CRISPR/dCas9-Mediated DNA Methylation Editing on emx2 in Chinese Tongue Sole (Cynoglossus semilaevis) Testis Cells

DNA methylation is a key epigenetic mechanism orchestrating gene expression networks in many biological processes. Nonetheless, studying the role of specific gene methylation events in fish faces challenges. In this study, we validate the regulation of DNA methylation on empty spiracles homeobox 2 (emx2) expression with decitabine treatment in Chinese tongue sole testis cells. We used the emx2 gene as the target gene and developed a new DNA methylation editing system by fusing dnmt3a with catalytic dead Cas9 (dCas9) and demonstrated its ability for sequence-specific DNA methylation editing. Results revealed that utilizing dCas9-dnmt3a to target emx2 promoter region led to increased DNA methylation levels and decreased emx2 expression in Chinese tongue sole testis cells. More importantly, the DNA methylation editing significantly suppressed the expression of MYC proto-oncogene, bHLH transcription factor (myc), one target gene of emx2. Furthermore, we assessed the off-target effects of dCas9-dnmt3a and confirmed no significant impact on the predicted off-target gene expression. Taken together, we developed the first DNA methylation editing system in marine species and demonstrated its effective editing ability in Chinese tongue sole cells. This provides a new strategy for both epigenetic research and molecular breeding of marine species.


Introduction
Empty spiracles homeobox 2 (EMX2), a member of the Homeobox gene family, plays an important role as a developmental regulator in both growth and sex differentiation/determination [1].In mice, EMX2 transcripts are detectable in the "bipotential" gonad before the emergence of morphological differences between the sexes [2].Specifically, gonadal regression occurs during the typical onset of sexual differentiation in both male and female knockout (KO) mice deficient in EMX2 [3,4].EMX2 also regulates Wnt/β-catenin signaling [5], an important pathway for the development of gonads [6], while represses the expression of stem cell regulatory genes, such as KLF4 and MYC [7].In the marine fish Chinese tongue sole (Cynoglossus semilaevis), emx2 is considered a potential target of gonadal DNA methylomes [8].Variations across pseudomale (The genotype shows ZW, but the gonads develop into testis), female, and male fish DNA methylomes of emx2 have showed an opposite trend to the gene expression levels.It suggests that the expression of emx2 may be affected by DNA methylation.However, the precise regulatory mechanism of DNA methylation on emx2 during sex determination and differentiation remains unclear.
Recent studies involving DNA methyltransferase (DNMT) mutations have confirmed that changes in global DNA methylation levels can lead to male sterility in mice, highlighting its importance for normal spermatogenesis [9].In mammals, a targeted addition of 5mC (5-methylcytosine, a common epigenetic modification in DNA) has been achieved by using human DNMT3A catalytic domain (DNMT3Acd) [10][11][12].Moreover, 5-Aza-2 ′ -deoxycytidine (5-Aza-dC), a DNA methyltransferase inhibitor employed to induce hypomethylation of the epigenome [13], and widely used in tissue culture and in animal models to facilitate promoter demethylation and reactivate gene expression was used.[14,15].These regulatory mechanisms may induce widespread global changes in methylation levels throughout the entire genome, leading to challenges in achieving methylation sequence-specific methylation modifications.
In recent years, the development of CRISPR/Cas9 has provided a new strategy for manipulating sequence-specific methylation.A combination of catalytically dead Cas9 nuclease (dCas9) and DNA methyltransferase has been employed to enhance sequencespecific methylation, while demethylase is used for DNA demethylation [16].This system has demonstrated promise in gene function research and disease models, such as targeting the promoter of amyloid precursor protein (APP) via dCas9-Dnmt3a treatment to alter its DNA methylation for Alzheimer's disease (AD) [11] and removing the 5mC target at the FLOWERING WAGENINGEN (FWA) promoter in the genome of Arabidopsis [17].In this context, these studies suggest that the epigenetic editing system based on dCas9 is a powerful tool for DNA methylation studies and has broad research applications.Nevertheless, while previous studies predominately focused on human and model species, limited attention has been given to marine species.
In this study, we aimed to construct a system for single-gene targeted DNA methylation editing in Chinese tongue sole.For this purpose, we validate the regulatory role of DNA methylation on empty spiracles homeobox 2 (emx2) expression with decitabine treatment in Chinese tongue sole testis cells.And then, we developed a novel DNA methylation editing system by fusing dnmt3a with catalytically dead Cas9 (dCas9) and demonstrated its capability for sequence-specific DNA methylation editing.To the best of our knowledge, we developed the first DNA methylation editing system in marine species and demonstrated its effective editing ability in Chinese tongue sole cells.This approach offers a new strategy for advancing both epigenetic research and molecular breeding in marine species.

