Rab proteins - core.ac.uk · Rab proteins form the largest branch of the Ras superfamily of GTPases. ... vary from one cell type to another. Nevertheless, some of them are cell type-

Rab proteins

Olivier Martinez, Bruno Goud *Laboratory `Molecular mechanisms of intracellular transport', UMR 144, Institut Curie, 26 rue d'Ulm, 75248 Paris Cedex 05, France

Received 20 November 1997; accepted 11 December 1997

Abstract

Rab proteins form the largest branch of the Ras superfamily of GTPases. They are localized to the cytoplasmic face oforganelles and vesicles involved in the biosynthetic/secretory and endocytic pathways in eukaryotic cells. It is now wellestablished that Rab proteins play an essential role in the processes that underlie the targeting and fusion of transport vesicleswith their appropriate acceptor membranes. However, the recent discovery of several putative Rab effectors, which are notrelated to each other and which fulfil diverse functions, suggests a more complex role for Rab proteins. At least two Rabproteins act at the level of the Golgi apparatus. Rab1 and its yeast counterpart Ypt1 control transport events through earlyGolgi compartments. Work from our laboratory points out a role for Rab6 in intra-Golgi transport, likely in a retrogradedirection. ß 1998 Elsevier Science B.V. All rights reserved.

Keywords: Rab GTPase; Intracellular transport; Golgi apparatus

1. Introduction

The ¢rst rab gene (YPT1) was identi¢ed in 1983 byD. Gallwitz et al. as an ORF between actin andtubulin genes in Saccharomyces cerevisiae [1]. Fouryears later, A. Salminen and P. Novick provided the¢rst evidence that a Ras-like GTPase (Sec4) was di-rectly involved in a vesicular transport process. Thesame year, A. Tavitian and colleagues cloned the ¢rsthomologs of SEC4/YPT in mammals, termed rab(ras-like in rat brain) genes. Fifteen years on, smallGTPases of the Rab/Ypt family form the largestbranch of the p21 ras superfamily and are recognizedas key players of the protein machinery involved invesicular transport and organelle dynamics in eu-karyotic cells. Many of them have been localized to

various compartments of both the biosynthetic/secre-tory and endocytic pathways. In vivo and in vitrostudies performed in mammalian cells with Rab pro-teins locked in their GDP- or GTP-bound conforma-tions, as well as genetic studies in yeast, have led tothe hypothesis that Rab proteins are involved indocking/fusion of transport vesicles with their targetmembranes. Good evidence also exists that Rab pro-teins ful¢l their function by cycling between a GDP-bound cytosolic and a GTP-bound membrane form.Nevertheless, their exact function is still poorlyunderstood. Rab proteins may participate in, ormodulate the formation of, the v/t-SNARE (vesicle/target-soluble N-ethylmaleimide-sensitive factor ac-ceptor receptor) complexes. These complexes, formedof proteins found on the membrane of transportvesicles (VAMP/synaptotagmin family) and on themembrane of acceptor compartments (syntaxin fam-ily), are involved in the speci¢c recognition of trans-port vesicles with their target membranes [2]. In ad-

0167-4889 / 98 / $19.00 ß 1998 Elsevier Science B.V. All rights reserved.PII: S 0 1 6 7 - 4 8 8 9 ( 9 8 ) 0 0 0 5 0 - 0

* Corresponding author. Fax: +33 (1) 42 34 63 82;E-mail : [email protected]

BBAMCR 14325 5-8-98 Cyaan Magenta Geel Zwart

Biochimica et Biophysica Acta 1404 (1998) 101^112

dition, several proteins, unrelated to each other, havebeen recently identi¢ed as putative Rab e¡ectors.This suggests a more complex role than previouslysuspected for the Rab family of proteins. We willfocus here on recent advances in the Rab ¢eld,with special emphasis on Rab proteins involved intransport at the level of the Golgi apparatus.

2. Localization of Rab proteins

Most of the Rab proteins identi¢ed so far are

ubiquitous, although their level of expression mayvary from one cell type to another. Nevertheless,some of them are cell type- or tissue-speci¢c, suchas Rab3A which is only expressed in neurons andneuro-endocrine cells [3,4], or Rab17 which is onlydetected in epithelial cells [5]. Many Rabs have nowbeen localized using speci¢c antibodies by a combi-nation of cell fractionation, immuno£uorescence andimmunoelectron microscopy approaches. We sum-marize in Fig. 1 and Table 1 what is currently knownabout the intracellular localization of Rabs in mam-mals and Sec4/Ypt proteins in yeast. All of the or-

Fig. 1. Hypothetical eukaryotic cell showing the localization and function of Rab proteins. RER: rough endoplasmic reticulum; IC:intermediate compartment; TGN: trans-Golgi network; SV: synaptic vesicles; EE: early endosomes; LE: late endosomes; L: lyso-somes; RecyE: recycling endosomes.


O. Martinez, B. Goud / Biochimica et Biophysica Acta 1404 (1998) 101^112102

ganelles (except lysosomes) involved in biosynthetic/secretory and endocytic pathways carry at least oneRab protein on their cytoplasmic face. In most cases,Rabs are associated with two adjacent compartmentsand some of them have been found on trans-port vesicles, such as clathrin-coated vesicles. Thesedata support the hypothesis that Rab proteinsare involved in targeting and/or docking/fusionof transport vesicles with their acceptor compart-ments.

3. Rab cycle

The functional cycle of Rab proteins is currentlyrelatively well known (Fig. 2). Like other GTPases,Rabs cycle between a GDP-bound (ò¡') and a GTP-bound (òn') conformation. The GDP/GTP exchangereaction is likely catalyzed by GEF (guanine ex-change factor) proteins. Two putative exchange fac-tors (Dss4 and its mammalian counterpart Mss4)have been identi¢ed [6^8]. Dss4 is active on Sec4and Ypt1, but not on Rab3A. Mss4 displays an ex-

Table 1Localization and function of Rab proteins in yeast and mammals

Protein Localization Function References

Sec4 post-Golgi vesicles, plasma membrane Golgi to plasma membrane transport [72^74]Ypt1 ER, early Golgi ER to Golgi, intra-Golgi transport [48,75,76]Ypt31, Ypt32 Golgi? intra-Golgi, Golgi to vacuole transport? [66]Ypt51, Ypt52,Ypt53

endosomes transport in early endocytic pathway [77]

Ypt6 ? intra-Golgi, Golgi to vacuole transport [65,67]Ypt7 ? transport in late endocytic pathway,

homotypic fusion of vacuoles[78^80]

Rab1A, Rab1B intermediate compartment, CGN ER to Golgi and intra-Golgi transport [45,47,49,81^83]Rab2 intermediate compartment, CGN ER to Golgi transport [47,84]Rab3A synaptic vesicles, secretory granules regulated exocytosis in neurons and

neuro-endocrine cells[3,4,85,86]

Rab3B synaptic vesicles? regulated exocytosis in neurons? [87,88]tight junctions of polarized epithelial cells ?

