Half Smooth Tongue Sole (Cynoglossus semilaevis) Under Low Salinity Stress Can Change Hepatic igf2 Expression Through DNA Methylation

LI Siping, HE Feng, WEN Haishen, SI Yufeng, LIU Mingyuan, HUANG Yajuan,and WU Shuxian

Half Smooth Tongue Sole () Under Low Salinity Stress Can Change HepaticExpression Through DNA Methylation

LI Siping, HE Feng*, WEN Haishen*, SI Yufeng, LIU Mingyuan, HUANG Yajuan,and WU Shuxian

,,,266003,

Salinity is a crucial environmental stress that severely affects fish growth and survival. Under environmental stress, DNA methylation plays an important role in gene expression and genome function. To better understand the epigenetic regulation mechanism ofunder low salinity stress, we analyzed the DNA methylation at 5’UTR, exon1, intron1, and exon2, and analyzed the relationship of DNA methylation with mRNA abundance as well as the special single CpG sites methylation patterns ofd. When exposed to low salinity, DNA methylation at 5’UTR and exon2 remained stable, while it was up-regulated firstly and then down-regulated at exon1 and intron1. Some single CpG sites ofassociated with low salinity, and most of these sites with significantly changed methylation levels (<0.05) are located in intron1 area. The discrepant variation of single CpG sites methylation levels andexpression further revealed that females and males showed different response to low salinity. Remarkably, the 38-CpG site of intron1 servers as a sexual marker. Additionally, our integrative analysis demonstrated that regional DNA ofmethylation had highly complex interplay on gene expression. The single CpG sites in intron1 were indispensable epigenetic markers under external environmental stress. Above all, to resist the low salinity stress, half smooth tongue sole liver can regulate the expression ofthrough methylation of CpG sites in intron1.

salinity; hepatic; DNA methylation; gene expression; half smooth tongue sole

1 Introduction

Salinity is one of the most basic physicochemical char- acteristics of aquaculture water. It rapidly changes and pul- satory fluctuates due to the influence of thermodynamic properties, such as heat capacity, vapor pressure, and den- sity (Kültz, 2015). In aquaculture industry, habitat salinity could change drastically because of the flooding associated with coastal rainstorms, changes of tide, or terrige- nous drainage (Drake,2013; Duggan., 2014). This abiotic stress causes indirect drought in physiology, and alter osmotic pressure regulation and metabolism, as well as biochemical processesandof the cells (Choi and An, 2008; Hasenbein., 2013; Nie., 2017). It stresses on fish growth, survival, development, and reproduction. Sometimes it even results in huge econo- mic losses. For its economic importance, salinity has been widely studied by aquaculture science (Boeuf and Payan, 2001; Jeremiah and Joseph, 2008; Eddie and Norman, 2009; Martins., 2014; Viviana.,2015). Most fish spe-ciesare adapted to tolerate some degree of salinity stress. However, stenohaline species have narrow salinity tolerance range (Wurts and Stickney, 1989; Kültz, 2015). For example,is relatively sensitive to low salinity ex- posure. To cope with salinity changes, some euryhaline teleost developed their complex and unique osmotic regu- lation mechanism to adjust themselves to a wide range of external salinities (Jung., 2012; Kültz, 2015). Like, it survives at salinity from 0 to 38, while its optimal salinity is 16–17 (Du., 2013; Zhang., 2017).In response to salinity variation, these fish species mediate digestive enzyme activities, physiological functions of hormones (GH, TH, and IGFs) and transmembrane proteins (calcium-sensing receptor and ion channels) to regain ho- meostasis of osmoregulation (Barton and Iwama, 1991; Mu- stafayev and Mekhtiev, 2008; Martins., 2014; Kültz, 2015). In addition, euryhaline fish species increase cell pro- liferation and remodel extensive epithelial of gills in response to salinity changes (Laurent and Dunel, 1980; Chre- tien and Pisam, 1986).

As a kind of typical euryhaline fish species, half smooth tongue sole () mainly distribute in Bohai Sea and Yellow Sea, China. It gains rapidly appeal as aquaculture candidate in China for its high commercial value and good taste.It lives in the aquatic environment with a salinity from 14 to 37, while the optimum salinity is 26 (Wang., 2003). Half smooth tongue sole is an optional model for studying salinity adaption. It has been proved that when subjected to different salinity conditions, half smooth tongue sole would obtain adaptability by chang- ing its gene expression, non-specific immunity enzymes activities, key carbohydrate metabolism enzymes, body com- position, and physical and chemical indexes of blood (Fang, 2013; He, 2016; Li., 2017a, 2017b; Si., 2018). Moreover, we have found that when subjected to low sa- linity of 15, the number of white blood cells, red blood cells, albumin, and alkaline phosphatase levels became significantly higher to enhance the immune function to adapt the low salinity environment (He, 2016). Increas- ing evidence have demonstrated the involvement of the GH/liver-IGF1 axis in osmoregulation when fish are under salinity stress. For example, the expression ofmRNA level was significantly deceased whileandmRNA levels increased, causing weight gain rate significantly decreased when half smooth tongue sole was under low salinity (Si, 2019). Meanwhile, our previous study al- so demonstrated that hepaticmRNA expression sig- nificantly decreased under long-term low salinity treatment (Li., 2017a). Karl(2010) found that both IGF1 and IGF2 in the liver crucially participated in fish os- moregulation with organ-specific manners in tilapia. Mean- while, IGF2 seemed to mimic the osmoregulation function of IGF1 with different manners when fish was under salinity stress (Reinecke and Collet, 1998; Reinecke, 2005; Codina,2008; Karl, 2010). Moreover, Nor- man. (2011) found that IGF2 may also affect salinity tolerance capacity as suggested by a genome-wide QTL on linkage group 19 in. However, the potential physiology role of IGF2 in fish osmoregulation still need further research.

Recently, the dynamic process of DNA methylation un- der environmental stresses has evoked the interest in epigenetic adaptive regulation in the abnormal conditions. When situating in the adversity environment, aquatic organisms depend on the gene expression and reprogramming metabolism to regain the physiology equilibrium for development, growth, and survival (Takei,2014; Yang,2016).DNA methylation plays important roles in gene expression and cellular differentiation, it may adjust genome function to adapt to changed environment (Szyf, 2012; Alvarado, 2014). Alterations in DNA methylation patterns induced by salt could enable hatchery- reared trout to acclimate to seawater conditions and increase their survival rate (Morán., 2013). Associated with gene silence or super-activity, DNA methylation can respond to stress quickly with a diversity without changing the DNA sequences (Habu, 2001; Bird, 2002). DNA methylation regulates gene expression by repressing the binding of transcriptional factors or incorporating some specific proteins into methylated CpG sites to modify chromatin structure (Ziller,2013). Anastasiadi(2017) found that temperature increase resulted in stage-dependent alterations in global DNA methylation and gene expression levels in European sea bass. In addition, we also had explored that DNA methylation of,2017a, 2017b).

