Transcriptional analysis reveals the suppression of RAD51 and disruption of the homologous recombination pathway during PEDV infection in IPEC-J2 cells

Abstract

PEDV is a highly contagious enteric pathogen that can cause severe diarrhea and death in neonatal pigs. Despite extensive research, the molecular mechanisms of host’s response to PEDV infection remain unclear. In this study, differentially expressed genes (DEGs), time-specific coexpression modules, and key regulatory genes associated with PEDV infection were identified. The analysis revealed 2,275, 1,492, and 3,409 DEGs in infected vs. mock-treated pigs at 12 h, 24 h, and 48 h, respectively. Time series analysis revealed that the upregulated genes were involved mainly in antiviral pathways such as the viral defense response and the regulation of immune system processes. Protein–protein interaction network analysis identified the top 20 core genes in the interaction network, which included six upregulated genes (TFRC, SUOX, RMI1, CD74, IFIH1, and CD86) and 14 downregulated genes (FOS, CDC6, CDCA3, PIK3R2, TUFM, VARS, ASF1B, POLD1, MCM8, POLA1, CDC45, BCS1L, RAD51, and RPA2). In addition, GSEA enrichment analysis revealed that pathways such as DNA replication and homologous recombination involving RAD51, CDC6, and RPA2 were significantly inhibited during viral infection. Our findings not only reveal dynamic changes in the transcriptome profile of PEDV-infected IPEC-J2 cells but also provide novel insights into the mechanism of PEDV infection of the host.

Introduction

Porcine epidemic diarrhea virus (PEDV) is a highly contagious enteric pathogen that induces severe diarrhea, dehydration, and high mortality in neonatal piglets [1]. PEDV, a member of the family Coronaviridae [2, 3], is a single-stranded positive-sense RNA virus with a genome size of approximately 28 kb, comprising seven open reading frames encoding the ORF1a, ORF1b, S, ORF3, E, M, and N proteins [4]. Since the identification of the prototype strain in 1978 [3], PEDV infections have been reported annually globally, resulting in significant economic losses for the swine industry [5,6,7,8,9]. Despite extensive research, the complete pathological mechanisms of PEDV and the molecular mechanisms of host responses remain elusive.

Genome integrity is critical for cell survival and function, and DNA double-strand breaks (DSBs) pose one of the most serious threats to genome structure [10]. Both endogenous factors, such as replication errors, and exogenous factors, such as ionizing radiation and chemicals, can cause DSBs. To cope with this damage, cells have evolved complex DNA repair processes, among which homologous recombination (HR) is one of the major mechanisms. HR relies on a highly conserved protein network that acts in concert to repair damaged DNA [11]. RAD51 is a core protein in HR; it is a homolog of bacterial RecA and plays a crucial role in eukaryotes [12]. By forming filamentous nucleoprotein structures, RAD51, an ATP-dependent DNA-binding protein, mediates homologous pairing and strand exchange reactions, enabling the repair of broken DNA fragments via the use of undamaged homologs as templates [13]. RAD51 function is critical for maintaining genomic stability, and defects in its function can lead to chromosomal instability and increased cancer incidence [14]. In addition to its role in DNA repair, RAD51 is also associated with virus infection. Increasing evidence suggests that RAD51 is involved in the viral replication cycle and may influence viral pathogenicity. For example, studies have shown that RAD51 can interact with the HIV-1 integrase, inhibiting its activity and limiting HIV-1 replication [15, 16]. RAD51 also plays a part in HBV infection by protecting the genome and helping to fix homologous DNA to support HBV replication [17].

In this study, we used comprehensive bioinformatics analyses, including differential expression analysis, time sequence clustering, weighted gene coexpression network analysis (WGCNA), and protein–protein interaction (PPI) network analysis, to investigate the transcription cluster time profiles of IPEC-J2 cells infected with PEDV at multiple time points (12 h, 24 h, and 48 h after infection) and identify the host genes and regulatory pathways that play critical roles in viral replication and transmission. These findings reveal significant inhibition of the homogenic reorganization pathway during the post-PEDV infection period, with RAD51 playing a crucial role in this pathway. These findings suggest that PDEV may disrupt the source reorganization pathway by suppressing RAD51 expression, thereby promoting virus reintegration and transmission. This study provides new insights into the molecular mechanisms of PEDV infection and potential targets for developing treatment strategies for PEDV infection. In the future, we can more closely study the specific mechanisms of the role of RAD51 in PEDV infection and investigate whether pharmacological interventions targeting RAD51 can effectively inhibit the replication and spread of PEDV.