Rab3C synaptic vesicles similar to rab3A? [89]Rab4A, Rab4B early endosomes recycling from early endosomes to plasma

membrane[90]

Rab5A, Rab5B, plasma membrane, clathrin-coated vesicles, plasma membrane to early endosome [84,91^95]Rab5C early endosomes transport

fusion of early endosomesRab6 medial/trans-Golgi, TGN intra-Golgi retrograde transport? [54,55,57^59,61,63]Rab7 late endosomes early to late endosome transport [96,97]

late endosome to lysosome transportRab8 post-Golgi basolateral vesicles, tight junction Golgi to basolateral plasma membrane [98,99]

in polarized cells transport in polarized cellsRab9 late endosomes, TGN late endosome to TGN transport [100,101]Rab10 Golgi and post-Golgi vesicles ? [68]Rab11 TGN, recycling endosomes, post-Golgi transport through recycling endosomes [102,103]

vesicles endosomes to TGN transport?Rab12 Golgi complex? ? [104]Rab13 tight junctions in polarized epithelial cells ? [105]Rab17 basolateral plasma membrane in epithelial

cells, apical endosomestranscytosis? [5]

Rab24 ER, cis-Golgi, late endosomes ? [104]Rab27 retinal cells ? [106]


O. Martinez, B. Goud / Biochimica et Biophysica Acta 1404 (1998) 101^112 103

change activity towards a subset of Rab proteinswhich, interestingly, are all involved in the biosyn-thetic/secretory pathway. However, exchange activ-ities of both Dss4 and Mss4 are very low comparedto values reported for GEFs of other members of theRas superfamily. Recently, proteins with higher ex-change activities have been characterized. Takai andcoworkers have puri¢ed from rat brain a 178 kDacytosolic protein speci¢c for prenylated Rab3A [9].Sec2p, an essential protein required for Golgi to plas-ma membrane tra¤c, mediates nucleotide exchangeon Sec4 [10]. Sec2p shares sequence homology withRabin-3, a 50 kDa protein that speci¢cally interactswith GDP-bound forms of Rab3A and Rab3D [11].However, Rabin-3 has no detectable exchange factoractivity. Finally, Zerial and coworkers have isolateda 60 kDa protein that forms a tight complex withrabaptin-5, a Rab5 e¡ector (see below), and catalyzesGDP/GTP exchange upon delivery of Rab5 to mem-branes [12]. This protein, termed Rabex-5, is homol-

ogous to Vps9, a yeast protein involved in endocy-tosis.

The intrinsic GTPase activity of Rab proteins isvery low, suggesting the involvement of GAP(GTPase activating) proteins in order to stimulateGTP hydrolysis. Rab GAP activities have been de-tected in various extracts [13^15]. However, only twocDNAs encoding for Rab GAPs have been charac-terized so far. One of them encodes the S. cerevisiaeGyp6 protein that stimulates GTP hydrolysis byYpt6, but not by Rhy1 or Rab6, the S. pombe andmammalian counterparts of Ypt6 ([16]; P. Mollatand B. Goud, unpublished results). More recently,a 130 kDa cytosolic GAP protein active on all mem-bers of the Rab3 family has been puri¢ed from ratbrain cytosol [17].

A characteristic of Rab proteins is that a cycle ofassociation with and dissociation from membranes issuperimposed onto their GDP/GTP cycle. In the cy-tosol, Rab proteins are maintained in their GDP-

Fig. 2. Rab cycle. Cytosolic Rab:GDP/GDI complexes are targeted to their speci¢c compartments (organelles and/or transportvesicles). After transient association of complexes, GDI is released through interaction with GDF and GDP/GTP exchange is cata-lyzed by GEF. Rab:GTP interacts with various e¡ectors (see text). GTP is then hydrolyzed after interaction with a GAP protein.GDI can remove Rab:GDP from membranes and Rab:GDP/GDI complexes are reutilized in a new cycle. GDI: GDP dissociation in-hibitor; GDF: GDI displacement factor; GEF: exchange factor; GAP: GTPase activating protein.



bound conformation through an interaction withGDI (GDP dissociation inhibitor) proteins. Onlyone GDI (Gdi1p) has been identi¢ed in yeast [18],but at least three (Rab3A GDI/GDIK/GDI-1, GDI-2and GDIL) and probably ¢ve GDIs exist in mam-mals [19^21]. The precise function of these GDI iso-forms is at present unknown, but they may serve to¢ne tune Rab function. GDI proteins play an impor-tant role in speci¢c delivery of Rab proteins to theirtarget membranes. For instance, Rab5-GDI andRab9-GDI complexes have been shown to mediatein vitro the binding of Rab5 and Rab9 to organellesin which these proteins normally reside [22,23]. An-other important property of GDIs is their capacity toextract GDP-bound forms of Rab proteins frommembranes. Altogether, these data suggest thatGDIs are chaperones that accompany Rab duringtheir cycle between membrane and cytosol.

The molecular mechanisms involved in membraneattachment of Rab proteins are still largely unclear.GDP/GTP exchange does not seem to be a prereq-uisite for Rab binding, since Rabs in their GDP-bound conformation can be transiently detected inmembranes after delivery of Rab-GDI complexes[22,23]. More likely, GDP/GTP exchange followsbinding of Rab:GDP-GDI complexes and releaseof GDI. A GDI displacement factor active on a sub-set of Rab proteins and devoid of exchange activityhas recently been characterized [24]. An importantbut still unsolved issue is whether saturable `recep-tors' for Rabs are present on membranes. The evi-dence for such `receptors', that may interact with theC-terminal hypervariable domain of Rab proteins[25], has been recently documented in the case ofRab4 [26].

4. Rab/Ypt e¡ectors

Within the past few years, a lot of e¡ort has beenput into identifying Rab/Ypt e¡ectors, i.e. proteinsthat would preferentially bind to Rabs in their GTP-bound conformation.

In mammalian cells, biochemical approaches andyeast two-hybrid screens have been extensively usedwith Rab3A, Rab5, Rab6, Rab8 and Rab9 proteins.The best characterized e¡ectors so far are Rabphilinand Rabaptin-5, which interact with Rab3A and

Rab5, respectively. Rabphilin, ¢rst isolated bycross-linking of bovine brain extracts, has domainhomology with synaptotagmin (two C2-like domainsinvolved in binding of Ca2� and phospholipids) [27].Its localization at the surface of synaptic vesicles isconsistent with that of Rab3A [28]. In addition, Rab-philin expression is decreased in mice lacking Rab3A[29]. The exact function of Rabphilin is still unclear.It has been shown to inhibit GAP-induced GTP hy-drolysis (although Rabphilin alone weakly stimulatesGTP hydrolysis by Rab3A), suggesting that Rabphil-in function is to maintain Rab3A in its GTP-boundconformation [30]. Rab3A:GTP may also recruitRabphilin on membranes, although this view hasbeen challenged [27,31]. Recently, Rabphilin wasshown to stimulate the activity of K-actinin tocross-link actin ¢laments (formation of bundles)[32]. This points to a function of Rab3A in the re-organization of the actin cytoskeleton, a processknown to be essential for exocytosis.