Effects of 5-Aza-dC on emx2 Methylation and Expression
To investigate the connection between emx2 gene expression and DNA methylation, we exposed dissociated testis cells from Chinese tongue sole to different concentrations of 5-Aza-dC.Bisulfite sequencing revealed a reduction in the number of methylated CpGs in the promoter region of emx2 (Figure 1A-D, Figure S1) following 5-Aza-dC treatment.Bisulfite sequencing PCR (BSP) showed that DNA methylation of the non-treated cell group was approximately 25.63% (Figure 1E).A decrease in DNA methylation levels was evident in the emx2 promoter region of the 5-Aza-dC-treated groups compared to the control group (Figure 1E).Particularly, treatment with 20 µM of 5-Aza-dC showed the most substantial reduction, with methylation levels of 10.63% (Figure 1E).
To clarify how the DNA methylation regulates gene expression of emx2, we used qPCR to confirm changes in gene expression following 5-Aza-dC treatment.The results showed an elevated expression of emx2 after 48 h of treatment with different concentrations of 5-Aza-dC.The concentration 20 µM showed the most significantly increase (Figure 1F).

Construction of CRISPR/dCas9-dnmt3a System in Chinese Tongue Sole
In order to make the building of the DNA methylation editing system so that it can be applied to the Chinese tongue sole, we cloned and analyzed the dnmt3a of Chinese tongue sole as the effector of editing system.We performed multiple sequence alignment analysis and phylogenetic tree construction for dnmt3a and conducted a comparative analysis between dnmt3a of Chinese tongue sole and other species, while all proteins contain conserved PWWP and ADDz domains and differ in their catalytic domains (Figures S2-S4).
To construct a flexible system for targeting emx2, we used dnmt3a of Chinese tongue sole as an effector enzyme, fused with dCas9 protein.This construct was guided to the emx2 promoter region using small guide RNAs (Figure 2A).The coding sequence (CDS) of dnmt3a was cloned and fused with dCas9 under the control of the constitutive ubiquitin C (UbC) promoter.Additionally, nuclear localization signals (NLS) were incorporated into both dCas9 and dnmt3a [12] (Figure 2B, Figure S5).Furthermore, we designed To clarify how the DNA methylation regulates gene expression of emx2, we used qPCR to confirm changes in gene expression following 5-Aza-dC treatment.The results showed an elevated expression of emx2 after 48 h of treatment with different concentrations of 5-Aza-dC.The concentration 20 µM showed the most significantly increase (Figure 1F).

Construction of CRISPR/dCas9-dnmt3a System in Chinese Tongue Sole
In order to make the building of the DNA methylation editing system so that it can be applied to the Chinese tongue sole, we cloned and analyzed the dnmt3a of Chinese tongue sole as the effector of editing system.We performed multiple sequence alignment analysis and phylogenetic tree construction for dnmt3a and conducted a comparative analysis between dnmt3a of Chinese tongue sole and other species, while all proteins contain conserved PWWP and ADDz domains and differ in their catalytic domains (Figures S2-S4).
To construct a flexible system for targeting emx2, we used dnmt3a of Chinese tongue sole as an effector enzyme, fused with dCas9 protein.This construct was guided to the emx2 promoter region using small guide RNAs (Figure 2A).The coding sequence (CDS) of dnmt3a was cloned and fused with dCas9 under the control of the constitutive ubiquitin C (UbC) promoter.Additionally, nuclear localization signals (NLS) were incorporated into both dCas9 and dnmt3a [12] (Figure 2B, Figure S5).Furthermore, we designed six sgRNAs targeting emx2, with each sgRNA driven by an independent U6 promoter (Figure 2B,C; Figure S6).Then, we designed a pair of methylation-specific PCR primers for the CpG Island of the emx2 gene to amplify this DNA sequence.The position and the number of CpGs contained are shown in Figure 2D.
six sgRNAs targeting emx2, with each sgRNA driven by an independent U6 promoter (Figure 2B,C; Figure S6).Then, we designed a pair of methylation-specific PCR primers for the CpG Island of the emx2 gene to amplify this DNA sequence.The position and the number of CpGs contained are shown in Figure 2D.

Targeted Methylation of the emx2 Promoter Region by dCas9-dnmt3a System
In order to investigate whether the dCas9-dnmt3a system can change the methylation level of the emx2 promoter region, we transfected dCas9-dnmt3a plasmid with the sgRNA vector in Chinese tongue sole testis cells.Results of fluorescence intensity and relative expression of dnmt3a showed high transfection efficiency (Figure S7).Results of BSP showed that emx2 methylation status differs after sgRNA-guided DNA methylation editing (Figure 3A-J).Among them, transfection with dCas9-dnmt3a/sgRNA1 and dCas9-dnmt3a/sgRNA3 resulted in a significant increase in DNA methylation in the emx2 promoter region, rising about 30% and 15%, respectively (Figure 3J).However, other sgRNAs guiding DNA methylation editing showed no significant change (Figure 3D,F-H).Additionally, we tested whether dCas-dnmt3a/sgRNA1+3 had a be er methylation effect.The BSP results indicate that the combined methylation effect of sgR-NA1+sgRNA3 was similar to that of sgRNA1 and sgRNA3 individually, approximately 17.5% (Figure 3I,J).In the subsequent experiments, we targeted Chinese tongue sole cells with sgRNA1, sgRNA3 and sgRNA1+3.The qPCR analysis detected that emx2 expression was dramatically decreased in sgRNAs' guiding compared with the blank control and dCas9-dnmt3a with non-sgRNAs (Figure 3K), suggesting a successful addition of methylation at the emx2 promoter and, consequently, resulting in significantly reduced expression of emx2.

