Plant Vacuolar Trafficking O

Current Biology 24, 1375–1382, June 16, 2014 ª2014 Elsevier Ltd All rights reserved http://dx.doi.org/10.1016/j.cub.2014.05.004

Reportccurs

through Distinctly Regulated Pathways

Kazuo Ebine,1,8 Takeshi Inoue,1,8 Jun Ito,2 Emi Ito,1

Tomohiro Uemura,1 Tatsuaki Goh,3 Hiroshi Abe,4 Ken Sato,5

Akihiko Nakano,1,6 and Takashi Ueda1,7,*1Department of Biological Sciences, Graduate School ofScience, TheUniversity of Tokyo, Bunkyo-ku, Tokyo 113-0033,Japan2Graduate School of Biological Sciences, Nara Institute ofScience and Technology, Ikoma, Nara 630-0101, Japan3Department of Biology, Graduate School of Science, KobeUniversity, Kobe 657-8501, Japan4Antibiotics Laboratory, RIKEN, and RIKEN Center forSustainable Resource Science (CSRS), Gene DiscoveryResearch Group, 2-1 Hirosawa, Wako, Saitama 351-0198,Japan5Department of Life Sciences, Graduate School of Arts andSciences, The University of Tokyo, Komaba, Meguro-ku,Tokyo 153-8902, Japan6RIKEN Center for Advanced Photonics, Live Cell MolecularImaging Research Team, Extreme Photonics Research Group,2-1 Hirosawa, Wako, Saitama 351-0198, Japan7Japan Science and Technology Agency (JST), PRESTO, 4-1-8Honcho Kawaguchi, Saitama 332-0012, Japan

Summary

The multifunctional vacuole is the largest organelle in plant

cells, andmany proteins are transported to and stored in thisorganelle; thus, the vacuole has great physiological and

agronomical importance. However, the molecular mecha-

nism and regulation of plant vacuolar traffic remain largelyunknown. In this study, we demonstrate that multiple vacu-

olar trafficking pathways operate in plants. RAB5 andRAB7 are evolutionarily conserved subfamilies of Rab

GTPase [1–3], whose animal and yeast counterparts regulatevacuolar/endosomal trafficking in a sequential manner [4–7].

Functional analyses of a putative activating complex forRAB7 indicated that this complex is responsible for matura-

tion from RAB5- to RAB7-positive endosomes in plant cells.Moreover, these machinery components are recruited to a

more complex trafficking network. Mutations in RAB5 andRAB7 conferred counteracting effects on the vti11 mutant.

Furthermore, impairment of RAB5- and RAB7-dependentpathways differentially affected the transport of distinctive

cargos. These results indicate that plants have developeda complex vacuolar transport system distinct from that of

nonplant systems by assigning evolutionarily conservedmachinery to unique trafficking pathways. These pathways

provide a fundamental basis for plant development at thecellular and higher-ordered levels.

Results and Discussion

The diverse functions of plant vacuoles are fulfilled throughtight regulation of trafficking to and from the vacuoles, whichinvolves evolutionarily conserved machinery components