Rabaptin-5 was found in a yeast two-hybrid screenusing the GTPase mutant of Rab5 (Rab5 Q79L) as abait [33]. It is a 115 kDa, mainly cytosolic protein.Rabaptin-5 is recruited from the cytosol to endoso-mal membranes by Rab5:GTP. In addition, its over-expression induces the formation of large endosomes,as does overexpression of Rab5 Q79L. Rabaptin-5also stabilizes Rab5 in its GTP-bound conformation.Interestingly, Rabaptin-5 is part of a large cytosoliccomplex required for membrane docking/fusion. Oneprotein of this complex (p60) shares homology withyeast Vps9 and acts as an exchange factor for Rab5[12].

Three other putative Rab e¡ectors have been re-cently characterized: a serine/threonine protein kin-ase, closely related to the GC kinase found in lym-phoid germinal centers [34]; Rim, a zinc ¢ngerprotein localized to presynaptic active zones in syn-apses and to synaptic ribbons in ribbon synapses[35], and a novel protein termed p40 [36]. These pro-teins interact speci¢cally with GTP-bound forms ofRab8, Rab3A and Rab9, respectively.

In the case of Rab6, a yeast two-hybrid screenwith the GTPase mutant Rab6 Q72L has allowedus to identify ¢ve novel proteins that interact withRab6:GTP. One of them corresponds to a 150 kDacytosolic protein. Like Rabaptin-5, this protein has alarge coiled-coil domain, but unlike Rabaptin-5, it



does not seem to be recruited on membranes by ac-tive forms of Rab6 (I. Janoueix-Lerosey, F. Jollivetand B. Goud, unpublished results). Another proteinwas identi¢ed as a new member of the kinesin-likeprotein family. This protein, which we called Rab-kinesin-6, is found associated with the Golgi appara-tus and plays a role in the dynamics of this organelle[37].

In yeast, genetic studies point out a role for Sec4/Ypt proteins in the interaction between v- and t-SNARES (see other reviews in this issue). A searchfor genes that compensate the loss of YPT1 has al-lowed the identi¢cation of a SLY1 allele (SLY1-1), amember of the SEC1 gene family whose productsinteract with t-SNARES (in that case Sed5) [38,39].In a YPT1 null mutant, the overexpression of v-SNAREs involved in endoplasmic reticulum (ER)to Golgi transport (Sec22/Sly2 and Bet1/Sly12) com-pensates for the Ypt1 defect [38]. The suppressivee¡ect is stronger when Bos1 (another v-SNARE ofER-derived vesicles, [40]) is co-expressed with eitherSec22 or Bet1 [41]. Although the exact role of Ypt1 isnot yet fully understood, it seems that Ypt1 functionis crucial for the formation of the v/t SNARE com-plex [42]. Ypt1 could promote the association be-tween Sec22 and Bos1, two v-SNARES, suggestinga role in the `priming' of transport vesicles [41]. Ypt1may also promote the dissociation of the t-SNARESed5 from Sly1 and therefore favors the associationof Sed5 with v-SNARES [43].

Genetic evidence also exists that Sec4 is involvedin the formation of v/t-SNARE complexes requiredfor exocytosis of Golgi-derived transport vesicles.For instance, the overexpression of Sec9, a memberof the SNAP 25 family that interacts with plasmamembrane t-SNARES, suppresses the e¡ect of a mu-tation in the e¡ector domain of Sec4 [44].

5. Rabs and Golgi

Two Rab proteins, Rab1 and Rab6, and theiryeast counterparts Ypt1 and Ypt6 have been shownto play a role in transport events at the level of theGolgi apparatus. Rab1 is a ubiquitous Rab foundassociated with pre-Golgi elements, as shown by itscolocalization with the p58 protein [45]. Such a local-ization is consistent with a role of Rab1 in ER to

Golgi transport. For instance, overexpression of theGDP-bound mutant of Rab1 (Rab1 S17N) stronglyinhibits the delivery of VSV-G to cis-Golgi compart-ments [46,47]. Similarly, it is now well establishedthat Ypt1 is involved in targeting/fusion of ER-de-rived vesicles to Golgi compartments in yeast. Inter-estingly, Rab 1 can also be detected on the cytoplas-mic face of one or two cisternae at the cis-face of theGolgi stacks [45]. In addition, it has been docu-mented that Rab1 as well as Ypt1 functions in earlyintra-Golgi transport in addition to ER to Golgitransport [48,49]. This illustrates the fact that at leastsome Rab proteins can be involved in more than onevesicular transport step. However, it should bepointed out that two isoforms of Rab1 (Rab1Aand Rab1B) exist in mammals and more work isneeded to determine whether these isoforms are func-tionally redundant.

Like Rab1, Rab6 is a ubiquitous protein expressedin virtually all cell types or tissues tested so far inmammals. Rab6 homologs have been found in awide variety of eukaryotic cells, including yeast S.cerevisiae (Ypt6) and S. pombe (Rhy1) [50,51], plantcells [52] and the protozoan malaria parasite Plasmo-dium falciparum [53]. Rab6 is found associated withmedial and trans-Golgi cisternae as well as withmembranes of the trans-Golgi network (TGN)[54,55]. The fractionation of pancreas microsomalfractions on sucrose density gradients indicates thatRab6 can be well separated from Rab1 in these gra-dients, Rab6 being enriched in the low density trans-Golgi fraction [45]. It should be pointed out, how-ever, that Rab6 was found associated with all theGolgi stack in liver cells [56]. In addition, Rab6can also be detected in post-Golgi vesicles in somespecialized cell types such as mouse hypothalamicneurons, Torpedo marmorata electrocytes or retinalcells [54,55,57^59].

