Effects of the type V secretion system eYadA on the biological properties and pathogenicity of avian pathogenic Escherichia coli

TL;DR

This study investigates how a specific protein in a type of E. coli affects its ability to cause disease in birds and mice.

Contribution

The study reveals the role of eYadA in APEC pathogenicity, including its effects on motility, biofilm formation, and cytokine regulation.

Findings

eYadA influences motility and biofilm formation in APEC.

eYadA affects pathogenicity in mice and ducklings.

eYadA regulates inflammatory cytokines during infection.

Abstract

Avian pathogenic Escherichia coli (APEC) mainly causes local or systemic infections in chickens, ducks, and other avian species. These infections clinically present as various types of inflammation and septicemia, resulting in significant economic losses in the poultry industry. The type V secretion system (T5SS) is prevalent in APEC and serves a crucial role as a virulence factor during the infection process. Trimeric autotransporter adhesins (TAAs), which belong to the Vc subtype of T5SS, are proteins located on the outer membrane of Gram-negative bacteria. Yersinia adhesin A (YadA) is an important model for studying TAA structure, function, and biogenesis; it has also been proven to be a significant virulence factor in other Gram-negative bacteria. To explore the role of eYadA, an autotransporter adhesin belonging to the YadA family, in the pathogenesis of APEC, we constructed both an eyadA deletion strain and its complemented strain. We analyzed their biological characteristics and pathogenicity and found that eyadA influences motility and biofilm formation in APEC as well as its pathogenicity towards both mice and ducklings. Furthermore, it affects the regulation of inflammatory cytokines; changes in these cytokine levels were found to correlate with both infection dose and duration of exposure. This study provides new insights into understanding the pathogenic mechanisms associated with APEC while also suggesting effective strategies for prevention and control measures against this pathogen.

The online version contains supplementary material available at 10.1186/s13567-025-01705-7.

Figures

Full text

Introduction

Avian pathogenic Escherichia coli (APEC) is a pathogenic bacterium that widely infects various poultry species, including chickens, turkeys, ducks, and geese. It can lead to systemic infections such as sepsis, peritonitis, perihepatic inflammation, and meningitis [1, 2]. The intensive production of poultry and the rapidly expanding free-range systems have increased the exposure of these birds to pathogens and stress factors, resulting in a higher incidence of APEC-related diseases in poultry and significant economic losses for the poultry farming industry. The predominant serotypes among clinical isolates are O1, O2, and O78 [3]. However, owing to China’s vast territory with diverse geographical climates and breeding practices across different regions, the dominant serotypes also vary regionally [1, 4]. Moreover, APEC serves as a virulence reservoir for human pathogenic Escherichia coli, sharing numerous similarities in terms of serotypes, virulence factors, and pathogenic mechanisms. APEC can be transmitted to humans through contaminated poultry products, thus posing a potential threat to human health while carrying risks associated with zoonotic diseases [5]. At present, APEC infections in poultry primarily involve vaccination programs and antibiotic therapy. However, the misuse of antibiotics is prevalent in certain regions of the poultry industry. Such abuse not only disrupts the balance of normal flora in avian populations but also complicates treatment efforts against APEC infections. In addition, residual antibiotics may enter the human body via the consumption of contaminated poultry products, further endangering public health [1, 2, 5]. Nevertheless, our understanding of the mechanism of this disease remains limited; only a handful of APEC isolates have been utilized in experimental studies thus far [6]. Therefore, conducting in-depth research on the pathogenic mechanisms of pathogenic Escherichia coli in poultry and enhancing long-term monitoring and investigation of APEC are of significant scientific importance for ensuring the development of the poultry breeding industry and safeguarding public health.

The type V secretion system (T5SS) is widely present in APEC and plays an important role as a virulence factor during the infection process [7, 8]. Trimeric autotransporter adhesins (TAAs), belonging to the Vc subtype of T5SS, are proteins located on the outer membrane of Gram-negative bacteria [9]. These proteins perform multiple functions, including cell adherence, biofilm formation, and contributions to pathogenicity [8]. Yersinia adhesin A (YadA) is a homologous trimer composed of subunits, with each monomer exhibiting an approximate molecular weight of 47 kDa. It serves as an important model for studying TAA structure, function, and biogenesis [10, 11]. Since classical autonomous transporters (monomers) and their trimer counterparts are not homologous, YadA has established itself along with its homologs as a distinct subfamily among autonomous transporters known as TAAs [12]. It was initially identified in Yersinia enterocolitica (Y. enterocolitica), and Yersinia pseudotuberculosis (Y. pseudotuberculosis) [13, 14]. Studies have indicated that YadA mediates biofilm formation and motility in these two bacteria. It promotes bacterial adhesion while enhancing resistance to phagocytosis by macrophages, thereby increasing the pathogenic potential of these bacteria [15–17]. Previous studies have indicated that Yersinia pestis (Y. pestis) also harbors the yadA gene; however, owing to a frameshift mutation affecting TAA expression, further investigation into its function was not pursued by researchers [16]. In recent years, YadA has been detected in a variety of Gram-negative bacteria. In this study, we designated the YadA identified in the APEC TW-XM strain as eYadA to differentiate it from the YadA found in Yersinia. We successfully aligned the amino acid sequence of the eYadA protein with that of the YadA family autotransport adhesin coding sequence, specifically at positions 519882–520898 in the APEC TW-XM genome (the accession number for this protein in Genbank is AUG63398.1, NCBI Reference Sequence: WP_000859812.1). Furthermore, starting from its encoding gene eyadA, we investigated the effect of eYadA on various biological characteristics and pathogenic traits of APEC. These include motility, biofilm formation ability, cell adhesion capacity, and resistance to macrophage phagocytosis. Our aim is to provide new insights that contribute to a deeper understanding of the pathogenesis of APEC TW-XM.

