SodA Contributes to the Virulence of Avian Pathogenic Escherichia coli O2 Strain E058 in Experimentally Infected Chickens

Avian colibacillosis, caused by strains of avian pathogenic Escherichia coli, is a major bacterial disease of severe economic significance to the poultry industry worldwide. The virulence mechanisms of APEC are not completely understood. This study investigated the influence of an antioxidant protein, SodA, on the phenotype and pathogenicity of APEC O2 strain E058. This is the first report demonstrating that SodA plays an important role in protecting a specific APEC strain against hydrogen peroxide-induced oxidative stress and contributes to the virulence of this pathotype strain. Identification of this virulence factor will enhance our knowledge of APEC pathogenic mechanisms, which is crucial for designing successful strategies against associated infections and transmission.

ABSTRACT

Strains of avian pathogenic Escherichia coli (APEC), the common pathogen of avian colibacillosis, encounter reactive oxygen species (ROS) during the infection process. Superoxide dismutases (SODs), acting as antioxidant factors, can protect against ROS-mediated host defenses. Our previous reports showed that the sodA gene (encoding a Mn-cofactor-containing SOD [MnSOD]) is highly expressed during the septicemic infection process of APEC. sodA has been proven to be a virulence factor of certain pathogens, but its role in the pathogenicity of APEC has not been fully identified. In this study, we deleted the sodA gene from the virulent APEC O2 strain E058 and examined the in vitro and in vivo phenotypes of the mutant. The sodA mutant was more sensitive to hydrogen peroxide in terms of both its growth and viability than was the wild type. The ability to form a biofilm was weakened in the sodA mutant. The sodA mutant was significantly more easily phagocytosed by chicken macrophages than was the wild-type strain. Chicken infection assays revealed significantly attenuated virulence of the sodA mutant compared with the wild type at 24 h postinfection. The virulence phenotype was restored by complementation of the sodA gene. Quantitative real-time reverse transcription-PCR revealed that the inactivation of sodA reduced the expression of oxidative stress response genes katE, perR, and osmC but did not affect the expression of sodB and sodC. Taken together, our studies indicate that SodA is important for oxidative resistance and virulence of APEC E058.

IMPORTANCE Avian colibacillosis, caused by strains of avian pathogenic Escherichia coli, is a major bacterial disease of severe economic significance to the poultry industry worldwide. The virulence mechanisms of APEC are not completely understood. This study investigated the influence of an antioxidant protein, SodA, on the phenotype and pathogenicity of APEC O2 strain E058. This is the first report demonstrating that SodA plays an important role in protecting a specific APEC strain against hydrogen peroxide-induced oxidative stress and contributes to the virulence of this pathotype strain. Identification of this virulence factor will enhance our knowledge of APEC pathogenic mechanisms, which is crucial for designing successful strategies against associated infections and transmission.

INTRODUCTION

As a subset of extraintestinal pathogenic Escherichia coli (ExPEC), avian pathogenic Escherichia coli (APEC) strains cause a wide range of localized and systemic infections collectively called avian colibacillosis, leading to great economic losses in poultry (1–3). APEC strains use virulence factors for colonizing and invading the host, including adhesins, invasins, outer membrane proteins, iron acquisition systems, and others (4–11).

APEC strains usually enter the host by crossing the respiratory epithelia and penetrate deeply into the mucosa and submucosa to reach the bloodstream, causing septicemia (12, 13). Phagocytic cells, such as macrophages and heterophils, are elicited into the respiratory tract and play a key role in the initial host defense against APEC (14, 15). An important killing mechanism of phagocytes involves the production of highly microbicidal reactive oxygen metabolites during the so-called oxidative burst, which is generally induced by the engulfment of bacteria (16). Thus, within the host, APEC faces reactive oxygen species (ROS) from the oxidative burst of macrophages. ROS, including superoxide anions (O₂⁻), hydroxyl radicals (OH·⁻), and hydrogen peroxide (H₂O₂), are known to damage DNA, RNA, proteins, and lipids (16, 17).

