Introduction

As a result of global warming, heat stress has become a factor negatively affecting animal performance and health and may jeopardize the development and economy of animal husbandry1,2. Exposure to high ambient temperatures has been reported to be detrimental to pig production and meat quality3. Moreover, heat stress affects protein anabolism and catabolism as well as satellite cell proliferation, all of which play important roles in determining muscle growth and development4. In addition, heat stress activates signaling pathways associated with the regulation of skeletal muscle growth and development5,6.

Because the influence of stress on the success of the pig industry has increased, scientists have studied the specific effects of heat stress on pigs, particularly meat quality7,8,9,10. Heat stress is thought to affect gene transcription; appropriate regulation of gene transcription is crucial for maintenance of cell identity and physiological function. In preliminary studies, we identified some differentially expressed genes that were closely related to skeletal development and meat quality parameters between constant heat stress and normal temperature in pigs (Hao et al. 2015, unpublished data). However, until now, no systematic studies have evaluated the underlying molecular mechanisms mediating the development of skeletal muscle under heat stress.

DNA methylation is a well-studied epigenetic regulatory mechanism that plays a key role in the regulation of gene expression11. Recently, multiple studies have been conducted to identify genome-wide methylation profiles of economically important animals12,13,Table 1).

Table 1 Data generated by genome-wide bisulfite sequencing.
Full size table

DNA methylation patterns

In each group, approximately 4% of all genomic C sites were methylated. Methylation in pig was found to exist in three sequence contexts: CG, CHG (where H is A, C, or T) and CHH. We observed overall genome-wide levels of 67.62% CG, 0.59% CHG and 0.58% CHH methylation in the CN group and 67.54% CG, 0.21% CHG and 0.20% CHH methylation in the HE group (Table 2). Compared with the CN group, the rate of methylated CG18 in the HE group was increased from 86.04% to 94.45%. However, the rate of mCHH was decreased 2.5-fold, from 10.70% to 4.24% (Fig. 1).

Table 2 Genome-wide methylation levels of the two groups.
Full size table
Figure 1
figure 1

Comparison of DNA methylation patterns in the two groups.

Full size image

In the genome of each group, over 60% of the CpG sites were methylated, which is the primary DNA sequence context of cytosine methylation (Fig. 2). In the CN group, we detected abundant DNA methylation in CHH contexts. Although heat stress did not change the methylation level of the CpG sites in the genome, heat stress decreased the methylation level of the CHH contexts, which accounted for most of the difference between the groups.

Figure 2
figure 2

Density plot of 5-methylcytosine in various sequence contexts (mCG, mCHG and mCHH).

mC signifies 5-methylcytosine. H = A, C, or T. Chromosome numbers and scales are indicated on the periphery.

Full size image

Sequence preferences analysis for methylation

For CG context methylation, certain methylated C sites are defined as hypermethylation sites, at which the methylation level is over 75%; and some others are defined as sites of hypomethylation, at which the methylation level is less than 75%. For non-CG context methylation, hyper- and hypomethylation sites are defined as those at which the methylation levels are over or under 25%, respectively. In this study, we analyzed the relationship between sequence context and methylation preference. We calculated the percentage methylation of all possible 9-mer sequences in which the methylated cytosine was either in the fourth position (allowing an analysis of three nucleotides upstream of CG, CHG and CHH methylation) or in the first position (allowing analysis of five nucleotides following the methylated cytosine; Fig. 3). In heavily methylated regions, there was no distinct difference between the two groups in the sequence enrichment based on the genomic regions. In contrast, however, in poorly methylated regions, a high level of sequence context preferences for methylation was observed, especially in non-CG contexts.

Figure 3
figure 3

Sequence preferences for methylation in CG, CHG and CHH contexts.

Logos of sequence contexts that are preferentially methylated at the highest or lowest levels for 9-mer sequences in which the methylated cytosine is in the fourth position. (A) Regions of high methylation. (B) Regions of low methylation.

Full size image

DNA methylation levels of different functional regions

To study the global DNA methylation profile differences in the pig genome between control and constant heat stress conditions, we analyzed the DNA methylation levels of different genomic regions (Fig. 4). The comparison of average methylation levels showed that there were differential methylation levels in different components of the genome. A major proportion of methylated sites were present in the regions of promoters and introns. Among all the classes, the average methylation level of exons was the lowest. Furthermore, all five regions in the HE group exhibited lower methylation levels than did those in the CN group.

