Biocontrol of Soft Rot Dickeya and Pectobacterium Pathogens by Broad-Spectrum Antagonistic Bacteria within Paenibacillus polymyxa Complex

Polymyxin-producing bacteria within the Paenibacillus polymyxa complex have broad-spectrum activities against fungi and bacteria. Their antibacterial activities against soft rot Dickeya and Pectobacterium phytopathogens containing multiple polymyxin-resistant genes were not clear. Here, we selected nine strains within the P. polymyxa complex having broad-spectrum antagonistic activities against phytopathogenic fungi and a polymyxin-resistant D. dadantii strain causing stem and root rot disease of sweet potato and did antagonistic assays on nutrient agar and sweet potato tuber slices. These strains within the P. polymyxa complex showed clear antagonistic activities against D. dadantii in vitro and in vivo. The most effective antagonistic strain P. polymyxa ShX301 showed broad-spectrum antagonistic activities against all the test Dickeya and Pectobacterium strains, completely eliminated D. dadantii from sweet potato seed tubers, and promoted the growth of sweet potato seedlings. Cell-free culture filtrate of P. polymyxa ShX301 inhibited D. dadantii growth, swimming motility, and biofilm formation and disrupted D. dadantii plasma membranes, releasing nucleic acids and proteins. Multiple lipopeptides produced by P. polymyxa ShX301 may play a major role in the bactericidal and bacteriostatic actions. This study clarifies that the antimicrobial spectrum of polymyxin-producing bacteria within the P. polymyxa complex includes the polymyxin-resistant Dickeya and Pectobacterium phytopathogens and strengthens the fact that bacteria within the P. polymyxa complex have high probability of being effective biocontrol agents and plant growth promoters.


Introduction
Gram-negative Pectobacterium and Dickeya bacteria produce multiple pectinases and cause destructive soft rot diseases of numerous crops and ornamental plants and great economic losses worldwide [1,2]. While the worldwide distribution of soft rot Pectobacterium was well-known, soft rot Dickeya was previously described as being present mainly in tropical and subtropical regions but is expanding its global distribution. For example, Dickeya spp. associated with blackleg of potato and foot rot of rice spread rapidly in Europe as far north as Finland near the Arctic Circle [3,4] and in China from South China to Northeast China [5]. Importantly, soft rot Dickeya is threatening staple food crops, including potato [3,6], rice [7], maize [8], sweet potato [9,10], and banana [11,12].
Management of soft rot diseases on crops has not been successful due to the lack of resistant crop varieties and the end of large-scale use of effective antibiotics for the risk of introducing resistance to bacterial pathogens in humans or animals [1]. Biological control is an alternative to the breeding of resistant crops and chemical control the diseases and has been increasingly tested to control Pectobacterium and Dickeya pathogens, particularly using bacteriophages with restricted bacterial hosts [13] and broad-spectrum antagonistic bacteria [14,15].
Gram-positive Bacillus and Paenibacillus bacteria, which produce multiple antimicrobial compounds against broad phytopathogens and produce endospores resistant to heating and dryness for the formulation of stable products, are high-profile biological control agents [16][17][18]. Paenibacillus polymyxa is well-known for its production of the antibiotic polymyxins, which are cyclic lipopeptides and used as last-resort agents against notorious Gram-negative multi-drug resistant human pathogens, including Acinetobacter baumannii, Pseudomonas aeruginosa, and Klebsiella pneumoniae [19]. P. polymyxa forms a monophyletic species complex with P. peoriae, P. kribbensis, P. ottowii, P. brasilensis, P. terrae, and "P. maysiensis," most of which also produce the cyclic lipopeptides-fusaricidins against broad-spectrum fungi, oomycetes, and Gram-positive bacteria, including important phytopathogens [16,20].
Polymyxins are a group of decapeptides containing five to six residues of the nonproteogenic amino acid 2,4-diaminobutyric acid, resulting in a high positive charge density. Polymyxins bind to the lipid A component of lipopolysaccharides on the outer membrane of Gram-negative bacteria and disrupt the outer membrane and then permeabilize and disrupt the inner membrane [16,21]. The general mechanism of bacterial resistance to polymyxins involves modification of lipid A, such as the addition of 4-aminoarabinose by the arnB operon products, the addition of phosphoethanolamine by EptA, hydroxylation of lipid A acyl chains by LpxO, and acylation or deacylation of lipid A by PagP or PagL [22], thereby reducing the net negative charge and polymyxin-binding affinity [16,21]. In Gram-positive bacteria, resistance to antimicrobial peptides occurs in the esterification of phosphate with alanine of teichoic acids by the dltXABCD operon products [23]. In all the sequenced genomes of Dickeya and Pectobacterium, not only arnB operon and eptA are present similar to other enterobacteria, but also dlt operon in Gram-positive bacteria is present [24]. These genes confer resistance to polymyxins.
Polymyxin-resistant Dickeya and Pectobacterium may resist the antagonistic activity from polymyxin-producing strains within the P. polymyxa complex. However, previous studies have not intentionally tested this hypothesis. Here, we tested if the antagonism of the P. polymyxa complex is negative on the polymyxin-resistant Dickeya and Pectobacterium and see if the antagonistic spectrum of the P. polymyxa complex can extend to the polymyxin-resistant Dickeya and Pectobacterium. We selected strains within the P. polymyxa complex previously screened out for their broad-spectrum antagonistic activities against phytopathogenic fungi [20,25] to test their antagonistic activities against Dickeya and Pectobacterium. A polymyxin-resistant D. dadantii strain was used for the first-round screening. Contrary to the hypothesis, all the test strains within the P. polymyxa complex showed clear antagonistic activities against D. dadantii, and the most effective strain P. polymyxa ShX301 showed antagonism to all the tested Dickeya and Pectobacterium strains. We further studied the antibacterial profiles of P. polymyxa ShX301 and showed its application potential to control diseases in fields.

