Aspergillus RabB Rab5 integrates acquisition of degradative identity with the long distance movement of early endosomes

Abstract

Aspergillus nidulans early endosomes display characteristic long-distance bidirectional motility. Simultaneous dual-channel acquisition showed that the two Rab5 paralogues RabB and RabA colocalize in these early endosomes and also in larger, immotile mature endosomes. However, RabB-GTP is the sole recruiter to endosomes of Vps34 PI3K (phosphatidylinositol-3-kinase) and the phosphatidylinositol-3-phosphate [PI(3)P] effector AnVps19 and rabB Delta, leading to thermosensitivity prevents multivesicular body sorting of endocytic cargo. Thus, RabB is the sole mediator of degradative endosomal identity. Importantly, rabB Delta, unlike rabA Delta, prevents early endosome movement. As affinity experiments and pulldowns showed that RabB-GTP recruits AnVps45, RabB coordinates PI(3)P-dependent endosome-to-vacuole traffic with incoming traffic from the Golgi and with long-distance endosomal motility. However, the finding that Anvps45 Delta, unlike rabB Delta, severely impairs growth indicates that AnVps45 plays RabB-independent functions. Affinity chromatography showed that the CORVET complex is a RabB and, to a lesser extent, a RabA effector, in agreement with GST pulldown assays of AnVps8. rabB Delta leads to smaller vacuoles, suggesting that it impairs homotypic vacuolar fusion, which would agree with the sequential maturation of endosomal CORVET into HOPS proposed for Saccharomyces cerevisiae. rabB Delta and rabA Delta mutations are synthetically lethal, demonstrating that Rab5-mediated establishment of endosomal identity is essential for A. nidulans.

Figure 1.

RabB localizes to motile punctate cytosolic structures. (A) Epifluorescence image of GFP-RabB corresponding to a still frame of Supplementary Movie 1. The kymograph illustrating the bidirectional motility of early endosomes corresponds to a 21-μm-long, 18-pixel-wide line covering the whole width of the hypha. (B) Epifluorescence image of mRFP-RabB (Supplementary Movie 2). (C) GDP-locked GFP-RabB^Ser31Asn mutant localizes to the cytosol and to faintly labeled reticulate structures corresponding to ER. Two nuclei are indicated with arrows. (D) Mutant GFP-RabB carrying Ala substitutions of C-terminal Cys residues (Cys216 and Cys218) that undergo geranylgeranylation is strictly cytosolic. Expression of GFP-tagged mutant and wild-type RabBs was driven by the alcA^p. Bar, 5 μm.

Figure 2.

RabB and RabA colocalize on endosomes. (A) Red and green channels of a cell coexpressing mRFP-RabB and GFP-RabA that had been treated with benomyl, which prevents EE movement and leads to aggregation of EEs. The arrow in the magnified “merge” box indicates one rare example of a structure containing undetectable levels of RabB. (B) Dual-View still frames corresponding to Supplementary Movie 3. The merge of GFP and mRFP channels is shown. The kymographs below, corresponding to the dotted line, illustrate the essentially complete colocalization of RabA and RabB in motile endosomes. Note that a minor proportion of mRFP-RabB, seen after contrasting the images for color merge, is sorted to the vacuoles. Faintly labeled red vacuoles are seen in kymographs as fuzzy vertical bars, contrasting with strongly labeled, static mature endosomes. (C) Frames of a time-lapse sequence showing an arrowed endosome that contains RabA and RabB, moving retrogradely at ∼3 μm/s.

Figure 3.

Growth phenotypes of null alleles of genes involved in early endosome function. (A) Strains were cultured on complete medium (MCA) plates that were incubated for 72 h at the indicated temperatures before being photographed. (B) Growth phenotype of the same strains on synthetic complete (SC) medium. Left, pictures were taken after a 72-h incubation at 42°C (for 37°C see also Figure 9G); right, radial colony growth of the indicated strains at 42°C. Diameters of colonies of the indicated genotypes were measured after 24, 48, and 72 h incubation at 42°C. Horizontal lines indicate average diameters (in mm) for each time point and genetic condition. Differences were statistically significant (p <0.001) for all pair combinations. (C) Apical extension rate of wild-type (n = 77) and rabBΔ (n = 51) tips at 42°C. Differences are statistically significant; *p < 0.001, Student's t test. (D) Nomarski images of wild-type and rabBΔ cells cultured at 42°C.

