Introduction

Bacterial pustule is caused by the bacterium Xanthomonas axonopodis pv. glycines and is one of the most serious diseases affecting soybean. It is common in areas with warm and wet weather, where the bacterium is spread by water or windblown rain entering the plant through the stomata or via wounds [1, 2]. The pathogen causes premature defoliation and chlorophyll degradation, both of which result in decreased seed size and yield [3]. Tests using artificial infection of susceptible soybean reported average yield losses of 4.3%, of which 86% was attributed to reduced seed yield and 14% to reduced seed size. Prathuangwong and Khandej [4] reported yield reductions from natural dissemination to a susceptible soybean cultivar (SJ4) of 20–35%. The most susceptible age was 6–7 weeks after emergence.

In order for the pathogen to be successfully established in the host plant, it must adhere to and invade the intercellular spaces of the leaf tissue. The entry of bacteria into host plant is complex and plant somata or hydathodes have been shown to function in innate immunity against bacterial plant infection [5, 6]. Experimental evidence suggests that the presence of mechanical forces, cuticle degrading enzymes, cell wall degrading enzymes (CWDE), adhesins, extracellular hydrolytic enzymes, and other translated effectors, play a role in the penetration of plant tissue [7,8,9,10,11,12,13,14,15]. Enzymes produced by Xanthomonas species, including pectinase, polygalacturonase, cellulase, endoglucanase, and xylanase, have been reported to stimulate cellulose and hemicellulose degradation of host plants [16,17,18,19,20]. Similarly, X. axonopodis pv. glycines produces extracellular polysaccharides, toxin, bacteriocins, cellulase, and protopectinase, which support plant infection [21,22,23]. Indole acetic acid and cytokinin phytohormones are also thought be involved in pustule formation of X. axonopodis pv. glycines [24]. Thowthampitak et al. [23] reported that extracellular diffusible factor (DSF), which is synthesized by X. axonopodis pv. glycines, is related to a well-characterized quorum-sensing molecule and virulence factor on soybean. Subsequently, the genome of X. axonopodis pv. glycines 12-2 has been reported (CP015972), only limited information is available on the ompA gene as an outer membrane lipoprotein for pore formation and transport of specific substrates [25].

Outer membrane protein (OmpA) is a structural protein of the outer membrane that is present in many gram-negative bacteria [26]. It is known to contribute to the structural integrity of the outer membrane and is also implicated in bacterial adhesion and invasion, and translocation of effector molecules [26, 27]. Current evidence suggests that OmpA is involved in at least one secretory system and serves as an additional accessory protein. [28]. Since pathogenicity often depends on these systems, OmpA may be considered a virulence factor. More importantly, OmpA and OmpA-equivalent proteins have been found in pathogenic bacteria. These include OprF in Pseudomonas aeruginosa, P5 in Haemophilus influenzae, OmpA in Klebsiella pneumoniae, major outer membrane protein (MOMP) in Chlamydia trachomatis, and OmpA1 in Xanthomonas albilineans [27,28,29,30]. Several studies have examined the functions of OmpA. OmpA-deleted mutants of Cronobacter sakazakii and Escherichia coli were 87% less invasive in both human cells and C6 glioma cells [31, 32]. Whereas the role of OmpA in animal pathogens is well established, little is known about its role in phytopathogenic bacteria. Bruening [33] showed an association between outer-membrane proteinB (MopB) and OmpA in a chlorosis strain of Xylella fastidiosa on quinoa plant (Chenopodium quinoa). The virulence of X. albilineans has been altered by knocking out the OmpA gene [30]. Similarly, transposon mutagenesis of MopB in X. campestris pv. campestris resulted in an altered OMP composition, impaired xanthan production, increased sensitivity, reduced adhesion, and motility to stressful conditions including SDS, elevated temperature, and changes in pH. The modified bacterium could no longer cause black rot disease in Brassica sp. [34]. Taken together, these results suggest that OmpA is a promising target for the treatment of phytopathogenic bacterial infections. The goal of the present study was to elucidate the role played by ompA in X. axonopodis pv. glycines and in pathogenesis to soybean.

