Negatively Regulated Aerobactin and Desferrioxamine E by Fur in Pantoea ananatis Are Required for Full Siderophore Production and Antibacterial Activity, but Not for Virulence

ABSTRACT

Pantoea ananatis is an emerging plant pathogen that causes disease in economically important crops such as rice, corn, onion, melon, and pineapple, and it also infects humans and insects. In this study, we identified biosynthetic gene clusters of aerobactin and desferrioxamine E (DFO-E) siderophores by using the complete genome of P. ananatis PA13 isolated from rice sheath rot. P. ananatis PA13 exhibited the strongest antibacterial activity against Erwinia amylovora and Yersinia enterocolitica (Enterobacterales). Mutants of aerobactin or DFO-E maintained antibacterial activity against E. amylovora and Y. enterocolitica, as well as in a siderophore activity assay. However, double aerobactin and DFO-E gene deletion mutants completely lost siderophore and antibacterial activity. These results reveal that both siderophore biosynthetic gene clusters are essential for siderophore production and antibacterial activity in P. ananatis PA13. A ferric uptake regulator protein (Fur) mutant exhibited a significant increase in siderophore production, and a Fur-overexpressing strain completely lost antibacterial activity. Expression of the iucA, dfoJ, and foxA genes was significantly increased in the Δfur mutant background, and expression of these genes returned to wild-type levels after fur compensation. These results indicate that Fur negatively regulates aerobactin and DFO-E siderophores. However, siderophore production was not required for P. ananatis virulence in plants, but it appears to be involved in the microbial ecology surrounding the plant environment. This study is the first to report the regulation and functional characteristics of siderophore biosynthetic genes in P. ananatis.

IMPORTANCE Pantoea ananatis is a bacterium that causes diseases in several economically important crops, as well as in insects and humans. This bacterium has been studied extensively as a potentially dangerous pathogen due to its saprophytic ability. Recently, the types, biosynthetic gene clusters, and origin of the siderophores in the Pantoea genus were determined by using genome comparative analyses. However, few genetic studies have investigated the characteristics and functions of siderophores in P. ananatis. The results of this study revealed that the production of aerobactin and desferrioxamine E in the rice pathogen P. ananatis PA13 is negatively regulated by Fur and that these siderophores are essential for antibacterial activity against Erwinia amylovora and Yersinia enterocolitica (Enterobacterales). However, siderophore production was not required for P. ananatis virulence in plants, but it appears to be involved in the microbial ecology surrounding the plant environment.

INTRODUCTION

Iron is an essential nutrient for nearly all forms of life, including pathogenic organisms, their hosts, and pathogenic microbes surrounding the hosts (1, 2). However, the bioavailability of iron is very limited, making it a crucial factor in the competition between host and pathogenic bacteria (3, 4). To acquire ferric iron under low-iron conditions, most bacteria produce and secrete high-affinity iron-binding chelators known as siderophores (5).

To date, many types of siderophores have been reported, and they are classified into five major groups based on the moiety that confers iron-binding capacity: catecholates, hydroxamates, phenolates, carboxylates, and mixed (6). Aerobactin is a mixed siderophore that is an important factor in many virulent Escherichia coli strains and Enterobacteriaceae. Aerobactin was originally isolated from Aerobacter aerogenes (7). Genes encoding aerobactin biosynthesis and uptake are clustered as an operon containing five genes (iucABCD and iutA). Four genes (iucABCD) encode aerobactin biosynthetic proteins, while iutA encodes a high-affinity outer membrane receptor (8). Genetic studies have established the function of iucABCD enzymes for the aerobactin biosynthesis pathway over the past decade (9). Desferrioxamines (DFOs) are siderophores of hydroxamate groups, which were first isolated from Streptomyces sp. DFO-B and DFO-E are produced by Streptomyces coelicolor (10). DFO-B has been commercialized as Desferal to treat iron overload and neuroblastoma in humans (11, 12). The biosynthetic gene clusters and putative biosynthetic pathways of DFO have been investigated in previous studies (13, 14). DFO has also been found in Enterobacterales, including Pantoea agglomerans, Pseudomonas stutzeri, and Erwinia amylovora (15–17).

Recent studies have conducted genome comparative analysis of the genus Pantoea to identify siderophore types, biosynthetic clusters, and their genetic origins. For example, aerobactin was common to Pantoea stewartii, Pantoea ananatis, and Pantoea septica and was horizontally acquired from human-pathogenic Enterobacterales members (18). DFO-E was confirmed in more species of genus Pantoea and was found to be horizontally acquired from its close relative Erwinia (14, 19). However, few reports have experimentally demonstrated the genetic and functional roles of siderophores in Pantoea, except for aerobactin in P. stewartii (20).