8Co-first author

*Correspondence: tueda@bs.s.u-tokyo.ac.jp

including Rab GTPases [8]. Although the basic framework ofthe Rab GTPase action is well conserved in eukaryotic cells[8–13], recent comparative genomic studies suggest thateach eukaryotic lineage has acquired a unique repertoire ofRab GTPases during evolution [1, 3, 14]. Plants also harbor aunique set of RabGTPases partly characterized by diversifica-tion of the RAB5 group acting in endosomal/vacuolar traf-ficking pathways [15, 16]. RAB7 is also proposed to regulatevacuolar traffic in plants [17]. In animal cells, a sequentialaction of RAB5 and RAB7 mediated by the effector complexHOPS [18] and a guanine nucleotide exchange factor (GEF)for RAB7 consisting of SAND1/Mon1 and CCZ1 [19, 20] isresponsible for the maturation from early to late endosomalcompartments [7, 21]. Conversely, how the vacuolar traffickingpathway is organized by RAB5- and RAB7-like GTPases inplant cells remainsmostly unknown. To elucidate the contribu-tion of RAB5 and RAB7 (aka RABF and RABG in Arabidopsis,respectively) in the targeting and biogenesis of vacuoles, wefirst characterized the Arabidopsis homologs of SAND1 andCCZ1. TheArabidopsis genome contains oneSAND1 homolog(SAND) and two CCZ1 homologs (CCZ1a and CCZ1b). BothCCZ1a and CCZ1b interacted with SAND in yeast cells(Figure 1A), which was confirmed by copurification of theseproteins from lysates of E. coli (Figure 1B) and coimmunopre-cipitation from the lysates of transgenic plants expressingGFP-tagged proteins (Figure S1A available online). Weattempted to detect nucleotide exchange in a RAB7 member,RABG3f, mediated by the SAND-CCZ1 complex and foundthat soluble RABG3f was not activated by this complex. Fora more efficient interaction between the RAB protein and theSAND-CCZ1 complex, we constructed C-terminally His6-tagged RABG3f and immobilized it on liposomes, which wasexpected to mimic the interaction between RABG3f and themembrane, and was successfully used for detecting the GEFactivity toward the yeast Ypt7p [22]. This strategy allowed usto detect concentration-dependent specific GEF activity ofSAND-CCZ1 for plant RAB7 (Figures 1C and S1B), thusdemonstrating that the function of the SAND-CCZ1 complexis conserved between plants and other kingdoms. The lossof function of SAND resulted in a severe growth defect, andthe double mutant ccz1a ccz1b exhibited a slightly weakergrowth defect than the sand mutants (Figures 1D and S1C).However, single ccz1 mutants were indistinguishable fromthe wild-type plant (Figure S1D), indicating redundant func-tions of CCZ1a and CCZ1b, which are coexpressedthroughout the plant (http://bar.utoronto.ca/efp/cgi-bin/efpWeb.cgi). Fluorescently tagged SAND, CCZ1a, andCCZ1b, all of which rescued the growth defects found for therespective mutants (Figure S1E), colocalized to punctatecompartments that were stained by the endocytic tracerFM4-64 in root epidermal cells (Figures 1E and 1F). Some ofthe SAND- or CCZ1-positive compartments also colocalizedwith RAB7/RABG3f in lateral root cap cells, which is consistentwith a function as a GEF for RAB7 (Figure 1G). Intriguingly,better colocalization with CCZ1a was observed for theRAB5 members ARA7/RABF2b and ARA6/RABF1 in rootepidermal cells. This result could reflect recruitment of theSAND-CCZ1 complex by active GTP-bound RAB5 to theendosome as previously reported in an animal system. Thisnotion was supported by the interaction between CCZ1b and

(legend on next page)

Current Biology Vol 24 No 121376

Multiple Trafficking Pathways for Plant Vacuoles1377

a constitutively active form of ARA7, but not RAB11-likeRABA1b, which was shown by an immunoprecipitation anal-ysis after chemical crosslinker treatment (Figure S1F). TheGolgi marker ST-mRFP localized to distinct compartmentsfrom CCZ1a, and the trans-Golgi network (TGN) marker VHA-a1-mRFP exhibited distinct but occasionally associated local-ization with CCZ1a (Figures S1G and S1H).

We then evaluated the localization of RAB7/RABG3f andRAB5 members in the relevant mutants. RABG3f was mainlylocalized to the vacuolar membrane in lateral root cap cellsof wild-type plants, and some of it was found in punctate endo-somes (Figure 2A). In ccz1a ccz1b, RABG3f failed to localize tothe vacuolar membrane but was found in the cytosol in addi-tion to punctate compartments that were not stained byFM4-64, although these compartments were frequently asso-ciated with FM-stained compartments (Figure 2A). We thenexamined the localization of RAB5 members (RHA1 andARA6) in root epidermal cells of this mutant. RAB5-positivecompartments with dilated ring-shaped structures clusteredin the cytoplasm and were stained by FM4-64, indicating thatthese structures retain endosomal properties. These resultsindicate that CCZ1 is essential for localization of RAB7 to thevacuolar membrane. Given that RAB5 localizes to small punc-tate endosomes in wild-type plants (Figure S2A), the dilatedmorphology of the RAB5-positive compartments suggestedthat transport and/or maturation from RAB5-positive endo-somes to the next destination, probably the late endosomeor vacuole, was impaired in the ccz1a ccz1b mutant. Theselines of evidence strongly support the existence of endosomaltrafficking/maturation from RAB5- to RAB7-positive endo-somes through sequential action of RAB5 and RAB7, as previ-ously proposed in both animal and plant cells [4, 18, 23, 24].RAB7/RABG3f was correctly localized to the vacuolarmembrane in the lateral root cap cells of the vps9a-2 mutant(a mutation allele of a RAB5 GEF) (Figure S2B), which mostlikely indicates that the residual GEF activity on RAB5 invps9a-2 is sufficient for RAB7 recruitment to the vacuolarmembrane. Alternatively, there could be another activatingsystem for RAB7 that is independent of RAB5. We also notedthat endocytic transport of FM4-64 to the vacuolar membranewas not remarkably impaired in root epidermal cells of ccz1accz1b (Figure S2C), which suggests the existence of a RAB7-independent vacuolar trafficking pathway in Arabidopsis.

