Phosphorylation of G protein-coupled receptors (GPCRs) by GPCR kinases (GRKs) is a major mechanism of desensitization of these receptors. GPCR activation of GRKs involves an allosteric site on GRKs distinct from the catalytic site. Although recent studies have suggested an important role of the N- and C-termini and domains surrounding the kinase active site in allosteric activation, the nature of that site and the relative roles of the RH domain in particular remain unknown. Based on evolutionary trace analysis of both the RH and kinase domains of the GRK family, we identified an important cluster encompassing helices 3, 9, and 10 in the RH domain in addition to sites in the kinase domain. To define its function, a panel of GRK5 and -6 mutants was generated and screened by intact-cell assay of constitutive GRK phosphorylation of the beta(2)-adrenergic receptor (beta 2AR), in vitro GRK phosphorylation of light-activated rhodopsin, and basal catalytic activity measured by tubulin phosphorylation and autophosphorylation. A number of double mutations within helices 3, 9, and 10 reduced phosphorylation of the beta2AR and rhodopsin by 50 to 90% relative to wild-type GRK, as well as autophosphorylation and tubulin phosphorylation. Based on these results, helix 9 peptide mimetics were designed, and several were found to inhibit rhodopsin phosphorylation by GRK5 with an IC(50) of approximately 30 microM. In summary, our studies have uncovered previously unrecognized functionally important sites in the regulator of G-protein signaling homology domain of GRK5 and -6 and identified a peptide inhibitor with potential for specific blockade of GRK-mediated phosphorylation of receptors.