Many species of phytopathogenic bacteria secrete siderophores for iron acquisition, often in competition with the host plant. In some phytopathogenic bacteria, siderophores have been reported to play an important role in virulence. In Erwinia chrysanthemi strain 3937, which causes soft rot disease in various plants, two types of siderophores, chrysobactin and achromobactin, are required for disease development (19). E. amylovora causes fire blight in apple and produces DFO, which is required for effective infection (16). The phytopathogen Pantoea stewartii, which causes Stewart’s wilt in susceptible corn, produces aerobactin, which is essential for virulence and systemic colonization of the host (20). In Xanthomonas campestris pv. campestris, which causes black rot disease in crucifers, the siderophore xanthoferrin is an important virulence factor (21). However, some phytopathogens do not require siderophores for virulence. In Pseudomonas syringae pv. syringae B301D, which causes bacterial diseases in stone fruit trees, and P. syringae pv. tomato DC3000, which causes bacterial speck disease in tomatoes, siderophore production is unnecessary for virulence within the host (22, 23). Siderophore-deficient mutants of Ralstonia solanacearum (24), Agrobacterium tumefaciens (25), and Pectobacterium carotovorum subsp. carotovorum (26) were found to retain full virulence, indicating considerable variation in the roles of siderophores in the virulence of plant-pathogenic bacteria.

Pantoea ananatis is an emerging plant pathogen that causes diseases in economically important crops such as rice, corn, onion, melon, and pineapple (27), and it also infects humans and insects (28, 29). P. ananatis PA13 causes rice grain rot, sheath rot, and onion center rot diseases in Korea (30, 31). In previous work, we found that virulence, exopolysaccharide production, and hypersensitive responses are regulated by quorum sensing in P. ananatis PA13 (32) and demonstrated that carotenoid production by P. ananatis PA13 depends on RpoS, which is positively regulated by Hfq-controlled quorum sensing (33). Previous studies have reported that siderophore-mediated iron acquisition plays an important role in virulence factors such as motility and biofilm formation (18, 20, 34). Recently, a genome-wide survey of Pantoea strains was conducted to identify known and candidate siderophore biosynthetic clusters using comparative genomics (14, 35). However, the role of siderophores in P. ananatis remains unknown. In the present study, we examined the biosynthetic gene clusters of aerobactin and DFO-E, as well as their regulation in P. ananatis PA13.

RESULTS

Identification of genes responsible for siderophore production in P. ananatis PA13.

The three siderophore biosynthetic clusters in environmental and host-associating strains of Pantoea were previously analyzed based on a genome-wide survey (35, 36). Using a scan of the genome sequences (GenBank accession no. CP003085) to identify siderophore biosynthetic clusters, we found that P. ananatis PA13 possesses gene clusters comprising two types of siderophores: aerobactin and DFO-E. The genetic orientations of the aerobactin and DFO-E biosynthetic gene clusters are shown in Fig. 1. The aerobactin biosynthetic gene cluster was an 8.0-kb DNA region comprising five open reading frames: iucABCD and iutA. Comparative analyses of the iucABCD-iutA gene cluster among P. ananatis strains revealed that iucABCD-iutA are identical to the aerobactin biosynthetic genes of P. ananatis PNA97-1R, which causes onion center rot (GenBank accession no. CP020943) (37). The iucABCD-iutA operon has been reported in many pathogenic E. coli strains and Pantoea species (20, 38). The DFO-E biosynthetic gene cluster is an 8.7-kb DNA region comprising five open reading frames: dfoJACS and foxA. The dfoJACS and foxA genes exhibit high similarity to those of P. ananatis LMG20103, which causes eucalyptus blight (GenBank accession no. CP001875) (39). The proteins DfoJ, DfoA, DfoC, DfoS, and FoxA are highly similar to l-2,4-diaminobutyrate decarboxylase Ddc (identity, 99%; similarity, 99%), the alcaligin biosynthesis protein AlcA (identity, 99%; similarity, 100%), the rhizobactin siderophore biosynthesis protein RhbF (identity, 99%; similarity, 99%), the multidrug translocase Cmr (identity, 99%; similarity, 100%), and the ferrioxamine receptor precursor FoxA (identity, 99%; similarity, 100%), respectively, in P. ananatis LMG20103. The DFO-E biosynthetic gene cluster was genetically close to those in the genera Erwinia and Pantoea (14). The Fur box consensus sequence (TGATAAT) was identified in the promoter regions of the iucA, dfoJ, and foxA genes. These results suggest that the iucABCD-iutA and dfoJACS-foxA operons may be regulated by Fur in response to intracellular iron levels.

FIG 1.

Genetic organization of the Pantoea ananatis PA13 loci responsible for iron acquisition and regulation. Arrows indicate the positions and orientations of the genes responsible for siderophore production and regulation. (A) P. ananatis iucABCD-iutA gene cluster involved in aerobactin biosynthesis and transport. (B) P. ananatis dfoJACS-foxA gene cluster involved in desferrioxamine E (DFO-E) biosynthesis and transport. (C) P. ananatis fur gene encoding a ferric uptake regulator protein. Vertical bars with arrows in the map indicate the position and orientation of lacZY insertion. The Fur-binding sequence (TGATAAT) is indicated in the promoter regions of the iucA, dfoJ, and foxA open reading frames. Gene numbers are indicated below the siderophore gene map.

