Skip to main content
NCBI home page
Science Advances logoLink to Science Advances
. 2024 Aug 7;10(32):eadn9519. doi: 10.1126/sciadv.adn9519

Epitranscriptomic m5C methylation of SARS-CoV-2 RNA regulates viral replication and the virulence of progeny viruses in the new infection

Hongyun Wang 1,†, Jiangpeng Feng 1,†, Zhiying Fu 1, Tianmo Xu 1, Jiejie Liu 1, Shimin Yang 1, Yingjian Li 1, Jikai Deng 1, Yuzhen Zhang 1, Ming Guo 1, Xin Wang 1, Zhen Zhang 1,2, Zhixiang Huang 1,2, Ke Lan 1,2, Li Zhou 1,2, Yu Chen 1,2,*
PMCID: PMC11305390  PMID: 39110796

Abstract

While the significance of N6-methyladenosine (m6A) in viral regulation has been extensively studied, the functions of 5-methylcytosine (m5C) modification in viral biology remain largely unexplored. In this study, we demonstrate that m5C is more abundant than m6A in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and provide a comprehensive profile of the m5C landscape of SARS-CoV-2 RNA. Knockout of NSUN2 reduces m5C levels in SARS-CoV-2 virion RNA and enhances viral replication. Nsun2 deficiency mice exhibited higher viral burden and more severe lung tissue damages. Combined RNA-Bis-seq and m5C-MeRIP-seq identified the NSUN2-dependent m5C-methylated cytosines across the positive-sense genomic RNA of SARS-CoV-2, and the mutations of these cytosines enhance RNA stability. The progeny SARS-CoV-2 virions from Nsun2 deficiency mice with low levels of m5C modification exhibited a stronger replication ability. Overall, our findings uncover the vital role played by NSUN2-mediated m5C modification during SARS-CoV-2 replication and propose a host antiviral strategy via epitranscriptomic addition of m5C methylation to SARS-CoV-2 RNA.

The host m5C methyltransferase NSUN2 regulates m5C modification, viral replication, and pathogenesis of SARS-CoV-2.

INTRODUCTION

COVID-19 caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has become a globe health challenge since December 2019 and is still going on (1–3). SARS-CoV-2 belongs to betacoronaviruses, containing a single-stranded positive-sense genomic RNA (gRNA) of approximately 30 kb with 5′-cap structure and 3′-poly-A tail (4). The positive-sense gRNA is used as the template to translate polyprotein 1a/1ab (pp1a/pp1ab) (5). The two large open reading frames (ORF1a/1b) near the 5′ terminus of gRNA encode pp1a/pp1ab, which is then cleaved into 16 viral nonstructural proteins (nsp1 to nsp16) by nsp3 harboring papain-like protease (PLPpro) activity and nsp5 harboring 3C-like protease (3CLpro) activity (6). The nsps form the replication-transcription complex. Subsequently, the subgenomic RNAs (sgRNAs) are generated by ORFs the near the 3′ terminus of gRNA through transcription-regulating sequence (TRS)–dependent template switching by the replication-transcription complex and encode four main structural proteins, including spike (S), envelope (E), membrane (M), nucleocapsid (N), and several accessory proteins (7). The positive-sense gRNA also serves as the template for replication and transcription to produce the negative-sense RNA intermediates by nsp12 harboring RNA-dependent RNA polymerase activity. The negative-sense RNA intermediates subsequently serve as the templates to produce positive-sense gRNA and sgRNAs (8).

More than 170 types of RNA chemical modifications to various types of RNAs have been characterized thus far, which are involved in physiological and pathological processes via regulating RNA metabolism, including stability, translation efficiency, alternative splicing, transport, and localization (9–12). N6-methyladenosine (m6A) and 5-methylcytosine (m5C) are particularly notable and extensively existed (13, 14). m6A has been reported to be involved in the modification and regulation of a large variety of viruses, including HIV (Retroviridae) (15–19), hepatitis B virus (HBV, Hepadnaviridae) (20), Zika virus (Flaviviridae) (21, 22), human metapneumovirus (Pneumoviridae) (23), Kaposi’s sarcoma-associated herpesvirus (Herpesviridae) (24–26), enterovirus 71 (Picornaviridae) (27), influenza A virus (Orthomyxoviridae) (28, 29), human respiratory syncytial virus (Paramyxoviridae) (30, 31), endogenous retroviruses (Retroviridae) (32), polyomavirus simian virus 40 (Polyomaviridae) (33), and so on. Especially recently, m6A were reported to be widely distributed in SARS-CoV-2 gRNA. Although inconsistent results of the METTL3-mediated regulation to the replication of SARS-CoV-2 have been demonstrated, all these studies indicate that m6A modification plays an important regulatory function in SARS-CoV-2 life cycle (34–40). However, compared to m6A, little is known about the functions of m5C modification machinery in most physiological and pathological contexts upon various viral infections. So far, only a few viruses, including HIV (41, 42), HBV (43), murine leukemia virus (Retroviridae) (44, 45), and Epstein-Barr virus (Herpesviridae) (46), have been reported to be modified and regulated by m5C modification machinery. However, whether there is m5C methylation modification in SARS-CoV-2 gRNA or the m5C modification participates in the regulation of SARS-CoV-2 replication and pathogenic processes remains to be investigated.

In this study, we characterized the m5C RNA modifications present in SARS-CoV-2 using mass spectrometry (MS) and next-generation sequencing. Our findings revealed that NSUN2, a typical RNA m5C methyltransferase, functions as a negative regulator of SARS-CoV-2 replication. We profiled the widespread distribution of m5C methylation sites in positive-sense SARS-CoV-2 gRNA. Mechanistically, NSUN2 mediated the methylation of SARS-CoV-2 RNA transcripts and facilitated their degradation. Knockout of NSUN2 notably reduced m5C levels in SARS-CoV-2 virion RNA and increased viral infection. By using the transcription and replication-competent SARS-CoV-2 virus–like particles (trVLP) system, we found that the SARS-CoV-2 green fluorescent protein (GFP)/ΔN trVLPs with low NSUN2-mediated m5C modification exhibited higher replication ability. Consistently, the progeny SARS-CoV-2 virions from Nsun2+/− mice (with low m5C level) exhibited a stronger replication ability in both Caco-2 cells and K18-hACE2 mice than the progeny SARS-CoV-2 virions from Nsun2+/+ mice (with normal m5C level). Overall, our findings highlight the vital role of NSUN2-mediated m5C methylation during SARS-CoV-2 replication and pathogenesis. We outline a paradigm that host has evolved an antiviral strategy via epitranscriptomic addition of m5C methylation to SARS-CoV-2 RNA, which will be an insight valuable for developing new anti-coronavirus drugs.

RESULTS

SARS-CoV-2 gRNA contains more abundant m5C modification compared to m6A and other modifications

To investigate RNA modifications that SARS-CoV-2 RNA contains, we performed liquid chromatography tandem MS (LC-MS/MS) of SARS-CoV-2–infected Caco-2 cells or purified virions (Fig. 1A). We identified several modified ribonucleosides in SARS-CoV-2 RNA, including 2′-O–methylated derivatives of four canonical nucleosides (Am, Cm, Um, and Gm), m5C, and m6A. Compared to total RNA of uninfected Caco-2 cells (Fig. 1B), m5C is the most abundant RNA modification both in total RNA of SARS-CoV-2–infected Caco-2 cells (Fig. 1C) or purified SARS-CoV-2 virion RNA (Fig. 1D) among these modifications. Next, we developed a biosafety level 2 (BSL-2) cell culture system for the production of trVLP, which expresses GFP protein replacing viral N protein (SARS-CoV-2 GFP/ΔN trVLP). We also purified SARS-CoV-2 GFP/ΔN trVLP virion and identified many modified ribonucleosides in its RNA. Consistently, m5C is the most abundant modification compared with many other RNA modifications (fig. S1). Furthermore, we performed m5C-methylated RNA immunoprecipitation sequencing (m5C-MeRIP-seq) in Caco-2 cells, which were infected with SARS-CoV-2 for 48 hours to profile the m5C landscape along SARS-CoV-2 transcriptome. As shown in Fig. 1E, the m5C is enriched in the region near the 3′ untranslated region.

Fig. 1. SARS-CoV-2 gRNA contains abundant m5C modification.

Fig. 1.

Open in a new tab

(A) Schematic diagram of LC-MS/MS of purified SARS-CoV-2 virions or SARS-CoV-2–infected Caco-2 cells at multiplicity of infection (MOI) = 0.02 for 48 hour. (B) Analysis of m5C/C ratio using LC-MS/MS of total RNAs extracted from uninfected Caco-2 cells. (C and D) Analysis of m5C/C ratio using LC-MS/MS of total RNAs extracted from SARS-CoV-2–infected Caco-2 cells (C) or purified SARS-CoV-2 virion RNA (D). (E) Visualization of m5C-MeRIP-seq results shows regions of enrichment for m5C immunoprecipitation (upper) over input (lower) from SARS-CoV-2–infected Caco-2 cells. Integrative Genomics Viewer tracks displaying read distributions from m5C-MeRIP-seq (red panel) and Input (blue panel) along SARS-CoV-2 positive-sense RNA. Data are representative of three independent experiments and were analyzed by two-tailed unpaired t test. Graphs show the means ± SD (n = 3) derived from three independent experiments [or two independent experiments for (E)].

NSUN2 inhibited SARS-CoV-2 replication and regulated m5C modification of SARS-CoV-2 RNA

To further investigate whether m5C machinery regulates SARS-CoV-2 infection and replication, we knocked down currently known writers or readers (47, 48) of mRNA m5C modification and NOP2/SUN RNA methyltransferase family members (49) in Caco-2 cells using small interfering RNAs (siRNAs) and detected the endogenous viral RNA levels of SARS-CoV-2. Knockdown of NSUN2 enhanced endogenous SARS-CoV-2 N and E RNA levels, while the other currently known writers or readers did not (Fig. 2A and fig. S2). Notably, knockdown of METTL3 or METTL14 only mildly enhanced endogenous SARS-CoV-2 N and E RNA levels. We thus speculated that NSUN2 may specifically and notably regulate SARS-CoV-2 replication compared to other currently known methyltransferases or readers. We then constructed NSUN2 knockout cell lines in Caco-2 cells using CRISPR-Cas9 (fig. S3) and found that knockout of NSUN2 showed notable elevation of the RNA levels of SARS-CoV-2 N and E intracellularly and extracellularly (Fig. 2B). Consistently, knockout of NSUN2 dramatically elevated the protein levels of SARS-CoV-2 N (Fig. 2E).

Fig. 2. NSUN2 regulated m5C modification and replication of SARS-CoV-2.

Fig. 2.

