Biochimica et Biophysica Acta 1793 (2009) 615–624

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

j ourna l homepage: www.e lsev ie r.com/ locate /bbamcr

Review

Delivery of endocytosed membrane proteins to the lysosome

Paul R. Pryor a,⁎, J. Paul Luzio b,⁎a Department of Biology (Area 9), University of York, York, YO10 5YW, UKb Cambridge Institute for Medical Research and Department of Clinical Biochemistry, University of Cambridge, Addenbrooke's Hospital, Hills Road, Cambridge, CB2 0XY, UK

⁎ Corresponding authors. P.R. Pryor is to be contactJ.P. Luzio, tel.: +44 1223 336780.

E-mail addresses: pp517@york.ac.uk (P.R. Pryor), jpl1

0167-4889/$ – see front matter © 2008 Elsevier B.V. Adoi:10.1016/j.bbamcr.2008.12.022

a b s t r a c t

a r t i c l e i n f o

Article history:

The delivery of proteins fro Received 22 July 2008Received in revised form 1 December 2008Accepted 12 December 2008Available online 8 January 2009

Keywords:EndocytosisLate endosomesLysosomesMultivesicular bodyMembrane fusion

m the plasma membrane to the lysosome for degradation is essential for normalcellular function. There is now a good understanding of the protein complexes involved in sorting proteins atthe plasma membrane and into the intralumenal vesicles of the multi-vesicular body. A combination of cellfree content mixing assays and live-cell imaging has dissected out the final step in delivery ofmacromolecules to the lysosome from the multi-vesicular body and provided insights into the molecularmechanisms by which late endosomes and lysosomes exchange lumenal contents. The endocytic pathwayhas provided a platform with which to understand the autophagic and phagocytic pathways, but the finedetails of how traffic through these pathways is regulated remain to be determined.

© 2008 Elsevier B.V. All rights reserved.

1. Introduction

Lysosomes are highly dynamic membrane-bound organelles thatact as the terminal degradative compartment of the endocytic,phagocytic and autophagocytic pathways (Fig. 1). Many of theproteins required for protein sorting and endocytic delivery tolysosomes in mammalian cells are orthologues of those used by theyeast Saccharomyces cerevisiae for delivery to the vacuole (the yeastequivalent of the lysosome). Studies on vacuolar delivery and fusionhave been extremely informative in providing insights to the universalmechanisms governing sorting and delivery in eukaryotic cells.Additionally, some cell types contain lysosome-related organelles(LROs) that include melanosomes, lytic granules, MHC-II compart-ments, platelet dense granules and neutrophil azurophil granules [1].Some LROs are simply themodified lysosome, for instance in cytotoxicT lymphocytes the lytic granules are the only lysosome-type organellepresent. In other cell types such as melanocytes, both the melano-somes and the ‘conventional’ lysosomes co-exist. Studies of LROs, inparticular from patients with Hermansky–Pudlak syndrome that havedeficiencies in melanosomes and platelet-dense granules, haveidentified mutations in genes encoding proteins (some that are notfound in yeast) that assemble into five distinct complexes (adaptorprotein-3 (AP-3), homotypic fusion and vacuole protein sorting(HOPS) complex and biogenesis of lysosome-related protein complex(BLOC)-1, -2 and -3) that have provided further insights into deliveryof endocytic cargo to the lysosome [2]. This review thereforesummarises recent insights into endocytic sorting of plasma mem-

ed at tel.: +44 1904 328563.

0@cam.ac.uk (J.P. Luzio).

ll rights reserved.

brane proteins destined to be delivered to mammalian lysosomes. Itincludes descriptions of sorting at the plasma membrane and inendosomes and the plethora of proteins required to regulate and sortcargo into the intralumenal vesicles (ILVs) of multivesicular bodies(MVBs). Finally, as all membrane fusion events may be defined ashaving tethering and docking and fusion steps, the review willindicate the key regulators in the membrane fusion between MVBsand lysosomes.

2. Clathrin mediated endocytosis at the plasma membrane

A plasma membrane protein that is destined to be endocytosedneeds to be recognised by the endocytic machinery. Although otheruptake pathways have been described [3], the best understoodendocyticmechanism is clathrinmediated endocytosis. At its simplest,thismay be regarded as a co-ordinated series of events involving cargorecruitment into small regions of the plasma membrane, clathrin cageassembly, membrane bending, vesicle formation, vesicle fission andfinally vesicle uncoating (for reviews see [4,5]). Cargo recruitmentrequires clathrin adaptors that interact with sequence motifs,structural features or added ubiquitin molecules in the cytosolic tailof the membrane protein. Some membrane proteins require thebinding of extracellular ligand before rapid internalisation proceeds,but others do not need the addition of ligand. In the latter case,endocytosis occurs to maintain low levels of the protein at the plasmamembrane and/or to redirectmis-sorted proteins to intracellular sites.Efficient incorporation of some endocytosed membrane proteins intoclathrin- coated-pits and -vesicles (CCVs) is mediated by short linearamino acid sequences, such as the YXXΦ and [DE]xxxLLmotifs, that arerecognised by the clathrin adaptor AP2 (reviewed in [6–8]). Othermembrane proteins, including several growth factor receptors, are

Fig. 1. Intracellular pathways to the lysosome. Endocytic cargo is internalised from the plasma membrane and delivered to early endosomes. Maturation of the early endosome givesrise to a late endosome/multi-vesicular body where the cargo destined for degradation has been sorted into intralumenal vesicles. The limiting membranes of late endosomes andlysosomes can fuse to form a hybrid organelle where degradation of endocytosed macromolecules commences. Lysosomes are reformed from the hybrid organelle by membraneretrieval and condensation reactions. Autophagic and phagocytic pathways both feed into the endocytic pathway with the lumenal contents of the autophagosome and phagosomeeventually being delivered to the lysosome for degradation.

616 P.R. Pryor, J.P. Luzio / Biochimica et Biophysica Acta 1793 (2009) 615–624

ubiquitylated, such that they carrymultiple single ubiquitins (multiplemonoubiquitylation), or short polyubiquitin chains, added to specificlysines in the cytosolic tail (reviewed in [9]). The presence of manyubiquitins allows efficient capture into CCVs by specific clathrinadaptors such as EPS15 and epsin, despite the relatively low affinity ofubiquitin binding domains in these adaptors for single ubiquitinmolecules.

Recently, a third mechanism for efficient incorporation into CCVshas been described, and is used by at least one soluble N-ethylmaleimide-sensitive factor (NSF) attachment protein receptor(SNARE) at the plasma membrane. SNAREs are critical for drivingmembrane fusion events on the secretory and endocytic pathways(reviewed in [10]). However, the mechanisms by which SNAREs areactually sorted to different membranes are largely unknown. Most ofthe SNAREs do not possess short linear endocytic sequence motifs andare not known to be ubiquitylated. However, the folded N-terminaldomain of the SNARE VAMP7 binds to a predicted unstructured region

of the clathrin adaptor HIV Rev-binding protein (HRB) allowingendocytosis of VAMP7, but only when VAMP7 is incorporated into acis-SNARE complex [11]. VAMP7 is found on lysosomes and is requiredfor lysosome fusionwith the plasma membrane. Its folded N-terminaldomain thus acts as a retrieval signal from the cell surface and theinteraction with an unstructured region of HRB has provided aninsight into the potential role of the unstructured regions of clathrinadaptors in cargo recognition.