Double-operon gene mutations in the iucABCD-iutA and dfoJACS-foxA operons caused complete loss of antibacterial activity.

To determine the requirements of the iucABCD-iutA and dfoJACS-foxA operon genes for siderophore production, we constructed the ΔiucABC aerobactin biosynthetic gene operon mutant containing an in-frame deletion, the ΔdfoA, ΔdfoC, and ΔdfoAC DFO-E biosynthetic gene operon mutants, and the ΔiucABC ΔdfoA, ΔiucABC ΔdfoC, and ΔiucABC ΔdfoAC double-operon gene mutants. Siderophore production can be assessed on antibacterial activity assay plates, with visually distinct results. We performed an antibacterial activity assay against the phytopathogenic and food poisoning bacteria listed in Table 1. P. ananatis PA13 exhibited the strongest antibacterial activity against E. amylovora and Y. enterocolitica (Enterobacterales) (Table 1). Compared with the wild type, the ΔiucABC mutant exhibited a slight reduction in antibacterial activity, whereas the ΔdfoA, ΔdfoC, and ΔdfoAC mutants exhibited similar antibacterial activities against both E. amylovora and Y. enterocolitica. However, the ΔiucABC ΔdfoA, ΔiucABC ΔdfoC, and ΔiucABC ΔdfoAC double-operon gene mutants showed a complete loss of antibacterial activity (Fig. 2). These data indicate that both aerobactin and DFO-E are required for full antibacterial activity in P. ananatis PA13.

TABLE 1.

Antibacterial activity of P. ananatis PA13^a

Bacterial strain	Inhibition diam (mm)
Acidovorax citrulli AAC02	0.0
Agrobacterium tumefaciens C58	13.0 ± 0.5
Erwinia amylovora ATCC 19383	39.5 ± 0.8
Pantoea ananatis HY02	11.8 ± 0.2
Pectobacterium carotovorum subsp. carotovorum KACC 10057	11.3 ± 0.2
Pseudomonas aeruginosa PAO1	0.0
Xanthomonas fragariae XF1	21.0 ± 0.5
Yersinia enterocolitica ATCC 9610	22.5 ± 1.0

^aCultures were grown at 28°C. Values are the mean ± SD of three replicates.

FIG 2.

Antibacterial activity of P. ananatis wild-type PA13 and single-operon and double-operon gene mutants of aerobactin and DFO-E biosynthetic operons against Erwinia amylovora ATCC 19383 and Yersinia enterocolitica ATCC 9610 on agrobacterium (AB) minimal agar medium. The graph shows the mean halo diameter around bacterial colonies. Values are means ± standard deviation (SD) of results of three independent experiments. The plates were incubated for 24 h at 28°C and photographed.

Antibacterial activity-deficient mutants show impaired siderophore production.

Next, we evaluated siderophore production on King’s B and chrome azurol S (King’s B-CAS) agar indicator plates. The positive strain Pseudomonas fluorescens Pf-5 produced a strong yellow halo on King’s B-CAS; however, the wild-type PA13 also produced a yellow halo, which was weaker than that of Pf-5. The ΔdfoA, ΔdfoC, and ΔdfoAC mutants produced yellow haloes similar to that of the wild type. The ΔiucABC aerobactin gene mutant showed a slight change in halo color from blue to yellow compared with that of the wild type. No halo was observed in the ΔiucABC ΔdfoA, ΔiucABC ΔdfoC, and ΔiucABC ΔdfoAC double-operon mutants (Fig. 3A). Because the color changes visualized in King's B-CAS analysis were subtle, we attempted to control image variability by quantifying these color changes (Fig. 3B). Relative siderophore production can be quantified as the relative value of the removed blue pixels. The relative siderophore production levels of the single-operon gene mutants except for the iucABC mutant were not significantly different from that of the wild type. The iucABC mutant showed about 3-fold-lower relative siderophore production than the wild type (Fig. 3B). The relative nonproduction of siderophores of the double-operon gene mutants was at a level similar to that of the bacterial colony-free region (Fig. 3B). These results were consistent with the visual images of siderophores in Fig. 3A. These results are consistent with those of antibacterial activity assays. Together, these findings indicate that both aerobactin and DFO-E are required for full siderophore production and antibacterial activity in P. ananatis PA13.

FIG 3.

Siderophore production of P. ananatis wild-type PA13 and single-operon and double-operon gene mutants of aerobactin and DFO-E biosynthetic operons on a chrome azurol S (CAS) indicator plate. (A) Yellow halos around bacterial colonies indicate siderophore production. The plates were incubated for 2 days at 28°C. Pseudomonas fluorescens Pf-5 used as a positive control strain overproduced the yellow halo. The photograph shows the plate after bacteria were removed but with colony margins indicated by dotted lines. (B) Relative siderophore production. The average value of the blue pixels in the circled area indicated by the dotted line in panel A was obtained, and using the Pf-5 value as a reference value, the average value of the blue pixels of each strain was relatively displayed. Values are averages of results of triplicate assays; error bars represent the range. Asterisks denote significant differences from the wild-type PA13 (***, P < 0.001; Student's t test).

