N6-Methyladenosine RNA Modification in Host Cells Regulates Peste des Petits Ruminants Virus Replication

Peste des petits ruminants virus (PPRV) causes a severe disease in sheep and goats. PPRV infection is a major problem, causing significant economic losses to small ruminant farmers in regions of endemicity. ABSTRACT N6-methyladenosine (m6A) modification is a major RNA epigenetic regulatory mechanism. The dynamics of m6A levels in viral genomic RNA and their mRNAs have been shown to have either pro- or antiviral functions, and therefore, m6A modifications influence virus-host interactions. Currently, no reports are available on the effect of m6A modifications in the genome of Peste des petits ruminants virus (PPRV). In the present study, we took PPRV as a model for nonsegmented negative-sense single-stranded RNA viruses and elucidate the role of m6A modification on viral replication. We detected m6A-modified sites in the mRNA of the virus and host cells, as well as the PPRV RNA genome. Further, it was found that the level of m6A modification in host cells alters the viral gene expression. Knockdown of the METTL3 and FTO genes (encoding the m6A RNA modification writer and eraser proteins, respectively) results in alterations of the levels of m6A RNA modifications in the host cells. Experiments using these genetically modified clones of host cells infected with PPRV revealed that both higher and lower m6A RNA modification in the host cells negatively affect PPRV replication. We found that m6A-modified viral transcripts had better stability and translation efficiency compared to the unmodified mRNA. Altogether, from these data, we conclude that the m6A modification of RNA regulates PPRV replication. These findings contribute toward a way forward for developing novel antiviral strategies against PPRV by modulating the dynamics of host m6A RNA modification. IMPORTANCE Peste des petits ruminants virus (PPRV) causes a severe disease in sheep and goats. PPRV infection is a major problem, causing significant economic losses to small ruminant farmers in regions of endemicity. N6-methyladenosine (m6A) is an important RNA modification involved in various functions, including virus-host interactions. In the present study, we used stable clones of Vero cells, having knocked down the genes encoding proteins involved in dynamic changes of the levels of m6A modification. We also used small-molecule compounds that interfere with m6A methylation. This resulted in a platform of host cells with various degrees of m6A RNA modification. The host cells with these different microenvironments were useful for studying the effect of m6A RNA modification on the expression of viral genes and viral replication. The results pinpoint the level of m6A modifications that facilitate the maximum replication of PPRV. These findings will be useful in increasing the virus titers in cultured cells needed for the economical development of the vaccine. Furthermore, the findings have guiding significance for the development of novel antiviral strategies for limiting PPRV replication in infected animals.

ABSTRACT N 6 -methyladenosine (m 6 A) modification is a major RNA epigenetic regulatory mechanism. The dynamics of m 6 A levels in viral genomic RNA and their mRNAs have been shown to have either pro-or antiviral functions, and therefore, m 6 A modifications influence virus-host interactions. Currently, no reports are available on the effect of m 6 A modifications in the genome of Peste des petits ruminants virus (PPRV). In the present study, we took PPRV as a model for nonsegmented negative-sense single-stranded RNA viruses and elucidate the role of m 6 A modification on viral replication. We detected m 6 A-modified sites in the mRNA of the virus and host cells, as well as the PPRV RNA genome. Further, it was found that the level of m 6 A modification in host cells alters the viral gene expression. Knockdown of the METTL3 and FTO genes (encoding the m 6 A RNA modification writer and eraser proteins, respectively) results in alterations of the levels of m 6 A RNA modifications in the host cells. Experiments using these genetically modified clones of host cells infected with PPRV revealed that both higher and lower m 6 A RNA modification in the host cells negatively affect PPRV replication. We found that m 6 A-modified viral transcripts had better stability and translation efficiency compared to the unmodified mRNA. Altogether, from these data, we conclude that the m 6 A modification of RNA regulates PPRV replication. These findings contribute toward a way forward for developing novel antiviral strategies against PPRV by modulating the dynamics of host m 6 A RNA modification. IMPORTANCE Peste des petits ruminants virus (PPRV) causes a severe disease in sheep and goats. PPRV infection is a major problem, causing significant economic losses to small ruminant farmers in regions of endemicity. N 6 -methyladenosine (m 6 A) is an important RNA modification involved in various functions, including virus-host interactions. In the present study, we used stable clones of Vero cells, having knocked down the genes encoding proteins involved in dynamic changes of the levels of m 6 A modification. We also used small-molecule compounds that interfere with m 6 A methylation. This resulted in a platform of host cells with various degrees of m 6 A RNA modification. The host cells with these different microenvironments were useful for studying the effect of m 6 A RNA modification on the expression of viral genes and viral replication. The results pinpoint the level of m 6 A modifications that facilitate the maximum replication of PPRV. These findings will be useful in increasing the virus titers in cultured cells needed for the economical development of the vaccine. Furthermore, the findings have guiding significance for the development of novel antiviral strategies for limiting PPRV replication in infected animals. origin of the transcript, whether from the host or from the invading virus. Taken together, the results also indicate that m 6 A-mediated effects are not uniform and appear to depend on the nature of viruses and differences in their life cycle stages. The mechanistic details of the m 6 A-mediated regulation of viral replication need to be understood separately for important pathogenic viruses.