Superoxide dismutases (SODs) are metalloenzymes that can directly convert deleterious superoxide to molecular oxygen and hydrogen peroxide; hydrogen peroxide is in turn metabolized by catalases and/or peroxidases (18). Hence, these enzymes play an important role in the pathogenesis of some pathogens (19–22). SODs are classified into the following three types depending on the metal cofactor used: Cu,ZnSOD (SodC), MnSOD (SodA), and FeSOD (SodB). Previously, we used an analysis of global gene expression profiles to show that the expression levels of the sodA gene in APEC strain E058 were clearly increased during infection in vivo (23). However, the role of sodA in the pathogenesis of APEC is not fully understood. Therefore, in this work, we constructed a sodA deletion mutant of the virulent APEC O2 strain E058 and evaluated its phenotype. The effects of sodA on serum and environmental stress resistance, biofilm formation, intracellular survival, virulence, and gene expression were evaluated to fully identify the contribution of SodA to the virulence of APEC O2 strain E058.

RESULTS

Bactericidal effects of serum and environmental stress on sodA mutant.

Serum resistance and environmental stress adaptation correlate with the virulence of APEC during infection in vivo (5, 24). In a serum bactericidal assay, the bacterial numbers of the sodA mutant inoculated in serial serum dilutions were similar to those of wild-type strain E058. The control strain DH5α was sensitive to serum, which demonstrated that the serum was active and bactericidal (Fig. 1A). Thus, the deletion of sodA did not appear to affect serum resistance.

FIG 1.

Bacterial resistance to SPF chicken serum and environmental stress. (A) Resistance to SPF chicken serum. Bacteria were incubated with the diluted SPF chicken serum at 37°C for 30 min. Bacteria were counted by plating on LB agar at serial dilutions. HI, heat-inactivated SPF chicken serum. (B) Resistance to environmental stress. Bacteria were examined for adaptation to environmental stresses, including alkalinity, acidity, and high osmolarity. Survival was calculated as the number of bacteria remaining after the stress exposure divided by the initial number of bacteria. Data represent averages of the results from three independent assays (*, P < 0.05).

To identify whether SodA plays a role in adapting to environmental stresses, including alkalinity, acidity, and high osmolarity, we compared the resistance phenotype of the sodA mutant and its wild-type strain under these stressful conditions. The results showed that the survival of the sodA mutant resembled that of the wild-type strain (Fig. 1B), indicating that SodA was not required for resistance to these environmental stresses.

Effect of sodA mutation on sensitivity to oxidative stress.

We examined the protective effects of SodA in APEC E058 against the oxidant hydrogen peroxide. Growth rate and survival in the presence of 10 mM H₂O₂ were tested in the sodA mutant, complementation, and wild-type strains. The growth of the sodA mutant was similar to that of the wild-type in Luria-Bertani broth (LB) medium but was severely inhibited in LB medium containing 10 mM H₂O₂ (Fig. 2A). The sodA mutant was poorly viable during exposure to 10 mM H₂O₂ compared to the wild-type and complementation strains (Fig. 2B).

FIG 2.

Effect of sodA deletion from APEC strain E058 on sensitivity to oxidative stress. (A) Growth curves of sodA mutant, complementation, and wild-type strains. Strains E058 (wild type), E058 ΔsodA (sodA deletion mutant), and ReE058 ΔsodA (complementation strain) were grown in Luria-Bertani broth with or without 10 mM H₂O₂ at 37°C, and their growth was determined by measurement of the optical density at 600 nm (OD₆₀₀). (B) The survival rates of strains in 10 mM H₂O₂ for 30 min at 37°C. Data represent the averages of the results from three independent assays (**, P < 0.01).

Ability of sodA mutant to form biofilm.