Figure 4.

rabBΔ results in vacuolar fragmentation. (A) Conidiospores of wild-type and rabBΔ strains were germinated at 25°C. The resulting germlings were stained with CMAC, which labels the lumen of vacuoles, and photographed by epifluorescence microscopy using a Leica D filter set. The contours of the germlings are indicated with gray lines. Insets, close-up (2× magnifications of the large fields) of the regions (dotted squares) corresponding to the basal conidiospores (conidium, indicated with arrows). (B) Quantification of vacuolar size. Germlings of similar length were segmented into four different regions. One was the basal conidiospore, and the three other were 10-μm-long regions increasingly distant from the basal conidiospore, as indicated. The average vacuolar diameter for each region was measured. Error bars, SE.

Figure 5.

rabB, but not rabA, is required for early endosome motility. (A) Cells were imaged within 5–10 min after FM4-64 loading to label endosomes. In the wild-type, generally smaller early endosomes are motile, whereas larger endosomes are static. rabBΔ reduced the proportion of endosomes moving in either basipetal or acropetal direction nearly six times. This marked reduction is illustrated by the kymographs on the right (time dimension, 15 s), corresponding to Supplementary Movies 4 (wild-type) and 5 (rabBΔ). For unknown reasons, FM4-64 appears to be either less fluorescent or more susceptible to photobleaching in the plasma membrane of the mutant than in the wild-type. Note that because FM4-64 labels both early and late endosomes, the proportion of static structures is higher than that seen when early endosome-specific GFP-RabA or GFP-RabB are used as reporters. (B) GFP-RabA and GFP-RabB EEs visualized in wild-type and mutant germlings, all shown at the same magnification (size bar in left image). In this panel, kymographs correspond to 10–13 s Movies (Supplementary Movies 6 through 9) and are shown at the same scale as the corresponding still image.

Figure 6.

AnVps19 and AnVps34 are RabB, but not RabA, effectors. GST pulldown assays with the indicated GST-Rab protein baits. Extracts from cells expressing AnVps19-(HA)3 (in S. cerevisiae Vps19p is also known as Vac1p) and AnVps34-(HA)3 at physiological levels were incubated with GST-RabA or GST-RabB glutathione-Sepharose beads, loaded with GTPγS or GDP. The proteins copurifying with the baits were analyzed by anti-HA Western blots. Input lanes contain 5% of the total material used for pulldowns.

Figure 7.

rabBΔ early endosomes are deficient in PI(3)P. (A) The PI(3)P reporter consists of a C-terminal fusion of GFP to two tandem copies of the AnVps27 FYVE domain. (B) (FYVE^Vps27)₂::GFP labels faintly the membrane of basal vacuoles (open arrowheads), which frequently showed closely associated (FYVE^Vps27)₂::GFP punctae (arrows). (C) Tip-proximal region of a wild-type hypha showing localization of (FYVE^Vps27)₂::GFP to small punctae. The kymograph insert (Supplementary Movie 10) illustrates how a proportion of these punctae show the characteristic bidirectional motility of EEs. Relatively immotile ones represent, in all likelihood, late endosomes. (D and E) In rabBΔ cells (FYVE^Vps27)₂::GFP localizes mainly to the cytosol and, in tip-distal regions, also to vacuolar membranes. The few punctae that were discernible against the cytosolic fluorescence haze were immotile (inserted kymograph). (F) rabAΔ cell showing the localization of (FYVE^Vps27)₂::GFP to punctate structures, as in the wild type. Some punctate structures show characteristic EE motility (kymograph).

Figure 8.

RabB is required for MVB sorting of the endocytic degradation cargo AgtA but not for the endocytic recycling of the SynA synaptobrevin homologue. (A) Wild-type cells expressing AgtA-GFP were cultured on 5 mM glutamate synthetic complete medium to induce the synthesis of the permease (0-min time point) and then shifted to the same medium containing 10 mM NH₄⁺, which shuts off agtA transcription and promotes the endocytic down-regulation of the transporter. The positions of vacuoles were revealed with CMAC. (B) As above, using rabBΔ cells. Note the nearly complete localization of AgtA-GFP to the plasma membrane in glutamate-cultured cells (0 min). On shifting cells to ammonium, the permease is internalized to punctate cytosolic structures, which do not show any overlap with vacuoles stained with CMAC (merge; vacuoles shown in magenta; AgtA-GFP, green). Images are maximal intensity projections of z-stacks. (C) Western blot analyses of AgtA-GFP in cells cultured on glutamate (0 min) and shifted to ammonium for the indicated time points. The blot was reacted with anti-GFP antibody. (D) Western blot analyses of AgtA-(HA)3 cells cultured on glutamate (0 min) and shifted to ammonium for the indicated time points. The blot was reacted with anti-HA antibodies. The graph shows a quantitative analysis of AgtA-(HA)3 signals as a function of time. Approximately equal loading in the different lanes was confirmed by Ponceau staining of the nitrocellulose membrane (E) rabBΔ does not prevent the polarization of SynA^Snc1-GFP to the apical plasma membrane. Shown are maximal intensity projections of z-stacks contrasted with the “sharpen” function of Metamorph. Bars, (A, B, and E) 5 μm.

