This study was approved by the Animal Experiment Committee of Shanxi Agricultural University (ethics approval number: SXAU-EAW-2022CE.GH.003007156). A total of fifty 14-day-old specific pathogen-free (SPF) White Leghorn chicks (mixed sex, balanced across groups) were obtained from Beijing Meri Avigon Laboratory Animal Technology Co., Ltd. (Beijing, China) and reared under strictly controlled pathogen-free conditions.

The virulent E. tenella strain (SX010323, Tsx) and the attenuated precocious strain (SX010323P15, PTsx) were obtained from the Laboratory of Veterinary Pathology at the College of Veterinary Medicine, Shanxi Agricultural University in Shanxi, China. The Tsx strain was originally isolated in 2008 from the cecal tissues of broiler chickens showing clinical coccidiosis, characterized by bloody diarrhea and severe cecal lesions, on a commercial poultry farm in Shanxi Province. It had a prepatent period of 141 hours. The PTsx strain was derived from Tsx after 15 successive passages involving precocious development and maintained a stable prepatent period of 120 h. Oocysts were incubated in 2.5% potassium dichromate at 28°C for 3 days. After grinding the oocysts, sporozoites were released and incubated with 0.25% trypsin and 0.5% bile salt at 37°C for 2 h, then washed three times with phosphate-buffered saline (PBS, pH 7.4). Sporozoite viability was checked by blood cell counting and trypan blue exclusion, with activity ≥90%. Excystation of Tsx and PTsx sporozoites was performed as described previously [15].

Chick embryo cecal epithelial cells (2 × 10⁵ per well, in 6-well plates, with 100 U/mL penicillin and 100 μg/mL streptomycin) were isolated from sixty 15-day-old SPF chick embryos (Merial Vital Corp., Beijing, China) following established protocols [16]. These cells were cultured in Dulbecco’s modified Eagle’s medium medium supplemented with 2.5% fetal bovine serum and 0.1 μg/mL epidermal growth factor at 41°C in 8% CO₂.

When cell adherence reached approximately 90%, chick embryo cecal epithelial cells were randomly divided into five experimental groups. The control group was not infected with E. tenella. Treatment groups were exposed to PTsx or Tsx sporozoites at low (2.0 × 10⁵ per well) or high (4.0 × 10⁵ per well) doses. Cells were collected at 4, 24, 72, and 120 hours after infection. Each experiment was independently repeated five times, with culture media refreshed every 48 hours. Infected cells were identified using H&E staining. For each sample, 500 cells were counted in 10 randomly chosen fields at ×400 magnification by a blinded observer. The infection rate was calculated as: (number of infected cells/total cells counted) × 100%.

Fifty 14-day-old SPF chicks with comparable body weights were randomly divided into five groups. The blank control group was not inoculated with E. tenella. Experimental groups received sporulated oocysts (1 mL) of PTsx or Tsx by oral gavage at low (4,500 oocysts/chicken) or high (45,000 oocysts/chicken) doses. Chickens were maintained at 30°C ± 1°C with 55% ± 5% relative humidity under a 12 h light/12 h dark photoperiod and had ad libitum access to anticoccidial-free commercial feed (crude protein ≥18%) and water.

Cell monolayers were fixed with 4% paraformaldehyde at 25°C for 20 min, permeabilized with 0.4% Triton X-100 for 5 min, and blocked with 5% goat serum for 60 min. The cells were then incubated overnight at 4°C with anti-LC3B rabbit antibody (Sigma-Aldrich; lot: 046M4787V) diluted 1:2000, followed by five PBS washes. Subsequently, they were incubated at 37°C for 1 h with Fluorescein isothiocyanate-conjugated goat anti-rabbit IgG secondary antibody (Bioss, Beijing, China) diluted 1:500. After washing five times with PBS, nuclei were stained with Hoechst 33342 (Beyotime, Shanghai, China) for 10 min. The samples were mounted with an anti-fluorescence quenching agent and analyzed under a fluorescence microscope (Olympus BX53, Japan). Nuclei showed blue fluorescence, while autophagosomes appeared as green puncta [17]. LC3II puncta were quantified by counting 100 cells per group using ImageJ software.