To determine the effect of sodA deletion on biofilm production, the biofilm formation ability of the sodA mutant, complementation, and wild-type E058 strains was examined using the microtiter plate method. Crystal violet staining of bacteria grown in LB rich or M9 minimum medium showed that the sodA mutant produced a decreased amount of biofilm compared with the wild-type and the complementation strains (Fig. 3), even though their growth rates were similar. Higher production of biofilm was observed under conditions using M9 minimal medium (Fig. 3). These results indicate that sodA is involved in biofilm formation by APEC E058.

FIG 3.

Quantification of the biofilms produced by the sodA mutant, the complementation strain, and wild-type APEC strain E058. Biofilm formation was measured for bacteria inoculated in LB rich and M9 minimum media in 96-well microtiter plates for 24 h. OD₅₉₅ values for stained biofilm are the mean ± standard deviation (SD) of the results from three independent biological replicates (**, P < 0.01).

SodA promotes the ability of APEC E058 to survive within chicken macrophages.

An ingestion assay was used to analyze the ingestion of the sodA mutant by avian macrophage HD-11 cells. The inactivation of sodA rendered the bacteria more vulnerable to macrophages. The sodA mutant showed a highly increased ingestion ratio compared to the wild-type and complementation strains (P < 0.01) (Fig. 4A).

FIG 4.

Survival of bacteria in HD-11 chicken macrophages. (A) Ingestion of wild-type strain E058, E058 ΔsodA isogenic mutant, and ReE058 ΔsodA complementation strain by HD-11 cells. The values represent the average data from three independent experiments. The error bars indicate standard deviations. (B) Intracellular growth of bacteria in chicken macrophage HD-11 cells. The intracellular growth of wild-type strain E058, the E058 ΔsodA isogenic mutant, and the ReE058 ΔsodA complementation strain were compared over a 12-h period. Standard errors of the mean for the results from three independent experiments are shown. The asterisks indicate statistically significant differences (**, P < 0.01).

To determine whether SodA affects bacterial survival within macrophages, we compared the bacterial yields of sodA mutant, complementation, and wild-type strains recovered at 2, 4, 6, 8, and 12 h postinfection (beginning at time zero [T₀]). The proportions of wild-type bacteria recovered from macrophages were 0.19%, 0.18%, 0.22%, 0.17%, and 0.15% of the inoculum at 2, 4, 6, 8, and 12 h postinfection (h p.i.), which were 1.1- to 1.4-fold higher than those of the primary ingested bacteria (T₀) (Fig. 4B). However, the sodA mutant survived poorly in macrophages, with proportions of 0.17%, 0.16%, 0.12%, 0.07%, and 0.04% of the inoculum recovered at 2, 4, 6, 8, and 12 h p.i., which were 1.2- to 4.8-fold lower than those of the primary ingested bacteria (T₀) (Fig. 4B). Moreover, the survival ability within macrophages was restored for the complementation strain. Thus, it can be concluded that sodA is involved in the survival of APEC E058 within macrophages.

Competition between the wild-type strain E058 and the sodA mutant in vitro and in vivo.

The results of competition experiments comparing mutant and wild-type strains reflect the effects of the mutation on the bacterial growth or virulence. In vitro, in LB medium, the sodA mutant competed equally with the wild-type strain (data not shown). However, during coinfection in vivo, the sodA mutant was weakened and showed a significantly reduced competitive index (CI), with mean decreases of 33-fold, 24-fold, 22-fold, and 36-fold compared to the wild-type strain in blood, liver, lung, and spleen, respectively (all P < 0.01) (Fig. 5). Meanwhile, complementation of the mutation restored competitive fitness. These results indicate that SodA confers APEC E058 a survival advantage in an in vivo competitive colonization model.

FIG 5.

In vivo competition assays. Comparative colonization levels, presented as the log₁₀ competitive index, in the blood, liver, lung, and spleen of SPF chickens 24 h after inoculation with a mixture of wild-type APEC strain E058 and the sodA mutant or complementation strain. Horizontal bars indicate the mean log₁₀ CI values. Statistically significant decreases in CI values are indicated with asterisks (**, P < 0.01).