Targeted Methylation of the emx2 Promoter Region by dCas9-dnmt3a System
In order to investigate whether the dCas9-dnmt3a system can change the methylation level of the emx2 promoter region, we transfected dCas9-dnmt3a plasmid with the sgRNA vector in Chinese tongue sole testis cells.Results of fluorescence intensity and relative expression of dnmt3a showed high transfection efficiency (Figure S7).Results of BSP showed that emx2 methylation status differs after sgRNA-guided DNA methylation editing (Figure 3A-J).Among them, transfection with dCas9-dnmt3a/sgRNA1 and dCas9-dnmt3a/sgRNA3 resulted in a significant increase in DNA methylation in the emx2 promoter region, rising about 30% and 15%, respectively (Figure 3J).However, other sgRNAs guiding DNA methylation editing showed no significant change (Figure 3D,F-H).Additionally, we tested whether dCas-dnmt3a/sgRNA1+3 had a better methylation effect.The BSP results indicate that the combined methylation effect of sgRNA1+sgRNA3 was similar to that of sgRNA1 and sgRNA3 individually, approximately 17.5% (Figure 3I,J).In the subsequent experiments, we targeted Chinese tongue sole cells with sgRNA1, sgRNA3 and sgRNA1+3.The qPCR analysis detected that emx2 expression was dramatically decreased in sgRNAs' guiding compared with the blank control and dCas9-dnmt3a with non-sgRNAs (Figure 3K), suggesting a successful addition of methylation at the emx2 promoter and, consequently, resulting in significantly reduced expression of emx2.

Off-Target Effects of CRISPR/dCas9-dnmt3a System
To assess potential off-target effects, we analyzed four predicted off-target genes in the Chinese tongue sole genome using the sgRNAs' sequence (sgRNA1 and sgRNA3) and the PAM sequence of Cas9 (NGG): sodium-dependent phosphate transporter 2 (slc20a2), DNA damage recognition and repair factor (xpa), DEAD-box helicase 17 (ddx17), and lysine acetyltransferase 6B (kat6b).BSP sequencing from the promoter of these genes indicated no significant change between the control and treatment groups (Figures 4A-D  and S8), except for a notable upregulation in the expression of xpa with sgRNA3 treatment (Figure 4E-H).However, since no methylation levels were altered in the xpa promoter region (Figure 4D), we do not a ribute such change in expression to DNA methylation levels.

Off-Target Effects of CRISPR/dCas9-dnmt3a System
To assess potential off-target effects, we analyzed four predicted off-target genes in the Chinese tongue sole genome using the sgRNAs' sequence (sgRNA1 and sgRNA3) and the PAM sequence of Cas9 (NGG): sodium-dependent phosphate transporter 2 (slc20a2), DNA damage recognition and repair factor (xpa), DEAD-box helicase 17 (ddx17), and lysine acetyltransferase 6B (kat6b).BSP sequencing from the promoter of these genes indicated no significant change between the control and treatment groups (Figures 4A-D and S8), except for a notable upregulation in the expression of xpa with sgRNA3 treatment (Figure 4E-H).However, since no methylation levels were altered in the xpa promoter region (Figure 4D), we do not attribute such change in expression to DNA methylation levels.between the blank control and sgRNA3 groups, while lines below the graph highlight that the sgRNAs were co-transfected with the CRISPR/dCas9-dnmt3a vector.The asterisks (*) are used to denote significant differences between groups using two-way ANOVA and Tukey's multiple comparisons.*** p < 0.001.

Methylation Editing on emx2 Affects Expression of Growth-Related Gene
Furthermore, we examined the changes in the regulation of downstream genes implicated in gonadal development and the maintenance of stem cell pluripotency after modulating emx2 gene expression with the CRISPR/dCas9-dnmt3a system plus sgRNAs (Figure 5A-C).Results showed that qPCR analysis of wingless-type MMTV integration site family, member 1 (wnt1), Kruppel-like factor 4 (klf4), and myc expression showed that myc was downregulated in the groups where the sgRNAs were co-transfected with the CRISPR/dCas9-dnmt3a vector (Figure 5A), while the expression of klf4 and wnt1 was not significantly affected (Figure 5B,C).

Discussion
CRISPR/dCas9-based epigenetic editing technology has been widely used in mammals and plants due to its powerful editing and precise targeting ability combined with between the blank control and sgRNA3 groups, while lines below the graph highlight that the sgRNAs were co-transfected with the CRISPR/dCas9-dnmt3a vector.The asterisks (*) are used to denote significant differences between groups using two-way ANOVA and Tukey's multiple comparisons.*** p < 0.001.