It should be noted that clathrin-mediated endocytosis may play animportant role in lysosome biogenesis. In most cell types studied,many newly synthesised acid hydrolases are delivered to lysosomesvia an intracellular route that requires addition of mannose 6-phosphate and binding to mannose 6-phosphate receptors, whichtraffic between the trans-Golgi network and endosomes. However, inproximal kidney tubule cells the lysosomal enzyme cathepsin B isdelivered to lysosomes following secretion of non-mannose 6-phosphate-labelled pro-cathepsin B, that then binds to the cell surface

617P.R. Pryor, J.P. Luzio / Biochimica et Biophysica Acta 1793 (2009) 615–624

receptor megalin, which is then internalised by clathrin mediatedendocytosis [12]. Some newly synthesised lysosomal membraneproteins, such as lysosomal acid phosphatase are also mainlydelivered to lysosomes by an indirect route via the plasmamembrane.However, the major lysosomal membrane proteins LAMP-1 andLAMP-2, that together account for over 50% of the membrane proteinin a lysosome, aremainly delivered to endosomes/lysosomes from thetrans-Golgi network by a direct intracellular route.Whilst LAMP-1 andLAMP-2, like lysosomal acid phosphatase, contain cytosolic tail YXXΦmotifs and are efficiently endocytosed by clathrin mediated endocy-tosis, this is thought to be mainly a retrieval route for these twoproteins (reviewed in [13]).

3. Sorting at the recycling endosome

Endocytosed cargo is first delivered to early endosomes, then tolate endosomes, which have the morphology of MVBs and finally tolysosomes. In the early endosome some bound ligands dissociate fromtheir receptors in the acidic lumen allowing recycling of the receptorsto the cell surface. Recycling may also occur via the tubular recyclingendosome derived from the early endosome (reviewed in [14]). It isalso likely that there are retrieval routes from the recycling endosometo the late endosome/lysosome. One membrane protein that appearsto use such a route is the SNARE VAMP7. In addition to binding HRB, itsN-terminal folded longin domain also binds to the clathrin adaptorAP3 and this interaction is required for delivery to late endosomes/lysosomes [15]. Given the localisation of AP3 to recycling endosomes[16], an attractive hypothesis is that VAMP7 endocytosed from the cellsurface is delivered to early endosomes, sorted into recyclingendosomes and then retrieved by AP3 to late endosomes/lysosomes.Although the precise mechanism of binding of VAMP7 to AP3 has notbeen resolved. There is evidence that in the direct route ofintracellular delivery of newly synthesised LAMP-1 and LAMP-2 tolysosomes one sorting step is an AP3 dependent sorting at tubularrecycling endosomes [16].

4. Multi-vesicular body formation

The multivesicular body (MVB) or late endosome is characterisedby intralumenal vesicles (ILVs). These ILVs are formed through inwardbudding of the endosomal membrane, a process that commences atthe early endosome. The sorting of cargo from the endosomal limitingmembrane to these ILVs eventually allows endocytosed cargo destinedfor degradation to be efficiently degraded by lysosomal enzymes. Thebest characterised signal for sorting of proteins into the internalvesicles is ubiquitylation. Hence, ubiquitylation not only represents anendocytosis signal but it also acts as a signal for degradation.Originally identified in yeast, the soluble class E vacuolar proteinsorting (Vps) proteins organise into four endosomal sorting com-plexes required for transport (ESCRT) complexes, namely ESCRT-0,ESCRT-I, ESCRT-II and ESCRT-III (reviewed in [17]). Current modelssuggest that the ESCRT complexes are required for both formation ofthe ILVs and for sorting ubiquitylated cargo into them. Certainlydepletion of components of ESCRT-0 and ESCRT-I in mammalian cellsreduces ILV formation [18–20]. The role of ESCRT-III proteins in ILVformation is not so clearly understood. Recent structural studies haveshown that components of ESCRT-III can form lattices on membranes[21,22] and, in the presence of a dominant negative VPS4, over-expression of VPS32 (CHMP4) in mammalian cells can promotenegative membrane curvature [23]. In addition, it has been shown invitro that filaments formed by ESCRT-III proteins VPS2 (CHMP2) andVPS24 (CHMP3) or a chimeric VPS2–VPS24 protein can be disas-sembled by VPS4 [24,25] suggesting that helical VPS2 and VPS24structures could assemble at the base of the neck of an inwardlybudding vesicle leading to membrane constriction and eventuallyabscission when coupled to VPS4. These data suggest that ESCRT-III

components are directly involved in the generation of ILVs. However,depletion of the ESCRT-III component VPS24 inhibits EGFR degrada-tion but not the kinetics of signal termination [26]. Similarly a CHMP5knock-out mouse forms MVBs, but degradation of receptors and fluidphase markers is reduced [27,28]. These latter observations suggestthat some of the components of ESCRT-III may be dispensable for ILVformation but may have a role in downstream events.

Rab interacting lysosomal protein (RILP) is a RAB7 binding proteinknown to act in events late on in the delivery of cargo to the lysosome.However, it has also been observed that RILP interacts with themammalian ESCRT-II proteins VPS22 (EAP30) and VPS36 (EAP45)[29,30]. Depletion of RILP by RNA interference reduces ILV formationand degradation of EGF receptors in a manner similar to VPS22depletion [31]. Depletion of RILP causes a reduction of cellular VPS22concentration and vice versa, so some of the effects of RILP depletionmight be related to the accompanying loss of VPS22. However, RILPdepleted endosomes are morphologically different to VPS22-depletedendosomes [31]. Given that RILP also interacts with the dynein–dynactin motor complex [32,33], it has been suggested that RILPmight co-ordinate the formation of MVBs with dynein-mediatedmotility along microtubules.

5. Delivery to the lysosome

Several hypotheses have been proposed for the mechanism oftransfer of endocytosed material from endosomes to lysosomes(reviewed in [34]). These include a maturation model where theendosome ‘matures’ into a lysosome [35], a vesicular transport modelwhere vesicles carry cargo from the endosome to the lysosome [36], akiss-and-run model where endosomes and lysosomes engage inrepeated transient fusions [37] and a model where endosomes andlysosomes completely fuse giving rise to a hybrid compartment fromwhich lysosomes reform [13,38–41]. The maturation model cannotexplain how lysosomes remain accessible to endocytic enzymereplacement therapies [42] nor how they undergo content mixingwith late endosomes [38,39]. Whilst there is no formal evidence for avesicular transport model, cell-free content mixing assays [38,39] andmore recently time-lapse confocal microscopy experiments [43] haveprovided evidence for both kiss-and-run and direct fusion events asthe major mechanism for delivery of endocytosed macromolecules tothe lysosome.

Fusion of membranes in the secretory and endocytic pathwaysrequires tethering, SNARE complex formation and then phospholipidbilayer fusion (recently reviewed in [44]). There is a high degree ofcomplexity in the molecular events preceding phospholipid bilayerfusion and the clear distinction between initial tethering events andmembrane fusion is becoming less apparent as the molecularmechanisms of these processes are unravelled. Because there is ahigh degree of conservation in the mechanism of membrane fusionbetween yeast and higher eukaryotes, studies of vacuolar fusion inSaccharomyces cerevisiae have aided our understanding of themolecular events required for delivery of endocytic cargo to thelysosome. The following sections summarise our current under-standing of tethering, SNARE complex formation and fusion eventsbetween the late endosome/MVB and the mammalian lysosome andprovide a model of delivery to the mammalian lysosome (Fig. 2).