The relatively wide distribution of Rab6 has led tospeculation that there may be several roles for thisprotein [60^62]. Rab6 seems, however, to be primar-ily involved in intra-Golgi transport. Speci¢cally, theoverexpression of the GTPase de¢cient mutant Rab6Q72L and to a lesser extent that of wild-type Rab6 inHeLa cells and mouse ¢broblasts has been shown toseverely impair transport of secretory markers be-tween cis/medial and late Golgi compartments, with-out a¡ecting their transport between the TGN and



the plasma membrane [61]. In addition, a strikinge¡ect of the overexpression of GTP-bound forms ofRab6 is the loss of Golgi structures and the redis-tribution of Golgi resident enzymes into the endo-plasmic reticulum [63]. This e¡ect was found to bealmost indistinguishable from those induced by bre-feldin A (including the need for microtubule integ-rity), with the notable di¡erence that even TGNmarkers redistribute to the ER in cells overexpressingGTP-bound forms of Rab6. The redistribution ofGolgi proteins seen in cells overexpressing GTP-bound forms of Rab6 may be a consequence of ablock of anterograde transport that would unbalanceanterograde and retrograde £ow in favor of retro-grade transport. Alternatively, Rab6 may be directlyinvolved in the retrograde transport (Fig. 3). In favorof this latter hypothesis, we have recently observedthat GDP-bound forms of Rab6 blocks the transportof Shiga toxin B fragment (a protein shown to be

transported retrogradely throughout the Golgi appa-ratus, [64]) at the TGN level, i.e. its site of entry intothe Golgi apparatus (L. Johannes and B. Goud, un-published results).

It is still unclear whether yeast Ypt6 has the samefunction as mammalian Rab6, although Rab6 cancomplement a YPT6 null strain [51]. The YPT6null strain is viable at 30³C, indicating that YPT6is not an essential gene in yeast. At higher temper-ature, vacuoles are fragmented and a partial missort-ing of carbopeptidase Y can be observed, suggestingthat Ypt6 is involved in Golgi to vacuole transport[51]. However, a nonsense mutation in YPT6 gives aslightly di¡erent phenotype than the null allele, i.e.an inhibition of both early Golgi function and ribo-some biosynthesis [65]. As discussed above, we be-lieve that Rab6 is primarily involved in retrogradetransport between Golgi and ER and we obtainedno evidence for a role of Rab6 in Golgi to endosome

Fig. 3. Model of Rab6 function. Rab6 is localized to medial/trans-Golgi cisternae and membranes of the TGN [54,55]. The overex-pression of Rab:GTP (wt and the GTPase mutant Rab6 Q72L) induces a redistribution of Golgi resident proteins into the ER and aloss of Golgi structures [63]. Such an e¡ect may be explained by a stimulation of retrograde transport that results in an imbalance be-tween anterograde and retrograde intra-Golgi membrane £ow. In contrast, the GDP-bound mutant Rab6 T27N does not redistributeGolgi membranes into the ER, but may inhibit the recycling of Golgi resident enzymes [63]. Several putative Rab6 e¡ectors have beenidenti¢ed in our laboratory. One of them is a new Golgi-associated kinesin-like protein, termed Rabkinesin-6. One of the functions ofRab6 could be to regulate the association/dissociation of Rabkinesin-6 to/from microtubules (MT) and, as a consequence, the bindingof Golgi membranes and/or Golgi-derived vesicles to microtubules [37].



transport by following a marker of this pathway inHeLa cells (a chimeric protein made of the ectodo-main of hemagglutinin and the cytoplasmic tail ofmannose-6P receptor) (O. Martinez, B. Goud andB. Ho£ack, unpublished results). Without rulingout the possibility that the apparent discrepancy be-tween the function of Rab6/Ypt6 in yeast and mam-mals can be explained by di¡erent experimental strat-egies, this might also be due to functional diversity ofRab/Ypt proteins in yeast and mammals. For in-stance, some evidence exists that Ypt5 and Rab5do not regulate the same transport pathway. Thesame holds true for Ypt31/Ypt32, which share stronghomology to Rab11 [66]. In favor of a hypothesis offunctional diversity between Rab and Ypt proteins, itis also interesting to note that Rab6 partners that wehave recently identi¢ed in our laboratory using atwo-hybrid screen display no sequence homologywith suppressor genes of YPT6 (IMH1 and SYSgenes) [65,67].

Two other Rab proteins, Rab10 and Rab12, andtwo isoforms of Ypt3 (Ypt31 and 32) are likely to beinvolved in transport events at the level of the Golgiapparatus [66,68]. However, very little is yet knownabout the precise function of these proteins.

6. What function for Rab proteins?

Rab GTPases were ¢rst considered proteins thatensured that a transport vesicle leaving one intracel-lular compartment is deposited in the appropriateacceptor compartment [69]. In its simplest form,this model predicts that Rab proteins act as molec-ular tags to specify the right acceptor compartmentfor a given transport vesicle. The ¢nding that a chi-meric protein of Sec4 and Ypt1 can function as ei-ther Sec4 or Ypt1 without grossly a¡ecting the cor-rect functioning of the yeast secretory pathway hasrendered this model less likely [70]. In addition, thespeci¢city of docking/fusion of vesicles with their tar-get compartments is believed to be mediated by otherfamilies of proteins (v/t-SNARES) [2]. The functionof Rab proteins now appears more complex thanpreviously thought. As discussed above, good evi-dence exists that in yeast, Sec4 and Ypt1 play animportant role in the formation of v/t-SNARE com-plexes. Rab3A has also been shown to stabilize

SNARE synaptic complex in Aplysia [71]. However,it is striking to note that Rab e¡ectors identi¢ed sofar are unrelated proteins with diverse functions. Forinstance, the recent discovery that Rab6 interactswith a kinesin-like protein raises the possibility ofcoordinated action between members of these twofamilies of proteins [37]. In addition, Rab proteinsmay be part of large complexes (such as the èxocyst'in the case of Sec4). Much more work is needed toidentify all the proteins of these complexes for aparticular Rab. However, a tentative hypothesis isthat Rab proteins coordinate spatially and tempo-rally all the events underlying movement and target-ing/fusion of transport vesicles and membranes with-in eukaryotic cells. If true, one can predict that Rabproteins will be (re)considered in the near future toplay a central role in organelle dynamics and intra-cellular transport.

Acknowledgements

We thank Ludger Johannes and Gordon Langsleyfor critical reading of the manuscript.

References

[1] D. Gallwitz, C. Donath, C. Sander, A yeast gene encoding aprotein homologous to the human c-has/bas proto-oncogeneproduct, Nature 306 (1983) 704^707.

[2] J.E. Rothman, Mechanisms of intracellular protein trans-port, Nature 372 (1994) 55^63.

[3] F. Darchen, A. Zahraoui, F. Hammel, M.P. Monteils, A.Tavitian, D. Scherman, Association of the GTP-binding pro-tein Rab3A with bovine adrenal chroma¤n granules, Proc.Natl. Acad. Sci. USA 87 (1990) 5692^5696.

[4] G. Fischer von Mollard, G.A. Mignery, M. Baumert, M.S.Perin, T.J. Hanson, P.M. Burger, R. Jahn, T.C. Su«dhof,rab3 is a small GTP-binding protein exclusively localizedto synaptic vesicles, Proc. Natl. Acad. Sci. USA 87 (1990)1988^1992.

