvation conditions, is an essential enzyme for PHB degradation. The PHB degradation pathway, as described in most bacteria, begins with the depolymerization of PHB into d-β-hydroxybutyrate oligomers by PHB depolymerase (PhaZ)20. There are two major types of PhaZs: intracellular PhaZ (iPhaZ) and extracellular PhaZ (ePhaZ). Under starvation conditions, microorganisms degrade the accumulated PHAs using iPhaZ. In Cupriavidus necator, the degradation of PHB by intracellular PhaZ1 results in (R)-3-HB-CoA, which is sequentially converted to crotonyl-CoA, (S)-3-HB-CoA, acetoacetyl-CoA, and two molecules of acetyl-CoA via the β-oxidation pathway; the acetyl-CoA molecule can enter central metabolism21,22. Moreover, microorganisms can utilize exogenous PHAs by secreting ePhaZ to hydrolyze PHAs23. However, whether PHA metabolism affects bacterial growth rate, swarming ability, EPS production, and harpin production remains unknown.

In this study, we generated a knock-out mutant (ΔphaC), a complemented mutant (C: ΔphaC), and an overexpression strain (O: ΔphaC) of PXO99A and assessed their phenotypes. We compared various physiological and biological functions with a positivity control strain (ΔphaZ and ΔphaC/ΔphaZ), respectively. Our results suggested that PHA metabolism plays a critical role in modulating diverse physiological and biological processes, including growth rate, utilization of different sugar sources, EPS production, harpin production, and swarming ability, which are related to Xoo virulence in plants. Furthermore, the results provide a background of the biological functions of PHA in Xoo.

Materials and methods

Construction of strains and plasmids

Bacterial strains and plasmids used in this study are listed in Table S1. PXO99A genome complete nucleotide sequence is accessible in the data bank (GenBank: CP000967.2). A fragment from the PXO99A genome containing the upstream and downstream of phaC and phaZ was amplified by PCR. Primers are listed in Table S2. The knockout mutant was generated in PXO99A by marker-free exchange mutagenesis using the suicide vector pK18mobsacB13. PCR products were digested by HindIII/BamHI or BamHI/EcoRI and pK18mobsacB were digested by HindIII/BamHI or BamHI/EcoRI, then PCR products were inserted in pK18mobsacB to form a new plasmid pK18mobsacB-PhaC and pK18mobsacB-PhaZ. All recombinant plasmids were transformed into Escherichia coli DH5α by electroporation and then introduced into PXO99A by electroporation14. The single homologous recombination mutant was screened on kanamycin-containing medium and 15% sucrose-containing. Clones that could grow on the kanamycin-containing medium but not on the sucrose-containing medium were selected for the next step. The second homologous recombination occurred during the culturing in a liquid medium without any antibiotics or sucrose. The clones that could grow on sucrose-containing medium but not on kanamycin-containing medium were chosen as phaC or phaZ gene-deficient mutant candidates for further PCR characterization (Fig. S1). To construct complementary strain or overexpressed strain, pHMphaC containing the coding sequence of phaC gene was transformed in ΔphaC or PXO99A by electroporation to make the complementary strain C: ΔphaC and overexpressed strain O: ΔphaC. The ΔphaZ and ΔphaC/ΔphaZ mutant is positivity control.

Growth medium and culture conditions

Escherichia coli DH5α was grown routinely for DNA manipulations in lysogeny broth (LB) medium (yeast extract 5 g L−1, tryptone 10 g L−1 and NaCl 5 g L−1) at 37 °C, Xoo was grown in PSA medium (tryptone 10 g L−1, l-Glutamic acid sodium salt 1 g L−1 and sucrose 10 g L−1), NYG medium (yeast extract 5 g L−1, tryptone 10 g L−1 and sucrose 10 g L−1)24, M210 medium (sucrose 5 g L−1, casein enzymatic hydrolysate 9 g L−1, yeast extract 4 g L−1, K2HPO4 3 g L−1 and MgSO4⋅7H20 3 g L−1), PGA medium (tryptone 10 g L−1, l-Glutamic acid sodium salt 1 g L−1 and glucose 10 g L−1), PSSU medium (tryptone 10 g L−1, l-Glutamic acid sodium salt 1 g L−1 and sucrose 150 g L−1), XOM3 medium (sucrose 10 g L−1, dl-methionine 0.097 g L−1, l-Glutamic acid sodium salt 1.87 g L−1, KH2PO4 2.04 g L−1, MnSO4 0.0067 g L−1, ethylenediaminetetraacetic acid monosodium ferric salt 0.101064 g L−1 and MgCl2 1.0165 g L−1)25 at 28 °C. Antibiotics were used at the following final concentrations, as required: 100 μg/ml ampicillin (Amp), 50 μg/ml kanamycin (Kan).

Sugar utilization medium is based on NYG medium by replacing glycogen to detect the ability of bacteria to utilize different glycogen. In sugar utilization experiments, several sugar sources were used, containing glucose, fructose, sucrose, or glycerol. For Xoo PSA pre-cultures, cells were adjusted to an optical density at 600 nm (OD600) of 0.3.

Bacterial growth rate assays

The wild-type PXO99A, ΔphaC, C: ΔphaC, O: ΔphaC, positivity control (ΔphaZ, and ΔphaC/ΔphaZ) were cultured in NYG with different sugar or PSA liquid medium overnight at 28 °C. Then bacterial strains were adjusted to an optical density (OD600) of 0.1. Determined the spectrophotometric value every 2 h and drew a line chart. The experiment was repeated three times, independently.

