|
Figure 2.
Fig. 2. Superposition of the putative effector binding loops
(  2-  4, 3- 5, and 4- 6) and the
5 helix
from G[s][  ]onto
G[i][ ]^(42).
The side chains from residues of G[s][ ]are
drawn as^ stick models with the use of conventional coloring.
The backbone^ and side chains of G[i][ ]are
illustrated in transparent rose.^ The model of G[i][ ]is
derived from the structure of the G[i][ ][1]·RGS4^
complex (17), which has a completely ordered 5 helix.
The superposition^ is essentially the same as that shown in Fig.
1B. The 2- 4 loops^ of
each subunit
are essentially identical. The 3- 5 loop of^
G[s][ ],
although structurally similar to that of G[i][ ], is^
rotated downward in the figure. This rotation creates a
hydrophobic^ pocket on the back side of the sheet,
which is filled by the^ side chain of Met^386 from the 5 helix,
and moves the residue at position 282 in G[s][ ]^toward
the conserved Phe^238. In the G[s] subfamily, residue 282 is a
leucine, which helps to^ accommodate the shift of the 3- 5 loop. The
4- 6 loop of
G[s][ ]^is
longer than and shares no sequence identity with its
counterpart^ in G[i][ ]. The
3- 5 and 4- 6 loops are
supported by a stacking^ interaction between Trp^277 and
His^357, both of which are invariant in the G[s] subfamily. The
5 helix^
of G[s][ ]is
bent, whereas that of G[i][ ]extends
straight into^ solvent. The large differences observed in the
4- 6 and 5
structures^ may help account for receptor specificity among
closely related^ subunits.
|