In the present study, we used small ruminant morbillivirus, or peste des petits ruminants virus (PPRV), as a model for nonsegmented negative-sense single-stranded RNA viruses belonging to the family Paramyxoviridae. PPRV causes an acute, highly contagious viral disease in sheep and goats, leading to severe economic losses for small ruminant farmers in regions of endemicity.
Here, we report the role of m 6 A RNA modifications in the PPRV-host interaction. We show that PPRV infection affects the levels of m 6 A modification of the host cells. The infection also alters the expression of m 6 A writer (METTL3 and WTAP) and eraser (FTO and ALKBH5) proteins. The reduction of m 6 A modification in host cells using 3-deazaadenosine (3-DAA) decreases PPRV replication. Further, we demonstrate that knockdown of writer (METTL3) and eraser (FTO) genes affects PPRV replication in a stable clone of host cells. Taken together, our results indicate that PPRV replication is highest in host cells, with relatively optimum m 6 A RNA modifications. Finally, we show that the presence of m 6 A in viral mRNA facilitates both stability and translation efficiency for better PPRV gene expression, and these events contribute to viral replication.

RESULTS
PPRV contains m 6 A modifications in its RNA. PPRV genomic RNA isolated from ultrapurified virus showed the presence of m 6 A modifications in the dot blot assay (Fig. 1A). Identification of m 6 A modifications in the PPRV transcripts was performed using Northern blot and m 6 A-meRIP (methylated RNA immunoprecipitation) sequencing. Here, we used RNA from two sources: (i) total RNA from PPRV-infected Vero cells and (ii) in vitro T7-transcribed (IVT) PPRV RNA. In the Northern blot, both the input and m 6 A antibody-enriched RNA originating from the PPRV-infected Vero cells displayed the presence of m 6 A modifications. However, only the input RNA but not the m 6 A antibody-enriched RNA originating from IVT showed a positive result, as the IVT RNA  Six m 6 A peaks were identified across different viral genes. The highest fold change was found in the region where the viral matrix (M) and fusion protein (F) genes are located. Other peaks were observed in the coding regions of the nucleocapsid (N) and phosphoprotein (P) genes (Fig. 1D). Overall, these results indicate that PPRV contains m 6 Amodified nucleotides in their mRNA and genomic RNA. PPRV infection affects the level of m 6 A modification and its related proteins. To understand the effect of PPRV infection on host m 6 A modification, we analyzed PPRVinfected host cells for m 6 A levels and also for the expression of genes related to m 6 A modification. It was found that there was reduced expression of m 6 A reader proteins (METTL3 and WTAP) at 24 h postinfection (hpi). We also observed similar changes in the m 6 A eraser protein (FTO). Further, expression of WTAP and FTO was found to have increased at 48 hpi. ( Fig. 2A and B). PPRV infection did not cause similar patterns of changes to the protein levels of these genes as observed in reverse transcription-quantitative PCR (RT-qPCR). The changes observed in the protein levels were not significant. Expectedly, the viral nucleocapsid (N) protein significantly increased at 48 hpi and 72 hpi (Fig. 2C), indicating the establishment and temporal progression of PPRV infection in the host cells. The m 6 A levels in the PPRV-infected host cells decreased at 24 hpi, followed by a subsequent increase, with the highest levels attained at 72 hpi. The phenomenon was accompanied by the expected patterns of m 6 A writer and eraser gene expression (Fig. 2D).
