Roles of B739_1343 in iron acquisition and pathogenesis in Riemerella anatipestifer CH-1 and evaluation of the RA-CH-1ΔB739_1343 mutant as an attenuated vaccine

Abstract

Iron is one of the most important elements for bacterial survival and pathogenicity. The iron uptake mechanism of Riemerella anatipestifer (R. anatipestifer, RA), a major pathogen that causes septicemia and polyserositis in ducks, is largely unknown. Here, the functions of the putative TonB-dependent iron transporter of RA-CH-1, B739_1343, in iron utilization and pathogenicity were investigated. Under iron-starved conditions, the mutant strain RA-CH-1ΔB739_1343 exhibited more seriously impaired growth than the wild-type strain RA-CH-1, and the expression of B739_1343 in the mutant strain restored growth. qRT-PCR results showed that the transcription of B739_1343 was not regulated by iron conditions. In an animal model, the median lethal dose (LD₅₀) of the mutant strain RA-CH-1ΔB739_1343 increased more than 10⁴-fold (1.6×10¹² CFU) compared to that of the wild-type strain RA-CH-1 (1.43×10⁸ CFU). In a duck co-infection model, the mutant strain RA-CH-1ΔB739_1343 was outcompeted by the wild-type RA-CH-1 in the blood, liver and brain of infected ducks, indicating that B739_1343 is a virulence factor of RA-CH-1. Finally, immunization with live bacteria of the mutant strain RA-CH-1ΔB739_1343 protected 83.33% of ducks against a high-dose (100-fold LD₅₀) challenge with the wild-type strain RA-CH-1, suggesting that the mutant strain RA-CH-1ΔB739_1343 could be further developed as a potential live attenuated vaccine candidate for the duck industry.

Introduction

Iron is an essential element for most bacteria [1, 2]. The ability of pathogenic bacteria to obtain iron from their host is a key determinant of virulence [3]. Conversely, to protect against invading pathogens that steal iron, hosts sequester iron using host iron-binding proteins, such as transferrin, lactoferrin and ferritin, and hemin-containing protein, such as hemoglobin, hemopexin, myoglobin and leghemoglobin [4]. To overcome host iron-withholding defenses, most bacterial pathogens have evolved highly sophisticated systems to acquire iron for successful infection [5, 6].

Riemerella anatipestifer (R. anatipestifer, RA) is the etiological agent of acute septicemia and infectious polyserositis in ducks, chickens, geese, and other avian species [7]. R. anatipestifer infection can give rise to high contagiosity and mortality among farm ducks, resulting in major economic losses in the poultry industry [8]. According to research reports, at least 21 serotypes of R. anatipestifer without cross-protection have been identified around the world [9–12]. Among clinically isolated strains, R. anatipestifer serotypes 1, 2 and 10 are the most prevalent, with a high level of virulence in China [13]. However, little is known about the pathogenicity or virulence of this pathogen based on nutritional metabolism associated with iron acquisition. The TonB-dependent receptor TbdR1 (Riean_1607) of R. anatipestifer strain CH3 has been reported to be involved in iron acquisition and virulence [14]. The ferric iron utilization gene B739_1208 of R. anatipestifer CH-1 was also recently shown to be involved in virulence [15]. However, the R. anatipestifer genome encodes at least 31 TonB-dependent receptors (TbdRs) [14], and the physiological roles of these putative TbdRs are mostly unknown.

Here, B739_1343 is demonstrated to be required for R. anatipestifer CH-1 growth under iron-limited conditions. Investigation of the virulence of R. anatipestifer RA-CH-1ΔB739_1343 showed that this strain is attenuated in ducks and outcompeted by the wild-type strain RA-CH-1. Furthermore, an examination of protective efficacy revealed that the mutant can be used as a live attenuated vaccine candidate for protecting ducks from R. anatipestifer infection.

Results

In silico analyses of the R. anatipestifer B739_1343 gene

It was shown that R. anatipestifer GSM15868 encodes at least 31 putative TbdRs [14]. The homologues of them are present in the genome of R. anatipestifer CH-1, one of which is B739_1343. B739_1343 of R. anatipestifer CH-1 has low identity with TbdR1 of R. anatipestifer CH-3 [14] (7.29% identity) and B739_1208 of R. anatipestifer CH-1 [15] (3.4% identity). Similar to other characterized outer membrane iron transporters, the N-terminal region of B739_1343 contains a putative TonB interaction site (ETVVV, residues 95–99), and the conserved structure of the protein comprises a beta-barrel and a plug domain occupying the pore inside the barrel [16, 17]. According to the NCBI, the B739_1343 protein is annotated as an outer membrane receptor involved in inorganic ion utilization. Additionally, the results of BLAST analyses revealed that the B739_1343 protein sequence shares over 98% identity in all sequenced R. anatipestifer strains, indicating that this sequence is highly conserved among different R. anatipestifer isolates.

