|
Figure 31.
Ser 31A, Gin 32A and the side chain of Glu 54A in
relation to the native enzyme. These changes are not
observed n the 2'-GMP complex as the ribose ring
of the inhibitor is directed towards the outside of he
enzyme molecule and less structural change is neces-
sary for 2'-GMP to bind t the enzyme. The largest
changes caused by 2'-GMP binding are observed in
the positions of residues His 85 and Tyr 86 which
move towards the phosphate group in order to bind
to it. These two residues are very flexible and show a
degree of disorder in the native structure also.
The puckers adopted by the ribose moiety in the A
an B molecules in EMBL 2'-GMP and the A mol-
ecule in the UCLA '-GMP are essentially identical.
They are not ideal C(3')-endo conformations, ut
closely similar, with the C2' atom lyng slightly
above the plane of the CI', 04' and C4' atoms rather
than somewhat below as in th ideal conformation
(Fig. 12a). The torsion angle around the glycosyl link
is in the syn conformation, in contrast to the 3'-GMP
complex where a C(2')-endo pucker and the anti
conformation was adopted (Fig. 12b). Indeed the
riboses lie in distinctly different positions in the 2'-
and 3'-GMP complexes (Fig. 10h). The pucker for
the 2'-GMP in the B molecule of the UCLA model
refines much closer to the ideal C(3')-endo conforma-
tion (Fig. 12c). This is almost certainly a esult of the
low occupacy (~) and overall higher B factor for this
structure, rather than a real structral difference.
The occuancy of 2'-GMP is 1 in both A molecules,
and ~ m the EMBL B molecule. This again
emphasizes the need of accurate high-resolution data
for such detailed analyses.
The binding of the phosphate group is not so
specific as the binding of the base. The phosphate
binding ligands are the side chains of Glu 54, Arg 65,
Arg 69, His 85 and Tyr 86 (Fig. 13), which lie in a
relatively flexible part of the structure equipped with
a nmber of potential binding sites capable of
|
