Iron-regulated gene ireA in avian pathogenic Escherichia coli participates in adhesion and stress-resistance

Background Avian pathogenic Escherichia coli (APEC) causes avian colibacillosis, which results in economic and welfare costs in the poultry industry worldwide. The pathogenesis of avian pathogenic E. coli strains is not well defined. Here, the function of an outer membrane protein encoded by the ireA gene of avian pathogenic E. coli strain DE205B was investigated. Results The ireA gene was distributed in 32.9 % (46/140) of tested E. coli strains, with high percentages in the phylogenetic ECOR groups B2 (58.8 %, 10/17) and D (55.9 %, 19/34). The gene expression level of ireA of APEC strain DE205B in high Fe M9 media was 1.8 times higher (P < 0.05) than that in low Fe M9 media. An ireA deletion mutant and complementary strain were constructed. Compared with the wild-type strain DE205B, the expression of most ferric uptake genes in the ireA deletion mutant were significantly upregulated (P < 0.05). The adhesion ability of the ireA deletion mutant to DF-1 cells was significantly decreased. The survival rate of ireA deletion mutant was reduced 21.17 % (P < 0.01), 25.42 (P < 0.05) and 70.0 % (P < 0.01) under alkali, high osmolarity, and low temperature (4 °C) conditions, respectively, compared with the wild-type strain. Conclusions The results suggested that the protein encoded by the iron-regulated gene ireA has roles in adhesion and stress resistance in avian pathogenic E. coli. Electronic supplementary material The online version of this article (doi:10.1186/s12917-016-0800-y) contains supplementary material, which is available to authorized users.


Background
Avian pathogenic Escherichia coli (APEC), a subgroup of extra-intestinal pathogenic E. coli (ExPEC) causes avian colibacillosis and imposes economic losses on the poultry industry worldwide [1]. However, the pathogenesis of APEC is poorly understood. Many virulence genes have been studied to identify virulence factors in APEC, including those involved in adhesion, iron-regulation, toxin/cytotoxin production and serum resistance [2]. Iron is an essential element involved in important biological processes [3]. Biological activities in cells, such as peroxide reduction, nucleotide biosynthesis and electron transport, are facilitated by iron ions [4]. Extra-intestinal sites have low iron contents; therefore, ExPEC strains struggle to take up iron from the host during infection [5]. During natural infection, the initiation, progression and transmission of most bacterial infections depend on the ability of the invading pathogen to acquire iron from the complicated environment [6]. During iron acquisition, the cell must produce transmembrane receptors for siderophores that chelate iron ions [7]. There are various receptors that chelate iron ions encoded by bacterial genes, such as chuA, the SitABCD system, iron, iha, iutA, and ireA. Outer membrane protein ChuA participates in heme acquisition in enterohemorrhagic E. coli and uropathogenic E. coli (UPEC) strains, and is important for the pathogenicity of APEC [8,9]. The SitABCD system, identified in the APEC strain MT512 by comparative genomic analysis, was reported to be associated with the pathogenicity of APEC [9,10].
IreA was suggested to be involved in Fe acquisition and to act as an iron-regulated virulence gene in the blood-or urine-derived ExPEC E. coli isolated from humans [11]; however, its exact role in APEC strains remains unknown. Herein, an ireA deletion mutant was constructed to study the ireA gene function in the APEC strain DE205B.

Expression of the ireA gene
The ireA gene expression was tested by Immunoblotting. Western blotting was performed with anti-His serum, showing expected fusion protein bands for ireA (39 kDa) from strains DE205B. However, only fusion his protein (18 kDa) was detected from the blank plasmid control (Fig. 1). These results indicated that ireA was expressed under laboratory conditions.
Gene expression of ireA in M9 media with different iron content The relative gene expression of ireA in low Fe M9 media was 1.8 times higher than that in high Fe M9 media (P < 0.01) (Fig. 2).

Growth curve
The CFU of the wild-type strain DE205B, mutant strain DE205BΔireA and the complementary strain DE205BCΔireA was monitored for 12 h. There was no significant difference between the growth curves of the wild-type and mutant strains (Fig. 3), which indicated that deletion of ireA had no effects on the growth of DE205B.

Expression variations of Ferric uptake system and adherence genes
The expressions of fepC, feoB, chuT, fyuA, irp1, irp2, chuA and fepA were detected in the mutant and compared with their expressions in the wild-type strain DE205B. Most of the ferric uptake genes in the mutant strain DE205BΔireA, except irp1, irp2 and chuA, were significantly upregulated (P < 0.05) (Fig. 4). Irp1, irp2 and chuA were non-significantly upregulated. Furthermore, the expressions of the eight ferric uptake genes in the complementary strain DE205BCΔireA were restored to wild-type levels.