From the aspect of how epigenetic regulation was involved in fish growth and adaptability under adverse environment, we investigated the methylation levels of regional(including 5’UTR, exon1, intron1, and exon2) in the liver and analyzed its function on mRNA expression, as well as some special single CpG sites variation when half smooth tongue sole were subjected to low salinity exposure. These results explain the function offish is under low salinity stress. Additionally, the analysis of differential methylation levels of four regional DNA and its correlation with mRNA expression provided additional evidence for the transcriptional regulation of genomic DNA. The methylation of specific single CpG sites in hepaticfurther improve our understanding of the function ofin regard of the epigenetic regulation under low salinity stress.

2 Materials and Methods

2.1 Animal Maintenance and Stressing

The 10-month old half smooth tongue sole () (body weight 101.58±32.07g; body length 25.39±3.52cm) were maintained in a local commercial fish farm. Three groups of fish were selected. In each groupforty healthy fish were randomly raised in a 5m×5m×1m tank filled with salinity 30 seawater. By mixing seawater with fresh well-water at the speed of decreasing 5 every day, the low salinity 15 environment was achieved (Wang,2003). The fish obtained from natural seawater (sa- linity 30), namely treated with salinity 15 for 0d (D), were defined as the control group. The fish further exposed to salinity 15 for 7d (D) and 60d (D) respectively were considered as treatment groups. The fish were cultured with appropriate temperature (22–24℃) and continuous aeration (DO>6mgL−1). Additionally, compound feed with 5% of body weight was used to feed fish twice a day before water was replaced to ensure the optimal aquaculture conditions. For each treatment, six individuals from each tank were anesthetized with 0.15% MS-222 (Sigma, St. Louis, MO) and the livers were collected. Three females and three males were used for further analysis in this study. The study was approved by the respective Animal Research and Ethics Committees of Ocean University of Chi- na. The field studies did not involve endangered or protected species.

2.2 Analysis of igf2 mRNA Abundance Under Stress Treatments by qRT-PCR

Total RNA from the liver of fish under different salinity stresses was extracted with RNAiso Reagent Kit. The PrimeScriptTMRT Reagent Kit with gDNA Eraser (Takara, Japan) was applied to synthesize the first-strand cDNA at 37℃ for 15min. Then(Accession: NC_024312.1) ex- pression was determined by qRT-PCR using specific pri- mers,-F and-R (Table1). Meanwhile, the(Accession: KP033459.1) (Table 1) was used as an endogenous reference (Liu., 2014). The specified amplification efficiency was ensured by conducting five 10- fold serial dilutions of cDNA to produce the standard cur- ves of cDNA samples.Then 1μL of 100× diluted cDNA template with three triplicates, as well as endogenous re- ference gene was simultaneously subjected to the Roche LightCycler 480 (Germany) to run the qRT-PCR process. The 2−∆∆CTmethod was applied to analyze the relative quantification (Livak and Schmittgen, 2001).

Table 1 Nucleotide sequences of primers used in the experiment

2.3 Analysis of igf2 DNA Methylation in Control and Experimental Groups by BSP Sequencing

Genomic DNA in the liver tissue was isolated by using Marine Animal Genomic DNA Kit (TransGen, China). The quality and integrity were guaranteed by the nucleic acid analyzer (Biodropsis BD-1000, OSTC, China) and 1.5% agarose gel. The BisulFlash DNA Modification Kit (EpiGentek, USA) could deaminate the unmethylated Cyto- sine (C) to become Uracil (U) while the methylated Cytosine (mC) remained unchanged. Therefore, we can distinguish the U from C to analyze the methylated C sites inby sequencing. One microgram of DNA was modified to serve as methylation-specific amplification templates. Five pairs of primers were specifically designed by online MethPrimer design software (http://www.urogene. org/methprimer/) to amplify the 5’UTR, exon1, intron1, and exon2 regions of, which are predicted CpG-rich regions. The PCR products were separated from agarose gel and purified by EasyPure Quick Gel Extraction Kit (Trans- Gen, China). Then the amplification fragments were cloned into a vector by pClone007 Simple Vector Kit (TsingKe, Beijing) and sequenced. In each sample, 7–10 individual clones were tested to present the CpG sites and the methylation level ofAll the primers used in the experiment are listed in Table 1.

2.4 The Prediction of Genetic Structure and CpG-Rich Regions

Thesequence was submitted to Splign software (http://www.ncbi.nlm.nih.gov/sutils/splign) and the four function areas, including 5’UTR, exon1, intron1 and exon2were analyzed. The MethPrimer design software (http://www.urogene.org/methprimer/) was employed to forecast the CpG- rich regions. Moreover, the open reading frame (ORF) was found by online ORF finder (http://www.ncbi.nlm.nih.gov/ gorf/orfig.cgi). The transcription factor and the binding site ofwere analyzed by PATCH™ public 1.0 online software (http://www.gene-regulation.com/cgi-bin/pub/programs/ patch/bin/patch.cgi).

2.5 Statistical Analysis

The data ofmRNA expression and methylation le- vels in the liver of half smooth tongue sole under salinity stress were analyzed by SPSS 19.0 software, followed by one-way ANOVA within Duncan’s multiple range test (≤0.05). Independenttest was performed to compare the difference between female and male fish. The Pearson cor- relation coefficient () was conducted to compute the cor- relation of the DNA methylation and the mRNA expression level.

3 Results

3.1 Prediction of igf2 Structure and CpG-Rich Regions

The structure schematic ofwas shown in Fig.1. With the length of 5727bp,contains four exons and three introns, and the initiation codon and termination codon are located at positions 141bp and 4957bp, respectively. The three predicated CpG islands inare 484bp, 210 bp, and 124bp in length and are embraced in 5’UTR, exon1,intron1 and exon2 areas. Analyzing the sequences of these four functional areas, there are 3, 3, 47, and 10 CpG sites, respectively. The open reading frame (ORF, locating at 1025–1165bp) encodes a 46-amino acid IGF-like superfamily domain. Additionally, as shown in Fig.2, a large number of forecasted transcription factors are gathered around the CpG sites, including GR, Sp1, AP-1, and NF. They play im- portant roles on the cellular proliferation and differentiation, metabolism and cell apoptosis, as well as regulating organism development and immunization.

