RESUMO
BACKGROUND: Heterotrimeric G proteins are important for numerous signaling events in eukaryotes, serving primarily to transduce signals that are initiated by G protein-coupled receptors. It has recently become clear that nonreceptor activators can regulate the level of heterotrimeric G protein signaling and, in some cases, drive cycles of receptor-independent G protein activation. In this study, we used a yeast expression cloning strategy to identify novel nonreceptor activators of heterotrimeric G proteins in a human adipocyte cDNA library. RESULTS: The human transcription factor E2F8 was found to activate heterotrimeric G proteins, suggesting a specific biological role for this recently described member of the E2F family. Epistasis studies showed that E2F8 acted at the level of G proteins and was specific for G alpha(i) over Gpa1. E2F8 augmented receptor-driven signaling, but also activated G proteins in the absence of a receptor. The GTPase-activating protein RGS4 antagonized the effect of E2F8, showing that E2F8's effect on G alpha involved nucleotide turnover. The entire E2F8 protein was required for full activity, but the majority of the signaling activity appeared to reside in the first 200 residues. CONCLUSION: In yeast, E2F8 is a guanine nucleotide exchange factor (GEF) for the alpha subunit of heterotrimeric G proteins. The molecular mechanism and biological significance of this effect remain to be determined.
RESUMO
The N terminus of G protein-coupled receptors has been implicated in binding to peptide hormones. We have used random saturation mutagenesis to identify essential residues in the N terminus of the human complement factor 5a receptor (C5aR). In a library of N-terminal mutant C5aR molecules screened for activation by C5a, residues 24-30 of the C5aR showed a marked propensity to mutate to cysteine, most likely indicating that sulfhydryl groups at these positions are appropriately situated to form disulfide interactions with the unpaired Cys(27) of human C5a. This presumptive spatial constraint allowed the ligand to be computationally docked to the receptor to form a model of the C5a/C5aR interaction. When the N-terminal mutant C5aR library was rescreened with C5a C27R, a ligand incapable of disulfide interactions, no individual position in the N terminus was essential for receptor signaling. However, the region 19-29 was relatively highly conserved in the functional mutants, further demonstrating that this region of the C5aR makes a productive physiologic interaction with the C5a ligand.
