Membrane trafficking between organelles is fundamental to the existence of eukaryotic cells. A multitude of proteins is involved in membrane trafficking, acting as building blocks for transport carriers, regulators of transport, and targeting and fusion factors. One important group of regulators are the Rab GTPases. They serve as multifaceted organizers of almost all membrane trafficking related processes in eukaryotic cells. In their active state, Rab proteins bind to effectors to mediate their function. One of these effector proteins is Rabaptin5, an early endosome protein with binding sites for Rab4, Rab5, the Rab5 GDP/GTP exchange factor Rabex5, as well as for the clathrin coat adaptor AP1. Rabaptin5 is considered to be the prototype of a Rab effector mediating a positive feedback loop by binding to active Rab5 and bringing along Rabex5, which activates further Rab5, thus maintaining endosomal fusion activity. Via the separate Rab4 interaction domain, Rabaptin5 has been proposed to function as a molecular linker between Rab5 and Rab4 to coordinate endocytic and recycling traffic. In the present study, we analysed the function of Rabaptin5 in more detail by mutagenesis of the different interaction domains or motifs and expression of the mutant proteins in HeLa cells. We identified two independent Rab4 binding domains in the N-terminal half of the protein and two cooperating sequences binding to Rab5a. Deletion of the Rab4 and Rabex5 binding domains, respectively, abolished endosome recruitment of Rabaptin5 mutants. Inactivation of Rab4a and Rabex5 by siRNA-mediated silencing, respectively, completely prevented membrane binding of wildtype Rabaptin5, confirming the requirement for Rab4a and Rabex5 and excluding indirect structural effects of the deletions. Interestingly, deletion of either one of the two Rab5 binding domains showed no effect on endosome recruitment, but induced giant endosomes positive for markers of early endosomes like Rab4a, Rab5a, and transferrin, but also for the late endosomal markers Rab7a and the ESCRT component CHMP2B, suggesting the formation of early/late endosomal chimeras. The complete disruption of the Rabaptin5/Rab5a interaction produced giant endosomes with only late endosomal properties. Our results clearly contradict the widely accepted feedback model, in which Rab5 controls its own activity. They rather indicate that Rabaptin5 is recruited to endosomes by Rab4a-GTP and Rabex5, which locally activates Rab5a by nucleotide exchange. At the same time, activated Rab5a appears to inhibit or moderate Rabaptin5 driven endosome maturation, since deletion of the Rab5 binding domains on Rabaptin5 induces a premature maturation process. The mechanism of Rabaptin5 driven endosome maturation remains to be clarified by further investigation.