5.1. Tethering complexes

Homotypic vacuolar tethering in yeast is mediated by the HOPScomplex (reviewed in [45]). This complex consists of six subunits, fourclass C proteins Vps11p,16p,18p and 33p and two class B proteinsVps41p/Vam2p and Vps39p/Vam6p (vps mutants are divided into sixclasses,A–F, basedupon several criteria, including themorphologyof themutant vacuoles [46,47]). The individual HOPS complex subunits havebeen studied in detail but their precise functions still require further

Fig. 2. The molecular machinery required for delivery of endocytosed material to lysosomes. The delivery of endocytosed material to the lysosome involves maturation of the earlyendosome to form a MVB/late endosome. Maturation events involve several complexes whose activity must be co-ordinated to give rise to an organelle with a membrane that isfusion competent with the lysosome. Complexes involved in priming the endosomal membrane ready for delivery to the lysosome include tethering factors such as the Class-C VPScomplex, which in yeast can interact with different proteins to form a CORVET or HOPS complex. These complexes regulate early Vps21p-dependent (mammalian RAB5-dependent)or later Ypt7p-dependent (mammalian RAB7-dependent) events, respectively. Sorting of membrane proteins into lumenal vesicles, a pre-requisite for efficient degradation by thelysosome, requires ESCRT machinery. The ESCRT machinery is recruited to membranes by cargo and specific phosphoinositides, but probably also by other proteins such as thelysosomal protein RILP that interacts with components of ESCRT-II. The ESCRTmachinery is also likely to act as a platform for recruitment of other proteins to the limiting membrane.All the complexes are likely to be integrated with each other. Interactions include, Class C-VPS subunits and ESCRT subunits (unpublished observations); RILP with ESCRT-II; RAB7with RILP and the HOPS complex. Endosomal maturation is likely to occur at the same time as traffic along the microtubules towards the microtubule organising centre. Finally,membrane fusion between the MVB and the lysosome occurs as a SNARE-dependent event. Yeast vacuolar fusion studies indicate that this too might be subjected to a proof-readingby the HOPS complex. Both membrane fusion to form a hybrid organelle and reformation of the lysosome are calcium-dependent processes where the calcium is derived from thelumen of the organelles.

618 P.R. Pryor, J.P. Luzio / Biochimica et Biophysica Acta 1793 (2009) 615–624

elucidation. Vam6p has been observed to have guanyl-nucleotideexchange factor (GEF) activity for Ypt7p (the yeast equivalent ofRAB7) [47] and binding of the HOPS to Ypt7p occurs when it is in itsGTP-bound form [48]. Furthermore, binding of the HOPS complex to thevacuole is diminished in the absence of Ypt7p [49]. HOPS binding to thevacuolar membrane is also likely to mediated by an affinity of the HOPScomplex for both phospholipids [50] and vacuolar SNAREs [50–53].

At the core of everymembrane fusion reaction is the formation of aspecific tetrameric trans-SNARE complex. Each type-II trans-mem-brane SNARE protein contains a 16 turn ‘SNARE’ helix and can bedefined as either a Q- or R-SNARE depending on the residue at layer 0in the SNARE helix [54]. The Q-SNAREs can be further sub-divided intoQa, Qb and Qc SNAREs with the tetrameric trans-SNARE complexconsisting of one of each class of Q-SNARE and a single R-SNARE [55].Like other membrane fusion reactions vacuolar fusion requires threeQ-SNAREs (Vti1p, Vam3p, Vam7p) and one R-SNARE (Nyv1p), but invitro vacuolar fusion can still occur when the Q or R residues aremutated to give a 4Q or 2Q:2R complex [56]. It has been shown thatvacuolar fusion is inhibited upon the addition of exogenous HOPS toSNARE complexes with altered 0-layers, or containing truncatedVam7p, This has suggested a ‘proofreading’ role for the HOPS complex,but precisely how this is achieved is unclear [57].

In mammalian cells, there are orthologues of the HOPS subunitsand there is considerable evidence that they also serve a role indelivery of endocytosed macromolecules to the lysosome. Thus,overexpression of VPS18, VAM6 and VPS33B cause lysosomalclustering [58–60], whereas depletion of VPS18 by RNA interference

causes organelle dispersion [58]. However, it is not only in deliveryto the vacuole/lysosome where components of the HOPS complexappear to be important, and in both yeast and mammalian cellscomponents of the HOPS complex also play a role early in theendocytic pathway [61–63]. Recent work in yeast has helped toresolve how components of the HOPS complex can act at both earlyand late endosomes. Ungermann and colleagues have shown that thefour Class C proteins Vps11p,Vps16p,Vps18p and Vps33p candifferentially interact with two other subunits [64]. At the endosomethe Class C Vps proteins interact with Vps8p and Vps3p to form aclass C core vacuole/endosome tethering complex (CORVET),whereas at the vacuole they interact with Vam6p and Vps41p toform the HOPS complex. Intermediate protein complexes betweenCORVET and HOPS have been isolated but the mechanisms thatregulate CORVET to HOPS conversion, and vice versa remain to bedetermined. Interestingly, Vps3p binds to Vps21p (yeast RAB5) inboth its GTP and GDP-bound forms and may therefore be a Vps21pGEF [64]. As described above, Vam6p is a GEF for Ypt7p. Thiscorresponds with mammalian cells in which RAB5 is required forearly endocytic traffic [65] and RAB7 in late endocytic traffic [66–68].Taken together with data where RAB5 and RAB7 associate with thesame endosome [69] it suggests that the transition from an earlyendosome to a mature MVB ready for fusion with the lysosome isregulated by tethering proteins and RABs (Fig. 2). Another linkbetween RAB5 and the mammalian orthologues of yeast class C VPSproteins has also been described. The autophagy protein UVRAG(also known as beclin-1 binding autophagy protein), which is also

619P.R. Pryor, J.P. Luzio / Biochimica et Biophysica Acta 1793 (2009) 615–624

important for endocytic traffic [70], binds to both class C VPSproteins and GTP-bound Rab5 [71]. A further example of the multi-functional role of the mammalian class C VPS proteins is theobservation that class C VPS proteins can interact with ESCRTsubunits (unpublished observation from the Luzio laboratory),consistent with tethering proteins also regulating MVB formation.This is not surprising, since MVB formation needs to be co-ordinatedwith organelle maturation. When other accessory proteins are addedinto the system, including RILP that binds both ESCRT subunits[29,30] and RAB7 [72], it becomes difficult to assign a linear order ofmolecular events to endosome maturation.

In mammalian cells there is more than one homologue of bothVps16p and Vps33p, namely VPS16A, VPS16B and VPS33A andVPS33B. The function of these is also related to the function of Dro-sophila orthologues. Drosophila eye pigment granules and mammalianmelanosomes are both lysosome related organelles (LRO) and defectsin trafficking to these organelles results in changes in eye colour in thefly and coat colour changes in mice [1]. Mutations in DmVps33 resultsin a mutant fly called carnation [73], whereas in the mouse mutationsin VPS33A result in a buff mouse [74]. Genomic analyses havesuggested that the carnation gene is more similar to VPS33A thanVPS33B [75]. Overexpression of human VPS33B in cultured cellscauses lysosomal clustering, but lysosomes are not clustered byoverexpression of VPS33A or DmVps33 [75]. These data suggest thatVPS33A may be important for LRO biogenesis and VPS33B for deliveryto the lysosome. Additionally, phagocytosis and endocytosis sharecommon intracellular machinery for delivery of material to thelysosome and a recent report has shown that the phagocytosedpathogenMycobacterium tuberculosis, which delays progression to thelysosome and thereby improves its ability to survive, secretes thetyrosine phosphatase PtpA into its host cell cytosol which depho-sphorylates VPS33B [76]. VPS33B is a self-phosphorylating tyrosine-kinase and the phosphorylation state of VPS33B may help to explainolder observations showing that yeast Vps33p binds ATP and itsassociation with membranes is regulated by cellular ATP levels [77].One caveat to VPS33B being more important for delivery to thelysosome compared to VPS33A is that a proteomic study of lysosomemembrane proteins identified the presence of VPS33A but not VPS33B[78]. VPS33B is not the only HOPS subunit to undergo phosphoryla-tion, because yeast Vps41p is also phosphorylated by the casein kinaseYck3p [79], Yck3p deletion leads to an increase of Vps41p at thevacuole and an increase in vacuolar fusion, whereas Yck3p over-expression reduces fusion [79].