The mutant strain RA-CH-1ΔB739_1343 exhibits reduced iron utilization in iron-starved conditions

To identify the function of B739_1343, the mutant strain RA-CH-1ΔB739_1343 was constructed and verified as described in the “Materials and Methods” section. After the B739_1343 gene was knocked out, its ability to affect the growth of R. anatipestifer CH-1 in tryptone soy broth (TSB) liquid medium was evaluated. As shown in Fig 1, disruption of B739_1343 did not damage the growth of R. anatipestifer CH-1 in TSB liquid medium. Moreover, the addition of the iron chelator 2,2'-dipyridyl at a final concentration of less than 80 μM impaired the optimal growth of both RA-CH-1ΔB739_1343 and the wild-type strain at the same level (data not shown). However, the addition of the iron chelator 2,2'-dipyridyl at a final concentration of 120 μM impaired the maximal growth rate, e.g., the slope of the log-transformed growth curve during the exponential phase, of RA-CH-1ΔB739_1343 more significantly than that of wild type. Complementation of RA-CH-1ΔB739_1343 with a plasmid expressing B739_1343 restored growth to wild-type levels. In addition, although R. anatipestifer CH-1 and RA-CH-1ΔB739_1343 did not grow well in TSB containing 120 μM Dip, the addition of 300 μM iron (III) chloride to cultures restored the growth of all the strains (Fig 1). The two strains expressed equivalent levels of the downstream gene (B739_1342) of B739_1343 (S1 Fig), confirming that the growth defect of RA-CH-1ΔB739_1343 was due to a specific B739_1343 mutation rather than a polar effect on gene expression levels. Overall, these data suggest that B739_1343 plays a pivotal role in R. anatipestifer CH-1 iron acquisition under iron starvation.

Fig 1. Growth curves for RA-CH-1pLMF03, RA-CH-1ΔB739_1343pLMF03, and RA-CH-1ΔB739_1343pLMF03::B739_1343 in TSB, TSB supplemented with 120 μM Dip and TSB supplemented with 120 μM Dip and 300 μM iron (III) chloride.

Cells were grown in 20 mL of TSB medium or TSB medium supplemented with Dip or TSB medium supplemented with Dip and iron (III) chloride at 37°C, starting at OD600 = 0.1. OD600 values were measured every 2 h for 14 h. Data were analyzed using two-way ANOVA. The error bars represent the standard deviations of three independent experiments and two replicate samples for each experiment (n = 3).

To obtain further support for this hypothesis, the bacteria were also tested for growth on TSA plates with or without Dip. As shown in Fig 2A, deletion of B739_1343 had no effect on the growth of the bacteria on TSA plates. However, R. anatipestifer CH-1ΔB739_1343pLMF03 was not able to grow on TSA plates containing 50 μM Dip, and the RA-CH-1ΔB739_1343pLMF03::B739_1343 complementation strain restored growth (Fig 2B). These results obtained on a solid medium agree with those obtained in a liquid medium, indicating that the B739_1343 gene of R. anatipestifer CH-1 is involved in ferric iron utilization.

Fig 2. Growth of RA-CH-1pLMF03, RA-CH-1ΔB739_1343pLMF03, and RA-CH-1ΔB739_1343pLMF03::B739_1343 on TSA and TSA supplemented with 50 μM Dip.

The R. anatipestifer strains (clockwise from top left) RA-CH-1pLMF03, RA-CH-1ΔB739_1343pLMF03, and RA-CH-1ΔB739_1343pLMF03::B739-1343 were grown on TSA plates containing cefoxitin (1 μg/mL) and 0 μM Dip (A) or 50 μM Dip (B). Growth was assessed by the appearance of bacterial colonies on plates. Pictures were taken after 48 h of growth at 37°C. All the experiments were repeated three times. Representative plates are presented.

The transcription of B739_1343 is not regulated by iron

In other bacteria, such as Vibrio vulnificus and Corynebacterium diphtheriae, iron uptake-related genes are negatively regulated by iron [18–21]. Thus, we sought to determine whether B739_1343 is also regulated according to this model. The transcription levels of B739_1343 in TSB (iron-rich medium) and TSB supplemented with 200 μM Dip (iron-depleted medium) were measured via qRT-PCR as described in the “Materials and Methods” section. The results showed no significant effect on the transcription of B739_1343 under iron depletion (S2 Fig). However, the control gene, B739_0103, was upregulated ~30-fold under the same conditions (S2 Fig). Sequence analysis showed that the B739_0103 promoter region possesses classic R. anatipestifer Fur boxes (ATTTATTTTTATTCTAAAT) [12], whereas the promoter of B739_1343 does not. These results are consistent with the results of a previously reported RNA-Seq analysis of RA-CH-1 in TSB and TSB supplemented with Dip [22].

