The animal experiment was approved by the Yichun Laboratory Animal Welfare and Ethics Committee (Approval Number: YCU-2025024), following the principles of animal welfare. Twelve healthy adult Ganxi goats (30-month-old, 30–40 kg) were purchased from Yichun Local Black Goat Breeding Co., Ltd., with a 7-day acclimation period before the experiment to exclude pre-existing infections. The goat-derived LM strain used in this study was isolated from a clinical case of caprine neurological listeriosis in our laboratory (serotype 4b, confirmed by multiplex PCR). The infection group (n = 6) received a single jugular injection of 10 mL LM suspension (1 × 10⁷ CFU/g); controls (n = 6) received 10 mL sterile 0.9% NaCl (Drevets et al., 2004; Oevermann et al., 2010). Animals were monitored daily for clinical signs (lethargy, ataxia, neurological symptoms). When typical neurological signs appeared (72–96 h post-infection), all goats were anesthetized prior to euthanasia via intravenous injection of pentobarbital sodium (60 mg/kg body weight; Sigma-Aldrich, St. Louis, United States) administered through the jugular vein. Euthanasia was then performed by exsanguination of the carotid artery under deep anesthesia, in accordance with the American Veterinary Medical Association (AVMA) Guidelines for the Euthanasia of Animals (2020 edition). Brain tissues (cerebrum, cerebellum, midbrain, pons, medulla oblongata) were collected, with portions fixed in 4% paraformaldehyde (for histology) or stored at −80 °C (for molecular analysis).

Briefly, the brain tissue homogenates (10% w/v in PBS) were serially diluted and plated on TSA agar. Colonies were counted after 24 h incubation at 37 °C, and results were expressed as CFU/g tissue.

Paraffin-embedded sections (5 μm) were deparaffinized in xylene, antigen-retrieved in gradient ethanol, and antigen-retrieved in citrate buffer (pH 6.0, 95 °C for 20 min). Sections were blocked with 5% BSA (Solarbio, Beijing, China) for 1 h at 37 °C, then incubated with primary anti-LM antibody (1:800; Abcam, Cambridge, UK) at 4 °C overnight. After 3 washes with PBS (0.01 M, pH 7.4), sections were incubated with Alexa Fluor® 594-conjugated secondary antibody (1:500; Thermo Scientific, Waltham, United States) for 1 h at 37 °C. Nuclei were stained with DAPI (Beyotime, Shanghai, China) for 5 min. Images were captured using an inverted fluorescence microscope (Ti-S, Nikon, Tokyo, Japan) at 200× magnification, with 5 random fields per section analyzed.

Apoptotic cells were detected using the TUNEL BrightGreen Apoptosis Detection Kit (Vazyme) according to the manufacturer’s protocol. TUNEL-positive cells were counted in 5 fields/section and expressed as density (the percentage of TUNEL-positive cells).

According to the published article (Hong et al., 2024), immunohistochemical staining was performed on the paraffin-embedded brain tissue sections. Sections were incubated with primary antibodies against E-cadherin (1:200; Abcam, Cambridge, UK), c-Met (1:300; Abcam, Cambridge, UK), ZO-1 (1:100; Santa Cruz Biotechnology, Dallas, United States), Claudin-1 (1:200; Santa Cruz Biotechnology, Dallas, United States), Occludin (1:150; Santa Cruz Biotechnology, Dallas, United States), LC3B (1:400; Bioss, Beijing, China), PINK1 (1:200; Bioss, Beijing, China), Parkin (1:200; Bioss, Beijing, China), Bax (1:200; Abcam, Cambridge, UK) and Bcl-2 (1:200; Abcam, Cambridge, UK) at 4 °C overnight. After HRP-conjugated secondary antibody incubation, signals were visualized with 2, diaminobenzidine (DAB, Vector Laboratories, Inc.). The results were observed under an optical microscope.

Total proteins from brain tissues were extracted using the total protein extraction kit (Solarbio, Beijing, China) and the protein concentration was determined using the BCA protein assay kit (Boster, Wuhan, China). Equal amounts of protein (50 μg) were separated on a 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a polyvinylidene fluoride (PVDF) membrane. The membrane was blocked at room temperature for 1 h with 5% skimmed milk powder, then incubated with the primary antibodies at 4 °C overnight, followed by a standard washing procedure and incubation with the secondary antibody at room temperature for 2 h. The dilution ratios of the primary antibodies (Table 1) for different target proteins were as follows: Claudin-1 (1:1000), Occludin (1:1000), Bax (1:1000), Bcl-2 (1:1500), Pink1 (1:1000), Parkin (1:1500), LC3B (1:1500). All horseradish peroxidase (HRP)-conjugated secondary antibodies were purchased from Bioss Biotechnology. The blots were visualized using the ECL ultra-sensitive chemiluminescence detection kit (NCM, Suzhou, China), and the protein bands were displayed using the AI600 multi-functional imaging system (GE Healthcare, United States). The gray value analysis was performed using ImageJ software.

Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, United States) and reverse-transcribed into cDNA using the miScript reverse transcription kit (Vazyme, Nanjing, China). qRT-PCR was performed using ChamQ SYBR qPCR premix (Vazyme, Nanjing, China) on a StepOnePlus system (Applied Biosystems) (Table 2). Primer sequences for target genes (E-cadherin, c-Met, Bcl-2, Bax, LC3B, Pink1, Parkin) and reference gene (β-actin) are listed in Supplementary Table S1. Relative expression was calculated using the 2⁻ΔΔCt method.

Data were analyzed using GraphPad Prism 10.1. Differences between groups were assessed by unpaired t-test. Results are presented as mean ± SD. p < 0.05 (*) was considered statistically significant, p < 0.01 (**) as highly significant, and p > 0.05 (ns) indicates no significant difference.

Clinical signs in infected goats included ataxia, head tilt, and lethargy within 72–96 h post-infection. Successful establishment of systemic infection was confirmed by characteristic pathological lesions in visceral organs, as detailed in our previous report using this model (Wu et al., 2019). Specifically, infected animals exhibited multifocal hepatic necrosis and splenic inflammatory cell infiltration, verifying the hematogenous dissemination of the pathogen. To objectively assess the distribution of Listeria monocytogenes (LM), we performed a quantitative analysis of bacterial signals across multiple brain regions (Figure 1). Immunofluorescence showed LM was localized in the cytoplasm of neurons (NeuN⁺) and glial cells (GFAP⁺), with focal aggregation. Quantitative analysis demonstrated significantly higher bacterial densities in the midbrain, pons, and medulla oblongata (p < 0.01) compared to the brain and cerebellum, indicating a marked infection tropism for the brainstem. Quantitative culture revealed an average LM load of (6.52 ± 0.83) × 10⁷ CFU/g in infected brain tissue, with colony density showing a significant positive correlation with LM fluorescence intensity (r = 0.89, p < 0.01). These data indicate that in this caprine infection model, LM is associated with blood–brain barrier disruption and subsequent CNS colonization, exhibiting a characteristic predilection for the brainstem.

To further evaluate the impact of LM-induced neuroinvasion on blood–brain barrier (BBB) integrity, we examined the expression of tight junction (TJ) proteins (ZO-1, Claudin-1, and Occludin) using immunohistochemistry and Western blotting. Western blot analysis revealed regional heterogeneity in TJ protein levels (Figure 2A): Claudin-1 and Occludin expression in the cerebrum, midbrain, and pons of the infected group was significantly lower than that of the control group (p < 0.05), while a significant increase was observed in the medulla oblongata (p < 0.05). Quantitative immunohistochemistry (IHC) analysis further substantiated these regional differences (Figure 2B). Claudin-1 expression was significantly reduced across all examined brain regions (p < 0.01). For ZO-1 and Occludin, significantly reduced signal intensity and structural continuity were observed in most regions (p < 0.05), with the exception of ZO-1 in the medulla and Occludin in the pons, where no statistically significant changes were detected (p < 0.05). In the control group, TJ proteins were widely distributed in neurons and glial cells, with ZO-1 primarily localized to cortical astrocytes and cerebellar Purkinje cells. Following infection, the structural continuity of these proteins was markedly disrupted in the cerebral cortex, cerebellum, midbrain, and pons.

These findings suggest that LM infection is associated with region-specific dysregulation of BBB tight junction proteins, potentially compromising the integrity of the blood–brain barrier.

To objectively evaluate the involvement of the E-cadherin/c-Met pathway during LM infection, we performed a comprehensive quantitative analysis of their protein and mRNA levels (Figure 3). Immunohistochemical analysis revealed that both E-cadherin and c-Met were widely distributed in the cell membranes and cytoplasm of neurons and glial cells in both infected and control goat brain tissues (Figure 3A). Notably, E-cadherin exhibited characteristic high expression in cortical astrocytes, cerebellar Purkinje cells, and brainstem trigeminal ganglia, while c-Met was enriched in brainstem vascular endothelial cells. Quantitative IHC scoring demonstrated that E-cadherin expression was significantly upregulated in the brain, cerebellum, midbrain, and pons (p < 0.01), with the exception of the medulla oblongata, where no significant change was observed. Similarly, c-Met expression was significantly increased in the brain, cerebellum, pons, and medulla oblongata (p < 0.01), while no statistically significant change was detected in the midbrain.

qRT-PCR analysis further substantiated these findings at the transcriptional level (Figure 3B): c-Met mRNA in the brain, cerebellum, midbrain, and brainstem of the infected group was significantly upregulated (p < 0.05), with an increase also observed in the medulla oblongata; E-cadherin mRNA was significantly upregulated in the brain and midbrain (p < 0.05), while there was a non-significant increase in other regions (p > 0.05).