Materials and methods

Ethics statement

Animal experiments were conducted in accordance with the guidelines established by the Regulations for the Administration of Affairs Concerning Experimental Animals, as approved by the State Council of the People’s Republic of China. All procedures received approval from Yangzhou University’s Animal Care and Ethics Committee (SYXK202502093). BALB/c mice (3-week-old) were obtained from Yangzhou University’s Comparative Medicine Center. The mice had unrestricted access to food and water under a 12-h light/dark cycle, with monitoring conducted twice daily. Gaoyou Ducklings (7-day-old) were purchased from Jiangsu Gaoyou Duck Development Company (Yangzhou, China) and provided with antibiotic-free balanced feed along with water throughout the experimental period. Tissue samples were collected from euthanized mice and ducklings following established protocols, with euthanasia carried out using isoflurane anesthesia to minimize suffering.

Bacterial strains, cell lines, and culture conditions

APEC TW-XM strain (O2: K1) and the isogenic ΔeyadA mutant were grown in Luria Bertani (LB) media at 37 °C with continuous agitation (180 rpm). Plasmids pKD3, pKD46, and pCP20 were utilized in the λ-Red-mediated recombination system to generate the eyadA deletion mutant. Plasmid pBR322 was employed to construct the complemented mutant APEC TW-XMΔeyadA/PeyadA, with ampicillin (Amp) selection at a concentration of 100 μg/mL (Solarbio, China). Other bacterial strains and plasmids used in this study are listed in Table 1. hBMEC (human brain microvascular endothelial cells, Procell, China, CP-H124) were cultured in Dulbecco’s modified eagle medium (DMEM, Gibco, Australia) supplemented with 10% fetal bovine serum (FBS, Vazyme, China) and endothelial cell growth supplement (1%, Sciencell, USA), under conditions of 37 °C and 6% CO_2_. RAW264.7 cells (mouse mononuclear macrophages, Procell, China, CL-0190), immortalized bone marrow-derived macrophages (iBMDM, OriCell, China, M3-1001), and HD11 (chicken macrophage cells, IMMOCell, China, IM-C072) cells were cultured in DMEM (Gibco, Australia) supplemented with 10% FBS at 37 °C, 6% CO_2_. Table 1Strains and plasmids used in this studyStrains and plasmidsCharacteristicsSource referencesStrainsAPEC XMWild-type, O2:K1:H7, GenBank: CP025328.1Donated by Prof. Chengping LuAPEC XMΔeyadA**eyadA deletion in APEC XMIn this studyAPEC XMΔeyadA/PeyadAAPEC XMΔeyadA carrying the vector pBR322-eyadA, Amp^r^In this studyPlasmidspKD3Cm^r^; Cm cassette template[18]pKD46Amp^r^; λ-red recombinase expression[18]pCP20Amp^r^; Cm^r^; Flp recombinase expression[18]pBR322-eyadAAmp^r^, pBR322 carrying the entire eyadA nucleotide sequence (GenBank: CP025328.1, 519882–520898)In this study

Construction of eyadA deletion mutant and the complement mutant

The eyadA deletion mutant was constructed using the previously described λ-Red mediated recombination system [18]. Briefly, primers P1 and P2 (Additional file 1) were employed to amplify the chloramphenicol resistance (Cm^r^) cassette from plasmid pKD3. The purified polymerase chain reaction (PCR) product was subsequently introduced into APEC TW-XM containing plasmid pKD46 to generate Cm^r^ recombinant bacteria. Following this, the Cm^r^ cassette was excised using plasmid pCP20. PCR screening with primers P3 and P4 (Additional file 1), along with subsequent DNA sequencing, confirmed that eyadA was completely absent in the APEC TW-XMΔeyadA mutant strain. To construct complementary strains, the full-length sequence of eyadA was cloned into plasmid pBR322 using primer pairs P5 and P6 (Additional file 1). The resulting recombinant plasmid, designated as pBR322-eyadA, was then transformed into the eyadA deletion mutant to create a complementary mutant.

Assessment of bacterial growth and motility

Overnight cultures of APEC TW-XM and its mutant strains were diluted 1:100 into Luria–Bertani (LB) medium and incubated at 37 °C for 12 h. The OD_600_ values were measured using a spectrophotometer (BioTek, Winooski, USA) every hour. Each experiment was conducted in triplicate. The overnight-cultured bacteria were resuspended to achieve a uniform concentration. Subsequently, 0.5 μL of the bacterial suspension was applied onto the surface of standard semi-solid medium and incubated at 37 °C for 36 h. The diameter of the motility ring was measured and photographed for documentation purposes to assess motility levels. A semi-solid medium was prepared containing 1% tryptone, 0.25% NaCl, and 0.25% agar. The outer diameter of the wild-type (WT) strain’s motility ring served as a reference point, designated as 100%. The motility rate of other strains relative to this reference point is calculated using the following formula: (Actual outer edge diameter)/(Average outer edge diameter of WT) × 100%. Each experiment consisted of three parallel runs and was repeated three times.

