Dynamic interplay between RNA N⁶-methyladenosine modification and porcine reproductive and respiratory syndrome virus infection

Abstract

N⁶-methyladenosine (m⁶A) has attracted significant attention for its role in regulating the complex interaction between viruses and host cells. Porcine reproductive and respiratory syndrome virus (PRRSV) is a significant pathogen affecting swine health worldwide. Here, we first identified seven m⁶A-enriched peaks in PRRSV genomic RNA by m⁶A RNA immunoprecipitation sequencing (m⁶A-seq). Moreover, functional analyses revealed a positive correlation between the m⁶A modification level and PRRSV replication. Treatment with the universal methylation inhibitor 3-deazaadenosine (3-DAA) effectively suppressed PRRSV replication in a dose-dependent manner. Furthermore, m⁶A-seq was also used to determine the m⁶A landscape of the transcriptome in PAMs infected with pandemic or highly pathogenic PRRSV strains. Among the 4677 transcripts exhibiting altered m⁶A modification levels, the MAPK14 gene and the p38/MAPK signalling pathway emerged as preliminary targets of m⁶A-mediated epigenetic regulation during PRRSV infection. These findings provide new insights into the epigenetic mechanisms underlying PRRSV infection and may facilitate the development of anti-PRRSV therapeutics.

Introduction, methods and results

The role of RNA modifications as critical regulators of gene expression has been explored in the context of viral infections over the past several years [1,2,3]. Among these modifications, N⁶-methyladenosine (m⁶A) has attracted significant attention for its role in various biological processes, including RNA stability, translation, and the immune response [4, 5]. m⁶A is the most prevalent internal modification in eukaryotic messenger RNA (mRNA), and its dynamic nature allows rapid alterations in mRNAs in response to intra- and extracellular insults [6,7,8]. The importance of m⁶A extends beyond cellular RNA to include viral RNA, as many viruses exploit this modification to increase their replication and evade host immune defenses [9]. m⁶A may regulate the antiviral immune response via several potential mechanisms. For example, m⁶A modifications can influence the expression of key immune-related genes by dynamically regulating mRNA stability, splicing, translation, and degradation. m⁶A modifications also play critical roles in antiviral responses by modulating the stability and translation of pattern recognition receptor (PRR) transcripts and downstream signalling components (e.g., RIG-I and IFN-β). Moreover, viruses can exploit m⁶A modifications to evade immune responses, as viral RNA often acquires or manipulates m⁶A marks to increase replication or suppress host defenses [10].

Porcine reproductive and respiratory syndrome virus (PRRSV) is a significant pathogen affecting swine health worldwide, leading to considerable economic losses in the pig industry [11]. Understanding the mechanisms of PRRSV infection is essential for developing effective control strategies. The emergence of PRRSV was first documented in the United States and Europe in the late 1980s [12]. The situation intensified in 2006 with the epidemic of a highly pathogenic variant of PRRSV (HP-PRRSV) in China, which led to more severe outbreaks characterized by higher morbidity and mortality rates [13,14,15]. PRRSV NADC30-like strains first emerged in China in 2013 [16]. These strains share genetic similarity with the NADC30 strain isolated in the United States, which has a 131-amino acid deletion within the hypervariable region of the nsp2 protein [17]. Lin et al. [18] and Gong et al. [19] suggested that m⁶A modifications are important for PRRSV infection and host antiviral immune regulation, but neither study directly validated the functional roles of specific m⁶A sites on PRRSV RNA or host immune-related transcripts. A comprehensive view of the networks regulated by m⁶A during PRRSV infection is lacking. Our study employed functional analysis to identify the exact roles of m⁶A modification during PRRSV infection and identify specific m⁶A sites on PRRSV genomic RNA. MeRIP-seq also provides a comprehensive overview of the gene networks in host cells regulated by m⁶A during PRRSV infection. These findings may provide new insights into the epigenetic mechanisms underlying viral infection and identify potential antiviral targets.

