The binding site of the dopa
mine D2 receptor, like that of ho
mologous G-protein-coupledreceptors (GPCRs), is contained within a water-accessible crevice for
med a
mong its seven trans
me
mbraneseg
ments (TMSs). Using the substituted-cysteine-accessibility
method (SCAM), we are
mapping the residuesthat contribute to the surface of this binding-site crevice. We have
mutated to cysteine, one at a ti
me, 21consecutive residues in the fourth TMS (TM4). Eleven of these
mutants reacted with charged sulfhydryl-specific reagents, and bound antagonist protected nine of these fro
m reaction. For the
mutants in whichcysteine was substituted for residues in the cytoplas
mic half of TM4, treat
ment with the reagents had noeffect on binding, consistent with these residues being inaccessible and with the low-resolution structureof the ho
mologous rhodopsin, in which TM3 and TM5 occlude the cytoplas
mic half of TM4. Althoughhydrophobicity analysis positions the C-ter
minus of TM4 at 4.64, Pro-Pro and Pro-X-Pro
motifs, whichare known to disrupt
mages/gifchars/alpha.gif" BORDER=0>-helices, occur at position 4.59 in a nu
mber of ho
mologous GPCRs. The SCAMdata were consistent with a C-ter
minus at 4.58, but it is also possible that the
mages/gifchars/alpha.gif" BORDER=0>-helix extends one additionalturn to 4.62 in the D2 receptor, which has a single Pro at 4.59. In ho
mologous GPCRs, the high degreeof sequence variation between 4.59 and 4.68 is
more characteristic of a loop do
main than a helical seg
ment.This region is shown here to be very conserved within functionally related receptors, suggesting an i
mportantfunctional role for this putative nonhelical do
main. This inference is supported by observed ligand-specificeffects of
mutations in this region and by the predicted spatial proxi
mity of this seg
ment to known ligandbinding sites in other TMs.