We next examined vacuolar transport in embryos. Weobserved mature seeds with ccz1a ccz1b or vps9a-2 express-ing GFP-CT24 [25, 26] and found that GFP-CT24 was misse-creted to the extracellular space in both ccz1a ccz1b andvps9a-2 (Figure 2C). Immuno-EM further demonstrated that12S globulin was also missecreted to the extracellular spacein these mutants. This common phenotype is consistent withthe notion that GFP-CT24 and 12S globulin are transported

Figure 1. Characterization of the SAND-CCZ1 Complex

(A) The interaction between SAND and CCZ1a/b was detected by the yeast tw

(B) His-tagged CCZ1b was copurified with GST-tagged SAND from the E. coli

His-CCZ1b.

(C) Nucleotide exchange on RABG3f/RAB7 was measured in the absence (bu

VPS9a.

(D) Phenotypes of sand and ccz1a ccz1b double mutants. The vps9a-2 mutan

(E) SAND and CCZ1a/b were localized to punctate compartments and the cyto

(F) GFP-CCZ1a and SAND-TagRFP colocalized to punctate compartments (top

(middle). These punctate compartments were stained by FM4-64. Scale bars r

(G) GFP-CCZ1a partially colocalized with Venus-RABG3f (top), and RAB5 (m

(middle and bottom). Scale bars represent 5 mm.

See also Figure S1.

via the trafficking pathway regulated by the sequential actionof RAB5 and RAB7, although we cannot eliminate the possibil-ity that other pathways also contribute to the targeting of theseproteins. We also observed large aggregating structures withGFP-CT24 fluorescence in ccz1a ccz1b. EM observationindicated that these structures were large paramural bodieswith numerous exosome-like vesicles (Figure 2D). We did notobserve such structures in vps9a-2 seeds. Neither PSVmorphology nor 12S globulin processing was markedlyaffected in the ccz1 single mutants (Figures S2D and S2E).The difference in phenotype suggests that RAB7 and RAB5

have at least partially distinct functions. To confirm this possi-bility, we generated combinations of multiple mutantscorresponding to RAB7/RABG members and compared theirphenotypes with that of vps9a-2. Eight RAB7 homologs existin theArabidopsis genome,which are further divided into threesubgroups (RABG1, G2, and G3a-f) [27]. We collected T-DNAinsertion mutants corresponding to six RABG3 members inwhich expression of the corresponding genes was abolishedor substantially reduced (Figure S3A), and we generatedmultiple mutants by crossing. Some of the quadruple, quin-tuple, and sextuple mutants exhibited semidwarfism in earlydevelopmental stages, although these mutants were fertileand eventually grew to a similar size as the wild-type(Figure S3B).We then examined the processing of 12S globulin in these

mature mutant seeds to monitor the defect in trafficking tothe PSVs, and we found that precursors of 12S globulin accu-mulated in quintuple rabg3b,c,d,e,f and sextuple rabg3a,b,c,d,e,f mutants but not in the quintuple rabg3a,b,c,d,f mutant(Figure 3A). The different phenotypes of the quintuple mutantsmost likely reflect distinct expression patterns of the RABG3isoforms (Figure S3C). The vps9a-2 mutation conferred asimilar effect, which was, however, substantially strongerthan the rabg3 quintuple mutations (Figure 3A). A similarweaker effect of rabg3 mutations compared to vps9a-2 wasalso observed in the transport of GFP-CT24 (Figures 2C and3B). Intriguingly, themorphology of the PSV inmature embryosexhibited distinct sensitivity to these mutations. The PSVs inrabg3 mutants, even in mutants such as rabg3a,b,c,d,f inwhich the transport of 12S globulin was not markedly affected,exhibited severe fragmented (or unfused) morphology,whereas the vps9a-2 mutation did not markedly affect PSVmorphology (Figures 3C and S3D). A similar effect of rabg3mutations was also observed for lytic vacuoles in leafepidermal cells (Figure 3D). Thus, mutations in RAB5 andRAB7/RABG3 confer distinct effects on vacuolar transportand vacuolar morphology; the transport of soluble cargo tothe PSV is more sensitive to impairment of RAB5, and rab7/rabg3 mutations primarily affect the vacuolar morphology.These results suggest that RAB5 and RAB7 act in genetically

o-hybrid method.

