Salt Stress Induced Variation in DNA Methylation Pattern and Its Influence on Gene Expression in Contrasting Rice Genotypes

Extent and Pattern of DNA Methylation Under Control and Salinity Stress

Thirty two primer pair combinations were used to determine the cytosine methylation status in the root and shoot of four contrasting rice genotypes at seedling stage after 24 h of exposure to salinity stress and control (non-stress) conditions. Based on the presence or absence of bands (Figure 1), 1582–1635 clear and reproducible DNA fragments were amplified from root or shoot of each genotype of rice under control or salinity stress (Table 1). Under control condition in shoot, total methylation of CCGG sequences was 46.4% in Nipponbare, 48.6% in Geumgangbyeo, 57.2% in Pokkali, and 59.3% in IR29 while in root, total methylation ranged between 19.8% (Geumgangbyeo) and 32.6% (Nipponbare). In shoot, salinity stress decreased the percentage of total methylated bands in IR29 (from 59.3% to 51.7%) and Geumgangbyeo (from 48.6% to 34.5%) but increased in Nipponbare (from 46.4% to 57.6%) and Pokkali (from 57.2% to 65.3%) in comparison with the control indicating alteration of DNA methylation by salinity stress in different rice genotypes (Table 1A). However, in root, the number of fully methylated bands was reduced under salt stress compared with the control in all the four rice genotypes but it was more pronounced in Nipponbare (Table 1B). Interestingly, a general trend of higher level of fully methylated and hemimethylated bands was observed in shoot than root under both control and salinity stress in all four genotypes of rice. The fully methylated loci were always more than hemi-methylated loci (Table 1).

Open in a separate window

Figure 1

A representative MSAP gel using the primer combination EcoRI-ACG/MspI-AATC.

Both control and salinity stressed root and shoot of rice genotypes, IR29, Nipponbare (Nipp), Pokkali (Pokk), Geumgangbyeo (Geum) were used for MSAP analysis. EH and EM refer to digestion with EcoRI+HpaII and EcoRI+MspI, respectively. RC: root control; RS: root stress; SC: shoot control; SS: shoot stress.

Salinity Induced Methylation/Demethylation Changes in Contrasting Genotypes of Rice

All possible banding patterns between control and salinity stress in shoot or root of the four rice genotypes were compared to find out the changes in cytosine methylation patterns under salinity stress. Sixteen different banding patterns between control and salinity stress were observed in the MSAP gels (Table 2). The pattern A-D represented monomorphic class in which methylation pattern was same in control and salinity stress treatment. The patterns E–J indicated cytosine demethylation patterns whereas possible cytosine methylation events induced by salt stress were represented by the patterns K-P. Around 44.9%–66.7% of the CCGG sites in shoot and 84.8%–90.9% in root remain unchanged under salinity stress (Table 2). The percentage of demethylated bands under salt stress was 29.7%, 17.5%, 11.5%, and 25.1% in shoot while it was 6.8%, 8.1%, 8.3%, and 6.4% in root of IR29, Nipponbare, Pokkali, and Geumgangbyeo, respectively, indicating relatively more DNA demethylation events in salt stressed shoot than in the root of all four rice genotypes (Table 2). In general, salinity stress induced more DNA demethylation and methylation in shoot of all four genotypes.

The test of independence between three different methylation patterns and salt treatments, control and salt stress conditions, was carried out using chi-square and likelihood ratio statistics (Table S1). The large value of chi-square test and likelihood ratio statistic as well as small P-value for both shoot and root methylation level in IR29 and Nipponbare (both salt susceptible) and only for shoot in Pokkali (tolerant) and Geumgangbyeo (moderately tolerant) suggested association between stress conditions and the level of methylation. IR29 and Geumgangbyeo showed reduced level of hemi- and full methylation whereas it was increased in Pokkali and Nipponbare in shoot under salt stress compared to control. In root, more loci were hemi methylated but fewer loci were fully methylated in response to salt stress.

Analysis of the Differentially Methylated DNA Sequences

A random set of ten polymorphic fragments showing unique banding pattern in individual genotype for methylation/demethylation under salinity stress compared to control were sequenced to identify the nature of DNA sequences involved in methylation-demethylation under salinity (Table 3). The size of polymorphic DNA fragments varied between 60 to 324 bp and these fragments were distributed on the rice chromosomes 1, 2, 3, 4, and 11. The positions of cytosine alteration due to salinity stress were located in exon and UTR region of affected loci suggesting the gene-body specific methylation of MSAP loci. BLAST analysis of sequenced fragments revealed that the sequences are homologous to retrotransposons, peptidyl prolyl cis trans isomerase, calcineurin B, histone methyl transferase, rho guanine nucleotide exchange factor, cystathionine gamma synthase, DUF domain containing protein, methyl CG binding domain containing protein, and pentatricopeptide repeat domain containing protein (Table 3).These results suggest that the genes involved in wide range of cellular functions were affected mostly by gene-body specific cytosine methylation-demethylation due to salinity stress in rice.