Total RNA was isolated from chick embryo cecal epithelial cells at 4, 24, 72, and 120 hours post-infection with E. tenella, as well as from chicken cecal tissues on day 5 after inoculation, using RNAiso Plus (Takara, Code No. 9109, lot: AHF1822A). The RNA samples underwent DNase treatment and were reverse-transcribed into cDNA with the PrimeScript™ RT reagent kit (Takara, Code No. RR047A, lot: AHF1207A). The concentration and purity of the cDNA were determined using a μDrop Duo Ultra microplate reader and a Multiskan SkyHigh spectro-photometer (Thermo Scientific™, USA). Quantitative real-time polymerase chain reaction (PCR) was carried out using TB Green® Premix Ex Taq™ GC (Takara, Code No. RR071A, lot: AHF1068A) on a CFX Connect™ Real-Time PCR Detection System (Bio-Rad). The primers for Beclin-1 were newly designed (F: 5′-TGGCTTTCTT GGACTGTGTG-3′; R: 5′-ACCACCACTGCCACCTGTAT-3′; reference sequence: NM-001006332.1). β-actin was used as the internal control (reference sequence: L08165.1; F: 5′-CACCACAGCCGAGAGAGAAAT-3′; R: 5′-TGACCATCA GGGAGTTCATA GC-3′). The amplification protocol included an initial denaturation at 95°C for 30 seconds, followed by 40 cycles of 95°C for 15 seconds and 59°C for 34 seconds. Relative gene expression levels were calculated using the 2^ΔΔCt method.

Total protein was extracted from chick embryo cecal epithelial cells and chicken cecal tissues using RIPA lysis buffer (Beyotime Biotechnology; lot: 051018180521). Equal amounts of protein (20 μg per sample) were separated on 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis gels under reducing conditions, transferred to nitrocellulose membranes via tank transfer at 120 V for 30 min in Tris-glycine buffer with 20% methanol. Membranes were blocked with Tris-buffered saline with Tween-20 containing 5% skim milk for 2 h at room temperature, then incubated overnight at 4°C with primary antibodies: anti-LC3B rabbit antibody (Sigma-Aldrich; lot: 046M4787V; 1:2000) and anti-β-actin rabbit antibody (CST; lot: 6; 1:2000). Afterward, membranes were incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG secondary antibody for 1 hour, and immunoreactive bands were visualized using an enhanced chemiluminescence substrate (Boster; lot: 13E18B96). Densitometric analysis was performed with ImageJ software. Antibody specificity was confirmed by detecting single bands at ~14 kDa and ~16 kDa for LC3, and ~42 kDa for β-actin, aligning with previous research (Zhang et al., 2023). The LC3II/I ratio was calculated as (LC3II/β-actin) ÷ (LC3I/β-actin).

Data were assessed for normality with the Shapiro–Wilk test and for homogeneity of variance with Levene’s test. One-way analysis of variance followed by Tukey’s multiple comparison test was conducted using SPSS 19.0 (SPSS Inc., Chicago, IL, USA). Data are presented as mean ± SE. Differences were regarded as not significant if p > 0.05, significant if p < 0.05, and highly significant if p < 0.01. Graphs were created with GraphPad Prism 9.2.0.

No significant differences were observed in the infection rates of cecal epithelial cells between groups inoculated with the same dose of PTsx or Tsx at 4, 24, and 72 h post-inoculation (p > 0.05). However, at 120 h, infection rates in the PTsx groups were significantly lower than those in the Tsx groups (p < 0.01). In addition, infection rates in the high-dose groups (4.0 × 10⁵/well) were significantly higher than those in the low-dose groups (2.0 × 10⁵/well) at corresponding time points following E. tenella inoculation (p < 0.01) (Figure 1; Table 1).

The number of punctate LC3II dots in host cells infected with PTsx or Tsx at both low (2.0 × 10⁵/well) and high (4.0 × 10⁵/well) doses was significantly higher than that in the non-inoculated control group at 4–120 h post-infection (p < 0.01). At equivalent doses, PTsx-inoculated cells exhibited significantly fewer LC3II puncta than Tsx-inoculated cells (p < 0.01). Moreover, within each strain, the high-dose groups showed significantly greater LC3II puncta accumulation compared with the corresponding low-dose groups (p < 0.01) (Figure 2; Table 1).

Beclin-1 mRNA expression was significantly upregulated in the PTsx high-dose group (4.0 × 10⁵/well) and in both Tsx-inoculated groups (2.0 × 10⁵ and 4.0 × 10⁵/well) compared with the non-inoculated control at 4–120 h post-infection (p < 0.01). At the same inoculation dose, Beclin-1 mRNA expression was significantly lower in PTsx-infected cells than in Tsx-infected cells (p < 0.01). Furthermore, Beclin-1 mRNA levels were significantly higher in high-dose groups than in low-dose groups for the same strain (p < 0.01) (Figure 3; Table 1).