In yeast, a complex of two proteins, Mon1p and Ccz1p, appears toalso bind the Class C Vps subunits of the HOPS complex, with vacuolefusion reduced if either protein is missing [80]. In mammalian cellsthere has been a gene duplication event resulting in two orthologuesof Mon1p,MON1A andMON1B. A knock-outmouse for MON1A showstrafficking defects through the secretory pathway. However, nochanges in endocytic trafficking to the lysosome, lysosome size orlysosome distribution were observed [81]. A proteomic study hasidentified MON1B associated with lysosomes [78], suggesting that inmammalian cells MON1Bmay be important for traffic to the lysosome.The precise roles of MON1B and its lysosomal target in mammaliancells are yet to be established. There is no obvious orthologue of Ccz1p,however Ccz1p does have an N-terminal CHiPS domain that hashomology to the Hermansky–Pudlak protein HPS4, one of thesubunits of BLOC-3 [82].

Recently a large phenotypic screen was undertaken in yeast toidentify proteins on the endosomal pathway based upon thehypothesis that genes with similar functions on a particular pathwaygive a similar phenotype when mutated [83]. To assess whetherendocytosis had been perturbed various criteria were used includingtrafficking of the vacuolar hydrolase carboxypeptidase Y (CPY) whichis secreted when traffic to the vacuole is impaired, processing of theyeast hormone alpha-factor which if secreted in its pro-alpha-factor

form indicates a defect in recycling of the pro-alpha factor-processingenzyme Kex2p between the Golgi and the endosome and sensitivity tochemicals, such as caffeine and SDS. In this screen, known complexesregulating endocytosis were found including the ESCRT complexes,but a new complex consisting of two proteins Vps55p and Vps68pwasalso identified [83]. Both Vps55p and Vps68p are predicted to containfour transmembrane domains, and in yeast both proteins localise tothe vacuole limiting membrane, their stability depending upon eachother, suggesting that they form a stable protein complex. Mutation ofeither protein slows the traffic of both endocytic and biosyntheticcargo through the MVB, but does not alter MVB formation itself [83].These data suggest that Vps55p and Vps68p regulate endosomematuration at a step downstream of MVB formation. Both Vps55p andVps68p have two human homologues, OB-RGRP and LEPROTL1 forVps55p and SMP1 and C21ORF4 for Vps68p [83]. OB-RGRP expressionin yeast can complement the carboxypeptidase Y trafficking defect in avps55 mutant [84], but a role of OB-RGRP in delivery of macro-molecules, or Vps68p orthologues, to the mammalian lysosome is yetto be established.

5.2. trans-SNARE complexes

In mammalian cells a trans-SNARE complex consisting of Q-SNAREs SYNTAXIN7, VTI1B and SYNTAXIN8 together with the R-SNARE VAMP8 has been shown to be responsible for homotypic fusionof late endosomes [85]. But by exchanging the R-SNARE VAMP8 forVAMP7 (also known as tetanus neurotoxin insensitive VAMP (TI-VAMP) or synaptobrevin-like-1 (SYBL1)), heterotypic membranefusion between late endosomes and lysosomes is achieved (Fig. 3)[41]. To date, all lysosomal fusion events in which the trans-SNAREcomplex has been defined, appear to require the R-SNARE VAMP7(reviewed in [34]), thus the high concentration of VAMP7 on thelysosomal membrane may help to define the lysosome. VAMP7, yeastNyv1p [86] along with the yeast and mammalian orthologues ofSec22p [87] and Ykt6p [88] comprise the ‘longin’ sub-family of R-SNAREs since they possess an N-terminal, profilin-like, longin domain(LD) [89]. As described above the LD of VAMP7 is important for itsintracellular trafficking, because it binds the endocytic adaptor HRB atthe plasmamembrane [11], thus enabling SNARE endocytosis, and AP3at the recyling/sorting endosome [15] to sort the SNARE to lysosomes.The LD of VAMP7 can bind back to its own SNARE helix but not whenthe SNARE helix is complexed to other SNAREs [11]. VAMP7 is the onlyprotein known to be differentially required for heterotypic lateendosome–lysosome fusion when compared to homotypic lateendosome fusion. Thus, the LD of VAMP7, that will be free to bindother proteins when a SNARE complex has formed, may also serve as adomain to recruit further proteins to regulate the final stages oflysosomal fusion.

The interaction of tethers and SNAREs may play a role inregulating fusion because there is good evidence to suggest thattethering complexes are capable of directly binding the SNAREsinvolved in lysosomal fusion. In both yeast and mammalian cellsClass C VPS complex subunits have been found to bind to thevacuolar SNARE Vam3p and its mammalian orthologue SYNTAXIN7[53,63,90]. It is possible that the binding of the Class C VPS complexto SYNTAXIN7 may keep SYNTAXIN7 in a closed conformation or in aconformation that prevents trans-SNARE pairing. The interactionbetween SYNTAXIN7 and the Class C VPS complex interaction maybe regulated by phosphorylation since a recent report in yeast hasshown that both Vps41p and Vam3p may be phosphorylated [91].The phosphorylation of Vps41p and Vam3p is reduced in thepresence of GTP-bound Ypt7p, the RAB7 orthologue [91]. Ypt7p hasalso been shown to bind Vps41p in addition to Vps39p (the Ypt7pGEF) [91]. One hypothetical scenario for how tethering and trans-SNARE pairing may be co-ordinated is shown by Fig. 4. Here,exchange of RAB5 for RAB7 would allow for recruitment of VPS39

Fig. 3. Schematic model of SNARE dependent fusion between late endosomes and lysosomes. Tethering reactions between organelles are likely to precede trans-SNARE pairing.Tethers between late endosomes and lysosomes are likely to be the HOPS complex and may regulate the fusion process itself. Fusion between a late endosome and a lysosomerequires the R-SNARE VAMP7 and three Q-SNAREs (SYNTAXIN7, SYNTAXIN8 and VTI1B) on the opposingmembrane. VAMP7 has an N-terminal longin domain (green oval) that bindsback to its SNARE domain. Upon trans-SNARE pairing the longin domain is released, but regulatory proteins may bind the longin domain to regulate the fusion process. Followingtrans-SNARE pairing membrane fusion occurs allowing content mixing. Membrane retrieval and condensation reactions allow reformation of the lysosome. Homotypic lateendosome fusion uses the same three Q-SNAREs as late endosome–lysosome fusion but a different R-SNARE, namely VAMP8.

620 P.R. Pryor, J.P. Luzio / Biochimica et Biophysica Acta 1793 (2009) 615–624

and VPS41. VPS39 and VPS41 could bind to the Class C VPS complex,bound to SYNTAXIN7, on an opposing membrane. The GEF activity ofVPS39 would activate RAB7 leading to de-phosphorylation eventsand release of the Class C VPS complex from SYNTAXIN7 and therebyallow for trans-SNARE pairing and membrane fusion.

5.3. Organelle fusion

In mammalian cells there are events after trans-SNARE pairing thatare necessary for content mixing to occur, as there are for yeasthomotypic vacuole fusion. In a similar manner to homotypic vacuolefusion [92], late endosome–lysosome content mixing can be inhibitedby the calcium chelator BAPTA [40]. The terminal step of lateendosome–lysosome fusion is also inhibited by calmodulin antago-nists, which implies that calcium/calmodulin is required for comple-tion of the fusion process. However, BAPTA inhibition of yeast vacuolefusion can be relieved byMgCl2 [93] and this has led to the suggestionthat in yeast the BAPTA effects may not entirely be due to calciumchelation but that divalent cations may have a more direct‘biophysical’ role in the membrane fusion process [94]. Using animmobilised phosphatase inhibitor (microcystin LR) as a bait, proteinphosphatase-1 (PP-1) in yeast, Glc7p, was identified as beingimportant for vacuole fusion [95], and the use of microcystin LRimplied that the Glc7p reaction was post SNARE pairing. However, ithas subsequently been shown thatmicrocystin-LR inhibits the vacuolefusion assay but not vacuole fusion itself [96,97], so the substrate ofGlc7p and the point in the vacuole fusion process where it is involved

remain to be determined. In mammalian late endosome–lysosomefusion microcystin-LR does not inhibit content mixing (unpublishedobservations from authors' research).