lysate. The arrowheads indicate full-length GST-SAND. The arrow indicates

ffer) or presence of 200 pmol or 400 pmol of SAND-CCZ1a or 200 pmol of

t is also presented. Scale bar represents 1 cm.

sol. Scale bars represent 5 mm.

). Colocalization between GFP-CCZ1a and CCZ1b-TagRFP was confirmed

epresent 5 mm.

RFP-ARA7 and ARA6-mRFP) exhibited better colocalization with CCZ1a

Figure 2. Sequential Action of RAB5 and RAB7,

which Is Required for the Transport of Soluble

Cargos to the PSV

(A) RAB7, which was localized to the vacuolar

membrane and endosomes in wild-type plants

(WT), relocalized to disc-like organelles in the

ccz1a ccz1b double mutant. FM4-64 stained the

punctate compartments, which were associated

but not overlapping with the puncta bearing

RABG3f in ccz1a ccz1b. Scale bars represent

5 mm.

(B) RAB5 (RHA1 or ARA6)-positive endosomes

were dilated to form large ring-shaped struc-

tures, which were stained by FM4-64. Scale

bars represent 5 mm.

(C) GFP-CT24 was missecreted to the extracel-

lular space in the ccz1a ccz1b and the vps9a-2

mutants (upper panels). Immuno EMalso demon-

strated that 12S globulin is missecreted to the

extracellular spaces in these mutants (lower

panels). Large aggregating structures with GFP

fluorescence were also observed in ccz1a

ccz1b mutant embryos (asterisk). Scale bars

represent 5 mm (upper panels) and 0.5 mm (lower

panels). CW, cell wall; IC, intercellular space.

(D) Large paramural bodies with numerous exo-

some-like vesicles were observed in ccz1a

ccz1b mutant embryos by TEM (asterisks). Scale

bar represents 2 mm. OB, oil body; PSV, protein

storage vacuole. Arrowheads indicate the cell

wall.

See also Figure S2.

Current Biology Vol 24 No 121378

separable membrane trafficking events, in addition to the traf-ficking pathway involving both RAB5 and RAB7.

Additional genetic evidence supported this notion. Weexamined the genetic interaction between RAB5 or RAB7/RABG3 and VTI11 (also called ZIG), which encodes a

vacuolar/PVC Qb-SNARE. Mutations inconventional RAB5s synergisticallyexaggerated the dwarfism of zig-1, aloss-of-function mutant of VTI11 [28](Figure 4A), suggesting that RAB5 andVTI11 act in the same traffickingpathway. The ara6-1 zig-1 doublemutant was indistinguishable from thezig-1 mutant (Figure 4A). Surprisingly,the rabg3b,c,d,e,f quintuple mutationpartly suppressed abnormal pheno-types of zig-1 such as zig-zag-shapedinflorescence (Figures 4B and S4A).We also found that overexpression ofARA7, but not ARA6, partially but sig-nificantly suppressed the zig-1 pheno-types (Figures S4B and S4C). Thus,mutations in RAB5 and RAB7 exertopposite effects on zig-1. Togetherwith the distinct effects on the targetingand morphology of the PSVs mentionedabove, these results describe a vacuolartrafficking pathway in which RAB5 andRAB7 fulfill different and probably coun-teracting functions in plants.Mutations in components of the

adaptor protein 3 (AP-3) complex alsosuppress the zig-1 mutation [13]. We

therefore asked whether RAB7/RABG and AP-3 function inthe same trafficking pathway. We expressed two vacuolarmembrane proteins, GFP-SYP22 and GFP-VAMP713, inmutants defective in the RAB7- (sand) or AP-3- (ap-3dand zip4/ap-3m) dependent trafficking pathway. Vacuolar

Figure 3. Different Effects of Mutations in RAB5 and RAB7 on the Traffic to and the Morphology of the Vacuole

(A) Immunoblotting of seed proteins from wild-type (WT) and mutant plants with the indicated genotypes using the anti-12S globulin antibody.