SodA contributes to bacterial colonization during systemic infection in vivo.

To investigate the involvement of sodA in APEC E058 virulence, colonization by the sodA mutant and its complementation strain, as well as the wild-type strain, was determined in challenged chickens. Bacterial loads in various tissues were compared between groups of challenged chickens at 24 h postinfection. The sodA mutant colonized poorly in vivo in comparison to the corresponding parental strain, with the bacterial loads recovered significantly lower than those of the wild-type strain E058 (P < 0.01) (Fig. 6). Complementation of the sodA mutant restored bacterial colonization capacity, with bacterial loads reaching the level of the wild-type strain. These results showed that inactivation of sodA may lead to reduced bacterial numbers in the internal organs and limit systemic invasion, indicating that SodA is involved in APEC E058 virulence and contributes to survival during systemic infection in the chicken infection model.

FIG 6.

Colonization by the wild-type strain E058, E058 ΔsodA mutant, and ReE058 ΔsodA complementation strain during systemic infection of chickens. The data are presented as log₁₀ CFU per milliliter of heart blood or log₁₀ CFU per gram of tissue. The horizontal bars indicate the mean values. Each data point represents a sample from an individual chicken. Statistically significant differences, as determined by the Mann-Whitney test, are indicated by asterisks (**, P < 0.01).

Expression profile of genes involved in oxidant resistance.

The transcription levels of several oxidant resistance genes were analyzed by quantitative real-time reverse transcription-PCR (qRT-PCR) in the sodA mutant. The expression of the oxidative stress response genes katE, perR, and osmC was significantly reduced in the sodA mutant strain, while the sodB and sodC genes were not affected compared to the wild-type strain. The expression levels of the tested genes were restored in the complementation strain (Fig. 7).

FIG 7.

Quantitative real-time reverse transcription-PCR (qRT-PCR) analysis of gene expression. Expression of genes involved in oxidative resistance in APEC was measured by qRT-PCR for each strain. Data were normalized to the housekeeping gene gapA. Results are relative expression ratios compared to wild-type strain E058 (**, P < 0.01). The error bars indicate standard deviations.

DISCUSSION

APEC, like most pathogens, may experience oxidative stress caused by ROS derived from aerobic metabolism, environmental sources, and the host immune response (16, 25). Bacterial antioxidants play a critical role in the detoxification of endogenously and host-derived oxidative radicals during host-pathogen interactions (26, 27). Mn-cofactor-containing superoxide dismutase SodA can destroy the toxic superoxide anion, which is produced both endogenously and within the oxidative burst of host phagocytes. For a comprehensive understanding of the influence of sodA on the phenotype and pathogenicity of APEC, the sodA-deficient mutant and complementation strains from the highly virulent APEC O2 strain E058 were constructed in this study. The deletion of sodA did not affect the growth or serum or environmental stress resistance of APEC E058. However, the sodA mutant displayed an increased susceptibility to killing by exogenous hydrogen peroxide (Fig. 2). The sensitivity of the sodA mutant to H₂O₂ may be due to increased cellular damage from hydroxyl radicals generated by iron-mediated Fenton reaction (28). This phenotype displayed by the sodA mutant is what would be expected of a bacterial strain that has lost an important cytoplasmic defense against oxidative stress and is similar to the phenotypes of other bacterial sodA mutants (19, 29).

Biofilms play an important role in protecting pathogens from the host immune system (30). Previous research found that a higher level of expression of the sodA gene was induced during the transition to biofilm growth both in E. coli and Staphylococcus aureus (31, 32). This prompted us to study the effect of the sodA deletion on biofilm production by APEC E058. We tested the biofilm formation ability of the bacteria under nutrient-rich or -poor culture conditions (i.e., LB rich or M9 minimum medium, respectively). The results showed that a mutation of sodA decreased biofilm formation in both LB and M9 media (Fig. 3). Additionally, biofilm formation in M9 minimal medium provides greater amounts of biofilm, which is probably caused by cell starvation, which triggers biofilm formation. These results suggest that the sodA gene involved in the bacterial oxidative stress responses may play a significant role in the development of biofilm.

