1 Introduction

DNA methylation has been proposed to be a predominant epigenetic mechanism underlining the response of organisms to environmental change (Bird, 2002). In eukaryotes, methylation is confined to cytosine and has been described in CG, CHG and CHH (H = A, T or C). DNA methylation may render genes inaccessible, and inactivate their transcription (Doerfler, 1983). Methylation is essential for development (Smith and Meissner, 2013) and associates with a number of key processes such as genomic imprinting and polyploidization (Li et al., 1993; Jiang et al., 2016), X-chromosome inactivation (Csankovszki et al., 2001), aging (Maegawa et al., 2010) and environmental adaptation (Christensen et al., 2009). Environmental stimuli may cause heritable phenotypic variation through methylation (Angers et al., 2010). DNA methylation plays an important role in adapting to environmental stresses in, for example, cold (Shan et al., 2013), heavy metals (Ou et al., 2012), temperature (Navarromartín et al., 2011) and salinity (Wang et al., 2014). Osmotic-stressed plants hold higher level demethylated DNA than the non-stressed (Zhang et al., 2013). In addition, DNA methylation level of Cynoglossus semilaevis significantly decreased at salinity 15 (Li et al., 2017).

The salinity of coastal aquaculture area is sensitive to climate, especially rainfall and typhoon (Feng et al., 2005), which may significantly affect the survival and growth of marine animals. Swimming crab (Portunus trituberculatus) widely inhabiting the coasts of China, Japan and Korea is one of the most important commercial species in China (Chen et al., 2006). It can adapt to low salinity environments (Ruscoe et al., 2004; Pan et al., 2007); however, its survival and growth are significantly affected by low salinity.

Methylation-sensitive amplified polymorphisms (MSAP) is effective for describing genome methylation status (Xiong et al., 1999). MSAP is a modification of amplified fragment length polymorphism (AFLP). Osmoregulation has been well documented in crustaceans; however DNA methylation of crustacean is scarcely studied. The present study was carried out to investigate the change of DNA methylation pattern in the muscle, hepatopancreas and gill of P. trituberculatus at low salinities in order to understand the potential role of DNA methylation when P. trituberculatus is low salinity stressed.

2 Materials and Methods 2.1 Crab Rearing and Sampling

The crab individuals used in this study were obtained from XinYi Aquatic Products Limited Company, Zhejiang, China. The crab individuals were acclimated at 25℃ for a week, during which the salinity of the aerated seawater was maintained at 24. During the entire experiment, crab individuals were fed with fresh fish with the uneaten removed, and one third of the water was changed daily. The low salinity of the seawater was adjusted by adding aerated tap water.

One hundred twenty individuals with an average weight of 25.32 g ± 3.12 g were placed in six pools (8 m × 2 m × 1 m), 20 each. The pools were divided into the control (salinity 24) and experimental (salinity 12) groups, three pools each. The crab was reared for 30 days with gill, muscle and hepatopancreas sampled for DNA extraction with a standard phenol/chloroform protocol.

2.2 MSAP Procedure and Fragment Isolation

MSAP was performed following the method of Xiong et al. (1999). DNA (200 ng) was double-digested with one of the methylation-sensitive enzymes, either Hpa Ⅱ or Msp Ⅰ, and the methylation-insensitive EcoR Ⅰ. Hpa Ⅱ and Msp Ⅰ recognize tetranucleotide sequence, 5'-CCGG-3', differentially. Hpa Ⅱ is inactive when two cytosines are methylated but cuts the sequence when any one of two cytocines is methylated (hemimethylated), whereas Msp Ⅰ cuts 5'-C^5mCGG-3' but not 5'-^5mCCGG-3'. Therefore, comparisons of EcoR Ⅰ/Hpa Ⅱ and EcoR Ⅰ/Msp Ⅰ digestion profiles allows us to detect the methylation status at this site.