We performed double immunofluorescence staining for PPRV nucleocapsid (N) and m 6 A machinery proteins with specific antibodies at 24 and 48 hpi. METTL3 was found more in the nucleus in uninfected cells. However, it was mainly redistributed to the cytoplasm at 24 hpi, and at 48 hpi, it was found equally in the nucleus and the cytoplasm. WTAP was found in both the cytoplasm and the nucleus in uninfected cells and remained the same in the virus-infected cells (at both 24 hpi and 48 hpi). In the case of FTO, the protein was found in both the nucleus and the cytoplasm of uninfected cells. However, it was redistributed to the nucleus after infection (24 hpi and 48 hpi). ALKBH5 protein was found distributed in both the nucleus and the cytoplasm in uninfected cells, and no changes in distribution were observed after infection (Fig. 3). Inhibition of m 6 A modification reduces PPRV replication. A previous report indicated that 3-deazaadenosine (3-DAA) reduces m 6 A modification by decreasing the formation of SAM (S-adenosyl methionine) (23). Initially, the maximum concentration of 3-DAA (50 mM) with no significant impact on the proliferation of cultured Vero cells was determined (Fig. 4A). Later, we confirmed that the m 6 A levels in host cells progressively decreased as the concentration of 3-DAA in the cell culture was increased (Fig. 4B). We treated the cultured Vero cells with 3-DAA, followed by infection with PPRV, and analyzed the viral gene expression and replication. It was found that the expression of PPRV nucleocapsid protein progressively decreased in 3-DAA-treated cells ( Fig. 4C to 4E), indicating less virus present in these cells. The PPRV titer was also significantly reduced at a 3-DAA concentration of 50 mM (Fig. 4F). Overall, these results indicate that m 6 A modification in host cells is essential for PPRV replication.  PPRV replication increased with higher m 6 A RNA modifications. Meclofenamic acid (MA) is a small-molecule inhibitor of FTO (m 6 A eraser protein). For cell culture, we found 60 mM to be the highest concentration of MA that did not significantly affect the proliferation of cultured Vero cells (Fig. 5A). MA treatment increased the levels of m 6 A RNA modification in the host cells in a dose-dependent manner (Fig. 5B). In order to assess the impact of MA treatment on PPRV replication, we treated the cultured Vero cells with MA and then infected them with PPRV. Next, we analyzed the viral gene expression and also quantified the virus titer. We found that MA treatment increased the expression of PPRV nucleocapsid mRNA and protein, with the maximum expression at 37.5 mM MA ( Fig. 5C to 5E). However, at a concentration of 50 mM MA, we observed an inhibitory effect. Similar results were also found for PPRV replication, showing higher PPRV titers at 37.5 mM MA and significantly lower PPRV titers after treatment with 50 mM MA (Fig. 5F). MA is also known to inhibit COX enzymes, and its effect on viral replication may be due to COX enzyme inhibition in host cells. To test this possibility, we treated the host cells with a known COX inhibitor (indomethacin), followed by PPRV infection. The results indicated that indomethacin treatment did not significantly alter PPRV gene expression or replication (see Fig. S1 in the supplemental material). Hence, the effect of MA on PPRV gene expression and viral replication was due to its inhibition of FTO (m 6 A eraser) protein, which increased the m 6 A levels in the host cells. m 6 A methyltransferase and demethylase regulate PPRV replication. METTL3 is a major component of the m 6 A methyltransferase complex involved in creating m 6 A modifications. In order to understand the effects of reduction of METTL3 protein in the host cell on PPRV replication, we generated a stable knockdown clone of the METTL3 gene in Vero cells (METTL3_KD). The stable