Bacterial Strains and Media
Three strains isolated from the rhizosphere soils of cotton plants [25] and six strains isolated from legume nodules [20] were recently screened out for their broad-spectrum antagonistic activities against phytopathogenic fungi. Dickeya dadantii strain CZ1501 is a causal agent of the bacterial stem, and the root disease of sweet potato occurred in Hangzhou, Zhejiang province, China. Its whole genome sequence (GenBank accession number MPDL00000000.1) contains the arnB, eptA, and dlt operons for resistance of polymyxins. It can grow in nutrient broth with 2 µg·mL −1 of polymyxin B sulfate. Six other Dickeya and two Pectobacterium strains were obtained from culture collections ( Table 1).
Bacterial strains were cultured in nutrient broth (10 g tryptone, 3 g beef extract, 2.5 g glucose, 5 g NaCl per liter, pH 7.0) or on nutrient agar (nutrient broth with 20 g agar per liter). Bacterial cultures in nutrient broth were washed with sterile water and adjusted to an optical density of 0.6 at 600 nm (OD600); the cell number of the suspension was counted by serial dilution and plating on nutrient agar. Bacterial suspensions were finally adjusted to the concentration of 1 × 10 8 CFU·mL −1 for use.

In Vitro and In Vivo Screening of Antagonistic Paenibacillus Strains
Paenibacillus strains against D. dadantii CZ1501 were screened using the in vitro overly culture assay on nutrient agar plates and the in vivo tuber slice assay as previously described [15]. Antagonistic activities of the P. polymyxa stain ShX301 were further tested on other Dickeya and Pectobacterium strains (Table 1) using the in vitro overlay culture assay on both the nutrient agar and the M9 minimal agar [15]. The in vitro inhibition rate (%) on target strains by Paenibacillus was calculated by [1-(diameter of Paenibacillus colony/diameter of inhibition zone)] × 100. The in vivo inhibition rate (%) was calculated by [1-(diameter of maceration zone by D. dadantii with Paenibacillus/diameter of maceration zone by D. dadantii without Paenibacillus] × 100.