Siderophore-deficient mutants retain virulence on rice and onion.

To determine the role of aerobactin and DFO-E siderophores in the virulence of P. ananatis PA13, we inoculated wild-type PA13 and the ΔiucABC, ΔdfoA, ΔdfoC, ΔiucABC ΔdfoA, ΔiucABC ΔdfoC, ΔdfoA ΔdfoC, and ΔiucABC ΔdfoAC siderophore mutants into onion bulbs and rice seedlings. The wild type caused severe soft rot in onion bulbs and sheath rot in rice at 15 and 18 days after inoculation, respectively. Interestingly, all aerobactin and DFO-E gene mutants exhibited severe rot symptoms on onion bulbs and rice sheaths, respectively, with no differences from the wild type (Fig. 4). These results indicate that aerobactin and DFO-E siderophores do not play a role in P. ananatis PA13 virulence.

FIG 4.

Virulence of P. ananatis wild-type PA13 and single-operon and double-operon gene mutants of aerobactin and DFO-E biosynthetic operons on onion bulbs and rice seedlings. Bacterial suspensions (OD₆₀₀ = 0.8) were inoculated into onion bulbs and rice sheaths using a syringe. Water was used as a negative control. Rice seedlings and onion bulbs were observed for disease symptoms for 18 and 15 days, respectively, after inoculation and photographed. The experiment was repeated at least three times independently.

The role of Fur in siderophore production and antibacterial activity.

To investigate the role of Fur in siderophore production and antibacterial activity, we constructed an in-frame deletion mutant of the fur gene. We complemented the Δfur mutant with a plasmid harboring P_lac-fur. The yellow halo of the Δfur mutant was significantly increased on King’s B-CAS compared with that of the wild type. The fur-complemented strain Δfur(fur⁺) with a plasmid harboring P_lac-fur showed a diminished yellow halo similar to that of the wild type (Fig. 5A). These results indicate that increased siderophore production was caused by the absence of the fur gene.

FIG 5.

Effects of Fur on siderophore production and antibacterial activity in P. ananatis. (A) Effect of Fur on siderophore production in P. ananatis wild-type PA13, Δfur mutant, and fur complementation strain [Δfur(fur⁺)]. Siderophore production was visualized and measured as the presence of a halo around bacterial colonies on CAS medium. The plates were incubated for 2 days at 28°C. (B) Effect of Fur on antibacterial activity in P. ananatis wild-type PA13 and fur-overexpressing strain (fur⁺). Antibacterial activity was visualized and measured as the diameter of the inhibition zone around bacterial colonies on AB minimal agar medium. The plates were incubated at 28°C for 24 h and photographed. The experiment was repeated at least three times independently.

As the Δfur mutant barely grew on minimal medium, antibacterial activity could not be evaluated. Therefore, to determine the role of Fur in antibacterial activity, we constructed a Fur overexpression strain harboring P_lac-fur in the wild-type background. The Fur overexpression strain completely lost its antibacterial activity against E. amylovora and Y. enterocolitica (Fig. 5B). These results indicate that Fur in P. ananatis negatively regulates siderophore production and antibacterial activity.

Fur represses the expression of the iucABCD-iutA and dfoJACS-foxA operons.

The presence of the Fur box in the promoter regions of iucA, dfoJ, and foxA suggests that the expression of these genes may be under the control of Fur (Fig. 1). To determine whether iucA, dfoJ, and foxA gene expression is under the control of the Fur protein, we constructed Campbell insertions iucA::pCOK194, dfoJ::pBS148, and foxA::pBS147 in both the PA13L and PA13LΔfur mutant backgrounds (Fig. 1). The expression of the iucA, dfoJ, and foxA genes determined using a β-galactosidase activity assay was significantly increased in the Δfur mutant background (Fig. 6) and returned to wild-type levels after fur complementation (fur⁺) (Fig. 6). These results indicate that Fur negatively regulates iucA, dfoJ, and foxA expression.

FIG 6.

Effects of Fur on the expression of iucA, dfoJ, and foxA in P. ananatis. Expression of iucA, dfoJ, and foxA in the backgrounds of the wild type, Δfur mutant, and fur complementation strain (fur⁺). Values are means ± standard deviation (SD) of results of three independent experiments.

DISCUSSION

P. ananatis is a bacterial pathogen that causes diseases in various plants, as well as in insects and animals. This bacterium is increasingly being studied as a potential pathogen due to its saprophytic ability (27). However, few studies have reported on the virulence factors of this bacterium as a phytopathogen. Many phytopathogenic bacteria synthesize siderophores as iron-chelating compounds to take up iron in environments where supply is limited. Several studies have reported that siderophores play important roles in virulence (16, 19–21). In the present study, we investigated the function of siderophores in competition with other bacteria and plant pathogenicity in P. ananatis PA13.