Expression Analysis of MSAP loci in Root and Shoot of Rice Genotypes

Expression analysis of five MSAP polymorphic genes was done using quantitative RT-PCR in shoot and root of six contrasting rice genotypes under control and salinity stress to determine the effect of methylation changes on the gene expression (Figure 2). Gene Os02g44330 (P4), methylated in the shoot of IR29, showed downregulation under salinity stress in shoot of IR29 and Nonabokra, whereas its expression remain unchanged in the shoot of the remaining four genotypes under salinity stress in comparison to control (Figure 2A). However in the root, its expression was enhanced in IR29, Nonabokra, and Geumgangbyeo but downregulated in Nipponbare and Pokkali. Bengal showed a slight increase in expression. Similarly, expression changes of Os03g32490 (P5) which was demethylated in the shoot of IR29, was minimal with increased expression in the shoot of IR29, Bengal, Nipponbare, and Pokkali, but reduced expression in Nonabokra and Geumgangbyeo under salinity stress (Figure 2A). In roots of IR29 and Geumgangbyeo, salt stress enhanced the expression of P5 but reverse was the case in Bengal, Nipponbare, Pokkali, and Geumgangbyeo (Figure 2B). One of the demethylated gene, Os11g23900 (P8) in the shoot of Nipponbare, showed 6–8 fold increase expression in shoot of Bengal, and Pokkali but was downregulated in IR29, while remain unchanged in Nipponbare, Nonabokra, and Geumgangbyeo under salinity (Figure 2A). It was upregulated in roots of IR29 and Geumgangbyeo but down regulated in other genotypes (Figure 2B).The gene Os02g52290 (P1) which was methylated in shoot of IR29 and homologous to peptidyl-prolyl cis-trans isomerase, showed increased expression in the shoot and root of IR29 and Geumgangbyeo under salinity stress. The expression of P1 was enhanced in the shoot of other four genotypes under salinity stress than control, but was downregulated in the root of Bengal, Nipponbare, Pokkali, and Nonabokra. Another gene, Os01g60309 (P9), a retrotransposon, demethylated in the shoot of IR29, showed downregulation in shoot of IR29, Bengal, Nipponbare, Pokkali, and Geumgangbyeo whereas it was upregulated in Nonabokra. In root, expression of P9 was enhanced under salinity stress in IR29, Pokkali, and Geumgangbyeo, but remained unchanged or slightly increased in Nipponbare, Bengal, and Nonabokra. Therefore, the methylation pattern within the gene-body region did not always correspond to its expression changes under salinity stress and expression variation was genotype specific.

Open in a separate window

Figure 2

Expression profiles of MSAP loci in rice under salt stress.

Salinity stressed shoot (A) and root (B) of six genotypes of rice, Bengal (Beng), IR29, Nipponbare (Nipp), Pokkali (Pokk), Nonabokra (Nona), and Geumgangbyeo (Geum) were used for quantifying the expression of MSAP loci relative to control. Real-time PCR analysis was performed using gene-specific primers. The expression of each gene in different RNA samples was normalized with the expression of internal control gene, rice elongation factor 1 α (eEF1 α). Fold changes in mRNA expression for each candidate gene in different genotypes were calculated relative to its control using ddCt method. The values represented are the mean of two biological replicates, each with three technical replicates. Error bars indicate the standard deviation.

DNA Extraction and Methylation Sensitive Amplified Polymorphism (MSAP) Analysis

Genomic DNA from root and shoot were isolated by method described by Murray and Thompson [47]. MSAP technique was employed following the procedure of Subudhi et al. [48] with little modification, using a pair of methylation-sensitive restriction enzymes, MspI and HpaII in combination with EcoRI. The adapter, preamplification, and selective amplification primers were listed in table S2. In this study, we used 700 and 800 fluorescently labeled EcoRI primers during selective amplification step. The denatured PCR-amplified products were separated on 6% denaturing polyacrylamide gels on a LiCor 4300 DNA Analyzer (LiCOR Inc., Lincoln, NE) at 1000 V for 2.5 h. Electrophoregrams of all AFLP profiles were visually scored for polymorphisms for the presence (1) or absence (0) of fragments ranging from 60 bp to 350 bp in a binary matrix. Regular polyacrylamide gels were run in a BioRad Sequi-Gen unit ® (38×50 cm) using the selected primer combinations showing changes in methylation/demethylation pattern using non labeled EcoRI and HpaII/MspI primer pairs followed by silver staining for visualization and elution of desired band.