Methylation Editing on emx2 Affects Expression of Growth-Related Gene
Furthermore, we examined the changes in the regulation of downstream genes implicated in gonadal development and the maintenance of stem cell pluripotency after modulating emx2 gene expression with the CRISPR/dCas9-dnmt3a system plus sgRNAs (Figure 5A-C).Results showed that qPCR analysis of wingless-type MMTV integration site family, member 1 (wnt1), Kruppel-like factor 4 (klf4), and myc expression showed that myc was downregulated in the groups where the sgRNAs were co-transfected with the CRISPR/dCas9-dnmt3a vector (Figure 5A), while the expression of klf4 and wnt1 was not significantly affected (Figure 5B,C).

Methylation Editing on emx2 Affects Expression of Growth-Related Gene
Furthermore, we examined the changes in the regulation of downstream genes implicated in gonadal development and the maintenance of stem cell pluripotency after modulating emx2 gene expression with the CRISPR/dCas9-dnmt3a system plus sgRNAs (Figure 5A-C).Results showed that qPCR analysis of wingless-type MMTV integration site family, member 1 (wnt1), Kruppel-like factor 4 (klf4), and myc expression showed that myc was downregulated in the groups where the sgRNAs were co-transfected with the CRISPR/dCas9-dnmt3a vector (Figure 5A), while the expression of klf4 and wnt1 was not significantly affected (Figure 5B,C).

Discussion
CRISPR/dCas9-based epigenetic editing technology has been widely used in mammals and plants due to its powerful editing and precise targeting ability combined with