Mapping the m⁶A peaks in PRRSV genomic RNA

In our study, we utilized the highly pathogenic HuN4 strain and pandemic NADC-30, similar to the HeN-L1 strain (Additional file 1). We first employed an ELISA-based approach to assess the presence of m⁶A modifications in PRRSV genomic RNA. Porcine alveolar macrophages (PAMs) were infected with either the PRRSV HuN4 strain or the PRRSV NADC30-like strain, and after 48 h, the cells were subjected to three freeze‒thaw cycles. The supernatants were collected by centrifugation, followed by ultrafiltration to concentrate the viral particles, from which the viral RNA was extracted and fragmented into 80–200-nt fragments. The presence of m⁶A was then measured as a percentage of all adenosines in the viral RNA sample using ELISA. As shown in Figure 1A, the percentage of viral RNA with m⁶A modification was approximately 0.6–0.8%, which was twice as high as the 0.3–0.4% observed in total cellular RNA (Figure 1A). Figure 1A highlights the baseline differences in global m⁶A levels between porcine alveolar macrophages (PAMs) infected with the PRRSV HuN4 strain and those infected with the NADC30-like strain. Using m⁶A-specific ELISA, we detected significantly greater m⁶A enrichment in both HuN4- and HeN-L1-infected cells than in uninfected cells, indicating the common upregulation of m⁶A modification during infection with diverse PRRSV strains. These findings provide the foundation for investigating the distinct regulatory roles of m⁶A in the host–virus interplay and highlight the importance of further exploration of the pathways modulated by m⁶A. To further map the m⁶A-modified regions within the viral genome, MeRIP-seq was conducted by DIATRE Biotechnology, Shanghai, China. The experimental flow of MeRIP-seq is shown in Additional files 1 and 2. Briefly, PAMs were infected with the HeN-L1 strain, and at 48 hours post-infection (hpi), the cells were subjected to three freeze‒thaw cycles, followed by centrifugation and ultrafiltration to concentrate the viral particles. The extracted viral RNA was fragmented and immunoprecipitated with a m⁶A-specific antibody. Reverse transcription and subsequent deep sequencing of the RNA‒antibody complexes were applied to identify the m⁶A-modified region in PRRSV genomic RNA. The stringent peak calling method (FDR < 0.01) was applied to determine the statistically significant peaks (p values < 1E-10). A correlation test between three biological replicates of viral RNA revealed a high correlation (0.996), which confirmed the replicability of the results. Seven m⁶A-enriched regions were revealed in the viral genome, with one located in the N protein-coding region and the others distributed across nonstructural protein-coding regions. Notably, the Nsp2-encoding region contained the highest m⁶A peak, the length of which was approximately 178 nt (Figures 1B and C).

Figure 1

Mapping the m⁶A peaks in the PRRSV genome and analysing the correlation between m⁶A modification and PRRSV replication. A Elisa-based methods were used to determine the ratio of m⁶A/A in the mRNAs of PRRSV-infected or uninfected PAM cells. B and C Mapping the m⁶A-enriched peaks in PRRSV genomic RNA. The stringent peak calling method (FDR < 0.01) was applied to determine the statistically significant peaks (p values < 1E-10). D Determination of the viral titre via the TCID₅₀ assay. Mrac145 cells were treated with 3-DAA or PBS prior to PRRSV infection. E and (F) Overexpression of m⁶A enzymes in PRRSV-infected Marc145 cells. The viral titre was determined via plaque assay. G IFA detection of the PRRSV N protein (green) in m⁶A enzyme-overexpressing Marc145 cells. DAPI was used to stain the cell nucleus. Statistical relevance was determined with an unpaired Student’s t‐test, **p < 0.001.