Fig.1 Structure distribution of igf2 within BSP-amplifica- tion. The igf2 gene contains four exons (blue frame) and three introns (black line); the initiation and termination codons were located at 141bp and 4957bp, respectively, labeled in red font. The yellow frame exhibits the amplified igf2 with 5’UTR, exon1, intron1 and exon2, with threepredicated CpG islands depicted in blue boxes. The red box indicates the open reading frame (ORF), locating 1025– 1165bp.

Fig.2 The CpG sites and forecasted transcription factors binding sites on the four functional areas of igf2. The underlined red letters show the CpG dinucleotide sites on 5’UTR, eoxn1, intron1 and exon2. The stained frames indicate the binding sequences of forecasted transcription factors with transcription factors marked upon it.

3.2 The Changes of igf2 DNA Methylation in the Liver Under Salinity Stress

As a particular indicator, DNA methylation of CpG sites is susceptible and dynamic to the external environment (Szyf, 2012). Since the predicted CpG islands main- ly covered the 5’UTR, exon1, intron1, and exon2 regions, we carried out BSP sequencing to analyze the methylation levels of these four functional areas to investigate how the methylation level ofDNA specifically responded to low salinity stress. Considering the adult females are two to three times larger than male ones (Chen, 2007), we analyzed the females and males respectively to eliminate the sexual difference. Interestingly, the DNA methylation levels among 5’UTR, exon1, intron1, and exon2 ofhad no significant difference. Meanwhile, the methylation levels of four functional areas showed the same change tendency. They all increased firstly and then decreased when fish were exposed to short- and long-term low salinity stresses.

In female fish, when under normal salinity environment, the CpG methylation status of the four functional areas was quite low, averaging between 4.95% and 8.89% with no significant difference (Fig.3A,>0.05). Subjected to low salinity stress (Fig.4A), 5’UTR CpG methylation le- vels increased initially and then decreased at a certain le- vel (>0.05), which were 8.89%, 23.33%, and 6.67%, res- pectively. Exon1 methylation ratio was 5.55% under low salinity at the beginning (0D), and then significantly up- regulated to 23.33% (<0.05) under 7D stress, followed by a dramatical decline under 60D stress (<0.05). The CpG methylation in intron1 significantly raised (<0.05) and then reduced along with stress time prolonged (4.95%, 20.75%, and 10.00%, respectively). Subjected to salinity 15, the methylation level of exon2 was 8.33% at 0D, 8.81% at 7D and 6.33% at 60D, showing a slightly upper and then lower changes with no significance (>0.05). In ge- neral, the methylation level ofwas 5.70% under at 0D, significantly increased to 18.99% (7D,<0.05) and then gently decreased to 8.99% (60D,>0.05).

In male fish, the methylation levels of all these four regionalhad no difference, which were 3.33%, 2.22%, 4.74% and 6.22%, respectively (Fig.3B,>0.05). In Fig.4B, 5’UTR methylation occupied 3.33% under 15 salinity at 0D. Then it increased to 5.33% and then recovered to 3.33%(>0.05). When the time of stress prolonged, the methylation status of exon1 showed a significant increase followed by a significant decrease (<0.05), presenting 2.22%, 12.22%, and 3.33% respectively. Intron1 had the methylation level of 4.74% on day 0, 9.50% on day 7 (<0.05) and 4.16% (>0.05) on day 60 with the consistent tendency of firstly rising and then dropping. There was litter discrepancy of exon2 methylation under stress (>0.05), which were 6.22%, 8.67%, and 7.33% respectively. Above all, with 0D salinity stress, the methylation level ofwas 4.78%; when the stress prolonged to 7D and 60D, it apparently was up-regulated to 9.21% and down-regulate to 4.60% (<0.05) with a recovery.

Fig.3 The presentation of the CpG methylation levels in four function areas of igf2 in the liver of untreated females (A) and males (B). One fish representative of the methylation level is demonstrated. A filled or open circle indicate CpG positions methylated or unmethylated in the CpG site, respectively. Ten clones per fish were used to determine the average methylation levels, which was specifically calculated outside the bar. Data represents mean±SD, P>0.05 shows no significant differences between function areas with Duncan’s t test.

Fig.4 The changes of igf2 methylation in female (A) and male (B) half smooth tongue sole under short-(7D) and long- (60D) term of low salinity stress. Different lowercase letters indicate the significances when subjected to stress (P<0.05, Duncan’s t test).

3.3 The Influence of Low Salinity Stress on Single CpG Site of igf2

Furthermore, the single CpG sites ofwere found to be sensitive to the low salinity stress. When we analyzed the methylation of single CpG sites (Fig.5), we found they were quite constant in 5’UTR area, while exhibited dramatic changes in exon1, intron1, and exon2 areas under treatments. Overall, the methylation levels of these significantly changed CpG sites mainly displayed firstly in-creased under stress for 7D and then decreased to recover under stress for 60D.

In female fish (Fig.5A), low salinity stress had a great effect on some particular single CpG sites in exon1 (two CpG sites), intron1 (12 CpG sites) and exon2 (one CpG site) of. The methylation level of the 37-CpG site in intron1 exhibited a continuous rising when subjected to salinity stress and showed a significant change with 60D stress (<0.05). Moreover, the other CpG sites all exhibited the same methylation tendency that firstly upgraded and then descended with significant difference (<0.05). Particularly in intron region, the methylation levels of 2-, 21-, 27-, 39-, 43- and 45-CpG sites were all 0.00% under 15 salinity at 0D. And the methylation levels of 1-, 21-, 31-, 35-, 39- and 40-CpG sites declined to 0.00% after 60D under low salinity stress.

In male fish (Fig.5B), five CpG sites in intron1 altered under stress, while the low salinity stress showed no influence on the methylation of CpG sites in 5’UTR, exon1, and exon2. The methylation of 1-, 5-, 8- and 11-CpG sites in intron1 all increased firstly and then declined with significant differences (<0.05). The methylation level of 38- CpG site kept continually declined to 0.00% under stress for 7D and 60D (<0.05). Additionally, the methylation le- vels of 5- and 8-CpG sites presented 0.00% at 0D of stress.

Fig.5 The effect of low salinity on the methylation of igf2 single CpG sites in female (A) and male (B) half smooth tongue sole. Different lowercase letters indicate the signi- ficances when subjected to stress (P<0.05, Duncan’s t test).