Figure 4
figure 4

DNA methylation levels of different functional regions between the Longissimus dorsi muscles of the two groups.

(A) CG regions. (B) CHG regions. (C) CHH regions. H = A, C, or T.

Full size image

DMR analysis

To characterize the differences of genome methylation levels between samples under different environmental conditions, DMRs and differentially methylated genes (DMGs) were detected. A total of 57,147 DMRs were identified between the two groups, which corresponded to 1,422 DMGs (Table 3 and Supplementary Table). The methylated regions were mainly located in introns, which proportion exceeded 70% (Fig. 5). The results of a boxplot analysis of DMRs showed that the methylation level of the HE group was lower than that of the CN group (Fig. 6). Furthermore, we also found that some of these DMGs were involved in biological processes important for skeletal muscle development: energy metabolism (such as 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 1 [PFKFB], phosphoglycerate kinase 1 [PGK1] and pyruvate dehydrogenase kinase, isozyme 3 [PDK3]), transcription factors (such as cAMP responsive element binding [CREB], NF-κB- activating protein [NKAP] and signal transducer and activator of transcription 2 [STAT2]), Ca2+ homeostasis (chloride intracellular channel 2 [CLIC2], ryanodine receptor 3 [RYR3] and calcium channel, voltage-dependent, alpha 2/delta subunit 2 [CACNA2D]), protein kinases (such as calpain 2 [CAPN2] and protein kinase C [PRKCA]) and repair and stabilization of stressed proteins (such as αB-crystallin [CRYAB] and DnaJ [Hsp40] homolog, subfamily C, member 5 [DNAJC5]). However, gene ontogeny (GO) assignments and KEGG pathway analysis showed that these methylated genes were not significantly enriched in predicted process and signaling pathway (Supplementary Fig. 1 and Fig. 2).

Table 3 Differentially methylated genes shared by HE and CN groups.
Full size table
Figure 5
figure 5

The distribution of DMR regions.

DMR, differentially methylated region.

Full size image
Figure 6
figure 6

Methylation levels of DMRs in different groups.

Boxes, quartiles 25–75%; black lines within boxes, median of the distribution (quartile 50%). DMR, differentially methylated region. HE, heat-exposed group. CN, control group.

Full size image

Association analysis between the DMGs and the differentially expressed genes (DEGs)

To explore the relationship between these DMGs and the DEGs found at the transcriptome level, an association analysis was performed. This identified four overlapped genes: RYR3, PGK1, CRYAB and four and a half LIM domains 1 protein, isoform C (FHL1C; Fig. 7). In addition, DEGs were divided into highly regulated types and lowly regulated types, according to their gene expression level. For each type of gene, we analyzed the methylation level in five functional element regions (Fig. 8). For highly regulated genes, we observed a visible difference in methylation level between the CN and HE groups in the promoter regions for both CG and non-CG contexts. However, whereas a visible difference in methylation level was found in the exon regions of CHG contexts, differences in methylation levels were found in the 5′UTR, intron and 3′UTR regions of CHH contexts. On the other hand, for lowly regulated genes, there was a visible difference in methylation level between CN and HE groups in the 5′UTR regions of CG context. A visible difference in methylation level was also found in the 3′UTR regions of CHG contexts and a difference in methylation level was found in the promoter, 5′UTR and 3′UTR regions of CHH contexts.

Figure 7
figure 7

Numbers of genes that were differentially expressed in the two groups.

DMG, differentially methylated gene; DEG, differentially expressed gene.

Full size image
Figure 8
figure 8

Association analysis of DMGs and DEGs.

DEGs were divided into highly and lowly regulated types, according to their gene expression level. For each type of gene, methylation levels were analyzed in five functional element regions. (A) Highly regulated genes. (B) Lowly regulated genes. DEG, differentially expressed gene; DMG, differentially methylated gene. HE, heat-exposed group. CN, control group.