In Vivo Biocontrol of D. dadantii by P. polymyxa ShX301
The biocontrol potential of the P. polymyxa stain ShX301 was determined using the in vivo assay with sweet potato seed tubers as previously described [15]. Surface-sterilized seed tubers were immersed in sterile distilled water (control), D. dadantii suspension (1 × 10 8 CFU·mL −1 ), P. polymyxa suspension (1 × 10 8 CFU·mL −1 ), or both D. dadantii and P. polymyxa at 1 × 10 8 CFU·mL −1 for 4 h at 25 • C and then kept under 28 • C, a photoperiod of 12-h light and 12-h dark, and 80% relative humidity for 21 d. CFCS effects on D. dadantii growth in wells of sterile polystyrene flat-bottom 96-well microplates, biofilm formation on the surface of the microplate wells, and swimming motility in 0.3% (w/v) agar were determined as previously described [15].
CFCS effects on D. dadantii cell integrity were observed by transmission electron microscopy on D. dadantii cells (1 × 10 8 CFU·mL −1 ) at mid-exponential phase grown with CFCS (50% volume) for 4 h. D. dadantii cells were washed twice with 0.1 mol·L −1 phosphate buffer (pH 7) and fixed in 2.5% (v/v) glutaraldehyde in the phosphate buffer at 4 • C overnight. After washing with the phosphate buffer, D. dadantii cells were fixed in 1% (w/v) OsO 4 dissolved in the phosphate buffer for 1 h at room temperature. D. dadantii cells were then washed with distilled water and dehydrated by a graded series of ethanol. Dehydrated D. dadantii cells were infiltrated by Spurr's resin at room temperature and embedded in Spurr's resin at 70 • C for 9 h. Ultrathin sections were cut with glass knives on an ultramicrotome (Reichert-Jung, Vienna, Austria), collected on copper grids, stained with uranyl acetate and lead citrate, and observed with the JEM-1230 transmission electron microscope (JEOL, Tokyo, Japan).

Detection of Lipopeptides Produced by P. polymyxa ShX301
Lipopeptides in the CFCS of P. polymyxa ShX301 were detected by matrix-assisted laser desorption/ionization coupled with time-of-flight mass spectrometry (MALDI-TOF MS) as previously described [15].

Statistical Analysis
Data were subjected to a one-way analysis of variance, and means were compared by Duncan's multiple range test using the SPSS software version 16 (SPSS, Chicago, IL, USA). The significance was set at p < 0.05.

Strains within the P. polymyxa Complex Inhibited D. dadantii Growth and Maceration of Sweet Potato Tuber Slices
All tested strains within the P. polymyxa complex inhibited D. dadantii growth in the nutrient agar and generated clearing zones around the Paenibacillus colonies. P. polymyxa ShX301 generated relatively larger clearing zones and showed a significantly higher rate (53%) of in vitro inhibition of D. dadantii growth than other Paenibacillus strains did.
D. dadantii CZ1501 degraded plant cell walls and generated maceration zones about 36 mm in diameter at 24 h after inoculation into the sweet potato tuber slices. All tested Paenibacillus strains did not macerate the tuber tissues but inhibited the maceration from D. dadantii. The in vivo inhibition of the maceration from D. dadantii was consistent with the in vitro inhibition of D. dadantii growth in the nutrient agar ( Table 2). P. polymyxa ShX301 showed the highest in vivo inhibition rate (80%) at the D. dadantii maceration of sweet potato tuber slices. Therefore, P. polymyxa ShX301 was the most effective strain against D. dadantii and was used for further analyses.  (Table 3). It showed more potent antagonistic activities on the M9 minimal agar than on the nutrient agar. * The letters a and b following the mean value ± standard error in the same row indicate a significant difference between the treatments at p < 0.05.