Assay for biofilm formation capability

Biofilm formation assays were conducted as previously described [19]. Bacteria in the logarithmic growth phase were diluted 1:100 in biofilm-inducing medium, which consisted of tryptone 10.0 g/L, yeast extract 5.0 g/L, NaCl 2.5 g/L, KH_2_PO_4_ 3.0 g/L, K_2_HPO_4_ 7.0 g/L, (NH_4_)2_SO_4 2.0 g/L, FeSO_4_ 0.5 mg/L, MgSO_4_ 1.0 g/L, and thiamine hydrochloride 2.0 g/L. The cultures were statically incubated at 30 ℃ for 36 h [20]. After incubation, the culture medium was carefully discarded, and the tubes were rinsed slowly and gently several times with ultrapure water. Subsequently, 5 mL of a filtered 1% crystal violet solution was added to each tube and allowed to stand at room temperature for 30 min. Following this step, the crystal violet solution was carefully discarded, and the tubes were rinsed thoroughly yet gently with ultrapure water. The biofilm rings formed by each strain were observed and documented photographically on a whiteboard, with experiments repeated three times for consistency. For quantitative analysis, diluted cultures were inoculated into 96-well microtiter plates (Corning, NY, USA) and incubated for 72 h at 30 °C. Following incubation, the culture was carefully removed from the wells and rinsed gently three times with ultrapure water. Bacterial biofilms were then stained with 0.1% crystal violet solution for 20 min and subsequently dissolved in a 95% ethanol solution. The absorbance was measured using a spectrophotometer at a wavelength of 600 nm. Each experiment included eight replicates and was conducted in triplicate to ensure reliability.

Bacterial adherence assay

The binding specificity of APEC TW-XM and different mutants to hBMEC cells was determined by adhesion assay as previously described [21]. A monolayer of about 1 × 10^5^ hBMEC cells (Procell, China, CP-H124) was cultured in 96-well cell culture plates (Corning, NY, USA), and subsequently incubated with bacteria in the logarithmic phase at a multiplicity of infection (MOI) of 100 at 37 ℃ for 1 h. Following this incubation, the cells were washed three times with phosphate buffered saline (PBS) and lysed with 100 μL of 0.5% Triton X-100 at room temperature for 30 min. The total number of cell-associated bacteria was then diluted to a ratio of 1:100 in PBS, plated on LB agar plates, and incubated at 37 ℃ for bacterial enumeration. Each experiment was performed in triplicate.

Assessment of macrophage phagocytosis

The assessment of bacterial resistance to phagocytosis in APEC TW-XM, APEC TW-XMΔeyadA, and APEC TW-XMΔeyadA/PeyadA was conducted as previously described [22]. RAW264.7 cells (Procell, China, CL-0190) were cultured in 48-well plates to form a monolayer and subsequently incubated with bacteria in the logarithmic phase at an MOI of 100 at 37 ℃ for 1 h. After discarding the bacterial solution, DMEM containing gentamicin (200 μg/mL, 150 μL per well) was added to eliminate extracellular bacteria, followed by an additional incubation at 37 °C for another hour. Following this incubation, the cells were washed three times with PBS and lysed using 100 μL of 1% Triton X-100 at room temperature for 30 min. The total number of cell-associated bacteria was then diluted to a ratio of 1:100 in PBS, plated onto LB agar plates, and incubated at 37 ℃ for bacterial enumeration. Each experiment was performed in triplicate. The procedure is the same for both iBMDM (OriCell, China, M3-1001) and HD11 cells (IMMOCell, China, IM-C072).

Median lethal dose, survival rate, and bacterial blood load in mice

Overnight cultures of APEC TW-XM and its mutant strains were diluted 1:100 into LB medium and incubated at 37 °C until the APEC TW-XM, APEC TW-XMΔeyadA, and APEC TW-XMΔeyadA/PeyadA reached the logarithmic phase. The bacteria were then harvested, resuspended in PBS, and individually adjusted to concentrations of 1 × 10^5^ CFUs (100 μL), 1 × 10^6^ CFUs (100 μL), 1 × 10^7^ CFUs (100 μL), 1 × 10^8^ CFUs (100 μL), and 1 × 10^9^ CFUs (100 μL) for inoculation into 3-week-old BALB/c mice (n = 5), respectively. The control group received sterile PBS treatment. Mice were infected by intraperitoneal injection and monitored for 7 days post-infection to assess the median lethal dose (LD_50_). For the survival rate assay, APEC TW-XM, APEC TW-XMΔeyadA, and APEC TW-XMΔeyadA/PeyadA were standardized to a concentration of 1 × 10^8^ CFUs (100 μL) for inoculation into 3-week-old BALB/c mice (n = 10). The clinical symptoms of the mice were systematically monitored following infection, and their morbidity and mortality rates were recorded until all mice had succumbed. To evaluate the bacterial loads in the blood, APEC TW-XM and APEC TW-XMΔeyadA strains were individually adjusted to a concentration of 1 × 10^8^ CFUs (100 μL) during the logarithmic phase for inoculation into 3-week-old BALB/c mice (n = 4) via intraperitoneal injection. At 10-h post-infection, blood samples were collected from the mice through the orbital vein at hourly intervals. Following multiple dilutions, 10 μL of each diluted sample was plated onto LB agar plates for bacterial enumeration after incubation at 37 ℃.