Discussion
CRISPR/dCas9-based epigenetic editing technology has been widely used in mammals and plants due to its powerful editing and precise targeting ability combined with low off-target effects [17][18][19][20].The use of this tool provides new strategies for controlling the occurrence of diseases and in animal growth studies [21,22].Nonetheless, while previous studies predominately focused on human and mammals model species, limited attention has been given to marine species.Chinese tongue sole is considered a promising species for Chinese aquaculture due to its large size and high-quality meat [23], holding great potential to investigate gene regulation and inheritance mechanisms related to epigenetic mechanisms [8].emx2 is recognized for regulating embryonic development as well as for modulating transcriptional factors in different biological pathways [24][25][26] and epigenetic modification in fish sexual reversal [8].Nonetheless, its regulation mechanisms involving DNA methylation remain largely unclear in fish gonads.
In our study, the relationship between emx2 expression and DNA methylation was determined by treatment with 5-Aza-dC.This is a relatively common reagent to study the relationship between gene methylation level and expression.We are to build an emx2 DNA methylation editing system in the process of finding that, in other species, the effector enzyme that performs DNA methylation editing, dnmt3a, is commonly of human origin.However, due to the differences in the catalytic domain of dnmt3a in different species, the use of human DNMT3Acd as an effector was unable to function in our system.So, we cloned the dnmt3a of Chinese tongue sole as the effector (Figures S2-S4).By fusing the dnmt3a gene of Chinese tongue sole to dCas9 to form a DNA methylation editing system (Figure 2), we targeted the emx2 promoter region to directly change its DNA methylation levels.In addition, the potential off-target effects and the impact of editing on emx2 downstream genes were also analyzed.
In this context, our initial findings indicated that the DNA methylation levels in cells derived from Chinese tongue sole testis were approximately 25.63% in the promoter region of emx2 of the control group (Figure 1A).Following treatment with 5-Aza-Dc, the DNA methylation levels in this region decreased to approximately 10.63% (Figure 1E), accompanied by an upregulation of emx2 expression across all concentrations (Figure 1F).As previously described in the literature, DNA methylation plays a crucial role in modulating the chromatin structure of regulatory regions in various genes, leading to a reduction in their expression levels [27,28].Interestingly, the concentration of 20 µM 5-Aza-dC induced the most significant demethylation in the promoter region of emx2, resulting in the highest increase in transcript expression of this gene.
Additionally, we successfully demonstrated that our DNA methylation editing system co-transfected with specific guides for Chinese tongue sole primary testis cells.Primary cells have a short time in vitro and do not undergo an immortalization process, so their biological characteristics have not changed much, and they still maintain the original genetic characteristics, which can better reflect the growth state of cells in vivo, so as to obtain data closer to the physiological function in vivo.In both the study by Michael Chavez and the study by Kanut Laoharawee, primary cells were used for CRISPR-based gene editing [29,30].Thus, primary testis cells of Chinese tongue sole are of great significance for the study of targeted DNA methylation.After DNA methylation editing of emx2, we found that the sgRNA1, sgRNA3, and their combination sgRNA1/3, led to an increased methylation level compared to the control and dCas9-dnmt3a alone (Figure 3J), followed by a significant decrease in emx2 expression (Figure 3K).The results showed that the DNA methylation level of CpG islands in the emx2 promoter region was increased about 30% by the sgRNA1-guided editing system, about 15% by the sgRNA3-guided editing system, and about 17.5% by multiplexing of several sgRNA-guided editing systems.This indicated that the sgRNA3-guided editing system had a higher editing efficiency.The single gRNA leads to efficient and widespread methylation of the promoters.This conclusion is consistent with the work of Peter Stepper et al. on the editing efficiency of DNA methylation editing systems [10].This expression pattern of emx2 strongly suggests the success of our editing system in methylating the promoter region of emx2.To our knowledge, this study presents the first DNA methylation editing system in marine fish cells through an in vitro approach, which holds significant importance in various applications by elucidating the functional importance of DNA methylation in gene repression and controlling cellular differentiation status.
Given the widespread implementation of epigenetic editing, it is crucial to assess the specificity of chromatin editors, especially with CRISPR technologies, considering their potential to induce off-target effects [31].To eliminate potential aberrant chromatin modifications at unrelated genomic loci, we screened the four most efficient potential off-target genes in the genome of Chinese tongue sole.These genes are based on Vanja Tadić et al.'s study about dCas9-based tools of off-target efficiency of the review, in accordance with the factors causing off-target to screen [32].Our results show a conserved percentual methylation level at the promoter region of transfected cells compared to the control (Figure 4A-D) for all three guides.Despite the elevated levels of CpG methylation at untargeted sites throughout the genome of different species with dCas-fused DNMT3A reported in the literature [33], our results with the species-specific dCas9-dnmt3a/sgRNA system show low to no occurrence of off-target methylation, as also reported in some studies [34].Furthermore, we also measured the expression levels of these genes to corroborate our off-target findings.Except for xpa expression following dCas9-dnmt3a/sgRNAs3 transfection (Figure 4H), no other gene modulation was observed (Figure 4A-G), confirming the high specificity of our dCas9-dnmt3a/sgRNAs3 system.Since no methylation levels were altered in the xpa promoter region (Figure 4D), we suggest that the downregulation of myc expression is not directly mediated by changes in DNA methylation levels within the xpa promoter region.These findings imply the existence of alternative regulatory mechanisms governing the relationship between emx2 and myc expression.Further investigation is warranted to elucidate these alternative regulatory pathways, which may involve transcription factors, chromatin remodeling complexes, or other epigenetic modifications.
In this study, we also investigated the influence of dCas-fused dnmt3a/sgRNAs' epigenetic effects in downstream genes and pathways related to Chinese tongue sole sexual reversal.Our findings indicated that the expression levels of klf4 and wnt1 remained stable following the methylation of the promoter region of emx2.In contrast, a notable downregulation of myc expression was observed (Figure 5).As the target gene of emx2, the altered expression of myc further verified the effectiveness of DNA methylation editing of the emx2 gene.In addition, after editing for DNA methylation, testis cells' mortality rate significantly increased (~30%) in the editing group, blank control group, and dCas9-dnmt3a with the non-sgRNA group without changing the number of cells significantly.We speculate that this may be caused by myc-regulated pathways, which we will further explore in subsequent studies.
In the context of fish, the inheritability of epigenetic editing technology remains a crucial topic for disease treatment, quality enhancement, and ornamental purposes.Therefore, further studies are needed to improve the editing efficiency, allowing for the practical application of epigenetic editing technology in fish.Taken together, we developed the first DNA methylation editing system in marine species and demonstrated its effective editing ability in Chinese tongue sole cells.This provides a new strategy for both epigenetic research and molecular breeding of marine species.

Experimental Fish Preparation
One-year-old Chinese tongue sole were obtained from Qingdao, China, and acclimatized to laboratory conditions for 48 h.Genotypic females and males were distinguished using specific markers [35], and physiological sex was confirmed post-dissection.Only male samples were retained for experimentation.Tissue excision was performed under MS222 anesthesia to minimize discomfort, adhering to Institutional Animal Care and Use Committee (IACUC) regulations (Approval No. YSFRI-2024011).

Cell Culture
One part of the gonad tissue was acquired from the Chinese tongue sole of three individuals and immediately placed in Leibovit's L-15 with L-glutamine (Solarbio, LA9510, Beijing, China).Others were placed in EP tubes, quickly frozen in liquid nitrogen, and then frozen for long-time storage at −80 • C until DNA and RNA extraction.The gonad tissues were minced into small pieces and washed three times using a serum-free medium to remove attached blood and other impurities in a biosafety cabinet.Subsequently, pieces of gonad tissues were digested with 0.25% trypsin-EDTA (Gibco™, 25200072, Thermo Fisher Scientific, Waltham, MA, USA).The degree of cell digestion was observed under a microscope.The complete medium was added to terminate the digestion.Cells were seeded in cell culture flasks and transferred to the 24 • C cell culture incubator.About one week later, the growing adherent cells were observed under a microscope.The cells were incubated at 24 • C and replenished with fresh medium every 3 days.After 10 days, the cells migrated and grew into confluent monolayers and were subcultured by 0.25% trypsin-EDTA at a 1:2 split ratio.Cells grew into a confluent monolayer again after 3 days [36].We used semi-quantitative RT-PCR to determine the predominant cell types in monolayer cells.We examined the gonadal somatic cell marker genes (foxl2, sox9, and wt1a) and germ cell marker genes (dmrt1 and vasa) [37][38][39][40][41], the RT-PCR results were confirmed in at least 3 batches of independent experiments, and representative results are shown (Figure S9).The cells were used for subsequent experiments after passage 5.