P. polymyxa ShX301 Protected Seed Tubers and Promoted Seedling Growth
Seed tubers inoculated with only D. dadantii did not germinate and were rotten; D. dadantii was isolated from the rotten seed tubers at 21 d after inoculation. In contrast, seed tubers inoculated with only P. polymyxa ShX301 germinated and grew seedlings significantly higher than the control seedlings. Seed tubers inoculated with D. dadantii along with P. polymyxa ShX301 germinated and grew seedlings slightly higher than the control seedlings ( Figure 1). D. dadantii was not isolated from the seedlings inoculated with P. polymyxa ShX301.

CFCS of P. polymyxa ShX301 Inhibited D. dadantii Growth, Biofilm Formation, and Swimming Motility
D. dadantii growth in nutrient broth in the microplate wells and biofilm formation on the surface of the microplate wells were significantly inhibited when the CFCS of P. polymyxa ShX301 was present at 50% volume (Figure 2A,B). D. dadantii CZ1501 swam via flagella in 0.3% (w/v) agar and formed haloes about 28 mm in diameter after 48 h. The swimming motility of D. dadantii was almost lost and formed haloes about 11 mm in diameter when the CFCS was present at 50% volume ( Figure 2C).

CFCS of P. polymyxa ShX301 Inhibited D. dadantii Growth, Biofilm Formation, and Swimming Motility
D. dadantii growth in nutrient broth in the microplate wells and biofilm formation on the surface of the microplate wells were significantly inhibited when the CFCS of P. polymyxa ShX301 was present at 50% volume (Figure 2a,b). D. dadantii CZ1501 swam via flagella in 0.3% (w/v) agar and formed haloes about 28 mm in diameter after 48 h. The swimming motility of D. dadantii was almost lost and formed haloes about 11 mm in diameter when the CFCS was present at 50% volume (Figure 2c).

CFCS of P. polymyxa ShX301 Breached D. dadantii Cells
Transmission electron microscopy showed that control D. dadantii cell envelopes were intact and enclosed electron-dense cytoplasm ( Figure 3A) while cell envelopes were convoluted ( Figure 3B) or beached and cytoplasm was clearing ( Figure 3C) when the CFCS of P. polymyxa ShX301 were present for 4 h. Under the CFCS of P. polymyxa ShX301, damage to D. dadantii cells was also indicated by the release of nucleic acids (128 µg·mL −1 ) and proteins (2.75 mg·mL −1 ) determined by the increases of the OD260 value (0.33) and OD280 value (0.31), respectively. oorganisms 2023, 11, x FOR PEER REVIEW 7 of Transmission electron microscopy showed that control D. dadantii cell envelop were intact and enclosed electron-dense cytoplasm (Figure 3a) while cell envelopes w convoluted (Figure 3b) or beached and cytoplasm was clearing (Figure 3c) when the CF of P. polymyxa ShX301 were present for 4 h. Under the CFCS of P. polymyxa ShX301, da age to D. dadantii cells was also indicated by the release of nucleic acids (128 μg·ml −1 ) a proteins (2.75 mg·ml −1 ) determined by the increases of the OD260 value (0.33) and OD2 value (0.31), respectively.  [26,27]. The mass peak m/z 1144-1225 was in the m/z range for polymyxins. T mass peak m/z 1101 may correspond to pelgipeptin B [28,29]. The mass peak m/z 1122 w close to m/z 1122. 6 [M+K] + of C17-iturin [30]. The mass peaks m/z 802-865, m/z 1559, a m/z 1753 were not identified.  (Figure 4). The mass peaks m/z 904-946 and m/z 984-1052 were in the m/z range for fusaricidins [26,27]. The mass peak m/z 1144-1225 was in the m/z range for polymyxins. The mass peak m/z 1101 may correspond to pelgipeptin B [28,29]. The mass peak m/z 1122 was close to m/z 1122. 6 [M+K] + of C17-iturin [30]. The mass peaks m/z 802-865, m/z 1559, and m/z 1753 were not identified.