Adherence assays
The number of mutant DE205BΔireA bacteria adhering to DF-1 cells was significantly decreased compared with   the number of DE205B wild-type adherent cells (P < 0.05) (Fig. 6). The adhesion to DF-1 cells of complementary strain DE205BCΔireA was partly restored to wildtype levels.

Animal infection
The LD 50 of the wild-type, deletion mutant and complementary strains were 1.74 × 10 5 , 2.45 × 10 5 , and 3.16 × 10 5 , respectively ( Table 2). The results showed that there was no significant difference between the LD 50 of the wildtype and deletion mutant strains, indicating that the deletion of ireA gene has no obvious effect on the virulence of DE205B toward ducks.

Determination of resistance to environmental stress
Compared with the wild-type strain, the survival rate of the ireA deletion mutant was reduced by 21.17 % (P < 0.01) and 25.42 % (P < 0.05) under alkali and high osmolarity conditions, respectively (Fig. 7). The resistance to alkali and high osmolarity conditions of the complementary strain DE205BCΔireA was almost restored to wild-type levels.
For the temperature challenge, compared to the wild type strain, the ireA deletion mutant was reduced 70.0 % (P < 0.01) under the low (4°C) temperature (Fig. 8). The resistance to low temperature of the complementary strain DE205BCΔireAwas partly restored. The results showed that presence of the ireA gene increased the stress-resistance of APEC strain DE205B.

Discussion
The ireA gene is an iron-regulated gene and is involved in iron acquisition in human pathogenic E. coli isolates and our study proved this protein functions in APEC. Additionally, we identified two new functions of ireA using the deletion mutant. In the present study, ireA was demonstrated to contribute to the adhesion to DF-1 cells. Moreover, the expressions of several adhesion genes were tested and the results showed no significant differences between wild-type and mutant strain, indicating that the ireA gene indeed plays a role in adhesion. Tarr et al., identified adhesin Iha from an O157:H7 strain of E. coli [12]. This adhesin shared high similarities with several identified or putative siderophore receptors. Siderophore receptor IrgA was reported to contribute to growth in the rabbit ileal loop model in vivo and to enhance virulence in an infant mouse model, suggesting a possible role in colonization [13,14]. Here, we proved that iron-regulated gene ireA plays a role in the adhesion of APEC strains. The ireA gene also increased stress-resistance under alkali and high osmolarity conditions, as well as underlow temperature. Thus, the E. coli strains were reported to be classified into four main phylogenetic groups (A, B1, B2, and D) [15]. Virulent ExPEC strains mainly belong to phylogroup B2 and D, whereas most commensal strains belong to phylogroup A [16]. The ireA gene was distributed more frequently in the B2 (58.8 %) and D (55.9 %) groups than in the A (19 %) and B1 (19.2 %) groups, indicating that ireA might be associated with the virulence of the APEC strain DE205B. Thus, the result correlated with that Russo's report: ireA was detected in 13 (26 %) of 50 random clinical isolates from patients and in none (0 %) of 14 fecal isolates, which presumably represented commensal strains [11]. Taken together, these results indicated that ireA might be a virulence gene in both human and avian ExPEC E. coli strains.
The expression level of ireA in DE205B was decreased in high Fe M9 media compared with that in low Fe media, indicating that ireA is involved in iron-regulation in this APEC strain. This result agreed with the report of Russo et al., who found that ireA was a iron-regulated virulence gene in the blood-or urine-derived isolates of ExPEC E. coli [11]. Most ferric uptake genes, such as fepC, feoB, chuT, fyuA and fepA were upregulated in the ireA deletion mutant strain. This might represent a compensatory function for ireA gene deletion. Fe acquisition is important for many microorganisms, especially for pathogens that grow in the host, which attempts to limit Fe availability. It is thought that pathogens harbor  multiple Fe acquisition systems to ensure that Fe is gained from the host cells to provide a selective advantage. Alternatively, certain siderophores, and their cognate receptors, might be more active in certain environments, such as inside or outside the gastrointestinal tract [17,18]. Moreover, multiple systems might represent 'alternatives' that protect against the disruption of one system caused by genomic rearrangements or mutations.
In the duck infection experiments, the LD 50 showed no significant difference between the wild-type DE205B and the ireA deletion mutant strain. Thus, it seemed that ireA deletion had no obvious effect on the virulence of DE205B. However, DE205B has several Fe acquisition systems. We showed that most of the other Fe acquisition genes were upregulated in the ireA gene deletion mutant. Thus, the ireA gene might indeed contribute to the virulence of the APEC strain DE205B, while other Fe acquisition genes displayed compensatory functions when the ireA gene was deleted.