Effect of 2-deoxy-d-glucose (2-DG) on the ability of the strains to utilize different sugar sources

Bacteria were cultured in PSA liquid medium overnight at 28 °C. Then the wild-type PXO99A, ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) were adjusted to an optical density (OD600) of 0.1. Dilution coating on 2-DG-added plates, and PSA plates without 2-DG addition were used as control. After 48 h of cultured at 28 °C, the growth status of the wild-type PXO99A, ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) were photographed.

Bacterial swarming and swimming assays

For swarming assays, the wild-type PXO99A, ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) were grown in PSA medium for 2 days. Bacterial cells were then grown in PSA medium for 1 day, harvested by centrifugation, re-suspended in sterilized distilled water, and adjusted to an optical density (OD600) of 0.1. 1 μL of bacterial suspension was then dropped onto the plates containing 0.8% (for swarming) or 0.3% (for swimming) agar PSA medium. Plates were incubated at 28 °C for 3 days. The diameters of the swarming zones indicated the ability of bacterial movement. The experiment was repeated three times, independently.

Bacterial extracellular polysaccharides (EPS) measurement

Bacteria were cultured in PSA liquid medium overnight at 28 °C. Then the wild-type PXO99A, ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) were adjusted to an optical density (OD600) of 0.1. The culture medium was transferred to a new PSA medium of 200 mL at a proportion of 1%, and the culture medium was incubated at 30 °C for 2 days at 200 r/min. Then the supernatant was harvested by centrifugation, evaporated concentrate to 1/3 of the original volume, added three times the volume of 95% ethanol, and precipitated overnight at 4 °C. The precipitate was harvested by centrifugation and freeze-dried, determined the extracellular polysaccharides. Under the action of concentrated sulfuric acid, sugars can be dehydrated to form uraldehyde, which can react with anthrone to form blue-green uraldehyde derivatives. The blue-green uraldehyde dv erivatives in the visible region is 630 nm, and colorimetry at this wavelength can be used for sugar quantification19. To facilitate a more direct observation of color discrepancies, 1mL of each reaction product was dispensed into 12 white quartz wells and subsequently photographed.

Determination of PHB content

Chloroform and sodium hypochlorite were added 1:1 to the test tubes containing the Xoo prepared by lyophilization, then mixed thoroughly for 2 h. The solution in the lower chloroform phase was collected after low-speed centrifugation and heated to evaporate the chloroform completely. Add concentrated sulfuric acid, seal with a lid, and digest at 100 °C for 10 min. Allowed the solution to cool and measured the absorbance value at 235 nm. The PHB content in the sample was obtained by comparison with the PHB standard curve.

Virulence assay in rice and analysis of sugar content in rice leaves

In this experiment, the host plant selected the indica rice IR24 and the japonica rice Nipponbare. Transplanted 60 each of IR24 and Nipponbare seedlings that have just grown their fourth leaves to the pool, the rice leaves that grow 15 days after transplanting are called young leaves, and mature rice leaves that grow 45 days after transplanting are called old leaves. Two different periods of rice leaves of IR24 and Nipponbare were used for virulence assays. Around 10 fully expanded leaves of each plant were inoculated with wild-type PXO99A, ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) at OD600 of 0.5 via the leaf-clipping method, each experiment was repeated 3 times. Lesion lengths were measured 14 days after inoculation, and leaves of each rice plant with distinctive disease spots were picked and photographed.

Cut 1 cm2 of leaf from healthy rice leaves, grind them thoroughly, add 1 mL ddH2O to dissolve the sugar in the leaves, then centrifuge briefly, and transfer the supernatant to tubes. Three leaves of each type of rice were collected as replicates. Add assay buffer, water, ATP and NADP in sequence to the supernatant. Four identical replicates were set for each sample, and the absorbance value at 340 nm was measured immediately after adding the supernatant in the first sample, which was recorded as A1. Add Hexokinase (HK) and glucose-6-phosphatedehydrogenase (G6P) to the second part, keep at 30 °C for 30 min, and then measure the absorbance at 340 nm, which is recorded as A2. Continue to add phosphoglucose isomerase (PGI) to the third part, keep at 30 °C for 60 min, and then measure the absorbance at 340 nm, which is recorded as A3. Continue to add β-d-fructosidase (INV) to the 4th portion, keep at 30 °C for 60 min, and then measure the absorbance at 340 nm, which is recorded as A4. The change of OD = 340 nm of the solution represent the amount of NADPH generated, and calculate the content of sucrose, glucose and fructose in the sample according to the amount of NADPH generated26.

ΔAG = A2 − A1, ΔAF = A3 – A2, ΔAS = A4 – A3, Where A1 is OD340 value of solution before reaction, A2 is OD value of solution after adding HK and G6P, A3 is OD value of solution after adding PGI, A4 is OD value of solution after adding INV, ΔAG is OD340 value of the solution after complete oxidation of glucose, ΔAF is OD340 value of the solution after complete oxidation of fructose, ΔAS is OD340 value of the solution after complete oxidation of sucrose.

Hypersensitive response in tobacco

In this experiment, the non-host plant N. tabacum was maintained in a growth cabinet set to 16-h light/25 °C and 8-h darkness/23 °C. Two-month-old N. tabacum was used for HR assays. We extracted the harpin proteins of the wild-type PXO99A, ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) cultured in XOM3 medium, and injected 100 μL soluble proteins in the back of leaves and labeled with them, each experiment was repeated 3 times27. The 1× PBS buffer was as negative control and observed the induction of HR in tobacco leaves after 3 days.

