Biology Reference
In-Depth Information
B. DNA Binding
MutL homologs bind both single- and double-stranded DNA in a
sequence- and mismatch-independent manner.
46,108,109
High-affinity binding
depends on DNA length, suggesting its cooperative nature.
110
While the
N-terminal domain of MutL is sufficient to bind DNA, the presence of the
dimerization domain enhances binding.
42
Similarly, studies using purified
yMutL
a
and the N-terminal domains of yMLH1 and yPMS1 reveal that the
two ATPase domains of the heterodimer bind DNA independently, but suggest
that the C-terminal region of the protein enhances binding.
110,111
The structure of the N-terminal domain of
E. coli
MutL bound to AMPPnP
resembles a saddle (
Fig. 2
).
101
Mutation of Arg266, which resides in the inner face
of the saddle, abrogates DNA binding, and hence it has been proposed that this
groove is responsible for DNA binding (
Figs. 2 and 3
).
112
Accordingly, mutation of
Arg273/Arg274 in yMLH1 or Lys197/Arg198 in yPMS1, which also reside on the
inner face of the saddle, also reduces DNA-binding affinity and causes loss of
mismatch repair activity.
107,111
Binding of both ATP and DNA protects the
N-terminal domains of human and yeast MutL proteins from proteolysis, reinfor-
cing the idea that the two
a
/
b
subdomains at the N-terminal region of MutL
function cooperatively.
103,110,113
DNA binding is enhanced by ATP binding and,
reciprocally, the ATPase activity of MutL is stimulated by single- and double-
stranded DNA, though binding to single-stranded DNA probably does not occur
in physiological conditions.
25,98,101,106
However, the specific effect of DNA bind-
ing on the ATPase activity of the protein is still poorly understood.
C. The Dimerization Domain
The structures of the dimerization domains of the
E. coli
,
Bacillus subtilis
,
and
Neisseria gonorrhoeae
MutL (referred to as MutL-CTD) have been de-
termined.
42,96,114
The N and C termini of the domain mediate dimerization,
while the intervening region defines an external subdomain that protrudes to
the solvent (
Fig. 4
). The dimerization subdomain consists of a four-stranded
antiparallel
b
-sheet. One side of this
b
-sheet defines the dimerization interface,
while the other is packed against helix
a
E. The C terminus defines a short
a
-helix that crosses the dimerization interface and is packed against the dimer-
ization subdomain of the neighboring protomer (
Fig. 4
). The external subdo-
main is formed by three tightly packed
a
-helices and a three-stranded
antiparallel
b
-sheet that has a sharp bend in the middle and is completely
exposed to the solvent. The two subdomains are connected via helix
a
A and the
loop joining helices
a
D and
a
E. In organisms lacking the latent endonuclease
MutH, MutL homologs contain a metal-binding motif with the DQHAX
2
EX
4
E
consensus sequence.
67,99
This motif is located in helix
a
at the interface
Α
between the two subdomains.
96,114