Targeted DNA N⁶-methyladenine editing by dCas9 fused to METTL4 in the lepidopteran model insect Bombyx mori

We have established a novel CRISPR–dCas9–METTL4 epigenome editing tool that can methylate target regions to achieve site-specific DNA 6mA methylation in both hypermethylated and hypomethylated genes. Targeted methylation on genes by dCas9–METTL4 results in misexpression, allowing for the functional investigation of target genes of interest in silkworm. Dear Editor, N6-Methyladenine (6mA) DNA modification is an important epigenetic mechanism with roles in regulating gene expression, nucleosome positioning, DNA damage repair, and cell cycle progression (Heyn & Esteller, 2015; Luo et al., 2015; Boulias & Greer, 2022). Despite progress in understanding the biological functions of 6mA, the contribution of individual 6mA installations on site-specific target genes is largely unknown, and therefore deciphering the molecular mechanism of 6mA in target gene expression has been difficult. The clustered regularly interspaced short palindromic repeat (CRISPR)–CRISPR-associated protein 9 (Cas9) system has promoted the functional research of genes by enabling precision genome editing (Doudna & Charpentier, 2014; Hsu et al., 2014; Komor et al., 2017). The engineering of the nuclease-deficient version of Cas9 (dCas9) has been used to tether epigenetic effector domains, such as ten-eleven translocation 1, DNA methyltransferase 3A, RNA methyltransferase-like 3, RNA methyltransferase-like 14, and ALKB homolog 5, for site-specific modifications to achieve programmable targeted epigenetic editing on DNA and RNA (Gilbert et al., 2013; Morita et al., 2016; Vojta et al., 2016; Pulecio et al., 2017; Liu et al., 2019a). We constructed a CRISPR–dCas9–METTL4 system that expresses the Bombyx mori (silkworm) 6mA methyltransferase METTL4 (SilkDB3.0: BMSK0011038) fused with the dCas9 protein to achieve site-specific methylation on DNA adenines, and investigated the regulatory role of 6mA on gene expression in the lepidopteran model silkworm (International Silkworm Genome Consortium, 2008; Xia et al., 2014). We first constructed a dCas9–METTL4 fusion protein by linking the full-length silkworm METTL4 (BmMETTL4) to the C-terminus of dCas9 with a polypeptide linker under the control of the IE2 promoter (Fig. 1A, B; Appendix S1) (Liu et al., 2019b). To analyze the expression of dCas9–METTL4 in cultured B. mori embryonic (BmE) cells, we extracted the proteins after dCas9 or dCas9–METTL4 transfection for western blot analysis. We found that the dCas9–METTL4 fusion protein was expressed, and that the protein size was larger than that of the dCas9 protein and was similar to the predicted size (Fig. 1C), which indicated the successful expression of the dCas9–METTL4 system in silkworm. To assess whether the dCas9–METTL4 expression system induced site-specific 6mA modification in silkworm, we targeted the proteasome subunit beta 1 gene (SilkDB3.0: BMSK0011382), which was previously found to be a DNA methylation target by 6mA-seq analysis (Wang et al., 2018). Moreover, the expression of proteasome subunit beta 1 was shown to increase upon knockdown of BmMETTL4 in cultured silkworm cells (Wang et al., 2018). We analyzed the structure and methylation status of proteasome subunit beta 1 in the silkworm genome and found that 6mA was significantly enriched on the upstream promoter of the proteasome subunit beta 1 gene (Wang et al., 2018) (Fig. 1D), suggesting that the 6mA modification may regulate the expression of target genes. We also found that RNA m6A modification on proteasome subunit beta 1 occurred mainly on the gene body, which differs from the site for the DNA 6mA modification on proteasome subunit beta 1 (Fig. 1D). This finding implies that 6mA and m6A installations may have distinct regulatory roles in gene expression (Li et al., 2019). We therefore designed guide RNAs (gRNAs) that target the 5′ untranslated region and gene body of proteasome subunit beta 1 (Fig. 1D; Table S1). and cloned it into a gRNA expression system. After transfection in BmE cells, we collected the cells and recovered methylated DNA fragments enriched using a 6mA antibody and methylated DNA immunoprecipitation (6mA-MeDIP). We measured the 6mA levels by PCR. We found that 6mA increased on the proteasome subunit beta 1 gene upon induction of dCas9–METTL4 expression, compared with 6mA on the dCas9 control (Figs. 1E and S1), implying that dCas9–METTL4 and gRNA targeted the DNA region for 6mA installation. Indeed, a study of the genome-wide distribution of 6mA modifications in silkworm showed that 6mA was negatively associated with active gene expression (Wang et al., 2018). However, direct evidence for the effect of this modification on gene expression remains to be determined. To determine the transcriptional effect of dCas9–METTL4 targeting on the gene, we isolated the proteasome subunit beta 1 RNA and carried out reverse transcription quantitative real-time PCR (RT-qPCR) analysis. We found that the expression of proteasome subunit beta 1 in cells treated with dCas9–METTL4 was much lower than its expression with