Open in a new tab

(A) qPCR analysis of RNA levels of N or E in Caco-2 cells transfected with siControl or siRNAs for 36 hours, with infection by SARS-CoV-2 at an MOI = 0.02 for another 24 hours. (B and C) qPCR analysis of RNA levels of N or E in cell or supernatant from WT Caco-2 cells or NSUN2 knockout cells, with infection by SARS-CoV-2 at an MOI = 0.02 for 24 hours, with (B) or without (C) ruxolitinib treatment. (D) qPCR analysis of RNA levels of N or E in cell or supernatant from WT Vero E6 cells or NSUN2 knockout cells, with infection by SARS-CoV-2 at an MOI = 0.02 for 24 hours. (E) Immunofluorescence analysis of N protein (red) in WT Caco-2 cells or NSUN2 knockout cells, with infection by SARS-CoV-2 at an MOI = 0.02 for 24 hours. Scale bar, 50 μm. (F) Immunofluorescence analysis of N protein (red) in WT Vero E6 cells or NSUN2 knockout cells, with infection by SARS-CoV-2. Scale bar, 50 μm. (G) Northern blot results from RNA probes targeting (+) sgRNAs (N probe) in WT Caco-2 cells or NSUN2 knockout cells, with infection by SARS-CoV-2. (H) Analysis of m5C/C ratio using LC-MS/MS of purified SARS-CoV-2 virion RNA from WT cells or NSUN2 knockout cells. (I) Bis-seq results of unique m5C-methylated sites along SARS-CoV-2 positive-sense RNA in SARS-CoV-2–infected WT Caco-2 cells versus NSUN2 knockout Caco-2 cells. Data are representative of three independent experiments and were analyzed by two-tailed unpaired t test. Graphs show the means ± SD (n = 3) derived from three independent experiments [or two independent experiments for (I)]. NS, not significant for P > 0.05, **P < 0.01, ***P < 0.001.

Our previous study identified that NSUN2 participates in the regulation of virus-induced innate immunity responses (50). To block out the effects of interferon, we used ruxolitinib, a potent and selective Janus kinase 1/2 inhibitor that blocks signaling downstream of type I interferon receptors. As shown in Fig. 2C, the promotion of SARS-CoV-2 propagation in NSUN2-knockout cells was not affected by ruxolitinib treatment. This result further confirmed that promotion of SARS-CoV-2 propagation in the NSUN2-deficient cells was independent of the interferon response. Moreover, we also constructed NSUN2 knockout cell lines in Vero E6 cell, which is deficient in interferon secretion, and the same results were also obtained in NSUN2-knockout Vero E6 cell lines compared with wild-type (WT) Vero E6 cells (Fig. 2, D and F). Meanwhile, we observed that the promotion of SARS-CoV-2 propagation in NSUN2-knockout cells was mildly increased when blocking out the effects of interferon, further demonstrating the notable direct NSUN2-mediated regulation of SARS-CoV-2 replication (Fig. 2, B to D). In addition, to explore the global landscape change of sgRNAs after NSUN2 knockout, we verified the presence of sgRNAs in SARS-CoV-2–infected Caco-2 cells using Northern blotting. The result showed that the global landscape of sgRNAs in NSUN2-knockout cells notably increased compared to WT Caco-2 cells (Fig. 2G). The replication of SARS-CoV-2 takes place at endoplasmic reticulum–derived double-membrane vesicles (ER-DMVs). We further checked whether NSUN2 exists in ER-DMVs. We purified the ER-DMVs from SARS-CoV-2–infected cells, and NSUN2 protein was identified in the ER-DMVs (fig. S4A). Moreover, we also perform nuclear and cytoplasmic protein extraction to investigate the subcellular localization of endogenous NSUN2 upon SARS-CoV-2 infection. Consistent with previous results, NSUN2 were detected mostly in the nucleus in uninfected cells. However, upon SARS-CoV-2 infection, the distribution of NSUN2 has changed and is more distributed in the cytoplasm (fig. S4B). These results further demonstrate the colocalization and critical role of NSUN2 in regulation of SARS-CoV-2 replication.

To assess the m5C level of SARS-CoV-2 gRNA in virion from WT or NSUN2 knockout cell lines, we further purified SARS-CoV-2 virions released from Vero E6 cell line and Caco-2 cell line and corresponding NSUN2 knockout cell lines. The LC-MS/MS results provided direct evidence of a notable decrease in m5C levels in virion RNA after NSUN2 knockout (Fig. 2H). To further confirm the specific m5C modification sites, we performed RNA-bisulfite sequencing (RNA-Bis-seq) both in WT Caco-2 cells and NSUN2 knockout Caco-2 cells that were infected with SARS-CoV-2 (Fig. 2I and figs. S5 and S6). We identified a large amount of m5C-methylated sites among SARS-CoV-2 gRNA. Here, we showed the unique m5C-methylated sites of positive-sense SARS-CoV-2 gRNA in WT groups versus NSUN2 knockout groups (Fig. 2I). These sites were mediated by NSUN2 and scattered across positive-sense gRNA. The relatively high methylation sites were mainly distributed in nsp14, nsp15, nsp16, S, ORF3a, E, M, and N.

The screened m5C-methylated cytosines in SARS-CoV-2 were identified to regulate RNA degradation

We then sought to study the function of these relatively high m5C methylation sites. We prepared RNA segments of each region of SARS-CoV-2 gRNA by in vitro transcription. The RNAs were used for in vitro methylation assays using recombinant glutathione S-transferase (GST)–NSUN2 and 3H-labeled S-adenosyl methionine (SAM). These transcribed RNA segments can be methylated by NSUN2 (Fig. 3, A and B). Meanwhile, the data showed that NSUN2 can more efficiently mediate the methylation of N in vitro compared with other regions. Furthermore, we prepared RNA segments with C or m5C. These transcribed RNA segments with C or m5C were transfected into cells after capping (Fig. 3, A and C, and figs. S7 and S8). The quantitative polymerase chain reaction (qPCR) results in the expression levels of these transcribed RNA segments with m5C in nsp14, nsp15, nsp16, S, E, M, and N being lower than those corresponding transcribed RNA segments with C (Fig. 3C and figs. S7 and S8). However, the incorporation of m5C mildly promoted the RNA stability of ORF3a but did not affect the RNA degradation of nsp6 (with no notable NSUN2-dependent m5C site), indicating that not all m5C cause degradation (fig. S7). We speculated that incorporation of m5C into these transcribed RNA segments may affect the stability of RNA of nsp14, nsp15, nsp16, S, E, M, and N. We then tested whether these identified highly methylated cytosines were indeed involved in the regulation of mRNA stability. We constructed expression plasmids containing either WT or site mutations (table S2) of methylated cytosines of nsp14, nsp15, nsp16, S, E, M, or N and found that most of the site mutations of methylated cytosines increased the corresponding RNA expression levels (Fig. 3D). We then measured the stability of these transcripts and found that the transcripts of site mutation of methylated cytosines were consistently more stable than the WT transcripts, indicating that methylation of these cytosines by NSUN2 was indeed critical for regulating their mRNA stability and degradation (Fig. 3, E to K, and fig. S9).

Fig. 3. NSUN2 regulated SARS-CoV-2 RNA degradation.

Fig. 3.

Open in a new tab

(A) Schematic diagram of in vitro–transcribed RNAs with C or m5C using in vitro methylation assays or transfection. (B) In vitro m5C methylation assays using recombinant GST-NSUN2 and the in vitro transcripts as indicated. (C) qPCR analysis of RNA levels of the corresponding gene in Huh7 cells transfected with transcribed RNAs with C or m5C for 10 hours. (D) qPCR analysis of RNA levels of the corresponding gene in Huh7 cells transfected with WT or m5C modification site-mutations (mut) of each gene. (E to K) Stability analysis of RNA of each gene in Huh7 cells transfected with WT or m5C modification site-mutations (mut) of each gene for 24 hours, with treatment of actinomycin D (ActD) for another 0, 3, 6, and 9 hours. Data are representative of three independent experiments and were analyzed by two-tailed unpaired t test. Graphs show the means ± SD (n = 3) derived from three independent experiments. NS, not significant for P > 0.05, **P < 0.01, ***P < 0.001.

Coronaviruses are known to generate sgRNAs through TRS-dependent template switching. The N region is located near the 3′ terminus of gRNA. Most of the sgRNAs contain the N region, and the N sgRNA is the most abundant among the various sgRNAs. The results of RNA-Bis-seq (Fig. 2I and fig. S5) and m5C-MeRIP-seq (Fig. 1E) both revealed that the N region was highly methylated. Therefore, we chose to further investigate the methylation and biological function of NSUN2 to N. On the one hand, we found that overexpression of NSUN2 inhibited the RNA levels of N, while knockout of NSUN2 elevated the RNA levels of N both in Vero E6 cells and human embryonic kidney (HEK) 293T cells (Fig. 4A and fig. S10A). Reconstitution of WT NSUN2 but not the methyltransferase inactivated mutant (50) in NSUN2 knockout cells restored the inhibition function to N RNA levels. Of note, the N transcripts in NSUN2 knockout cells showed higher RNA stability compared to WT cells (Fig. 4B and fig. S10B). Correspondingly, reconstitution of WT NSUN2 but not the methyltransferase inactivated mutant in NSUN2 knockout cells restored the inhibitory function on RNA stability of N transcripts. On the other hand, we developed a BSL-2 cell culture system for production of trVLP, which expresses GFP protein replacing viral N protein (SARS-CoV-2 GFP/ΔN trVLP). We then used the lentiviral system to generate stable WT SARS-CoV-2-N or SARS-CoV-2-N-C28840-mut cell lines in Caco-2 cells, which can be infected by SARS-CoV-2 GFP/ΔN trVLP. By this system, we identified that the SARS-CoV-2 GFP/ΔN trVLP in the SARS-CoV-2-N-C28840-mut Caco-2 cell line showed higher replication level than that in the SARS-CoV-2-N-WT Caco-2 cell line (Fig. 4, C to F). Both intracellular (Fig. 4, C and E) and extracellular viruses (Fig. 4F) were notably elevated in the SARS-CoV-2-N-C28840-mut Caco-2 cell line. Because of the gain of function of SARS-CoV-2 caused by the mutation of the m5C methylation sites, reverse genetics cannot be used to make mutant recombinant virus. Together, the results verified that this regulation of N transcripts by NSUN2 was dependent on its m5C methyltransferase activity, and the mutation of m5C site on N promoted the trVLP replication level.

Fig. 4. NSUN2 regulated RNA level of SARS-CoV-2 N gene depending on its methyltransferase activity.

Fig. 4.