Tissue loading of bacteria and changes in inflammatory factors in ducklings

APEC TW-XM, APEC TW-XMΔeyadA, and APEC TW-XMΔeyadA/PeyadA in the logarithmic phase were individually adjusted to a concentration of 1 × 10^8^ CFUs (200 μL) for inoculation into 7-day-old Gaoyou Ducklings (n = 10) by tracheal injection. The control group (n = 5) received sterile PBS treatment. At 20-h post-infection, the organs (heart, liver, spleen, kidney, pancreas, and brain) of the ducklings were harvested, weighed, and homogenized. Subsequently, they were diluted in a tenfold gradient series and plated onto LB agar plates for bacterial enumeration after incubation at 37 ℃. Each experiment was performed in triplicate. Serum samples were obtained for the subsequent testing. In another set of experiments, ducklings inoculated with 5 × 10^6^ CFUs (200 μL) of bacteria were euthanized at intervals of 24 h, 36 h, 48 h, 60 h, 72 h, and 84 h (n = 3). Aseptically collected tissue samples included the heart, liver, spleen, kidney, pancreas, and brain. Following weighing, the tissue samples were homogenized and subsequently used for bacterial enumeration as previously described. Serum samples were collected from infected ducklings at 36 h and 72 h post-infection. The changes in inflammatory cytokines among different groups were evaluated using enzyme-linked immunosorbent assay (ELISA) according to the manufacturer’s protocol (mlbio, Shanghai, China).

Statistical analyses

All statistical analyses were performed using GraphPad Prism 9.5 software (GraphPad Software, San Diego California USA). LD_50_ values were analyzed using Spearman–Kaeber correlation; survival rates were represented as Kaplan–Meier curves, while other datasets underwent one-way analysis of variance (ANOVA). All figures and data are presented as means ± standard deviations (SD) throughout the text. Significant differences are indicated by p values, with p > 0.05 denoting nonsignificance (ns) and p < 0.05 indicating significance.

Results

Genome sequence and protein analysis of eYadA

We conducted a comprehensive investigation into the YadA sequence across Escherichia coli, Yersinia pseudotuberculosis, and Yersinia enterocolitica. We successfully aligned the eYadA sequence with the coding sequence for the YadA family autotransport adhesin at positions 519882–520898 within the APEC TW-XM genome (GenBank: CP025328.1). Furthermore, phylogenetic analysis of the most prevalent YadA variants revealed that eYadA (Genbank: AUG63398.1) demonstrates a high degree of homology with the YadA protein from Escherichia coli (Additional file 2).

Deletion of eyadA enhances the motility of APEC TW-XM bacteria

Before investigating the role of the eyadA gene in the infection caused by the APEC TW-XM strain, we initially constructed deletion (APEC TW-XMΔeyadA) and complementation (APEC TW-XMΔeyadA/PeyadA) mutants (Figure 1A). The analysis of the growth curve indicated that the deletion of eyadA did not significantly affect bacterial growth characteristics (Figure 1B). However, all three strains exhibited the capability to form distinct motility rings on semi-solid culture media. The deletion of eyadA significantly enhanced the motility of these bacteria (Figures 1C, D), resulting in an increase of approximately 210% compared with that observed in APEC TW-XM (p = 0.001). The motility of APEC TW-XMΔeyadA/PeyadA was found to be 1.35 times greater than that of APEC TW-XM (p = 0.1132). These findings indicate that the absence of eyadA leads to a significant enhancement in motility for strain APEC TW-XM, suggesting that eYadA acts as an inhibitor of motility in this strain.Figure 1The effect of eyadA gene deletion on the motility of the APEC TW-XM strain. A PCR verification of the deletion and complementation mutants of the eyadA gene in the APEC TW-XM strain. M: Trans2K plusII marker, Lane 1: PCR product from APEC TW-XM, Lane 2: PCR product from APEC TW-XMΔeyadA, Lane 3: PCR product from APEC TW-XMΔeyadA/PeyadA. The results of the PCR identification confirmed the successful construction of both the eyadA deletion strain and its complement. B Bacterial growth curves. APEC TW-XM (WT), APEC TW-XMΔeyadA ΔeyadA, and APEC TW-XMΔeyadA/PeyadA (PeyadA) were incubated in LB broth at 37 °C for 14 h with continuous agitation, and OD_600_ optical density was measured hourly throughout the incubation period. C,D Motility of APEC TW-XM, APEC TW-XMΔeyadA, and APEC TW-XMΔeyadA/PeyadA strains. Bacterial motility assays were conducted on semi-solid plates to evaluate the impact of eyadA deficiency on APEC TW-XM motility. The length of the motility loop for each bacterium was measured.

Deletion of eyadA promotes biofilm formation in APEC TW-XM

The biofilm formation capacity of APEC TW-XM, APEC TW-XMΔeyadA, and APEC TW-XMΔeyadA/PeyadA was assessed through qualitative evaluation of biofilm in tubes as well as by quantifying surface-adherent biofilms on 96-well microtiter plates. As shown in Figure 2, the eyadA deletion mutant exhibits a significant enhancement in biofilm formation compared to the APEC TW-XM strain in 30 °C. Under these temperature conditions, all strains were capable of forming circular biofilms in the biofilm induction medium. Notably, the biofilm ring formed by APEC TW-XMΔeyadA exhibited a deeper staining intensity than that observed for APEC TW-XM. The quantitative analysis of biofilms supported the qualitative observations, indicating that the absence of eyadA promotes enhanced biofilm formation in APEC TW-XM. At 30℃, APEC TW-XMΔeyadA showed an approximate increase of 114% compared with its counterpart APEC TW-XM (p = 0.0006). No significant difference was observed between the biofilm formation levels of APEC TW-XMΔeyadA/PeyadA and those observed for APEC TW-XM (p = 0.8665). Therefore, eYadA appears to possess limited inhibitory effects on the formation of biofilms by TW-XM strains.Figure 2Biofilm formation of APEC TW-XM, APEC TW-XMΔeyadA**, and APEC TW-XMΔeyadA/PeyadA was evaluated at temperatures of 30 °C**. A Qualitative analysis of biofilm in tubes (APEC TW-XM, APEC TW-XMΔeyadA, and APEC TW-XMΔeyadA/PeyadA). B The surface-adhered biofilm on 96-well microtiter plates was quantified by measuring OD_600_ of ethanol-solubilized CV (2%) following biofilm staining. Data are shown as mean ± standard deviation from triplicate experiments (n = 8). Using the biofilm formation level of APEC TW-XM as a reference point (i.e., 100%), we will assess the expression levels of biofilms for both APEC TW-XMΔeyadA and APEC TW-XMΔeyadA/PeyadA. The horizontal axis denotes the strain names, while the vertical axis represents the percentage of biofilm expression.