clones showed lower METTL3 mRNA and protein ( Fig. 6A and B). Further, METTL3_KD cells expectedly showed a reduction in m 6 A modifications in the mRNA compared to wild-type (WT) Vero cells (Fig. 6C). Next, to assess the role of host METTL3 on viral replication, both METTL3_KD and WT Vero FTO is a demethylase which, along with the ALKBH5, is involved in removing m 6 A modifications. In the present study, we generated a stable knockdown of FTO in Vero cells (FTO_KD). FTO_KD cells showed reduced FTO mRNA and protein ( Fig. 7A and B). Due to reduced demethylase activities, as expected, FTO_KD cells showed higher m 6 A levels than the WT Vero cells (Fig. 7C). To assess viral gene expression and replication, next, FTO_KD and WT Vero cells were infected with PPRV. The results indicated that both the mRNA of the PPRV nucleocapsid (N) gene and the viral nucleocapsid protein were lower in the FTO_KD than in the WT Vero cells at different time points of infection ( Fig. 7D and E). It was also revealed that the PPRV load increased significantly over a period of time in the WT Vero cells but not in the FTO_KD cells (Fig. 7F). Further, at all time points (24 hpi, 48 hpi, and 72 hpi), viral replication was significantly lower in the FTO_KD cells than in the WT Vero cells. Taken together, these data make it evident that m 6 A methyltransferase (METTL3) and demethylase (FTO) regulate PPRV replication in Vero cells.
A certain level of m 6 A RNA modification enhances PPRV replication. In "m6A methyltransferase and demethylase regulate PPRV replication" (above), we showed that knockdown of methyltransferase (METTL3_KD) and demethylase (FTO_KD) in host cells resulted in decreased PPRV replication. Next, we tested how different degrees of m 6 A modification in host cells affected PPRV replication. For this purpose, we used FTO_KD cells, which had a higher level of m 6 A modifications than the wild-type Vero cells. The m 6 A modifications in the FTO_KD cells were reduced by treating the cells  (Fig. 8B). The viral nucleocapsid protein levels showed similar patterns ( Fig. 8C and D). We separately subjected FTO_KD cells and WT Vero cells to treatment with 12.5 mM 3-DAA for 24 h, followed by PPRV infection. We found that the viral nucleocapsid protein levels and viral titers were significantly higher in FTO_KD cells treated with 3-DAA (12.5 mM) than in WT Vero cells (Fig. 8E and F). Taken together with the results observed in the previous section, this indicates that neither lower m 6 A modification (as in METTL3_KD) or higher m 6 A modification (as in FTO_KD) support PPRV replication. These data indicate that a certain level of m 6 A modification higher than the basal levels of the wild-type cells is required to facilitate increased PPRV replication. m 6 A modification improves PPRV mRNA stability and translation efficiency. We investigated the impact of m 6 A modifications on mRNA stability and translation efficiency by incorporating them with or without m 6 -ATP during in vitro-transcribed mRNA synthesis. For in vitro transcription, we used a partial PPRV nucleocapsid gene sequence and other essential elements for its translation, along with a His tag. Cultured Vero cells were transfected separately with the m 6 A-modified mRNA and unmodified mRNA. These cells were analyzed for stable mRNA copies that remained at subsequent time points after transfection. The results indicated that the modified mRNA was more stable than the unmodified mRNA from 24 h to 72 h (Fig. 9A). The translation was higher from the m 6 A-modified mRNA than from the unmodified mRNA (Fig. 9B). Similar observations were made about the immunofluorescence staining (Fig. 9C) and flow cytometric ( Fig. 9D and E) results. These findings indicate that m 6 A modification facilitates both higher stability and translation efficiency of the viral transcripts.