Experiments using in-frame deletion mutants of siderophores ostensibly revealed that aerobactin in PA13 is central to siderophore activity, whereas DFO-E is secondary. However, in this study, siderophores were measured using a CAS assay, and considering the difference in the effect of siderophore structures according to bacteria, it cannot be concluded that aerobactin is central and DFO-E is secondary. It could be an interesting follow-up study to further investigate why bacteria make these multiple siderophores. When both siderophores were abolished, siderophore production and antibacterial activity were completely eliminated (Fig. 2 and Fig. 3). We reasoned that the antibacterial activity of PA13 results from the growth inhibitory effect of competitive bacteria by effectively preempting the surrounding iron by producing two types of siderophores under iron-poor environmental conditions. Antibacterial activity against competing bacteria can be explained in two ways. Bacteria can be killed directly by the production of antibiotics and indirectly by inhibition of the growth of competing bacteria by preempting limited resources (40, 41). The best example of a limited resource for bacteria is iron, which is an essential nutrient for their growth and survival (42). Numerous studies have demonstrated that various animal-pathogenic, phytopathogenic, and biocontrol bacteria benefit from inhibiting the growth of competing bacteria via the antibacterial activity of siderophores under limited iron conditions (43–46). We propose that PA13 also inhibits the growth of E. amylovora and Y. enterocolitica by producing two types of siderophores and scavenging the surrounding iron, suggesting favorable competition with the host.

Siderophores are key virulence factors in many animal-pathogenic bacteria (47). In phytopathogenic bacteria, some siderophores act as virulence factors, whereas others are unrelated to pathogenicity. In Agrobacterium tumefaciens C58, a siderophore-deficient mutant formed tumors in potato and tomato seedlings, suggesting that iron-rich conditions within the plant allowed C58 to form tumors in the absence of siderophore production. Alternatively, C58 may still cause disease by directly pirating the iron chelator produced by the plant (25). Siderophore-deficient mutants of Ralstonia solanacearum and P. carotovorum subsp. carotovorum remain pathogenic for these reasons (24, 26). In P. syringae pv. tomato DC3000, a mutant unable to produce all three of its natural siderophores caused disease in tomatoes. This finding suggests that siderophore-deficient mutants can pirate plant-produced iron chelators such as hemo/hemin or iron-nicotianamine. An alternative explanation is that siderophore production is not essential for pathogenesis because bacteria can cause disease in iron-replete environments (23). We also found that the siderophore-deficient mutant of PA13 caused the disease for these reasons. Siderophore-deficient mutants of PA13 still caused disease, suggesting that the host plant represents an iron-replete environment and that siderophore production is not required for disease development. Thus, it appears that siderophore-deficient mutants use trace amounts of iron contained in plant tissues or phytosiderophores, allowing the mutants to overcome iron deficiency (24, 25). In these cases, siderophores do not appear to affect pathogenicity but may play an important role in survival and competition with other bacteria in iron-deficient plant environments (48–50).

Although iron is an essential nutrient for bacterial growth and survival, it can also be toxic at excessive cellular concentrations. Therefore, iron metabolism in bacteria is tightly regulated according to environmental conditions (51–53). The Fur protein is an important global regulator that maintains intracellular iron homeostasis (54). In the presence of ferrous ions, the Fur-ferrous ion complex binds to the Fur box, a conserved DNA sequence in target gene promoters, and regulates many genes (55). Fur results in the transcriptional repression of iron acquisition and storage-related genes (56). In particular, Fur negatively regulates the biosynthesis of siderophores (52, 53). In this study, siderophore production was significantly increased in a PA13 Δfur mutant, and the transcriptional expression of the iucA, dfoJ, and foxA genes was also significantly increased (Fig. 5A and Fig. 6). These results demonstrate that Fur in PA13 negatively regulates the synthesis and production of siderophores. This result is consistent with those of fur deletion mutants in many other bacteria (52, 53). In addition, the Fur-overexpressing strain completely lost antibacterial activity, indicating that Fur is directly involved in antibacterial activity via siderophore regulation (Fig. 5B). Taken together, we proposed a schematic regulation of the production of aerobactin and DFO-E in P. ananatis PA13 under iron-depleted and -replete conditions (Fig. 7). When iron is depleted, Fur without iron (Apo-Fur) does not bind to the Fur-binding site in the promoter regions of iucA, dfoJ, and foxA, thereby allowing access of RNA polymerase (RNAP) to the promoter regions to initiate transcription (Fig. 7A). The redox state of iron in the surrounding environment is likely dominated by ferric iron (Fe³⁺). Ferric-siderophore complexes are internalized via specific uptake systems composed of outer membrane receptors, a periplasmic binding protein, and inner membrane transporters (57). Once inside the bacterial cell, the ferric-siderophore complexes are dissociated by reduction (57). Under iron-replete conditions, Fe²⁺-Fur complex binds to the Fur-binding site in the promoter regions of iucA, dfoJ, and foxA, disrupting RNAP transcription, resulting in no siderophore biosynthesis (Fig. 7B).

FIG 7.