Deletion of eyadA reduces the adhesion of bacteria to hBMEC, while simultaneously hindering bacterial resistance against phagocytosis by macrophages

Considering that APEC TW-AM can induce meningitis in both mice and ducks, thereby posing a potential public health risk, we selected hBMECs, which are essential components of the blood–brain barrier, as a model to assess the impact of eyadA on the adhesion capacity of APEC TW-XM. The results are presented in Figure 3. Compared with APEC TW-XM, APEC TW-XMΔeyadA exhibited a 30% reduction in its adhesion ability to hBMECs (p = 0.0053), while the adhesion ability of APEC TW-XMΔeyadA/PeyadA was comparable to that of APEC TW-XM. To investigate the role of eyadA in resisting macrophage phagocytosis, we conducted experiments using RAW264.7, iBMDM, and HD11 cell lines. Our results indicated that the absence of eyadA led to extensive phagocytosis of bacteria by macrophages, resulting in diminished resistance to phagocytosis. In RAW264.7 cells, APEC TW-XMΔeyadA exhibited an increased phagocytosis rate reaching up to 166% compared with APEC TW-XM. In iBMDM cells, the phagocytosis rate increased by 137%, while in HD11 cells it reached 124%. These findings confirm that eYadA possesses the ability to enhance pathogen resistance against macrophage-mediated clearance.Figure 3Effect of eyadA gene deletion on adhesion and resistance to phagocytosis in strain APEC TW-XM. A Adhesion of APEC TW-XM, APEC TW-XMΔeyadA, and APEC TW-XMΔeyadA/PeyadA to hBMEC cells. Bacteria were coincubated with hBMEC (MOI = 100) at 37 ℃ for 1 h. The cell-associated bacteria were subsequently diluted on LB agar plates for bacterial enumeration. B Phagocytosis of APEC TW-XM, APEC TW-XMΔeyadA, and APEC TW-XMΔeyadA/PeyadA by RAW264.7, iBMDM, and HD11 cells. Bacteria were coincubated with these three cells (MOI = 100) respectively at 37 ℃ for 1 h. Unphagocytosed bacteria were eliminated using gentamicin before dilution on LB agar plates for bacterial enumeration. Each experiment was performed in triplicate.

The deletion of the eyadA gene reduces the pathogenicity of APEC TW-XM in BALB/c mice

APEC TW-XM has been reported to induce systemic infections in BALB/c mice. Therefore, we employed BALB/c mice as an infection model to investigate the impact of the eyadA gene on the pathogenicity of APEC TW-XM strains. Given that no clinical symptoms were observed within 72 h following intragastric infection of mice with varying doses of the APEC TW-XM strain, we opted for intraperitoneal injection to establish a mouse infection model. The results of the LD_50_ determination are presented in Table 2. The LD_50_ for APEC TW-XM is calculated at 5.01 × 10^6^ CFUs, while that for APEC TW-XMΔeyadA is 2 × 10^8^ CFUs. In addition, the LD_50_ for APEC TW-XMΔeyadA/PeyadA is recorded at 3.16 × 10^8^ CFUs. Compared with APEC TW-XM, there is a notable increase in the LD_50_ of APEC TW-XMΔeyadA and a corresponding decrease in fatality rate, indicating that deletion of the eyadA gene reduces the virulence of APEC TW-XM. Subsequently, we conducted a mouse challenge study to assess survival rates and blood bacterial loads following infection. Within 24 h post-infection, all mice in the APEC TW-XM group exhibited severe illness with no survivors; conversely, all mice in the APEC TW-XMΔeyadA group demonstrated a survival rate of 100% (Figure 4A). Assessments of blood bacterial load confirmed that the absence of eyadA reduces the survival capability of the APEC TW-XM strain in murine circulation. Following entry into circulation, proliferation trends for the APEC TW-XM strain were observed over time; however, bacterial loads in mice from the ΔeyadA group decreased significantly within 1 h post-infection and remained low for up to 9 h thereafter (Figure 4B). Furthermore, deletion of eyadA genes significantly mitigates pathological damage inflicted by APEC TW-XM on mouse tissues, as illustrated in Additional file 3. The changes in inflammatory factors among these mice infected with different bacteria also indicated that deletion of eyadA resulted in a significant reduction in inflammatory lesions induced by APEC TW-XM (Additional file 4). These findings suggest that eyadA deletion mutants exhibit reduced pathogenicity in mice. Table 2Determination of the LD50 for APEC TW-XM, APEC TW-XMΔeyadA**, and APEC TW-XMΔeyadA/PeyadAStainDead number/total number1 × 10^5^1 × 10^6^1 × 10^7^1 × 10^8^1 × 10^9^LD_50_APEC TW-XM0/52/52/55/55/55.01 × 10^6^ΔeyadA0/50/50/51/55/52 × 10^8^ΔeyadA/PeyadA0/50/50/51/54/53.16 × 10^8^Control0/50/50/50/50/5–Three-week-old BALB/c mice were infected via intraperitoneal injection with APEC TW-XM, APEC TW-XMΔeyadA, and APEC TW-XMΔeyadA/PeyadA in the logarithmic growth phase. The number of mouse fatalities was recorded over a period of 7 days post-infection, and the LD_50_ of each strain was determined using a modified version of Karber’s method.Figure 4Survival curves and bacterial blood load in mice**. A Survival curves of APEC TW-XM, APEC TW-XMΔeyadA, and APEC TW-XMΔeyadA/PeyadA infected mice. The 3-week-old BALB/c mice (n = 10) were intraperitoneally infected with an equal dose of bacteria, and their morbidity and mortality rates were monitored until all mice had succumbed. The horizontal axis represents the time interval, while the vertical axis indicates the cumulative survival rate. B Bacterial load dynamics in mouse blood. Blood samples were collected from mice via the orbital vein at hourly intervals following infection with APEC TW-XM and APEC TW-XMΔeyadA (n = 4). After performing multiple dilutions, 10 μL of each diluted sample was plated onto LB agar plates for bacterial enumeration, which was conducted after incubation at 37 ℃.