[5] A. Lutcke, S. Jansson, R.G. Parton, P. Chavrier, A. Valen-cia, L.A. Huber, E. Lehtonen, M. Zerial, Rab17, a novelsmall GTPase, is speci¢c for epithelial cells and is inducedduring cell polarization, J. Cell Biol. 121 (1993) 553^564.

[6] M. Moya, D. Roberts, P. Novick, DSS4-1 is a dominantsuppressor of sec4-8 that encodes a nucleotide exchangeprotein that aids Sec4p function, Nature 361 (1993) 460^463.

[7] J. Burton, D. Roberts, M. Montaldi, P. Novick, P. De Ca-



milli, A mammalian guanine-nucleotide-releasing protein en-hances function of yeast secretory protein Sec4, Nature 361(1993) 464^467.

[8] J.L. Burton, M.E. Burns, E. Gatti, G.J. Augustine, P. DeCamilli, Speci¢c interactions of Mss4 with members of theRab GTPase subfamily, EMBO J. 13 (1994) 5547^5558.

[9] M. Wada, H. Nakanishi, A. Satoh, H. Hirano, H. Obaishi,Y. Matsuura, Y. Takai, Isolation and characterization of aGDP/GTP exchange protein speci¢c for the Rab3 subfamilysmall G proteins, J. Biol. Chem. 272 (1997) 3875^3878.

[10] C. Walch-Solimena, R.N. Collins, P.J. Novick, Sec2p medi-ates nucleotide exchange on Sec4p and is involved in pola-rized delivery of post-Golgi vesicles, J. Cell Biol. 137 (1997)1495^1509.

[11] W.H. Brondyk, C.J. McKiernan, K.A. Fortner, P. Stabila,R.W. Holz, I.G. Macara, Interaction cloning of Rabin3, anovel protein that associates with the Ras-like GTPaseRab3A, Mol. Cell. Biol. 15 (1995) 1137^1143.

[12] H. Horiuchi, R. Lippe, H. McBride, M. Rubino, P. Wood-man, H. Stenmark, V. Rybin, M. Wilm, K. Ashman, M.Mann, M. Zerial, A novel Rab5 GDP/GTP exchange factorcomplexed to Rabaptin-5 links nucleotide exchange to e¡ec-tor recruitment and function, Cell 90 (1997) 1149^1159.

[13] E. Burstein, I. Macara, Characterization of a guanine nu-cleotide-releasing factor and a GTPase-activating proteinthat are speci¢c for the ras-related protein p25Arab3A, Proc.Natl. Acad. Sci. USA 89 (1992) 1154^1158.

[14] T.J. Tan, P. Vollmer, D. Gallwitz, Identi¢cation and partialpuri¢cation of GTPase-activating proteins from yeast andmammalian cells that preferentially act on Ypt1/Rab1 pro-teins, FEBS Lett. 291 (1991) 322^326.

[15] N.C. Walworth, P. Brennwald, A.K. Kabcenell, M. Garrett,P. Novick, Hydrolysis of GTP by Sec4 protein plays animportant role in vesicular transport and is stimulated by aGTPase-activating protein in Saccharomyces cerevisiae, Mol.Cell. Biol. 12 (1992) 2017^2028.

[16] M. Strom, P. Vollmer, T.J. Tan, D. Gallwitz, A yeastGTPase-activating protein that interacts speci¢cally with amember of the Ypt/Rab family, Nature 361 (1993) 736^739.

[17] K. Fukui, T. Sasaki, K. Imazumi, Y. Matsuura, H. Naka-nishi, Y. Takai, Isolation and characterization of a GTPaseactivating protein speci¢c for the Rab3 subfamily of small Gproteins, J. Biol. Chem. 272 (1997) 4655^4658.

[18] M.D. Garrett, J.E. Zahner, C.M. Cheney, P.J. Novick,GDI1 encodes a GDP dissociation inhibitor that plays anessential role in the yeast secretory pathway, EMBO J. 13(1994) 1718^1728.

[19] N. Nishimura, H. Nakamura, Y. Takai, K. Sano, Molecularcloning and characterization of two rab GDI species fromrat brain: brain-speci¢c and ubiquitous types, J. Biol. Chem.269 (1994) 14191^14198.

[20] A. Shisheva, T.C. Sudhof, M.P. Czech, Cloning, character-ization, and expression of a novel GDP dissociation inhib-itor isoform from skeletal muscle, Mol. Cell. Biol. 14 (1994)3459^3468.

[21] I. Janoueix-Lerosey, F. Jollivet, J. Camonis, P.N. Marche,

B. Goud, Two-hybrid system screen with the small GTP-binding protein Rab6. Identi¢cation of a novel mouseGDP dissociation inhibitor isoform and two other potentialpartners of Rab6, J. Biol. Chem. 270 (1995) 14801^14808.

[22] O. Ullrich, H. Horiuchi, C. Bucci, M. Zerial, Membraneassociation of Rab5 mediated by GDP-dissociation inhibitorand accompanied by GDP/GTP exchange [see comments],Nature 368 (1994) 157^160.

[23] T. Soldati, A.D. Shapiro, A.B. Svejstrup, S.R. Pfe¡er, Mem-brane targeting of the small GTPase Rab9 is accompaniedby nucleotide exchange [see comments], Nature 369 (1994)76^78.

[24] A.B. Dirac-Svejstrup, T. Sumizawa, S.R. Pfe¡er, Identi¢ca-tion of a GDI displacement factor that releases endosomalRab GTPases from Rab-GDI, EMBO J. 16 (1997) 465^472.

[25] P. Chavrier, J.P. Gorvel, E. Stelzer, K. Simons, J. Gruen-berg, M. Zerial, Hypervariable C-terminal domain of rabproteins acts as a targeting signal, Nature 353 (1991) 769^772.

[26] N. Ayad, M. Hull, I. Mellman, Mitotic phosphorylation ofrab4 prevents binding to a speci¢c receptor on endosomesmembranes, EMBO J. 16 (1997) 4497^4507.

[27] H. Shirataki, T. Yamamoto, S. Hagi, H. Miura, H. Oishi, Y.Jin-no, T. Senbonmatsu, Y. Takai, Rabphilin-3A is associ-ated with synaptic vesicles through a vesicle protein in amanner independent of Rab3A, J. Biol. Chem. 269 (1994)32717^32720.

[28] A. Mizoguchi, Y. Yano, H. Hamaguchi, H. Yanagida, C.Ide, A. Zahraoui, H. Shirataki, T. Sasaki, Y. Takai, Local-ization of Rabphilin-3A on the synaptic vesicle, Biochem.Biophys Res. Commun. 202 (1994) 1235^1243.