DISCUSSION m 6 A modification of viral RNA was first detected about 4 decades ago (27,28). However, recent observations have shown the importance of m 6 A modifications in virus-host interactions, which has stimulated interest in finding m 6 A-modified sites in different viruses (29,30). In the present study, we observed m 6 A modifications in both PPRV genomic RNA and viral mRNAs (tested for the N, P, F, M, and H genes). Previously, m 6 A modifications were reported for a variety of viruses, including HIV, influenza A virus, SV-40, RSV, HSV-1, HMPV, and SARS-CoV-2 (16,(19)(20)(21)(22)(23)(24)(25). Earlier, different studies on the same virus showed variations in the m 6 A-modified regions. For example, 14 distinct m 6 A methylation peaks were located in the splicing junctions, coding regions, and noncoding regions of HIV-1 (14). However, later, another group identified only four clusters of m 6 A modifications containing 2 or 3 m 6 A peaks in the HIV-1 genome (16). The variations in the results were likely due to differences in the Viruses can impact host m 6 A modification machinery to create a favorable microenvironment for their replication. We wanted to know whether PPRV infection alters the host m 6 A modification levels and has any effect on the expression of m 6 A writer and eraser proteins. It was found that m 6 A levels in the PPRV-infected host cells decreased initially (24 hpi) but increased at later time points (48 hpi and 72 hpi). The expression patterns of m 6 A writer (METTL3 and WTAP) and eraser (ALKBH5) proteins appeared to support these findings. Previously, it was reported that HIV-1 modulates the dynamics of m 6 A methylation in host T cells. At the active stage of HIV-1 replication (72 hpi), human MT4 CD4 1 T cells showed significantly higher m 6 A levels than the uninfected control cells (14). Our observations also reflect similar higher levels of m 6 A in PPRVinfected cells at 72 hpi. Interestingly, in Kaposi's sarcoma-associated herpesvirus (KSHV)-infected cells, the level of m 6 A methylation was increased when latent KSHV was stimulated to undergo lytic replication (13). Taken together, these findings indicate that virus-induced higher m 6 A modification in host cells may facilitate active viral replication. We also found virus-induced changes in the subcellular localization of methyltransferase (METTL3) and demethylase (FTO). PPRV infection causes redistribution of predominantly nuclear METTL3 to the cytoplasm. After PPRV infection, FTO protein becomes more concentrated in the nucleus. Similar virus-induced changes in the subcellular localization of methyltransferase and demethylase were previously reported in enterovirus-infected Vero cells (31). It has been shown that cells subjected to heat shock display redistribution of m 6 A modification proteins to the cytoplasm to facilitate the translation of stress proteins (32). Viral infection induces a host cell stress response whereby m 6 A modification might favor translation by changes in the subcellular localization of METTL3.
Treatment with 3-deazaadenosine (3-DAA) reduces the formation of S-adenosyl methionine (SAM) and inhibits all types of methylation reactions. Hence, reduction of the levels of m 6 A of mRNA is one of the effects of 3-DAA treatment. We found that 3-DAA treatment decreased m 6 A methylation in the host Vero cells, and it also reduced PPRV gene expression and its replication. The mechanism of host m 6 A reduction is a recent understanding that explains the earlier findings of 3-DAA-mediated antiviral effects against multiple viruses (33)(34)(35). This inhibitory mechanism of m 6 A editing by 3-DAA and its effect on the replication of viruses have been further explored in recent studies. It was reported that 3-DAA treatment significantly reduced the viral gene expression and replication of HSV-1 and HIV-1 (23,36). Our results similarly indicated that reducing m 6 A RNA methylation levels has an inhibitory effect on PPRV gene expression and replication. These findings also corroborate our data showing that knockdown of METTL3 (specific reduction of m 6 A modification levels) caused a reduction in PPRV gene expression and replication. Initially, it was reported that FTO protein uses m 6 Am (N 6 ,29-O-dimethyladenosine)modified RNA as its substrate (37). However, later studies clearly indicated that the major substrate for FTO demethylation activity is m 6 A modification of mRNA (38). Meclofenamic acid (MA) was found to specifically inhibit the m 6 A eraser function of FTO, and its competitive inhibition was confirmed through detailed structure-based results (39). Inhibition of the m 6 A eraser function of FTO protein leads to impairment in different biological functions (40,41). We used MA to understand the effect of increased m 6 A methylation levels in host cells on PPRV replication. MA treatment increased PPRV gene expression and its replication in a dose-dependent manner but only to a certain extent. The results suggest that higher m 6 A than the basal levels of host Vero cells facilitates viral replication.