In Vitro Treatment with 5-Aza-dC
5-Azacitidine-2 ′ -deoxycytidine (5-Aza-dC, Macklin ® , C10749023, Shanghai, China) was dissolved in 100% dimethyl sulfoxide (DMSO) to create a stock concentration of 10 5 µM, which was stored at 4 • C. The stock solutions were then diluted with culture medium to obtain concentrations of 10 µM, 20 µM, and 40 µM.Cells were seeded in 6-well plates and allowed to grow overnight in a growth medium.Subsequently, cells were treated with various concentrations of 5-Aza-dC for 2 days before being isolated for further DNA and RNA extraction.

Molecular Cloning of dnmt3a and CRISPR/dCas9 Vector Construction
RNA from 1-year-old Chinese tongue sole was extracted and quantitatively reversetranscribed using the HiScript ® III 1st Strand cDNA Synthesis Kit (+gDNA wiper) (Vazyme #R312, Nanjing, China), following the instruction manual.The amplification primers were designed according to the dnmt3a X1 sequence of Chinese tongue sole from NCBI (Gene ID: 103380388) [42,43] (Table S1).The PCR cloning reactions were performed by using the high-fidelity enzymes PrimeSTAR ® Max DNA Polymerase (TAKARA, R045Q, Osaka, Japan).The reaction system of PCR was 50 µL, containing 25 µL of PrimeSTAR Max Premix (2×), 2 µL of cDNA, 1.5 µL of primer F, 1.5 µL of primer R, and 20 µL of DEPC water.The PCR program was as follows: 5 min at 95 • C for predenaturation, followed by 35 cycles of 95 • C for 15 s, 60 • C for 1 min, and 72 • C for 15 s.The PCR products were subjected to 1% agarose gel electrophoresis to separate them, and bands of the desired sizes were excised, purified, and ligated with the pEAZY-T1 vector for Sanger sequencing.

The sgRNA Vectors' Construction
The online tool MethPrimer (https://www.urogene.org/methprimer/,accessed on 10 April 2023) was used to analyze the size and locations of CpG islands of the emx2 promoter region [44].With the location of the CpG Island and TSS of emx2, six sgR-NAs at different positions were designed by using the online website tool CRISPOR (http://crispor.gi.ucsc.edu/,accessed on 8 May 2023).We used a reverse PCR method to construct the designed forward and reverse primers into the plasmid pGL3-U6-sgRNA-PGK (Addgene plasmid # 51133).The reaction system of PCR was 20 µL, containing 10 µL of KDO (2 × Master Mix, TOYOBO, Osaka, Japan), 1 µL of sgRNA plasmid, 0.5 µL of primer F, 0.5 µL of primer R, and 8 µL of DEPC water.The PCR program was as follows: 5 min at 95 • C for predenaturation, followed by 35 cycles of 95 • C for 15 s, 50 • C for 1 min, and 72 • C for 15 s.Primer sequences are described in detail in Table S2.Plasmid was extracted with a TAKARA plasmid extraction kit without endotoxin (TIANGEN, DP118) and stored at −20 • C.

Cell Transfection
The cells were seeded in growth medium in 6-well plates and incubated overnight.Experiments were conducted when the cell density reached 80% and replenished with fresh medium with 50% serum reduced.To assess potential off-target effects of sgRNA-guided dCas9-dnmt3a, two control groups were established: a blank control and a dCas9-dnmt3a with a non-sgRNA control, each with three replicates.We used Lipofectamine™ 3000 (Thermo Fisher, Waltham, MA, USA) for transfection by liposome-mediated transfection.Each well was transfected with 1000 ng of both sgRNAs and dCas9-dnmt3a plasmids.We also transfected an additional 500 ng of plasmid expressing green fluorescent protein (GFP) to characterize transfection efficiency.Following transfection for approximately 6-8 h, the medium was reverted to full serum medium and cultured for three consecutive days, during which the growth status and fluorescence intensity of the cells were observed.