The B739_1343 deletion attenuates the virulence of RA-CH-1 in ducklings

B739_1343 is involved in iron utilization by RA-CH-1; therefore, it was hypothesized that B739_1343 may also be involved in the virulence of the R. anatipestifer CH-1 strain. Thus, the LD₅₀ was evaluated by infecting 3-day-old ducklings with RA-CH-1pLMF03, RA-CH-1ΔB739_1343pLMF03 and RA-CH-1ΔB739_1343pLMF03::B739_1343, as described in the “Materials and Methods” section. The mortality of the ducks was observed for 7 days post-challenge. The calculated LD₅₀ value of the RA-CH-1ΔB739_1343 mutant was 1.6×10¹² CFU, whereas the LD₅₀ value of RA-CH-1pLMF03 was 1.43×10⁸ CFU. The LD₅₀ of the complementation strain was 8.66×10⁹ CFU, indicating that the B739_1343 gene plays an important role in the virulence of R. anatipestifer CH-1.

B739_1343 contributes to the colonization of R. anatipestifer CH-1 in vivo

To further investigate whether B739_1343 contributes to the colonization dynamics of R. anatipestifer during systemic infection, a competitive experiment was conducted in which each duck was subjected to intramuscular inoculation with a 1:1 ratio of the wild-type strain RA-CH-1 and the mutant strain RA-CH-1ΔB739_1343, as described in the “Materials and Methods”. At 24 h post-inoculation, the bacterial load of the mutant strain in the heart blood, livers and brains of the ducks was significantly reduced compared with that of the wild-type strain (Fig 3A). At 48 h post-inoculation, the difference in the bacterial load between the wild-type strain and the mutant strain was more apparent in the heart blood (5×10³-fold reduction, P<0.0001), liver (3.9×10³-fold reduction, P = 0.0001) and brain tissue (10⁵-fold reduction, P<0.0001) (Fig 3B). Thus, the RA-CH-1ΔB739_1343 mutant was significantly outcompeted by the RA-CH-1 wild-type strain in these tissues, suggesting that B739_1343 contributes to the ability of R. anatipestifer CH-1 to colonize the blood and liver and disseminate to the brain.

Fig 3. Competition assay of wild-type RA-CH-1 (filled symbols) and the mutant strain RA-CH-1ΔB739_1343 (open symbols) in vivo.

The wild-type strain RA-CH-1 (10⁹ CFU) and RA-CH-1 ΔB739_1343 (10⁹ CFU) were mixed at a 1:1 ratio and injected into the leg muscles of 3-day-old ducklings. At 24 h (A) and 48 h (B) post-infection, bacteria were isolated from the livers, brains and blood according to the method described in the “Materials and Methods” section. The data points represent the CFU/g of individual animals in the indicated organs; the bars show the median values (n = 6).

Immunization and determination of serum antibody levels

Since the mutant strain RA-CH-1ΔB739_1343 showed significantly attenuated pathogenicity, the potential use of RA-CH-1ΔB739_1343 as a live attenuated vaccine against infection by the virulent RA-CH-1 wild-type strain was evaluated. First, injection of the mutant strain was evaluated to determine whether it affected the health of ducklings. Ducklings immunized with RA-CH-1ΔB739_1343 showed no visible differences in habits or appetites compared with uninjected ducklings. The average body weights and body weight gains during the observation period of the ducklings immunized with RA-CH-1ΔB739_1343 (group 2) did not differ significantly from those of the ducks in the group injected with PBS (group 1), the group immunized with the inactivated RA-CH-1 vaccine (group 3) or the group receiving no immunization (group 4) (Table 1). Thus, the mutant strain RA-CH-1ΔB739_1343 is a suitable attenuated vaccine candidate.

Table 1. Duck body weight before challenge and body weight gain (mean±SD).

Group	Immunization	Body weight before vaccination (g)a	Body weight gain after vaccination (g)
			D3	D6	D9	D12
1	PBS	86.25±7.02 (p = 0.9264)	66.1±5.38 (p = 0.1236)	141.1±11.48 (p = 0.5214)	245.4±19.97 (p = 0.2792)	331.9±27.01 (p = 0.3925)
2	RA-CH-1ΔB739_1343	85.65±7.6 (p = 0.9936)	66.85±5.94 (p = 0.1154)	143.8±12.77 (p = 0.3988)	229.5±20.37 (p = 0.8611)	337.1±29.93 (p = 0.4647)
3	Inactivated RA-CH-1 vaccine	85.5±7.54 (p = 0.9742)	61.1±5.39 (p = 0.5150)	141±12.44 (p = 0.5452)	229.9±20.28 (p = 0.8419)	326.1±28.77 (p = 0.3034)
4	-	85.7±6.67	58.2±4.53	134.8±10.48	226.6±17.62	353.2±27.47

^aNumber of ducks in a group (n = 20).

Next, the ducks inoculated with RA-CH-1ΔB739_1343 were evaluated for a specific humoral immune response. To determine the serum antibody response after vaccination, serum samples were collected from vaccinated ducks at days 7, 14, 21, 28, 35, 42, and 49 after primary immunization, and serum antibodies against R. anatipestifer CH-1 were tested using indirect ELISA, as described previously [23]. As a control, negative sera samples were also collected from pre-immunized ducks. As shown in Fig 4, starting on day 7 and day 14 after immunization, the RA-CH-1ΔB739_1343 vaccine elicited higher IgY titers than the inactivated RA-CH-1 vaccine (Fig 4). Furthermore, the serum antibody levels of both the RA-CH-1ΔB739_1343 vaccine group and the inactivated RA-CH-1 group remained high for at least 49 days after immunization. However, the IgY antibody titer elicited by inactivated RA-CH-1 was higher than that elicited by RA-CH-1ΔB739_1343 at day 21 (Fig 4). These results suggested that the RA-CH-1ΔB739_1343 vaccine enhanced the specific humoral immune response in ducks.