Conclusions
In summary, the ireA gene was mainly distributed in the more virulent phylogenetic ECOR group B and D. Compared with the wild-type strain, the adhesion and resistance to environmental stress of the ireA deletion mutant were significant decreased. This indicated that ireA is a Fe iron-regulated gene that aids adhesion and stress-resistance in the APEC strain DE205B.

Bacterial strains, plasmids and growth conditions
The APEC strain DE205B, which was characterized previously, was isolated from the brain of a duck with neurological symptoms in our previous work [19][20][21][22]. Its serotype is O2:K1, which is one of the predominant serotypes of APEC [9,23]. All E. coli strains were grown in Luria-Bertani (LB) medium at 37°C with shaking or on LB plates with 2 % agar. When necessary, the LB medium was supplemented with appropriate antibiotics: ampicillin (Amp; 100 μg/mL) and kanamycin (Kan; 50 μg/mL), unless otherwise specified. Information on the main strains and plasmids used in this study are listed in Table 3. M9 minimal medium was used to assess the expression of ireA, which was prepared as

Prevalence of ireA among E. coli Strains
The distribution of ireA among 140 E. coli strains (Additional file 1: Table S1) maintained in our laboratory was tested using PCR. The PCR primers were designed using Primer Premier 5. The specific primers ireA-F and ireA-R are listed in Table 4. The PCR conditions were: 5 min at 95°C for initial denaturing; 30 cycles of 1 min at 95°C, 30 s at 50°C and 1 min at 72°C; followed by 10 min at 72°C for extension. The DNA of DE205B and ddH 2 O were used as positive and negative templates, respectively. The PCR products were electrophoresed through 1 % agarose gels. Phylogenetic E. coli reference (ECOR) groups of single isolates of the 140 E. coli strains were determined using triplex PCR, as described previously [16,24].

Expression of the ireA gene
The expression of the ireA gene was tested by fusion expression and western blotting as previously reported [25,26]. The fusion fragment of ireA (including the ireA promoter and 579bp of ireA sequence) and His tag were inserted into plasmid pET32a(+). The fusion PCR primers P1 and P2 were listed in Table 4. Plasmid pET32a(+) without the fusion fragment was used as a blank control. For immunoblotting, protein samples were subjected to sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene fluoride membranes (Amersham Pharmacia Biotech, Piscataway, NJ, USA), as described previously [27]. Anti-His serum was the primary antibody, horseradish peroxidase-conjugated goat anti-mouse IgG was the secondary antibody and 3,3′-diaminobenzidine was used as the substrate.

Regulation of ireA expression in M9 media
M9 minimal medium was used to assess the expression of ireA.M9 medium or M9 medium with Fe (0.1 mM Fe(NO 3 ) 3 ) were used as low and high Fe content media, respectively. DE205B was cultured in both M9 media to the mid-log phase and the expression of ireA was detected by quantitative real-time reverse transcription PCR(qRT-PCR). The real -time PCR primers were designed by PrimerQuest Tool IDT (http://sg.idtdna.com/ primerquest/Home/Index). Briefly, total RNA was extracted from 1 ml of bacteria culture using the Trizol RNA isolation protocol (Invitrogen, Shanghai, China) and cDNA was amplified by reverse transcription according to the instructions of the primeScript RT reagent Kit (Takara). Quantitative real-time PCR (qPCR) was carried out using the ABI Prism 7300 and Sequence Detection System software version 1.4 (Applied Biosystems, Foster City, CA, USA), according to the instructions of the SYBR Premix Ex Taq (Takara, Dalian, China). QPCR primers for ireA (QireA-F and QireA-R) are listed in Table 4. DnaE was used as a reference gene. Assays were performed three times. The relative expressions of ireA in different media were calculated using the 2 -△△Ct method [28]. Statistical analysis was performed using an unpaired t test in Graphpad Prism 5.0.

Construction of the ireA deletion mutant and complementary strain
An ireA knockout strain of DE205B was constructed using the lambda red recombinase system described by Datsenko and Wanner [29]. The specific primers ireAMu-F and ireAMu-R were designed to amplify the target gene ireA. The kanamycin resistance gene, which contained sequences homologous to the 5′ and 3′ ends of the target sequence, was amplified using plasmid pKD4 as a template. The PCR products were then transformed by electroporation into DE205B containing the lambda red recombinase expression plasmid pKD46. The transformed bacterial cells were first incubated at 30°C for 2 h in super optimal broth with catabolite repression (SOC) broth, and then grown on LB agar containing kanamycin at 37°C. Mutants were confirmed by PCR and sequenced using primers k1 and k2 (Table 4) in combination with primers ireA-1 and ireA-2 (Table 4) flanking the ireA region. To remove the kanamycin resistance gene, plasmid pCP20 was transformed into the mutant and a kanamycin sensitive mutant strain was selected. Finally, the ireA deletion mutant strain without kanamycin resistance was named as DE205BδireA (Additional file 2: Figure S1).