[29] M. Geppert, V.Y. Bolshakov, S.A. Siegelbaum, K. Takei, P.De Camilli, R.E. Hammer, T.C. Sudhof, The role of Rab3Ain neurotransmitter release, Nature 369 (1994) 493^497.

[30] S. Kishida, H. Shirataki, T. Sasaki, M. Kato, K. Kaibuchi,Y. Takai, Rab3A GTPase-activating protein-inhibiting activ-ity of Rabphilin-3A, a putative Rab3A target protein, J. Biol.Chem. 268 (1993) 22259^22261.

[31] B. Stahl, J.H. Chou, C. Li, T.C. Sudhof, R. Jahn, Rab3reversibly recruits rabphilin to synaptic vesicles by a mecha-nism analogous to raf recruitment by ras, EMBO J. 15(1996) 1799^1809.

[32] M. Kato, T. Sakaki, T. Ohya, H. Nakanishi, H. Nishioka,M. Imamura, Y. Takai, Physical and functional interactionof rabphilin-3A with alpha-actinin, J. Biol. Chem. 271 (1996)31775^31778.

[33] H. Stenmark, G. Vitale, O. Ullrich, M. Zerial, Rabaptin-5 isa direct e¡ector of the small GTPase Rab5 in endocyticmembrane fusion, Cell 83 (1995) 423^432.

[34] M. Ren, J. Zeng, C. De Lemos-Chiarandini, M. Rosenfeld,M. Adesnik, D.D. Sabatini, In its active form, the GTP-binding protein rab8 interacts with a stress-activated proteinkinase, Proc. Natl. Acad. Sci. USA 93 (1996) 5151^5155.

[35] Y. Wang, M. Okamoto, F. Schmitz, K. Hofmann, T. Sud-horf, Rim is a putative Rab3 e¡ector in regulating synaptic-vesicle fusion, Nature 388 (1997) 593^598.



[36] E. Diaz, F. Schimmoller, S. Pfe¡er, A novel rab9 e¡ectorrequired for endosome to TGN transport, J. Cell Biol. 138(1997) 283^290.

[37] A. Echard, F. Jollivet, O. Martinez, J.-J. Lacape©re, A. Rous-selet, I. Janoueix-Lerosey, B. Goud, Interaction of a Golgi-associated kinesin-like protein with Rab6, Science 79 (1998)580^587.

[38] C. Dascher, R. Ossig, D. Gallwitz, H.D. Schmitt, Identi¢ca-tion and structure of four yeast genes (SLY) that are able tosuppress the functional loss of YPT1, a member of the RASsuperfamily, Mol. Cell. Biol. 11 (1991) 872^885.

[39] M. Sogaard, K. Tani, R.R. Ye, S. Geromanos, P. Tempst, T.Kirchhausen, J.E. Rothman, T. Sollner, A rab protein isrequired for the assembly of SNARE complexes in the dock-ing of transport vesicles, Cell 78 (1994) 937^948.

[40] J.P. Lian, S. Ferro-Novick, Bos1p, an integral membraneprotein of the endoplasmic reticulum to Golgi transportvesicles, is required for their fusion competence, Cell 73(1993) 735^745.

[41] J.P. Lian, S. Stone, Y. Jiang, P. Lyons, S. Ferro-Novick,Ypt1p implicated in v-SNARE activation, Nature 372(1994) 698^701.

[42] M. SÖgaard, K. Tani, R.R. Ye, S. Geromanos, P. Tempst, T.Kirchhausen, J.E. Rothman, T. So«llner, A rab protein isrequired for the assembly of SNARE complexes in the dock-ing of transport vesicles, Cell 78 (1994) 937^948.

[43] V.V. Lupashin, G. Waters, t-SNARE activation throughtransient interaction with a Rab-like GTPase, Science 276(1997) 1255^1258.

[44] P. Brennwald, B. Kearns, K. Champion, S. Keranen, V.Bankaitis, P. Novick, Sec9 is a SNAP-25-like componentof a yeast SNARE complex that may be the e¡ector ofSec4 function in exocytosis, Cell 79 (1994) 245^258.

[45] J. Saraste, U. Lahtinen, B. Goud, Localization of the smallGTP-binding protein rab1p to early compartments of thesecretory pathway, J. Cell Sci. 108 (1995) 1541^1552.

[46] H. Plutner, A.D. Cox, S. Pind, R. Khovravi-Far, J.R.Bourne, R. Schwaninger, C.J. Der, W.E. Balch, Rab1b reg-ulates vesicular transport between the endoplasmic reticulumand successive Golgi compartments, J. Cell Biol. 115 (1991)31^43.

[47] E.J. Tisdale, J.R. Bourne, R. Khosravi-Far, C.J. Der, W.E.Balch, GTP-binding mutants of rab1 and rab2 are potentinhibitors of vesicular transport from the endoplasmic retic-ulum to the Golgi complex, J. Cell Biol. 119 (1992) 749^761.

[48] G. Jedd, C. Richardson, R. Litt, N. Segev, The Ypt1GTPase is essential for the ¢rst two steps of the yeast secre-tory pathway, J. Cell Biol. 131 (1995) 583^590.

[49] C. Nuo¡er, H.W. Davidson, J. Metteson, J. Meinkoth, W.E.Balch, A GDP-bond form of rab1 inhibits protein exportfrom the endoplasmic reticulum and transport between Gol-gi compartments, J. Cell Biol. 125 (1994) 225^237.

[50] L. Hengst, T. Lehmeier, D. Gallwitz, The ryh1 gene in the¢ssion yeast Schizosaccharomyces pombe encoding a GTP-binding protein related to ras, rho and ypt: structure, ex-

pression and identi¢cation of its human homologue, EMBOJ. 9 (1990) 1949^1955.

[51] L. Hengst, PhD Thesis, Philipps-Universita«t Marburg/Lahn,1992.

[52] S.Y. Bednarek, T.L. Reynolds, M. Schroeder, R. Grabow-ski, L. Hengst, D. Gallwitz, N.V. Raikhel, A small GTP-binding protein from Arabidopsis thaliana functionally com-plements the yeast YPT6 null mutant, Plant Physiol. 104(1994) 591^596.

[53] F. Alves de Castro, G.E. Ward, R. Jambou, G. Attal, V.Mayau, G. Jaureguiberry, C. Braun-Breton, D. Chakrabarti,G. Langsley, A family of Rab G-proteins in Plasmodiumfalciparum, Mol. Biochem. Parasitol. 80 (1996) 77^78.

[54] B. Goud, A. Zahraoui, A. Tavitian, J. Saraste, Small GTP-binding protein associated with Golgi cisternae, Nature 345(1990) 553^556.