In view of the role of sodA in bacterial resistance to the oxidative burst of host phagocytes, we tested the ability of the sodA mutant to survive within chicken macrophage HD-11 cells. The sodA mutant was more easily ingested by the macrophages, while the wild-type and complemented bacteria appeared to resist ingestion. The deletion of sodA also resulted in decreased intracellular growth of APEC E058 in macrophages (Fig. 4), while no differences were found between the survival within macrophages of the wild-type and complemented strains. Thus, the lack of a functional sodA gene alters the interaction of APEC E058 with macrophages. Our results indicate that Mn-cofactor-containing SOD (MnSOD), by its ability to scavenge the primary product of the oxidative burst, is able to effectively impair the host response to APEC infection and thus may represent an important virulence factor in APEC E058.

To determine the contribution of sodA to APEC infection, we performed avian experimental infection tests. The sodA mutant was significantly outcompeted by the wild-type strain during in vivo coinfection (Fig. 5). Coinfection experiments can show fitness defects as well as decreased virulence, depending on the type of mutation. To test whether or not the drop in competitive fitness also occurred in vitro or if it was only seen during infection, we also conducted in vitro competition test. The sodA mutant competed equally with the wild-type strain in LB medium, which excluded the possibility of a fitness defect and demonstrated the decreased virulence phenotype of the mutant. As we have shown a potential role for SodA in survival in a macrophage cell line, we further investigated the survival of the mutant strain in the blood. The decreased numbers for the sodA mutant in the blood correlate with less of a capacity to survive in phagocytic cells. Moreover, the sodA mutant showed significantly decreased colonization compared with the wild-type strain in various organs tested in a single-strain challenge model (Fig. 6), implying that SodA plays an important role in the virulence of APEC E058 in chickens. The complementation strain showed restored virulence and colonized the internal organs of inoculated chickens to the same extent as the wild-type strain. Thus, the disruption of the sodA gene in APEC E058 leads to attenuation of virulence in systemic infection in chickens.

A previous report described a lack of a role for MnSOD in an extraintestinal pathogenic E. coli O18 strain in a newborn rat meningitis model (33). Their results indicated, however, that the elimination of sodA had little or no effect on either gastrointestinal tract colonization or the transition to bacteremia following oral inoculation in neonatal rats. Their results seem to be contradiction to ours, since APEC strains are closely related to the ExPEC strains that are associated with human neonatal meningitis. The contradiction might be due to the differences in the strains (APEC O2 versus NMEC O18) or the different inoculation routes (chickens infected via the air sac versus neonatal rat infected via oral administration). In their study, they determined the proportions of E. coli bacteremia in neonatal rats after oral inoculation but did not quantify the bacterial loads in the bloodstream, which may not fully reflect the virulence difference between the wild type and isogenic mutant. As the authors of that study indicated, their results did not rule out the possibility that SOD activity is important for survival in the bloodstream. Alternatively, previous work has shown that the virulence mechanisms of ExPEC strains can vary. Some may be invasive and internalize within cells and survive in phagocytes, whereas others may be highly refractory to phagocytosis and remain mainly extracellular. Such differences in pathogenic lifestyle could also result in differences in the importance of certain genes or pathways that would be dependent on the pathogenic mechanisms of the particular APEC strain or ExPEC strain. Thus, the previous results on ExPEC suggest that SodA may not always be important for certain types of ExPEC strains.