6. Lysosome reformation

Delivery of endocytic material to the lysosome results in a hybridorganelle, that hasmarkers of both late endosomes and lysosomes andis of intermediate density [38]. We have proposed that it is in thehybrid organelle where degradation of endocytosed macromoleculescommences and not the lysosome per se, which may be regarded asstorage organelles of mature lysosomal enzymes. Once digestion oflumenal material has occurred there needs to be a mechanismwhereby limiting membrane proteins, such as the SNAREs, arerecycled for further rounds of fusion with the lysosome and amechanism for lysosome re-formation/content condensation. Hybridorganelles both from rat liver and from late endosome–lysosomecontent mixing assays can be isolated on discontinuous Nycodenz/Ficoll gradients and lysosome reformation/content condensationexamined [38]. It has been shown that content condensation isindependent of cytosol, but is dependent upon ATP. Furthermore, theuse of bafilomycin A1 and EGTA-AM showed that lysosome reforma-tion is dependent upon lumenal pH and Ca2+ [40]. Evidence formembrane retrieval from hybrid organelles has come from live-cellstudies where vesicles have been observed budding from hybridcompartments [43,98]. Machinery that may be involved in membraneretrieval could be the retromer complex, which mediates retrograde

Fig. 4. Hypothetical model for regulation of tethering and SNARE pairing. Exchange of RAB5 for RAB7 and binding of RAB7 to RILP on membranes leads to the recruitment of VPS39and VPS41. VPS39 and VPS41 can bind to Class C VPS subunits on an opposing membrane that are themselves are bound to the SNARE SYNTAXIN7. Activation of RAB7 by the GEFactivity of VPS39 leads to changes in protein phosphorylation of both VPS41 and SYNTAXIN7 (based upon data in yeast and changes in phosphorylation of Vps41p and Vam3p) whichin turn allows SYNTAXIN7 to form a trans-SNARE complex with VTI1B, SYNTAXIN8 and VAMP7.

621P.R. Pryor, J.P. Luzio / Biochimica et Biophysica Acta 1793 (2009) 615–624

transport of transmembrane cargo from endosomes to the trans-Golginetwork (recently reviewed in [99,100]). Whilst there is little furtherinformation on how lysosomes are regenerated from hybrid orga-nelles, there is a clear need to understand further the lysosomereformation process, since a failure to re-form lysosomesmay underliesome diseases. For instance, mutations in the lysosomal ion channelmucolipin-1 results in the debilitating disease mucolipidosis type-IV(MLIV) [101–104], a rare lysosomal storage disease, where unlikeother lysosomal storage diseases there does not appear to be a defectin lysosomal enzymes but a defect in trafficking of material out of thehybrid organelle [105–108].

7. Intracellular ions

As has been mentioned above, fusion between the late endosomeand the lysosome and lysosome reformation are dependent uponlumenal calcium concentrations [40]. Precisely what channelsregulate lumenal calcium are unknown but possible lysosomalcalcium channels include mucolipin-1 [106], an NAADP regulatedcalcium channel [109] and the ATP-regulated P2X4 cation channel[110]. Whilst mucolipin-1 does not appear to be the lysosomalreceptor for NAADP a recent study has shown that antibodies tomucolipin-1 can inhibit NAADP-induced calcium release from lyso-somes suggesting that the NAADP-sensitive calcium release channel isassociated with mucolipin-1 [111].

The increased lumenal acidity in the endocytic pathway is mainlyaccomplished by V-type H+-ATPases. As these pumps are electro-genic they would create a lumenal-positive voltage in the absence of aneutralising current. A family of 9 chloride channels has beenidentified [112], four of which (ClC-3, -4, -5, -6 and -7) reside on theendocytic pathway (reviewed in [113]). Disruption in mice of ClC-3, -6

and -7 gives phenotypes similar to a subset of human lysosomalstorage diseases grouped as neuronal ceroid lipofuscinosis (NCL)[114–116]. It has recently been shown that ClC-7 is the major chloridetransporter in lysosomes and the transporter acts as a Cl−/H+

antiporter [117]. Depletion of ClC-7 by RNA interference stronglydiminishes the ability of lysosomes to acidify [117]. Together thesedata indicate that chloride transport in the endocytic pathway iscritical for normal endocytic functions and membrane dynamics.

In yeast, of the 18 proteins encoded by the class E VPS genes, whichincludes the ESCRT proteins, only one, Vps44p/Nhx1p, is a transmem-brane protein. Nhx1p shows high homology to mammalin Na+/H+exchangers of the NHE family and is thought to transport sodium ionsinto the yeast prevacuolar compartment in exchange for protons [118].It is unclear which isoform of NHE is important for transport throughthe endocytic pathway and whilst no NHE isoform has been found inlysosomes, NHE6 has been localised to recycling endosomes and NHE9to late recycling endosomes [119].

8. Conclusions

There is now a wealth of information on the molecular mechan-isms of delivery of endocytosed molecules to the lysosome. However,whilst we have a greater understanding of these mechanisms there isstill some way to go to understand precisely how these proteincomplexes integrate with one another. Additionally, it is clear that theproteins regulating protein sorting and membrane fusion arethemselves subject to regulation by phosphorylation/de-phosphor-ylation (as described above for VPS33B and Vps41p). Indeed, a studysilencing each kinase of the human kinome using siRNAwas shown tohave profound effects on the endocytic pathway [119]. A review of therole of kinases on regulating the endocytic machinery has recently

622 P.R. Pryor, J.P. Luzio / Biochimica et Biophysica Acta 1793 (2009) 615–624

been published [120] and clearly we still have much to learn on howendocytosis is regulated by phosphorylation and de-phosphorylationevents. In this review, we have not discussed the role of lipids, somewhich may also be phosphorylated [121], in the endocytic pathwayand how they contribute to protein recruitment to membranes,membrane curvature and membrane fusion.

As shown in Fig. 1, the final destination of the autophagic andphagocytic pathways is the lysosome. It is clear that these two otherpathways to the lysosome use similar machineries to the endocyticpathway, for instance both autophagy and phagocytosis utilise theESCRTmachinery [122,123]. A challenge now is not only to understandthe endocytic pathway in greater detail, but to determine where theautophagic and phagocytic pathways integrate into the endocyticpathway. This will improve our understanding of autophagic path-ways which may help in treatment of diseases such as Huntington'sand Parkinson's disease [124]. Understanding phagocytosis will aid inunderstanding how intracellular pathogens, such as Mycobacteriumtuberculosis, survive. But the reverse is also true, as research intophagocytosis and autophagy expands, we will learn more about theendocytic pathway as has recently been seen in the example of thesecretion of PtpA by Mycobacterium tuberculosis and the depho-sphorylation of VPS33B [76].

Acknowledgements

We would like to acknowledge the support of the Wellcome Trust(P.R.P is in receipt of a Wellcome Trust Career Development Fellow-ship (079705) and the CIMR is in receipt of a Wellcome Trust StrategicAward 07895) and the MRC (J.P.L. is in receipt of a MRC programmegrant G9310915).

References

[1] E.C. Dell'Angelica, C. Mullins, S. Caplan, J.S. Bonifacino, Lysosome-relatedorganelles, FASEB J. 14 (2000) 1265–1278.

[2] E.C. Dell'Angelica, The building BLOC(k)s of lysosomes and related organelles,Curr. Opin. Cell Biol. 16 (2004) 458–464.