[55] C. Antony, C. Cibert, G. Geraud, A. Santa Maria, B. Maro,V. Mayau, B. Goud, The small GTP-binding protein rab6pis distributed from medial Golgi to the trans-Golgi networkas determined by a confocal microscopic approach, J. CellSci. 103 (1992) 785^796.

[56] G. Feldmann, A.-M. Durand-Schneider, B. Goud, Behav-iour of the small GTP-binding protein rab6 in the liver ofnormal rats and rats presenting an acute in£ammatory reac-tion, Biol. Cell. 83 (1995) 121^125.

[57] D. Deretic, D.S. Papermaster, Rab6 is associated with acompartment that transport rhodopsin from the trans-Golgito the site of rod outer segment disk formation in retinalphotoreceptors, J. Cell Sci. 106 (1993) 803^813.

[58] B. Jasmin, B. Goud, G. Camus, J. Cartaud, The low molec-ular weight guanosine triphosphate-binding protein Rab6passociates with distinct post-Golgi vesicles in Torpedo mar-morata electrocytes, Neuroscience 49 (1992) 849^855.

[59] A. Tixier-Vidal, A. Barret, R. Picart, V. Mayau, D. Voqt, B.Wiedenman, B. Goud, The small GTP-binding protein rab6pis associated with both Golgi and post-Golgi synaptophysin-containing membranes during synaptogenesis of hypothal-amic neurones in culture, J. Cell Sci. 105 (1993) 935^947.

[60] S.M. Jones, J.R. Crosby, J. Salamero, K.E. Howell, A cyto-solic complex of p62 and rab6 associates with TGN38/41and is involved in budding of exocytic vesicles from thetrans-Golgi network, J. Cell Biol. 122 (1993) 775^788.

[61] O. Martinez, A. Schmidt, J. Salamero, B. Ho£ack, M. Roa,B. Goud, The small GTP-binding protein rab6 functions inintra-Golgi transport, J. Cell Biol. 127 (1994) 1575^1588.

[62] T. Mayer, N. Touchot, Z. Elazar, Transport between cis andmedial Golgi cisternae requires the function of the Ras-re-lated GTPase Rab6, J. Biol. Chem. 271 (1996) 16097^16103.

[63] O. Martinez, C. Antony, G. Pehau-Arnaudet, E.G. Berger,J. Salamero, B. Goud, GTP-bound forms of rab6 induce theredistribution of Golgi proteins into the endoplasmic retic-ulum, Proc. Natl. Acad. Sci. USA 94 (1997) 1828^1833.

[64] D. Johannes, D. Tenza, C. Antony, B. Goud, Retrogradetransport of KDEL-bearing B-fragment of Shiga toxin,J. Biol. Chem. 272 (1997) 19554^19561.

[65] B. Li, J.R. Warner, Mutation of the Rab6 homologue of



Saccharomyces cerevisiae, YPT6, inhibits both early Golgifunction and ribosome biosynthesis, J. Biol. Chem. 271(1996) 16813^16819.

[66] M. Benli, F. Doring, D.G. Robinson, X. Yang, D. Gallwitz,Two GTPase isoforms, Ypt31p and Ypt32p, are essential forGolgi function in yeast, EMBO J. 15 (1996) 6460^6475.

[67] M. Tsukada, D. Gallwitz, isolation and characterization ofSYS genes from yeast, multicopy suppressors of the func-tional loss of the transport GTPase Ypt6p, J. Cell Sci. 109(1996) 2471^2481.

[68] Y.T. Chen, C. Holcomb, H.P. Moore, Expression and local-ization of two low molecular weight GTP-binding proteins,Rab8 and Rab10, by epitope tag, Proc. Natl. Acad. Sci.USA 90 (1993) 6508^6512.

[69] H.R. Bourne, Do GTPases direct membrane tra¤c in secre-tion?, Cell 55 (1988) 59^61.

[70] P. Brennwald, P. Novick, Interactions of three domains dis-tinguishing the ras-related GTP-binding proteins Ypt1 andSec4, Nature 362 (1993) 560^563.

[71] L. Johannes, F. Doussau, A. Clabecq, J.-P. Henry, F.Darchen, B. Poulain, Evidence for a functional link betweenRab3 and the SNARE complex, J. Cell Sci. 109 (1996) 2875^2884.

[72] B. Goud, A. Salminen, N.C. Walworth, P.J. Novick, AGTP-binding protein required for secretion rapidly associ-ates with secretory vesicles and the plasma membrane inyeast, Cell 53 (1988) 753^768.

[73] A. Salminen, P.J. Novick, A ras-like protein is required for apost-Golgi event in yeast secretion, Cell 49 (1987) 527^538.

[74] N. Walworth, B. Goud, A. Kabcenell, P. Novick, Mutation-al analysis of SEC4 suggests a cyclical mechanism for theregulation of vesicular tra¤c, EMBO J. 8 (1989) 1685^1693.

[75] N. Segev, J. Mulholland, D. Botstein, The yeast GTP-bind-ing YPT1 protein and a mammalian counterpart are associ-ated with the secretion machinery, Cell 52 (1988) 915^924.

[76] D. Baker, L. Wuestehube, R. Schekman, D. Botstein, N.Segev, GTP-binding Ypt1 protein and Ca2+ function inde-pendently in a cell-free protein transport reaction, Proc.Natl. Acad. Sci. USA 87 (1990) 355^359.

[77] B. Singer-Kruger, H. Stenmark, A. Dusterhoft, P. Philipp-sen, J.S. Yoo, D. Gallwitz, M. Zerial, Role of three rab5-likeGTPases, Ypt51p, Ypt52p, and Ypt53p, in the endocytic andvacuolar protein sorting pathways of yeast, J. Cell Biol. 125(1994) 283^298.

[78] H. Wichmann, L. Hengst, D. Gallwitz, Endocytosis in yeast:evidence for the involvement of a small GTP-binding protein(Ypt7p), Cell 71 (1992) 1131^1142.

[79] F. Schimmoller, H. Riezman, Involvement of Ypt7p, a smallGTPase, in tra¤c from late endosome to the vacuole inyeast, J. Cell Sci. 106 (1993) 823^830.

[80] A. Mayer, W. Wickner, Docking of yeast vacuoles is cata-lyzed by the ras-like GTPase Ypt7p after symmetric primingby Sec18p (NSF), J. Cell Biol. 136 (1997) 307^317.

[81] H. Plutner, H.W. Davidson, J. Saraste, W.E. Balch, Mor-phological analysis of protein transport from the ER toGolgi membranes in digitonin-permeabilized cells : role of

the P58 containing compartment, J. Cell Biol. 119 (1992)1097^1116.