Discussion
Contrary to the hypothesis that polymyxin-resistant Dickeya and Pectobacterium may resist the antagonism from polymyxin-producing strains within the P. polymyxa complex, all the test strains within the P. polymyxa complex, which have broad-spectrum antagonistic activities against fungal phytopathogens [20,25], showed clear antagonistic activities against the polymyxin-resistant D. dadantii strain CZ1501. Moreover, the most effective strain P. polymyxa ShX301 showed clear antagonistic activities against all the tested Dickeya and Pectobacterium strains. The extents of the in vitro inhibition of D. dadantii growth in nutrient agar and the in vivo inhibition of D. dadantii maceration in sweet potato tuber slices are consistent. P. polymyxa ShX301 showed more potent antibacterial activities against Dickeya and Pectobacterium strains on the nutrient-limited minimal medium than on the nutrient-rich medium. Thus, it may inhibit soft rot pathogens in nutrient-limited niches, such as soils, leaf surfaces, and seed tuber surfaces.
CFCS of P. polymyxa ShX301 inhibited D. dadantii growth, indicating that P. polymyxa released antibacterial compounds into the CFCS. The antibacterial compounds in the CFCS appeared to distort and breach D. dadantii cell envelopes, particularly plasma membranes, leading to the release of cell contents, such as nucleic acids and proteins, which were revealed by transmission electron microscopy and spectrophotometry. Disrupting plasma membranes indicates the involvement of amphiphilic lipopeptides in the antibacterial mechanisms of P. polymyxa.
MALDI-TOF-MS revealed that P. polymyxa ShX301 produced fusaricidins, polymyxins, and pelgipeptin-like and iturin-like lipopeptides. Fusaricidins have potent antimicrobial activities against fungi and Gram-positive bacteria but no or weak activity against Gram-negative bacteria [21,31] and thus may not contribute to the cytotoxic action on D. dadantii. Polymyxins consist of up to 30 closely related lipopeptides against Gram-negative bacteria [19], while Dickeya and Pectobacterium have evolved multiple pathways (arnB, eptA, and dlt operons) to resist polymyxins [24]. However, polymyxin P was found to inhibit the growth of Pectobacterium carotovorum (formerly Erwinia carotovora) [32]. Here, although D. dadantii CZ1501 resists 2 μg·ml −1 of polymyxin B, multiple polymyxins produced by P. polymyxa ShX301 may accumulate to a locally high concentration, contributing to the