Open in a new tab

(A) qPCR analysis of RNA levels of N gene in WT Vero E6 cells (NSUN2+/+) or NSUN2 knockout Vero E6 cells (NSUN2−/−), transfected with SARS-CoV-2 N and WT NSUN2 (NSUN2-WT) or I302A/C321A double mutation (NSUN2-DM) for 36 hours. (B) Stability analysis of RNA levels of N gene in WT Vero E6 cells (NSUN2+/+) or NSUN2 knockout Vero E6 cells (NSUN2−/−), transfected with SARS-CoV-2 N and WT NSUN2 (NSUN2-WT) or I302A/C321A double mutation (NSUN2-DM) for 24 hours, with treatment of actinomycin D (ActD) for another 0, 2, and 4 hours. (C) Immunofluorescence analysis of SARS-CoV-2 GFP/ΔN trVLP (green) in Caco-2 cells reconstituted with WT N (N-WT) or N-C28840-Mut by lentiviral system, with infection by SARS-CoV-2 GFP/ΔN trVLP at an MOI = 0.05 for 48 to 72 hours. Scale bar, 50 μm. (D) Western blot analysis of N protein expression levels in Caco-2–N-WT and Caco-2–N-C28840-Mut stable cell lines. (E) qPCR analysis of RNA levels of N, E, and S genes in Caco-2 cells reconstituted with WT N (N-WT) or N-C28840-Mut by lentiviral system, with infection by SARS-CoV-2 GFP/ΔN trVLP at an MOI = 0.05 for 72 hours. (F) TCID50 assay of SARS-CoV-2 GFP/ΔN trVLP in supernatant from Caco-2 cells reconstituted with WT N (N-WT) or N-C28840-Mut by lentiviral system, with infection by SARS-CoV-2 GFP/ΔN trVLP at an MOI = 0.05 for 72 hours. Data are representative of three independent experiments and were analyzed by two-tailed unpaired t test. Graphs show the means ± SD (n = 3) derived from three independent experiments. NS, not significant for P > 0.05, **P < 0.01, ***P < 0.001.

The SARS-CoV-2 GFP/ΔN trVLPs with low m5C modification showed a stronger replication ability

By using the trVLP system, we further explored the effect of the presence or absence of m5C modifications by NSUN2 on the replication ability of these trVLPs (Fig. 5A). Consistent with the previous results, knockout of NSUN2 elevated the replication of SARS-CoV-2 GFP/ΔN trVLP both in protein level and RNA level (Fig. 5, B and C). In addition, the median tissue culture infectious dose (TCID50) of SARS-CoV-2 GFP/ΔN trVLP amplified from NSUN2 knockout Caco-2 cells (low m5C) was higher than that from WT cells (normal m5C) (Fig. 5D). Further, we used the SARS-CoV-2 GFP/ΔN trVLP with NSUN2-mediated m5C methylation (normal m5C) and SARS-CoV-2 GFP/ΔN trVLP without NSUN2-mediated m5C methylation (low m5C) to infect Caco-2–N-mCherry cells and NSUN2 knockout Caco-2–N-mCherry cells at the same multiplicity of infection (MOI) (Fig. 5A). Strikingly, SARS-CoV-2 GFP/ΔN trVLP without NSUN2-mediated m5C methylation showed a stronger replication ability compared to SARS-CoV-2 GFP/ΔN trVLP without NSUN2-mediated m5C methylation, both in Caco-2–N-mCherry cells and NSUN2 knockout Caco-2–N-mCherry cells (Fig. 5, E to H). Caco-2–N-mCherry cells infected with SARS-CoV-2 GFP/ΔN trVLP with NSUN2-mediated m5C showed the lowest replication ability, while the NSUN2 knockout Caco-2–N-mCherry cells infected with SARS-CoV-2 GFP/ΔN trVLP without NSUN2-mediated m5C methylation showed the highest replication ability. Collectively, these results further demonstrated that NSUN2-mediated m5C modification negatively regulated SARS-CoV-2 replication, and SARS-CoV-2 GFP/ΔN trVLPs without NSUN2-mediated m5C showed a stronger replication ability.

Fig. 5. The SARS-CoV-2 GFP/ΔN trVLPs with low m5C modification showed a stronger replication ability.

Fig. 5.

Open in a new tab

(A) Schematic diagram of virus amplification and identification of SARS-CoV-2 GFP/ΔN trVLP from WT Caco-2 cells or NSUN2 knockout Caco-2 cells. (B) Immunofluorescence analysis of SARS-CoV-2 GFP/ΔN trVLP (green) in WT Caco-2 cells or NSUN2 knockout Caco-2 cells reconstituted with WT N (N-mCherry, red) by lentiviral system, with infection by SARS-CoV-2 GFP/ΔN trVLP at an MOI = 0.05 for 72 hours. Scale bar, 50 μm. (C) qPCR analysis of RNA levels of N, E, and S genes in WT Caco-2 cells or NSUN2 knockout Caco-2 cells reconstituted with WT N (N-mCherry, red) by lentiviral system, with infection by SARS-CoV-2 GFP/ΔN trVLP at an MOI = 0.05 for 72 hours. (D) TCID50 assay of SARS-CoV-2 GFP/ΔN trVLP in supernatant from WT Caco-2 cells or NSUN2 knockout Caco-2 cells reconstituted with WT N (N-mCherry, red) by lentiviral system, with infection by SARS-CoV-2 GFP/ΔN trVLP at an MOI = 0.05 for 72 hour. (E to H) SARS-CoV-2 GFP/ΔN trVLP with m5C or without m5C from Fig. 5D was used to infect WT Caco-2 cells or NSUN2 knockout Caco-2 cells reconstituted with WT N (N-mCherry, red) by lentiviral system, at the same MOI. Immunofluorescence analysis (E), qPCR analysis of RNA levels of E and S genes in cell, and TCID50 assay of SARS-CoV-2 GFP/ΔN trVLP in supernatant were detected 72 hours after infection. Data are representative of three independent experiments and were analyzed by two-tailed unpaired t test. Graphs show the means ± SD (n = 3) derived from three independent experiments. NS, not significant for P > 0.05, **P < 0.01, ***P < 0.001.

NSUN2 regulated SARS-CoV-2 replication in two mice models

To further investigate the role of NSUN2 in the regulation of SARS-CoV-2 infection in vivo, we created targeted deletions of NSUN2 in mice by removing 10 base pairs (bps) in exon 3 of Nsun2 genome by CRISPR-Cas9, which resulted in a frameshift mutation (50). However, the homozygous Nsun2−/− mice died in utero. Therefore, we chose Nsun2+/− mice as Nsun2 deficiency mice (fig. S11). Then, we used an adenoviral transduction–based mouse model, which can be infected with SARS-CoV-2 in C57BL/6J mice (Fig. 6A) (51). Nsun2+/+ and Nsun2+/− mice were anesthetized with isoflurane and intranasally transduced with Ad5-hACE2. Five days after transduction, the mice were then infected intranasally with SARS-CoV-2. The Nsun2+/− mice lost more weight than the Nsun2+/+ mice (Fig. 6B), and there was a higher viral burden of SARS-CoV-2 in lungs of Nsun2+/− mice than in WT mice at the RNA levels of N and E (Fig. 6C). Meanwhile, Nsun2+/− mice showed a higher viral burden of SARS-CoV-2 (N protein) in lungs and more severe lung tissue damages compared with their WT littermates (Fig. 6, D and E).

Fig. 6. Nsun2 deficiency mice exhibited higher viral burden and more severe lung tissue damages.

Fig. 6.

Open in a new tab

(A) Timeline of Ad5-hACE2–mediated SARS-CoV-2 WT strain infection mice model (Nsun2+/+ mice and Nsun2+/− mice). (B) Body weight changes of SARS-CoV-2 WT strain–infected Nsun2+/+ mice (n = 5) and Nsun2+/− mice (n = 5). (C) qPCR analysis of RNA levels of N and E genes in lungs from Ad5-hACE2–mediated SARS-CoV-2 WT strain–infected Nsun2+/+ mice (n = 5) or Nsun2+/− mice (n = 5). (D) The hematoxylin and eosin (H&E) stain of lungs of uninfected mice group and Ad5-hACE2–mediated SARS-CoV-2 WT strain–infected mice group (Nsun2+/+ mice and Nsun2+/− mice). (E) Immunofluorescence analysis of Ad5-hACE2–mediated SARS-CoV-2 WT strain–infected mice group (Nsun2+/+ mice and Nsun2+/− mice). SARS-CoV-2 N (green). ACE2 (red). 4′,6-diamidino-2-phenylindole (DAPI, blue). (F) Timeline of SARS-CoV-2 Beta variant (B.1.351) infection mice model (Ifnar−/−Nsun2+/+ mice and Ifnar−/−Nsun2+/− mice). (G) Body weight changes of SARS-CoV-2 Beta variant (B.1.351)–infected Ifnar−/−Nsun2+/+ mice (n = 5) and Ifnar−/−Nsun2+/− mice (n = 5). (H) qPCR analysis of RNA levels of N and E genes in lungs from SARS-CoV-2 Beta variant (B.1.351)–infected Ifnar−/−Nsun2+/+ mice (n = 5) and Ifnar−/−Nsun2+/− mice (n = 5). (I) The H&E stain of lungs of SARS-CoV-2 Beta variant (B.1.351)–infected Ifnar−/−Nsun2+/+ mice and Ifnar−/−Nsun2+/− mice. (J) Immunofluorescence analysis of lungs of SARS-CoV-2 Beta variant (B.1.351)–infected Ifnar−/−Nsun2+/+ mice and Ifnar−/−Nsun2+/− mice. SARS-CoV-2 N (red). DAPI (blue). **P < 0.01, ***P < 0.001.

To block the effect of interferon responses induced by NSUN2 deletion, we crossed the Nsun2+/− mice with Ifnar−/− and got the Ifnar−/−Nsun2+/+ mice and Ifnar−/−Nsun2+/− mice. Then, we used SARS-CoV-2 Beta variant (B.1.351) infection mice model to further verify the NSUN2-mediated regulation of SARS-CoV-2 infection in vivo (Fig. 6F). The Ifnar−/−Nsun2+/− mice lost up to about 10% of their weight in the first 3 to 4 days of infection, while the Ifnar−/−Nsun2+/+ mice did not show notable weight change (Fig. 6G). Accordingly, Ifnar−/−Nsun2+/− mice showed a higher viral burden of SARS-CoV-2 in lungs and more severe lung tissue damages compared with Ifnar−/−Nsun2+/+ mice both at the protein levels and RNA levels (Fig. 6, H to J). Together, the results of the two SARS-CoV-2–infected mice models suggest that NSUN2 was a negative regulator of SARS-CoV-2 and quite pivotal for the replication and pathogenicity of SARS-CoV-2 in vivo, which may serve as a strategy for the host to resist the replication of SARS-CoV-2.