Fig 4. Serum antibody titers of ducks vaccinated with RA-CH-1ΔB739_1343 and the inactivated RA-CH-1 vaccine (n = 5).

Serum was collected from the ducks at 7-day intervals from day 7 to day 49 post-vaccination, and serum antibodies against RA-CH-1 were quantified using indirect ELISA. The antibody titers represent the highest dilutions that produced positive results. The data represent the average value from five serum samples for each group. The statistical significance of the data was ascertained with Student’s T test. This experiment was performed using three independent experiments and three replicate samples in each experiment (n = 3).

Evaluation of RA-CH-1ΔB739_1343 as an attenuated vaccine

To examine whether the immunization of ducks with RA-CH-1ΔB739_1343 can prevent infection by wild-type RA-CH-1, challenge experiments were performed as described in the “Materials and Methods” section. After challenge with the wild-type strain RA-CH-1, the survival rates of the ducks in groups 1, 2 and 3 were 25%, 85% and 100%, respectively (Table 2). The surviving ducks in groups 2 and 3 remained healthy, whereas three of the five surviving ducks in group 1 showed clinical signs that included loss of weight and appetite, opisthotonus and drowsiness. Thus, the morbidity of group 1 was 90% (Table 2). After challenge, the average body weights and body weight gains of groups 2, 3 and 4 did not differ significantly (data not shown). Given the above results, RA-CH-1ΔB739_1343-vaccinated ducks were 83.3% protected from challenge by the RA-CH-1 wild-type strain, indicating that the RA-CH-1ΔB739_1343 attenuated mutant strain can be used as a live vaccine candidate.

Table 2. Animal challenge experiment.

Group	Immunization	Challenge straina	No. of deaths (total)	No. showing morbidity (total)	Mortality (%)	Morbidity (%)	Protectionb (%)
1	PBS	RA-CH-1	15(20)	18(20)	75%	90%	-
2	RA-CH-1ΔB739_1343	RA-CH-1	3(20)	3(20)	15%	15%	83.33%
3	Inactivated RA-CH-1 vaccine	RA-CH-1	0(20)	0(20)	0	0	100%
4	-	-	0(20)	0(20)	0	0	-

^aThe challenge strain dose was 100 LD₅₀.

^bThe protection rate was calculated as [1−(% Morbidity in vaccinate/% Morbidity in control)]×100

Discussion

Iron acquisition is an important aspect of pathogenesis for many pathogens, and bacterial strategies for the acquisition of iron have been described for many decades [24]. It has been shown that the inactivation of genes involved in iron acquisition in Shigella [25], Y. pestis [26], and P. luminescens [27] attenuates virulence in animal infection models. However, relatively little is known about the iron uptake machinery and the role that iron plays in the physiology and virulence of R. anatipestifer. The major objective of this study was to explore the role of B739_1343 (annotated as a putative TonB-dependent iron transporter) in the ferric iron utilization and pathogenesis of R. anatipestifer CH-1.

Our previous research showed that knockout of the ferric iron utilization gene B739_1208 in RA-CH-1 impaired growth in both iron-rich and iron-limited media [15]. Similarly, knockout of the ferric iron utilization gene tbdR1 (Riean_1607) in R. anatipestifer strain CH3 also impaired growth in both iron-rich and iron-limited media [14]. In contrast, B739_1343 deletion affected growth only in iron-limited conditions, suggesting that there are multiple putative TonB-dependent iron transporters involved in iron utilization in RA-CH-1. The components of the iron uptake system and the mechanism involved are under investigation. Nevertheless, through growth analysis, we demonstrated that iron is required for optimal RA-CH-1 growth and that B739_1343 is important for iron uptake in an iron-limited environment. These conclusions were strengthened by the results of the growth assay on iron-limited TSA plates. However, in contrast to classic iron utilization-related genes, the transcription of B739_1343 was not upregulated under iron-limited conditions. The “Fur box” of R. anatipestifer was recently identified [12] but was not found in the promoter region of B739_1343. In addition to B739_1208 [15], B739_1343 is the second iron utilization-related gene that is not regulated by iron.

For most bacterial pathogens, including R. anatipestifer, iron acquisition in an iron-limited host is critical to virulence during infection [15, 28]. In R. anatipestifer CH-3, mutation of the putative TonB-dependent iron transporter tbdR1 attenuated virulence by approximately 45-fold [14], whereas mutation of the siderophore-interacting protein sip attenuated virulence by approximately 35-fold [28]. In R. anatipestifer CH-1, the putative TonB-dependent iron transporter B739_1208 was attenuated approximately 15-fold [15]. In this study, the LD₅₀ of the RA-CH-1ΔB739_1343 mutant was found to be increased by more than 10⁴-fold in a duck infection model, and expression of the B739_1343 gene provided a competitive advantage for the colonization of tissues in a duck co-infection model. Thus, B739_1343 is likely to play a more important role than other TbdRs in the virulence of R. anatipestifer, although without any growth effect in an iron-rich environment.