Methods

Bacterial Strains and Culture Conditions

Xanthomonas axonopodis pv. glycines strain KU-P-SW005 was cultured at 28 °C on nutrient broth yeast extract agar (NBY) [19]. An ompA mutant was cultured on NBY containing 50 μg/ml kanamycin (Sigma Chemical Co., Saint Louis, MO). Complemented ompA was cultured on NBY containing 50 μg/ml kanamycin and 40 μg/ml gentamycin (Sigma). Escherichia coli was cultured on LB medium (Difco) containing antibiotics (Sigma) (Table 1) [20]. DNA isolation and manipulation were performed following an earlier study [35]. All bacterial proteins were separated using SDS-PAGE (12% w/v), followed by staining of protein bands with Coomassie brilliant blue R-250.

Table 1 Bacterial strains and plasmids
Full size table

Construction of ompA Mutant and Complemented Strain

Though the genome of X. axonopodis pv. glycines strain 12-2 has been reported (CP015972), is available on the ompA gene as an outer membrane lipoprotein for pore formation and transport of specific substrates. Sequences of ompA gene from X. axonopodis pv. glycines strain 12-2 were used as the genetic template to develop an ompA mutant of X. axonopodis pv. glycines strain KU-P-SW005. Knockout mutagenesis of the ompA homologs in the KU-P-SW005 cells was accomplished using site-directed replacement with a kanamycin resistant marker [36]. OmpA_upstream_F and upstream_R and OmpA_downstream_F and downstream_R primers were used to amplify 500 bp DNA fragments upstream and downstream of the ompA gene (Table 2). Both amplicons used as the template to amplify 1 kb DNA fragment by OmpA_upstream_F and OmpA_downstream_R primers. The 1 kb amplicons were ligated into the AscI site of the kanamycin fragment, then into the EcoRI site of the E. coli vector pUC19, followed by transformation using E. coli TOP10 competent cells. After sequence verification, pUC-ompA was introduced into the KU-P-SW005 cells by electroporation, as described in [37]. Two μl of the construct and 0.5 μl of TypeOneTM restriction inhibitor were added to 50 μl of KU-P-SW005 electrocompetent cells and subjected to electroporation at 2.5 kV, 25 µF, and 200 Ω. The electroporated cells were then added to 1 ml of fresh NBY broth and incubated at 28 °C for 1 h. The transformants were selected using NBY medium supplemented with 50 μg/ml kanamycin. Disruption of the genetic loci in the marker-exchange mutant was confirmed by sequencing, PCR, and Southern blots.

Table 2 Oligonucleotides used
Full size table

For genetic complementation, the entire ompA gene and its promoter was amplified by PCR using KU-P-SW005 chromosomal DNA as the template and cloned into the broad-host-range vector pBBR1MCS-5. After verification of the sequence, the plasmid pBBR-ompA was introduced into the ompA mutant, ompA:: Kmr, to form the complemented strain. Verification was accomplished by sequencing and PCR with the primer OmpA_F and OmpA_R as described in Tables 1 and 2.

Plate Assay for Cellulase Activity

The relative level of cellulase production was assessed using radial diffusion assays. KU-P-SW005, the ompA mutant (ompA:: Kmr), and the complemented strain [ompA:: Kmr (pBBR-ompA)] were grown in LB at 28 °C overnight. A cork borer was used to make 0.5-cm wells in the agar medium and 10 μl of bacterial suspension at OD600 = 0.2 (OD600 = 0.2, ca. 108 cfu per ml of each) were placed in each well. Inoculated plates containing assay medium (0.5% carboxymethyl cellulose (CMC), 0.1% NaNO3, 0.1% K2HPO4, 0.1% KCl, 0.05% MgSO4, 0.05% yeast extract, 0.1% glucose, and 1.7% agar) were incubated at room temperature overnight, then stained with 1% Congo red for 20 min and washed twice with 1 M NaCl. CMC degradation appeared as white halos surrounding the wells [20, 23]. Carboxymethyl cellulase (CMCase) activity was analyzed using the (H2–C2)/C2 ratio (H—diameter of the halo, C—diameter of the colony). All experiments were performed three times with ten replicate plates.