NSUN2 expression was antagonized by SARS-CoV-2 infection

To investigate the biological role of NSUN2 during SARS-CoV-2 infection, we performed RNA sequencing (RNA-seq) to systematically analyze the expression levels of mRNAs in Caco-2 cells or Caco-2 cells with SARS-CoV-2 infection for 24 hours. NSUN2 mRNA levels were consistently decreased in SARS-CoV-2–infected Caco-2 cells compared to mock Caco-2 cells (Fig. 7, A and B). Using the SARS-CoV-2 Beta variant (B.1.351)–infected mouse model, we found that endogenous Nsun2 mRNA levels also decreased in SARS-CoV-2 Beta variant (B.1.351)–infected group [4 days post infection (dpi)] compared to the uninfected group (Mock) (Fig. 7C). However, there was no notable difference in the endogenous Nsun2 mRNA levels between the infected group (10 dpi) and the uninfected group. According to our results, the SARS-CoV-2 Beta variant (B.1.351)–infected mouse model showed mild symptoms, and, in the infected group (10 dpi), the mice regained weight and recovered with low viral loads in their lung tissues. These results suggested that when SARS-CoV-2 replicates to a higher level, it suppresses the expression level of NSUN2 through some unknown mechanism. In addition, we carried out transcriptome sequencing of the RNAs isolated from the bronchoalveolar lavage fluid (BALF) of two patients with COVID-19. NSUN2 mRNA levels were also reduced in patients with COVID-19 compared with healthy individuals (Fig. 7D) (52). From our above results, NSUN2 negatively regulated the replication of SARS-CoV-2 via epitranscriptomic addition of m5C modification to SARS-CoV-2 transcripts. This mechanism may be exploited by the host as an antiviral strategy. Therefore, SARS-CoV-2 will antagonize NSUN2 expression via unknown mechanism. We have systematically screened the proteins encoded by SARS-CoV-2 but did not find a specific regulatory factor (fig. S12A). Viral infections will induce the production of interferons and proinflammatory factors as a common feature. Therefore, we speculated whether interferons or proinflammatory factors decreased NSUN2 expression. Then, we screened interferon-β (IFN-β), tumor necrosis factor–α (TNF-α), and interleukin-1β (fig. S12B). However, NSUN2 mRNA levels did not show notable change with treatment of these factors. On the basis of the current data, we cannot give a clear mechanism at this time. We will further explore the specific mechanism of this degradation in follow-up work. The expression levels of the two m5C readers, ALYREF and YBX1, also decreased upon SARS-CoV-2 infection, although they were not involved in the regulation of SARS-CoV-2 (Fig. 2A). Still, this result may uncover that SARS-CoV-2 infection will have an impact on host m5C machinery and m5C methylome. However, the expression levels of m6A machinery genes did not show notable or consistent change compared with m5C machinery (Fig. 7, B and D). We further collected BALF from patients with severe and mild COVID-19 with the SARS-CoV-2 Omicron subvariant BA.5.2 and found that NSUN2 expression levels and m5C levels decreased in patients with severe compared to mild COVID-19 (Fig. 7, E and F). Unfortunately, two of the patients with severe COVID-19 passed away. We then collected the blood of these convalescent patients. The NSUN2 expression levels and m5C levels of these convalescent patients restored to the same level as in healthy individuals (Fig. 7, G and H), which is consistent with the above results of SARS-CoV-2 Beta variant (B.1.351)–infected mouse model (Fig. 7C). These results provide further evidence that m5C machinery is involved in the regulatory process of SARS-CoV-2 replication and NSUN2 expression was antagonized by SARS-CoV-2 infection via unknown mechanism.

Fig. 7. NSUN2 expression decreased during SARS-CoV-2 infection.

Fig. 7.

Open in a new tab

(A) RNA-seq signals for NSUN2 mRNA in uninfected Caco-2 cells (Mock) and SARS-CoV-2 (WT strain, 24 hours)–infected Caco-2 cells (Infected). The scale on the y axis indicates the read density per million of total normalized reads. (B) Heatmap showed the expression levels of m5C and m6A machinery genes in uninfected Caco-2 cells (Mock) and SARS-CoV-2–infected Caco-2 cells (Infected). (C) qPCR analysis of mRNA levels of Nsun2 in lungs from the uninfected C57BL/6J mice, the C57BL/6J mice infected with SARS-CoV-2 Beta variant (B.1.351) for 4 days with a high viral load, and the C57BL/6J mice infected with SARS-CoV-2 Beta variant (B.1.351) for 10 days with a low viral load. (D) Heatmap showed the expression levels of m5C and m6A machinery genes in BALF of patients with COVID-19 (Patient1 and Patient2) and healthy controls (Ctrl1, Ctrl2, and Ctrl3). Data were obtained from the analysis of our previous results (52). (E) qPCR analysis of RNA levels of NSUN2 in BALF from seriously ill patients (n = 5) and mildly ill patients (n = 5) with the SARS-CoV-2 Omicron subvariant BA.5.2. (F) Analysis of m5C/C ratio using LC-MS/MS of BALF from seriously ill (n = 5) and mildly ill patients with COVID-19 (n = 5). (G) qPCR analysis of RNA levels of NSUN2 in blood from convalescent patients with severe COVID-19 (n = 5) and convalescent patients with mild COVID-19 (n = 5) with the SARS-CoV-2 Omicron subvariant BA.5.2. (H) Analysis of m5C/C ratio using LC-MS/MS of blood from the convalescent patients with severe COVID-19 (n = 3), the convalescent patients with mild COVID-19 (n = 5), and the healthy individuals (n = 5). NS, not significant for P > 0.05, **P < 0.01, ***P < 0.001.

We further measured different time points to understand the kinetics of m5C modification mediated by NSUN2. NSUN2 expression levels gradually decreased upon SARS-CoV-2 infection, while the m5C levels of total RNAs gradually increased to the highest value and then decrease (fig. S13). Because NSUN2 is constitutively expressed in uninfected cells, in the early stage of viral infection, the level of NSUN2 is relatively abundant and enough to modify large amounts of the newborn SARS-CoV-2 RNAs. Therefore, the m5C levels of total RNAs will gradually increase because of the newborn m5C-modified SARS-CoV-2 RNAs. However, in the late stages of viral infection, because of the decrease of NSUN2 caused by SARS-CoV-2 infection, there are fewer NSUN2 proteins in the cell, and large amounts of progeny viruses containing m5C modifications are released outside the cell. Then, the m5C modification level of total RNAs will begin to decline.

The SARS-CoV-2 virions with low m5C modification from Nsun2+/− mice exhibited a stronger replication ability and caused more severe lung tissue damages in K18-hACE2 mice

To further verify the differences in the replication ability and virulence of SARS-CoV-2 virions with normal or low m5C modification in vivo, we used SARS-CoV-2 Beta variant (B.1.351) to infect Nsun2+/+ mice and Nsun2+/− mice, respectively, and gently dissociated the lungs to collect the progeny SARS-CoV-2 virions from Nsun2+/+ mice or Nsun2+/− mice (Fig. 8A). We quantified the progeny SARS-CoV-2 virions from Nsun2+/+ mice or Nsun2+/− mice, analyzed the m5C/C ratio using LC-MS/MS of the progeny SARS-CoV-2 virions, and then used them to infect Caco-2 cells and K18-hACE2 mice, respectively. The LC-MS/MS results showed that the m5C level of the progeny SARS-CoV-2 virions from Nsun2+/− mice reduced by about half (Fig. 8B). Consistent with the previous SARS-CoV-2 GFP/ΔN trVLP results in Fig. 5, the progeny SARS-CoV-2 virions from Nsun2+/− mice (with low m5C level) showed a stronger replication ability both in Caco-2 cells (Fig. 8C) and K18-hACE2 mice (Fig. 8D) than the progeny SARS-CoV-2 virions from Nsun2+/+ mice (with normal m5C level). Meanwhile, the progeny SARS-CoV-2 virions from Nsun2+/− mice (with low m5C level) caused more severe lung tissue damages in K18-hACE2 mice and a higher viral burden of SARS-CoV-2 (N protein) in lungs (Fig. 8, E and F).

Fig. 8. The progeny SARS-CoV-2 virions from Nsun2+/− mice with low m5C modification showed a stronger replication ability in Caco-2 cells and K18-hACE2 mice.

Fig. 8.

Open in a new tab

(A) Schematic diagram of virus amplification and identification of SARS-CoV-2 virions from Nsun2+/+ mice or Nsun2+/− mice. (B) Analysis of m5C/C ratio using LC-MS/MS of SARS-CoV-2 virion RNA from lungs of Beta variant (B.1.351)–infected Nsun2+/+ mice (n = 5) or Nsun2+/− mice (n = 5). (C) The purified SARS-CoV-2 virions from lungs of Beta variant (B.1.351)–infected Nsun2+/+ mice or Nsun2+/− mice were further used to infect Caco-2 cells at the same MOI for 24 and 48 hours, followed by qPCR analysis of RNA levels of E and N genes from the infected Caco-2 cells. (D to F) The purified SARS-CoV-2 virions from lungs of Beta variant (B.1.351)–infected Nsun2+/+ mice or Nsun2+/− mice were further used to infect K18-hACE2 mice (n = 5) at the same MOI for 24 and 48 hours, followed by qPCR analysis (D) of RNA levels of E and N genes, H&E stain of lungs (E), and immunofluorescence analysis of lungs (F) from the infected K18-hACE2 mice. SARS-CoV-2 N (red). DAPI (blue). **P < 0.01, ***P < 0.001.

Together, these results provide strong evidence that m5C modification is an important way to regulate SARS-CoV-2 replication and virulence, and the progeny SARS-CoV-2 virions with low m5C modification replicates more strongly inside the host in vivo. NSUN2-mediated m5C methylation of SARS-CoV-2 RNA negatively regulates viral replication and the virulence of progeny viruses in the new infection.

DISCUSSION

Inconsistent results of the m6A-mediated regulation of SARS-CoV-2 replication have been reported. Two groups reported that m6A RNA methylation negatively regulates the SARS-CoV-2 life cycle (36, 37), while three groups demonstrated that m6A positively regulates SARS-CoV-2 replication (38–40). We found that knockdown of m6A writers mildly increased SARS-CoV-2 replication (Fig. 2A). However, knockdown or knockout of m5C writers NSUN2 notably increased SARS-CoV-2 replication. Moreover, according to our results of LC-MS/MS, m5C is more abundant than m6A in SARS-CoV-2 virion RNA (Fig. 1D and fig. S1). Together, the above results demonstrate that m5C RNA methylation showed more notable regulation of SARS-CoV-2 than m6A.