[82] S.N. Pind, C. Nuo¡er, J.M. McCa¡ery, H. Plutner, H.W.Davidson, M.G. Farquhar, W.E. Balch, Rab1 and Ca2� arerequired for the fusion of carrier vesicles mediating endo-plasmic reticulum to Golgi transport, J. Cell Biol. 125(1994) 239^252.

[83] B.S. Wilson, C. Nuo¡er, J.L. Meinkoth, M. McCa¡ery, J.R.Feramisco, W.E. Balch, M.G. Farquhar, A Rab1 mutanta¡ecting guanine nucleotide exchange promotes disassemblyof the Golgi apparatus, J. Cell Biol. 125 (1994) 557^571.

[84] P. Chavrier, R.G. Parton, H.P. Hauri, K. Simons, M. Zerial,Localization of low molecular weight GTP binding proteinsto exocytic and endocytic compartments, Cell 62 (1990) 317^329.

[85] R.W. Holz, W.H. Brondyk, R.A. Senter, L. Kuizon, I.G.Macara, Evidence for the involvement of Rab3A inCa(2+)-dependent exocytosis from adrenal chroma¤n cells,J. Biol. Chem. 269 (1994) 10229^10234.

[86] L. Johannes, P.M. Lledo, M. Roa, J.D. Vincent, J.P. Henry,F. Darchen, The GTPase Rab3a negatively controls calcium-dependent exocytosis in neuroendocrine cells, EMBO J. 13(1994) 2029^2037.

[87] P.-M. Lledo, P. Vernier, J.-D. Vicent, W.T. Mason, R. Zor-ec, Inhibition of Rab3B expression attenuates Ca2�-depend-ant exocytosis in rat anterior pituitary cells, Nature 364(1993) 540^544.

[88] E. Weber, G. Berta, A. Tousson, P. St John, M.W. Green,U. Gopalokrishnan, T. Jilling, E.J. Sorscher, T.S. Elton,D.R. Abrahamson et al., Expression and polarized targetingof a rab3 isoform in epithelial cells, J. Cell Biol. 125 (1994)583^594.

[89] G. Fischer von Mollard, B. Stahl, A. Khokhlatchev, T.C.Sudhof, R. Jahn, Rab3C is a synaptic vesicle protein thatdissociates from synaptic vesicles after stimulation of exocy-tosis, J. Biol. Chem. 269 (1994) 10971^10974.

[90] P. van der Sluijs, M. Hull, P. Webster, P. Male, B. Goud, I.Mellman, The small GTP-binding protein rab4 controls anearly sorting event on the endocytic pathway, Cell 70 (1992)729^740.

[91] J. Gorvel, P. Chavrier, M. Zerial, J. Gruenberg, Rab5 con-trols early endosome fusion in vitro, Cell 64 (1991) 915^925.

[92] C. Bucci, R.G. Parton, I.H. Mather, H. Stunnenberg, K.Simons, B. Ho£ack, M. Zerial, The small GTPase rab5 func-tions as a regulatory factor in the early endocytic pathway,Cell 70 (1992) 715^728.

[93] G.P. Li, P.D. Stahl, Structure-function relationship of thesmall GTPase rab5, J. Biol. Chem. 268 (1993) 24475^24480.

[94] H. Stenmark, R.G. Parton, O. Steele-Mortimer, A. Lutcke,J. Gruenberg, M. Zerial, Inhibition of rab5 GTPase activitystimulates membrane fusion in endocytosis, EMBO J. 13(1994) 1287^1296.

[95] H. Horiuchi, A. Giner, B. Ho£ack, M. Zerial, A GDP/GTPexchange-stimulatory activity for the Rab5-RabGDI com-plex on clathrin-coated vesicles from bovine brain, J. Biol.Chem. 270 (1995) 11257^11262.



[96] P. Chavrier, M. Vingron, C. Sander, K. Simons, M. Zerial,Molecular cloning of YPT1/SEC4-related cDNAs from anepithelial cell line, Mol. Cell. Biol. 10 (1990) 6578^6585.

[97] E. Papini, B. Satin, C. Bucci, M. de Bernard, J.L. Telford,R. Manetti, R. Rappuoli, M. Zerial, C. Montecucco, Thesmall GTP binding protein rab7 is essential for cellularvacuolation induced by Helicobacter pylori cytotoxin,EMBO J. 16 (1997) 15^24.

[98] L.A. Huber, S. Pimplikar, R.G. Parton, H. Virta, M. Ze-rial, K. Simons, Rab8, a small GTPase involved in vesiculartra¤c between the TGN and the basolateral plasma mem-brane, J. Cell Biol. 123 (1993) 35^45.

[99] J. Peranen, P. Auvinen, H. Virta, R. Wepf, K. Simons,Rab8 promotes polarized membrane transport through re-organization of actin and microtubules in ¢broblasts,J. Cell. Biol. 135 (1996) 153^167.

[100] D. Lombardi, T. Soldati, M.A. Riederer, Y. Goda, M.Zerial, S.R. Pfe¡er, Rab9 functions in transport betweenlate endosomes and the trans Golgi network, EMBOJ. 12 (1993) 677^682.

[101] M.A. Riederer, T. Soldati, A.D. Shapiro, J. Lin, S.R.Pfe¡er, Lysosome biogenesis requires rab9 function and

receptor recycling from endosomes to the trans-Golgi net-work, J. Cell Biol. 125 (1994) 573^582.

[102] S. Urbe, L.A. Huber, M. Zerial, S.A. Tooze, R.G. Parton,Rab11, a small GTPase associated with both constitutiveand regulated secretory pathways in PC12 cells, FEBS Lett.334 (1993) 175^182.

[103] O. Ullrich, S. Reinsch, S. Urbe, M. Zerial, R.G. Parton,Rab11 regulates recycling through the pericentriolar recy-cling endosome, J. Cell Biol. 135 (1996) 913^924.

[104] V.M. Olkkonen, P. Dupree, I. Killisch, A. Lutcke, M. Ze-rial, K. Simons, Molecular cloning and subcellular local-ization of three GTP-binding proteins of the rab subfamily,J. Cell Sci. 106 (1993) 1249^1261.

[105] A. Zahraoui, G. Joberty, M. Arpin, J.-J. Fontaine, R.Hellio, A. Tavitian, D. Louvard, A small rab GTPaseis distributed in cytoplasmic vesicles in non polarizedcells but colocalizes with the tight junction marker ZO-1in polarized epithelial cells, J. Cell Biol. 124 (1994) 101^115.

[106] M.C. Seabra, Y.K. Ho, J.S. Anant, De¢cient geranylgera-nylation of Ram/Rab27 in choroidemia, J. Biol. Chem. 270(1995) 24420^24427.



Documents

Rab proteins - core.ac.uk · Rab proteins form the largest branch of the Ras superfamily of GTPases. ... vary from one cell type to another. Nevertheless, some of them are cell type-