Full size image

Discussion

DNA methylation profiles in skeletal muscle

Although global DNA methylation surveys have been performed on pigs15,17, this study is the first to systematically compare the genome-wide skeletal muscle methylation profiles between normal and heat-stressed pigs, which show different overall meat quality. Skeletal muscle is the major portion the animal used for human food and its features have a direct impact on meat quality. Moreover, the DNA methylation status of promoter and gene body regions can affect gene expression via changes in chromatin structure or transcription efficiency18,19,20; therefore, we aimed to compare the genome-wide methylation patterns between heat-exposed and normal pigs to identify differentially methylated genes that may affect the quality of meat exposed to heat stress. From these analyses, we found that the intronic regions of the pig genome comprised a large proportion of DMRs (>70%) and that only a small proportion of DMRs were located in the 3′UTR, 5′UTR and promoters. Furthermore, our previous study showed that the average daily gain (ADG) and average daily feed intake (ADFI) of heat-exposed pigs were both significantly decreased compared with those of control pigs21. These results could affect the observed changes in DNA methylation, in addition to direct heat stress. For decreased feed intake in heat-exposed pigs, food composition may be altered to compensate for decreased nutrients, which could offset a proportion of the observed methylation changes.

In mammals, most DNA methylation occurs at CpG sites and asymmetric non-CpG methylation has been detected at appreciable levels in a few cell types22,23,24. Methylation in symmetrical sequences is preserved through cycles of DNA replication by maintenance DNA methyltransferases, which show a preference for hemimethylated substrates and methylate cytosines in the newly synthesized strand. Maintenance mechanisms for asymmetric methylation patterns are unknown, but must include de novo methylation after each cell division. Results of a recent study by Guo et al.25 showed that CpG methylation is always symmetrically methylated, whereas non-CpG sites are strand-specifically methylated in introns, short interspersed nuclear elements and long interspersed nuclear elements25. They also showed that the skew of non-CpG methylation in introns is more pronounced at the boundaries and more significant for highly expressed genes. The results of this study provide further details regarding the biological significance of the various methylation patterns (CG, CHG and CHH), particularly in the context of pigs.

Key differentially methylated genes related to muscle development and meat quality

Porcine skeletal muscle development and growth have been shown to be associated with the differential expression of genes. In this study, several genes involved in muscle growth and development, including small muscle protein X-linked (SMPX), myosin, heavy chain 11 (MYH11), collagen, type XVI, alpha 1 (COL16A1), and collagen, type IV, alpha 3 (COL4A3), exhibited hypomethylation following constant heat stress. This result indicated that the expression of these genes might be upregulated. SMPX is upregulated in skeletal muscle by stretching and has been suggested to serve as both a transcription factor and a mechanosensor, possibly giving rise to changes in muscle fiber size and type. Thus, SMPX is thought to be a candidate gene for regulating muscle size. Schindeler et al.26 also found that SMPX might participate in the regulation of cytoskeletal dynamics through the Rac1/p38 pathway26. Additionally, differentiating C2C12 myoblasts overexpressing SMPX increase their susceptibility to fuse, forming large “myosacs” in an IGF-1 dependent manner27 further indicating a role for SMPX in the regulation of fiber size.

COL16A1 is a gene suggested to have importance for cell adhesion28. COL4A3 was also identified as a structural component of collagen, which is known to affect meat tenderness. In turn, myosin is a major structural protein of the thick filament of the sarcomere28. A previous study demonstrated that MyHC isoform composition affects the quality of the meat originating from bovine muscles55 was used to perform alignments of bisulfite-treated reads to a reference genome using the default parameters. The reference genome was first transformed into a bisulfite-converted version (C-to-T and G-to-A conversions) and then indexed using bowtie256. Sequence reads were also transformed into fully bisulfite-converted versions (C-to-T and G-to-A conversions) before they were aligned to similarly converted versions of the genome in a directional manner. Sequence reads that produced a unique best alignment from the two alignment processes (original top and bottom strand) were then compared to the normal genomic sequence and the methylation state of all cytosine positions in the read was inferred. The same reads that aligned to the same regions of the genome were regarded as duplicated reads. The sequencing depth and coverage were summarized using deduplicated reads. The sodium bisulfite non-conversion rate was calculated as the percentage of cytosines sequenced at cytosine reference positions in the lambda genome.