Total RNA and DNA Extraction
Half of the cells in each well were harvested by scraping with a cell scraper, and total cell DNA was extracted using the Vazyme FastPure Blood/Cell/Tissue/Bacteria DNA Isolation Mini Kit (Vazyme #dC112), following the manufacturer's instructions.After extraction, the DNA was dissolved in sterile enzyme-free water, stored at −20 • C, and long-term-stored at −80 • C refrigerator.Trizol reagent (Invitrogen 15596026CN, Thermo Fisher, Waltham, MA, USA) was added at a volume of 500 µL to each well, and total RNA was extracted using the Trizol manufacturer's protocol.The Trizol, chloroform, and isopropanol were mixed at a ratio of approximately 5:1:2.The resulting mixture was used for total RNA extraction.Subsequently, the extracted RNA was dissolved and diluted with water, and its concentration and quality were assessed.The RNA samples were then stored at −80 • C in the refrigerator.

RT-qPCR
A PrimeScript™ RT reagent Kit with gDNA Eraser (Perfect Real Time) was used for cDNA synthesis for reverse-transcriptase qPCR, following the instruction manual (Takara RR047A, Osaka, Japan).Primer sequences for qPCR were designed by NCBI Blast tool (Table S3).A 2 × QuantiFast SYBR Green RT-PCR Kit (Qiangen, Cat.No./ID: 204156, Shanghai, China) was used to amplify the target gene.Qualitative detection and melting curve analysis was performed by the LightCycler ® 96 Instrument (Roche LightCycler ® 96 Instrument, Basel, Switzerland), a precise 96-well real-time PCR device.The reaction system of qPCR was 20 µL, containing 10 µL of SYBR, 2 µL of cDNA, 1 µL of primer F, 1 µL of primer R, and 6 µL of DEPC water.Melting curves were used to determine amplification specificity.After the reaction, according to the internal genes (beta-actin) and Ct value of the target genes, the 2 −∆∆ Ct method was used to determine the relative expression level of target genes [45].All experiments were set up with three biological and three technical replicates.

Bioinformatic Analysis and Phylogenetic Tree Construction
DNAMAN was used to compare the sequencing sequence and the CDS sequence of dnmt3a X1 in the NCBI database, and the sequence information was analyzed.An NCBI online tool (NCBI Conserved Domain Search (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi,accessed on 3 March 2024)) was used to predict the dnmt3a domain of Chinese tongue sole with the help of NCBI's Conserved Domain Database [47][48][49].The amino acid sequences of closely related species of Chinese tongue sole and the model species at a key position in the evolution were selected for multiple sequence alignment and phylogenetic tree construction.Phylogenetic tree construction was performed using the IQ-TREE 2 (http://www.iqtree.org,last accessed on 6 February 2020) tool, and phylogenetic inference was performed using the maximum likelihood (ML) criterion.We ran 10,000 ultrafast bootstrap (UFBoot) replicates [50].Multiple sequence alignments were performed using the tool EMBL-EBI (https://www.ebi.ac.uk/, accessed on 3 March 2024) [51].

Conclusions
In this study, we validated the regulation of DNA methylation on emx2 expression with decitabine treatment in Chinese tongue sole testis cells.We developed a new DNA methylation editing system and demonstrated its ability for sequence-specific DNA methylation editing.It targets the emx2 promoter region and led to increased DNA methylation levels and decreased emx2 expression in Chinese tongue sole testis cells.DNA methylation editing significantly suppressed the expression of myc, one target gene of emx2.Furthermore, we assessed the off-target effects of dCas9-dnmt3a and confirmed no significant impact on the predicted off-target gene expression.Taken together, we developed the first DNA methylation editing system in marine species and demonstrated its effective editing ability in Chinese tongue sole cells.This provides a new strategy for both epigenetic research and molecular breeding of marine species.

Patents
This manuscript describes inventions covered in the Chinese patent application No. 202410463573.1 (C.S. and Y.S., inventors).

Figure 1 .
Figure 1.Effects of 5-Aza-dC-mediated demethylation on the emx2 promoter followed by qPCR gene expression.CpGs' methylation number was assessed for the (A) blank control and at three different concentrations of 5-Aza-dC treatment: (B) 10 µM, (C) 20 µM, and (D) 40 µM.(E) Percentage of DNA methylation levels at emx2 promoter region with different 5-Aza-dC concentrations.(F) Relative expression of emx2 with different concentrations of 5-Aza-dC.Filled (black) circles correspond to methylated CpGs; unfilled (white) circles correspond to unmethylated CpGs.Upper black lines indicate statistical differences among groups, while lines below the graph highlight the treated group compared to the control.The asterisks (*) are used to denote significant differences between groups using two-way ANOVA and Tukey's multiple comparisons.* p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 1 .
Figure 1.Effects of 5-Aza-dC-mediated demethylation on the emx2 promoter followed by qPCR gene expression.CpGs' methylation number was assessed for the (A) blank control and at three different concentrations of 5-Aza-dC treatment: (B) 10 µM, (C) 20 µM, and (D) 40 µM.(E) Percentage of DNA methylation levels at emx2 promoter region with different 5-Aza-dC concentrations.(F) Relative expression of emx2 with different concentrations of 5-Aza-dC.Filled (black) circles correspond to methylated CpGs; unfilled (white) circles correspond to unmethylated CpGs.Upper black lines indicate statistical differences among groups, while lines below the graph highlight the treated group compared to the control.The asterisks (*) are used to denote significant differences between groups using two-way ANOVA and Tukey's multiple comparisons.* p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 2 .
Figure 2. Schematic overview of the CRISPR/dCas9-dnmt3a editing system for the emx2 promoter region of Chinese tongue sole testis cells.(A) Edit system schematic diagram.(B) dCas9-dnmt3a plasmid structure diagram and sgRNA plasmid structure diagram.(C) The sequence of the sgR-NAs.(D) Bisulfite sequencing positions and sgRNA sites of emx2; lollipop chart on the left represents bisulfite sequencing of the CpGs' position.The arrows indicate the target sequence of the designed sgRNAs, with the direction of the arrows aligning with the target sequence.