(B) Localization of GFP-CT24 in mature WT and mutant embryos with the indicated genotypes.

(C) Morphology of protein storage vacuoles in mature WT and mutant embryos with the indicated genotypes.

(D) Vacuolar morphology visualized by GFP-VAMP713 in leaf epidermal cells of WT and mutant plants with the indicated genotypes.

Scale bars represent 10 mm. See also Figure S3.

Multiple Trafficking Pathways for Plant Vacuoles1379

localization of GFP-SYP22 was not affected by the loss offunction of AP-3d or ZIP4/AP-3m in root epidermal cells (Fig-ures 4C and S4D), suggesting that this protein reaches vacu-oles via an AP-3-independent pathway. In yeast, transport ofthe ortholog of SYP22, Vam3p, to the vacuole is known torequire AP-3 [29]. This difference is a good example of thediversification of vacuolar trafficking pathways between plantand yeast cells. The vacuolar localization of XFP-SYP22 wasalso insensitive to the sand-2 and ccz1 mutations, and frag-mentation of the vacuole was evident in these mutants

(Figures 4C and S4E). This result indicates that RAB7/RABGdoes not play a critical role in SYP22 transport. In contrast,GFP-VAMP713 was not correctly localized to the vacuolarmembrane in ap-3d and zip4, in which it accumulated at theplasma membrane and in punctate compartments in the cyto-plasm in addition to bulb-like structures inside vacuoles (Fig-ures 4C and S4D), indicating an indispensable function ofAP-3 in the proper vacuolar targeting of VAMP713. Intriguingly,the sand-2 and ccz1 mutations did not substantially alter thelocalization of VAMP713 (Figures 4C and S4E), which suggests

Figure 4. Distinct Trafficking Pathways to the Vacuole for Distinctive Cargos

(A) Wild-type (WT) and mutant plants with the indicated genotypes at

30 days of age are presented.

(B) Wild-type and mutant plants at 30 days of age and the 7th and 8th leaves

of the plants (inset) are indicated. The quintuple rabg3b,c,d,e,f mutation

(rabg3) suppressed the phenotypes of zig-1.

(C) Localization of GFP-SYP22 and GFP-VAMP713 in root epidermal cells of

wild-type (WT), sand-2, ap-3d, and vps9a-2 plants.

(D) Model of membrane trafficking to the vacuole in plant cells. The vacuolar

trafficking pathway in Arabidopsis comprises at least three trafficking

Current Biology Vol 24 No 121380

that RAB7 could be dispensable for transport of this protein tothe vacuole. These results strongly suggest that RAB7 andAP-3 function in different trafficking pathways, althoughimpairments in these pathways resulted in similar effects inzig-1 mutants.We then investigated the effects of vps9a-2 on the locali-

zation of these two proteins. GFP-VAMP713 was not markedlyaffected by vps9a-2 (Figure 4C), indicating that activation ofRAB5 is not critical for targeting of VAMP713 to the vacuole.Thus, the trafficking pathway responsible for the vacuolarlocalization of VAMP713 depends on AP-3, but not on RAB7or RAB5. Conversely, the majority of GFP-SYP22 was mistar-geted to the plasma membrane and punctate compartmentsin the cytoplasm in vps9a-2 (Figure 4C). Thus, transport ofGFP-SYP22 to the vacuole requires RAB5 activation, which,however, does not depend on AP-3 andRAB7. The localizationof VTI12, a close homolog of VTI11 residing on the TGN, alsodistinctly required RAB5- and RAB7-dependent pathways.VTI12 was partially relocalized to the vacuolar membrane inthe root epidermal cells of ccz1a ccz1b, whereas VTI12 wasmistargeted to the plasma membrane in vps9a-2 (Figure S4F).These different effects could account for the effects of themutations in RAB5 and RAB7/RABG3 on zig-1 counteractingeach other, because VTI12 relocalized to the vacuolar mem-brane can replace the function of VTI11 [30].Analyses of the SAND-CCZ1 complex indicated that plant

