Structure of the Mon1-Ccz1 complex reveals molecular basis of membrane binding for Rab7 activation

Activation of the GTPase Rab7/Ypt7 by its cognate guanine nucleotide exchange factor (GEF) Mon1-Ccz1 marks organelles such as endosomes and autophagosomes for fusion with lysosomes/vacuoles and degradation of their content. Here, we present a high-resolution cryogenic electron microscopy structure of the Mon1-Ccz1 complex that reveals its architecture in atomic detail. Mon1 and Ccz1 are arranged side by side in a pseudo-twofold symmetrical heterodimer. The three Longin domains of each Mon1 and Ccz1 are triangularly arranged, providing a strong scaffold for the catalytic center of the GEF. At the opposite side of the Ypt7-binding site, a positively charged and relatively flat patch stretches the Longin domains 2/3 of Mon1 and functions as a phosphatidylinositol phosphate–binding site, explaining how the GEF is targeted to membranes. Our work provides molecular insight into the mechanisms of endosomal Rab activation and serves as a blueprint for understanding the function of members of the Tri Longin domain Rab-GEF family.

Rab GTPases are molecular switches that function as markers of organelle identity and coordinate intracellular trafficking as part of the conserved fusion machinery (1). The cycling of Rab GTPases between the inactive GDP-bound and the active GTP-bound form is tightly controlled by GTPase-activating proteins (GAPs) and guanine nucleotide exchange factors (GEFs) (2). While GAPs promote the intrinsically low GTP hydrolysis rate of Rabs to switch them off, GEFs stimulate nucleotide release and the loading of Rab with GTP to convert the GTPase to its active conformation. Inactive Rabs are kept cytosolic by the GDP dissociation inhibitor (GDI), which binds the Rab prenyl anchor (3). GDI can be removed from Rab GTPases by the GDI displacement factor (4), and the exchange of GDP with GTP also couples the association of Rab GTPases with membranes. Thus, the spatiotemporal regulation of Rab GTPases and downstream fusion events ultimately depend on the activation of their cognate GEFs.Tri Longin domain (TLD) Rab-GEFs comprise one of many GEF families known. They contain at least two subunits, each of which is predicted to consist of three Longin domains (LDs) (5, 6). The TLD GEF family comprises the universally conserved Rab7-GEF Mon1-Ccz1 (MC1) (7–9) and two other complexes specific to metazoans, namely BLOC-3 (biogenesis of lysosome-related organelles complex-3, which includes Hps1 and Hps4) and Inturned-Fuzzy. BLOC-3 is the GEF for Rab32 and Rab38 on lysosome-related organelles (10), and mutations in BLOC-3 cause the genetic disease Hermansky–Pudlak syndrome (11). Inturned-Fuzzy, which acts as the GEF of Rab23 (5), has been implicated in the establishment of planar cell polarity and ciliogenesis and was described as part of the planar cell polarity effector complex in flies and the CPLANE (ciliogenesis and planar cell polarity effector) complex in mammalian cells (12). Several studies showed that in metazoans, TLD RabGEFs do not consist of two subunits but have additional non-TLD proteins bound to the heterodimeric core (13–17). The function of these additional subunits in the “enlarged” TLD RabGEF complexes is currently not clear.The best-studied TLD Rab-GEF is MC1, which activates Rab7 (Ypt7 in yeast) in endosomal maturation and autophagy. In yeast, it has been demonstrated that MC1-dependent recruitment of Ypt7 to both late endosomes/multivesicular bodies and autophagosomes is required for the fusion of these organelles with the vacuole and degradation of the respective cargo (7, 18). This process is conserved in plants and mammalian cells (9, 14, 19). The metazoan MC1 complex contains a third subunit, namely Bulli/RMC1; however, the function of this protein remains elusive (14–16). Importantly, Bulli/RMC1 is not required for Rab5-dependent Rab7 activation and is thus not involved in regulating the GEF activity of MC1 (20). Recently, the uncharacterized protein C5orf51 was identified as an interactor of MC1 that links the GEF complex to mitophagy (17), yet the underlying mechanism remains to be determined.We have previously identified the structure of Ypt7 bound to the MC1 core, which comprises a heterodimeric complex of the first Longin domains of Mon1 and Ccz1, respectively (21). The structure revealed a mechanism that involves remodeling of the GTPase switch regions. In the GEF-bound conformation, MC1 binding opens the nucleotide-binding pocket of Ypt7, which directs a lysine residue of Ypt7 into the Mg²⁺-binding pocket, thus favoring displacement of the bound nucleotide.Although the MC1 core is required and sufficient for the GEF activity of the complex (21), functional studies in yeast showed that the catalytic core complex was unable to rescue the vacuolar fragmentation phenotype of mon1Δ or ccz1Δ strains and did not properly localize in cells. Thus, LD2 and LD3 as well as the rest of the complex are likely involved in correct recruitment of MC1 to the proper organelle membrane. Previous studies have identified GTPases of the Rab5 family and phosphatidylinositol phosphate (PIP) lipids as recruiting factors that promote binding of MC1 to endosomal membranes (20, 22, 23). On autophagosomes, Atg8 supports Mon1-Ccz1 function (16, 18). However, the mechanistic basis underlying these processes remains unclear.To gain molecular insight into the targeting mechanism of MC1 and to understand how the complex is built in three dimensions, we determined the cryogenic electron microscopy (cryo-EM) structure of a stable MC1 complex from Chaetomium thermophilum. We observe a unique arrangement of the three LDs of each subunit within the complex and identify a conserved basic surface on MC1 that defines the orientation of the complex on lipid bilayers. Based on this, we developed a model of the function of Mon1-Ccz1 on membranes. The structure of Mon1-Ccz1 thus provides a blueprint for the architecture and function of the TLD RabGEF family.