Positive correlation between m⁶A modification levels and PRRSV replication

Previous studies have shown that the PRRSV genome does not encode kinases that mediate m⁶A modification. Therefore, we first interfered with the expression of endogenous m⁶A-related enzymes [20, 21]. Marc145 cells were seeded in a six-well plate, and when the cell confluence reached 60%, 3-deazaadenosine (3-DAA, 25 µM) was used to inhibit the methylation of adenosine [22]. Twenty-four hours after 3-DAA treatment, the cells were infected with HeN-L1 (MOI = 0.1). Eight to 48 h after virus infection, the cells were subjected to repeated freeze‒thaw cycles, and the virus in the supernatant was collected for the TCID₅₀ assay to determine the virus titre. As shown in Figure 1D, compared with PBS treatment, 3-DAA treatment significantly decreased the virus titre from 24 to 48 h. To investigate the specific effects of m⁶A enzymes on PRRSV replication, Marc145 cells overexpressing (OE) m⁶A writers or erasers were used. As shown in Figure 1E, when ALKBH5 or FTO was overexpressed, the virus titre was significantly reduced at 12 hpi, whereas in cells overexpressing METTL14 or METTL3, the virus titre increased to approximately 10^7.0 PFU/mL. To further validate the effects of m⁶A enzymes on viral replication kinetics, the viruses were collected at 8, 12, 24, and 48 hpi. As shown in Figure 1F, the replication of PRRSV began to be affected by manipulation of endogenous m⁶A enzymes after 8 hpi, and this effect lasted until 48 hpi. To detect the effects of m⁶A modification on the expression of virus-encoded proteins, Marc145 cells were seeded in 24-well plates, and m⁶A enzymes were overexpressed via lentiviral vectors. Eight hours post-transfection, the cells were infected with HeN-L1 (MOI = 0.1). At 24 hpi, the cells were fixed, and an indirect immunofluorescence assay (IFA) was used to detect the expression of the PRRSV N protein. As shown in Figure 1G, compared with the backbone vector, transient overexpression of METTL3 increased virus replication, whereas ALKBH5 expression decreased virus replication. These results preliminarily indicate that m⁶A modification may positively regulate PRRSV replication.

The m⁶A landscape of the transcriptome altered by PRRSV infection

Given the profound impact of PRRSV infection on gene expression within host cells, we further used MeRIP-seq to investigate changes in m⁶A modification levels across the transcriptome of PAM cells after PRRSV infection. As shown in Figure 2A, approximately 27 329 m⁶A peaks, among the 4677 transcripts, exhibited significant differences between HeN-L1-infected cells and uninfected cells, with m⁶A levels significantly increased in 2258 genes and decreased in 2419 genes. Similarly, in HuN4-infected cells, 25 641 m⁶A peaks presented significant changes, with m⁶A levels upregulated in 2557 genes and downregulated in 2510 genes. Profiling the distribution of m⁶A peaks across the full length of mRNA revealed that, similar to those in uninfected cells, m⁶A modifications in virus-infected PAM cells predominantly appeared in coding regions (Additional file 2), indicating that m⁶A may be indirectly involved in regulating mRNA expression after viral infection. Mapping the reads to the porcine genome using IGV software revealed that genes with differential m⁶A levels were distributed among multiple cellular signalling pathways, including the PI3K-AKT pathway and cell cycle regulation (Additional file 2). In addition to mRNAs, m⁶A modifications have also been detected in noncoding RNAs, where miRNAs, in particular, play a key role in gene expression. Therefore, we conducted motif prediction analysis for differentially modified m⁶A peaks and identified several potential miRNAs. For example, miR-181c has been reported to counteract the PRRSV ORF4 gene to hinder viral replication, while miR-505 regulates viral infection by interacting with the PRRSV ORF7 gene. These findings suggest that m⁶A may serve as a regulatory mechanism for host miRNA expression in the context of PRRSV infection (Additional file 2).

Figure 2

Altered m⁶A landscape in the transcriptome of PRRSV-infected cells. m⁶A-seq was used to detect the m⁶A modification levels in the mRNAs of PRRSV-infected PAMs. A Volcano maps illustrating the genes with different m⁶A modification levels between HuN4-infected PAMs (left lane) and HeN-L1-infected PAMs (right lane). B and C Using the KEGG database, cluster analysis was performed on genes with differential m⁶A modification levels. Bubble charts showing the clustering results of differentially expressed genes between HuN4-infected and uninfected cells (B) or HeN-L1-infected and uninfected cells C. The figure displays only the ten pathways with the smallest p-adjusted values (p-adjv). D Bubble charts were generated to display the genes with more than three m⁶A peaks and the specific signalling pathways to which they belong. The figure displays only the seven pathways with the smallest p-adjusted values (p-adjv). E Profiling of the m⁶A peaks in genes belonging to the MAPK signalling pathway via IGV software. The red lines indicate the m⁶A peaks, and the blue lines indicate the input signals. F and G RT‒qPCR was used to detect the expression level and stability of MAPK14 mRNA in HuN4- and HeN-L1-infected cells. H and I The online tool SRAMP was used to predict the m⁶A modification site in MAPK14 mRNA.