Schematic representation of ferric uptake regulator (Fur)-mediated iron uptake regulation via aerobactin and DFO-E in P. ananatis PA13 under iron-depleted (A) or -replete (B) conditions. In iron-depleted conditions, Fur without iron (Apo-Fur) does not bind to the Fur-binding site in the promoter region of the siderophore gene operons, allowing RNA polymerase (RNAP) to initiate transcription of siderophore genes and resulting in higher siderophore production. The redox state of iron in the surrounding environment is likely dominated by ferric iron (Fe³⁺). Ferric-siderophore complexes are internalized via specific uptake systems composed of outer membrane receptors, a periplasmic binding protein, and inner membrane transporters. Once inside the bacterial cell, the ferric-siderophore complexes are dissociated by reduction. In the presence of iron, Fe²⁺-Fur complex, which binds to the Fur-binding sites in the promoter region of the siderophore gene operons, blocks RNAP transcription.

In summary, we demonstrated that P. ananatis PA13 produced aerobactin and DFO-E siderophores and that these siderophores exhibited direct antibacterial activity. We also revealed that Fur negatively regulates the expression of siderophore biosynthesis genes. However, siderophore production in PA13 did not affect plant pathogenicity. This study is the first to report the biosynthesis, regulation, and functional characterization of siderophores in the plant pathogen P. ananatis. The results of this study provide essential data for controlling competition among various bacteria for iron acquisition in P. ananatis.

MATERIALS AND METHODS

Bacterial strains and growth conditions.

The bacterial strains and plasmids used in this study are listed in Table 2. E. coli strains were cultured on Luria-Bertani (LB) medium at 37°C. Then, P. ananatis PA13 was cultivated at 28°C on LB medium or agrobacterium (AB) minimal medium supplemented with 0.2% glucose. Antibiotics were used at the following concentrations: ampicillin, 100 μg/mL; kanamycin, 50 μg/mL; rifampin, 50 μg/mL; tetracycline, 10 μg/mL; gentamicin, 25 μg/mL. We used 40 μg/mL 5-bromo-4-chloro-3-indoyl-β-d-galactopyranoside (X-Gal) when necessary.

TABLE 2.

Bacterial strains and plasmids used in this study

Strain/genotype or plasmid	Relevant genotypea	Reference or source
Strains
Acidovorax citrulli AAC02	Wild type	66
Agrobacterium tumefaciens C58	Wild type, nopaline-type, genomospecies G8	67
Erwinia amylovora ATCC 19383	Wild type	63
Escherichia coli
DH5α λpir	Cloning host	Gibco BRL
S17-1 λpir	IncP conjugal donor	Promega
Pantoea ananatis
HY02	Wild type	31
PA13	Wild type, Rifr	30
PA13L	PA13 in-frame deletion of lacZY	33
ΔiucABC	PA13 in-frame deletion of iucABC, Rifr	This study
ΔdfoA	PA13 in-frame deletion of dfoA, Rifr	This study
ΔdfoC	PA13 in-frame deletion of dfoC, Rifr	This study
ΔdfoA ΔdfoC	PA13 in-frame deletions of dfoA and dfoC, Rifr	This study
ΔiucABC ΔdfoA	PA13 in-frame deletions of iucABC and dfoA, Rifr	This study
ΔiucABC ΔdfoC	PA13 in-frame deletions of iucABC and dfoC, Rifr	This study
ΔiucABC ΔdfoA ΔdfoC	PA13 in-frame deletions of iucABC, dfoA and dfoC, Rifr	This study
Δfur	PA13 in-frame deletion of fur, Rifr	This study
iucA::pCOK194	PA13L iucA-lacZY transcriptional fusion, Kmr	This study
dfoJ:pBS148	PA13L dfoJ-lacZY transcriptional fusion, Kmr	This study
foxA::pBS147	PA13L foxA-lacZY transcriptional fusion, Kmr	This study
iucA::pCOK194 Δfur	PA13L Δfur iucA-lacZY transcriptional fusion, Kmr	This study
dfoJ:pBS148 Δfur	PA13L Δfur dfoJ-lacZY transcriptional fusion, Kmr	This study
foxA::pBS147 Δfur	PA13L Δfur foxA-lacZY transcriptional fusion, Kmr	This study
Pectobacterium carotovorum subsp. carotovorum KACC 10057	Wild type	64
Pseudomonas aeruginosa PAO1	Wild type	64
Xanthomonas fragariae XF1	Wild type	68
Yersinia enterocolitica ATCC 9610	Wild type	64
Plasmids
pGEM-T Easy	PCR cloning vector, Ampr	Promega
pVIK112	R6K suicide vector, lacZY for transcriptional fusion, Kmr	60
pNPTS138-R6KT	R6K sacB; suicide plasmid for in-frame deletions, Kmr	61
pBBR1MCS5	pBBR1MCS-5-derived broad-host-range expression vector, Gmr	62
pCOK190	pGEM-T Easy, iucA internal fragment, Ampr	This study
pBS145	pGEM-T Easy, foxA internal fragment, Ampr	This study
pBS146	pGEM-T Easy, dfoJ internal fragment, Ampr	This study
pJY18	pGEM-T Easy, fur deletion fragment, Ampr	This study
pJY61	pGEM-T Easy, iucABC deletion fragment, Ampr	This study
pJY62	pGEM-T Easy, dfoC deletion fragment, Ampr	This study
pJY72	pGEM-T Easy, dfoA deletion fragment, Ampr	This study
pJY73	pGEM-T Easy, dfoA and dfoC deletion fragment, Ampr	This study
pCOK194	pVIK112, iucA internal fragment, transcriptional fusion, Kmr	This study
pBS147	pVIK112, foxA internal fragment, transcriptional fusion, Kmr	This study
pBS148	pVIK112, dfoJ internal fragment, transcriptional fusion, Kmr	This study
pJY19	pNPTS138-R6KT, fur deletion fragment, Kmr	This study
pJY63	pNPTS138-R6KT, iucABC deletion fragment, Kmr	This study
pJY64	pNPTS138-R6KT, dfoC deletion fragment, Kmr	This study
pJY80	pNPTS138-R6KT, dfoA deletion fragment, Kmr	This study
pJY85	pNPTS138-R6KT, dfoA and dfoC deletion fragment, Kmr	This study
pLY291	pBBR1MCS5::Plac-fur; fur complementation	This study