Pectate Lyase Activity

KU-P-SW005 cells, including the wildtype, ompA mutant, and a complemented strains, were evaluated for their ability to produce pectate lyase by separately placing 10 μl of each bacterial suspension at OD600 = 0.2 into 5 mm diameter wells on a polygalacturonate (PGA) solid medium. The medium contained basal medium (per liter) of 2.0 g (NH4)2SO4, 0.3 g MgSO4·7H2O, 0.3 g CaCl2·2H2O, 0.5 g FeSO4·7H2O, 10.0 g KH2PO4), and 4 g of polygalacturonic acid (pH 7) [39,40,41]. Subsequently, plates were incubated at 30 °C for 24 h, followed by flooding with a cetyl-trimethyl ammonium bromide solution (10 g/l). Colonies that produced pectate lyase were surrounded by clear haloes as a result of substrate degradation.

All bacterial strains were grown in PGA broth and incubated on a rotary shaker at 150 rpm at 30 °C for 24 h. After removal of cells by centrifugation at 8000×g for 10 min, the clear supernatant was used as the crude extra cellular enzyme source and the amount of pectate lyase produced was investigated. Pectate lyase activity was assayed by adding 0.3 ml of diluted sample to a solution containing 1 ml of 0.9% polygalacturonic acid and 0.7 ml of 50 mM glycine buffer containing 0.5 mM CaCl2 (pH 8.5). The mixture was incubated at 30 °C for 60 min. After incubation, the reaction was terminated by adding 4 ml of 0.01 M HCl to the mixture. Crude enzyme was inactivated by boiling in water for 10 min and used as the control. Absorbance was measured at 232 nm and the galacturonic acid concentration of all samples was estimated using a calibration curve constructed from known concentrations (0–2.5 mM). One unit of pectate lyase activity was defined as the amount of enzyme needed to release 1 μmol of reducing sugar (galacturonic acid) per min from the polygalacturonic acid [41]. All experiments were performed three times with ten replicate plates.

Polysaccharide Production

Cells of wildtype, KU-P-SW005, ompA mutant, and the complemented strain from the overnight culture were harvested by centrifugation and inoculated at an initial OD550 of 0.35 into XOLN medium [42] containing 2% (w/v) glucose. After 72 h, cultures were diluted 2- to 10-fold with distilled water and centrifuged at 12,000×g for 20 min to remove the cells. Xanthan polysaccharides in the supernatant were precipitated by adding 40 mM NaCl and 70% (v/v) of ethanol and incubated at 20 °C overnight. The suspensions were centrifuged at 20,000×g for 30 min. The pellet was washed once with 70% (v/v) ethanol and resuspended in distilled water. Xanthan levels were determined using a modified anthrone method [43] and calculated using a standard curve constructed from known concentrations of purified xanthan polysaccharides [42]. All experiments were performed three times with ten replicate plates.

Biofilm Assessment

Biofilm development was assessed using 96-well polystyrene culture plates. X. axonopodis pv. glycines wildtype, ompA mutant, and the complemented strain cell suspensions were grown to an initial OD600 nm of 0.1 (OD600 = 0.1, ca. 106 cfu per ml of each) in NBY broth and 150 ml were added to the wells. Cultures were maintained for 7 days at 28 °C with agitation at 200 rpm. A Synergy 2 plate reader (Biotek) was used to quantify the OD600 nm in each well, and this was considered the ‘total’ measurement. The plates were rinsed three times with water. For measurement of attached cells, 200 ml of aqueous 0.1% crystal violet were added to each well and the plates were kept at room temperature for 20 min. Subsequently, the plates were washed three times with water, 200 ml of 6:4 of acetone:ethanol were added and the plates were agitated for 5 min. The ‘attached cell’ measurement was performed by measuring the acetone:ethanol–dye solution at OD600 [44].

Electron Microscopy

KU-P-SW005, ompA mutant, and the complemented strain cells were collected from the periphery of the 18-hour-old colonies. The cells were mixed in distilled water, deposited on Formvar-coated grids, stained with either phosphotungstic acid or uranyl acetate, and examined using a JEOL-1230 Kv80 transmission electron microscope (TEM).