To evaluate whether the B739_1343 mutant can serve as an attenuated vaccine candidate, different doses of the B739_1343 mutant were inoculated into ducklings through the leg. Here, the highest dose that did not affect the health of ducklings was chosen for immunization. Unexpectedly, at days 7 and 14 after immunization, the IgY titers of these ducklings were higher than those of ducklings immunized with the inactivated vaccine. However, the inactivated vaccine elicited higher IgY titers than the mutant strain after day 21. These findings suggest that the live bacteria stimulated the immune system of ducks more quickly than the inactivated vaccine, although the rate of protection was lower than that when the ducks were immunized with the inactivated vaccine. As a next step, the RA-CH-1ΔB739_1343 mutant could be combined with adjuvant or immunopotentiator to increase the potential protection. In addition, it will be interesting to challenge with heterologous serotypes to determine whether the candidate vaccine protects against one or multiple different serotypes. Studies should also evaluate whether the RA-CH-1ΔB739_1343 mutant stimulates the cell-mediated immune response to obtain further insight into the mechanism of protection. This work is the first to report a live attenuated R. anatipestifer vaccine candidate harboring a deletion in a gene involved in iron homeostasis.

In summary, compared with the wild-type strain RA-CH-1, R. anatipestifer CH-1 lacking B739_1343 is significantly deficient in growth under iron-starved conditions. The B739_1343 gene is required for bacterial virulence, and the RA-CH-1ΔB739_1343 mutant can be used as a live attenuated vaccine for protecting ducks against R. anatipestifer CH-1.

Materials and methods

Bacterial strains, plasmids and primers

The bacterial strains and plasmids used in this study are described in Table 3. The primers used in this study are described in S1 Table.

Table 3. Strains and plasmids used in this study.

Strains and plasmids	Genotype or serotype	Source or reference
Strains
XL1-BLUE	F- supE44 hdsR17 recA1 endA1 gyrA46 thi relA1 lac- F’ proAB- lacIq lacZΔM15 Tn10, Tetr	Laboratory collection
S17-1	Thi-1 thr leu tonA lac Y supE recA::RP4-2-Tc::Mu Kanr	[29]
S17-1 pEX18GM::B739_1343usd	S17-1 carrying pEX18GM::B739_1343usd, Kanr, Genr	This study
RA-CH-1	R. anatipestifer serotype 1	Laboratory collection
RA-CH-1ΔB739_1343	R. anatipestifer CH-1 B739_1343::spcR, Spcr	This study
RA-CH-1ΔB739_1343 pLMF03::B739_1343	R. anatipestifer CH-1 B739_1343::spcR carrying pLMF03::B739_1343	This study
Plasmids
pEX18GM	oriT+, sacB+, gene replacement vector with MCS from pUC18, Genr	[30]
pAM238	pSC101 origin, Spcr	[31]
pEX18GM::B739_1343usd	pEX18GM carrying B739_1343usd from R. anatipestifer CH-1 Genr	This study
pLMF03	B739_0921 promoter, oriColE1, ori pRA0726, Ampr, Cfxr	[32]
pLMF03::B739_1343	pLMF03 carrying B739_1343 from R. anatipestifer CH-1, Ampr, Cfxr	This study

Amp^r, ampicillin resistance; Gen^r, gentamicin resistance; Kan^r, kanamycin resistance; Spc^r, spectinomycin resistance; Cfx^r, cefoxitin resistance.

Media and growth conditions

The preparation of 2,2’-dipyridyl (Dip) (Sigma–Aldrich, St. Louis, MO, USA) has been described elsewhere [32]. E. coli strains were grown on LB medium (Sigma-Aldrich, Product Number: L3522) at 37°C. Solid media contained 1.5% agar (Difco). The R. anatipestifer strains were grown on LB plates supplemented with 5% defibrinated sheep blood or TSA plates (tryptone soy broth, TSB, containing 1.5% agar) at 37°C. Iron-depleted medium was obtained by the addition of Dip. Antibiotics were added at the following final concentrations: ampicillin (Amp): 100 μg/mL, kanamycin (Kan): 50 μg/mL, and gentamicin (Gen): 20 μg/mL, for E. coli; and spectinomycin (Spec): 80 μg/mL, and cefoxitin (CfxA): 1 μg/mL for R. anatipestifer CH-1.