The eyadA deletion mutant elicits a mild inflammatory response in ducklings, which is influenced by both the infection dose and the duration of exposure

Considering that APEC TW-XM was isolated from the brains of ducks exhibiting symptoms of sepsis and meningitis, we conducted infection experiments to investigate the impact of eyadA gene deletion on the pathogenicity of APEC TW-XM using 7-day-old Gaoyou ducklings. In the group infected with 1 × 10^8^ CFUs (200 μL) of APEC TW-XM, the ducklings exhibited symptoms including lethargy, unsteady posture or paralysis, depression, diarrhea within 20 h post-infection. Subsequently, these affected ducklings succumbed to the infection in rapid succession. Pathological changes included fibrinous pericarditis characterized by marked epicardial congestion, perihepatitis associated with inflammatory adhesions, and multifocal whitish necrotic foci distributed across the splenic surface. The pancreas displayed severe liquefactive necrosis accompanied by diffuse congestion. In addition, cerebral edema was observed alongside vascular engorgement. Collectively, these findings suggest a systemic infectious process with significant inflammatory and necrotizing components. In contrast, no congestion was observed in the hearts and livers of ducklings from the APEC TW-XMΔeyadA group. There were no apparent necrotic lesions present in their spleens. The pancreas retained a normal shape and color without any signs of congestion or edema in the brain. Ducklings did not exhibit any obvious clinical symptoms within 72 h following infection with 5 × 10^6^ CFUs (200 μL) of bacteria. The ducklings maintained normal neurobehavioral status and physiological parameters, as evidenced by well-groomed plumage, consistent feed and water intake, absence of gastrointestinal disturbances (particularly diarrhea), appropriate alertness and responsiveness to environmental stimuli, and excellent viability with minimal mortality. The results of bacterial load assessments in various tissues and organs of the ducklings indicated that, compared with APEC TW-XM, the bacterial load in the tissues of ducklings from the APEC TW-XMΔeyadA group was significantly reduced (Figure 5A). In the low-dose infection group (5 × 10^6^ CFUs), bacterial loads were observed only in the hearts of ducklings from the APEC TW-XMΔeyadA group and in both the hearts and pancreas of those from the APEC TW-XMΔeyadA/PeyadA group at 48–60 h post-infection (Figure 5B). In summary, the absence of eyadA appears to reduce colonization and survival rates of APEC TW-XM in duckling tissues while attenuating its pathogenicity towards these animals.Figure 5Bacterial tissue load and serum levels of inflammatory cytokines in ducklings. A Tissue load assessment in ducklings infected with 1 × 10^8^ CFUs (200 μL) bacteria. APEC TW-XM, APEC TW-XMΔeyadA, and APEC TW-XMΔeyadA/PeyadA strains were inoculated into 7-day-old Gaoyou ducklings (n = 10) via tracheal injection during their logarithmic growth phase. At 20 h post-infection, the organs (heart, liver, spleen, kidney, pancreas, and brain) were harvested from the ducklings. These samples were weighed and homogenized for further analysis. Following a series of dilutions, 10 μL from each diluted sample was plated onto LB agar plates to enumerate bacterial colonies after incubation at 37 °C. B Tissue load assessment in ducklings infected with 5 × 10^6^ CFUs (200 μL) of bacteria. Tissue samples were collected at intervals of 24 h, 36 h, 48 h, 60 h, 72 h, and 84 h post-infection (n = 3). Each sample was subsequently weighed and homogenized before being utilized for bacterial enumeration as previously described. All experiments were conducted in triplicate. C Changes in inflammatory cytokines in ducklings infected with 1 × 10^8^ CFUs (200 μL) of bacteria. Serum samples were collected 20 h post-infection and subsequently analyzed using an ELISA assay, in accordance with the manufacturer’s protocol, to assess the concentrations of interleukin (IL)−1β, IL-10, and tumor necrosis factor (TNF)-α in the serum. D Changes in inflammatory cytokines in ducklings infected with 5 × 10^6^ CFUs (200 μL) of bacteria. Serum samples were collected from infected ducklings at 36 h and 72 h post-infection. Statistical analysis was performed using one-way ANOVA, with significance determined by p value; specifically, a p value greater than 0.05 indicates non-significance (ns), while a p value of 0.05 or less denotes statistical significance.