We took another approach to reducing the major m 6 A machinery proteins to capture their effect on PPRV replication. Accordingly, we generated host cells with stable knockdown of METTL3 and FTO genes. These knockdown clones expectedly exhibited their respective effects on m 6 A modifications. Interestingly, we found reduced PPRV gene expression and replication in both the METTL3 and FTO knockdown conditions. Earlier studies reported that silencing of METTL3 reduced viral replication, whereas silencing FTO increased replication, for viruses such as RSV, HSV-1, SARS-CoV-2, and enterovirus 71 (21,(23)(24)(25)31). However, the direct correlation between the level of m 6 A editing and viral replication was not withstanding the conflicting results reported by some studies. For example, depletion of m 6 A methyltransferases or an m 6 A demethylase respectively increases or decreases viral replication in the case of flaviviruses (18). On the contrary, we found decreased PPRV replication in both METTL3 and FTO knockdown cells. These divergent results may be due to the type of host cells, infection stages, and strategies used by different viruses.
Further, to confirm the relationship between the levels of m 6 A modification and PPRV replication, we treated stable FTO knockdown cells (which had higher m 6 A levels than the wild-type cells) with 3-DAA and analyzed PPRV replication. Specifically, at certain m 6 A levels (achieved by treatment of FTO_KD cells with 12.5 mM 3-DAA), the PPRV gene expression and viral replication were increased. Our results indicate that the maximum viral replication occurs with optimal m 6 A editing, and this phenomenon may be unique for the virus and the host cells involved. The requirement of a specific m 6 A editing level may be essential for the viral RNA stability, specific localization, and translation that determine viral replication (42). Further, successful viral replication is potentially linked to the m 6 A alterations in the host cells (5). In the present study, we created a platform of host cells by stably knocking down major m 6 A machinery proteins, and this platform enabled us to set different levels of m 6 A modifications using small-molecule inhibitors. Our results indicated a relationship between the m 6 A RNA levels of the host cells and PPRV replication (Fig. 10).
Among the different molecular mechanisms that regulate mRNA half-life (stability and degradation), m 6 A modification is considered to play a major role (43). In conjunction with m 6 A-modified sites, YTH domain-containing proteins (YTHDF1, YTHDF2, and YTHDF3) regulate different aspects of the mRNA life cycle (44). The present study assessed the effect of m 6 A modification on the biology of viral mRNA. The results indicate that m 6 A-modified PPRV mRNA had a longer half-life and higher translation efficiency than the unmodified PPRV mRNA. Previously, it was reported that YTHDF2 binds m 6 A sites and recruits adaptor proteins to trigger the degradation of m 6 A-containing mRNA (45,46). YTHDF3, on the other hand, facilitates the translation of m 6 A-modified mRNA in collaboration with YTHDF1 and influences the YTHDF2-mediated decay process (47). This indicates that viral RNA utilizes m 6 A modification to promote translation and turnover, contributing to viral replication in the host cells.
In summary, the present study revealed that PPRV has m 6 A-modified sites in its RNA and that PPRV affects the dynamics of m 6 A modification in the infected host cells. Reducing the host m 6 A modification levels decreased the PPRV replication, whereas increasing the host m 6 A levels enhanced viral replication up to a certain limit. Further, host cells with stable knockdown of the METTL3 and FTO genes did not completely support PPRV replication compared to the wild-type cells. It was found that both higher and lower levels of m 6 A modification in the host cells may adversely affect viral replication. Higher PPRV gene expression and replication were observed at an optimal level of m 6 A editing in the host cells. Based on these results, we summarized the relationship between host m 6 A levels and PPRV replication (Fig. 10). Further, m 6 A-modified viral transcripts had better stability and translation efficiency compared to unmodified mRNA. Together, these results indicate that the process of m 6 A modification in host cells regulates PPRV replication. These findings contribute to a way forward to regulate viral replication and devise novel antiviral strategies by modulating the host m 6 A modification dynamics.