Virulence to Soybean

The virulence of KU-P-SW005, ompA mutant, and the complemented strain to the susceptible soybean cultivar Spencer was assessed, using a previously described quantitative method [20]. Aqueous cell suspensions (OD600 = 0.2, ca. 108 cfu per ml of each) were sprayed onto the leaves of plants in a greenhouse. At 7–10 days after spraying inoculation, the disease severity was assessed using the scoring method described in [4]. Three trifoliate leaves from the top, middle, and basal portions of three plants inoculated with each strain were evaluated.

Results and Discussion

The pathogenesis of pustule disease by X. axonopodis pv. glycines involves multiple steps, including leaf attachment, epiphytic fitness, entry, colonization, biofilm formation/dispersal, and invasion into the intercellular space of the host [45,46,47]. Virulence factors including adhesins, polysaccharides, lipopolysaccharides (LPS), and degradative enzymes such as cellulase are often found in the pathogenic strains of Xanthomonas species. Those virulence factors play a major role in the pathogenicity of the Xanthomonas species. Kim et al. [48] reported the presence of type secretion system III (T3SS) in X. axonopodis pv. glycines. Kladsuwan et al. [49] reported X. axonopodis pv. glycines strain SP4, Race 1 carried 6 transcription activator-like effector (TALe) genes including tal1a, tal1b, tal1c, tal2a, tal2b, and tal3. Strain 12-2, Race 2 carried 6 (same as strain SP4), and KU-P-SW005, Race 3 carried 5 (same as strain SP4 but tal1c absent) that tal genes correlated with pustule disease pathogenesis on different soybean cultivars. Although, genome of X. axonopodis pv. glycines strain 12-2 (CP015972) was reported and showed the position of the ompA similar gene from the T6SS cluster in the genome (locus_tag = ”A9D66_11115”). T6SS has not been reported for this species, this accessory protein may also be involved in the pathogenesis of X. axonopodis pv. glycines as previous reported that the gene encoding the OmpA protein is located near the T6SS cluster and, more importantly, may be an essential protein of the T6SS system [29].

Preparation of a Deletion ompA Mutant

Due to the high degree of within-species similarity of Xanthomonas, the ompA from X. axonopodis pv. glycines strain 12-2 (CP015972) was successfully used as the genetic template when amplifying ompA and developing the ompA mutant of X. axonopodis pv. glycines strain KU-P-SW005. The ompA gene was deleted from KU-P-SW005 using site-directed mutagenesis with a kanamycin resistant marker [36]. Deletion of the ompA gene was confirmed by PCR using multiple primer sets. The primers, OmpA_upstream_F and downstream_R (Table 2), amplified a 1725 bp band of ompA from the wildtype bacteria and a 2400 bp band from the deletion plasmid pUC19-ompA and the deletion mutant strain (ompA:: Kmr). The primers OmpA_F and OmpA_R are complementary to sequences within the ompA gene, producing a 725 bp band for wildtype cells and no fragments for the mutant bacteria or plasmid control. In addition, it was determined by Southern blot and sequencing analyses that the ompA mutant strain (ompA:: Kmr) had a single transposon insertion in ORF A9D66_11115 (CP015972), encoding a predicted outer membrane or secreted lipoprotein. A 35 kDa protein of OmpA protein was consistently the most abundant in wildtype strain KU-P-SW005 and the complemented strains ompA:: Kmr (pBBR-ompA) and was detectable by SDS-PAGE, but this was not the case with the ompA mutant ompA:: Kmr (data not shown). Previous SDS-PAGE analysis of outer membrane proteins (OMPs) from X. campestris showed that heterogeneity of the OMP profiles existed within individual pathovars, with a 37 kDa protein being consistently the most abundant in isolates of X. campestris [50]. The OmpAs from X. campestris also had a strong identity with those from other members of Xanthomonas (98%), the related genera Stenotrophomonas (86.7%), and Xylella (68.4%); however, OmpAs showed a low level of identity when compared with OprF of P. aeruginosa (30.3%) and OmpA of E. coli (20.8%) [50,51,52]. Mutation of ompA gene resulted in alteration of the OMP profile. OMPs serve a range of functions crucial to cell viability and activity, including structural support and catalysis, and are involved in both active transport and passive diffusion. All proteins are synthesized by ribosomes in the cytosol. OMP preproteins containing N-terminal signal peptides are first translocated across the inner membrane through translocons. Once reaching the periplasm, the leader signal peptide of OMP is removed by signal peptidase. The change in the surface layer is probably the cause of the multiple defects observed in the ompA mutant.