RNA isolation and qRT-PCR

Bacteria strains were grown either in nutrient broth (PSA) medium or in hrp-inducing medium XOM3 overnight at 28 °C25. 1 mL culture was collected for RNA isolation using SV Total RNA Isolation System (Promega, Madison, WI, United States). Reverse transcription was carried out using PrimeScript RT Master Perfect Real Time Kit (TakaRa Bio. Inc., Dalian). Quantitative real-time PCR (qRT-PCR) was performed using SYBR Premix Ex TaqTM Kit (TaKaRa Bio. Inc., Dalian). 16s rDNA was used as internal reference gene. The primer sets used are listed in Table S2.

Statistical analysis

All the data analyses were performed using SPSS® (version 19.0; SPSS Inc). The Duncan significant difference test was used for post-ANOVA pairwise tests for significance, set at 0.05 (P ≤ 0.05).

Results

Sequence analysis of Xoo phaC

In Xoo PXO99A, phaC were 1065 bp. Phylogenetic trees of PhaC gene was also established. As shown in Fig. 1A, Xoo PhaC is highly similar to those of X. oryzae pv. oryzicola YM15, X. campestris pv. raphani MAFF106181, and P. putida NBRC 14164. Figure 1B shows that PhaZ gene shares high similarity with those of X. oryzae pv. oryzicola YM15, X. campestris pv. raphani MAFF106181.

Figure 1
figure 1

(A) Phylogenetic analysis of Xoo based on phaC gene sequences and comparative analysis of protein sequences of phaC and its homologous proteins. (B) Phylogenetic analysis of Xoo based on phaZ gene sequences and comparative analysis of protein sequences of phaZ and its homologous proteins. (C) Wild-type PXO99A, mutants ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) growth in PSA liquid medium. The bacterial strains were inoculated with the final concentration of OD600 of 0.10, and the OD600 of the culture was measured at different time points. The experiment was repeated three times. (D) The relative expression levels of ftsZ gene in wild-type PXO99A, mutants ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) cultured in PSA medium at 72 h, after inoculation. Error bars represent ± SD. Columns with different letters above were significantly different by ANOVA (p < 0.05).

Full size image

phaC mutation reduced the growth rate of Xoo

To test the phaC gene effects on bacterial growth, we examined the growth curves of wild-type PXO99A and the ΔphaC, C: ΔphaC, O: ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) mutants in a nutrient-rich medium (PSA). When inoculated 8 h later, the growth rate of ΔphaZ slowed significantly (Fig. 1C). To further verify whether the deletion of phaC and phaZ affected the expression of cell division-related genes, we measured the expression levels of the ftsZ gene during the decline phase(72 h after inoculation). the transcriptional expression levels of ftsZ gene in ΔphaC and ΔphaC/ΔphaZ were significantly decreased, which were 0.05 and 0.02-fold, respectively (Fig. 1D).

Observation of the morphology by transmission electron microscope (TEM) and PHB content analysis

To verify the difference in the production of PHB by Xoo in the absence of key PHA synthesis and PHA depolymerization genes. Morphological differences were observed among the PXO99A, ΔphaC, C: ΔphaC, O: ΔphaC, ΔphaZ, and ΔphaC/ΔphaZ (Fig. 2). Wild-type PXO99A, C: ΔphaC, O: ΔphaC, and the ΔphaZ mutant exhibited ribosome aggregation, and round plaques (PHB) were generated in the cells. In ΔphaC and ΔphaC/ΔphaZ mutant cells, ribosomes moved toward the cell wall, and no PHB accumulation was observed.

Figure 2
figure 2

Transmission electron microscopy images of wild-type PXO99A, mutants ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) after 24h of growth in PSA liquid medium.

Full size image

The production of PHB using different sugar sources (sucrose, glucose, fructose, and glycerol) was determined. Glycerol is an important raw material for PHA synthesis28,29; So we chose glycerol as the carbon source for the experiment. Since the growth rates of the PXO99A, ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) on different sugar source medium showed considerable differences, we accurately weighed the freeze-dried cells to infer whether the reduction in PHB accumulation was due to reduced bacterial numbers or reduced PHB production. The results of the quantitative analysis are shown in Fig. 3A–F. The PHB production of ΔphaC and ΔphaC/ΔphaZ was significantly reduced when under PHB production-inducing conditions that glycerol was the only carbon source. On the medium with sucrose as the sole sugar source, the PHB production level of the C: ΔphaC and O: ΔphaC mutants returned back to that of PXO99A, while the PHB production of ΔphaC and ΔphaC/ΔphaZ was significantly lower than in PXO99A.

Figure 3
figure 3

PHB content of 6 strains in medium with different carbon sources (sucrose, glucose, fructose, and glycerol). (A) wild-type PXO99A, (B) mutant ΔphaC, (C) C: ΔphaC, (D) O: ΔphaC, (E) ΔphaZ and (F) ΔphaC/ΔphaZ. The PHB content was counted in terms of mg/mL. Error bars represent ± SD. Columns with different letters represent significant difference by ANOVA (p < 0.05).

Full size image

Deletion of phaC affects the ability of Xoo to utilize different sugar sources

Three different concentrations of sucrose, glucose, and fructose NYG medium were used to examine the changes in the metabolic utilization of different sugar sources by the ΔphaC, C: ΔphaC, O: ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) mutants. The results (Fig. 4) showed that in 2% fructose medium, the growth of PXO99A, C: ΔphaC, and O: ΔphaC were significantly inhibited, and the average absorbance at 600 nm was only 0.3 after 24 h of culture. The growth of the ΔphaZ and ΔphaC/ΔphaZ mutants in the 2% sucrose medium was significantly inhibited, and the bacterial density after 24 h of culture were only 46.6% and 57.2% that of ΔphaC, respectively. On the 1% concentration and 0.1% concentration medium with different sugar sources, the growth of ΔphaZ mutant was weaker than the PXO99A, ΔphaC, and ΔphaC/ΔphaZ. However, C: ΔphaC and O: ΔphaC showed no significant difference from wild-type.