MATERIALS AND METHODS
Cell culture and viruses. Vero cells (African green monkey kidney cells) were cultured in Dulbecco's modified Eagle's medium (12800-017; Gibco) supplemented with 10% fetal bovine serum (1600-044; Gibco) with 5% CO2 at 37°C. Peste des petits ruminants virus (PPRV) (strain Sungri/96) was obtained from the Division of Virology, Mukteshwar, Indian Veterinary Research Institute (IVRI), and was propagated and multiplied in Vero cells. Titration was determined as the 50% tissue culture infective dose (TCID 50 ) using the Reed-Muench formula (48). m 6 A dot blot analysis. Supernatant from the PPRV-infected Vero cell culture was collected and clarified to remove cell debris by centrifugation at 8,000 rpm for 10 min. The clarified fluid was filtered and subjected to ultracentrifugation to pellet the virus. RNA was isolated from the purified virus using TRIzol reagent (15596026; Invitrogen). In vitro PPRV RNA was transcribed from plasmid DNA template containing an N-gene sequence using the MEGAscript T7 kit (AM1333; Ambion) according to the manufacturer's instructions, with or without the addition of m 6 -ATP for modified or unmodified IVT RNA. For dot blot analysis, PPRV RNA and in vitro-transcribed RNA were adsorbed onto positively charged N1 Hybond membranes (11209299001; Roche), followed by UV cross-linking. Detection of m 6 A modifications was performed using anti-m 6 A antibody (202003; Synaptic Systems). There are always concerns regarding the specificity of anti-m 6 A antibody. However, the anti-m 6 A antibody used in the present study has been widely reported in other previous reports (24). We used anti-rabbit-alkaline phosphatase (AP) secondary antibody with a BCIP (5-bromo-4-chloro-3-indolyl phosphate)-NBT (nitro blue tetrazolium) detection system (B1911; Sigma-Aldrich). Further, the dot blot assay for m 6 A level estimation is a semiquantitative method, and it cannot capture subtle differences in m 6 A levels. Methylated RNA immunoprecipitation. MeRIP was conducted following a previously described protocol (49). Briefly, total RNA (250 mg) was isolated from Vero cells infected with PPRV at 48 hpi. Purification of mRNA from total RNA was performed using Dynabeads oligo(dT) 25 (61005; Life Technologies) according to the manufacturer's instructions. The purified mRNA was fragmented using Ambion RNA fragmentation reagents (AM8740; Life Technologies) and repurified with overnight ethanol precipitation. Part of the fragmented RNA was retained as an input control. m 6 A RNA enrichment was performed using the EpiMark N 6 -methyladenosine enrichment kit (E1610S; NEB) following the manufacturer's instructions. Briefly, m 6 A antibody was attached to protein G magnetic beads by incubation with orbital rotation for 30 min at 4°C. The prepared RNA was bound to the beads coated with m 6 A by incubation with orbital rotation for 1 h at 4°C. After careful washing, the enriched RNA was eluted with elution buffer, followed by cleanup and concentration of the eluted RNA. MeRIP and input control RNA were used for further analysis.
Northern blotting. Northern blotting was performed using the DIG Northern starter kit (12039672910; Roche), following the manufacturer's instructions. Briefly, input RNA and m 6 A-MeRIP RNA were separated on 2% agarose with 2.2-M formaldehyde gels in running buffer (20 mM MOPS [morpholinepropanesulfonic acid], 5 mM sodium acetate, 2 mM EDTA, pH 7.0). The RNAs were transferred to Hybond N1 hybridization membranes (11209299001; Roche) in 20Â SSC buffer (3.0 M NaCl, 0.3 M sodium citrate) overnight. The transferred RNA was UV cross-linked to the membrane and hybridized with a DIG-labeled PPRV N-gene probe (100 ng/mL) for 6 h at 68°C. The membrane was washed 2 times with 2Â SSC buffer and 2 times with 0.1Â SSC and 0.1% SDS. The membrane was blocked with blocking solution for 30 min at room temperature (RT) and incubated for 30 min in antibody solution. The membrane was washed with washing buffer and equilibrated for 2 to 5 min in detection buffer; the membrane was then placed in CDP-Star solution. Detection signals were developed on a ChemiDoc MP imaging system (Bio-Rad Laboratories, USA).