Materials and methods

Cell culture and virus infection

IPEC-J2 cells were maintained in DMEM (Gibco, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, USA) at 37 °C and 5% CO₂. The PEDV CV777 strain was generously provided by China Agricultural University. The IPEC-J2 cells were infected with PEDV at a multiplicity of infection (MOI) of 1 and then cultured in DMEM containing 2 µg/mL trypsin at 37 °C and 5% CO₂ for 1 h. After incubation, the cells were washed with phosphate-buffered saline (PBS) and then cultured in DMEM containing 2% FBS.

PEDV-M gene copy number detection

The fluorescent quantitative primers for the M gene were designed on the basis of the PEDV genome information, and the abundance (cycle number) of the PEDV M gene was detected through fluorescence quantitative analysis. The standard curve equation established in the early stage for the PEDV-CV777 type was used: y = -3.3354lg(x) + 37.832, R2 = 0.9994, to calculate the copy number of PEDV.

RNA extraction

Samples collected at 12, 24, and 48 h post-PEDV infection were subjected to RNA sequencing. IPEC-J2 cells were divided into four groups: cells infected with PEDV for 12 h, cells infected with PEDV for 24 h, cells infected with PEDV for 48 h, and control cells subjected to mock infection (Mock). Four biological replicates were prepared for each group. Total RNA was extracted from PEDV-infected and uninfected cells using TRIzol^® reagent (Invitrogen, USA) according to the manufacturer’s instructions. The RNA quality and concentration were tested using NanoDrop 2000 (Thermo Scientific, MA, USA) and Agilent 2100 Bioanalyzer (Agilent Technologies, CA, USA) instruments. Samples with an RNA integrity number (RIN) ≥ 7 were retained for further analysis.

Library construction and data analysis

For each sample, 1 µg of RNA was used as the input for RNA sequencing library preparation. Following the manufacturer’s instructions, the Epicenter Ribo-Zero rRNA Removal Reagent was used to deplete ribosomal RNA, and the NEBNext^® Ultra™ II Directional RNA Library Prep Kit for Illumina^® was used to generate mRNA sequencing libraries. The purity of the library products was then evaluated with the Agilent 2100 Bioanalyzer, and the libraries were sequenced as 150 bp paired-end reads using the NovaSeq 6000 platform. Raw reads were preprocessed with in-house Perl scripts to trim adapters, poly-N sequences, and low-quality bases, and were then aligned to the pig reference genome (Sscrofa11.1) using STAR.

Analysis of differentially expressed mRNAs

Differential gene expression analysis was performed between the groups (12 h vs. Mock, 24 h vs. Mock, 48 h vs. Mock, 24 h vs. 12 h, 48 h vs. 12 h, 48 h vs. 24 h) via the R package DESeq2 (version: 1.34.0) [18], and genes with p-adj < 0.05 (using the Benjamini-Hochberg method to adjust p-values) and |log2FC| >1 were selected as differentially expressed genes (DEGs).

Enrichment analyses

Functional enrichment analysis of the DEGs was performed with the Gene Ontology (GO) [19, 20], Kyoto Encyclopedia of Genes and Genomes (KEGG) [21, 22], and Reactome [23] databases. The hypergeometric distribution test was used with the enrichGO and enrichKEGG functions from the clusterProfiler R package (version 4.2.2) [24] to identify enriched pathways from the GO and KEGG databases. Additionally, the enrichPathway function from the ReactomePA package was used to perform enrichment analysis on the Reactome database. Pathways with p values less than 0.05 were retained for further analysis.

Time series analysis

Soft clustering analysis was performed via the fuzzy c-means algorithm provided by the Mfuzz R package (version 2.54.0) [25] to identify different expression patterns of genes in the time series experimental design. Two parameters, c (number of clusters) and m (fuzziness parameter), were required for this analysis. The value of parameter c was determined by evaluating the change in the sum of squared errors as the number of clusters increased, and the value of parameter m was obtained via the mestimate function from the Mfuzz package. After the two key parameters were determined, clustering was performed, and genes with a membership degree > 0.6 were retained to ensure similar expression trends within each group.