Construction of the B739_1343 gene deletion in R. anatipestifer CH-1

The B739_1343 gene of R. anatipestifer CH-1 was deleted via allelic exchange through a recombinant suicide vector, pEX18GM [30]. Here, the B739_1343 gene was replaced with a 1140-bp SpcR cassette according to a method described elsewhere [33]. Briefly, the 796-bp left flanking sequence and the 803-bp right flanking sequence of the B739_1343 gene of R. anatipestifer CH-1 were amplified via PCR using the primer pairs B739_1343upP1 plus B739_1343upP2 and B739_1343downP1 plus B739_1343downP2, respectively (S1 Table). The SpcR cassette was amplified from the plasmid pAM238 [31] using the primer pairs SpcRP1 plus SpcRP2. The three resultant PCR fragments (B739_1343 upstream, B739_1343 downstream and SpcR cassette) were ligated using overlap PCR and digested with KpnI and BamHI. The resultant fragments were cloned into the pEX18GM plasmid to generate pEX18GM::B739_1343usd. Then, pEX18GM::B739_1343usd was further introduced into cells of the CaCl₂-competent E. coli strain S17-1. pEX18GM::B739_1343usd was transferred to the recipient strain R. anatipestifer CH-1 through conjugation as described elsewhere [33]. The transconjugants were screened using blood agar plates supplemented with Kan (50 μg/mL) and Spec (80 μg/mL). The gene deletion mutant strains were identified by PCR by amplifying the conserved 16S rRNA gene of R. anatipestifer using the primers 16S rRNA P1 and 16S rRNA P2 and the deleted gene using the corresponding primers B739_1343compP1 and B739_1343compP2 (S3 Fig).

Construction of the RA-CH-1ΔB739_1343 pLMF03::B739_1343 complementation strain

To construct the RA-CH-1ΔB739_1343 complementation strain, the B739_1343 gene of R. anatipestifer CH-1, with its own promoter, was amplified using the primers B739_1343compP1 and B739_1343compP2, which contained SalI and XbaI restriction sites, respectively (S1 Table). The fragments were cut by these two enzymes and cloned into the shuttle plasmid pLMF03 [32]. The resulting plasmid, pLMF03::B739_1343, was transformed into cells of the CaCl₂-competent strain E. coli S17-1, and the recombinant plasmid was introduced into the RA-CH-1ΔB739_1343 mutant strain via conjugation as described elsewhere [32]. The transconjugants were selected using blood agar plates supplemented with Cfx (1 μg/mL) and Kan (50 μg/mL) and identified by PCR amplification of 16S rRNA and B739_1343 using the primer pairs 16S rRNA P1 plus 16S rRNA P2 and B739_1343compP1 plus B739_1343compP2, respectively (S1 Table and S3 Fig). The resulting strain was designated RA-CH-1ΔB739_1343pLMF03::B739_1343.

In vitro growth rate determination

The in vitro growth rates of the test strains were determined by measuring the optical density (OD) at 600 nm with a spectrophotometer (Eppendorf Biophotometer, Germany). Briefly, early exponential-phase cultures were inoculated into 20 mL of TSB or TSB supplemented with Dip (120 μM), Dip (120 μM) and Fe(NO₃)₃ (300 μM) at OD600 0.1, followed by incubation at 37°C with shaking at 180 rpm. The OD was determined at 600 nm every 2 h for 14 h. The experiment was performed using three independent experiments with two replicate samples for each experiment.

Iron utilization experiment on TSA plates

RA-CH-1pLMF03, RA-CH-1ΔB739_1343pLMF03 and RA-CH-1ΔB739_1343 pLMF03::B739_1343 were inoculated onto 5% sheep blood plates, which were then incubated overnight at 37°C. The bacterial strains were subsequently collected, re-suspended in 1 mL of PBS and centrifuged for 5 min at 6,000 rpm. This operation was repeated three times to wash the bacteria. The OD₆₀₀ values of the bacterial suspensions were then checked and adjusted to OD₆₀₀ = 1. Next, the standardized strains were diluted to 10⁴ bacteria/mL (1OD₆₀₀ = 6×10⁸ bacteria), and a 20-μL (approximately 200 bacteria) sample of each strain was inoculated onto a TSA plate or a TSA plate containing 50 μM Dip. Growth was recorded after a 2-day incubation at 37°C.

qRT-PCR

RA-CH-1 was inoculated into 20 mL of TSB medium and 20 mL of TSB medium with 200 μM Dip at an OD600 of 0.05 at 37°C (the glassware was deferrated by acid-washing). After 6–8 h of incubation (corresponding to mid-log growth phase), the bacteria were immediately mixed with a two-fold volume of RNA protect Bacteria Reagent (Qiagen: 76506) and centrifuged again at 5,000 g for 10 min. RNA extraction and reverse transcription were performed as described elsewhere [32]. qPCR was conducted using SYBR Green Master Mix (Vazyme:Q111-01) and primers at 0.2 μM. Each experiment consisted of three biological replicate samples with three technical replicates each. The fold change was calculated as described in reference [34] with the delta delta Ct method to consider the efficiency of the PCR reaction for each target, and recA served as the reference gene [32].