From the results of RNA-Bis-seq and m5C-MeRIP-seq (Figs. 1E and 2I and fig. S5), the m5C methylation sites were distributed across the entire gRNA, with no particularly hypermethylated m5C methylation sites. The Input signal and m5C-IP signal near the N region are very high because of the highest expression level of N sgRNAs, but the ratios of m5C-IP/Input in N region or other m5C-modified regions are close. The N gene contains NSUN2-dependent m5C sites, which play notable roles in regulating the function of N expression and the replication of SARS-CoV-2. Meanwhile, the other m5C sites were also identified in other regions, such as nsp14, nsp15, nsp16, E, M, and S. The m5C incorporation experiment suggested that m5C modification had a notable effect on the degradation of viral RNA fragments as a whole effect (Fig. 3C and fig. S8). The identified m5C methylation sites among nsp14, nsp15, nsp16, S, E, M, and N were proved to be functional in regulating SARS-CoV-2 RNA stability, and the mutations of these cytosines enhance RNA stability. However, we found that the mutation of the three high methylation sites in ORF3a decreased RNA expression levels, unlike nsp14, nsp15, nsp16, S, E, M, and N. We found that the methylation sites on ORF3a had different regulatory mechanisms from other regions. We are conducting further research to investigate this phenomenon.

Our previous work revealed that NSUN2 serves as a negative regulator of type I interferon responses in antiviral innate immunity during various viral infections (50). Knockout of NSUN2 can enhance type I interferon responses and downstream interferon-stimulated genes (ISGs) expression after some viral infections both in vitro and in vivo, thus inhibiting the replication of vesicular stomatitis virus (VSV), herpes simplex virus 1 (HSV-1), or Sendai virus (SeV). However, the interferon responses caused by SARS-CoV-2 infection is not so strong as VSV, HSV-1, and SeV. SARS-CoV-2 invests substantial resources to block the establishment of the antiviral response (53). Therefore, the enhanced interferon responses by NSUN2 knockout were not so strong in SARS-CoV-2–infected models. Of note, in this work, the regulation of NSUN2 on SARS-CoV-2 is independent of interferon responses but by directly affecting the stability of viral RNAs. This suggests that the direct effect of NSUN2 on SARS-CoV-2 is much greater than the effect on interferon responses during SARS-CoV-2 infection, unlike some other viruses.

The currently known readers of RNA m5C modification are not involved in regulating the replication of SARS-CoV-2 (Fig. 2A). The reason may be the irrelevance of their functions. ALYREF was reported to promote the export of m5C-modified mRNAs (47). However, SARS-CoV-2 replicates only in the cytoplasm. YBX1 maintains the stability of its target mRNA with m5C modification (48), but our results demonstrated that NSUN2 mediated the methylation of SARS-CoV-2 RNA transcripts and facilitated their degradation. These results imply the involvement of unidentified m5C reader proteins that may contribute to the regulation of SARS-CoV-2 replication. The specific degradation mechanism induced by m5C modification has not yet been clarified clearly and requires further investigation. Another study reported that m5C modification of Epstein-Barr virus RNA decreases its stability (46), which may also support the host antiviral strategy we propose via epitranscriptomic addition of m5C methylation to viral RNAs. Further work is required to delineate these different mechanisms and roles that different m5C readers play in the regulation of various viruses.

Since SARS-CoV-2 itself does not encode the NSUN2 protein, the original SARS-CoV-2 RNA may not contain so many m5C modifications. According to our results (Fig. 5), SARS-CoV-2 GFP/ΔN trVLP without NSUN2-mediated m5C methylation showed a stronger replication ability compared to SARS-CoV-2 GFP/ΔN trVLP without NSUN2-mediated m5C methylation. We propose a SARS-CoV-2–host interaction model. After SARS-CoV-2 infects the host, the gRNA and sgRNAs of SARS-CoV-2 will be modified with m5C modification gradually by the host methyltransferase NSUN2. The m5C modification can be recognized by the host as a “marker” and degraded by host factors, thus serving as a host antiviral strategy. This marker is quickly recognized and degraded when the progeny viruses with m5C modification enters another cell in the new round of infection. From our results (Fig. 7, D to F), the NSUN2 expression level in patients with severe COVID-19 was lower than that in patients with mild COVID-19. Correspondingly, m5C modification on viral and host RNA was reduced in patients with severe COVID-19. When the patients recovered, the NSUN2 expression level in patients with COVID-19 returned to normal as in healthy people (Fig. 7, G and H). Because it is not possible to conduct experiments on reinfection of SARS-CoV-2 (with low or normal m5C modification) in humans, we used a mouse model to simulate patients with severe and mild COVID-19 and revealed that progeny SARS-CoV-2 virions from Nsun2+/− mice (with low m5C level, simulating patients with severe COVID-19) showed a stronger replication ability and virulence in the new round of infection in a K18-hACE2 mouse model (Fig. 8) than the progeny SARS-CoV-2 virions from Nsun2+/+ mice (with normal m5C level, simulating patients with mild COVID-19 or healthy individuals). The above results may imply that the progeny SARS-CoV-2 viruses transmitted by patients with severe COVID-19 have a stronger replication ability and virulence in the new round of infection when spreading from person to person due to the lack of m5C RNA modification (Fig. 9). The mechanism of NSUN2 expression decline caused by SARS-CoV-2 remains to be further investigated, and we are conducting follow-up investigations.

Fig. 9. Schematic diagram of NSUN2-mediated m5C modification regulates SARS-CoV-2 replication and the virulence of progeny viruses in the new infection.

Fig. 9.

Open in a new tab

We used a mouse model to simulate patients with severe and mild COVID-19 and revealed that progeny SARS-CoV-2 virions from Nsun2+/− mice (with low m5C level, simulating patients with severe COVID-19) showed a stronger replication ability and virulence in the new round of infection in a K18-hACE2 mouse model than the progeny SARS-CoV-2 virions from Nsun2+/+ mice (with normal m5C level, simulating patients with mild COVID-19 or healthy individuals).

In conclusion, our investigation has revealed a role for NSUN2-mediated m5C modification in regulating SARS-CoV-2 replication and virulence. We outline a paradigm that the host has evolved an antiviral strategy via epitranscriptomic addition of m5C methylation to SARS-CoV-2 RNAs, which will help in the development of anti-coronavirus drugs or efficient therapeutic interventions.

MATERIALS AND METHODS

Viruses, cells, and reagents

The SARS-CoV-2 WT strain (IVCAS 6.7512) and Beta variant (B.1.351) (NPRC2.062100001) were provided by the National Virus Resource, Wuhan Institute of Virology, Chinese Academy of Sciences. Caco-2, Vero E6, HEK293T, and Huh7 cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS), penicillin (100 U/ml), and streptomycin (100 μg/ml) at 37°C in 5% CO2 incubator. Plasmids were transfected using Lipofectamine 3000 (Invitrogen) or Neofect DNA transfection reagent (Neofect, TF201201) following the manufacturer’s instructions, and siRNAs (RiboBio, Guangzhou, China) were transfected using RNAiMAX (Invitrogen) following the manufacturer’s instructions. Ruxolitinib and actinomycin D were purchased from MCE (MedChemExpress).

Mice

Nsun2+/− C57BL/6J mice were obtained from Gempharmatech Co. Ltd. (Nanjing, China) and housed and bred in specific pathogen–free conditions. The primers used for genotyping of Nsun2 mutant mice were F-TGTCCAACAGAACAGTGAACTGGAG and R-CCAAGCTCTTTAAGCCGACAGTG. Ifnar−/− mice were a gift from B. Zhong (Wuhan University). The primers used for genotyping of Ifnar−/− mice were F-CGAGGCGAAGTGGTTAAAAG, R1-ACGGATCAACCTCATTCCAC (WT reverse primer), and R2-AATTCGCCAATGACAAGACG (mutant reverse primer). Nsun2+/− C57BL/6J mice were crossed with Ifnar−/− mice to obtain Ifnar−/−Nsun2+/+ mice and Ifnar−/−Nsun2+/− mice. All animal experiments were conducted in accordance with the Regulations of Hubei Province Laboratory Animal Management and approved by Wuhan University Animal Experiment Ethics Committee.

SARS-CoV-2 live virus infection assay

All SARS-CoV-2 live virus–related experiments were approved by the level 3 Biosafety Committee (ABSL-3) of Wuhan University. All experiments involving SARS-CoV-2 were performed in the BSL-3 and ABSL-3 facilities of Wuhan University.

SARS-CoV-2 infection mouse models

The establishment of Ad5-hACE2 mouse model was performed as described previously (51). Briefly, 8-week-old Nsun2+/+ C57BL/6J mice and Nsun2+/− C57BL/6J mice were lightly anesthetized with isoflurane and transduced intranasally with 2.5 × 108 focus-forming units of Ad5-hACE2 in 75 μl of DMEM. Five days after transduction, the mice were infected intranasally with 1 × 105 plaque-forming units (PFU) of SARS-CoV-2 WT strain (IVCAS 6.7512) in 50 μl of DMEM. The mice were monitored and weighted daily. Seven days after infection, the mice were euthanized and lung tissues were taken.

For the establishment of SARS-CoV-2 Beta variant (B.1.351) (NPRC2.062100001) infection mouse model, 8-week-old Ifnar−/−Nsun2+/+ mice and Ifnar−/−Nsun2+/− mice were lightly anesthetized with isoflurane and intranasally with 1.25 × 105 PFU of SARS-CoV-2 Beta variant (B.1.351) in 50 μl of DMEM. The mice were monitored and weighted daily. Four days after infection, the mice were euthanized and lung tissues were taken.

Histopathological analysis and immunofluorescence assay for mice tissues

Histopathological analysis and immunofluorescence assay for mice tissues were performed by Wuhan Servicebio Technology Co. Ltd. For histopathological analysis, the mice tissues were immediately fixed in 4% paraformaldehyde (PFA), embedded in paraffin, sectioned, and stained with hematoxylin and eosin (H&E). For immunofluorescence assay, the sections were blocked with phosphate-buffered saline (PBS) containing 3% bovine serum albumin (BSA) at room temperature for 30 min and then incubated with primary antibody (diluted with PBS appropriately) overnight at 4°C. After being washed three times with PBS, the slides were incubated with fluorescent-labeled secondary antibody (appropriately respond to primary antibody in species) at room temperature for 50 min in dark condition. After being washed for another three times with PBS, the slides were incubated with 4′,6-diamidino-2-phenylindole (DAPI) solution at room temperature for 10 min in dark condition.