3.4 The igf2 mRNA Abundance Under Stress and Its Correlation with DNA Methylation

Surveys of epigenetic markers had proved that DNA me- thylation participates in transcriptional regulation (Ansel., 2006). However, the correlation of regional DNA methylation levels and mRNA expression were rarely dis- cussed. To test whether this epigenetic mechanism evokes plastic transcriptional programs of, especially the re- lationship of methylation level with mRNA expression, when half smooth tongue sole was under low salinity stress, we quantified themRNA expression by RT-qPCR tech- nology. Its correlation with the methylation level was fi- gured out by correlation coefficient () analysis. ThemRNA expression (Fig.6) significantly declined as low sa-linity stress prolonged. Compared with salinity 15 at 0D, themRNA expression levels in female fish liver significantly decreased to 3.6 times (<0.05) at 7D and 5.1 times (<0.05) at 60D, respectively. In male fish liver,mRNA expression significantly dropped under 7D stress, which was 2.5 times (<0.05) less than at 0D. When sub- jected to low salinity for 60D, the relative quantity slight- ly raised, which was 2.5 times (<0.05) less than for 0D, given the significance of DNA methylation in regulating gene expression we tested.

Fig.6 The relative expression of igf2 in half smooth tongue sole under salinity stress (salinity 15). Different lowercase letters indicate the significances when subjected to stress (P<0.05, s (P<0.05, Ducan’s t test).

In female fish liver (Figs.7A and B), the CpG methylation levels of 5’UTR, exon1, intron1, andnegatively correlated with mRNA expression, were −0.299, −0.319, −0.425, and −0.239, respectively. Among them, intron1 oc- cupied the greatest correlation coefficient. However, thebetween exon2 methylation and mRNA expression presented a positive relationship with 0.239. In male fish liver (Figs.7C and D), 5’UTR, exon1, intron1, exon2, andshowed a negative correlation with mRNA transcription with−0.126, −0.588, −0.313, −0.188, and −0.341, respectively. And theof exon1 exceeded other functional DNA areas.

3.5 Analysis of Gender Difference of igf2 and Single CpG Site in the Liver of Stressed Fish

Half smooth tongue sole exhibits significant sexual di- morphism as females are two to three times larger than males (Chen, 2007). Thus, it is necessary to analyze the sexual difference. In this study, the analysis manifested that the DNA methylation levels ofat its four functional areas (5’UTR, exon1, intron1, and exon2) in the liver under salinity stress had no gender difference (Table 2).Although theexpression under low salinity stress showed no significant sex difference under salinity 15 for 0D (>0.05), dramatic differences were observed under stress for 7D and 60D, as theexpression in the liver of male fish was significantly higher than female fish to 2 times (<0.05) and 3.3 times (<0.05), respectively.

We also detected some single CpG sites which showed significantly different methylation levels between stressed female and male fish (Fig.8). Under salinity 15 for 0D, the methylation level in males significantly exceeded females at the 38-CpG site of intron1 (<0.01). For 7D stress, the 2-CpG of 5’UTR, 20-, 27-, 35-, 43- and 45- CpG of intron1 and the 2-CpG of exon2 all had a significantly higher methylation status in female fish than in male fish (<0.05). As time prolonged to 60D, only the 37-CpG of intron1 hadan evident difference in methylation and females were sig- nificantly higher than in males (<0.05). Thus, the results implied thatmight play different osmotic-regulation roles in females and males.

Fig.7 The correlation between CpG methylation and mRNA expression of igf2 in female and male fish when exposed to low salinity treatments (salinity 15). A and C: The correlation between the four functional areas (5’UTR, exon1, intron, and exon2) methylation levels with igf2 mRNA expression in female (A) and male (C) fish. B and D: The correlation between igf2 DNA methylation level with igf2 mRNA expression in female (B) and male (D) fish.

Table 2 Comparisons of methylation and expression levels of igf2 between female and male fish under low salinity treatments

Notes: The difference between females and males in regards of methylation levels of 5’UTR, exon1, intron1, exon2 and total, as well as expression level ofmRNA under different salinity treatments are presented in the table. Data are expressed as mean±SD; different lowercase (a and b) or uppercase (A and B) letters represent significance difference between females and males when subjected to different salinity stress (<0.05, t test).

Fig.8 The significant difference between female and male fish in single CpG site under salinity stress. * indicates the sig- nificances when subjected to stress (* <0.05, ** <0.01, t test).

4 Discussion

Salinity is an important environmental stimulating factor. It irreversibly affects physiological and biochemical functions, endocrine systems, molecular receptor of osmo- tic pressure, as well as the stability of DNA and proteins in organism to affect fish growth, reproduction and ecological distribution (Barton and Iwama, 1991; Kültz, 2012, 2015; Martins,2014).By regulating different gene expressions for adjusting genomic functions, the potential role of DNA methylation in adaptation mechanism under environment contexts has been proved to be important (Szyf, 2012). As a commercial euryhaline marine fish in China, half smooth tongue sole significantly changed DNA methylation levels and gene expression of GH-IGF axis genesto adjust itself to low salinity stress (Li,2017a, 2017b; Si, 2019). However, how DNA methylation of heptic IGF2 changes and how it regulates mRNA expression under low salinity stress has not studied thoroughly. Based on the background, we carried out the analysis to figure out how regional DNA methylation ofregulates gene expression to serve as genome adaptation mechanism when half smooth tongue sole is exposed to low salinity.

4.1 The Diversity Characteristics of DNA Methylation in Four Functional Areas of igf2

DNA methylation has well established its role in tissue differentiation and development as a mechanism for the genome to express diverse phenotypes (Choe, 2008; Szyf, 2012). Additionally, different regional DNA presented its specific methylation pattern, which has been proved to greatly affect gene expression (David, 2014). Structure prediction showedcontains three CpG islands. For 5’UTR, exon1, intron1, and exon2 areas of, there were 3, 3, 47, and 10 CpG sites, respectively. The analysis indicated that the methylation levels offour functional areas were polymorphic in the liver of half smooth tongue sole. It is relatively higher in exon1 and is lower in exon2, which may correlate with the different abundances of CpG dinucleotide methylated. Based on the nine cell lines and 15 tissues of human methylation data, Ai(2016) found that different functional regions had diverse methylation distributions: first exon region kept in lower state, transcription start site region fluctuated within a certain range, while the medium exon, last exon, intron, and transcription termination site had comparative higher methylation levels. Our previous study also had manifested that methylation in three exons ofwas significantly different with exon1 prominently lower than exon2 and exon3 (Li, 2017a). DNA methylated occurs in all crea- tures and expresses the fundamental dynamics of epigenomes. The diversity of methylation states across individuals resulted in the plasticity, tissue-specific nature, as well as the variability of gene expression (Weiss., 1996; Yano.,2003; Christensen.,2009).