Genomic DNA (200 ng) was digested with 20 U Hpa Ⅱ or Msp Ⅰ and 10 U EcoR Ⅰ in 20 μL volume at 37℃ for 6 h. The digestion was stopped at 65℃ for 10 min. The digested DNA was ligated with EcoR Ⅰ and Hpa Ⅱ/Msp Ⅰ adaptors for 12 h. Pre-amplification was performed in a 20 μL volume containing 2 μL ligated products, 0.4 μL 10 mmol L^-1 dNTP (each), 2 μL 10×Buffer, 0.2 μL 5 U μL^-1 Taq DNA polymerase, 0.5 μL 10 μmol L^-1 HM adaptor and 0.5 μL 10 μmol L^-1 EcoR Ⅰ adaptor. PCR was performed on a master-cycler gradient thermal cycler (Eppendorf) by predenaturing at 94℃ for 2 min followed by 30 cycles of denaturing at 94℃ for 30 s, annealing at 56℃ for 30 s, and extending at 72℃ for 60 s, and a final extension at 72℃ for 10 min. The PCR product was used as the template for the second PCR by denaturing at 94℃ for 2 min, followed by 11 cycles of denaturing at 94℃ for 30 s, annealing at 65℃ for 30 s and extending at 72℃ for 60 s (the annealing temperature was decreased to 55℃, one degree each cycle, touchdown PCR), 25 cycles of denaturing at 94℃ for 30 s, annealing at 65℃ for 30 s and extending at 72℃ for 60 s and a final extension at 72℃ for 10 min. Primers and adaptors are listed in Table 1.

Amplification product was resolved via 8% denaturing polyacrylamide gel, and visualized by silver-staining. A 10-bp DNA ladder (Invitrogen Inc.) was used as markers of allele size.

2.3 Data Analysis

Only unambiguous and reproducible fragments were scored. The different types of bands were counted three times to minimize artificial error. Based on presence (1) or absence (0) of EcoR Ⅰ/Hpa Ⅱ and EcoR Ⅰ/Msp Ⅰ bands, respectively, four types of band patterns were identified, A, present in both (1-1), non-methylation; B, present in only EcoR Ⅰ/Hpa Ⅱ (1-0) and hemi-methylation; C, present in only EcoR Ⅰ/Msp Ⅰ (0-1) and full methylation; and D, absent in both (0-0), hyper-methylation or uninformative because of the absence of a fragment. The analysis of DNA methylation fragments was conducted using a chisquare test in SPSS19.0 (Fig. 1).

3 Results

The statistical analysis of the methylation epiloci in experimental and control groups are presented in Table 2. The methylation ratio of control ranged from 47.31% (muscle) to 22.94% (hepatopancreas) to 17.69% (gill). The ratio of experimental group varied similarly from 39.61% (muscle) to 25.19% (hepatopancreas) to 17.26% (gill). In comparison with control, the methylation ratio decreased in muscle (by 7.70%, P < 0.01) but increased in hepatopancreas (by 2.25%, P < 0.05).

The change of DNA methylation pattern under low salinity stress is listed in Table 3. Three changes of methylation epiloci were identified including 1) unchanged (A1-A3), 2) demethylated (B1-B6); and 3) methylated (C1-C6). The percentage of methylation pattern change is shown in Fig. 2. At low salinity, most of the epiloci remained unchanged, but demethylation and methylation occurred in three tissues simultaneously. The largest proportion of demethylation was detected in muscle, and the methylation ratio of these three tissues was highly similar.