The detection results for serum inflammatory cytokines were consistent with these findings. In the group infected with 1 × 10^8^ CFUs (200 μL) of bacteria, the contents of IL-1β, IL-10, and TNF-α were lower in the APEC TW-XMΔeyadA group compared with those in APEC TW-XM. Notably, significant differences were observed for IL-10 and TNF-α (p < 0.05), suggesting that eyadA plays a role in regulating inflammatory cytokine responses (Figure 5C). In another experimental group infected with bacteria at a concentration of 5 × 10^6^ CFUs (200 μL), two time points for detection were established on the basis of outcomes from tissue bacterial load experiments. At 36 h post-infection, the contents of inflammatory cytokines IL-1β, IL-10, and TNF-α in serum samples from ducklings infected with APEC TW-XMΔeyadA were lower than those observed in ducklings infected with APEC TW-XM; however, no significant difference was noted. At 72 h post-infection, serum concentrations of IL-1β and IL-10 remained lower among ducklings infected with APEC TW-XMΔeyadA compared with their counterparts infected with APEC TW-XM, while the content of TNF-α increased relative to that seen in ducklings infected with APEC TW-XM (Figure 5D). This observation may suggest that by this later stage post-infection (72 h), an imbalance within the immune response occurs among ducklings, characterized by an enhanced proinflammatory reaction coupled with diminished antiinflammatory regulation.

Discussion

Autotransporters (AT) represent the largest group of secretory proteins and outer membrane proteins found in Gram-negative bacteria, with trimeric autotransporter adhesins (TAAs) being significant members of this family [23, 24]. TAAs encompass a diverse array of surface adhesins secreted by pathogenic Gram-negative bacteria, playing a crucial role in the pathogenic process and facilitating evasion of the host immune response. Among TAAs, YadA stands out as a prominent representative; it effectively mediates bacterial biofilm formation and motility, promotes bacterial adhesion, and enhances resistance to macrophage phagocytosis [16, 22]. These functions significantly contribute to the overall pathogenicity of these bacteria.

The APEC TW-XM strain was initially isolated from the brain of a duck exhibiting septicemia and neurological symptoms [25]. It has also been reported to have potential zoonotic transmission, as it possesses the capacity to induce meningitis in mice [20, 25]. To investigate the function of eyadA in APEC TW-XM, deletion and complemented mutants for eyadA were constructed in this study, and the biological characteristics of ΔeyadA strains were assessed. It was observed that the absence of eyadA did not impede normal bacterial growth; however, it significantly enhanced the motility of APEC TW-XM. It is hypothesized that eYadA may exert a negative regulatory effect by inhibiting the expression of motility-related genes or modulating associated signaling pathways (such as those involved in flagellar synthesis [26, 27]); however, specific molecular mechanisms warrant further investigation. Through qualitative and quantitative analyses using crystal violet staining, we found that biofilm formation ability was markedly increased in eyadA deletion strains compared with wild-type strains. This finding suggests that eYadA may play a role in inhibiting biofilm formation by the APEC TW-XM strain.

Pathogen adhesion to host cells represents the initial step in the onset of infection [28]. Linke et al. [9] demonstrated that YadA plays a critical role in the colonization of intestinal mucosa and the establishment of infection. In our study, we also found that the deletion of eyadA resulted in reduced adhesive capabilities. Furthermore, we found that the survival rates of the eyadA deletion strain were significantly reduced in RAW264.7, iBMDM, and HD11 macrophage cells. Notably, the eyadA deletion strain exhibited a significant reduction in virulence compared with the wild-type strain. In the wild-type group, there was a continuous increase in blood bacterial load observed, whereas the deletion strain exhibited a decrease by two to three orders of magnitude. This finding aligns with observations made by Ackermann et al. [29], which demonstrated that YadA confers serum resistance and promotes invasion and colonization by pathogenic bacteria. Furthermore, analysis of inflammatory cytokine levels in serum from infected mice revealed that concentrations of IL-1β, IL-10, IL-6, and TNF-α in the eyadA deletion group decreased by 83%, 20%, 12.5%, and 19% respectively when compared with those in the wild-type strain. These results indicate that eYadA may exacerbate tissue damage through specific modulation of proinflammatory and antiinflammatory balance [30]. Considering that certain cytokines, such as IL-10, are subject to negative feedback regulation and can inhibit proinflammatory responses while simultaneously downregulating other cytokines, it is plausible for other cytokines to be upregulated to assume a compensatory role [31]. The differentiated regulation of inflammatory cytokines fundamentally arises from mutation specificity, the complexity of the host signaling network, and the dynamic balance between pathogen–host interactions. S100A8 and S100A9 serve as important clinical markers for assessing the degree of inflammation within the body [32, 33]. In this study, we measured the levels of S100A8 and S100A9 proteins in serum samples obtained from mice across different infection groups. The results indicated that the serum concentration of S100A8 in mice from the APEC TW-XMΔeyadA group was 115 pg/mL, representing a significant decrease compared with that observed in the APEC TW-XM group (p < 0.05). Furthermore, we found that the serum level of S100A9 in mice from the APEC TW-XMΔeyadA group was 1630 pg/mL (Additional file 4). This represents a significant reduction when compared with the levels found in the APEC TW-XM group (p < 0.001). These findings suggest that inflammation induced by APEC TW-XMΔeyadA is less severe than that caused by APEC TW-XM.