Figure 4
figure 4

Growth rate of 6 strains in medium with different carbon sources (sucrose, glucose and fructose). (A) wild-type PXO99A, (B) mutant ΔphaC, (C) C: ΔphaC, (D) O: ΔphaC, (E) ΔphaZ and (F) ΔphaC/ΔphaZ. The bacterial strains were inoculated with the final concentration of OD600 of 0.10, and the OD600 of the culture was measured at 24 h. The experiment was repeated three times. Error bars represent ± SD. Columns with different letters above were significantly different by ANOVA (p < 0.05).

Full size image

Exogenous addition of 2-deoxy-d-glucose (2-DG) affects the ability of the PXO99A, ΔphaC, ΔphaZ, and ΔphaC/ΔphaZ strains to utilize different sugar sources

2-deoxy-d-glucose is a phosphohexose isomerase inhibitor. As shown in Fig. 5A, wild-type PXO99A and the ΔphaC/ΔphaZ mutant grew as round yellow colonies of normal size in the nutrient-rich medium (PSA). The number of ΔphaC and ΔphaZ colonies increased significantly (1.75- and 1.53-fold the number of wild-type colonies), and the colonies were small. After adding 2-DG, the number of PXO99A colonies was reduced to 70.08% that of the control group, and the number of ΔphaZ colonies was significantly reduced to 32.55% that of the control group. In contrast, the addition of 2-DG had positive effects on the growth ofΔphaC and ΔphaC/ΔphaZ; the number of colonies increased by 6.54 and 23.05%, respectively (Fig. 5B). It is worth mentioning that the addition of 2-DG also affected the size of PXO99A and ΔphaC/ΔphaZ colonies. After the 2-DG treatment of PXO99A and ΔphaC/ΔphaZ, the colonies became significantly larger.

Figure 5
figure 5

(A) Colony morphology of wild-type PXO99A, mutants ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) on PSA and PSA medium added with 2-DG. (B) The number of colonies on the plate. Error bars represent ± SD. Columns with different letters above were significantly different by ANOVA (p < 0.05). Different uppercase and lowercase letters represent different groups of ANOVA. (p < 0.05).

Full size image

Deletion of phaC reduced swarming and swimming ability in different sugar sources

To determine whether a mutation in the phaC genes resulted in changes in the swarming and swimming ability of bacteria, the swarming and swimming ability of ΔphaC, C: ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) were compared with that of PXO99A on 0.8% or 0.3% semisolid agar plates, respectively. The swarming ability of ΔphaC on medium containing 0.2% sucrose and 0.2% glucose as the sole sugar sources was significantly lower than wild-type, but superior to that of the ΔphaZ and ΔphaC/ΔphaZ mutants; the swarming zones were 7.36 ± 0.05 and 7.60 ± 0.29 mm, respectively (Fig. 6). The swarming ability of C: ΔphaC was similar to that of PXO99A. On the medium with 0.2% sucrose as the sole sugar sources, the mutants ΔphaC (the swimming zone is 12.68 ± 0.96 mm) and ΔphaZ (the swimming zone is 14.79 ± 0.26 mm) exhibited reduced swimming ability compared with PXO99A (the swimming zone is 16.56 ± 0.65 mm), while ΔphaC/ΔphaZ (the swimming zone is 19.13 ± 0.55 mm) showed increased swimming ability.

Figure 6
figure 6

Swarming and swimming phenotypes of the PXO99A, ΔphaC, C: ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) strains. (A) Diameters of swarming motility halos, formed on semisolid 0.8% agar plates. Error bars represent ± SD. Columns with different letters above were significantly different by ANOVA (p < 0.05). (B) Diameters of swimming motility halos, formed on semisolid 0.3% agar plates. Error bars represent ± SD. Columns with different letters above were significantly different by ANOVA (p < 0.05).

Full size image

Deletion of phaC reduced the EPS production of Xoo in different sugar sources

To examine the effects of phaC mutagenesis on PXO99A morphology, we spotted wild-type PXO99A and the ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) mutants on nutrient-rich plates. Three days later, ΔphaC, ΔphaZ, and ΔphaC/ΔphaZ developed dry colonies with reduced slime content on the surface, whereas PXO99A had normal mucoid colonies (Fig. 7A). This indicates that phaC and phaZ might affect EPS production. The ability to produce EPSs is an essential factor in the virulence of Xoo6.

Figure 7
figure 7

Colony morphology and EPS production of the PXO99A, ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) on the plates. (A) Colony morphology of on the PSA plates. 1: PXO99A, 2: ΔphaC, 3: ΔphaZ, 4: ΔphaC/ΔphaZ. (B) EPS production. EPS from 20 ml bacterial cultures of the same OD600 value were precipitated with ethanol. The depth of the color indicates the weight of EPS. (C) The EPS productivity was counted in terms of mg. Error bars represent ± SD. Columns with different letters represent significant difference by ANOVA (p < 0.05). (D) The relative expression levels of gumD in wild-type PXO99A, mutants ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) cultured in PSA plates with 2% glucose at 72h after inoculation.

Full size image

Next, we investigated whether EPS production by the mutants on medium with different sugar sources was affected. The EPS test showed that with the same volume of culture supernatant (with different sugar sources), EPSs precipitated by ethanol from PXO99A, ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) significantly differed (Fig. 7B). Color depth in Fig. 7B reflects the sugar content of the solution to a certain extent. In the medium with 0.2% fructose (Fig. 7B), the darkest color was for the ΔphaC mutant. Figure 7C indicated that ΔphaC mutant produced the most EPS (3.58 mg/mL), in contrast, wild-type PXO99A produced the least EPS(0.89 mg/mL).