4.2 The Dynamic Response of Methylation and mRNA Abundance of igf2 in Liver to Salinity Treatments

As a crucial environmental factor, salinity greatly alters osmotic pressure regulation to threaten fish biochemical processes (Boeuf and Payan, 2001). To adapt to this specific environmental cue, the organism gains its ability to maintain tissue-specific function and regulate the gene transcriptional patterns (Alvarado., 2014). Compared with sequences adjustment, DNA methylation can quickly respond to environmental stimulus as it methylates different fragments to change gene expression quantity or activate new functional genes in specific environmental cues (Steward,2002; Mazzucotelli, 2008; Ou,2009). The dynamic response of DNA methylation had emphasized its integral role as an important adaptive mechanism under environment changes (Bird, 1986; Ha- shida,2006; Zhao,2009).

The four functional areas showed dynamic methylation levels in response to low salinity stress. As we have manifested, the methylation levels of exon1 and intron1 elevated significantly under 7D stress with mRNA expression significantly decreased. As low salinity prolonged to 60D, methylation levels gradually decreased to the original level, serving as a genome adaptation method. Mean- while, the conservation of DNA methylation suggests to be indispensable in maintaining the genomic stability as well (Alvarado,2014). In our study, the methylation levels of 5’UTR and exon2 ofhad no significant difference under low salinity stress. The single CpG sites, especially in 5’UTR, presenting no significant changes un- der treatments. These GC enrichment regions remained steady under stress which may involve in maintaining ge- nomic stability to ensure the fish survival and growth. Studies had provided that the loss of DNA methylation could result in the genomic instability and chromosomal aberrations which observed in cancer (Gama-Sosa, 1983; Eden,2003). We speculated the normal methylation level of 5’UTR may be integral forgene expression. As for exon2 methylation, the higher methylated CpG dinucleotide made it difficult to remethylate.Additionally, previous work indicated that the full-methy- lation in three tissues (gonad, kidney, and gill) under low salinity stress steadily ranged from 12.06% to 18.64% with- out significant change, which was considered to be necessary for the stability of gene expression(Li, 2017b).

Interestingly, by analyzing the methylation levels of to- tal 64 CpG sites in three CpG islands of, the results showed that the significant changes of single CpG sites under salinity stress were found in intron1 region. When studied the DNA methylation changes of rice under drought stress, Pan(2009) verified that both the coding and non-coding regions had the similar frequency of DNA me- thylation. Regarding the mutability of the length and sequence of intron fragments, Wang andLiu(2000) pointed out that most of the mutants occur in intron regions randomly, and intron can diminish the impacton organism toenhance the resistant ability of mutations.In our study, the significant changes of methylation in intron1, especially the single CpG sites further implied the unique adjustment ability and regulation of fish under salinity stress. Moreover, the variation tendency of single CpG sites methylation levels contributed to the functional areas methylation levels, causing the methylation level ofincreases firstly and then decreased.

DeChiara(1990) had illuminated the physiological role of IGF2 in cellular proliferation and differentiation, mediating metabolism and inhibiting apoptosis. Incon- siderationofits important role, the balanced expression ofis imperative for fish survival and growth.Karl(2010)had revealed theandmRNA expressions in tilapia were downregulated in parallel after seawater transfer and recovered ultimately, suggesting thatandare involved in fish osmoregulation with organ-spe- cific manner.In our study, as methylation ofhad no changes under 7D stress and significantly increased under 60D low salinity (Li,2017a), DNA methylation ofcould respond to low salinity more quickly and recoverits methylation level as stress time prolonged. Thus, thefunction regarding epigenetic adaptive mechanism ofhas superior timeliness to.Although half smooth tongue tolerates a certain degree of salinity stress, the low salinity down-regulates the expression ofwhen the time of stress becomes longer. Wang(2003) found that the growth rate was affected when half smooth ton- gue sole subjected to a lower (for example, 15 or 18) or higher salinity (for example, 32 or 35).Our previous studies also found that the expression ofandalso decreased and resulting in lower weight gain rate when half smooth tongue was subjected to 15 salinity for 60D (Li, 2017b; Si, 2019).

4.3 The Sexual Difference of igf2 DNA Methylation and mRNA Expression Levels

Half smooth tongue sole is an important commercial fish in China which exhibits significant sexual dimorphism as females are two to three times larger than males(Chen, 2007). The studies focused on the gonadal develop- ment and sexdetermination mechanism had been widely explored. Sex-related genes, including,,,,,andhad been characterized during gonadal differentiation (Deng,2009; Dong, 2011; Hu and Chen, 2013; Hu., 2014). Interestingly, methylation modification appeared to be indispensable for sex determinationby regulating gene expression (Shao,2014). In our study, the methylation levels in four functional areas ofhad no sexual difference under normal salinity environment, indicatingis not a sexdetermination gene.Interestingly,the DNA methylation in males significantly exceeded females in the intron1 38-CpG siteunder normal salinity environment,which could be considered as a candidate marker to character female and male half smooth tongue sole during breeding. When exposed to low salinity, the DNA methylation ofhad no difference between stressed male and female fish.

However, the number and the site of single CpG which had significantly changed methylation levels are different between female and male fish under salinity stress. Remarkably, we observed that the methylationlevels of single CpG sites under low salinity treatment have significant differencebetween sexes.Meanwhile, theexpression levels in males significantly exceed those in females under low salinity for 7D and 60D. These differences may reveal that female and male fish have different epigenetic responses to low salinity.Additionally, the sen- sibility of different methylated CpG sitesto low salinitymay be different in female and male fish. Furthermore,expression level showed the opposite change asfemales wassignificantly higher than that in males under salinity stress for 7D and 60D(Li,2017a), su- ggesting the discrepantexpressionofandthey are subjected to low salinity stress.

4.4 The Correlation Between DNA Methylation and Gene Expression

The role of DNA methylation has been extensively cha- racterized in altering transcription pattern temporally and spatially.Mechanistically, it was reported to cause steric hindrance that inhibited transcriptional activators binding to the DNA thus repressing gene expression particularly. Moreover, DNA methylation was suggested to combine withtranscriptional repressors such as methyl-binding do- main family. These complexes could recruit histone modi- fying complexes to silence the transcription(Razin and Riggs, 1980; Choy, 2010; Alvarado, 2014).Most work has focused on the importance of promoter region inrepressing gene expression. As the studies of regional DNA in regulating gene expression get deeper, the function of DNA methylation at different functional areas has attract- ed more attention. Given the comprehensive DNA methy- lation map of the entire genome in, Zhang(2006)reported that over one-third of expressed genes had methylated sites within transcribed regions and only approximately 5% of genes contain methylation sites within promoter regions.In addition, the high methylation ratio of exon1 involved in the inactivation of AR gene in Hela cells (Li,2011). Based on the background, we analyzed the correlation between DNA methylation levels in four regions and mRNA expression. The results showed that the methylation level in exon2 was positive with gene expression, while other regions presented a negative correlation with higher r values in exon1 and intron1. When Eckhardt(2006) explored the DNA methylation pro- file of human chromosomes 6, 20 and 22, they discovered that one-third of the differentially methylated sites in 5’UTR were inversely correlated with gene transcription. Brenet(2011) found that methylated exon1 region was tight- ly linked with transcriptional depression, much more than the promoter region. In our study, the methylation level of exon2 in female fish showed the positive correlation with transcription, which was consistent with the previous result that higher methylation in intragenic CpG islands led to active genes expression (Jjingo, 2012). Xiao(2014) also observed the positive correlation between intron2 methylation and expression ofgenein the subcutaneous adipose tissue and greater omentum of, while the DNA methylation might alter the chromosome structure to promote the gene transcription (Lorincz, 2004; Jjingo, 2012; Kulis, 2013).