Figure 9.

RabB and, to a lesser extent, RabA are able to recruit CORVET, but only RabB recruits AnVps45. (A) A. nidulans extracts were run through glutathione-Sepharose columns loaded with GTPγS- or GDP-RabB. Proteins retained by the columns were eluted and analyzed by SDS-PAGE and silver staining. The identity of the different bands was determined by MS of bands excised from preparative gels (Supplementary Table 4). (B) As in A but using GST-RabA–loaded affinity columns. (C) A region of an SDS-PAGE gel corresponding to the eluates of GTPγS- and GDP-RabA columns. The gel was run to allow resolution of the GST-RabA and GdiA bands. (D) Pulldown assays (as in Figure 6) of AnGdiA-(HA)3–containing cell extracts using the indicated protein baits. Input:,10% of the material used for the pulldown. (E) Pulldown assays of AnVps8-(HA)3–containing cell extracts using the indicated protein baits. Input, 5% of the material used for the pulldown. RabE is the A. nidulans orthologue of S. cerevisiae Ypt31p, and RabN the A. nidulans orthologue of S. cerevisiae Ypt7p. The faster mobility band detected with AnVps8-(HA)3 very likely results from partial proteolysis during protein extraction/incubation (AnVps8 is a large 177-kDa protein) (F) Pulldown assays of AnVps45-(HA)3, as above. (G) Strains of the indicated genotypes were cultured on complete (MCA) or synthetic complete (SC) medium.

Figure 10.

rabBΔ and rabAΔ are synthetically lethal. (A) Analysis of the progeny of a cross between parental strains with genotypes rabAΔ::pyrG^Af pyroA4 pyrG89 (thus pyridoxine requiring) and rabBΔ::pyroA^Af pyroA4 pyrG89 (thus pyrimidine requiring). Left, normal size colony arising from the progeny of this cross plated on synthetic complete medium (i.e., containing the nutritional supplements required by all parental and recombinant progeny genotypes). Middle and right, microcolonies of the progeny plated on double selective medium. Only rabAΔ::pyrG^Af, rabBΔ::pyroA^Af double mutants were expected to grow in the absence of both pyrimidines and pyridoxine. Note the minute size of the colonies (colonies in left and middle pictures are at the same magnification). The enlarged region shown on the right illustrates the submillimetric size of the colonies. (B) Protoplasts of the above rabBΔ::pyroA^Af pyrG89 pyroA4 strain were transformed with a rabAΔ::pyrG^Af deletion cassette. Heterokaryotic, prototrophic clones (hk) were mostly recovered on minus-pyrimidine plates alongside with spontaneous diploid/merodiploids (di) arising during transformation. (C) Single nuclei segregate into conidiospores. Conidiospore transfers from heterokaryons gave rise to mycelium on pyrimidine-supplemented but not on pyrimidine-deficient MCA plates, indicating that, in the rabBΔ background, the rabAΔ::pyrG^Af deletion is lethal (single rabAΔ or rabBΔ mutants are viable on complete medium, Figure 3). In contrast, merodiploid/diploid strains (di) grew normally on both media.

Figure 11.

A model of A. nidulans EE maturation. After membrane recruitment of RabB and its activation by GTP, early endosomal domains acquire degradative identity because RabB-GTP (red lollypops) engages the Vps34 complex, which synthesizes PI(3)P (gray circles). This leads to subsequent recruitment to EEs of “early” effectors including the FYVE domain protein Vps19 and the SM protein Vps45, which mediate fusion of Golgi-derived traffic with endosomes. The also FYVE domain protein Vps27, the gatekeeper of the MVB pathway, is incorporated from this early stage, and primes ESCRT-I–, -II–, and -III–mediated inward vesicle budding (this work; motile early endosomes contain Vps32; see Galindo et al., 2007). On formation of RabB domains, motile endosomes undergo long-distance movement on MTs. In a subsequent step, CORVET is recruited to Rab5 domains, mediating the Vps8-dependent fusion of endosomes as proposed (Markgraf et al., 2009), which leads to a progressive increase in size with a concomitant loss of long-distance motility. Once endosomes reach a certain degree of maturation, Ypt7 replaces RabB (Vps21p), HOPS is exchanged for CORVET (Peplowska et al., 2007), and LEs become competent to undergo homotypic fusion (not shown).