LD₅₀ determination

The bacterial LD₅₀ was measured to evaluate virulence as previously described [35]. Briefly, for each strain, 3-day-old Pekin ducks were randomly divided into four groups (10 ducks/group). The ducks were then injected intramuscularly with 10⁷, 10⁸, 10⁹, or 10¹⁰ CFU of each bacterial strain and were examined every 4–6 hours for seven days. Once the ducks exhibited signs of moribundity, including depression, lack of movement or refusal of food, they were euthanized via forced inhalation of CO₂, and identification of R. anatipestifer was subsequently performed. The mortality of the ducks was recorded daily for seven days post-challenge. The LD₅₀ was calculated using the Reed-Muench method [36].

In vivo competition assay

A competition assay was performed in vivo as previously described by Hagan [16] and Wang et al [15]. Briefly, bacterial strains were grown to exponential phase in TSB medium and collected via centrifugation at 6,000 g. This operation was repeated three times to wash the bacteria. The re-suspended strains were adjusted to 10¹⁰ CFU per mL. The standardized RA-CH-1 and RA-CH-1ΔB739_1343 mutant strains were mixed at a 1:1 ratio, and 200 μL of the mixture containing 10⁹ CFU of each strain was injected intramuscularly into the same 3-day-old Pekin ducks. At 24 h and 48 h post-inoculation, the ducks were euthanized by forced inhalation of CO₂. Heart blood, liver and brain tissue were collected at 24 h and 48 h post-inoculation (from six ducks at each time point). Liver and brain samples were weighed and homogenized in PBS. Dilutions of heart blood and the homogenates were then plated on TSA agar to determine the CFU of bacteria per mL of heart blood or per gram of tissue. Additionally, dilutions of heart blood and homogenate were also plated on TSA agar containing Spec (80 μg/mL) to differentiate RA-CH-1 and RA-CH-1ΔB739_1343.

Immunization and challenge

To investigate whether RA-CH-1ΔB739_1343 could be used as an attenuated live vaccine candidate, the RA-CH-1ΔB739_1343 strain and an inactivated RA-CH-1 vaccine (Chengdu Tecbond Biological Products Corporation, Sichuan, China) were prepared for immunization. Briefly, the mutant strain RA-CH-1ΔB739_1343 was cultured in TSB at 37°C to exponential phase. Then, RA-CH-1ΔB739_1343 was collected and re-suspended in PBS. The re-suspended strain was adjusted to 5×10⁸ CFU per mL. Each duck was intramuscularly injected with 200 μL of the standardized strain RA-CH-1ΔB739_1343 containing 10⁸ CFU of bacterial cells. The inactivated RA-CH-1 vaccine was subcutaneously injected, as a control, into the neck according to the manufacturer’s recommendations.

Three-day-old Pekin ducks were randomly divided into four groups (twenty ducks per group). Group 1 was injected with PBS; group 2 was immunized with RA-CH-1ΔB739_1343; and group 3 was immunized with the inactivated RA-CH-1 vaccine. Group 4 was not subjected to immunization and challenge and was used as a control. To evaluate the safety of RA-CH-1ΔB739_1343, the habits, appetites, mental status and other clinical manifestations of the vaccinated ducks were observed for 12 days post-vaccination. Furthermore, the average bodyweights of the ducklings in all four groups were recorded every three days until challenge.

On day 12 after immunization, the ducks in groups 1, 2 and 3 were challenged with wild-type R. anatipestifer CH-1 via intramuscular injection at a dose of 2.28×10¹⁰ CFU (100-fold LD₅₀) per duck. Deaths were recorded, and clinical manifestations were observed daily for 10 days after challenge to evaluate the protection rate of the vaccine. Similarly, moribund ducks were euthanized by forced inhalation of CO₂. The protection rate was calculated as described by Sandhu [37] with some modification as follows: [1−(%Morbidity in vaccinated/%Morbidity in control)]×100.

Determination of serum antibody titers via ELISA

A total of 10 3-day-old Pekin ducks were randomly assigned to two groups (five ducks per group) and immunized intramuscularly with RA-CH-1ΔB739_1343 or subcutaneously in the neck with the inactivated RA-CH-1 vaccine. Blood samples were collected before vaccination (D0) as a control and weekly thereafter until D49 after primary immunization.

For antigen preparation, cultured R. anatipestifer CH-1 was grown in TSB medium to exponential phase, then harvested via centrifugation at 8,000 g for 10 min at 4°C and washed twice with PBS buffer. The bacterial pellets were re-suspended in 25 mL of buffer (20 mM Tris–HCl pH 7.4, 10 mM EDTA, 1 mM TLCK) and lysed using a French press. Cellular debris was removed via centrifugation at 8,000×g for 30 min at 4°C, and the protein concentration of the R. anatipestifer CH-1 lysate was determined using a BCA Protein Assay Kit (Thermo Scientific, USA). ELISA was performed as described previously using 1 μg/well of R. anatipestifer CH-1 [23]. Briefly, 96-well ELISA plates were coated with the R. anatipestifer CH-1 lysate antigen in 100 μL of bicarbonate buffer (pH 9.6) and incubated at 4°C overnight. The plates were then washed with PBS containing 0.1% Tween-20 (PBST) three times and blocked with 1% BSA in PBST at 37°C for 1 h. After blocking, serial-diluted duck serum (from 1:20 to 1:20,480) was added to the wells, followed by incubation at 37°C for 2 h. Thereafter, the plates were washed three times with PBST, and horseradish peroxidase (HRP)-conjugated goat anti-duck IgY (1:5,000 dilution) (ab112771, Abcam) was added. The plates were then incubated at 37°C for 1 h, washed three times with PBST, and 100 μL of soluble TMB substrate solution (TIANGEN, China) was added to each well. The reaction was stopped by adding 100 μL of 2 M H₂SO₄, and the plates were read at 450 nm using a 680 microplate reader (Bio-Rad, USA). The highest dilutions of the sera with an OD₄₅₀ value 2.1 times that of the negative control wells were used as the ELISA titers. The experiment was performed using three independent experiments with three replicates samples for each experiment.