Antibodies and immunoblot analysis

The antibodies used were as follows: rabbit anti-NSUN2 (Proteintech, 20854-1-AP), rabbit anti-METTL3 (Proteintech, 15073-1-AP), mouse anti–SARS-CoV-2 N (Sino Biological, 40143-MM05), rabbit anti-human ACE2 (Sino Biological, 10108-RP01), and mouse anti–glyceraldehyde-3-phosphate dehydrogenase (GAPDH, Proteintech, 60004-1-Ig). The cells were washed once with PBS and lysed in radioimmunoprecipitation assay or NP-40 lysis buffer (Beyotime). SDS loading buffer (5×) was added to the protein sample and boiled for 10 min. The samples were resolved on SDS–polyacrylamide gel electrophoresis and transferred onto nitrocellulose membrane (GE Healthcare), followed by blocking with tris-buffered saline containing 0.1% Tween-20 (TBST) and 5% nonfat powdered milk or BSA and probing with different antibodies.

Immunofluorescence assay

Caco-2 or Vero E6 cells were seeded in 48-well plates and infected with SARS-CoV-2 for 24 hours. The cells were inactivated and fixed with 4% PFA at room temperature for 30 min, permeabilized with 0.2% Triton X-100 at room temperature for 10 min, and blocked with 1% BSA at 37°C for 30 min. Then, the cells were incubated with SARS-CoV-2 N antibody (40143-MM05, Sino Biological) diluted in 1% BSA at 37°C for 1 hour, followed by PBS washing. Next, the cells were incubated with Alexa Fluor 594–conjugated goat anti-mouse immunoglobulin G (Thermo Fisher Scientific, A11032) diluted in 1% BSA at room temperature for 30 min, followed by PBS washing. The nucleus was stained with Hoechst 33342 for 5 min. The images were captured with an inverted fluorescence microscope.

RNA isolation and qPCR

Total RNA was isolated using TRIzol reagent (Invitrogen) following the manufacturer’s instructions. The isolated RNA was reverse transcribed to cDNA using PrimeScript RT Reagent Kit (Takara, RR037A) or NovoScript Plus All-in-one first-strand cDNA Synthesis SuperMix (Novoprotein). Real-time qPCR was carried out through ABI 7500 Real Time PCR System by SYBR Green Master Mix (YEASEN, 11199ES03) or Taqman Probe Master Mix (YEASEN, 11205ES08). GAPDH was used in normalization via the ΔΔCt method. Primer sequences are shown in table S1.

Liquid chromatography tandem MS

LC-MS/MS was performed by Wuhan Metware Biotechnology Co. Ltd. Briefly, the digestion mixture contains 1 μg of RNA, 1 U of nuclease P1, 10 mM NaCl, and 2 mM ZnCl2 in a final volume of 30 μl. The mixture was incubated at 37°C for 3 hours. Then, 1 U of shrimp alkaline phosphatase and 2.5 μl of ammonium bicarbonate (1 M) were added into the mixture for another 2 hours and diluted to 100 μl. Three microliters of the mixture was injected into the LC-MS/MS. The nucleosides were separated by a C18 column and detected by triple-quadrupole MS (Shimadzu MS-8050 mass spectrometer, Tokyo, Japan). RNA modifications contents were detected by Wuhan Metware Biotechnology Co. Ltd. (http://metware.cn/) based on the AB Sciex QTRAP 6500 LC-MS/MS platform.

Construction of knockout cell line by CRISPR-Cas9

The gRNAs were NSUN2-gRNA-1: F-CACCGACGCGGAGGATGGCGCCGA and R-AAACTCGGCGCCATCCTCCGCGTC; NSUN2-gRNA-2: F-CACCACCGTG GCGTTTCAGCGGTT and R-AAACAACCGCTGAAACGCCACGGT. The gRNAs were constructed in lentiCRISPR-v2 plasmid (Addgene). Lentiviral backbone (1 μg), psPAX2 (0.5 μg), and pMD2.G (0.5 μg) were transfected into HEK293T cells, which were seeded on six-well plates using Neofect. Forty-eight hours later, supernatants were collected and filtered using a 0.45-μm filter to infect target cells with polybrene (8 μg/ml). The cells were infected twice to get higher transduction efficiency. Then, puromycin was used to screen positive cells, followed by seeding into 96-well plates (1 cell per well). After 2 weeks’ cultivation, single clones were selected following enlarged cultivation with puromycin selection. Single clones were identified by immunoblot analysis, and genomic DNA was extracted followed by PCR and sequencing.

In vitro transcription assays

RNA was in vitro transcribed with CTP or m5CTP (Syngenebio, Nanjing, China) as substrates using a MEGAscript T7 Kit (Ambion, USA) according to the manufacturer’s instructions. The transcribed RNAs were then capped with the Vaccinia Capping Enzyme (NEB, M2080) and mRNA Cap 2′-O-methyltransferase (NEB, M0366) in the presence of guanosine triphosphate and SAM if the RNAs were used for transfection.

In vitro methylation assays

Reaction mixtures (50 μl) containing 0.2 nM recombinant GST–tagged NSUN2, 0.01 nM in vitro–transcribed fragments of mRNA, 1 μCi of S-adenosyl [methyl-3H] methionine (0.5 μCi/μl, PerkinElmer) in reaction buffer [500 mM tris-HCl (pH 7.5), 5 mM EDTA, 10 mM dithiothreitol, 20 mM MgCl2], and 40 U of ribonuclease inhibitor were incubated for 60 min at 37°C. The 3H-labeled products were isolated using DEAE-Sephadex A-50 columns and quantitated by liquid scintillation counting (PerkinElmer).

MeRIP-seq

The m5C-MeRIP-seq was provided by Cloudseq Biotech Inc. (Shanghai, China). Briefly, m5C RNA immunoprecipitation was performed with the GenSeq m5C RNA IP Kit (GenSeq, China) by following the manufacturer’s instructions. Both the input samples without immunoprecipitation and the m5C IP samples were used for RNA-seq library construction with NEBNext Ultra II Directional RNA Library Prep Kit (NEB, USA). The library quality was evaluated with BioAnalyzer 2100 system (Agilent Technologies, USA). Library sequencing was performed on an illumina Novaseq 6000 instrument with 150-bp paired-end reads. The m5C-MeRIP-seq data have been deposited in the GSA database under the accession number: HRA002087.

RNA-Bis-seq

The RNA-Bis-seq was provided by Cloudseq Biotech Inc. (Shanghai, China). Briefly, total RNAs for each sample was ribosomal RNA (rRNA)–depleted using GenSeq rRNA Removal Kit (GenSeq Inc.). rRNA-depleted RNA was bisulfite converted and purified using the EZ RNA methylation Kit (Zymo Research). RNA libraries were then constructed with GenSeq Low Input RNA Library Prep Kit (GenSeq Inc.) according to the manufacturer’s instructions. The library was quality-controlled with BioAnalyzer 2100 system (Agilent Technologies Inc.) and then sequenced in an Illumina Novaseq 6000 instrument. Analysis of the spike-in showed C to T conversion rates >99%. The RNA-Bis-seq data have been deposited in the GSA database under the accession number (HRA006230).

RNA sequencing

Total RNAs were extracted from indicated cells using TRIzol reagent. DNA digestion was carried out after RNA extraction by deoxyribonuclease I. RNA quality was determined by examining A260/A280 with Nanodrop. RNA Integrity was confirmed by 1.5% agarose gel electrophoresis. Qualified RNAs were finally quantified by Qubit 3.0 with QubitTM RNA Broad Range Assay kit (Life Technologies, Q10210). Two micrograms of total RNAs were used for stranded RNA-seq library preparation using KCTM Stranded mRNA Library Prep Kit (Wuhan Seqhealth Co. Ltd. China, DR08402) for Illumina following the manufacturer’s instruction. PCR products corresponding to 200 to 500 bp were enriched, quantified, and finally sequenced on Novaseq 6000 sequencer (Illumina) with PE150 model. The RNA-seq data have been deposited in the GSA database under the accession number (HRA006231).

Preparation of bone marrow–derived dendritic cells and bone marrow–derived macrophages

Bone marrow cells were isolated from C57BL/6J mouse tibia and femur and then cultured for 7 to 9 days in 10% FBS DMEM containing mouse granulocyte-macrophage colony-stimulating factor (GM-CSF, 50 ng/ml, Peprotech) for bone marrow–derived dendritic cells or M-CSF (50 ng/ml, Peprotech) for bone marrow–derived macrophages. IFN-β, TNF-α, and IL-1β were from MCE (MedChemExpress) and used at a final concentration of 20 ng/ml.

Constructions

NSUN2 was constructed into the pCAGGS and pGEX6P-1 vector, respectively. SARS-CoV-2 N gene was constructed into the pCAG and pLVX vector, respectively. The other SARS-CoV-2 genes were constructed into the pCAG vector and pCDNA3.1 vector, respectively. The full-length genome of SARS-CoV-2 was divided into seven fragments and inserted respectively into a pUC57 vector. The gRNAs of NSUN2 were constructed into the lentiCRISPR-v2 vector, respectively.

ER-DMVs fractionation

The ER-DMVs purification method was based on a previous article (54) and performed using the Endoplasmic Reticulum Isolation Kit (Sigma-Aldrich, ER0100) according to the manufacturer’s instructions.

Construction of the trVLP

The construction method is based on the articles (55, 56). In brief, the full-length genome of SARS-CoV-2 was divided into seven fragments and inserted respectively into a pUC57 vector, with a replacement of SARS-CoV-2 N gene by GFP. The seven fragments were amplified using Phanta Super-Fidelity DNA Polymerase (Vazyme, P501-d1). The obtained PCR product was digested by type IIS restriction endonuclease BsaI or BsmBI and then ligated with T4 ligase (NEB) to obtain a full-length cDNA template. The full-length template and N gene were in vitro transcribed and purified, respectively. Then, 20 μg of full-length mRNA and 10 μg of N mRNA were both added into a 4-mm cuvette (Bio-Rad) containing 0.4 ml of Caco-2–N cells (8 × 106) in Ingenio Electroporation Solution (Mirus). A single electrical pulse was applied using a GenePulser apparatus (Bio-Rad) set at 270 V, 950 μF. The sample was placed at room temperature for 5 min and then cultured in a 10-cm cell culture dish for 3 days. Then, the supernatant was collected. This trVLP expresses a reporter gene (GFP) replacing SARS-CoV-2 N gene (SARS-CoV-2 GFP/ΔN trVLP). The complete viral life cycle can be achieved and confined in the cells ectopically expressing SARS-CoV-2 N proteins.

Nuclear and cytoplasmic protein extraction

Nuclear and cytoplasmic extracts were obtained using Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime, P0027) according to the manufacturer’s instructions.