MeRIP-Seq and analysis. MeRIP sequencing of RNA (input transcriptome sequencing [RNA-Seq] and MeRIP-Seq) was performed by preparing libraries using Illumina TruSeq stranded mRNA kits following the manufacturer's instructions. The samples were sequenced on a HiSeq 2500 instrument in singleread 50-base format. Bioinformatics analysis of the MeRIP-Seq data was carried out following standard methods. Briefly, RNA-Seq reads were aligned to the PPRV genome (GenBank accession no. KF727981) using BWA software, keeping only uniquely mapping reads. Peaks were called using exomePeak, with 20-bp windows to test for statistically significant enrichment in the Immunoprecipitation (IP) relative to the control input RNA with an adjusted (Benjamini-Hochberg [BH]) P value cutoff of 0.05. The aligned reads and coverage were visualized using IGV and other features of BEDTools 2.5.
Double immunofluorescence confocal microscopy. Vero cells were seeded into six-well plates 1 day before infection at ;50% confluence; then, the cells were infected with PPRV (multiplicity of infection [MOI], 1.0) and incubated for the indicated times. Indirect immunofluorescence assay was performed. In brief, the cells were washed three times with PBS, fixed in 4% paraformaldehyde (R143; G Biosciences) in PBS for 15 min, permeabilized in 0.5% Triton X-100 (MB031-100ML; HiMedia) for 15 min, and blocked in 5% bovine serum albumin (MB083-25G; HiMedia) for 30 min at room temperature. The cells were incubated with primary antibodies, diluted as suggested by the manufacturer, overnight at 4°C; the cells were then washed three times with PBS and stained with the corresponding secondary antibody for 1 h at room temperature. The nuclei were stained with DAPI (D9542; Sigma). The slides were observed under a confocal microscope (FV1000; Olympus).
Generation of stable knockdown Vero cells. Lentivirus transduction particles (Mission; SHCLNV; Sigma) expressing short hairpin RNA (shRNA) and specifically targeting the m 6 A writer protein, METTL3 (SCHLNV NM_019852; GPP Web Portal identifier, TRCN0000289812; target sequence, CGTCAGTATCTTGGG CAAGTT), and the m 6 A eraser protein, FTO (SCHLNV; GenBank accession no. NM_001080432; GPP Web Portal identifier, TRCN0000246250; target sequence, TCACCAAGGAGACTGCTATTT), were used for generating stable knockdown Vero cells. Cells (60% to 70% confluent) were transduced with particles following the manufacturer's instructions. Briefly, after 6 h of incubation, fresh medium was added with puromycin for selection, and the medium was replenished every 3 to 4 days. Individual resistant colonies were picked up and expanded. The clones were tested for knockdown of the target genes (METTL3 and FTO) using RT-qPCR and Western blot assay.
Reverse transcription-quantitative PCR. Total RNA was extracted using TRIzol reagent (15596026; Invitrogen). Reverse transcription was performed with 3 mg of total RNA using the RevertAid cDNA synthesis kit (K1622; Thermo Fisher). RT-qPCR was performed using SYBR green QuantiFast PCR master mix (204054; Qiagen) on a CFX Connect real-time system (Bio-Rad).
Statistical analysis. Statistical analysis of the RT-qPCR data and other observations was performed using a two-tailed unpaired t test using GraphPad Prism software (La Jolla, CA, USA). Data are presented as the means 6 standard error of the mean (SEM) (n = 3). All experiments were repeated at least three times.
Data availability. The MeRIP-Seq data have been submitted to GenBank under the BioProject accession no. PRJNA896352.

SUPPLEMENTAL MATERIAL
Supplemental material is available online only. SUPPLEMENTAL FILE 1, PDF file, 0.1 MB.

ACKNOWLEDGMENTS
This work was supported by the DBT/Wellcome Trust India Alliance (grant no. IA/E/ 17/1/503689 awarded to B.S.). We acknowledge support from IVRI for extending the facilities to carry out this research work.