cells are also equipped with a vacuolar trafficking pathway inwhich RAB5 and RAB7 sequentially act (Figure 4D), whichcontributes to the transport of storage proteins to PSVs (seeaccompanying paper by Singh et al. [31] in this issue ofCurrentBiology). Conversely, evaluation of RAB7- and RAB5-relatedmutants demonstrated that multiple trafficking pathways thatdistinctly require RAB5 and RAB7 also operate in plantvacuolar trafficking (Figure 4D). The unique property of thetrafficking pathway of GFP-SYP22 should be especiallyemphasized: it requires RAB5 but not RAB7 and AP-3, andsuch a pathway has never been reported in an animal system.The transport route of GFP-VAMP713 also offers interestinginsight; this trafficking pathway requires AP-3, but activationof RAB5 and RAB7 does not seem to be critical. AP-3 is local-ized to punctate compartments that are not located in theknown endocytic pathway [32]. Thus, in addition to the routecommonly used by endocytic and vacuolar trafficking path-ways, there could be a vacuolar trafficking pathway indepen-dent of the endocytic pathway in plants. Alternatively, ourresult may indicate that the functions of RAB5 and RAB7 areinterchangeable and that either of them is sufficient for vacu-olar targeting of VAMP713. SYP22 has been proposed to betransported to the vacuole in a RAB7-dependent and RAB5-independent manner in tobacco [33], whereas our results indi-cate that SYP22 does not require RAB7 to reach the vacuole inArabidopsis. The reason for this apparent discrepancy is notclear, but it could reflect distinct trafficking systems employedin different plant species or tissues.As shown above, plants have developed a unique vacuolar

trafficking system by recruiting conserved molecular machin-ery to trafficking pathways distinct from those in nonplantsystems. Our results indicate that each trafficking route isresponsible for the transport of particular cargos (Figure 4D),

routes: (1) a route depending on the sequential action of RAB5 and RAB7,

(2) an AP-3-dependent but RAB5- and RAB7-independent pathway, and

(3) a RAB5-dependent and AP-3-independent route.

See also Figure S4.

Multiple Trafficking Pathways for Plant Vacuoles1381

suggesting that the mechanism of cargo sorting should beinvestigated in parallel with studies on docking and fusionmachinery involving Rab GTPases. Further studies to unravelthemechanisms of cargo sorting at donor organelles includingthe ER, Golgi, and TGN in addition to detailed analyses ondocking and fusion mechanisms at the target vacuolarmembrane will help to construct a whole image of the vacuolartransport system in plants that was acquired to play funda-mental roles in various plant functions.

Supplemental Information

Supplemental Information includes four figures, Supplemental Experi-

mental Procedures, and one table and can be found with this article online

at http://dx.doi.org/10.1016/j.cub.2014.05.004.

Author Contributions

K.E. performed amajor part of the experiments presented in Figures 3 and 4,

and T.I. mainly performed the experiments presented in Figures 1 and 2. J.I.,

E.I., T. Uemura, T.G., and H.A. generated transgenic plants, multiple

mutants, and antibodies used in this study. K.S. performed the GEF assay

with T.I. T. Ueda designed the study and wrote the manuscript, and A.N.

and T. Ueda supervised this study.

Acknowledgments

We thank M.T. Morita (Nagoya University), S. Utsumi (Kyoto University), I.

Hara-Nishimura (Kyoto University), and T. Demura (NAIST) for sharingmate-

rials; H.T. Sakurai (University of Tokyo), E. Furuyama (University of Tokyo),

Y. Ichikawa (RIKEN), R. Nakazawa (RIKEN), and S. Fukushima (RIKEN) for

technical support; C. Ungermann (University of Osnabruck) for precious

advice about the GEF assay; and the SALK Institute, the Max Planck

Institute, and the ABRC for providing Arabidopsis mutants. This work was

supported by Grants-in-Aid for Scientific Research from the Ministry of

Education, Culture, Sports, Science, and Technology of Japan (to A.N.

and T. Ueda), JST, PRESTO, and a Grant-in-Aid for JSPS fellows (T.I.,

195010).

Received: February 12, 2014

Revised: April 17, 2014

Accepted: May 1, 2014

Published: May 29, 2014

References

1. Dacks, J.B., Poon, P.P., and Field, M.C. (2008). Phylogeny of endocytic

components yields insight into the process of nonendosymbiotic

organelle evolution. Proc. Natl. Acad. Sci. USA 105, 588–593.