We also measured the effects of sodA deletion on the expression of other oxidation resistance genes. All of the katE, perR, and osmC genes, which are known to participate in the antioxidant defense mechanism against H₂O₂-induced stress (34–36), had lower transcriptional levels in the sodA mutant (Fig. 7) than in the wild type, which may be due to the loss of function of SodA that results in decreased production of H₂O₂ from intracellular superoxide radicals. The transcription of sodB or sodC was not affected by the disruption of sodA, suggesting that no compensation effect exists between these isozymes. Our previous array data showed that the sodA gene was highly expressed in vivo, while the other two metal-cofactor-containing SODs (sodB and sodC) were normally expressed (23), which implied that SodA may play a more important role than the other two SODs in the extraintestinal infectious process of APEC.

Taken together, our results suggest that sodA plays an important role in protecting against hydrogen peroxide (H₂O₂)-induced oxidative stress and that it contributes to the virulence of APEC strain E058.

MATERIALS AND METHODS

Bacterial strains, plasmids, primers, and growth conditions.

The bacterial strains and plasmids used in this study are described in Table 1. Oligonucleotide primers obtained from Sangon (Shanghai, China) are listed in Table 2. APEC O2 serotype E058 (iss⁺ iut⁺ iroN⁺ tsh⁺ irp⁺ cvaC⁺ sit⁺ mntH⁺) is a highly virulent strain, isolated from a chicken with a respiratory infection (37). Bacteria were routinely grown at 37°C in Luria-Bertani broth (LB) with aeration. When necessary, ampicillin and chloramphenicol were added at 60 and 30 μg/ml, respectively.

TABLE 1.

Bacterial strains and plasmids used in this study

Strain, plasmid, or cell line	Characteristicsa	Source or reference
Strains
E058	Wild-type avian E. coli serotype O2	37
E058 ΔsodA	E058 ΔsodA::cat	This study
ReE058 ΔsodA	Complementation of E058 ΔsodA	This study
DH5α	endA1 hsdR17(rK− mK+) supE44 thi-1 recA1 gyrA (Nalr) RelA1 Δ(lacIZYA-argF)U169 deoR [ϕ80dlacΔ(lacZ)M15]	Invitrogen
Plasmids
pACYC184	Medium-copy-number vector, p15A ori, Cmr Tcr	Gifted
pACYC-sodA	pACYC184 carrying sodA ORF and its putative native promoter	This study
pKD46	Ampr; expresses λ Red recombinase	38
pKD3	cat gene, template plasmid	38
Cell line HD-11	Chicken macrophage line, chicken myelomonocytic transformed by the myc-encoding MC29 virus	42

^aNal^r, nalidixic acid resistance; Cm^r, chloramphenicol resistance; Tc^r, tetracycline resistance; ORF, open reading frame; Amp^r, ampicillin resistance.

TABLE 2.

Primers designed and used in this study

Primer by use	Primer sequence (5′–3′)	Target gene
Deletion
AF	ATGAGCTATACCCTGCCATCCCTGCCGTATGCTTACGATGCCCTGGAACCGCACTTCGATAAGCAGACCATTGTGTAGGCTGGAGCTGCT	sodA
AR	TTATTTTTTCGCCGCAAAACGTGCCGCTGCTTCGTCCCAGTTCACCACGTTCCAGAACTCTTTAATGTAGATGGGAATTAGCCATGGTCC
Complementation
RAF	CTCAAGCTTCCATCGTAATCGCGTTACa	sodA
RAR	CTCTCTAGATTATTTTTTCGCCGCAAAACGb
RT-qPCR
katE RT-F	TTGTGGGAAGCCATTGAA	katE
katE RT-R	GCGATTGAGCACCATTTT	katE
perR RT-F	AGATGACGCCACCCAATA	perR
perR RT-R	GGGCATACCAGTTTACCG	perR
osmC RT-F	CGGGATTCACGCCAACAT	osmC
osmC RT-R	CGGCACCGCAACTTCACT	osmC
sodB RT-F	TATCACTACGGCAAGCACC	sodB
sodB RT-R	CAGGCAGTTCCAGTAGAAAGTA	sodB
sodC RT-F	ACATGAAGGGCCAGAAGG	sodC
sodC RT-R	ATATTATCGCCGCCAACG	sodC

^aHindIII restriction site is in bold.