To further explore how genes with differential m⁶A levels affect antiviral responses, we conducted gene cluster analysis. A heatmap (Additional file 3) illustrated differential gene expression patterns across samples, with distinct expression profiles in cells infected with the epidemic strain compared with those infected with the highly pathogenic strain. For example, the m⁶A modification levels of genes such as TRIM16 and MAP3K8 increased in cells infected with HeN-L1 but decreased in cells infected with HuN4, whereas the m⁶A levels of genes such as NOD2 and SOX9 decreased in HeN-L1-infected cells but increased in HuN4-infected cells. The GO and KEGG databases were used to analyse the signalling pathways associated with genes with altered m⁶A modification levels between HeN-L1- or HuN4-infected and uninfected cells. In the GO analysis (Additional file 3), genes with upregulated m⁶A peaks were enriched primarily in pathways such as negative regulation of DNA-templated transcription, negative regulation of the RNA biosynthetic process, and negative regulation of transcription by RNA polymerase II. While the results revealed m⁶A level changes in genes involved in RNA replication and cellular stress processes, further analysis is necessary to clarify the correlation between changes in m⁶A levels and specific signalling pathways. In the KEGG analysis (Figures 2B and C), we observed significant enrichment of MAPK signalling pathways, as well as the PI3K-AKT, Hippo, and TNF signalling pathways, suggesting that m⁶A modification may be crucial in modulating interactions between PRRSV and host cells. Interestingly, we screened genes with more than three m⁶A peaks and analysed the signalling pathways to which they belong. Pattern recognition receptor (PRR) signalling pathways associated with antiviral innate immune responses, such as the RIG-I-like, NOD-like, and Toll-like receptor signalling pathways, were enriched (Figure 2D). Previous studies have suggested that m⁶A modification plays a key role in antiviral innate immune responses by influencing the recognition of viral-derived nucleic acids by cytoplasmic PRRs. Our findings indicate that this mechanism may also play a significant role in PRRSV-infected PAMs.

m⁶A modifications regulate the p38/MAPK signalling pathway during PRRSV infection

Among the diverse cellular signalling cascades, the p38 mitogen-activated protein kinase (MAPK) (p38/MAPK) signalling pathway plays a particularly pivotal role because it functions as a central metabolic hub that responds to various external and internal stress stimuli during PRRSV infection [23, 24]. This pathway is activated by a wide range of extracellular signals, such as proinflammatory cytokines, UV radiation, oxidative stress, and viral pathogens. Upon activation, p38/MAPK undergoes phosphorylation, which subsequently triggers a cascade of downstream signalling events that regulate cellular processes such as gene expression, apoptosis, differentiation, and cell cycle progression. Notably, we observed significant m⁶A enrichment in transcripts associated with the p38/MAPK signalling pathway during infection. These findings were identified through m⁶A-specific meRIP-seq analysis, followed by differential m⁶A peak comparisons between the infected and control groups. MAPK-related signalling pathways were significantly enriched in both the GO and KEGG analyses in the present study, with the MAPK9, MAPK11, and MAPK14 genes exhibiting multiple distinct m⁶A peaks. Mapping the m⁶A reads to the genome revealed that MAPK14 exhibited significantly different levels of m⁶A modification in both the HuN4 and CTR comparisons and the HeN-L1 and CTR comparisons, which indicated that MAPK14 may serve as a potential critical target for m⁶A-mediated regulation of PRRSV infection (Figure 2E).