^aAmp^r, ampicillin resistance; Gm^r, gentamicin resistance; Km^r, kanamycin resistance; Rif^r, rifampicin resistance.

DNA manipulation and data analyses.

DNA manipulation, cloning, restriction enzyme digestion, and agarose gel electrophoresis were performed as described previously (58). DNA sequencing was performed by Macrogen (Seoul, Republic of Korea). DNA sequences were analyzed using the BLAST program at the National Center for Biotechnology Information (59) and MegAlign (DNASTAR, Madison, WI, USA) and Genetyx-Win software (Genetyx, Tokyo, Japan) software.

Construction of Campbell insertion and in-frame deletion mutants.

For lacZY transcriptional integration mutagenesis (Campbell insertion), an internal DNA fragment of iucA was amplified using the primers IucAK and IucAS. The partial iucA fragment was purified, cloned into pGEM-T Easy (Promega, Madison, WI, USA), and confirmed by sequencing. For recombinational mutagenesis, the SmaI/KpnI-digested iucA fragment of pCOK190 was cloned into the pVIK112 suicide vector (60), creating pCOK194. The parent strain PA13 was conjugated with E. coli S17-1 harboring pCOK194, and kanamycin-resistant colonies were selected. The mutants were confirmed by PCR using a primer that anneals upstream of the truncated fragment and the primer LacFuse, followed by sequencing. We constructed the dfoJ and foxA null mutants using the same method as previously described with the primers and digestive enzymes listed in Table 3.

TABLE 3.

Primers used in this study

Primer	Description	Use	Reference
LacFuse	5′-GGGGATGTGCTGCAAGGCG-3′	Sequencing for lacZ fusion junctions	33
IucAS	5′-TTACCCGGGGTTACGCTGGTTTGCGGTGGA-3′	iucA-lacZY fusion	This study
IucAK	5′-GGCGGTACCGTTTGTCAGCCGCACGCTCAG-3′	iucA-lacZY fusion	This study
DfoJE	5′-GCTGTCGATCTTGACCAG-3′	dfoJ-lacZY fusion	This study
DfoJK	5′-GGTACCGCTCAGTCTGGAGGTCAA-3′	dfoJ-lacZY fusion	This study
FoxAE	5′- GCGATGGATGTGAATCAG -3′	foxA-lacZY fusion	This study
FoxAK	5′- GGTATGATCGTACTGCGTATC -3′	foxA-lacZY fusion	This study
IucABC 1	5′-GGCACTAGTAAGAAGTGCGCAGCCCAT-3′	iucABC deletion	This study
IucABC 2	5′-AAGCTTGGTACCGAATTCCGATTGAATCATTATCAAACAATT-3′	iucABC deletion	This study
IucABC 3	5′-GAATTCGGTACCAAGCTTACCAAGGACTGAATCATGACAAAC-3′	iucABC deletion	This study
IucABC 4	5′-GGCGGATCCAATGCCCAGACAGATGTT-3′	iucABC deletion	This study
DfoA 1	5′-GGCACTAGTGGCTGACCTTACGCATTA-3′	dfoA deletion	This study
DfoA 2	5′-AAGCTTGGTACCGAATTCCATTGCACCGTTCTCACTGGTTAG-3′	dfoA deletion	This study
DfoA 3	5′-GAATTCGGTACCAAGCTTCACTAATATGTCACAGTCCACACT-3′	dfoA deletion	This study
DfoA 4	5′-GGCGGATCCCTGAACATGTAGTCCATG-3′	dfoA deletion	This study
DfoC 1	5′-GGCACTAGTGCACCTTAATCAACGACA-3′	dfoC deletion	This study
DfoC 2	5′-AAGCTTGGTACCGAATTCCTGTGACATATTAGTGTTCCGTTT-3′	dfoC deletion	This study
DfoC 3	5′-GAATTCGGTACCAAGCTTCGGTAACGGGATAATGACCTCCTT-3′	dfoC deletion	This study
DfoC 4	5′-GGCGGATCCCTGCACCGATGAAACACA-3′	dfoC deletion	This study
Fur 1	5′-GGCACTAGTTATTGATGAAGATCGCCA-3′	fur deletion	This study
Fur 2	5′-AAGCTTGGTACCGAATTCCATGCGGATATTGTCCTGTTACTT-3′	fur deletion	This study
Fur 3	5′-GAATTCGGTACCAAGCTTGACAAATAAGCGCTTTCGGGTACA-3′	fur deletion	This study
Fur 4	5′-GGCGTCGACGTCGTGTTACATCAGCTT-3′	fur deletion	This study
FurX	5′-GGCCTCGAGCTGAAACAGGAAACAGCTATGACTGACAACAATACCGCA-3′	fur complementation	This study
FurP	5′-GGCCTGCAGTTATTTGTCGTGCAGCGTTTC-3′	fur complementation	This study

