204 Hornby
substrate can be monitored by subsequent restriction enzyme analysis.
In each case, the DNA substrate should be carefully chosen in order to
maximize the incorporation of methyl groups during the incubation.
For example, the rate of methylation of calf thymus DNA by a verte-
brate DNA Mtase will be slower than that of a bacterial DNA prepa-
ration, since the frequency of methylated CpG dinucleotides will be
much greater in the former substrate. Furthermore, it has been shown
that in the case of mammalian and some (but not all) prokaryotic DNA
Mtases, hemimethylated DNA is the preferred substrate. Therefore, a
judicious choice of DNA substrate can improve the sensitivity of the
assay considerably. In the case of dA methylation, a chromosomal
DNA preparation from a dam- strain of E. coli will be free of adenine
methylation and is therefore a convenient substrate for monitoring
activity during purification.
More defined substrates may be required in those cases where the
sequence specificity of the enzyme is known. For example, in order to
determine the specific activity of the enzyme EcoRI Mtase, a plasmid
or oligonucleotide duplex containing one or more GAATTC recogni-
tion sequences would be a suitable substrate. Incubation of the latter
enzyme preparation with a plasmid containing no EcoRI sites would
serve as a control against nonspecific methylation. This is especially
important during the early stages of a protein purification.
2.3. Cofactor Requirements
Restriction and modification enzymes are classically divided into
three groups that reflect their structural and functional complexity.
The Type I modification enzymes are the most complex and require
Mg 2÷ ions (-5 mM), and ATP ( 100 laM) for optimal activity. In general,
Type II and III enzymes catalyze methyltransfer in the absence of ATP
and Mg 2÷, although there are some exceptions, such as TaqI Mtase,
which do require divalent cations. Mammalian DNA Mtases require
the presence of neither of the above cofactors, although it has been
reported that Co 2÷ (100 ~M) ions stimulate some mammalian enzymes.
2.4. Ionic Strength
All enzymes have an intrinsic optimal ionic strength for activity, and
this can only be determined on an empirical basis. However, in general,
the optimum salt (usually NaC1 or KC1) concentration is that of the cor-