2. Klopper, T.H., Kienle, N., Fasshauer, D., and Munro, S. (2012).

Untangling the evolution of Rab G proteins: implications of a compre-

hensive genomic analysis. BMC Biol. 10, 71.

3. Mackiewicz, P., andWyroba, E. (2009). Phylogeny and evolution of Rab7

and Rab9 proteins. BMC Evol. Biol. 9, 101.

4. Bohdanowicz, M., and Grinstein, S. (2010). Vesicular traffic: a Rab

SANDwich. Curr. Biol. 20, R311–R314.

5. Solinger, J.A., and Spang, A. (2013). Tethering complexes in the endo-

cytic pathway: CORVET and HOPS. FEBS J. 280, 2743–2757.

6. Epp, N., Rethmeier, R., Kramer, L., and Ungermann, C. (2011).

Membrane dynamics and fusion at late endosomes and vacuoles—

Rab regulation, multisubunit tethering complexes and SNAREs. Eur. J.

Cell Biol. 90, 779–785.

7. Kinchen, J.M., and Ravichandran, K.S. (2010). Identification of two

evolutionarily conserved genes regulating processing of engulfed

apoptotic cells. Nature 464, 778–782.

8. Saito, C., and Ueda, T. (2009). Chapter 4: functions of RAB and SNARE

proteins in plant life. Int Rev Cell Mol Biol 274, 183–233.

9. Ackers, J.P., Dhir, V., and Field, M.C. (2005). A bioinformatic analysis of

the RAB genes of Trypanosoma brucei. Mol. Biochem. Parasitol. 141,

89–97.

10. Mizuno-Yamasaki, E., Rivera-Molina, F., and Novick, P. (2012). GTPase

networks in membrane traffic. Annu. Rev. Biochem. 81, 637–659.

11. Nakada-Tsukui, K., Saito-Nakano, Y., Husain, A., and Nozaki, T. (2010).

Conservation and function of Rab small GTPases in Entamoeba: anno-

tation of E. invadens Rab and its use for the understanding of

Entamoeba biology. Exp. Parasitol. 126, 337–347.

12. Kremer, K., Kamin, D., Rittweger, E., Wilkes, J., Flammer, H., Mahler, S.,

Heng, J., Tonkin, C.J., Langsley, G., Hell, S.W., et al. (2013). An overex-

pression screen of Toxoplasma gondii Rab-GTPases reveals distinct

transport routes to the micronemes. PLoS Pathog. 9, e1003213.

13. Niihama, M., Takemoto, N., Hashiguchi, Y., Tasaka, M., andMorita, M.T.

(2009). ZIP genes encode proteins involved in membrane trafficking of

the TGN-PVC/vacuoles. Plant Cell Physiol. 50, 2057–2068.

14. Dacks, J.B., and Field, M.C. (2007). Evolution of the eukaryotic

membrane-trafficking system: origin, tempo and mode. J. Cell Sci.

120, 2977–2985.

15. Ebine, K., Miyakawa, N., Fujimoto, M., Uemura, T., Nakano, A., and

Ueda, T. (2012). Endosomal trafficking pathway regulated by ARA6, a

RAB5 GTPase unique to plants. Small GTPases 3, 23–27.

16. Ebine, K., and Ueda, T. (2009). Unique mechanism of plant endocytic/

vacuolar transport pathways. J. Plant Res. 122, 21–30.

17. Pedrazzini, E., Komarova, N.Y., Rentsch, D., and Vitale, A. (2013). Traffic

routes and signals for the tonoplast. Traffic 14, 622–628.

18. Rink, J., Ghigo, E., Kalaidzidis, Y., and Zerial, M. (2005). Rab conversion

as a mechanism of progression from early to late endosomes. Cell 122,

735–749.

19. Nordmann, M., Cabrera, M., Perz, A., Brocker, C., Ostrowicz, C.,

Engelbrecht-Vandre, S., and Ungermann, C. (2010). The Mon1-Ccz1

complex is the GEF of the late endosomal Rab7 homolog Ypt7. Curr.

Biol. 20, 1654–1659.

20. Gerondopoulos, A., Langemeyer, L., Liang, J.R., Linford, A., and Barr,

F.A. (2012). BLOC-3 mutated in Hermansky-Pudlak syndrome is a

Rab32/38 guanine nucleotide exchange factor. Curr. Biol. 22, 2135–

2139.