[3] S. Mayor, R.E. Pagano, Pathways of clathrin-independent endocytosis, Nat. Rev.Mol. Cell Biol. 8 (2007) 603–612.

[4] T. Kirchhausen, Clathrin, Annu. Rev. Biochem. 69 (2000) 699–727.[5] H.T. McMahon, J.L. Gallop, Membrane curvature andmechanisms of dynamic cell

membrane remodelling, Nature 438 (2005) 590–596.[6] J.S. Bonifacino, L.M. Traub, Signals for sorting of transmembrane proteins to

endosomes and lysosomes, Annu. Rev. Biochem. 72 (2003) 395–447.[7] D.J. Owen, B.M. Collins, P.R. Evans, Adaptors for clathrin coats: structure and

function, Annu. Rev. Cell Dev. Biol. 20 (2004) 153–191.[8] T.J. Brett, L.M. Traub, Molecular structures of coat and coat-associated proteins:

function follows form, Curr. Opin. Cell Biol. 18 (2006) 395–406.[9] R.C. Piper, J.P. Luzio, Ubiquitin-dependent sorting of integral membrane proteins

for degradation in lysosomes, Curr. Opin. Cell Biol. 19 (2007) 459–465.[10] Y.A. Chen, R.H. Scheller, SNARE-mediated membrane fusion, Nat. Rev., Mol. Cell

Biol. 2 (2001) 98–106.[11] P.R. Pryor, L. Jackson, S.R. Gray, M.A. Edeling, A. Thompson, C.M. Sanderson, P.R.

Evans, D.J. Owen, J.P. Luzio, Molecular basis for the sorting of the SNARE VAMP7into endocytic clathrin-coated vesicles by the ArfGAP Hrb, Cell 134 (2008)817–827.

[12] R. Nielsen, P.J. Courtoy, C. Jacobsen, G. Dom, W.R. Lima, M. Jadot, T.E. Willnow,O. Devuyst, E.I. Christensen, Endocytosis provides a major alternative pathwayfor lysosomal biogenesis in kidney proximal tubular cells, Proc. Natl. Acad. Sci.U. S. A. 104 (2007) 5407–5412.

[13] J.P. Luzio, V. Poupon, M.R. Lindsay, B.M. Mullock, R.C. Piper, P.R. Pryor, Membranedynamics and the biogenesis of lysosomes, Mol. Membr. Biol. 20 (2003) 141–154.

[14] J.S. Bonifacino, R. Rojas, Retrograde transport from endosomes to the trans-Golginetwork, Nat. Rev., Mol. Cell Biol. 7 (2006) 568–579.

[15] S. Martinez-Arca, R. Rudge, M. Vacca, G. Raposo, J. Camonis, V. Proux-Gillardeaux,L. Daviet, E. Formstecher, A. Hamburger, F. Filippini, M. D'Esposito, T. Galli, A dualmechanism controlling the localization and function of exocytic v-SNAREs, Proc.Natl. Acad. Sci. U. S. A. 100 (2003) 9011–9016.

[16] A.A. Peden, V. Oorschot, B.A. Hesser, C.D. Austin, R.H. Scheller, J. Klumperman,Localization of the AP-3 adaptor complex defines a novel endosomal exit site forlysosomal membrane proteins, J. Cell Biol. 164 (2004) 1065–1076.

[17] R.L. Williams, S. Urbe, The emerging shape of the ESCRTmachinery, Nat. Rev. Mol.Cell Biol. 8 (2007) 355–368.

[18] C. Kanazawa, E. Morita, M. Yamada, N. Ishii, S. Miura, H. Asao, T. Yoshimori, K.Sugamura, Effects of deficiencies of STAMs andHrs, mammalian class E Vps proteins,on receptor downregulation, Biochem. Biophys. Res. Commun. 309 (2003) 848–856.

[19] M. Razi, C.E. Futter, Distinct roles for Tsg101 and Hrs in multivesicular bodyformation and inward vesiculation, Mol. Biol. Cell 17 (2006) 3469–3483.

[20] A. Doyotte, M.R. Russell, C.R. Hopkins, P.G. Woodman, Depletion of TSG101 formsa mammalian “Class E” compartment: a multicisternal early endosome withmultiple sorting defects, J. Cell. Sci. 118 (2005) 3003–3017.

[21] S. Shim, L.A. Kimpler, P.I. Hanson, Structure/function analysis of four core ESCRT-III proteins reveals common regulatory role for extreme C-terminal domain,Traffic 8 (2007) 1068–1079.

[22] T. Muziol, E. Pineda-Molina, R.B. Ravelli, A. Zamborlini, Y. Usami, H. Gottlinger, W.Weissenhorn, Structural basis for budding by the ESCRT-III factor CHMP3, Dev.Cell 10 (2006) 821–830.

[23] P.I. Hanson, R. Roth, Y. Lin, J.E. Heuser, Plasma membrane deformation by circulararrays of ESCRT-III protein filaments, J. Cell Biol. 180 (2008) 389–402.

[24] S. Lata, G. Schoehn, A. Jain, R. Pires, J. Piehler, H.G. Gottlinger, W. Weissenhorn,Helical structures of ESCRT-III are disassembled by VPS4, Science 321 (2008)1354–1357.

[25] S. Ghazi-Tabatabai, S. Saksena, J.M. Short, A.V. Pobbati, D.B. Veprintsev, R.A.Crowther, S.D. Emr, E.H. Egelman, R.L. Williams, Structure and disassembly offilaments formed by the ESCRT-III subunit Vps24, Structure 16 (2008) 1345–1356.

[26] K.G. Bache, S. Stuffers, L. Malerod, T. Slagsvold, C. Raiborg, D. Lechardeur, S.Walchli, G.L. Lukacs, A. Brech, H. Stenmark, The ESCRT-III subunit hVps24 isrequired for degradation but not silencing of the epidermal growth factorreceptor, Mol. Biol. Cell 17 (2006) 2513–2523.

[27] D.M. Ward, M.B. Vaughn, S.L. Shiflett, P.L. White, A.L. Pollock, J. Hill, R.Schnegelberger, W.I. Sundquist, J. Kaplan, The role of LIP5 and CHMP5 inmultivesicular body formation and HIV-1 budding in mammalian cells, J. Biol.Chem. 280 (2005) 10548–10555.

[28] J.H. Shim, C. Xiao, M.S. Hayden, K.Y. Lee, E.S. Trombetta, M. Pypaert, A. Nara, T.Yoshimori, B. Wilm, H. Erdjument-Bromage, P. Tempst, B.L. Hogan, I. Mellman, S.Ghosh, CHMP5 is essential for late endosome function and down-regulation ofreceptor signaling during mouse embryogenesis, J. Cell Biol. 172 (2006)1045–1056.

[29] C. Progida, M.R. Spinosa, A. De Luca, C. Bucci, RILP interacts with the VPS22component of the ESCRT-II complex, Biochem. Biophys. Res. Commun. 347(2006) 1074–1079.

[30] T. Wang,W. Hong, RILP interacts with VPS22 and VPS36 of ESCRT-II and regulatestheir membrane recruitment, Biochem. Biophys. Res. Commun. 350 (2006)413–423.

[31] C. Progida, L. Malerod, S. Stuffers, A. Brech, C. Bucci, H. Stenmark, RILP is requiredfor the propermorphology and function of late endosomes, J. Cell. Sci. 120 (2007)3729–3737.

[32] I. Jordens, M. Fernandez-Borja, M. Marsman, S. Dusseljee, L. Janssen, J. Calafat, H.Janssen, R. Wubbolts, J. Neefjes, The Rab7 effector protein RILP controlslysosomal transport by inducing the recruitment of dynein–dynactin motors,Curr. Biol. 11 (2001) 1680–1685.