These findings indicate that LM infection is associated with the synergistic enhancement of the E-cadherin/c-Met pathway at both the transcriptional and protein levels. This synergy may be linked to the pathogen’s ability to interact with host endothelial and neuronal cells during central nervous system (CNS) colonization.

To objectively assess the extent of programmed cell death, a quantitative analysis of TUNEL-positive cells was performed across multiple brain regions (Figure 4A). Statistical analysis revealed that the density of apoptotic cells (measured as the percentage of TUNEL-positive cells) was significantly higher in the infected group compared to the control group across the brain, cerebellum, midbrain, pons, and medulla oblongata (p < 0.01). Representative high-magnification images clearly illustrated typical apoptotic features, including nuclear shrinkage, chromatin marginalization, and the formation of apoptotic bodies (indicated by arrows in Figure 4A).

Western blot quantification (Figure 4B) confirmed: the Bcl-2 protein was significantly upregulated in the brain and midbrain of the infected group (p < 0.05), and significantly downregulated in the cerebellum (p < 0.01); the Bax protein was specifically upregulated in the midbrain (p < 0.01), and significantly downregulated in other brain regions (p < 0.05). The Bcl-2/Bax ratio was significantly increased in the brain, midbrain and medulla oblongata (p < 0.01), and significantly decreased in the cerebellum and midbrain (p < 0.05).

Immunohistochemical analysis indicated (Figure 4C) that Bcl-2 and Bax were widely co-localized in the cytoplasm of neurons, and showed strong positive signals in cortical pyramidal cells and cerebellar Purkinje cells. After infection, the expression intensity of Bcl-2 in motor neurons of the medulla oblongata was significantly upregulated (p < 0.01), while Bax was significantly downregulated in the cerebellar granular layer (p < 0.01).

qRT-PCR verification (Figure 4D) showed: Bcl-2 mRNA was significantly upregulated in the midbrain and pons (p < 0.01), and significantly downregulated in the cerebellum (p < 0.01); Bax mRNA was specifically upregulated in the midbrain (p < 0.01), and significantly decreased in other brain regions (p < 0.05); the transcription ratio of Bcl-2/Bax was highly consistent with the changes in protein levels.

These results indicate that LM infection can regulate the Bcl-2/Bax balance in a region-specific manner, and induce the activation of the apoptotic pathway in brain tissues.

To assess the potential involvement of autophagy during LM infection, we analyzed the expression of total LC3B, Pink1, and Parkin (Figure 5). Western blot quantification (Figure 5A) revealed significantly elevated LC3B protein levels in the brain, midbrain, and medulla oblongata of the infected group (p < 0.05), indicating enhanced autophagosome formation. Pink1 protein was significantly up-regulated in the cerebellum and medulla (p < 0.01) but down-regulated in the midbrain (p < 0.05). Parkin protein was significantly up-regulated in the cerebellum, pons, and medulla (p < 0.01).

Immunofluorescence revealed the co-localization of LC3B, Pink1, and Parkin within the neuronal cytoplasm (Figure 5B). Specifically, LC3B was predominantly localized in cerebellar Purkinje cells, granule cells, and Golgi cells, while Pink1 and Parkin were similarly enriched in Purkinje cells. Quantitative immunohistochemical analysis demonstrated that, with the exception of non-significant changes in LC3B in the midbrain, Pink1 in the cerebellum, and Parkin in the midbrain (p >0.05), the expression of LC3B, Pink1, and Parkin was significantly up-regulated in most brain regions (p < 0.05). Notably, in the infected group, the positive signal density of these proteins significantly increased in glial cells of the pons and medulla (p < 0.01), indicating a robust autophagic stress response in both neuronal and non-neuronal cells.

qRT-PCR verification (Figure 5C) indicated that LC3B mRNA was significantly upregulated in the brain, midbrain, and medulla oblongata (p < 0.05), and was highly positively correlated with the protein level (r = 0.927); Pink1 mRNA was extremely significantly upregulated in the medulla oblongata and pons (p < 0.01), and showed a downward trend in the midbrain; Parkin mRNA was extremely significantly upregulated in all brain regions (p < 0.01), suggesting a global activation at the transcriptional level.