The behaviour of 5-hydroxymethylcytosine in bisulfite sequencing

Abstract

Background: We recently showed that enzymes of the TET family convert 5-mC to 5-hydroxymethylcytosine (5-hmC) in DNA. 5-hmC is present at high levels in embryonic stem cells and Purkinje neurons. The methylation status of cytosines is typically assessed by reaction with sodium bisulfite followed by PCR amplification. Reaction with sodium bisulfite promotes cytosine deamination, whereas 5-methylcytosine (5-mC) reacts poorly with bisulfite and is resistant to deamination. Since 5-hmC reacts with bisulfite to yield cytosine 5-methylenesulfonate (CMS), we asked how DNA containing 5-hmC behaves in bisulfite sequencing.

Methodology/principal findings: We used synthetic oligonucleotides with different distributions of cytosine as templates for generation of DNAs containing C, 5-mC and 5-hmC. The resulting DNAs were subjected in parallel to bisulfite treatment, followed by exposure to conditions promoting cytosine deamination. The extent of conversion of 5-hmC to CMS was estimated to be 99.7%. Sequencing of PCR products showed that neither 5-mC nor 5-hmC undergo C-to-T transitions after bisulfite treatment, confirming that these two modified cytosine species are indistinguishable by the bisulfite technique. DNA in which CMS constituted a large fraction of all bases (28/201) was much less efficiently amplified than DNA in which those bases were 5-mC or uracil (the latter produced by cytosine deamination). Using a series of primer extension experiments, we traced the inefficient amplification of CMS-containing DNA to stalling of Taq polymerase at sites of CMS modification, especially when two CMS bases were either adjacent to one another or separated by 1-2 nucleotides.

Conclusions: We have confirmed that the widely used bisulfite sequencing technique does not distinguish between 5-mC and 5-hmC. Moreover, we show that CMS, the product of bisulfite conversion of 5-hmC, tends to stall DNA polymerases during PCR, suggesting that densely hydroxymethylated regions of DNA may be underrepresented in quantitative methylation analyses.

Figure 1. Reaction of sodium bisulfite with C, 5-mC and 5-hmC.

(A) Bisulfite-mediated deamination of cytosine. HSO₃ ⁻ reversibly and quickly adds across the 5,6 double bond of cytosine, promoting deamination at position 4 and conversion to 6-sulfonyluracil. 6-sulfonyluracil is stable under neutral conditions, but is easily desulfonated to uracil (U) at higher pH. (B) 5-methylcytosine is deaminated to thymine by bisulfite conversion, but the rate is approximately two orders of magnitude slower than that of cytosine. (C) Bisulfite quickly converts 5-hydroxymethylcytosine to form cytosine-5-methylenesulfonate (CMS). This adduct does not readily undergo deamination .

Figure 2. The bisulfite adduct of 5-hmC hinders PCR amplification.

(A) Sequence of oligonucleotide containing multiple cytosines (used in Figures 2B-D ). The yellow highlighted sequences and the red asterisks indicate the sequences and cytosine (putative CMS) residues at which DNA polymerases tended to stall when bisulfite-treated 5-hmC-containing DNA was used as template, whereas the grey highlighted sequences and black asterisks indicate the sequences and cytosine (putative CMS) residues that cause weak or no stalling by the DNA polymerases (see Figure 2D ). Cytosines in the first 106 bases of the oligonucleotide are difficult to distinguish via Sanger sequencing and thus are not annotated with regard to stalling. The underlined sequences correspond to the forward and reverse PCR primers used for PCR amplification. (B) Real-time PCR amplification curve of an oligonucleotide containing C, 5-mC or 5-hmC before and after bisulfite treatment. The sequence of the oligonucleotide is shown in Figure 2A . The small lag observed for the bisulfite-treated cytosine oligonucleotide is due in part to the fact that after conversion of cytosine to uracil, this oligonucleotide can only be amplified from one of the two strands. (C) Quantification of Ct value from experiments performed as in Figure 2B . The mean and standard deviation of three experiments is shown. (D) Primer extension assays for DNA containing different cytosine species, shown beside a Sanger sequencing ladder. Ladders of incomplete extension products were only observed in the 5-hmC-containing DNA after bisulfite treatment, at positions corresponding to G in the Sanger sequencing ladder (left lanes). Red asterisks: positions with the most significant stalling; black asterisks: positions with weak stalling or no stalling. The corresponding sequences are indicated on the left (please compare with Figure 2A ). The extension reaction performed with bisulfite-treated 5-hmC-containing DNA yielded less full-length product (arrow).

Figure 3. LC-MS analysis of conversion efficiency of 5-hmC to CMS.

(A) MS analysis of the nuclease P1 digestion products of the oligonucleotides used in Figure 2A , before (upper panel) and after (lower panel) bisulfite treatment. (B) To determine the conversion efficiency of 5-hmC to CMS in the oligonucleotide shown in Figure 2A , a standard curve was used to determine the unknown quantity of hmdCMP in the sample before and after treatment with sodium bisulfite (see text for details). The absolute value of the intercept of the best-fit line with the X-axis gives the concentration of hmdCMP remaining in the sample after bisulfite treatment as 4.69 nM. Given that the hmdCMP concentration before bisulfite treatment was 1.5 µM, this corresponds to a conversion efficiency as high as 99.7%.

Figure 4. 5-hmC did not undergo C->T transitions after bisulfite treatment and 5-mC antibody cannot recognize 5-hmC in DNA.

(A) Shown are sequencing traces of 5-hmC-containing oligonucleotide ( Figure 2A ) before and after bisulfite treatment (top and middle panels). The control C-containing oligonucleotide shows complete conversion of all C's in the top strand (highlighted sequences) to T's (lower panel). (B) Dot-blot assay of monoclonal anti-5-mC antibody detection of oligonucleotide ( Figure 2A ) containing 5-mC or 5-hmC ( Figure 2A ). Recognition on DNA by the anti-5-mC antibody is shown in the top panel, loading control is shown by the methylene blue stain in the bottom panel. The anti-5-mC antibody only recognizes the 5-mC oligonucleotide but not the 5-hmC oligonucleotide.

Figure 5. The bisulfite adduct of 5-hmC stalls Taq polymerase at CpG dinucleotides.

(A) Sequences of a set of five 158 bp oligonucleotides used in this assay. At the position marked XXXX (red font with yellow highlight), the CG oligonucleotide contains the sequence CGAT, the CGCG oligonucleotide contains two tandem CGs, and the CC and CCGG oligonucleotides contain CCAT and CCGG sequences respectively. The underlined sequences correspond to the forward and reverse PCR primers used for PCR amplification. (B) Primer extension assays of oligonucleotides shown in Figure 5A . The bands corresponding to stalled PCR reactions (red asterisks, see XXXX in Figure 5A ) were most prominent in 5-hmC-containing CC and CCGG oligonucleotides after bisulfite treatment, and were observed, though less obvious, in the CG and CGCG oligonucleotides. Full length product is indicated by an arrow. Right lanes, the Sanger sequencing was performed using the CCGG oligonucleotide as a template. (C) Quantification of Ct value of real-time PCR from experiments performed on the substrates used in Figure 5A . The mean and standard deviation of three experiments is shown.