Construction of WGCNA

Weighted gene co-expression network analysis (WGCNA) [26] was used to construct gene coexpression modules from the gene expression profiles. First, a gene relationship matrix was obtained from the gene expression profiles via Pearson correlation coefficients. By setting the soft threshold β to 14, the gene relationship matrix derived from the Pearson correlation coefficients was transformed into an adjacency matrix. The topological overlap matrix (TOM) was subsequently calculated to measure the interconnectivity of the network. We used the dissimilarity of the TOM as the clustering distance to divide the genes into different modules. Additionally, a dynamic tree-cutting algorithm with a threshold of 0.25 was applied to merge similar gene modules.

Protein–protein interaction network analysis

Protein–protein interaction information for the corresponding genes was retrieved from the STRING database (version 12.0) [27]. A minimum interaction confidence score threshold of 0.4 was set, i.e., only interactions with a confidence score greater than or equal to 0.4 were retained. The protein‒protein interaction network was constructed via Cytoscape software. Topological analysis of the network was performed, calculating node-specific metrics such as degree centrality and betweenness centrality. Furthermore, core genes were identified using the cytoHubba plugin within Cytoscape. We evaluated and selected core genes by employing seven commonly used algorithms: Degree, MCC, MNC, Closeness, Radiality, Stress, and EPC.

Quantitative real-time‑PCR (qPCR) verification

On the basis of the gene sequences published in the GenBank database, qPCR primers were designed via Primer Premier 5.0 software, and GAPDH was used as the reference gene. All primers were synthesized by Sangon Biotechnology (Shanghai, China), and the corresponding sequences are shown in Table S8. qPCR analysis was performed via a real-time fluorescence detection kit. All qPCRs were carried out in a 20 µL volume, with 10 µL of 2 × SYBR Premix ExTapTM II, 0.4 µL of 10 µmol/L PCR forward primer, 0.4 µL of 10 µmol/L PCR reverse primer, 0.4 µL of 50 × ROX Reference Dye II, 2.0 µL of cDNA, and RNase-free ddH₂O to a total volume of 20 µL. Three independent experimental replicates were set up for each sample. The qPCR amplification program was as follows: 95 °C for 5 min; 95 °C for 10 s; and 60 °C for 30 s, for a total of 40 cycles. To analyze the specificity of the amplification products, samples were collected at multiple points during the PCR amplification, and melting curve analysis was performed.

Statistical analysis and data visualization

All the statistical analyses were performed in the R environment, and plots were created with the R package ggplot2.

Results

Differential transcriptomic landscapes of PEDV infection at multiple time points

The normal IPEC-J2 cells exhibited irregular shapes with clear outlines and distinct boundaries and were evenly distributed on the cell plate. After infection with the classical PEDV strain CV777, the cells showed obvious shrinkage, became rounded, and lost their normal cellular morphology, indicating cytopathic effects typical of PEDV infection (Fig. 1A). The qPCR analysis revealed that the copy number of the PEDV-M gene was increased at 12 h, reached the highest expression level at 24 h, and was slightly decreased at 48 h postinfection (Fig. 1B).

Fig. 1

PEDV infection induces damage to IPEC-J2 cells at different time points. (A) Observation under an optical microscope. (B) PEDV copy numbers in IPEC-J2 cells at different time points postinfection. Data are presented as the mean ± SD, n = 3. Different letters indicate significant differences, P < 0.01

Transcriptome sequencing was subsequently performed on samples from different time points. A relatively uniform distribution of TPM values for all sample genes was observed (Fig. 2A). The PCA results also indicated that the genes within each group presented similar expression patterns, while the samples in the different groups were well distinguished (Fig. 2B). These results suggested that the sequencing data quality was satisfactory for further bioinformatics analysis. Differential analysis revealed 2,275 (611 upregulated and 1,664 downregulated), 1,492 (609 upregulated and 883 downregulated), 3,409 (2,093 upregulated and 1,316 downregulated), 2,231 (1,509 upregulated and 722 downregulated), 5,417 (3,398 upregulated and 2,019 downregulated), and 2,703 (1,951 upregulated and 752 downregulated) DEGs in the 12 h vs. Mock, 24 h vs. Mock, 48 h vs. Mock, 24 h vs. 12 h, 48 h vs. 12 h, and 48 h vs. 24 h comparison groups, respectively (Fig. 2C, D).