[33] M. Johansson, N. Rocha, W. Zwart, I. Jordens, L. Janssen, C. Kuijl, V.M. Olkkonen,J. Neefjes, Activation of endosomal dynein motors by stepwise assembly ofRab7-RILP-p150Glued, ORP1L, and the receptor betalll spectrin, J. Cell Biol. 176(2007) 459–471.

[34] J.P. Luzio, P.R. Pryor, N.A. Bright, Lysosomes: fusion and function, Nat. Rev. Mol.Cell Biol. 8 (2007) 622–632.

[35] B. Alberts, Molecular biology of the cell, Garland Science; [London: Taylor &Francis, distributor], New York, 2008.

[36] T. Vida, B. Gerhardt, A cell free assay allows reconstitution of Vps33p dependenttransport to the yeast vacuole/lysosome, J. Cell Biol. 146 (1999) 85–98.

[37] B. Storrie, M. Desjardins, The biogenesis of lysosomes: is it a kiss and run,continuous fusion and fission process? Bioessays 18 (1996) 895–903.

[38] B.M. Mullock, N.A. Bright, C.W. Fearon, S.R. Gray, J.P. Luzio, Fusion of lysosomeswith late endosomes produces a hybrid organelle of intermediate density and isNSF dependent, J. Cell Biol. 140 (1998) 591–601.

[39] N.A. Bright, B.J. Reaves, B.M. Mullock, J.P. Luzio, Dense core lysosomes can fusewith late endosomes and are re-formed from the resultant hybrid organelles,J. Cell. Sci. 110 (Pt 17) (1997) 2027–2040.

[40] P.R. Pryor, B.M. Mullock, N.A. Bright, S.R. Gray, J.P. Luzio, The role ofintraorganellar Ca(2+) in late endosome–lysosome heterotypic fusion and inthe reformation of lysosomes from hybrid organelles, J. Cell Biol. 149 (2000)1053–1062.

[41] P.R. Pryor, B.M. Mullock, N.A. Bright, M.R. Lindsay, S.R. Gray, S.C. Richardson, A.Stewart, D.E. James, R.C. Piper, J.P. Luzio, Combinatorial SNARE complexes withVAMP7 or VAMP8 define different late endocytic fusion events, EMBO Rep. 5(2004) 590–595.

[42] R.J. Desnick, E.H. Schuchman, Enzyme replacement and enhancement therapies:lessons from lysosomal disorders, Nat. Rev., Genet. 3 (2002) 954–966.

[43] N.A. Bright, M.J. Gratian, J.P. Luzio, Endocytic delivery to lysosomes mediated byconcurrent fusion and kissing events in living cells, Curr. Biol. 15 (2005) 360–365.

[44] W. Wickner, R. Schekman, Membrane fusion, Nat. Struct. Mol. Biol. 15 (2008)658–664.

[45] C.W. Ostrowicz, C.T. Meiringer, C. Ungermann, Yeast vacuole fusion: a modelsystem for eukaryotic endomembrane dynamics, Autophagy 4 (2008) 5–19.

[46] C.K. Raymond, I. Howald-Stevenson, C.A. Vater, T.H. Stevens, Morphologicalclassification of the yeast vacuolar protein sorting mutants: evidence for aprevacuolar compartment in class E vps mutants, Mol. Biol. Cell 3 (1992)1389–1402.

[47] L.M. Banta, J.S. Robinson, D.J. Klionsky, S.D. Emr, Organelle assembly in yeast:characterization of yeast mutants defective in vacuolar biogenesis and proteinsorting, J. Cell Biol. 107 (1988) 1369–1383.

623P.R. Pryor, J.P. Luzio / Biochimica et Biophysica Acta 1793 (2009) 615–624

[48] D.F. Seals, G. Eitzen, N. Margolis, W.T. Wickner, A. Price, A Ypt/Rab effectorcomplex containing the Sec1 homolog Vps33p is required for homotypic vacuolefusion, Proc. Natl. Acad. Sci. U. S. A. 97 (2000) 9402–9407.

[49] A. Price, D. Seals, W. Wickner, C. Ungermann, The docking stage of yeast vacuolefusion requires the transfer of proteins from a cis-SNARE complex to a Rab/Yptprotein, J. Cell Biol. 148 (2000) 1231–1238.

[50] C. Stroupe, K.M. Collins, R.A. Fratti, W. Wickner, Purification of active HOPScomplex reveals its affinities for phosphoinositides and the SNARE Vam7p,EMBO J. 25 (2006) 1579–1589.

[51] I. Dulubova, T. Yamaguchi, Y. Wang, T.C. Sudhof, J. Rizo, Vam3p structure revealsconserved and divergent properties of syntaxins, Nat. Struct. Biol. 8 (2001)258–264.

[52] R. Laage, C. Ungermann, The N-terminal domain of the t-SNARE Vam3pcoordinates priming and docking in yeast vacuole fusion, Mol. Biol. Cell 12(2001) 3375–3385.

[53] T.K. Sato, P. Rehling, M.R. Peterson, S.D. Emr, Class C Vps protein complexregulates vacuolar SNARE pairing and is required for vesicle docking/fusion,Mol. Cell 6 (2000) 661–671.

[54] D. Fasshauer, R.B. Sutton, A.T. Brunger, R. Jahn, Conserved structural features ofthe synaptic fusion complex: SNARE proteins reclassified as Q- and R-SNAREs,Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 15781–15786.

[55] J.B. Bock, H.T. Matern, A.A. Peden, R.H. Scheller, A genomic perspective onmembrane compartment organization, Nature 409 (2001) 839–841.

[56] R.A. Fratti, K.M. Collins, C.M. Hickey, W. Wickner, Stringent 3Q.1R composition ofthe SNARE 0-layer can be bypassed for fusion by compensatory SNARE mutationor by lipid bilayer modification, J. Biol. Chem. 282 (2007) 14861–14867.

[57] V.J. Starai, C.M. Hickey, W. Wickner, HOPS proofreads the trans-SNARE complexfor yeast vacuole fusion, Mol. Biol. Cell 19 (2008) 2500–2508.

[58] V. Poupon, A. Stewart, S.R. Gray, R.C. Piper, J.P. Luzio, The role of mVps18p inclustering, fusion, and intracellular localization of late endocytic organelles,Mol. Biol. Cell 14 (2003) 4015–4027.

[59] S. Caplan, L.M. Hartnell, R.C. Aguilar, N. Naslavsky, J.S. Bonifacino, HumanVam6p promotes lysosome clustering and fusion in vivo, J. Cell Biol. 154 (2001)109–122.

[60] P. Gissen, C.A. Johnson, N.V. Morgan, J.M. Stapelbroek, T. Forshew, W.N. Cooper,P.J. McKiernan, L.W. Klomp, A.A. Morris, J.E. Wraith, P. McClean, S.A. Lynch, R.J.Thompson, B. Lo, O.W. Quarrell, M. Di Rocco, R.C. Trembath, H. Mandel, S. Wali,F.E. Karet, A.S. Knisely, R.H. Houwen, D.A. Kelly, E.R. Maher, Mutations inVPS33B, encoding a regulator of SNARE dependent membrane fusion, causearthrogryposis-renal dysfunction cholestasis (ARC) syndrome, Nat. Genet. 36(2004) 400–404.

[61] M.R. Peterson, S.D. Emr, The class C Vps complex functions at multiple stages ofthe vacuolar transport pathway, Traffic 2 (2001) 476–486.

[62] A. Srivastava, C.A. Woolford, E.W. Jones, Pep3p/Pep5p complex: a putativedocking factor at multiple steps of vesicular transport to the vacuole ofSaccharomyces cerevisiae, Genetics 156 (2000) 105–122.

[63] S.C. Richardson, S.C. Winistorfer, V. Poupon, J.P. Luzio, R.C. Piper, Mammalian latevacuole protein sorting orthologues participate in early endosomal fusion andinteract with the cytoskeleton, Mol. Biol. Cell 15 (2004) 1197–1210.