4 Discussion

DNA methylation is an epigenetic process which helps to control gene transcription. It is important through the whole life in eukaryotes. Although various cell types have the same genome, these cells have different methylomes at different times and in different spaces. In rice (Oryza sativa L.), the DNA methylation in leaf and root was 16.43% and 22.13% (Wang et al., 2011). In chicken, the methylation level was about 29.7% in muscle, 27.5% in liver, 27.5% in heart, and 26.1% in kidney (Xu et al., 2011). In grass carp (Ctenopharyngodon idella), a high level of methylation was detected (75.9%) (Cao et al., 2007). DNA methylation has been well-studied in vertebrates and plants but less studied in most invertebrates. In the present study, the genomic baseline of DNA methylation of P. trituberculatus was 17.69% in gill, 22.94% in hepatopancreas, and 47.31% in muscle, being consistent with the reported in other marine invertebrates varing between 20% and 50% (Jiang et al., 2014; Sun et al., 2014; Zhao et al., 2015; Kong et al., 2017). The different DNA methylation levels in different tissues suggest a tissue-specific epigenetic regulation of gene expression in differentiation (Zhao et al., 2015). In our study, the higher methylation rate in the muscle (47.31%) may be due to the relatively low metabolic rate and low expression of genes in this tissue.

Salinity is one of the most important abiotic factors that influence not only the distribution and abundance of crustaceans, but also their general physiology and well being (Romano and Zeng, 2012). DNA methylation plays an important role in resisting abiotic stress (Bird, 2002). The changes of DNA methylation under low salinity stress have been documented recently in Cynoglossus semilaevis and Crassostrea gigas (Li et al., 2017; Zhang et al., 2017), both showed a downward trend in methylation in gill at earlier low-salinity stage (7 days and 24 h, respectively); however, they recovered to the original and control level at long term (60 days and 5 days, respectively). In the present study, when the crab subjected to low salinity stress for as long as 30 days, the methylation ratio was not significantly affected in gill (decreased by 0.43%, P > 0.05). Gill was reported as the direct regulation tissue for osmoregulation (Mcnamara and Faria, 2012). The relatively stable methylation level in gill reflects its greater adaptability to environmental salinity stress. Despite the lack of change in overall methylation ratio in gill of P. trituberculatus after low salinity stress, changes in the methylation pattern of gill tissue could still be observed (Table 3), indicating that DNA methylation may act as a dynamic balance to regulate the low salinity-resistance in the gill of P. trituberculatus.

Difference of methylation during low salinity stress mainly occurred in muscle (7.70%, P < 0.01) and hepatopancreas (2.25%, P < 0.05), both are tissue-specific. Muscle and hepatopancreas have been reported to suffer a significant alteration of amino acid composition and deformation of microstructure (Han et al., 2014; Koyama et al., 2018), which presumably link with a strong suppression or activation of transcription and translation under low salinity stress.

DNA methylation often associated with the change of gene expression; however, the relationship between DNA methylation and gene expression is more complicated than the initially predicted. The effect of DNA methylation on gene expression varies in different tissues, sequences and genome regions (Wang et al., 2014). To further understand the relationship between methylation and expression of a specific gene, the gene expression profiles and methylation sites should be carefully examined.

5 Conclusions

The results showed that the baseline DNA methylation is different in muscle, gill and hepatopancreas of P. trituberculatus (17.69% to 47.31%), demonstrating the important role of DNA methylation in tissue-specific epiloci. Furthermore, DNA methylation patterns of three tissues changed after P. trituberculatus was cultured at salinity 12 for 30 days, which suggests that DNA methylation may have an important role in salinity adaptation. The results provided a basis for further research on the potential role of DNA methylation in P. trituberculatus during adapting to low salinity.

Acknowledgements

This study was supported by the grants from the National Natural Science Foundation of China (No. 4147 6124), the Natural Science Foundation of Zhejiang Province (No. LY17C190005), the Major Agriculture Program of Ningbo (No. 2017C110007), the Ningbo Science and Technology Project (No. 2016C10037), the Open Fund of Ningbo University (No. xkzsc1505) and K C Wong Magana Fund in Ningbo University. We would like to thank Dr. Sarah-Louise Counter Selly (University of Stirling, UK) for her linguistic assistance during the preparation of this manuscript.