the dCas9 control (Fig. 1F), which is consistent with a previous finding that the knockdown of BmMETTL4 upregulated the expression of proteasome subunit beta 1 (Wang et al., 2018). To further verify the efficiency of the dCas9–METTL4 system, we selected several other 6mA target genes for expression analysis. All of the selected genes were clearly downregulated by the expression of dCas9–METTL4 (Fig. S2). Together, these results imply that dCas9–METTL4 increased the methylation of hypermethylated DNA targets in a gRNA-guided manner in silkworm. These findings further validate the high efficiency and specificity of the CRISPR–dCas9 system to install and remove DNA or RNA modifications (Gilbert et al., 2013; Morita et al., 2016; Vojta et al., 2016; Pulecio et al., 2017; Liu et al., 2019a, 2019b). To investigate the ability of the dCas9–METTL4 expression system to install 6mA modifications on a hypomethylated nontarget gene in silkworm, we used the BmMETTL4 gene, which was shown to have a low methylation level in 6mA-Seq and m6A-Seq analyses (Wang et al., 2018; Li et al., 2019). We designed gRNAs that targeted the gene body of BmMETTL4 (Fig. 2A). After transfection, 6mA-MeDIP-PCR was used to analyze the methylation status of this target gene. We found that the methylation levels on BmMETTL4 were significantly induced by the dCas9–METTL4 expression system, and none of this methylation was found with the dCas9 control (Figs. 2B and S3). To determine the effect of dCas9–METTL4 on the methylation of endogenous BmMETTL4, we analyzed its expression by RT-qPCR. The results showed that BmMETTL4 expression was downregulated by dCas9–METTL4, compared with its expression under dCas9 treatment (Fig. 2C). These results indicate that the dCas9–METTL4 system can install 6mA on a hypomethylated nontarget gene at a site-specific location in the silkworm genome. To examine whether the dCas9–METTL4-mediated regulation of target gene expression was functional, we investigated the effect of the dCas9–METTL4-mediated downregulation of BmMETTL4 on the cell cycle, because the RNA interference (RNAi)-mediated knockdown of BmMETTL4 was previously shown to result in the disordered formation of metaphase spindle in cultured silkworm cells (Wang et al., 2018). We found that the dCas9 and gRNA control treatments showed normal chromosome alignments during metaphase, whereas the downregulation of BmMETTL4 by dCas9–METTL4 and gRNA-mediated targeting resulted in abnormal metaphase cells with a disordered congression of chromosomes (Figs. 2D and S4). This finding confirmed that the dCas9–METTL4-mediated regulation of target gene expression was a functional system in silkworm. Although our results are informative for elucidating the installation and importance of DNA 6mA modification in cellular regulation, they remain limited in determining the details of the epigenetic mechanism. For instance, the epigenetic status for regulating gene expression can be inherited after cell division and maintained across generations (Huypens et al., 2016; Komor et al., 2017; Pulecio et al., 2017). In future studies, it may be interesting to investigate whether the transient establishment of 6mA modification by transfected dCas9–METTL4 on a target gene in cultured silkworm cells can be inherited after removing the dCas9–METTL4. We anticipate that using the dCas9–METTL4 system to directly create 6mA modifications will help in understanding the mechanism of epigenetic inheritance across generations. In summary, we have established a CRISPR–dCas9–METTL4 epigenome editing tool in silkworm that can methylate target regions to achieve site-specific 6mA installation and the regulation of gene expression in both hypermethylated and hypomethylated genes. Furthermore, the targeted methylation of genes resulted in altered expression, allowing for the functional investigation of target genes of interest. This study is a key step towards the epigenetic regulation of target gene expression by site-specific 6mA modification in cultured silkworm cells. In future work, the establishment of an individual silkworm line stably expressing the dCas9–METTL4 system will be valuable not only to regulate the expression of target genes in an epigenetic manner, but also to determine the mechanism of epigenetic inheritance in silkworm individuals. We would like to thank Prof. Sanyuan Ma (Biological Science Research Center of Southwest University) for kindly providing the CRISPR–dCas9 plasmid. This work was supported by National Key Research and Development Program of China (No. 2022YFD1201600), National Natural Science Foundation of China (Nos. 32030103 and 32172798), and Natural Science Foundation of Chongqing (No. cstc2020jcyj-cxttX0001). The authors declare they have no conflicts of interest. Fig. S1 6mA-MeDIP-PCR analysis of methylation status on proteasome subunit beta 1 gene. Fig. S2 Verification of dCas9–METTL4-mediated 6mA methylation on target gene. Fig. S3 6mA-MeDIP-PCR analysis of methylation status on BmMETTL4 gene. Fig. S4 Immunofluorescence staining of metaphase cells. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.