To eliminate the influence of different sugar sources on the growth, we selected the optimal sugar source for Xoo, glucose, for further studies and examined the relative expression levels of gumD (related to the biosynthesis of EPS) in wild-type PXO99A and the ΔphaC, ΔphaZ, and ΔphaC/ΔphaZ mutants cultured in PSA plates with 2% glucose 72 h post-inoculation. The results revealed that gumD in the ΔphaZ mutant was almost no change compared with the wild-type. However, gumD levels in the ΔphaC (0.36) and ΔphaC/ΔphaZ (0.03) mutants decreased significantly (Fig. 7D).

Deletion of phaC reduced the elicitation of HR in tobacco and decreased the expression of harpin-encoding gene hpa1

We extracted the harpin proteins of PXO99A, ΔphaC, C: ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) cultured on XOM3 medium, which mimics plant nutrient status. To compare the ability of the different strains to induce HR, proteins were infiltrated into tobacco (Nicotiana tabacum) leaves. Three days post-inoculation, ΔphaC, C: ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) induced low-intensity HR and produced a smaller necrotic area than that caused by wild-type PXO99A (Fig. 8A), whose necrotic area was the largest at 1.66 cm2.

Figure 8
figure 8

(A) Necrotic spots area caused by four harpinXoo proteins in tobacco leaves. The harpinXoo protein of wild-type PXO99A, mutants ΔphaC, C: ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) induced HP. Protein of the PXO99A, ΔphaC, C: ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) were infiltrated onto 6-week-old tobacco leaves. The PBS was used as control. (B) The relative expression levels of hpa1 in wild-type PXO99A, mutants ΔphaC, C: ΔphaC, ΔphaZ, and ΔphaC/ΔphaZ cultured in XOM3 medium at 48h after inoculation. At least four independent experiments were performed with similar results. Error bars represent ± SD. Columns with different letters represent significant difference by ANOVA (p < 0.05).

Full size image

RT-qPCR results showed that the hpa1 gene was expressed in PXO99A, ΔphaC, C: ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) strains within 48 h when cultured on XOM3 medium. Using the expression level of wild-type PXO99A at 48 h as a reference, the transcriptional expression levels of hpa1 in the ΔphaC, ΔphaZ, and ΔphaC/ΔphaZ mutants were significantly decreased, which were 0.18-, 0.13-, and 0.15-fold that of wild-type PXO99A, respectively (Figs. 8B).

Deletion of phaC and phaZ reduced the virulence of Xoo in rice

The pathogenicity of the wild-type PXO99A ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) on the old and young leaves of indica (IR24) and japonica (Nipponbare) rice was tested using the leaf clipping method, and lesion lengths were measured 14 days post-inoculation. The IR24 rice was used for the virulence test of mutants C: ΔphaC and O: ΔphaC (Fig. 9). As shown in Fig. 10E, on the old leaves of IR24 rice, the average lesion length after infection with wild-type PXO99A is 6.26 ± 0.40 cm, which is significantly longer than those of ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ). As shown in Fig. 10D, on the young leaves of IR24 rice, the average lesion length after PXO99A infection was 14.93 ± 0.51 cm, which was significantly longer than those of ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) at 4.30 ± 0.62, 5.57 ± 0.60, and 5.3 ± 0.56 cm, respectively. As shown in Fig. 9, the average lesion length caused by C: ΔphaC and O: ΔphaC inoculation was 10.80 ± 1.81, and 17.51 ± 1.98 cm. C: ΔphaC and O: ΔphaC showed no significant difference from the wild-type. The average lesion length on Nipponbare rice old leaves (Fig. 11E) following PXO99A infection was 6.70 ± 1.08 cm (Fig. 9B), significantly longer than those of ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) (2.34 ± 0.28, 0.71 ± 0.20, and 1.40 ± 0.24 cm, respectively), while that on young leaves was 4.15 ± 1.08 cm, also significantly longer than those of the ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) (Fig. 11D).

Figure 9
figure 9

Pathogenicity test of wild-type PXO99A, mutants ΔphaC, C: ΔphaC, and O: ΔphaC on the indica rice IR24. (A) PXO99A, ΔphaC, C: ΔphaC, and O: ΔphaC strains were inoculated into 15 days rice leaves by using the leaf-clipping method. The disease symptoms were observed at 14 days post-inoculation. (B) The lesion lengths were recorded from 5 inoculated young leaves for the PXO99A, ΔphaC, C: ΔphaC, and O: ΔphaC.

Full size image
Figure 10
figure 10

Pathogenicity test of wild-type PXO99A, mutants ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) on the indica rice IR24. The contents of (A) sucrose, (B) glucose and (C) fructose in old and young leaves of the indica rice IR24, respectively. (D) The lesion lengths were recorded from 5 inoculated young leaves for the PXO99A, ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ). (E) The lesion lengths were recorded from 5 inoculated old leaves for the PXO99A, ΔphaC, ΔphaZ, and ΔphaC/ΔphaZ. (F) PXO99A, ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) strains were inoculated into 15 days rice leaves by using the leaf-clipping method. The disease symptoms were observed at 14 days post-inoculation. (G) PXO99A, ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) strains were inoculated into 45 days rice leaves by using the leaf-clipping method. The disease symptoms were observed at 14 days post-inoculation.

Full size image
Figure 11
figure 11

Pathogenicity test of wild-type PXO99A, mutants ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) on the japonica rice Nipponbare. The contents of (A) sucrose, (B) glucose and (C) fructose in old and young leaves of the indica rice Nipponbare, respectively. (D) The lesion lengths were recorded from 5 inoculated young leaves for the PXO99A, ΔphaC, ΔphaZ, and ΔphaC/ΔphaZ. (E) The lesion lengths were recorded from 5 inoculated old leaves for the PXO99A, ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ). (F) PXO99A, ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) strains were inoculated into 15 days rice leaves by using the leaf-clipping method. The disease symptoms were observed at 14 days post-inoculation. (G) PXO99A, ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) strains were inoculated into 45 days rice leaves by using the leaf-clipping method. The disease symptoms were observed at 14 days post-inoculation.