Ethics statement

This study was approved by the Ethics Committee of the Renmin Hospital of Wuhan University. The analyses of BALF samples and blood samples were performed on existing samples collected during standard diagnostic tests, posing no extra burden to patients. The sample collection was applied by Renmin Hospital of Wuhan University with written consent under appropriate institutional review board approval and was deidentified.

Acknowledgments

We thank J. Zhao for providing Ad5-hACE2. We thank P. Wang for providing SARS-CoV-2–related expression plasmids.

Funding: This study was supported by grants from China NSFC projects (32188101), the National Key R&D Program of China (2021YFA1300800 and 2022YFC2604100), China NSFC projects (82341061 and 82172243), the Fundamental Research Funds for the Central Universities (2042022dx0003 and 2042023kf1028), China Postdoctoral Science Foundation (2023 M732707), and Hubei Province Postdoctoral Innovation Research Post.

Author contributions: Y.C. and H.W. conceived the research and experiments. H.W., J.F., Z.F., T.X., J.L., and S.Y. performed the major experiments and data analysis. Y.L. and J.D. assisted with SARS-CoV-2 GFP/ΔN trVLPs experiments. Y.Z. assisted with the ER-DMVs purification experiments. M.G., X.W., Z.Z., and Z.H. participated in cell and mice experiments of live SARS-CoV-2. K.L. and L.Z. provided critical advice. H.W. and Y.C. wrote and revised the manuscript with contributions from all the other authors.

Competing interests: The authors declare that they have no competing interests.

Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials.

Supplementary Materials

This PDF file includes:

Figs. S1 to S13

Tables S1 and S2

sciadv.adn9519_sm.pdf (3.1MB, pdf)