To further investigate whether altered MAPK14 expression patterns are correlated with changes in m⁶A modification levels, we first examined the dynamic changes in MAPK14 mRNA levels in PRRSV-infected PAM cells. Compared with the uninfected cells, both HuN4 and HeN-L1 infection in the early phase (0–24 h) led to a decrease in MAPK14 mRNA levels (Figure 2F). However, from 24 to 48 h postinfection, MAPK14 mRNA levels partially recovered, which was consistent with the trend of protein expression. These findings indicate that MAPK14 transcriptional levels are affected during the early stages of PRRSV infection. m⁶A modification may enhance cytoplasmic mRNA decay, therefore affecting mRNA stability. To examine whether the observed decrease in MAPK14 mRNA is associated with reduced mRNA stability due to m⁶A modification, we used actinomycin D (actD) to inhibit intracellular mRNA transcription. The results (Figure 2G) revealed fewer MAPK14 mRNAs in PRRSV-infected cells than in uninfected cells, indicating the decreased stability of MAPK14 mRNA. Therefore, we preliminarily hypothesized that m⁶A modification regulates MAPK14 expression by influencing its mRNA stability. To further confirm the m⁶A modification sites within MAPK14 mRNA, we used the online prediction tool SRAMP to identify potential m⁶A modification sites. The results suggest that multiple modification sites exist in the MAPK14 gene, including three sites with high or very high confidence (Figures 2H and I). These findings provide a crucial foundation for further studies on how m⁶A modification regulates MAPK14 expression following PRRSV infection. In summary, our study indicates that the MAPK signalling pathway, particularly the MAPK14 gene, may be a key target through which m⁶A modification regulates the interaction between PRRSV and host cells.

Discussion

The involvement of m⁶A in the viral genome has broader implications for our understanding of virus evolution and pathogenicity. The ability of viruses to exploit host mechanisms is key to their pathogenic success. As m⁶A is a type of epigenetic modification that can be manipulated by both the host and the virus, studying its role in viral infections may reveal new insights into viral evolution. In our study, seven m⁶A-enriched regions within the PRRSV genome were detected. Like those of other RNA viruses [25, 26], the m⁶A peaks are distributed across whole-genome RNA, which may indicate a general role for m⁶A in regulating PRRSV genomic RNA metabolism. Notably, the Nsp2 gene, which is the most variable gene of PRRSV [27, 28], contained the highest m⁶A peak, spanning approximately 178 nucleotides. The Nsp2 protein is a key viral nonstructural protein involved in PRRSV genomic RNA replication, viral polyprotein processing, and immune evasion. The presence of the highest m⁶A peak in the Nsp2 coding region suggests that this region is a potential target of m⁶A modification in the regulation of viral RNA synthesis. Given that the Nsp2 protein is also involved in modulating host immune responses and contributes to PRRSV pathogenesis, m⁶A modification in this region may also impact viral immune evasion strategies by regulating Nsp2 protein expression. Therefore, m⁶A modification may also serve as a strategy for PRRSV evolution.