Growth curve
The growths of the DE205B wild-type and mutant strains were compared in LB medium at 37°C over a course of 12 h, starting at 10 7 CFU/ml. Bacterial growth was estimated by plate counting as Colony Forming Units (CFU). Assays were performed three times.
The expression of the Ferric uptake system and adherence genes The effect of ireA deletion on the regulation of the ferric uptake system, including fepC, feoB, chuT, fyuA, irp1, irp2, chuA and fepA was detected using qRT-PCR. Seveal adherence genes, including yfcO, yfcQ, aufG, fmlD, fmlE, yadN and fimH were also selected to test their expression levels. The wild-type DE205B and the ireA deletion mutant were cultured in LB to mid-log phase and total RNA was extracted from 1 ml of bacterial culture using the Trizol RNA isolation protocol (Gibco BRL, USA, cat. no.15596-026). cDNA was reverse transcribed and real-time PCR was carried out as described above. The qRT-PCR primers for the ferric uptake system and adherence genes are listed in Table 4. The real-time PCR primers were designed unsing Pri-merQuest Tool IDT (http://sg.idtdna.com/primerquest/ Home/Index). The relative expression levels of the genes were calculated using the 2 -△△Ct method [28]. Assays were performed three times, and the statistical analysis was performed using an unpaired t test in Graphpad Prism 5.0. On the figures, error bars indicate the standard deviation.

Adherence assays
The adherence assay was performed as described previously [24]. Briefly, chicken embryo fibroblast (CEF) DF-1 cells were seeded at approximately 1 × 10 5 cells per well in 24-well tissue culture trays (TPP, Shanghai, China) and grown in Dulbecco's modified Eagle medium (DMEM) with 10 % fetal bovine serum at 37°C in a 5 % CO 2 humidified atmosphere without antibiotics. DF-1 cells were washed once with DMEM and then inoculated with 500 ul of 2 × 10 7 CFU/ml bacteria per well for 2 h at 37°C in the presence of 5 % CO 2 . The cells were washed, lysed with ddH 2 O, and the number of bacterial cells was calculated by plate counting. In all assays, wells only containing DF-1 cells were used as negative controls. The adherence assays were conducted three times. The statistical analysis was performed using an unpaired t test in Graphpad Prism 5.0.

Animal infections
Animal infections were performed as described previously [19,20]. We purchased 7-day-old ducklings and young duck feeds from Anhui Poultry Farm (Anhui, China). Bacterial strains were cultured to the exponential phase, harvested, washed three times in PBS and then adjusted to the appropriate doses. Twenty-five 7day-old ducks were inoculated intramuscularly 0.2 ml of each bacterial suspension (DE205B, DE205BΔireA or DE205BCΔireA) at four concentrations (5 × 10 8 CFU/ ml, 5 × 10 7 CFU/ml, 5 × 10 6 CFU/ml, and 5 × 10 5 CFU/ ml). Assays were performed three times. Seven ducks were used for each dose. Seven ducks were injected with PBS as a negative control. Death of the ducks was monitored for 7 days post infection. We calculated the LD 50 of each strain using the method described by Spearman-Karber [30].

Determination of resistance to environmental stress
Resistance to environmental stress was tested for the wildtype and the mutant strain, as described by La Ragione et al. [31]. Bacteria were cultured in LB broth overnight and harvested by centrifugation. The cells were resuspended in PBS and adjusted to 10 7 CFU/ml in PBS. For alkali challenge, 100 μl of adjusted cells were mixed with 100 μl Tris buffer (1 M, pH10.0) and 800 μl ddH 2 O (final concentration, 100 mM, pH10.0) and incubated at 37°C for 30 min. For high osmolarity endurance, cells were mixed with an equal volume of 4.8 M NaCl (final concentration, 2.4 M) and incubated at 37°C for 1 h. Bacteria were exposed to PBS (pH 7.0) as a control.
Temperature challenge was performed as previously described with modifications [32], each bacterial suspension, at a concentration of 10 7 CFU/ml (DE205B, DE205BΔireA or DE205BCΔireA) was incubated at 4°C for 7days. Assays were performed three times. The survival rates of wild-type and mutant strains were calculated by plate counting and compared using GraphPad Prism 5.