Animals and ethics statement

One-day-old Pekin ducks were purchased from Grimaud farms in Chengdu (Sichuan, China) and housed at our animal facilities with free access to food and water.

This study was performed in accordance with the recommendations of the local animal welfare bodies and the Sichuan Agricultural University ethics committee (SYXK2014-187). The protocol was approved by the Sichuan Agricultural University ethics committee.

Sequence analysis

The homology of the B739_1343 sequences was analyzed using the Basic Local Alignment Search Tool (BLAST) algorithm (http://blast.ncbi.nlm.nih.gov/). Prediction of protein structure was performed using Phyre² programme online (http://www.sbg.bio.ic.ac.uk/phyre2/html/page.cgi?id=index) [38]. Multiple sequence alignments of the B739_1343 sequences were performed using the program Clustal W2 [39].

Statistical analysis

Statistical analysis was performed using GraphPad Prism 6 software and SPSS statistics 20 for Windows. The statistical significance of the data was ascertained using Student’s T test. A value of P<0.05 was considered significant.

Supporting information

S1 File. ARRIVE checklist.

(PDF)

S1 Fig. The transcriptional levels of B739_1342 in RA-CH-1ΔB739_1343 and RA-CH-1.

Quantitative real-time PCR analysis of the relative expression of B739_1342 in RA-CH-1ΔB739_1343 and RA-CH-1 in TSB. The fold change was calculated with the delta delta Ct method to consider the efficiency of the PCR reaction for each target gene. The error bars represent the standard deviations of three independent experiments (n = 3).

(TIF)

S2 Fig. The fold change in the transcriptional levels of RA-CH-1 B739_1343 and B739_0103 in TSB and TSB supplemented with 200 μM Dip.

Quantitative real-time PCR analysis of the relative expression of RA-CH-1 B739_1343 (A) and B739_0103 (B) mRNA in TSB and in TSB supplemented with 200 μM Dip. The fold change was calculated with the delta delta Ct method to consider the efficiency of the PCR reaction for each target. The error bars represent the standard deviations of three independent experiments (n = 3).

(TIF)

S3 Fig. Characterization of the R. anatipestifer CH-1 mutant strain (RA-CH-1ΔB739_1343) and the complementation strain (RA-CH-1ΔB739_1343pLMF03::B739_1343) by PCR.

(A) Verification of the deletion of B739_1343 by PCR. Lane M, BM5000 DNA Marker (Biomed, Beijing, China). Lane 1 and Lane 2: 16S rRNA (960 bp) was amplified from RA-CH-1 and RA-CH-1ΔB739_1343 using the primers 16S rRNA P1 and 16S rRNA P2, respectively. Lanes 3–5: The SpcR cassette (1140 bp) was amplified from the plasmid pAM238, RA-CH-1 and RA-CH-1ΔB739_1343 using the primers SpcR P1 and SpcR P1, respectively. Lane 6 and Lane 7: The B739_1343 gene (2352 bp) was amplified from RA-CH-1 and RA-CH-1ΔB739_1343 using the primers B739_1343compP1 and B739_1343compP2, respectively. Lane 8, Lane 9 and Lane 10: The sacB gene (1422 bp) was amplified from the plasmid pEX18GM, RA-CH-1 and RA-CH-1ΔB739_1343 using the primers SacB P1 and SacB P2, respectively. (B) Verification of the complementation strain RA-CH-1ΔB739_1343pLMF03::B739_1343 by PCR. Lane M: BM5000 DNA Marker (Biomed, Beijing, China). Lane 1: 16S rRNA (960 bp). Lane 2: B739_1343 gene (2352 bp). Lane 3: SpcR cassette (1140 bp). Lane 4: CfxA resistance gene (638 bp).

(TIF)

S1 Table. Primers used in this study.

(DOCX)

Data Availability

All relevant data are within the paper and its Supporting Information files.

Funding Statement

This work was supported by the National Natural Science Foundation of China (Grant No. 31772772, http://www.nsfc.gov.cn/, Recipient: M.L.), International S&T Cooperation Program of Sichuan Province (Grant No. 2016HH0052, http://www.scst.gov.cn, Recipient: M.L.), and the China Agricultural Research System (CARS-42-17, http://www.moa.gov.cn, Recipient: A.C.). The funders (M.L. and A.C.) had a role in study design, decision to publish and perparation of the manuscript.