^bXbaI restriction site is in bold.

Construction of the sodA mutant and its complementation strain.

The sodA mutant was constructed using gene replacement methods based on the lambda Red recombinase system (38). The chloramphenicol resistance cassette, flanked by 5′ and 3′ sequences of sodA, was obtained by amplifying template plasmid pKD3 using primers AF and AR (Table 2). The PCR product was transformed into strain E058 containing pKD46. The mutant (E058 ΔsodA) was confirmed by PCR and verified by sequence analysis.

To construct the complementation strain, the sodA open reading frame and its upstream promoter were amplified and subcloned into pACYC184 using primer pairs RAF and RAR (Table 2). The purified recombinant plasmid pACYC-sodA was transformed into the sodA mutant strain to complement the sodA gene deletion to generate the ReE058 ΔsodA strain.

Oxidative stress experiments.

Overnight (18-h) cultures of the sodA mutant, complemented strain, and wild-type strain E058 were diluted 1:100 into LB broth and incubated until the optical density at 600 nm (OD₆₀₀) reached 0.4. The bacteria were challenged in LB containing 10 mM H₂O₂ for 30 min at 37°C. Viable cells were counted after the oxidative stress challenge, and the percent survival was calculated as the number of CFU per milliliter remaining after the oxidant treatment divided by the initial number of CFU per milliliter times 100.

To determine bacterial growth curves under oxidative stress, overnight cultures were washed twice in phosphate-buffered saline (PBS) and standardized to an OD₆₀₀ of 1.0, and approximately 10⁶ CFU were inoculated into 5 ml LB with or without 10 mM H₂O₂. Bacterial growth was measured every 30 min by spectrophotometry (OD₆₀₀). The experiments were performed in triplicate.

Adaptation to chicken serum and environmental stress.

Complement-sufficient specific-pathogen-free (SPF) chicken serum was prepared and pooled from 10 SPF chickens. A bactericidal assay was performed in a 96-well plate, as described previously, but with the following modifications (39). SPF chicken serum was diluted to 5, 12.5, 25, 50, and 100% in PBS (pH 7.2). Bacteria (10 μl containing 10⁶ CFU) were inoculated into wells containing 190 μl of the diluted SPF chicken serum and heat-inactivated SPF chicken serum and then incubated at 37°C for 30 min. Bacteria were counted by plating on LB agar at serial dilutions. E. coli DH5α was used as the control strain.

Bacterial adaptation to environmental stress was examined as described previously (5). The bacteria were suspended in PBS and diluted to 10⁸ CFU/ml. For the acidic and alkaline adaptation, 100-μl bacterial suspensions were exposed to 900 μl LB broth (pH 4.0) or 100 mM Tris (pH 10.0) for 30 min. For high-osmolarity stress, bacteria were mixed equally with 4.8 M NaCl and incubated for 1 h at 37°C. After environmental stress exposure, bacteria were diluted serially and plated on LB agar. Survival was calculated as the number of bacteria remaining after the stress exposure divided by the initial number of bacteria.

Biofilm assay.

The biofilm assay was carried out as previously described but with a slight modification (40). Overnight cultures were inoculated at 10⁸ CFU/ml in LB or M9 minimal medium (6 g Na₂HPO₄, 3 g KH₂PO₄, 0.5 g NaCl, 1 g NH₄Cl, 2 mM MgSO₄, and 0.1 mM CaCl₂ per liter of water [pH 7.0]) supplemented with 0.4% glucose. Then, 200 μl of bacterial cultures was incubated in 96-well plates at 37°C for 24 h. After three washes with PBS, the samples were stained for 30 min with 0.1% crystal violet, rinsed with sterile PBS, and air dried. Finally, the dye was dissolved in 95% ethanol. The solubilized biofilm was determined by measuring the OD₅₉₅. The experiment was performed in triplicate.

Ingestion and intracellular survival assay.