Estimating methylation levels

To identify the sites of methylation, we modeled the sum of methylated counts as a binomial (Bin) random variable with a methylation rate as follows:

We employed a sliding-window approach, which is conceptually similar to approaches that have been used for bulk bisulfite sequencing (http://www.bioconductor.org/packages/2.13/bioc/html/bsseq.html). Using a window size w = 3,000 bp and step size of 600 bp57, the sum of methylated and unmethylated read counts in each window were calculated. The methylation level (ML) for each C site showed the fraction of methylated Cs and was defined as:

The calculated ML value was further corrected using the bisulfite non-conversion rate according to previous studies58. Given the bisulfite non-conversion rate r, the corrected ML was estimated as:

Differentially methylated region analysis

Differentially methylated regions (DMRs) were identified using swDMR software (http://122.228.158.106/swDMR/), using a sliding-window approach. The window was set to 1000 bp and the step length was 100 bp. A Fisher test was then implemented to detect the DMRs. We defined regions as DMRs that had corrected P-value of less than 0.05, reads coverage of greater than 5, fold-change values of greater than 2 and false discovery rate of less than 0.05.

GO and KEGG enrichment analysis of DMR-related genes

GO enrichment analysis of genes related to DMRs was implemented by the GOseq R package59, in which gene length bias was corrected. KOBAS software60 was used to test the statistical enrichment of DMR-related genes in KEGG pathways. GO terms and KEGG with corrected P-values of less than 0.05 were considered significantly enriched by DMR-related genes.

Transcriptome analysis

Sample preparation for RNA-seq analysis

mRNA was purified from total RNA using poly-T oligo-attached magnetic beads. Fragmentation was performed using divalent cations under elevated temperatures in a proprietary Illumina fragmentation buffer. First-strand cDNA was synthesized using random oligonucleotides and SuperScript II (Life Technologies Inc, Rockville, MD, USA). Second-strand cDNA synthesis was then performed using DNA polymerase I and RNase H. Remaining overhangs were converted into blunt ends using exonuclease and polymerase. After the adenylation of the 3′ ends of the DNA fragments, Illumina PE adapter oligonucleotides were ligated in preparation for hybridization. To preferentially select cDNA fragments 200 bp in length, the library fragments were purified using the AMPure XP system (Beckman Coulter, Beverly, MA, USA). DNA fragments with adaptors ligated on both ends were selectively enriched using the Illumina PCR Primer Cocktail with a 10-cycle PCR protocol. The products were then purified (AMPure XP system) and quantified using Agilent high-sensitivity DNA assays on the Agilent Bioanalyzer 2100 system.

Clustering and sequencing

Clustering of the index-coded samples was performed on a cBot Cluster Generation System using the TruSeq PE Cluster Kit v3-cBot-HS (Illumina). After cluster generation, the library preparations were sequenced on an Illumina Hiseq 2000 platform and 100-bp paired-end reads were generated (Novogene, Bei**g, China).

Quantification of gene expression levels

After quality control and filtering of raw data, clean reads were aligned to the S. scrofa 10.2 reference genome using TopHat v1.4.061, which was chosen as it could generate a database of splice junctions based on the gene model annotation file to provide better map** results.

HTSeq v0.5.3 was used to count the read numbers mapped to each gene. The reads per kilobase of exon model per million mapped reads (RPKM) of each gene was calculated based on the length of the gene and the number of reads mapped to this gene.

Differential gene expression analysis

Differential gene expression between the control and heat stress groups was analyzed using the DESeq R package (1.8.3). The P-values were adjusted using the Benjamini & Hochberg method62. A corrected P-value of 0.05 was set as the threshold for significantly different expression.

Association analysis

We estimated the associations between the DMGs and the differentially expressed genes (DEGs) by Pearson correlation. In addition, DEGs were divided into highly regulated types and lowly regulated types, according to the results of transcriptome analysis. For each type of gene, we analyzed the methylation level in five functional element regions. Each gene function element was divided into 20 bins according to the location. The methylation level of each bin was calculated sequentially and the average methylation levels in bin1, bin2, bin3 … bin20 of all genes were then calculated.

Additional Information

How to cite this article: Hao, Y. et al. Genome-wide DNA methylation profiles changes associated with constant heat stress in pigs as measured by bisulfite sequencing. Sci. Rep. 6, 27507; doi: 10.1038/srep27507 (2016).