Figure 2 .
Figure 2. Schematic overview of the CRISPR/dCas9-dnmt3a editing system for the emx2 promoter region of Chinese tongue sole testis cells.(A) Edit system schematic diagram.(B) dCas9-dnmt3a plasmid structure diagram and sgRNA plasmid structure diagram.(C) The sequence of the sgRNAs.(D) Bisulfite sequencing positions and sgRNA sites of emx2; lollipop chart on the left represents bisulfite sequencing of the CpGs' position.The arrows indicate the target sequence of the designed sgRNAs, with the direction of the arrows aligning with the target sequence.

Figure 4 .
Figure 4. Off-target effects of CRISPR/dCas9-dnmt3a system.DNA methylation of the promoter region of (A) slc20a2; (B) ddx17; (C) katb6 and (D) xpa and relative expression of potential off-target genes (E) slc20a2; (F) ddx17; (G) katb6 and (H) xpa.Upper black lines indicate statistical differencebetween the blank control and sgRNA3 groups, while lines below the graph highlight that the sgRNAs were co-transfected with the CRISPR/dCas9-dnmt3a vector.The asterisks (*) are used to denote significant differences between groups using two-way ANOVA and Tukey's multiple comparisons.*** p < 0.001.

Figure 5 .
Figure5.Effects on downstream genes after the emx2 promoter region methylated.Relative expression of downstream genes (A) myc, (B) klf4, and (C) wnt1.Upper black lines indicate statistical difference among blank control and sgRNA groups for myc expression, while lines below the graph highlight that the sgRNAs were co-transfected with the CRISPR/dCas9-dnmt3a vector.The asterisks (*) are used to denote significant differences between groups using two-way ANOVA and Tukey's multiple comparisons.*** p < 0.001.

Figure 4 .
Figure 4. Off-target effects of CRISPR/dCas9-dnmt3a system.DNA methylation of the promoter region of (A) slc20a2; (B) ddx17; (C) katb6 and (D) xpa and relative expression of potential off-target genes (E) slc20a2; (F) ddx17; (G) katb6 and (H) xpa.Upper black lines indicate statistical differencebetween the blank control and sgRNA3 groups, while lines below the graph highlight that the sgRNAs were co-transfected with the CRISPR/dCas9-dnmt3a vector.The asterisks (*) are used to denote significant differences between groups using two-way ANOVA and Tukey's multiple comparisons.*** p < 0.001.

Figure 4 .
Figure 4. Off-target effects of CRISPR/dCas9-dnmt3a system.DNA methylation of the promoter region of (A) slc20a2; (B) ddx17; (C) katb6 and (D) xpa and relative expression of potential off-target genes (E) slc20a2; (F) ddx17; (G) katb6 and (H) xpa.Upper black lines indicate statistical differencebetween the blank control and sgRNA3 groups, while lines below the graph highlight that the sgRNAs were co-transfected with the CRISPR/dCas9-dnmt3a vector.The asterisks (*) are used to denote significant differences between groups using two-way ANOVA and Tukey's multiple comparisons.*** p < 0.001.

Figure 5 .
Figure 5. Effects on downstream genes after the emx2 promoter region methylated.Relative expression of downstream genes (A) myc, (B) klf4, and (C) wnt1.Upper black lines indicate statistical difference among blank control and sgRNA groups for myc expression, while lines below the graph highlight that the sgRNAs were co-transfected with the CRISPR/dCas9-dnmt3a vector.The asterisks (*) are used to denote significant differences between groups using two-way ANOVA and Tukey's multiple comparisons.*** p < 0.001.

Figure 5 .
Figure 5. Effects on downstream genes after the emx2 promoter region methylated.Relative expression of downstream genes (A) myc, (B) klf4, and (C) wnt1.Upper black lines indicate statistical difference among blank control and sgRNA groups for myc expression, while lines below the graph highlight that the sgRNAs were co-transfected with the CRISPR/dCas9-dnmt3a vector.The asterisks (*) are used to denote significant differences between groups using two-way ANOVA and Tukey's multiple comparisons.*** p < 0.001.