Western blot analysis demonstrated that the LC3II/I ratio in Tsx-inoculated groups was markedly higher than in both non-inoculated and PTsx-inoculated groups between 4 and 72 h after inoculation (p < 0.05 or p < 0.01). No significant difference was observed between PTsx-inoculated and non-inoculated groups during this period (p > 0.05). At 120 h, the LC3II/I ratio was significantly elevated in the high-dose PTsx group (4.0 × 10⁵/well) and in both Tsx-inoculated groups compared to controls (p < 0.05). At the same doses, PTsx-inoculated cells had significantly lower LC3II/I ratios than Tsx-inoculated cells (p < 0.05 or p < 0.01), and high-dose groups presented higher ratios than low-dose groups for the same strain (p < 0.05 or p < 0.01) (Figure 4; Table 1).

On day 5 post-inoculation, Beclin-1 mRNA expression in the cecal tissues of chickens inoculated with PTsx at 45,000 oocysts/chicken and with Tsx at either 4,500 or 45,000 oocysts/chicken was significantly higher than that in the non-inoculated control group (p < 0.01). Beclin-1 mRNA levels in PTsx-inoculated chickens were significantly lower than those in Tsx-inoculated chickens at corresponding doses (p < 0.01). Additionally, Beclin-1 mRNA expression in the high-dose groups (45,000 oocysts/chicken) was significantly greater than that in the low-dose groups (4,500 oocysts/chicken) for the same strain (p < 0.01) (Figure 5; Table 2).

The LC3II/I ratio in cecal tissues from the PTsx high-dose group and from both Tsx-inoculated groups was significantly higher than that in the non-inoculated group on day 5 post-infection (p < 0.05). At equivalent doses, PTsx-inoculated groups showed significantly lower LC3II/I ratios than Tsx-inoculated groups (p < 0.05 or p < 0.01). Furthermore, LC3II/I ratios were significantly higher in the 45,000 oocysts/chicken groups than in the 4,500 oocysts/chicken groups for the same strain (p < 0.05) (Figure 6; Table 2).

Autophagy markers, including Beclin-1 mRNA expression and LC3II/I ratios, showed consistent strain- and dose-dependent trends across time points in vitro (4–120 h) and in vivo (day 5 post-inoculation), although autophagy activation peaked earlier in vivo. The LC3II/I ratio was approximately 4.0–5.4-fold higher in vivo than in vitro, particularly in high-dose Tsx groups, likely reflecting contributions from inflammatory responses and microbial signals in the intestinal microenvironment. Basal autophagy levels were also higher in cecal tissues in vivo (LC3II/I = 2.9239 ± 0.1828) than in cultured epithelial cells in vitro (0.5706 ± 0.0116). Notably, PTsx-induced autophagy was comparatively mild in vivo and was associated with reduced intestinal damage, supporting the relative safety of PTsx as a live attenuated vaccine candidate (Tables 1–3).

Autophagy is a conserved cellular process in eukaryotic cells in which double-membrane vesicles sequester excess or damaged organelles and deliver them to lysosomes or vacuoles for degradation, thereby facilitating cellular recycling and removal of harmful components. In addition to apoptosis and necrosis, autophagy is recognized as a distinct form of programmed cell death [18]. Several intracellular protozoa, including Toxoplasma gondii and Cryptosporidium parvum, are known to induce host cell autophagy during infection [19, 20]. Previous work by our research group demonstrated that large numbers of double-membrane autophagosomes were formed in chick embryo cecal epithelial cells 72 h after E. tenella infection, indicating that E. tenella can actively induce autophagosome formation in host cells. In parallel, the transcriptional and protein expression levels of autophagy-related factors, including autophagy-related gene (ATG)5, Beclin-1, and LC3, were significantly upregulated, confirming that E. tenella promotes host cell autophagy [14].

The virulent E. tenella strain Tsx strongly activates autophagy and promotes autophagosome–lysosome fusion. Using drugs that modulate autophagy, such as the inducer rapamycin (RAPA) and the inhibitor 3-methyladenine (3-MA), which target the mTOR and Phosphatidylinositol 3-kinase pathways, respectively, can alter host cell autophagy levels, thereby decreasing or increasing E. tenella infection rates. Additionally, chloroquine significantly elevated LC3B-II and p62 levels in host cells infected with E. tenella at 72 h post-infection [14]. In this study, the expression levels of Beclin-1 mRNA, the number of punctate LC3II signals, and the LC3II/I ratio were notably higher in the high-dose precocious strain group and all virulent strain–inoculated groups compared to controls following 4–120 h of in vitro infection. At the same inoculation doses, these autophagy markers were consistently higher in virulent strain groups than in precocious strain groups and higher in high-dose groups than in low-dose groups. These results suggest that the virulent E. tenella strain induces stronger autophagy in host cells than the precocious strain at the same dose, and that autophagy induction correlates positively with infection dose.