Full size image

We also measured the content of different sugars in rice leaves at different times and found that the sucrose content of old and young IR24 rice leaves was similar (36.8 and 37.0 mg/g) (Fig. 10), respectively. However, old leaves (39.2 mg/g) contain more glucose than young leaves (38.0 mg/g), whereas young leaves (40.0 mg/g) contain more fructose than old leaves (38.7 mg/g). In contrast, the fructose and glucose contents of old and young Nipponbare rice leaves were similar (Fig. 11). However, old leaves contain more sucrose (39.0 mg/g) than that young leaves (36.8 mg/g).

Discussion

PHAs serve as intracellular carbon and energy storage materials, accumulating in the cytoplasm when bacteria encounter imbalanced growth conditions, thus, aiding in bacterial resistance to adversity23,30. Over 70 bacteria species are known to produce PHA, with most of them being gram-negative31,32. By analyzing the genome database of PXO99A, we identified gene clusters related to PHA biosynthesis and metabolism (Fig. 1). Our analysis revealed high homology between the phaC genes of Xoo PXO99A and those of Xanthomonas oryzae pv. oryzicola YM15 (Xoc). Long et al.19 demonstrated that the phaR gene in PXO99A significantly influences multiple bacterial characteristics associated with virulence in plants, such as EPS production, growth rate, harpin protein production, and motility. However, Mccool and Cannon15 found that phaR is located downstream of phaC, and sequence alignment revealed no significant sequence homology between the phaR and phaC genes. Consequently, our understanding of how phaC genes impact bacterial traits related to pathogenesis in Xanthomonas spp. remains limited.

Indeed, PHA degradation has been implicated in facilitating bacterial colonization and expansion in competitive environments, such as soil and the rhizosphere33,34. For instance, Eugenio et al.35 demonstrated that the ability to metabolize PHA confers greater viability to Pseudomonas putida wild-type strain KT2442 cells, with the number of viable mutant cells significantly reduced compared with those of the wild-type strain. Similarly, our study found that the growth rate of the ΔphaC and ΔphaZ mutants of Xoo strain PXO99A decreased after 8 h of culture, and the number of bacteria reaching the growth plateau was substantially lower than those of other strains. The mutant strain lacking the phaZ gene, which is responsible for PHA degradation, exhibited a slowed growth rate and reduced bacterial mass due to its inability to degrade the intracellular polymer. We hypothesize that the capacity for efficient PHA metabolism enhances the growth rate of Xoo and facilitates its colonization, expansion, and competitiveness, providing Xoo with a survival advantage in environmental settings.

The metabolic pathways and products derived from sugar sources serve as the primary steps and raw materials, respectively, for synthesizing both EPSs and PHA. PHA presence can render bacteria less susceptible to detrimental factors, allowing them to endure various stresses, such as nutrient deficiencies, and adverse physicochemical or biological conditions for extended periods36. This suggests that PHA can impart robust stress resistance to bacterial cells. Among PHAs, PHB is the most prevalent monomer8. Wang et al.37 engineered Escherichia coli strains capable of both synthesizing and degrading PHB, and observed that recombinant strains producing PHB exhibited greater resilience to adverse environments.

In our study, transmission electron microscopy revealed PHB accumulation in Xoo (Fig. 2). Given the challenging infection process for Xoo, the bacterium likely synthesizes PHA as a mechanism to cope with stress, compete for nutrients in rice leaves or the natural environment, resist unfavorable conditions, and establish growth advantages. Moreover, we quantified PHB production of wild-type PXO99A, the ΔphaC mutant, C: ΔphaC, O: ΔphaC, and the positivity controls (ΔphaZ and ΔphaC/ΔphaZ) using various sugar sources. Regardless of the sugar source they were cultured in, wild-type PXO99A, C: ΔphaC, O: ΔphaC, and ΔphaZ produced substantial amounts of PHB. Conversely, the ΔphaZ mutant only synthesized PHB without decomposing it due to the lack of PHA depolymerase. Consequently, PHB accumulation in ΔphaZ was significantly higher than that in wild-type PXO99A. The ΔphaC and ΔphaC/ΔphaZ mutants lacked PHA synthase and, therefore, could not synthesize PHB, resulting in greatly reduced production (Fig. 3).

In the intricate interplay between pathogens and hosts, a pivotal aspect lies in the pathogen's ability to acquire nutrients from host plants, as nutrient acquisition profoundly impacts pathogen cell division and DNA replication38. Among the nutrients garnered by pathogens, carbohydrates are the favored carbon source for microorganisms39. Pathogenic bacteria typically prioritize the utilization of monosaccharides over di- and polysaccharides, with fructose and glucose taking precedence over galactose39.

To assess the capacity of PXO99A, the ΔphaC mutant, and the positivity controls (ΔphaZ and ΔphaC/ΔphaZ), we utilized various concentrations of sucrose, glucose, and fructose. Notably, we observed differing degrees of inhibition at a 2% fructose concentration. Among these, the wild-type strain experienced the most pronounced inhibitory effect. The growth of the ΔphaC mutant in the 2% fructose medium was significantly superior to that of PXO99A and the positivity controls (ΔphaZ and ΔphaC/ΔphaZ). This discrepancy may be attributed to the pgi gene in Xoo, which encodes phosphoglucose isomerase, catalyzing the production of 6-phosphoglucose and rendering bacteria unable to utilize fructose40. Consequently, wild-type PXO99A cannot thrive normally in fructose. Changes in the ability of PHA synthesis and degradation may influence the expression of the pgi gene, thus, affecting the utilization of fructose by the ΔphaC mutant, C: ΔphaC, O: ΔphaC, and ΔphaC/ΔphaZ mutants (Fig. 12).