21. Poteryaev, D., Datta, S., Ackema, K., Zerial, M., and Spang, A. (2010).

Identification of the switch in early-to-late endosome transition. Cell

141, 497–508.

22. Cabrera, M., Nordmann, M., Perz, A., Schmedt, D., Gerondopoulos, A.,

Barr, F., Piehler, J., Engelbrecht-Vandre, S., and Ungermann, C. (2014).

The Mon1-Ccz1 GEF activates the Rab7 GTPase Ypt7 via a longin-fold-

Rab interface and association with PI3P-positive membranes. J. Cell

Sci. 127, 1043–1051.

23. Limpens, E., Ivanov, S., van Esse, W., Voets, G., Fedorova, E., and

Bisseling, T. (2009). Medicago N2-fixing symbiosomes acquire the

endocytic identity marker Rab7 but delay the acquisition of vacuolar

identity. Plant Cell 21, 2811–2828.

24. Bottanelli, F., Gershlick, D.C., and Denecke, J. (2012). Evidence for

sequential action of Rab5 and Rab7 GTPases in prevacuolar organelle

partitioning. Traffic 13, 338–354.

25. Nishizawa, K., Maruyama, N., Satoh, R., Fuchikami, Y., Higasa, T., and

Utsumi, S. (2003). A C-terminal sequence of soybean beta-conglycinin

alpha’ subunit acts as a vacuolar sorting determinant in seed cells.

Plant J. 34, 647–659.

26. Fuji, K., Shimada, T., Takahashi, H., Tamura, K., Koumoto, Y., Utsumi,

S., Nishizawa, K., Maruyama, N., and Hara-Nishimura, I. (2007).

Arabidopsis vacuolar sorting mutants (green fluorescent seed) can be

identified efficiently by secretion of vacuole-targeted green fluorescent

protein in their seeds. Plant Cell 19, 597–609.

27. Pereira-Leal, J.B., and Seabra, M.C. (2001). Evolution of the Rab family

of small GTP-binding proteins. J. Mol. Biol. 313, 889–901.

28. Kato, T., Morita, M.T., Fukaki, H., Yamauchi, Y., Uehara, M., Niihama,M.,

and Tasaka, M. (2002). SGR2, a phospholipase-like protein, and ZIG/

SGR4, a SNARE, are involved in the shoot gravitropism of

Arabidopsis. Plant Cell 14, 33–46.

29. Cowles, C.R., Odorizzi, G., Payne, G.S., and Emr, S.D. (1997). The AP-3

adaptor complex is essential for cargo-selective transport to the yeast

vacuole. Cell 91, 109–118.

30. Niihama, M., Uemura, T., Saito, C., Nakano, A., Sato, M.H., Tasaka, M.,

and Morita, M.T. (2005). Conversion of functional specificity in Qb-

SNARE VTI1 homologues of Arabidopsis. Curr. Biol. 15, 555–560.

31. Singh, M.K., Kruger, F., Beckmann, H., Brumm, S., Vermeer, J.E.M.,

Munnik, T., Mayer, U., Stierhof, Y.-D., Grefen, C., Schumacher, K., and

Jurgens, G. (2014). Protein delivery to vacuole requires SAND protein-

dependent Rab GTPase conversion for MVB-vacuole fusion. Curr.

Biol. 24, 1383–1389.

Current Biology Vol 24 No 121382

32. Feraru, E., Paciorek, T., Feraru, M.I., Zwiewka, M., De Groodt, R., De

Rycke, R., Kleine-Vehn, J., and Friml, J. (2010). The AP-3 b adaptin

mediates the biogenesis and function of lytic vacuoles in Arabidopsis.

Plant Cell 22, 2812–2824.

33. Bottanelli, F., Foresti, O., Hanton, S., and Denecke, J. (2011). Vacuolar

transport in tobacco leaf epidermis cells involves a single route for

soluble cargo and multiple routes for membrane cargo. Plant Cell 23,

3007–3025.

Note Added in Proof

A related paper by Cui et al. entitled ‘‘Rab7 activation by the MON1-CCZ1

complex is essential for PVC-to-vacuole trafficking and plant growth in

Arabidopsis’’ has been accepted for publication in The Plant Cell. In this

paper, the authors reported the sequential action of RAB5 and RAB7 medi-

ated by the SAND/MON1-CCZ1 complex in a vacuolar trafficking pathway,

which is consistent with our findings.