Salinity is a crucial environmental stimulus to stress an irreversible influence on fish growth, reproduction and eco- logical distribution (Martins, 2014). In response to salinity stimulus, DNA methylation occurs at different frag- ments to control gene expression as an important adaptive mechanism (Bird, 1986; Hashida,2006; Zhao, 2009). We had manifested the genes of GH-IGF axis chang- ed their expressions through DNA methylation when half smooth tongue sole was under low salinity stress. With a high identity of IGF1, IGF2 plays a role in the osmoregulation with a different manner (Reinecke and Collet, 1998; Reinecke,2005; Codina,2008; Karl, 2010).To further understand theosmoregulation function from the aspect of DNA methylation and mRNA expression, we sought to identify the different functional regulation of regional DNA methylation on mRNA expression and some special single CpG sites ofin the liver of half smooth tongue sole that was associated with low salinity. The 5’UTR and exon2 hold the stability of methylation statues under stress, and the significant methylation happened on single CpG sites of intron. The discrepant variation of single CpG sites methylation levels andexpression in female and male fish under salinity stress further reveal that female and male fish respond to low salinity variously. However,was not a sex determination gene as no methylation or expression difference was found between female and male half smooth tongue sole. Remarkably, the methylation levels of intron1 38-CpG could be considered as a candidate marker to differentiate gender during breeding. The exon2 in females showed a posi- tive correlation between DNA methylation and mRNA ex- pression, while DNA methylation in other regions showed negative functions ontranscription.

Overall, our study firstly indicated that by changing regional DNA methylation and mRNA expression,plays a complicated function in response to low salinity stress in half smooth tongue sole liver. More work still needs to be conducted to further understand the modification me- chanism of cytidine analog inhibitor of DNA methyltrans- ferases (DNAMTs) on DNA methylation under stress.

Acknowledgements

We thank Shandong Dongying Farm for providing the animals. This research was supported by the Key Laboratory of Mariculture of Ministry of Education, Ocean Uni- versity of China (No. KLM2018009) and the National Na- tural Science Foundation of China (No. 31672642). Editors and reviewers are highly appreciated for their comments to greatly improve our manuscript.

Ai, X., Zhang, Y. F., Liu, J. J., and Zhang, L. R., 2016. Analysis of DNA methylation in gene functional regions., 47 (3): 290-299 (in Chinese with English abstract).

Alvarado, S., Fernald, R. D., Storey, K. B., and Szyf, M., 2014. The dynamic nature of DNA methylation: A role in response to social and seasonal variation., 54 (1): 68-76.

Anastasiadi, D., Díaz, N., and Piferrer, F., 2017. Small ocean tem- perature increases elicit stage-dependent changes in DNA me- thylation and gene expression in a fish, the European sea bass., 7 (1): 12401.

Ansel, K. M., Lee, D. U., and Rao, A., 2003. An epigenetic viewof helper T cell differentiation., 4 (7): 616- 623.

Barton, B. A., and Iwama, G. K., 1991. Physiological changes in fish from stress in aquaculture with emphasis on the response and effects of corticosteroids., 1: 3-26.

Bird, A. P., 1986. CpG-rich island and the function of DNA me- thylation., 321 (2): 209.

Bird, A. P., 2002. DNA methylation patterns and epigenetic me- mory., 16: 6-21.

Boeuf, G., and Payan, P., 2001. How should salinity influence fish growth?–, 10 (4): 411-423.

Brenet, F., Moh, M., Funk, P., Feierstein, E., Viale, A. J., Socci, N. D., and Scandura, J. M., 2011. DNA methylation of the first exon is tightly linked to transcriptional silencing., 6 (1): e14524.

Chen, S. L., Li, J., Deng, S. P., Tian, Y. S., Wang, Q. Y., Zhuang, Z. M., Sha, Z. X., and Xu, J. Y., 2007. Isolation of female- specific AFLP markers and molecular identification of geno- typic sex in half-smooth tongue sole ()., 9: 273-280.

Christensen, B. C., Houseman, E. A., Marsit, C. J., Zheng, S. C., Wrensch, M. R., Wiemels, J. L., Nelson, H. H., Karagas, M. R., Padbury, J. F., Bueno, R., Sugarbaker, D. J., Yeh, R. F., Wiencke, J. K., and Kelsey, K. T., 2009. Aging and environmental exposures alter tissue-specific DNA methylation dependent upon CpG island context., 5 (8): e100 0602.

Codina, M., García de la Serrana, D., Sánchez-Gurmaches, J., Montserrat, N., Chistyakova, O., Navarro, I., and Gutiérrez, J., 2008. Metabolic and mitogenic effects of IGF-II in rainbow trout () myocytes in culture and the role of IGF-II in the PI3K/Akt and MAPK signaling pathways., 157: 116-124.

Choe, J., 2008. DNA Methylation in development., 5: 100-104.

Choi, C. Y., and An, K. W., 2008. Cloning and expression of Na+/ K+-ATPase and osmotic stress transcription factor 1 mRNAin black porgy,during osmotic stress.–, 149 (1): 91-100.

Choy, J. S., Wei, S., Lee, J. Y., Tan, S., Chu, S., and Lee, T. H., 2010. DNA methylation increases nucleosome compaction andrigidity., 132: 1782- 1783.

Chretien, M., and Pisam, M., 1986. Cell renewal and differentia- tion in the gill epithelium of fresh- or salt-water adapted euryhaline fish as revealed by [3H]-thymidine radioautography., 56: 137-150.

David, M. M., Hatice, O. A., Ryan, M. M., Peter, A. K., and Lawrence, C. B., 2014. Genomics of CpG methylation in developing and developed zebrafish., 4: 861-869.

DeChiara, T. M., Efstratiadis, A., and Robertson, E. J., 1990. A growth-deficiency phenotype in heterozygous mice carrying an insulin-like growth factor II gene disrupted by targeting., 345: 78-80.