Discussion
Contrary to the hypothesis that polymyxin-resistant Dickeya and Pectobacterium may resist the antagonism from polymyxin-producing strains within the P. polymyxa complex, all the test strains within the P. polymyxa complex, which have broad-spectrum antagonistic activities against fungal phytopathogens [20,25], showed clear antagonistic activities against the polymyxin-resistant D. dadantii strain CZ1501. Moreover, the most effective strain P. polymyxa ShX301 showed clear antagonistic activities against all the tested Dickeya and Pectobacterium strains. The extents of the in vitro inhibition of D. dadantii growth in nutrient agar and the in vivo inhibition of D. dadantii maceration in sweet potato tuber slices are consistent. P. polymyxa ShX301 showed more potent antibacterial activities against Dickeya and Pectobacterium strains on the nutrient-limited minimal medium than on the nutrientrich medium. Thus, it may inhibit soft rot pathogens in nutrient-limited niches, such as soils, leaf surfaces, and seed tuber surfaces.
CFCS of P. polymyxa ShX301 inhibited D. dadantii growth, indicating that P. polymyxa released antibacterial compounds into the CFCS. The antibacterial compounds in the CFCS appeared to distort and breach D. dadantii cell envelopes, particularly plasma membranes, leading to the release of cell contents, such as nucleic acids and proteins, which were revealed by transmission electron microscopy and spectrophotometry. Disrupting plasma membranes indicates the involvement of amphiphilic lipopeptides in the antibacterial mechanisms of P. polymyxa.
MALDI-TOF-MS revealed that P. polymyxa ShX301 produced fusaricidins, polymyxins, and pelgipeptin-like and iturin-like lipopeptides. Fusaricidins have potent antimicrobial activities against fungi and Gram-positive bacteria but no or weak activity against Gramnegative bacteria [21,31] and thus may not contribute to the cytotoxic action on D. dadantii. Polymyxins consist of up to 30 closely related lipopeptides against Gram-negative bacteria [19], while Dickeya and Pectobacterium have evolved multiple pathways (arnB, eptA, and dlt operons) to resist polymyxins [24]. However, polymyxin P was found to inhibit the growth of Pectobacterium carotovorum (formerly Erwinia carotovora) [32]. Here, although D. dadantii CZ1501 resists 2 µg·mL −1 of polymyxin B, multiple polymyxins produced by P. polymyxa ShX301 may accumulate to a locally high concentration, contributing to the cytotoxic action on the damage of D. dadantii plasma membrane and the inhibition of D.
dadantii growth in vitro, in vivo, and in planta. Pelgipeptins produced by Paenibacillus have broad-spectrum activities against Gram-positive and Gram-negative bacteria and fungi, including phytopathogens [28,29,33,34]. Iturins, which are produced by Bacillus and have potent antifungal activity and limited antibacterial activity [21], have also been found to inhibit the growth of phytopathogenic bacteria, including P. carotovorum [35]. The pelgipeptin-like and iturin-like lipopeptides from P. polymyxa ShX301 may also contribute to the cytotoxic action on D. dadantii.
Flagella-mediated motility and biofilm formation facilitate D. dadantii to colonize plant surface, intercellular spaces, and xylem vessels and complete disease cycles [36][37][38][39]. CFCS of P. polymyxa ShX301 inhibited D. dadantii swimming motility and biofilm formation. Fusaricidins and polymyxins from P. polymyxa may not play a major role in cytotoxic action on D. dadantii but may act as biosurfactants to inhibit surface attachment and biofilm formation [40], thus reducing D. dadantii infection and increasing D. dadantii susceptibility to cytotoxic lipopeptides. Multiple lipopeptides produced by P. polymyxa ShX301 play a major role in bactericidal and bacteriostatic actions on the phytopathogens.
From the view of plant-microbe interactions, plants can induce the expression of polymyxin and fusaricidin biosynthesis genes in P. polymyxa [41], while lipopeptides such as fusaricidins can induce plant systemic resistance to fungal and bacterial pathogens [42,43]. Notably, P. polymyxa ShX301 completely eliminated D. dadantii from sweet potato seed tubers at 21 d after the equivalent inoculation of the two bacteria. Bactericidal and bacteriostatic actions and induced plant systemic resistance mediated by multiple lipopeptides may play together to eliminate D. dadantii. The latent infection of seed tubers by Dickeya and Pectobacterium is a major source of the soft rot diseases of potato and sweet potato plants [2,44]. P. polymyxa ShX301 has shown the potential to control soft rot diseases from seed tuber-borne pathogens. Moreover, inoculating P. polymyxa ShX301 alone to sweet potato seed tubers promoted the growth of sweet potato seedlings. Likewise, P. polymyxa ShX301 promoted the growth of cotton seedlings and suppressed the soil-borne fungal pathogen Verticillium dahlia and the Verticillium wilt disease of cotton seedlings [25]. Therefore, P. polymyxa ShX301 is a promising biocontrol agent and plant growth promoter for future application in fields and study of the underlying mechanisms of the broad-spectrum antagonism and mutually beneficial interactions with plants.

Conclusions
This study clarifies that the antimicrobial spectrum of polymyxin-producing bacteria within the P. polymyxa complex includes the polymyxin-resistant soft rot Dickeya and Pectobacterium pathogens and strengthens the fact that bacteria within the P. polymyxa complex have high probability of being effective biocontrol agents and plant growth promoters. P. polymyxa ShX301 will be used to clarify the biocontrol mechanism of the P. polymyxa complex against broad-spectrum phytopathogens, including polymyxin-resistant Dickeya and Pectobacterium and to control plant diseases in fields. Data Availability Statement: Not applicable.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.