Fig. 2

Analysis of differentially expressed genes. (A) Box plot of log2(TPM) values for mRNAs across different time points. (B) PCA diagram of normalized mRNA expression values illuminating the general relationships among datasets. (C) Differential gene expression results showing up- and downregulated genes in the six comparison groups (12 h vs. Mock, 24 h vs. Mock, 48 h vs. Mock, 24 h vs. 12 h, 48 h vs. 12 h and 48 h vs. 24 h). The threshold used to define the DEGs was a |log2(FC)| >1 and an adjusted p value < 0.05. The blue dots indicate downregulated genes, and the red dots indicate upregulated genes. (D) Histograms of the number of differentially expressed genes

Enrichment analysis revealed that pathways such as the JAK-STAT signaling pathway, MAPK signaling pathway, cytokine‒cytokine receptor interaction, and PI3K‒Akt signaling pathway were enriched in the Mock group. Immune-related pathways, including regulation of the T-cell apoptotic process, regulation of lymphocyte differentiation, and regulation of the adaptive immune response, were enriched only in the 48 h vs. Mock comparison. Pathways such as carbohydrate metabolism, the HIF-1 signaling pathway, and positive regulation of lipid transport were enriched exclusively in the 12 h vs. Mock comparison (Fig. 3).

Fig. 3

GO, KEGG, and REACTOME enrichment analyses. The y-axis represents pathway entries, and the x-axis represents the grouping of differentially expressed genes. The shape of the plot represents different databases, with circles representing the GO database, triangles representing the KEGG database, and squares representing the REACTOME database. The color represents the magnitude of the p value, with redder color indicating a smaller p value. Group Specific: pathways enriched only in the differential genes of a single group. Group Two: pathways enriched in the differential genes of exactly two groups. Group Wide: pathways enriched in the differential genes of exactly three groups

Time-course transcriptomics uncovers transcriptional reprogramming during PEDV infection

To investigate the gene expression trends across different groups, we employed a soft-threshold clustering method based on the within-group sum of squares “gap” statistic (Fig. 4A) and classified the gene expression patterns into 5 clusters (Fig. 4C and Table S1). A gene expression heatmap was constructed to display the expression patterns of genes from Cluster 1 to Cluster 5 (Fig. 4B). The expression of genes in Cluster 1 tended to increase after PEDV infection, and enrichment analysis revealed that these genes were involved mainly in antiviral pathways, such as the defense response to viruses, regulation of immune system processes, and TNF signaling. Conversely, genes in Clusters 3 and 4 tended to be downregulated following PEDV infection and were enriched primarily in metabolism-related pathways, including the cellular amide metabolic process, liposaccharide metabolic process, and carbohydrate derivative metabolic process. Cluster 2 and Cluster 5 presented irregular expression patterns over time and were enriched mainly in pathways such as RNA splicing, rRNA processing, the cell cycle, and tight junctions (Figs. 4D and 5A and B and Tables S2, 3).

Fig. 4

Global changes in gene expression across multiple time points. (A) Analysis plot for determining the optimal number of gene clusters on the basis of the total within-cluster sum of squares and the “gap” statistic. (B) Heatmap of genes highlighting the different expression patterns. The colors on the y-axis indicate the different gene expression pattern clusters, whereas the colors on the x-axis indicate the different sample groups. The intensity of the heatmap colors indicates the relative expression levels, with red representing increased expression and blue representing decreased expression. (C) Line plots showing the average gene expression trends for different expression patterns. (D) Line plots of the differential gene expression patterns. The line colors represent the membership of genes in the clusters, with darker red indicating stronger membership and darker blue indicating weaker membership in the respective cluster

Fig. 5

Enrichment analysis of different gene expression pattern clusters. (A) Enrichment analysis based on the GO database. (B) Enrichment analysis based on the KEGG database. The x-axis represents the different gene expression pattern clusters, and the y-axis indicates the pathway entries. The color of the points also indicates the gene expression pattern clusters. The size of the points represents the number of genes in the indicated pathway, with larger points indicating a greater number of genes

WGCNA identifies time-specific coexpression modules in PEDV infection

This study utilized WGCNA to investigate gene coexpression networks associated with the PEDV infection process. The sample clustering tree revealed no outlier samples, indicating that all the samples could be further analyzed via WGCNA (Fig. 6A). With the soft threshold β set to 14 (Fig. 6B), WGCNA successfully divided the genes into eight coexpression modules represented by different colors (Fig. 6C), with each module exhibiting distinct expression patterns (Fig. 6D). Among these modules, the brown gene module (r=-0.64, p = 8e-03), cyan gene module (r = 0.76, p = 7e-04), dark red gene module (r=-0.85, p = 3e-05), and dark gray gene module (r = 0.72, p = 2e-03) were significantly correlated with the PEDV infection process (Fig. 6E) (Tables S4, S5, S6 and S7).