Deng, S. P., Chen, S. L., Xu, J. Y., and Liu, B. W., 2009. Mole- cular cloning, characterization and expression analysis of gonadal P450 aromatase in the half-smooth tongue sole,., 287: 211-218.

Dong, X. L., Chen, S. L., Ji, X. S., and Shao, C. W., 2011. Mole-cular cloning, characterization and expression analysis of Sox9a and Foxl2 genes in half-smooth tongue sole (C)., 30: 68-77.

Drake, P. L., Coleman, B. F., and Vogwill, R., 2013. The res- ponse of semi-arid ephemeral wetland plants to flooding: Link-ing water use to hydrological processes., 6: 852- 862.

Du, T., Huang, Y., Qin, X. Y., and Zhang, G. L., 2013. Difference analysis on growth characteristic of one year oldcultured at different salinity., 44: 337-341 (in Chinese with English abstract).

Duggan, M., Connolly, R. M., Whittle, M., Curwen, G., and Bur- ford, M. A., 2014. Effects of freshwater flow extremes on intertidal biota of a wet-dry tropical estuary., 502: 11-23.

Eckhardt, F., Lewin, J., Cortese, R., Rakyan, V. K., Attwood, J., Burger, M., Burton, J., Cox, T. V., Davies, R., Down, T. A., Haefliger, C., Horton, R., Howe, K., Jackson, D. K., Kunde, J., Koening, C., Liddle, J., Niblett, D., Otto, T., Pettett, R., See-mann, S., Thompson, C., West, T., Rogers, J., Olek, A., Berlin, K., and Beck, S., 2006. DNA methylation profiling of human chromosomes 6, 20 and 22., 38 (12): 1378- 1385.

Eden, A., Gaudet, F., Waghmare, A., and Jaenisch, R., 2003. Chro- mosomal instability and tumors promoted by DNA hypome- thylation., 300 (5618): 455.

Eddie, E. D., and Norman, Y. S. W., 2009. Modulation of fish growth hormone levels by salinity, temperature, pollutants and aquaculture related stress: A review., 19 (1): 97-120.

Fang, Z. H., 2013. Effect of salinity on the growth of juvenile tongue sole and its eco-physiological mechanism. Master thesis. Ocean University of China, Qingdao.

Gama-Sosa, M. A., Slagel, V. A., Trewyn, R. W., Oxenhandler, R., Kuo, K. C., Gehrke, C. W., and Ehrlich, M., 1983. The 5- methylcytosine content of DNA from human tumors., 11: 6883-6894.

Habu, Y., Kakutani, T., and Paszkowski, J., 2001. Epigenetic de- velopmental mechanisms in plants: Molecules and targets of plant epigenetic regulation., 11: 215-220.

Hasenbein, M., Komoroske, L. M., Connon, R. E., Geist, J., and Fangue, N. A., 2013. Turbidity and salinity affect feeding per- formance and physiological stress in the endangered delta smelt., 53 (4): 620-634.

Hashida, S. N., Uchiyama, T., Martin, C., Kishima, Y., Sano, Y., and Mikami, T., 2006. The temperature-dependent change in methylation of the Antirrhinum transposon Tam3 is controlled by the activity of its transposase., 18 (1): 104-118.

He, H. W., 2016. Effect of salinity on growth performance and physiological mechanism of. Master thesis. Ocean University of China, Qingdao.

Hu, Q. M., and Chen, S. L., 2013. Cloning, genomic structure and expression analysis of ubc9 in the course of development in the half-smooth tongue sole ().–, 165: 181-188.

Hu, Q. M., Zhu, Y., Liu, Y., Wang, N., and Chen, S. L., 2014. Cloning and characterization of wnt4a gene and evidence for positive selection in half-smooth tongue sole ()., 4: 7167.

Jeremiah, K., and Joseph, A. B., 2008. Effect of salinity on growth, feed utilization, and survival ofunder laboratory conditions., 20 (4): 256- 271.

Jung, D., Sato, J. D., Shaw, J. R., and Stanton, B. A., 2012. Ex- pression of aquaporin 3 in gills of the Atlantic killifish (): Effects of seawater acclimation.–, 161: 320-326.

Jjingo, D., Conley, A. B., Yi, S. V., Lunyak, V. V., and Jordan, I. K., 2012. On the presence and role of human gene-body DNA methylation., 3 (4): 462-474.

Karl, L., Giorgi, B., Natallia, S., Helena, D. C., Jean-Francois, B., Manfred, R., and Elisabeth, E., 2010. Seawater and fresh- water challenges affect the insulin-like growth factors IGF-I and IGF-II in liver and osmoregulatory organs of the tilapia., 327: 40-46.

Kulis, M., Queirós, A. C., Beekman, R., and Martin-Subero, J., 2013. Intragenic DNA methylation in transcriptional regulation, normal differentiation and cancer., 1829 (11): 1161-1174.

Kültz, D., 2012. The combinatorial nature of osmosensing in fishes., 27: 259-275.

Kültz, D., 2015. Physiological mechanisms used by fish to cope with salinity stress., 218: 1907-1914.

Laurent, P., and Dunel, S., 1980. Morphology of gill epithelia in fish., 238: R147-R159.

Li, Z. X., Zhao, G. S., Yue, Y. Y., Zhai, X. D., Ai, H. W., and Xue, X. Q., 2011. Methylation analysis of androgen receptor exon1 in Hela cells., 46 (6): 856 (in Chinese with English abstract).

Li, S. P., He, F., Wen, H. S., Li, J. F., Si, Y. F., Liu, M. Y., Huang, Y. J., and Meng, L. C., 2017a. Low salinity affects cellularity, DNA methylation, and mRNA expression of)., 43 (6): 1587-1602.

Li, S. P., He, F., Wen, H. S., Li, J. F., Si, Y. F., Liu, M. Y., He, H. W., Huang, Z. J., 2017b. Analysis of DNA methylation level by methylation-sensitive amplification polymorphism in half smooth tongue sole () subjected to salinity stress., 16 (2): 269- 278.

Livak, K. J., and Schmittgen, T. D., 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2−∆∆CTmethod., 25: 402e408.

Lorincz, M. C., Dickerson, D. R., Schmitt, M., and Groudine, M., 2004. Intragenic DNA methylation alters chromatin structure and elongation efficiency in mammalian cells., 11 (11): 1068-1075.

Martins, Y. S., Melo, R. M. C., Campos-Junior, P. H. A., Santos, J. C. E., Luz, R. K., Rizzo, E., and Bazzoli, N., 2014. Salinity and temperature variations reflecting on cellular PCNA, IGF- I and II expressions, body growth and muscle cellularity of a freshwater fish larvae., 202: 50-58.