Figure 12
figure 12

The relative expression levels of Pgi in wild-type PXO99A, mutants ΔphaC, C: ΔphaC, O: ΔphaC, and positivity control (ΔphaZ, and ΔphaC/ΔphaZ) cultured in PSA medium with xylose at 24h (A) and 48h (B) after inoculation. At least four independent experiments were performed with similar results. Error bars represent ± SD. Columns with different letters represent significant difference by ANOVA (p < 0.05).

Full size image

To examine whether the growth and morphology of Xoo are affected when the glucose metabolism pathway is blocked, we added 2-deoxy-d-glucose (2-DG) to the nutrient-rich medium (PSA). Especially, 2-DG is a synthetic analog of glucose that converts phosphoglucose to phosphofructose and inhibits phosphohexose isomerase, thereby blocking glycolysis at the initiation stage41,42. We observed a difference in colony morphology at 72 h after inoculation. The addition of 2-DG inhibited the pathway for converting glucose phosphate to fructose phosphate (Fig. 13), and the utilization of glucose by Xoo was reduced. In general, in a medium with sucrose, when the glycolytic pathway of glucose is inhibited, Xoo uses fructose, which is produced by breaking down sucrose, to maintain basic life activities. As seen from previous experiments in this study (Fig. 4), fructose produced by sucrose decomposition inhibited the growth of wild-type PXO99A and ΔphaZ; hence, the colony numbers of PXO99A and ΔphaZ were significantly reduced, with ΔphaZ being the most sensitive to 2-DG, followed by wild-type PXO99A, whereas mutants ΔphaC and ΔphaC/ΔphaZ grew better when the glucose metabolism pathway was inhibited. For the ΔphaC and ΔphaC/ΔphaZ mutants, despite experiencing reduced glucose utilization, they exhibited effective utilization of fructose generated through sucrose decomposition. As a result, their growth was not hindered, and in fact, the number of colonies was even higher. As the ΔphaZ mutant lacks enzymes necessary for PHA depolymerization, the number of colonies was significantly reduced after adding 2-DG, much lower than that of the wild-type PXO99A. These results are consistent with those observed for PXO99A, ΔphaC, and positivity control (ΔphaZ and ΔphaC/ΔphaZ) at a 2% fructose concentration, indicating that strains lacking the phaC gene demonstrate improved fructose utilization. While it is known that pathogenic bacteria typically prioritize the utilization of monosaccharides over di- and polysaccharides, with fructose and glucose being preferred over galactose40, the relationship between fructose utilization and PHA production remains understudied. Further investigations are warranted to elucidate how phaC influences bacterial traits related to fructose utilization in Xanthomonas spp.

Figure 13
figure 13

Carbon metabolism pathway, EPS production pathway, PHA synthesis, and metabolism pathway of Xoo.

Full size image

One factor important for colonization and pathogenesis is bacterial surface translocation, which can be achieved by swimming or swarming motility. The swarming ability of bacteria helps them move toward the host to obtain more nutrients and avoid adverse situations43,44. Swarming can increase the virulence of pathogens in the preliminary stages of infection and colonization45. Therefore, swarming ability is an essential virulence factor for phytopathogenic bacteria, such as Xanthomonas spp., Pseudomonas spp., Ralstonia spp., and Erwinia spp46,47,48,49. For example, Yang et al.50 constructed a mutant strain of the Xanthomonas campestris flgM gene and found that it had abnormal flagella and reduced motility, resulting in less virulence to the host.

In this study, we observed that in the optimal medium for Xoo (sucrose or glucose as the sole sugar sources), wild-type PXO99A, C: ΔphaC, and O: ΔphaC exhibited strong swarming ability. In contrast, the swarming ability of the ΔphaC and positivity control (ΔphaZ and ΔphaC/ΔphaZ) mutants was significantly reduced. Specifically, the swimming ability of ΔphaC and ΔphaZ was diminished compared with that of PXO99A, while ΔphaC/ΔphaZ exhibited increased swimming ability on medium with 0.2% sucrose as the sole sugar source. Contrastingly, fructose is not an optimal sugar source; therefore, wild-type PXO99A and the ΔphaC, ΔphaZ, and ΔphaC/ΔphaZ mutants could not sufficiently metabolize fructose to generate enough energy for bacterial swarming (Fig. 6). Based on these results, we speculate that eliminating the PHA synthesis or degradation function changes intracellular nutrient condition, thereby affecting flagella function.

EPSs are an essential virulence factor in plant pathogenic bacteria51. Previous studies have reported that EPSs can chelate Ca2+ in plant cells, thereby inhibiting the immune response of plants52. Reduced EPS production reduces biofilm formation, growth rate, and pathogen survival on leaf surfaces, resulting in reduced virulence and disease symptoms53. In Xanthomonas, the synthesis of EPSs plays a crucial role in pathogenicity. EPSs reportedly suppress callose formation54, contribute to biofilm formation55, and enhance pathogen survival and colonization in plants by providing protection against environmental stressors and host defenses49,56. Therefore, EPS synthesis is positively correlated with pathogenicity in Xanthomonas.