The amplified DNA fragments were divided into four types based on presence or absence of bands due to the differential sensitivity of MspI and HpaII restriction digestion. Type I represents the presence of bands in both enzymes combinations i.e. EcoRI/HpaII and EcoRI/MspI, type II bands appeared only in EcoRI/HpaII but not in the EcoRI/MspI, type III generated bands in EcoRI/MspI but not in the EcoRI/HpaII, and type IV represents the absence of band in both enzyme combinations. Type II indicates the hemimethylated state of DNA due to methylation in one DNA strand but not in its complementary strand [49]. Type III represents the case of full CG (internal cytosine) methylation, whereas type IV is the case of full methylation at both cytosines. Percentage of polymorphic MSAP bands in table 1 was calculated using the following formula:

DNA Sequencing and Homology Search of Differentially Expressed MSAP Fragments

DNA bands showing desirable polymorphism were excised from silver stained AFLP gel. DNA was eluted from the excised bands by adding 30 µl of deionized sterile water followed by heating at 100°C for 5 min. Eluted DNA was then reamplified using corresponding primer pairs and DNA band of desired size was excised from agarose gel and eluted using gel elution column (Qiagen, Valencia, USA). Eluted bands were sequenced at the Gene Lab of the School of Veterinary Medicine, Louisiana State University. Identity of obtained DNA sequence was confirmed by nucleotide-nucleotide homology search and the position of MSAP fragment within the rice gene and chromosome location was determined by alignment with homologous gene sequences in the rice genome database available at the Rice Genome Annotation Project Website (http://rice.plantbiology.msu.edu/).

RNA Isolation and Expression Analysis

Total RNA was isolated from both control and salt stressed tissues using total RNA isolation kit (Qiagen, USA). Reverse transcription for the synthesis of 1^st strand cDNA was carried out using single strand cDNA synthesis kit (BioRad, Hercules, USA) containing 2 µg of total RNA in a reaction volume of 20 µl. Primers for expression analysis of methylated genes in rice genotype were designed using Primer3 and manually verified by Oligoanalyzer (IDT Inc., USA) (Table S3). The transcript levels in different RNA samples were quantified by real-time PCR analysis employing MyiQ™ Real-Time PCR detection system (Bio-Rad, Hercules, USA). Diluted cDNA samples were used as template and mixed with 200 nM of each primer and SYBR Green PCR Master Mix (BioRad, Hercules, USA). PCR reactions were performed using the following parameters: 10 min at 95°C, 40 cycles of 15 s at 95°C and 1 min at 60°C in 96-well optical reaction plates. The identities of the amplified DNA and the specificity of the reaction were verified by agarose gel electrophoresis and melting curve analysis, respectively. Rice endogenous gene, eEF1 α (elongation factor 1α) was used to normalize variance in the quality of RNA and the amount of input cDNA. The relative mRNA levels for each of the genes in different RNA samples were computed by ddCt method [50]. At least two different RNA isolations and synthesized cDNA were used for quantification and each cDNA sample was subjected to real-time PCR in triplicate.

Bisulfite Sequencing

One micrograms of total DNA extracted from unstressed and salinity stressed tissues of six rice genotypes, collected 24 hours after imposition of salt stress, was modified by sodium bisulfite using Epitect Plus DNA Bisulfite Kit (Qiagen, USA) following the manufacturer’s protocol. An aliquot of 2 µl of bisulfite-treated DNA was used for each PCR reaction (25 µl) using EpiMark® Hot Start Taq DNA Polymerase (New England Biolab, USA). The primers for two MSAP loci corresponding to fragments, Os02g52290 (P1) and Os01g60309 (P9), were designed using Methyl Primer Express Software^R (Applied Biosystems) to amplify bisulfite-converted genomic DNA. The primers were: 5′ TGTAAGTTTGTTTTTGGTTTG-3′ (P1F), 5′-TACAACCCAAAACTTATTATC-3′ (P1R), 5′-ATTAAGGGTTGGTGTTATTTT-3′ (P9F), 5-CATCAAACCTTTTTCTTATAACTCC-3′(P9R). PCR amplified fragments were eluted from gel using QiaQuick Gel Extraction Kit (Qiagen, USA) and top strands of DNA were sequenced at the High Throughput Genomics Unit of the University of Washington, Seattle, USA, using gene specific forward primers. Methylation status of DNA was obtained by comparing the sequence of the bisulfite-treated DNA with that of untreated DNA, where conversion of a C to T indicated non-methylated C. In contrast, the absence of C to T conversion indicated methylation. Cytosine methylation status in the top strand of obtained nucleotide sequences were calculated using CyMATE v2 [51]. The methylation level for each of the three kinds of cytosines, CG, CHG, and CHH, (H represents A, T, or C) was calculated using following formula:

Methylated cytosine (%)  =  [Number of non-converted (methylated) cytosines/Total number of cytosines of each type]×100.

We thank Dr. Sabrina Taylor of the School of Renewable Natural Resources, LSU Agricultural Center for allowing us to use the LI-COR 4300 for MSAP analysis. The manuscript is approved for publication by the Director of Louisiana Agricultural Experiment Station, USA as manuscript number 2012-306-7274.