To further explore the functions of the gene modules significantly associated with the PEDV infection process, we performed enrichment analysis on the key genes (MM ≥ 0.8, GS ≥ 0.8) within the brown, cyan, dark red, and dark gray gene modules (Fig. 6F). GO enrichment analysis revealed that these genes were involved primarily in biological pathways such as T-cell proliferation, positive regulation of cell–cell adhesion, and regulation of cell proliferation (Fig. 7A). KEGG enrichment analysis revealed enrichment of pathways including the TNF signaling pathway, viral protein interaction with cytokines and cytokine receptors, the Toll-like receptor signaling pathway, and the JAK-STAT signaling pathway (Fig. 7B). Reactome enrichment analysis revealed enrichment in biological pathways such as extracellular matrix organization, homology-directed repair, and KK complex recruitment mediated by RIP1 (Fig. 7C).

Fig. 6

WGCNA identified gene coexpression modules at different time points after PEDV infection. (A) Sample clustering tree at different time points postinfection. (B) Selection of the soft-thresholding power (β). The left panel displays the scale-free fit index in relation to the soft-thresholding power. The right panel shows the mean connectivity versus the soft-thresholding power. A power of 14 was chosen because the fit index curve flattens out at higher values (> 0.8). (C) Hierarchical cluster dendrogram of samples at different time points postinfection, showing the coexpression modules generated via WGCNA. Modules belonging to branches are color-coded according to the interconnectedness of genes. Eight modules represented by colors in the horizontal bar were found via a 0.25 threshold for merging. (D) Heatmap showing the gene expression of the 8 modules across the four time points. (E) Relationships between modules and time points after PEDV infection. The correlation heatmap displays the correlations between modules, with the color intensity representing the strength of the correlation. A deeper red color indicates a stronger positive correlation between modules, whereas a deeper blue color indicates a stronger negative correlation. The lines connecting the time points and the 8 modules represent the associations between the traits and the coexpression modules. The line thickness represents the strength of the association, with thicker lines indicating stronger associations. The line colors represent the p values of the associations, with deep red for p < 0.01, orange for 0.01 < p < 0.05, and gray for p > 0.05. (F) Scatter plot of gene significance vs. module membership for modules associated with different postinfection time points. The y-axis represents gene significance, which indicates the degree of association between a gene and the infection time point trait. Genes with high significance may have particularly important biological functions or regulatory roles under specific phenotypic conditions. The x-axis represents module membership, with higher values indicating that a gene expression pattern is more closely correlated with the overall module. The color of the points corresponds to the four modules associated with each time point (deep red, cyan, brown, and dark gray)

Fig. 7

Enrichment analysis of genes within coexpression modules associated with PEDV infection time. (A) Bubble plot of the GO enrichment analysis results. (B) Bubble plot of the KEGG enrichment analysis results. (C) Bubble plot of the Reactome enrichment analysis results. The x-axis represents the enrichment factor, which is the ratio of the number of DEGs in a pathway to the total number of genes in that pathway. The size of the dots represents the number of genes, with larger dots indicating a greater number of genes. The color of the dots represents the magnitude of the p value

PPI network analysis reveals regulatory hubs in PEDV infection

To identify key core regulatory factors during the PEDV infection process, we conducted protein interaction network analysis. Among the 985 key genes associated with PEDV infection, the top 20 core genes in the interaction network included 6 upregulated genes (TFRC, SUOX, RMI1, CD74, IFIH1, and CD86) and 14 downregulated genes (FOS, CDC6, CDCA3, PIK3R2, TUFM, VARS, ASF1B, POLD1, MCM8, POLA1, CDC45, BCS1L, RAD51, and RPA2) (Fig. 8A, B). GSEA of the 48 h vs. Mock group revealed that among the 20 core genes, the only enriched genes were RAD51, CDC6, and RPA2, which participate primarily in pathways such as DNA replication and homologous recombination (Fig. 8C, D).

