3.4 m6A methylation was closely associated with actively expressed genes in BPH-infested plants
We analyzed the Euclidian distance coefficients among gene transcript profiles based on nanopore DRS. The distance coefficients between the replicates were lower than those of other objects, and the tile colors of the two were relatively close (Supporting Information: Figure S7), suggesting their reproducible patterns. These sequenced genes were divided into three groups: transcripts per kilobase of exon model per million mapped reads (TPM) < 1, 1 < TPM < 5, and TPM > 5 (Supporting Information: Table S5). The proportion of gene numbers in each category was calculated and shown as a heatmap (Supporting Information: Figure S8). Rice genes that could be annotated by m6A positions were referred to ‘m6A genes’ and those that could not be annotated were marked as ‘non-m6A genes’ for subsequent analyses. In the Nip and Nl-Nip groups, most m6A genes were distributed in the highly expressed gene category (TPM > 5). In contrast, non-m6A genes were evenly distributed across the three groups and were biased towards gene categories with low expression (TPM < 1). Compared to that in Nip, the proportion of non-m6A genes in low expression category was lower in Nl-Nip group (Figure 2a, Supporting Information: Figure S8). When the m6A and non-m6A genes were divided by gene expression categories—high (TPM ≥ 1) or low (TPM < 1)—the number of transcripts showing high/low expression was recorded for control Nip and Nl-Nip samples (Figure 2b). We found that most genes were m6A methylated and distributed in the high expression category in Nip and Nl-Nip groups; these gene numbers were enriched upon BPH infestation (Figure 2b). In both Nip and Nl-Nip groups, m6A-methylated genes were expressed at a higher level than non-m6A-methylated genes (Figure 2c). Thus, m6A methylation mainly occurred in highly expressed genes in the control and BPH-infested samples, while the m6A modified gene numbers increased in the high expression category with BPH infestation.
3.5 m6A methylation positively correlated with the transcript expression in BPH-infested rice
We found 21,718 (76.59%) transcripts that showed no significant difference (no change), 3,506 (12.36%) were upregulated and 3,131 (11.04%) were downregulated in BPH-infested plants compared with those in the control plants (Figure 2d, Supporting Information: Figure S9a, Table S6) . A total of 116,817 methylated positions were detected among all expressed transcripts. Among the m6A modifications, 63.99%, 17.43%, and 18.58% exhibited no change, or fell into the up- or down-m6A directions, respectively (Supporting Information: Table S6). A large proportion of m6A modification positions were found in CDS and 3′-UTR, with most showing no change in expression (Figure 2e, Supporting Information: Table S6). Among up-directed m6A-methylated transcripts, the number of upregulated transcripts was higher than that of downregulated transcripts (2,094 vs. 856) in the NI-Nip vs. Nip comparison; conversely, in down-directed transcripts, the number of downregulated ones was higher than that of upregulated ones (1,995 vs. 743) (Figure 2e). The correlational heat map showed a strong positive correlation between the m6A methylation direction and the corresponding transcript change, with many transcripts and m6A modifications exhibiting simultaneous up- or downregulation/direction (Supporting Information: Figure S9b). The abundance of upregulated transcripts that underwent 5′-UTR methylation was higher than that of the downregulated transcripts. However, this tendency was not evident among 3′-UTR methylated transcripts (Figure 2f). Taken together, the m6A-methylation differential types in BPH-infested samples were positively correlated with the corresponding transcript regulation types, with transcripts containing different m6A methylated functional elements likely involved in diverse transcript regulation.