In our study, by manipulating the expression of m⁶A enzymes, we found a positive relationship between m⁶A levels and PRRSV replication. As expected, further investigation revealed that treatment with 3-DAA significantly suppressed the replication of PRRSV. As shown in Additional file 2, 3-DAA inhibited virus replication from 20 µM to 25 µM but not from 5 µM to 10 µM at 48 hpi. Furthermore, the mRNA levels of N, GP5 and Nsp9 also decreased after 3-DAA treatment (Additional file 2), indicating that the methylation inhibitor may dampen virus replication via interference with viral mRNA transcription. Notably, such methylation inhibitors may serve as novel antiviral reagents in the context of acute PRRSV infection. Understanding how different viruses utilize m⁶A could help in the development of novel antiviral therapies that target these interactions. Emerging evidence highlights that the expression, localization, and functional dynamics of the host m⁶A machinery (writers, erasers, and readers) vary across species, which may shape diverse viral‒host interactions. For example, murine models may incompletely recapitulate human m⁶A-dependent viral replication owing to differences in METTL3 substrate specificity. This could explain why the Zika virus shows reduced m⁶A-dependent neurotropism in nonhuman primates than in humans. Additionally, variations in host cell types and immune responses may contribute to distinct m⁶A landscapes across different species. Viruses may evolve m⁶A as the optimized advantage for their primary hosts. For example, HIV-1 is enriched with m⁶A in regions critical for nuclear export in human T cells, whereas simian immunodeficiency virus (SIV) lacks analogous modifications, suggesting host-specific adaptation. Furthermore, species-specific m⁶A readers (e.g., YTHDF1/2/3) differentially regulate antiviral IFN responses. Compared with human coronavirus (HCoV-OC43), mouse hepatitis virus (MHV) replication is more sensitive to m⁶A depletion in murine cells, likely due to divergent YTHDF paralogue functions. Given that m⁶A modifications are widely utilized by various RNA viruses, including coronaviruses, flaviviruses, and paramyxoviruses, targeting m⁶A pathways could be a promising broad-spectrum antiviral strategy. For example, blocking METTL3 (e.g., with STM2457) reduces the replication of m⁶A-dependent viruses such as HIV-1 and SARS-CoV-2. Conversely, FTO inhibitors (e.g., FB23-2) enhance viral RNA sensing by stabilizing m⁶A-modified immunostimulatory RNAs, as shown in dengue and hepatitis C virus models. Moreover, CRISPR/Cas-based epitranscriptomic editing and m⁶A-regulated cis-elements in viral genomic RNAs (e.g., frameshift regions, IRESs) may further emphasize the potential applicability of m⁶A-based antiviral strategies to PRRSV and other pathogenic RNA viruses. Such therapies could disrupt the ability of viruses to exploit m⁶A modifications, potentially reducing their virulence and enhancing host defenses.

The p38/MAPK signalling pathway plays a pivotal role in the cellular response to various stresses, including viral infections. p38/MAPK is also critical in modulating the host immune response to viral infections. It activates transcription factors such as AP-1 and NF-κB, which are involved in the production of proinflammatory cytokines and interferons. The p38 MAPK pathway is involved in determining cell fate following viral infection. In response to viral infection, p38/MAPK can trigger apoptotic signalling pathways, which may limit viral replication by inducing cell death. However, p38 MAPK may also support viral survival by modulating antiapoptotic proteins, thereby enabling the virus to evade host immune responses and sustain infection. Our analysis suggested that the p38/MAPK signalling pathway may be a preliminary target for m⁶A modification in the regulation of PRRSV-host cell interactions. The term “preliminary targets” was used because our results indicate a strong association but do not yet establish causality. Functional validation experiments, such as m⁶A site mutagenesis or knockdown studies, are needed to confirm the direct regulatory role of m⁶A modifications on MAPK14 and the p38/MAPK pathway. Furthermore, because its activation serves both antiviral and proviral roles depending on the context, further research may be conducted to explore the dual functions of m⁶A regulation of the p38/MAPK signalling pathways in PRRSV infections. Moreover, other host genes, such as TRIM29, may also be involved in m⁶A regulation during PRRSV infection. TRIM29 plays a pivotal role in regulating immune responses against DNA and RNA viruses by modulating type I interferon signalling pathways, PERK-mediated ER stress immune responses, and inflammasome activation [29, 30]. Notably, TRIM29 expression has been shown to be upregulated via YTHDF1-mediated m⁶A modification [31]. Given the importance of TRIM29 in antiviral immunity, it is plausible that m⁶A modifications could similarly regulate TRIM29 expression during PRRSV infection, potentially impacting viral replication and immune evasion. Future studies might explore whether m⁶A methylation of TRIM29-related transcripts and the activity of YTHDF1 influence PRRSV pathogenesis and host defense mechanisms.

In summary, the importance of m⁶A in viral infections cannot be overstated. This modification serves as a crucial player in regulating various aspects of the viral lifecycle, from RNA stability and translation to immune evasion. By studying the role of m⁶A in specific viruses, researchers can gain valuable insights into the mechanisms of viral pathogenesis and host–virus interactions. Furthermore, understanding these processes opens new avenues for therapeutic interventions aimed at viral diseases. As our knowledge of m⁶A continues to expand, it is essential to explore its potential as a target for antiviral strategies, paving the way for innovative approaches to combat viral infections.