To generate in-frame deletion mutants, we amplified upstream and downstream fragments (approximately 450 bp) of the targeted gene region by PCR using the corresponding primer pairs (Table 3). After purification, the fragments were fused by overlap PCR. The final PCR products were cloned into pGEM-T Easy and confirmed by DNA sequencing. The fragments were excised using appropriate restriction enzymes and ligated into the suicide vector pNPTS138-R6KT (61). The resulting plasmids were introduced into PA13 by conjugative mating, and mating cells were spread on LB medium containing kanamycin and rifampicin. Single-crossover integrants were selected on these LB plates. Single colonies were grown overnight in LB with rifampin (25 μg/mL) and plated on LB containing 5% (wt/vol) sucrose to select for plasmid excision. Excision of the integrated plasmid was confirmed by patching onto LB supplemented with kanamycin or sucrose. Kanamycin-sensitive colonies were selected, and deletion of the targeted DNA was confirmed by diagnostic PCR and DNA sequencing of the product.

Gene complementation.

To generate target gene complementary strains, we cloned the intact fur gene into the broad-host-range plasmid vector pBBR1-MCS5 (62), generating pBBR1-MCS5::P_lac-fur, which was transferred to the corresponding mutant strains by conjugation (Table 2).

Antibacterial activity assay.

For the antibacterial activity assay, each overnight culture of the target bacterial strains including E. amylovora ATCC 19383 (63) or Y. enterocolitica ATCC 9610 (64) was added to warm AB minimal agar medium containing 0.2% glucose, mixed, and then immediately poured onto petri plates. After solidification of the agar, 10-μL overnight cultures of P. ananatis PA13 and the mutant strains were spotted onto the surface of the agar plate. The plates were incubated at 28°C until the inhibition zones were visible. The diameter of the inhibition zone around the bacterial colonies was measured. The experiment was repeated at least three times independently.

Siderophore production assay.

A siderophore production assay was performed using King’s B-CAS agar plates based on King’s B medium as described previously (65). Overnight cultures of P. ananatis PA13 and mutant strains were dropped onto a CAS agar plate and incubated at 28°C for 48 h. The diameter of the orange-yellow halo was measured. To assess the relative siderophore production, the average value of the blue pixels of the halo of each strain was obtained from King’s B-CAS assay using ImageJ software (version 1.45s). Using the value of strain Pf-5 as a reference value, the average value of the removed blue pixels of each strain was compared with that of each of the other strains. The average value of blue pixels in the area inside the dotted line of Pf-5 was used as a positive control, and the bacterial colony-free region was used as a negative control. The experiment was repeated at least three times independently.

Virulence assay.

Rice seedlings (Oryza sativa cv. Dongjin) were grown for 5 weeks at 28°C and 60% humidity. Onion bulbs (Allium cepa) were purchased from a commercial market. The P. ananatis PA13 and mutant strains were cultured in LB medium at 28°C overnight, centrifuged, washed, and suspended in sterilized distilled water (SDW) at an optical density at 600 nm (OD₆₀₀) of 0.8. Bacterial suspensions (100 μL) were injected into the sheaths of rice or onion bulbs using a syringe. Sterilized distilled water was used as a negative control. The inoculated rice seedlings were maintained in a greenhouse at 28°C and 60% humidity. The onion bulbs were placed into a plastic box with saturated paper at room temperature. Rice seedlings and onion bulbs were observed for disease symptoms for 18 and 15 days postinoculation, respectively. The experiment was repeated at least three times independently.

β-Galactosidase assay.

The wild-type and mutant background used in the β-galactosidase assays was PA13L. All test strains were grown for 20 h and subcultured in LB broth at 28°C. The assays were performed using exponential-phase cultures at an OD₆₀₀ of ∼0.4 as described previously (33).