REFERENCES AND NOTES

  • 1.Zhou P., Yang X. L., Wang X. G., Hu B., Zhang L., Zhang W., Si H. R., Zhu Y., Li B., Huang C. L., Chen H. D., Chen J., Luo Y., Guo H., Jiang R. D., Liu M. Q., Chen Y., Shen X. R., Wang X., Zheng X. S., Zhao K., Chen Q. J., Deng F., Liu L. L., Yan B., Zhan F. X., Wang Y. Y., Xiao G. F., Shi Z. L., A pneumonia outbreak associated with a new coronavirus of probable bat origin. Nature 579, 270–273 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Chen L., Liu W., Zhang Q., Xu K., Ye G., Wu W., Sun Z., Liu F., Wu K., Zhong B., Mei Y., Zhang W., Chen Y., Li Y., Shi M., Lan K., Liu Y., RNA based mNGS approach identifies a novel human coronavirus from two individual pneumonia cases in 2019 Wuhan outbreak. Emerg. Microbes Infect. 9, 313–319 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Chen Y., Liu Q., Zhou L., Zhou Y., Yan H., Lan K., Emerging SARS-CoV-2 variants: Why, how, and what’s next? Cell Insight 1, 100029 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Gordon D. E., Jang G. M., Bouhaddou M., Xu J., Obernier K., White K. M., O’Meara M. J., Rezelj V. V., Guo J. Z., Swaney D. L., Tummino T. A., Hüttenhain R., Kaake R. M., Richards A. L., Tutuncuoglu B., Foussard H., Batra J., Haas K., Modak M., Kim M., Haas P., Polacco B. J., Braberg H., Fabius J. M., Eckhardt M., Soucheray M., Bennett M. J., Cakir M., McGregor M. J., Li Q., Meyer B., Roesch F., Vallet T., Mac Kain A., Miorin L., Moreno E., Naing Z. Z. C., Zhou Y., Peng S., Shi Y., Zhang Z., Shen W., Kirby I. T., Melnyk J. E., Chorba J. S., Lou K., Dai S. A., Barrio-Hernandez I., Memon D., Hernandez-Armenta C., Lyu J., Mathy C. J. P., Perica T., Pilla K. B., Ganesan S. J., Saltzberg D. J., Rakesh R., Liu X., Rosenthal S. B., Calviello L., Venkataramanan S., Liboy-Lugo J., Lin Y., Huang X. P., Liu Y. F., Wankowicz S. A., Bohn M., Safari M., Ugur F. S., Koh C., Savar N. S., Tran Q. D., Shengjuler D., Fletcher S. J., O’Neal M. C., Cai Y., Chang J. C. J., Broadhurst D. J., Klippsten S., Sharp P. P., Wenzell N. A., Kuzuoglu-Ozturk D., Wang H. Y., Trenker R., Young J. M., Cavero D. A., Hiatt J., Roth T. L., Rathore U., Subramanian A., Noack J., Hubert M., Stroud R. M., Frankel A. D., Rosenberg O. S., Verba K. A., Agard D. A., Ott M., Emerman M., Jura N., von Zastrow M., Verdin E., Ashworth A., Schwartz O., d’Enfert C., Mukherjee S., Jacobson M., Malik H. S., Fujimori D. G., Ideker T., Craik C. S., Floor S. N., Fraser J. S., Gross J. D., Sali A., Roth B. L., Ruggero D., Taunton J., Kortemme T., Beltrao P., Vignuzzi M., García-Sastre A., Shokat K. M., Shoichet B. K., Krogan N. J., A SARS-CoV-2 protein interaction map reveals targets for drug repurposing. Nature 583, 459–468 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Perlman S., Netland J., Coronaviruses post-SARS: Update on replication and pathogenesis. Nat. Rev. Microbiol. 7, 439–450 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Chen Y., Liu Q., Guo D., Emerging coronaviruses: Genome structure, replication, and pathogenesis. J. Med. Virol. 92, 2249 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Kim D., Lee J.-Y., Yang J.-S., Kim J. W., Kim V. N., Chang H., The architecture of SARS-CoV-2 transcriptome. Cell 181, 914–921.e10 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Wang D., Jiang A., Feng J., Li G., Guo D., Sajid M., Wu K., Zhang Q., Ponty Y., Will S., Liu F., Yu X., Li S., Liu Q., Yang X.-L., Guo M., Li X., Chen M., Shi Z.-L., Lan K., Chen Y., Zhou Y., The SARS-CoV-2 subgenome landscape and its novel regulatory features. Mol. Cell 81, 2135–2147.e5 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Frye M., Harada B. T., Behm M., He C., RNA modifications modulate gene expression during development. Science 361, 1346–1349 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Pan Y., Ma P., Liu Y., Li W., Shu Y., Multiple functions of m6A RNA methylation in cancer. J. Hematol. Oncol. 11, 48 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Sun H., Li K., Liu C., Yi C., Regulation and functions of non-m6A mRNA modifications. Nat. Rev. Mol. Cell Biol. 24, 714–731 (2023). [DOI] [PubMed] [Google Scholar]
  • 12.Yang Y., Hsu P. J., Chen Y. S., Yang Y. G., Dynamic transcriptomic m6A decoration: Writers, erasers, readers and functions in RNA metabolism. Cell Res. 28, 616–624 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Squires J. E., Patel H. R., Nousch M., Sibbritt T., Humphreys D. T., Parker B. J., Suter C. M., Preiss T., Widespread occurrence of 5-methylcytosine in human coding and non-coding RNA. Nucleic Acids Res. 40, 5023–5033 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Shi H., Wei J., He C., Where, when, and how: Context-dependent functions of RNA methylation writers, readers, and erasers. Mol. Cell 74, 640–650 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Tirumuru N., Zhao B. S., Lu W., Lu Z., He C., Wu L., N(6)-methyladenosine of HIV-1 RNA regulates viral infection and HIV-1 Gag protein expression. eLife 5, (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Kennedy E. M., Bogerd H. P., Kornepati A. V. R., Kang D., Ghoshal D., Marshall J. B., Poling B. C., Tsai K., Gokhale N. S., Horner S. M., Cullen B. R., Posttranscriptional m6A editing of HIV-1 mRNAs enhances viral gene expression. Cell Host Microbe 19, 675–685 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Lichinchi G., Gao S., Saletore Y., Gonzalez G. M., Bansal V., Wang Y., Mason C. E., Rana T. M., Dynamics of the human and viral m6A RNA methylomes during HIV-1 infection of T cells. Nat. Microbiol. 1, 16011 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.McIntyre W., Netzband R., Bonenfant G., Biegel J. M., Miller C., Fuchs G., Henderson E., Arra M., Canki M., Fabris D., Pager C. T., Positive-sense RNA viruses reveal the complexity and dynamics of the cellular and viral epitranscriptomes during infection. Nucleic Acids Res. 46, 5776–5791 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Chen S., Kumar S., Espada C. E., Tirumuru N., Cahill M. P., Hu L., He C., Wu L., N6-methyladenosine modification of HIV-1 RNA suppresses type-I interferon induction in differentiated monocytic cells and primary macrophages. PLOS Pathog. 17, e1009421 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Imam H., Khan M., Gokhale N. S., McIntyre A. B. R., Kim G. W., Jang J. Y., Kim S. J., Mason C. E., Horner S. M., Siddiqui A., N6-methyladenosine modification of hepatitis B virus RNA differentially regulates the viral life cycle. Proc. Natl. Acad. Sci. U.S.A. 115, 8829–8834 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Lichinchi G., Zhao B. S., Wu Y., Lu Z., Qin Y., He C., Rana T. M., Dynamics of human and viral RNA methylation during Zika virus infection. Cell Host Microbe 20, 666–673 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Gokhale N. S., McIntyre A. B. R., McFadden M. J., Roder A. E., Kennedy E. M., Gandara J. A., Hopcraft S. E., Quicke K. M., Vazquez C., Willer J., Ilkayeva O. R., Law B. A., Holley C. L., Garcia-Blanco M. A., Evans M. J., Suthar M. S., Bradrick S. S., Mason C. E., Horner S. M., N6-methyladenosine in flaviviridae viral RNA genomes regulates infection. Cell Host Microbe 20, 654–665 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Lu M., Zhang Z., Xue M., Zhao B. S., Harder O., Li A., Liang X., Gao T. Z., Xu Y., Zhou J., Feng Z., Niewiesk S., Peeples M. E., He C., Li J., N(6)-methyladenosine modification enables viral RNA to escape recognition by RNA sensor RIG-I. Nat. Microbiol. 5, 584–598 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Ye F., Chen E. R., Nilsen T. W., Kaposi’s sarcoma-associated herpesvirus utilizes and manipulates RNA N(6)-adenosine methylation to promote lytic replication. J. Virol. 91, e00466-17 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Tan B., Liu H., Zhang S., da Silva S. R., Zhang L., Meng J., Cui X., Yuan H., Sorel O., Zhang S. W., Huang Y., Gao S. J., Viral and cellular N6-methyladenosine and N6,2′-O-dimethyladenosine epitranscriptomes in the KSHV life cycle. Nat. Microbiol. 3, 108–120 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Hesser C. R., Karijolich J., Dominissini D., He C., Glaunsinger B. A., N6-methyladenosine modification and the YTHDF2 reader protein play cell type specific roles in lytic viral gene expression during Kaposi’s sarcoma-associated herpesvirus infection. PLOS Pathog. 14, e1006995 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Hao H., Hao S., Chen H., Chen Z., Zhang Y., Wang J., Wang H., Zhang B., Qiu J., Deng F., Guan W., N6-methyladenosine modification and METTL3 modulate enterovirus 71 replication. Nucleic Acids Res. 47, 362–374 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Courtney D. G., Kennedy E. M., Dumm R. E., Bogerd H. P., Tsai K., Heaton N. S., Cullen B. R., Epitranscriptomic enhancement of influenza A virus gene expression and replication. Cell Host Microbe 22, 377–386.e5 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Zhu Y., Wang R., Zou J., Tian S., Yu L., Zhou Y., Ran Y., Jin M., Chen H., Zhou H., N6-methyladenosine reader protein YTHDC1 regulates influenza A virus NS segment splicing and replication. PLOS Pathog. 19, e1011305 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Xue M., Zhao B. S., Zhang Z., Lu M., Harder O., Chen P., Lu Z., Li A., Ma Y., Xu Y., Liang X., Zhou J., Niewiesk S., Peeples M. E., He C., Li J., Viral N6-methyladenosine upregulates replication and pathogenesis of human respiratory syncytial virus. Nat. Commun. 10, 4595 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Xue M., Zhang Y., Wang H., Kairis E. L., Lu M., Ahmad S., Attia Z., Harder O., Zhang Z., Wei J., Chen P., Gao Y., Peeples M. E., Sharma A., Boyaka P., He C., Hur S., Niewiesk S., Li J., Viral RNA N6-methyladenosine modification modulates both innate and adaptive immune responses of human respiratory syncytial virus. PLOS Pathog. 17, e1010142 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Chelmicki T., Roger E., Teissandier A., Dura M., Bonneville L., Rucli S., Dossin F., Fouassier C., Lameiras S., Bourc’his D., m6A RNA methylation regulates the fate of endogenous retroviruses. Nature 591, 312–316 (2021). [DOI] [PubMed] [Google Scholar]
  • 33.Tsai K., Courtney D. G., Cullen B. R., Addition of m6A to SV40 late mRNAs enhances viral structural gene expression and replication. PLOS Pathog. 14, e1006919 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Phillips S., Mishra T., Khadka S., Bohan D., Espada C. E., Maury W., Wu L., Epitranscriptomic N6-methyladenosine profile of SARS-CoV-2-infected human lung epithelial cells. Microbiol. Spectr. 11, e0394322 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Vaid R., Mendez A., Thombare K., Burgos-Panadero R., Robinot R., Fonseca B. F., Gandasi N. R., Ringlander J., Hassan Baig M., Dong J. J., Cho J. Y., Reinius B., Chakrabarti L. A., Nystrom K., Mondal T., Global loss of cellular m6A RNA methylation following infection with different SARS-CoV-2 variants. Genome Res. 33, 299–313 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Liu J., Xu Y.-P., Li K., Ye Q., Zhou H.-Y., Sun H., Li X., Yu L., Deng Y.-Q., Li R.-T., Cheng M.-L., He B., Zhou J., Li X.-F., Wu A., Yi C., Qin C.-F., The m6A methylome of SARS-CoV-2 in host cells. Cell Res. 31, 404–414 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Zhang T., Yang Y., Xiang Z., Gao C. C., Wang W., Wang C., Xiao X., Wang X., Qiu W. N., Li W. J., Ren L., Li M., Zhao Y. L., Chen Y. S., Wang J., Yang Y. G., N(6)-methyladenosine regulates RNA abundance of SARS-CoV-2. Cell Discov. 7, 7 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Zhang X., Hao H., Ma L., Zhang Y., Hu X., Chen Z., Liu D., Yuan J., Hu Z., Guan W., Methyltransferase-like 3 modulates severe acute respiratory syndrome coronavirus-2 RNA N6-methyladenosine modification and replication. mBio. 12, e0106721 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Li N., Hui H., Bray B., Gonzalez G. M., Zeller M., Anderson K. G., Knight R., Smith D., Wang Y., Carlin A. F., Rana T. M., METTL3 regulates viral m6A RNA modification and host cell innate immune responses during SARS-CoV-2 infection. Cell Rep. 35, 109091 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Burgess H. M., Depledge D. P., Thompson L., Srinivas K. P., Grande R. C., Vink E. I., Abebe J. S., Blackaby W. P., Hendrick A., Albertella M. R., Kouzarides T., Stapleford K. A., Wilson A. C., Mohr I., Targeting the m6A RNA modification pathway blocks SARS-CoV-2 and HCoV-OC43 replication. Genes Dev. 35, 1005–1019 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Courtney D. G., Tsai K., Bogerd H. P., Kennedy E. M., Law B. A., Emery A., Swanstrom R., Holley C. L., Cullen B. R., Epitranscriptomic addition of m(5)C to HIV-1 transcripts regulates viral gene expression. Cell Host Microbe 26, 217–227.e6 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Kong W., Biswas A., Zhou D., Fiches G., Fujinaga K., Santoso N., Zhu J., Nucleolar protein NOP2/NSUN1 suppresses HIV-1 transcription and promotes viral latency by competing with Tat for TAR binding and methylation. PLOS Pathog. 16, e1008430 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Feng J., Xu T., He M., Li J., Yao P., Ma C., Yang S., Xu Z., Yan K., Chen X., Wang H., Liu J., Zeng C., Xia Y., Yan H., Zhou L., Chen Y., NSUN2-mediated m5C modification of HBV RNA positively regulates HBV replication. PLOS Pathog. 19, e1011808 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Courtney D. G., Chalem A., Bogerd H. P., Law B. A., Kennedy E. M., Holley C. L., Cullen B. R., Extensive epitranscriptomic methylation of A and C residues on murine leukemia virus transcripts enhances viral gene expression. mBio. 10, e01209 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Eckwahl M., Xu R., Michalkiewicz J., Zhang W., Patel P., Cai Z., Pan T., 5-Methylcytosine RNA modifications promote retrovirus replication in an ALYREF reader protein-dependent manner. J. Virol. 94, e00544 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Henry B. A., Kanarek J. P., Kotter A., Helm M., Lee N., 5-methylcytosine modification of an Epstein-Barr virus noncoding RNA decreases its stability. RNA 26, 1038–1048 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Yang X., Yang Y., Sun B. F., Chen Y. S., Xu J. W., Lai W. Y., Li A., Wang X., Bhattarai D. P., Xiao W., Sun H. Y., Zhu Q., Ma H. L., Adhikari S., Sun M., Hao Y. J., Zhang B., Huang C. M., Huang N., Jiang G. B., Zhao Y. L., Wang H. L., Sun Y. P., Yang Y. G., 5-methylcytosine promotes mRNA export-NSUN2 as the methyltransferase and ALYREF as an m(5)C reader. Cell Res. 27, 606–625 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Chen X., Li A., Sun B. F., Yang Y., Han Y. N., Yuan X., Chen R. X., Wei W. S., Liu Y., Gao C. C., Chen Y. S., Zhang M., Ma X. D., Liu Z. W., Luo J. H., Lyu C., Wang H. L., Ma J., Zhao Y. L., Zhou F. J., Huang Y., Xie D., Yang Y. G., 5-methylcytosine promotes pathogenesis of bladder cancer through stabilizing mRNAs. Nat. Cell Biol. 21, 978–990 (2019). [DOI] [PubMed] [Google Scholar]
  • 49.Barbieri I., Kouzarides T., Role of RNA modifications in cancer. Nat. Rev. Cancer 20, 303–322 (2020). [DOI] [PubMed] [Google Scholar]
  • 50.Wang H., Feng J., Zeng C., Liu J., Fu Z., Wang D., Wang Y., Zhang L., Li J., Jiang A., He M., Cao Y., Yan K., Tang H., Guo D., Xu K., Zhou X., Zhou L., Lan K., Zhou Y., Chen Y., NSUN2-mediated M5c methylation of IRF3 mRNA negatively regulates type I interferon responses during various viral infections. Emerg. microbes infect. 12, 2178238 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Sun J., Zhuang Z., Zheng J., Li K., Wong R. L.-Y., Liu D., Huang J., He J., Zhu A., Zhao J., Li X., Xi Y., Chen R., Alshukairi A. N., Chen Z., Zhang Z., Chen C., Huang X., Li F., Lai X., Chen D., Wen L., Zhuo J., Zhang Y., Wang Y., Huang S., Dai J., Shi Y., Zheng K., Leidinger M. R., Chen J., Li Y., Zhong N., Meyerholz D. K., Mc Cray P. B. Jr., Perlman S., Zhao J., Generation of a broadly useful model for COVID-19 pathogenesis, vaccination, and treatment. Cell 182, 734–743.e5 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Xiong Y., Liu Y., Cao L., Wang D., Guo M., Jiang A., Guo D., Hu W., Yang J., Tang Z., Wu H., Lin Y., Zhang M., Zhang Q., Shi M., Liu Y., Zhou Y., Lan K., Chen Y., Transcriptomic characteristics of bronchoalveolar lavage fluid and peripheral blood mononuclear cells in COVID-19 patients. Emerg. microbes & infect. 9, 761–770 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Minkoff J. M., tenOever B., Innate immune evasion strategies of SARS-CoV-2. Nat. Rev. Microbiol. 21, 178–194 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Yin J. A., Gao G., Liu X. J., Hao Z. Q., Li K., Kang X. L., Li H., Shan Y. H., Hu W. L., Li H. P., Cai S. Q., Genetic variation in glia-neuron signalling modulates ageing rate. Nature 551, 198–203 (2017). [DOI] [PubMed] [Google Scholar]
  • 55.Ju X., Zhu Y., Wang Y., Li J., Zhang J., Gong M., Ren W., Li S., Zhong J., Zhang L., Zhang Q. C., Zhang R., Ding Q., A novel cell culture system modeling the SARS-CoV-2 life cycle. PLOS Pathog. 17, e1009439 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Li Y., Tan X., Deng J., Liu X., Liu Q., Zhang Z., Huang X., Shen C., Xu K., Zhou L., Chen Y., An optimized high-throughput SARS-CoV-2 dual reporter trans-complementation system for antiviral screening in vitro and in vivo. Virol Sin. 39, 447–458 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figs. S1 to S13

Tables S1 and S2

sciadv.adn9519_sm.pdf (3.1MB, pdf)

Articles from Science Advances are provided here courtesy of American Association for the Advancement of Science

RESOURCES