Two classes of overlaps can be described. The first class of overlap occurs with restriction endonucleases which recognize degenerate sequences and methylases which act on only one of the subsets. For example, HincII recognizes the degenerate sequence GTPyPuAC which represents the four sequences GTCGAC, GTCAAC, GTTGAC and GTTAAC. M.TaqI methylates the sequence TCGA at the adenine residue. Those HincII sequences (GTCGAC) which contain the internal sequence TCGA are resistant to cleavage by HincII after methylation by M.TaqI, but all other HincII recognition sequences are cleaved. The new cleavage specificity can be described as GTPyAAC. The second class of overlap occurs at the boundaries of the recognition sequence of a restriction endonuclease and a methylase. For example, a BamHI site (GGATCC) followed by GG (or preceded by CC) overlaps an M.MspI site (CCGG). M.MspI will methylate the 5´ C (mCCGG) and result in a BamHI site methylated at its internal cytosine residue (GGATmCCGG) that is now resistant to cleavage by BamHI.

Since the number of restriction endonucleases with recognition sequences long enough to create megabase DNA fragments is limited, multistep protection/cleavage reactions can be used to create rare cleavage sites. A well characterized example involves protection of DNA with M.FnuDII (mCGCG) followed by cleavage with NotI (GCGGCCGC). This combination blocks NotI cleavage at the overlapping NotI/M.FnuDII site CGCGGCCGC (equivalent to GCGGCCGCG). As a result, the subset of Not I sites preceded by a C or followed by a G will be resistant to cleavage thereby increasing the apparent NotI specificity approximately twofold (1).

An additional product available from New England Biolabs which takes advantage of methylation to alter apparent recognition specificities is the CpG Methylase (M.SssI). CpG methylase methylates the cytosine residue in the dinucleotide sequence 5´...CG...3´ and can be used to alter the cleavage of restriction endonucleases that overlap this sequence.

Reaction Conditions
The altered specificities are generated using a two-step in vitro procedure: methylation of DNA by a site-specific methylase, followed by cleavage of the DNA by a restriction endonuclease. DNA methylases, from the type II restriction/modification systems of bacteria, methylate under conditions similar to those of restriction endonucleases with the exception that methylases require S-adenosylmethionine (SAM) as a methyl group donor. Therefore it is generally acceptable to carry out the methylation reactions using standard restriction endonuclease buffers to which SAM (supplied with each vial of methylase) has been added. It should be noted that methylases do not require divalent cations for activity and that virtually all DNases (restriction endonucleases, exonucleases and non-specific endonucleases) require a divalent cation (usually Mg⁺⁺).

The Cleavage Products
In vitro methylated DNA fragments are in most respects indistinguishable from unmethylated DNA fragments. We have observed no difficulty in the ligation of these fragments. There are two notable exceptions with regard to DNA fragments that have been methylated by cytosine methylases. When sequencing DNA by the chemical method of Maxam and Gilbert (2), 5-methyl cytosines do not generate a band in the C reaction channel. This should not be a major concern because the absence of a band on a sequencing gel can be predicted to occur at the recognition sequence of the methylase. The more significant problem is a reduced transformation efficiency with certain methylcytosine-modified DNAs when cloning in most common strains of E. coli. This reduction in transformation efficiency may be caused by endogenous restriction systems in E. coli : mcrA (rglA) restricts HpaII methylated DNA (3), and mcrBC (rglB) restricts HaeIII, AluI, HhaI and MspI methylated DNA (3). The problem can be avoided by cloning into E. coli strains which lack the Mcr systems. For additional details, see Restriction of Foreign DNA by E.coli K-12.

Reference
1. Qiang, B-q. et al. (1990) Gene 88, 101­105.
2. Maxam, A. M. and Gilbert, W. Methods in Enzymology 65, p. 499.
3. Raleigh, E. A. and Wilson, G. (1986) Proc. Natl. Acad. Sci. USA  83, 9070-9074.