Mazzucotelli, E., Mastrangelo, A. M., Crosatti, C., Guerra, D., Stan- ca, A. M., and Cattivelli, L., 2008. Abiotic stress response in plants: When post-transcriptional and post-translational regu- lations control transcription., 174 (4): 420-431.

Morán, P., Marco-Riusa, F., Megías, M., Covelo-Soto, L., and Pérez-Figuero, A., 2013. Environmental induced methylation changes associated with seawater adaptation in brown trout.,392-395: 77-83.

Mustafayev, N. J., and Mekhtiev, A. A., 2008. Changes of the serotonergic system activity in fish tissues during an increase of water salinity., 44: 69-73.

Nie, H., Jiang, L., Chen, P., Huo, Z., Yang, F., and Yan, X., 2017. High throughput sequencing of RNA transcriptomes inidentifies genes involved in osmotic stress response., 7: 4953.

Norman, J. D., Danzmann, R. G., Glebe, B., and Ferguson, M. M., 2011. The genetic basis of salinity tolerance traits in Arctic charr ()., 12: 81.

Ou, X., Long, L., Zhang, Y., Xue, Y., Liu, J., Lin, X., and Liu, B., 2009. Spaceflight induces both transient and heritable alterations in DNA methylation and gene expression in rice ()., 662 (1-2): 44-53.

Pan, Y. J., Fu, B. Y., Wang, D., Zhu, L. H., and Li, Z. K., 2009. Spatial and Temporal Profiling of DNA Methylation Induced by Drought Stress in Rice., 42 (9): 3009-3018 (in Chinese with English abstract).

Razin, A., and Riggs, A. D., 1980. DNA methylation and gene function., 210: 604-610.

Reinecke, M., and Collet, C., 1998. The phylogeny of the insulin-like growth factors., 183: 1-94.

Reinecke, M., Björnsson, B. T., Dickhoff, W. W., McCormick, S. D., Navarro, I., Power, D. M., and Gutierrez, J., 2005. Growth hormone and insulin-like growth factors in fish: Where we are and where to go., 142: 20-24.

Shao, C. W., Li, Q. Y., Chen, S. L., Zhang, P., Lian, J. M., Hu, Q. M., Sun, B., Jin, L. J., Liu, S. S., Wang, Z. J., Zhao, H. M., Jin, Z. H., Liang, Z., Li, Y. Z., Zheng, Q. M., Zhang, Y., Wang, J., and Zhang, G. J., 2014. Epigenetic modification and inhe- ritance in sexual reversal of fish., 24: 604- 615.

Si, Y. F., 2019. Study on SNP and DNA methylation of growth- related genes and liver transcriptome in different salinities in half smooth tongue sole (). PhD thesis, Ocean University of China, Qingdao.

Si, Y. F., Wen, H. S., Li, Y., Feng, H., Li, J., Li, S. P., and He, H. W., 2018. Liver transcriptome analysis reveals extensive transcriptional plasticity during acclimation to low salinity in., 19: 464.

Steward, N., Ito, M., Yamaguchi, Y., Koizumi, N., and Sano, H., 2002. Periodic DNA methylation in maize nucleosomes and demethylation by environmental stress., 277 (40): 37741-37746.

Szyf, M., 2012. The early-life social environment and DNA me- thylation., 81 (4): 341-349.

Takei, Y., Hiroi, J., Takahashi, H., and Sakamoto, T., 2014. Di- verse mechanisms for body fluid regulation in teleost fishes.

, 307 (7): 778-792.

Viviana, L., Indianara, F. B., Luís, A. S., and Adalto, B., 2015. Effect of salinity on survival, growth and biochemical parame- ters in juvenile Lebranch mullet(Perciformes: Mu- gilidae)., 13 (2): 447-452.

Wang, X. B., and Liu, G. Y., 2000. New research progress on intron function., 17 (3): 211-212 (in Chinese with English abstract).

Wang, Z. S., Huang, J. T., and Peng, B., 2003. Studies on critical salinity of survival and suitable growth salinity of., 18 (12): 18-20 (in Chinese with English abstract).

Weiss, A., Keshetl, R. A., and Cedar, H., 1996. DNA demethylation: Involvement of RNA., 86: 709-718.

Wurts, W. A., and Stickney, R. R., 1989. Responses of red drum () to calcium and magnesium concentrations in fresh and salt water., 76 (1-2): 21- 35.

Xiao, J., Wang, H. M., Ma, J. D., He, M. N., Long, K. R., and Jiang, A. A., 2014. Methylation of CpG island within second intron increased G-protein coupled receptor 120 (GPR120) mRNA transcripts in porcine () subcutaneous and visceral adipose tissue., 22 (8): 992-1000.

Yang, L., Lin, T. T., Zhang, D., and Liu, X., 2016. Time course effect of low salinity on the plasma osmotic pressure, ion concentrations and Na+/K+-ATPase activity in the gill of juvenile lined seahorse,., 7: 11.

Yano, S., Ghosh, P., Kusaba, H., Buchholz, M., and Longo, D. L., 2003. Effect of promoter methylation on the regulation of human peripherial blood T cells into a Th2 population., 107 (5): 2510-2516.

Zhang, X. Y., Wen, H. S., Wang, H. L., Ren, Y. Y., Zhao, J., and Li,Y., 2017. RNA-Seq analysis of salinity stress-responsive trans- criptome in the liver of spotted sea bass ()., 12 (3): e0173238.

Zhang, X. Y., Yazaki, J., Sundaresan, A., Cokus, S., Chan, S. W. L., Chen, H., Henderson, I. R., Shinn, P., Pellegrini, M., Ja- cobsen, S. E., and Ecker, J. R., 2006. Genome-wide high-re- solution mapping and functional analysis of DNA methylation in., 126: 1189-1201.

Zhao, Y. L., Ye, W. W., Wang, J. J., Fan, B. X., and Song, L. Y., 2009. Review of DNA methylation and plant stress-tolerance., 29 (7): 1479-1489.

Ziller, M. J., Gu, H., Muller, F., Donaghey, J., Tsai, L. T., Kohl- bacher, O., DeJage, P. L., Rosen, E. D., Bennett, D. A., Bern- stein, B. E., Gnirke, A., and Meissner, A., 2013. Charting a dy-namic DNA methylation landscape of the human genome., 500: 477e81.

May 16, 2019;

July 16, 2019;

October 9, 2019

© Ocean University of China, Science Press and Springer-Verlag GmbH Germany 2019

E-mail: hefengouc@ouc.edu.cn

E-mail: wenhaishen@ouc.edu.cn

(Edited by Qiu Yantao)