Aneja et al.57 constructed a PHB synthase (phaC) mutant in Sinorhizobium meliloti Rm1021 and found that the EPS production of the mutant was significantly reduced compared with that of the wild type. Similarly, in this study, the EPS production of the ΔphaC, ΔphaZ, and ΔphaC/ΔphaZ mutants was significantly lower than that of the corresponding of wild-type PXO99A. The ΔphaC and positivity control (ΔphaZ and ΔphaC/ΔphaZ) mutants developed dry colonies with reduced slime on the surface, whereas wild-type PXO99A mutants C: ΔphaC and O: ΔphaC exhibited normal mucoid colonies on medium with sucrose as the sole sugar source. Moreover, data on bacteria cultured on medium with glucose, the optimal sugar sources for Xoo, showed that the EPS productivity of ΔphaC and positivity control (ΔphaZ and ΔphaC/ΔphaZ) was significantly impaired, indicating that both phaC and phaZ positively regulate EPS production. In addition, EPS production by the same strains grown using different sugar sources also differed considerably.

The growth rate of ΔphaC and ΔphaC/ΔphaZ in 2% fructose medium was significantly higher than that of the PXO99A and ΔphaZ (Fig. 4). ΔphaC and ΔphaC/ΔphaZ can use 2% fructose concentrations to produce energy substances (adenosine triphosphate [ATP]) for maintaining bacterial life activities and large amounts of EPSs. These results are similar to those for sugar metabolism, suggesting that eliminating PHA synthesis or degradation functions can affect EPS production. We speculate that sugar metabolism can also directly or indirectly determine the pathogenicity of Xoo, while the metabolic processes and products of sugar sources are the primary steps and synthetic raw materials, respectively, for synthesizing EPS and PHA. The reduced EPS in the mutants may be attributed to a decrease in ATP and an insufficient supply of sugar precursors, such as O-acetyl and acetonyl residues.

In Xanthomonas, the harpin protein is a vital protein secreted by the type III secretion system that enables Xanthomonas to trigger a rapid, localized, programmed HR in non-host plants and induce disease induction in host plants58. We observed that the harpin protein of wild-type PXO99A caused extensive blighted spots on N. tabacum leaves. However, under the same inoculation conditions, the lesion areas caused by the ΔphaC and positivity control (ΔphaZ and ΔphaC/ΔphaZ) mutants were significantly smaller than those caused by wild-type PXO99A. These results suggest that deleting phaC and phaZ reduces the harpin protein content of Xoo. As the intracellular energy storage substance PHA cannot be synthesized or degraded in these mutants, the energy supply may be insufficient for the bacteria to secrete and transport effector proteins to cells.

Our experiments revealed that the variation of virulence of Xoo is influenced by multiple factors rather than relying on a single determinant. The pathogenicity of Xoo depends on various virulence factors, including lipopolysaccharides, EPSs, extracellular enzymes, toxins, adhesions, and effectors injected into host rice by the T3SS59,60.

In the present study, differences in the virulence of PXO99A, ΔphaC, and positivity controls (ΔphaZ and ΔphaC/ΔphaZ) against IR24 and Nipponbare rice were compared. Regardless of the rice variety, the average lesion lengths caused by ΔphaC and positivity controls (ΔphaZ and ΔphaC/ΔphaZ) were significantly shorter than those caused by wild-type PXO99A. Rice infection is a challenge for Xoo, which needs to resist the self-defense response of rice and compete with host cells for nutrients. However, the mutants lacking phaC and phaZ genes could not maintain PHA metabolism, making them more sensitive to adverse conditions and unable to establish an advantage during the infection process.

More importantly, the ΔphaC and positivity control (ΔphaZ and ΔphaC/ΔphaZ) mutants, lacking phaC and phaZ genes, could not synthesize or depolymerase PHA. Consequently, their carbon supply was insufficient, resulting in decreases in virulence factors, such as bacterial swarming ability, EPS production, and effector protein secretion and transport, thus, greatly reducing virulence in rice.

Additionally, even the same type of rice shows different resistances at different stages of the reproductive growth period. Nipponbare before the tillering stage (in the seedling and transplanting stages) exhibits strong resistance to Xoo and is most susceptible to disease at the tillering and booting stages. In the present study, the contents of different sugars in rice leaves at different stages were determined, and the virulence of Xoo in those leaves was tested. The results showed that old IR24 rice leaves had more glucose and less fructose, whereas young leaves had less glucose and more fructose. In contrast, the fructose and glucose contents of the old and young Nipponbare leaves were similar, with more sucrose in the old leaves and less sucrose in the young leaves.

When comparing the pathogenicity of wild-type PXO99A across different growth stages of the same rice variety, the length of disease spots on old IR24 rice leaves was significantly shorter than the corresponding length on young leaves. In contrast, old Nipponbare rice leaves exhibited significantly longer disease spots than those of young leaves. The optimal sugar sources for Xoo are sucrose and glucose, while fructose cannot be effectively utilized. Therefore, when rice leaves contain more sucrose and glucose, Xoo can successfully infect them. ΔphaC lacks the phaC gene and activates the fructose metabolic pathway. We speculate that when rice leaves contain more fructose and less sucrose and glucose, the ∆phaC mutant causes longer disease spots on rice leaves than those of the positivity control mutants (∆phaZ and ∆phaC/∆phaZ). This indicated that the content of different sugars in rice leaves plays an important role in the process of Xoo infecting rice to a certain extent.

From these observations, it can be inferred that carbon metabolism directly or indirectly influences the pathogenicity of Xoo. The metabolic processes and products of sugar sources serve as the primary steps and raw materials for synthesizing EPSs and PHA.

Therefore, this study reveals the prominent role played by phaC in Xoo in the maintenance of various physiological and biological functions through the carbon supply, growth rate, swarming ability, EPS production, and harpin expression. It also underscores the indispensable role of PHA synthesis in Xoo virulence.