To accurately reflect the interaction characteristics between pathogens and hosts under natural infection conditions, a specific infection model for ducklings was constructed. By dynamically monitoring tissue bacterial loads, the pathogenic role of eYadA in avian hosts was elucidated. The virulence assessment results showed that the APEC TW-XMΔeyadA strain (LD_50_ = 3.16 × 10^6^ CFUs) exhibited a significant reduction in virulence compared with the APEC TW-XM strain (LD_50_ = 5.01 × 10^4^ CFUs). This finding is consistent with the trend observed in LD_50_ determination experiments conducted on mice. The experiment assessing bacterial load in duckling tissues demonstrated that at an infection dose of 1 × 10^8^ CFUs, the bacterial load across various tissues from ducklings infected with the APEC TW-XMΔeyadA group significantly decreased when compared with those infected with APEC TW-XM. At an infection dose of 5 × 10^6^ CFUs, a notable reduction in bacterial load was observed in both the heart and pancreas of ducklings from the APEC TW-XMΔeyadA group across six time points when compared with those infected with APEC TW-XM. However, by the fourth time point (60 h post-infection), increases in bacterial loads were noted in the liver, spleen, kidneys, and brain. This phenomenon may be attributed to the liver’s central role in metabolism, detoxification, and immunity [34]. The spleen serves as a peripheral lymphoid organ integral to immune function; it participates in blood filtration, storage, and metabolic regulation. In addition, the kidneys are responsible for filtering blood, excreting waste products, and regulating endocrine functions [35]. These organs play crucial roles in the body’s resistance when virulence factors colonize the host at various stages. In ducklings from both the APEC TW-XMΔeyadA group and those from the APEC TW-XMΔeyadA/PeyadA group, bacterial loads were only detected in their hearts at 48–60 h post-infection. This observation may be owing to a low infection dose coupled with a minimal number of bacteria present during the early stages of infection. Consequently, there exists a minimum detection limit when employing plate counting methods, which can impede accurate assessment of bacterial load at certain time points. The combination of these two infection doses indicates that the absence of eyadA significantly reduces both colonization and survival rates of APEC TW-XM within duckling tissues while also reducing its pathogenicity. This finding aligns with research conducted by Tsugo et al. [36] concerning the pathogenic effects of YadA in Yersinia enterocolitica using murine models, as well as our findings on eYadA presented in this study. It indicates that both YadA from Yersinia enterocolitica and eYadA from APEC TW-XM play significant roles in facilitating the invasion and colonization processes of pathogenic bacteria within host organisms. Furthermore, results from multi-tissue colonization assays (heart, liver, spleen, etc.) demonstrate a reduction in bacterial loads by 4–5 orders of magnitude for deletion strains. This suggests that eYadA enhances pathogen dissemination throughout host tissues during disease progression. In addition, analysis of inflammatory cytokine levels in serum samples from infected ducklings revealed that eyadA may influence the regulation of inflammatory cytokines. Given that various inflammatory cytokines are modulated by distinct signaling pathways and each cytokine does not exert a singular regulatory function [37], the absence of eyadA may disrupt some pathways involved in this regulation, resulting in altered expression levels of specific inflammatory cytokines. This modulation reflects an intricate interplay between the complexity inherent in host signaling networks and the dynamic balance that governs interactions between pathogens and their hosts. In addition, the alterations in the levels of inflammatory cytokines present in the serum of ducklings are closely associated with both the infection dose and the infection cycle. This specific regulatory mechanism is intricately linked to adhesin-mediated signal transduction at the pathogen–host interface, thereby providing a theoretical foundation for developing targeted prevention and control strategies aimed at blocking virulence factors.

In conclusion, the absence of eyadA can significantly reduce the virulence of the APEC TW-XM strain and effectively mitigate the proinflammatory response triggered by pathogen infection at both in vivo and cellular levels in animal models. This finding underscores that eYadA is a crucial virulence factor that facilitates infections not only within native animals but also across species due to APEC TW-XM strains. The results provide a theoretical foundation and technical support for effective prevention and treatment strategies against APEC TW-XM in clinical practice, along with valuable insights for the future development of efficient vaccines and antibacterial interventions.

Supplementary Information

**Additional file 1. Primers used in this study.**Additional file 2. Analysis of the sequence and Phylogenetic tree for YadA. A. Sequence analysis of YadA. The YadA sequences from Yersinia enterocolitica, Yersinia pseudotuberculosis, and Escherichia coli were retrieved from the National Center for Biotechnology Information (NCBI) and used for sequence analysis. A substantial number of consecutive identical YadA sequences are observed across various strains (such as blue, cyan, and other regions characterized by uniform color blocks), indicating that these areas have remained highly conserved throughout evolutionary history. B. Phylogenetic tree analysis of YadA. A phylogenetic analysis of the YadA protein sequences was conducted using MEGA 7.0 [38]. Initially, the ClustalW algorithm was employed for multiple sequence alignment, followed by manual refinement of conserved loi. The phylogenetic tree was constructed utilizing the Neighbor-Joining (NJ) method. The reliability of the branches was assessed through bootstrap analysis with 1000 repeated samplings. The target sequence (eYadA) demonstrates the highest homology with Escherichia coli YadA.Additional file 3. H&E staining of pathological tissue sections from infected mice. Following intraperitoneal infection with a concentration of 1 × 10^8^ CFUs (100 μL) of APEC TW-XM or APEC TW-XMΔeyadA, mice were observed for a duration of 12 h before the preparation of tissue samples for paraffin-embedding and subsequent hematoxylin and eosin (H&E) staining. Histopathological changes were examined using microscopy (BX53, Olympus Corporation, Japan).Additional file 4. **ELISA analysis of immune factor variations in mice. **The whole blood from infected mice was collected through the orbital vein and subsequently centrifuged at 1500 × g for 15 min at 4 ℃ to isolate serum. Serum samples from different groups (mice inoculated with APEC TW-XM, APEC TW-XMΔeyadA, and APEC TW-XMΔeyadA/PeyadA) were used for the ELISA test in accordance with the manufacturer’s protocol (Colorful-Gene Biotech, Wuhan, China).