Study of the Deletion ompA Mutant

Several factors play essential roles in the pathogenesis of Xanthomonas species. These include the presence of a T3SS [48], the avirulence gene avrXg1, which confers resistance to the soybean resistant cultivar Williams 82 [37], and the synthesis of exoenzymes and extracellular polysaccharide (EPS) [20, 23]. The ompA gene produces outer membrane proteins that are known to contribute to pore formation and the transport of specific substrates; however, the roles of these proteins in the virulence of X. axonopodis pv. glycines have not been explored. An ompA functional gene for virulence factor production in X. axonopodis pv. glycines KU-P-SW005 was produced by marker exchange of a cassette for resistance to kanamycin. In this study, we examined the production of cellulase and pectate lyase in the wildtype, ompA mutant (ompA:: Kmr), and complemented ompA mutant [ompA:: Kmr (pBBR-ompA)]. KU-P-SW005 and the complemented strain [ompA:: Kmr (pBBR-ompA)] colonies were hydrolyzed with CMC, as shown by the significantly large diameter of the halos surrounding the colonies, of 1.9 and 1.7 cm, respectively. In comparison, the ompA mutant (ompA:: Kmr) showed a 42.11% smaller inhibition zone, of 1.1 cm (Table 3). CMC degradation was confirmed by the modified CMCase activity. Selection criteria were changed from the intensity of the clear zones below the bacterial colonies on the CMC plates to the size of the clear-zones formed around the colonies. The CMCase activity was significantly lower in the ompA mutant (ompA:: Kmr) than in the wildtype and complemented strains [ompA:: Kmr (pBBR-ompA)]. As shown in Table 3, the wildtype, ompA mutant (ompA:: Kmr), and complemented strain [ompA:: Kmr (pBBR-ompA)] showed CMCase activity indexes of 6.72, 0.95, and 5.97, respectively. This suggested that ompA supports CMCase secretion in X. axonopodis pv. glycines.

Table 3 CMCase and pectate lyase activities of Xanthomonas axonopodis pv. glycines KU-P-SW005, ompA mutant, and complemented strain
Full size table

Based on their growth on PGA medium and production of clear zones, all strains showed positive pectate lyase activity. However, the wildtype and complemented strain had significantly larger clear zones, of 2.2 and 2.1 cm, respectively, compared with the 1.1 cm clear zone of the ompA mutant (Table 3). These results were further confirmed by pectate lyase production under liquid conditions. The wildtype and complemented strains showed significantly higher pectate lyase activity, at 45.55 and 44.15 Unit/ml, respectively, compared with 7.73 Unit/ml for the ompA mutant (ompA:: Kmr) (Table 3). Collectively, these data suggest that the mutation of ompA inhibited the production of pectate lyase and cellulase in the X. axonopodis pv. glycines strain KU-P-SW005.

Our study demonstrated that EPS production in X. axonopodis pv. glycines is also influenced by the presence of an OmpA protein. The wildtype produced 2.5 mg/ml dry weight EPS in NGB. In contrast, the ompA mutants (ompA:: Kmr) exhibited significantly reduced EPS production in the same medium (0.5 mg/ml dry weight). Complementation of the ompA mutant [ompA:: Kmr (pBBR-ompA)] by transformation with pBBR-ompA restored production of polysaccharide to levels similar to those of the wildtype, at 2.1 mg/ml dry weight EPS in NGB (Table 4).