Ingestion and intracellular survival assays were performed as previously described (41). The avian macrophage cell line HD-11 was cultured in Dulbecco’s modified Eagle’s medium (Gibco, NY) with 10% fetal bovine serum (PAA, Pasching, Australia). Cells were maintained at 37°C in a 5% CO₂ environment with 2 × 10⁵ cells per well in 24-well cell culture plates. Bacteria were added to the cells with a multiplicity of infection of 100 for 1 h to allow ingestion. Wells were washed with PBS, and the appropriate cell culture medium containing 100 μg/ml gentamicin was added for 1.5 h to kill extracellular bacteria. At this time (T₀) and after different incubation periods (2, 4, 6, 8, and 12 h), the macrophages were washed with PBS, lysed with 0.1% Triton X-100, diluted in PBS, and plated onto LB agar plates for APEC CFU determination. The ingestion ratio was determined by dividing the number of ingested bacterial cells at T₀ by the initial bacterial inoculation number. Intracellular growth was expressed as the fold change in the bacterial number at an additional incubation time point (i.e., 2, 4, 6, 8, or 12 h) relative to the number of ingested bacteria at T₀.

In vitro and in vivo competition assays.

For the in vitro competition assay, cultures of wild-type strain E058 and the sodA mutant were mixed in a ratio of 1:1, incubated in LB broth for 4 h at 37°C, and then plated onto LB medium with or without chloramphenicol. For the in vivo competition assay, 5-week-old SPF chickens (White Leghorn; Jinan SPAFAS Poultry Co. Ltd., Jinan, China) were inoculated with cultures of wild-type strain E058 and its E058 ΔsodA mutant or ReE058 ΔsodA complemented strain, mixed in a ratio of 1:1 (1 × 10⁸ CFU for each strain), via the left air sac. The chickens were provided with food and water ad libitum and treated in accordance with the regulations of the Administration of Affairs Concerning Experimental Animals; the protocol was approved by the Animal Care and Use Committee of Yangzhou University (approval ID SYXK [Su] 2007–0005, 21 September 2016). At 24 h after infection, cardiac blood samples (0.2 ml) were collected and suspended in 0.8 ml of PBS. Tissue samples (0.25 g) from livers, lungs, and spleens of the chickens were collected, weighed, triturated in 1 ml of PBS, and homogenized. Serial dilutions were plated on LB medium with or without chloramphenicol to select for the mutants and total bacteria, respectively. The results are shown as the log₁₀ competitive index (CI). The CI was calculated for each mutant by dividing the output ratio (mutant/wild type) by the input ratio (mutant/wild type).

Bacterial colonization ability during systemic infection.

Five-week-old SPF chickens were infected with the wild-type, mutant, and complementation strains via the left thoracic air sac, with each bacterial suspension containing 10⁸ CFU. After 24 h, 10 chickens from each group were euthanized by carbon dioxide asphyxiation, and organs were removed aseptically. Cardiac blood samples (0.2 ml) were collected and suspended in 0.8 ml of PBS. Tissue samples (0.25 g) from livers, lungs, and spleens of the chickens were collected, weighed, triturated in 1 ml of PBS, and homogenized. Tenfold serial dilutions of homogenates were plated onto LB agar plates to determine the numbers of bacteria.

Quantitative real-time reverse transcription-PCR.

To analyze how gene expression was affected by the disruption of sodA, we performed qRT-PCR analyses for various oxidation resistance genes in APEC E058. Overnight cultures of strains were diluted 1:100 in fresh LB medium and grown until logarithmic phase. Following RNA isolation and cDNA synthesis, qRT-PCR was performed to determine the transcription levels of the selected genes using SYBR premix Ex Taq and gene-specific primers (Table 2), and the data were normalized to the gapA transcript.

Statistical analysis.

Differences between groups were analyzed using the Statistical Package for the Social Sciences (SPSS version 15.0; SPSS, Chicago, IL). P values of <0.05 were considered significant.