The genes of Type I R-M Systems
Type I Restriction-Modification enzymes are composed of three
subunits, encoded by the three host specificity
determinant (hsd) genes; the hsdR gene is
absolutely required for restriction activity; the hsdS gene is responsible for DNA specificity and together
with hsdM is the minimal genetic requirement for methylation
activity.
However, the holoenzyme can also behave as a methylase,
the activity of the enzyme is dependent upon the
methylation status of the DNA substrate. Therefore, the
enzyme must control the
opposing functions of restriction and modification.
hsdR - restriction gene, phenotype of
mutations = r-m+.
hsdM - modification gene, phenotype of
mutations = r-m-.
hsdS - specificity gene, phenotype of
mutations = r-m-.
The domain structure of HsdS
The DNA-binding subunit of a Type I R-M system consists of
two target recognition domains separated by conserved domains, which are highly
conserved between family members.
For the EcoR124I R-M system the recognition sequence
(sR124) is 5'-GAAnnnnnnRTCG-3'.
TRD1 is responsible for recognition of the 5'-end of the recognition sequence (GAA),
while TRD2 recognises the 3'-end (RTCG).
The conserved regions contain two highly conserved repeats
(shown by the arrows) that were originally described by Argos
(1983) as the regions of the protein
likely to recognise the target DNA sequence. However, this has now been shown
not to be the case (e.g. Gubler, 1992).
Careful analysis of the arrangement of these repeated units
in all Type I HsdS subunits indicates that all the subunits consist of a similar
set of sub-repeats and this has led to a prediction of a
circular structure for the HsdS
subunit (Kneale 1994). To test
this model, Janscak & Bickle (1998)
circularly permuted the HsdS subunit of the Type IB R-M enzyme EcoAI, at the DNA
level, by direct linkage of codons for the original termini and introduction of
new termini elsewhere along the N-terminal and central conserved regions. By
analysing the activity of these mutant enzymes, two circularly permuted variants
of HsdS, which had termini located at equivalent positions in the N-terminal and
central repeats, respectively, were found to fold into a functional DNA
recognition subunit with wild-type specificity. This result suggests that the
N- and C-termini of the native protein are in close proximity.
 Recently, two crystal structures have been solved for HsdS
proteins from the uncharacterised Type I R-M systems MjaXIP (Kim
et al., 2005) and MgeORF438 (Calisto
et al., 2005) respectively. While these HsdS subunits are not known to be
components of functional restriction enzymes, and their DNA specificity is
unknown, the structure has confirmed the circular nature of the HsdS subunit and
the close proximity of the N- and C-termini as indicated by
Janscak & Bickle.
The central conserved region has been subjected to intensive
mutagenesis using deletion mutagenesis, site-directed mutagenesis and
PCR-misincorporation mutagenesis. This has allowed us to show that this region
is critical for protein-protein interactions with the other subunits of the
endonuclease (Abadjieva et
al., 1994). TRD1 and the central conserved region are sufficient to
assemble an active enzyme (Abadjieva
et al., 1993;
Macwilliams & Bickle, 1996). This reflects the circular structure of the
HsdS subunit and the deletion mutants assemble through
dimerisation of a M1S½
subunit. Since the recognition sequence of this deletion mutant is
GAAnnnnnnnTTC, and TRD1 recognises the GAA component of the sR124
sequence, it is apparent that each half subunit recognises the opposite strand
on the DNA strand.
The Type IC DNA methyltransferase M.EcoR124I
recognises the nonpalindromic DNA sequence GAAn6RTCG.
Taylor et al. (1994)
have used small angle X-ray scattering to investigate the solution structure of
the methyltransferase and of complexes of the enzyme with unmethylated and
hemimethylated 30 bp DNA duplexes containing the specific recognition sequence.
A major change in the quaternary structure of the enzyme was observed following
DNA binding, based on a decrease in the radius of gyration and a reduction in
the maximum dimension of the enzyme. The structural transition observed is
independent of the methylation state of the DNA. CD shows that there is no
change in the secondary structure of the protein subunits when DNA is bound. In
contrast, there is a large increase in the CD signal arising from the DNA,
suggesting considerable structural distortion which may allow access to the
bases targeted for methylation. We propose that DNA binding induces a large
rotation of the two HsdM subunits towards the DNA, mediated by hinge bending
domains in the specificity subunit HsdS.
van Noort et al.
(2004) have shown that the large structural distortion in the DNA is due to
DNA bending and that the MTase induces a DNA bend of 49º.
This bend is comparable to that observed with EcoKI, which is also the angle of
the 'mimic' protein OCR
(Walkinshaw et al 2002), which acts an antirestriction protein
(Atanasiu et al 2002), and is
presumed compete for the DNA target site within the enzyme.
DNA specificity
Type I restriction-modification enzymes recognise bipartite
sequences, which consist of a 5' specific region of 3-4 bp and a 3' specific
region of, usually, 4 bp, separated by a non-specific spacer of 6-8 bp (e.g.
EcoR124I recognises GAAn6RTCG - where R = purine and n = any
nucleotide). DNA cleavage by Type I R-M enzymes occurs at non-specific sites far
from their recognition sequence, which is mediated by ATP-dependent DNA
translocation past the enzyme (Yuan et
al., 1980; Studier & Banyopadhyay
1888; Dryden et al., 1997;
Szczelkun et al., 1997).
Known recognition sequences for Type I R-M enzymes:
| 
Restriction-Modification Systems