Fig. 8

Protein interaction network revealing core genes and GSEA enrichment analysis. (A) Protein–protein interaction network. The size of the nodes represents the importance of the genes in the protein interaction network, with larger nodes indicating proteins with more interaction partners. The top 20 core genes are labeled. (B) Heatmap of the top twenty core genes. The colors of the horizontal axis represent different sample groups. The depth of color in the heatmap represents the level of expression, with red indicating high gene expression and blue indicating low gene expression. (C-D) GSEA enrichment analysis plot. The x-axis represents the ranking of gene sets, with the ranking decreasing from left to right, whereas the y-axis represents the enrichment score. The enrichment score curve shows the degree of enrichment of a gene set during the PEDV infection process, with a larger absolute value of the curve peak indicating a greater degree of enrichment of the gene set during the PEDV infection process. (E) Bar plot of the fold change in core gene RNA-seq and qPCR expression. The y-axis shows the log2FoldChange, and the x-axis indicates the analyzed gene. Data are presented as the mean ± SD, n = 3

Discussion

PEDV causes substantial economic losses within the livestock industry, and existing commercial vaccines have not been able to offer complete protection for pigs. Despite advancements in PEDV research over recent decades, a comprehensive understanding of the pathogenic mechanisms of this virus remains elusive. In this investigation, we utilized bioinformatics analyses to examine the transcriptional alterations in PEDV-infected IPEC-J2 cells across various time intervals and identify DEGs, time-specific coexpression modules, and crucial regulatory elements linked to PEDV infection. Through WGCNA, we identified gene coexpression modules significantly associated with the progression of PEDV infection. The genes in these modules play pivotal roles in the immune response and metabolic regulation of host cells during PEDV infection. The findings of this study indicate that host immune responses are triggered following PEDV infection, resulting in the proliferation of immune cells and the modulation of immune-related pathways. The innate immune system acts as the primary defense against viral infections, with pattern recognition receptors (PRRs) identifying virus-associated molecular patterns, initiating downstream signaling and generating antiviral substances such as interferon [28, 29]. However, viruses have developed various strategies to evade innate immunity, including inhibiting PRR recognition, disrupting interferon signaling, and suppressing the expression of antiviral genes [30, 31]. Additionally, PEDV may manipulate the metabolic processes of host cells to create an environment that is more conducive to its replication and survival. The interaction between PEDV and host cells also influences cellular functions such as RNA splicing [32], cell cycle regulation [33], and cell‒cell interactions [34].

One significant discovery from this work is the notable inhibition of the HR pathway after PEDV infection. HR is a DNA repair mechanism essential for maintaining genome stability and integrity [35]. RAD51, a core HR protein responsible for homologous pairing and strand exchange reactions to ensure accurate DSB repair [36], is downregulated in PEDV-infected IPEC-J2 cells. This observation indicates that PEDV may disrupt the HR pathway, potentially facilitating viral replication and spread by suppressing RAD51 expression. Decreased RAD51 levels have been linked to increased chromosomal instability and cancer risk. In viral infections, RAD51 is associated with the replication cycle and pathogenicity of various viruses [37, 38]. For example, RAD51 interacts with the replication initiator protein of begomovirus to increase viral replication [39]. In astrocytes, elevated RAD51 levels increase HIV-1 L activity in conjunction with HIV-1 Tat, the transcription factor C/EBPβ, and CHOP [40]. Knocking down RAD51 significantly reduces HCV protein expression and intracellular and extracellular HCV RNA levels before HCV infection [37]. In the present study, protein‒protein interaction network analysis revealed key regulatory factors of PEDV infection. The top 20 genes in the network, including RAD51, CDC6, and RPA2, are involved in DNA replication, HR, and other pathways. These findings underscore the potential roles of these genes in the viral replication cycle and suggest that they could be targeted to inhibit PEDV replication and dissemination.

Conclusions

In summary, our study provides insights into the molecular mechanisms of PEDV infection and identifies potential targets for developing therapeutic strategies against PEDV infection. PEDV inhibits the HR pathway by downregulating the expression of RAD51, which may promote viral replication and dissemination. Furthermore, the dynamic changes in gene expression patterns, involvement of immune-related pathways, and regulation of host cellular processes highlight the complex interactions between PEDV and host cells.