Table 4 Polysaccharide production, cell aggregation, and virulence to soybean of Xanthomonas axonopodis pv. glycines KU-P-SW005, ompA mutant and complemented strain
Full size table

Bacterial Growth, Biofilm Assessment, Electron Microscopy, and Virulence to Soybean

The X. axonopodis pv. glycines cells containing ompA mutant (ompA:: Kmr), complemented ompA mutant [ompA:: Kmr (pBBR-ompA)], and wildtype (KU-P-SW005) grew equally well in NGB (data not shown), excluding the possibility that the ompA gene is responsible for the growth of this bacterium. However, the amount of biofilm formed by the ompA mutant was less than that of the wildtype cells (p < 0.05) (Fig. 1). TEM revealed that the ompA mutant (ompA:: Kmr) cells were not surrounded by a polysaccharide layer, in contrast with the KU-P-SW005 and complemented strains, both of which had thick polysaccharide walls (Fig. 2).

Fig. 1
figure 1

Biofilm formation of Xanthomonas axonopodis pv. glycines wildtype; KU-P-SW005, ompA mutant; ompA:: Kmr, and ompA complemented strain; ompA:: Kmr (pBBR-ompA) after 7 days in NBY broth at 28 °C and 120 rpm. Optical density of planktonic and attached cells was measured. Statistically significant differences (p < 0.05), as determined by Duncan’s new multiple range test, are indicated by letters “a” and “b”

Full size image
Fig. 2
figure 2

Electron microscopy of cells of aXanthomonas axonopodis pv. glycines wildtype; KU-P-SW005, bompA mutant; ompA:: Kmr, and compA complemented strain; ompA:: Kmr (pBBR-ompA). Bars = 0.5 µm

Full size image

No significant differences were observed in the growth of wildtype, ompA mutant, or complemented ompA mutant on Spencer leaves at 1, 2, 3, 4, and 10 days (data not shown). However, the ompA mutants showed significantly lower virulence. As shown in Fig. 3, the ompA mutants had virulence of 11.23%, compared with the 87.98% of the wildtype at 10 days after spray inoculation (Table 3). This suggested that OmpA is one of the essential proteins for pathogenicity of X. axonopodis pv. glycines.

Fig. 3
figure 3

Disease induction by spray inoculation at 108 cfu/ml with aXanthomonas axonopodis pv. glycines wildtype; KU-P-SW005, bompA mutant; ompA:: Kmr, and compA complemented strain; ompA:: Kmr (pBBR-ompA)

Full size image

Multiple defects were caused in the ompA mutant due to mutation. These multiple defects include altered biofilm formation, decreased cellulase and pectase lyase activity, reduced polysaccharide production, and possibly reduced adhesion and motility, all of these result in decrease of pathogenicity. Mutation of the ompA gene associated with OMP stabilizes in the periplasm and disturbs the folding and insertion into the outer membrane through the β-barrel assembly machinery of a variety of macromolecules, including proteins, polysaccharides, nucleic acids, and lipids. Motility and biofilm formation during pathogenesis affected by the resulting decrease of adhesion and cell-to-cell attachment. Virulence factors such as cellulase and pectize lyase are unable to translocate across the inner membrane. However, since OMP involves multiple genes, the precise function of the omp gene cluster must be elucidated through further studies.

Conclusions

The mutation in the X. axonopodis pv. glycines ompA gene had pleiotropic effects, including reduced formation of cell aggregates and biofilm, and reduced production of cellulase, pectate lyase, polysaccharide, and possibly xanthan. Since the mutation altered the synthesis of extracellular enzymes, the initial stages of pathogenesis may have been limited. It, therefore, seems reasonable to conclude that the reduction of pathogenicity in the ompA mutant was the result of increased sensitivity to stress. This may reduce epiphytic fitness and aggregate formation, as well as motility and adhesion, changing biofilm formation/dispersal and limiting bacterial colonization and spread. Our discovery of pleiotropic effects caused by mutation of the ompA gene, including decrease of pathogenicity, may improve understanding the pustule disease cycle and biology of X. axonopodis pv. glycines during pathogenesis on soybean and develop the control strategies for this disease.