Evidence shows that autophagy is essential for host defense against intracellular pathogens, but many pathogens have developed ways to evade or harness autophagy to survive and replicate within host cells [21, 22]. For instance, Plasmodium infection triggers autophagy in liver cells, yet the parasite can grow inside autophagic vesicles [23]. Enhancing host autophagy through drugs increases Plasmodium growth and infection rates, while blocking autophagy reduces parasite development and infection [24]. Similarly, Toxoplasma gondii avoids immune-related autophagy by creating parasitophorous vacuoles and also uses nutrients derived from autophagy to sustain persistent infection [25]. These findings align with our current results showing that the PTsx strain induces weaker autophagy activation compared to the virulent Tsx strain at the same dose, resulting in only mild and temporary activation of the autophagic pathway.

The differential induction of autophagy by Tsx and PTsx may involve multiple signaling pathways, such as mTOR/AMPK (AMP-activated protein kinase) and Ca²⁺-dependent signaling. Tsx’s greater virulence and ability to proliferate likely cause more nutrient stress in host cells, leading to the activation of AMPK and suppression of mTOR, which promotes autophagy [21]. Conversely, PTsx has roughly 40% lower reproductive capacity and shows reduced Ca²⁺ dysregulation [15], potentially diminishing endoplasmic reticulum stress and reactive oxygen species, thus limiting autophagy activation. Additionally, rhoptry proteins (ROPs), key virulence factors, are expressed at lower levels in attenuated strains, possibly affecting autophagy via changes in inflammatory signaling and nutrient requirements.

No notable differences in host cell infection rates were observed between precocious and virulent strains from 4 to 72 h after in vitro inoculation. However, at 120 h, the virulent strain group showed significantly higher infection rates. This may be due to incomplete schizogony of second-generation merozoites in the precocious strain, leading to fewer progeny, lower nutrient needs, and less host cell damage. These phenotypic differences could limit the precocious strain’s ability to induce long-term host cell autophagy, though further validation is needed. While autophagy acts as a cellular defense mechanism, sustained hyperactivation, ike that seen during Tsx infection from 4 to 120 h, may cause negative effects such as nutrient imbalance, cell damage, and immune dysregulation. In early E. tenella infection, the parasite seems to promote host cell autophagy while blocking apoptosis, ensuring nutrient supply and extending host cell survival to support its development.

On day 5 after inoculation in vivo, Beclin-1 mRNA levels and LC3II/I ratios increased in the high-dose precocious strain group and all virulent strain groups compared to non-inoculated controls, except for the low-dose precocious group. At similar doses, autophagy markers were consistently higher in chickens infected with virulent strains than in those with precocious strains, and higher in high-dose groups than in low-dose groups. Notably, Beclin-1 mRNA levels showed similar activation levels in vivo and in vitro (e.g., Tsx high-dose group: 3.1109 ± 0.0393 in vivo vs. 3.0895 ± 0.0412 in vitro), indicating a conserved, strain- and dose-dependent regulation of autophagy initiation. Conversely, LC3II/I ratios were significantly higher in vivo, with increases of 4.0-fold in the PTsx low-dose group and 5.4-fold in the Tsx high-dose group compared to in vitro measurements. This amplification probably reflects the impact of the intestinal microenvironment, including immune cells, microbiota-derived metabolites like lipopolysaccharides and short-chain fatty acids, as well as cytokines such as IFN-γ and IL-1β.

The relatively mild induction of autophagy by PTsx likely helps reduce nutrient depletion in host cells and prevents intestinal damage. This offers a mechanistic explanation for the lesser growth suppression seen in vaccinated chickens compared to those infected with Tsx [8, 9, 11]. The dose-dependent autophagy responses observed support optimizing PTsx vaccination doses (for example, 4,500 oocysts per chicken) to achieve a balance between immune response and safety, while also minimizing adverse effects related to autophagy. Although the study clearly links host cell autophagy, E. tenella strain virulence, and infection dose, it is still unknown whether PTsx’s limited ability to induce autophagy reflects co-evolution between host and parasite. PTsx might improve host cell adaptability by restricting second-generation schizont development and reducing autophagic stress, thus aiding parasite survival. This research primarily examined autophagy marker expression and did not measure autophagic flux with lysosomal inhibitors like chloroquine or bafilomycin A1. Future research incorporating flux assays and exploring interactions between autophagy, innate immune responses, and gut microbiota will be crucial to fully understand the mechanisms behind Eimeria–host interactions.