[64] K. Peplowska, D.F. Markgraf, C.W. Ostrowicz, G. Bange, C. Ungermann, TheCORVET tethering complex interacts with the yeast Rab5 homolog Vps21 and isinvolved in endo-lysosomal biogenesis, Dev. Cell 12 (2007) 739–750.

[65] S. Duclos, R. Corsini, M. Desjardins, Remodeling of endosomes during lysosomebiogenesis involves ‘kiss and run’ fusion events regulated by rab5, J. Cell. Sci. 116(2003) 907–918.

[66] Y. Feng, B. Press, A. Wandinger-Ness, Rab 7: an important regulator of lateendocytic membrane traffic, J. Cell Biol. 131 (1995) 1435–1452.

[67] C. Bucci, P. Thomsen, P. Nicoziani, J. McCarthy, B. van Deurs, Rab7: a key tolysosome biogenesis, Mol. Biol. Cell 11 (2000) 467–480.

[68] S. Meresse, J.P. Gorvel, P. Chavrier, The rab7 GTPase resides on a vesicularcompartment connected to lysosomes, J. Cell. Sci. 108 (Pt 11) (1995) 3349–3358.

[69] J. Rink, E. Ghigo, Y. Kalaidzidis, M. Zerial, Rab conversion as a mechanism ofprogression from early to late endosomes, Cell 122 (2005) 735–749.

[70] C. Liang, J.S. Lee, K.S. Inn, M.U. Gack, Q. Li, E.A. Roberts, I. Vergne, V. Deretic, P.Feng, C. Akazawa, J.U. Jung, Beclin1-binding UVRAG targets the class C Vpscomplex to coordinate autophagosome maturation and endocytic trafficking,Nat. Cell Biol. 10 (2008) 776–787.

[71] C. Liang, D. Sir, S. Lee, J.H. Ou, J.U. Jung, Beyond autophagy: the role of UVRAG inmembrane trafficking, Autophagy 4 (2008) 817–820.

[72] G. Cantalupo, P. Alifano, V. Roberti, C.B. Bruni, C. Bucci, Rab-interacting lysosomalprotein (RILP): the Rab7 effector required for transport to lysosomes, EMBO J. 20(2001) 683–693.

[73] E.A. Sevrioukov, J.P. He, N. Moghrabi, A. Sunio, H. Kramer, A role for the deeporange and carnation eye color genes in lysosomal delivery in Drosophila,Mol. Cell 4 (1999) 479–486.

[74] T. Suzuki, N. Oiso, R. Gautam, E.K. Novak, J.J. Panthier, P.G. Suprabha, T. Vida, R.T.Swank, R.A. Spritz, The mouse organellar biogenesis mutant buff results from amutation in Vps33a, a homologue of yeast vps33 and Drosophila carnation,Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 1146–1150.

[75] P. Gissen, C.A. Johnson, D. Gentle, L.D. Hurst, A.J. Doherty, C.J. O'Kane, D.A. Kelly,E.R. Maher, Comparative evolutionary analysis of VPS33 homologues: geneticand functional insights, Hum. Mol. Genet. 14 (2005) 1261–1270.

[76] H. Bach, K.G. Papavinasasundaram, D. Wong, Z. Hmama, Y. Av-Gay,Mycobacteriumtuberculosis virulence is mediated by PtpA dephosphorylation of human vacuolarprotein sorting 33B, Cell Host Microbe. 3 (2008) 316–322.

[77] B. Gerhardt, T.J. Kordas, C.M. Thompson, P. Patel, T. Vida, The vesicle transport

protein Vps33p is an ATP-binding protein that localizes to the cytosol in anenergy dependent manner, J. Biol. Chem. 273 (1998) 15818–15829.

[78] B. Schroder, C. Wrocklage, C. Pan, R. Jager, B. Kosters, H. Schafer, H.P. Elsasser, M.Mann, A. Hasilik, Integral and associated lysosomal membrane proteins, Traffic 8(2007) 1676–1686.

[79] T.J. LaGrassa, C. Ungermann, The vacuolar kinase Yck3 maintains organellefragmentation by regulating the HOPS tethering complex, J. Cell Biol. 168 (2005)401–414.

[80] C.W. Wang, P.E. Stromhaug, E.J. Kauffman, L.S. Weisman, D.J. Klionsky, Yeasthomotypic vacuole fusion requires the Ccz1–Mon1 complex during thetethering/docking stage, J. Cell Biol. 163 (2003) 973–985.

[81] F. Wang, P.N. Paradkar, A.O. Custodio, D. McVey Ward, M.D. Fleming, D.Campagna, K.A. Roberts, V. Boyartchuk, W.F. Dietrich, J. Kaplan, N.C. Andrews,Genetic variation in Mon1a affects protein trafficking and modifies macrophageiron loading in mice, Nat. Genet. 39 (2007) 1025–1032.

[82] M. Hoffman-Sommer, M. Grynberg, R. Kucharczyk, J. Rytka, The CHiPS Domain—ancient traces for the Hermansky–Pudlak syndrome, Traffic 6 (2005) 534–538.

[83] C. Schluter, K.K. Lam, J. Brumm, B.W. Wu, M. Saunders, T.H. Stevens, J. Bryan, E.Conibear, Global analysis of yeast endosomal transport identifies the Vps55/68sorting complex, Mol. Biol. Cell 19 (2008) 1282–1294.

[84] N. Belgareh-Touze, S. Avaro, Y. Rouille, B. Hoflack, R. Haguenauer-Tsapis, YeastVps55p, a functional homolog of human obesity receptor gene-related protein, isinvolved in late endosome to vacuole trafficking, Mol. Biol. Cell 13 (2002)1694–1708.

[85] W. Antonin, C. Holroyd, D. Fasshauer, S. Pabst, G.F. Von Mollard, R. Jahn, A SNAREcomplex mediating fusion of late endosomes defines conserved properties ofSNARE structure and function, EMBO J. 19 (2000) 6453–6464.

[86] W.Wen, L. Chen, H.Wu, X. Sun, M. Zhang, D.K. Banfield, Identification of the yeastR SNARE Nyv1p as a novel longin domain containing protein, Mol. Biol. Cell 17(2006) 4282–4299.

[87] L.C. Gonzalez Jr., W.I. Weis, R.H. Scheller, A novel snare N-terminal domainrevealed by the crystal structure of Sec22b, J. Biol. Chem. 276 (2001)24203–24211.

[88] H. Tochio, M.M. Tsui, D.K. Banfield, M. Zhang, An autoinhibitory mechanism fornonsyntaxin SNARE proteins revealed by the structure of Ykt6p, Science 293(2001) 698–702.

[89] F. Filippini, V. Rossi, T. Galli, A. Budillon, M. D'Urso, M. D'Esposito, Longins: a newevolutionary conserved VAMP family sharing a novel SNARE domain, TrendsBiochem. Sci. 26 (2001) 407–409.

[90] B.Y. Kim, H. Kramer, A. Yamamoto, E. Kominami, S. Kohsaka, C. Akazawa,Molecular characterization of mammalian homologues of class C Vps proteinsthat interact with syntaxin-7, J. Biol. Chem. 276 (2001) 29393–29402.

[91] C.L. Brett, R.L. Plemel, B.T. Lobinger, M. Vignali, S. Fields, A.J. Merz, Efficienttermination of vacuolar Rab GTPase signaling requires coordinated action by aGAP and a protein kinase, J. Cell Biol. 182 (2008) 1141–1151.

[92] C. Peters, A. Mayer, Ca2+/calmodulin signals the completion of docking andtriggers a late step of vacuole fusion, Nature 396 (1998) 575–580.

[93] V.J. Starai, N. Thorngren, R.A. Fratti, W. Wickner, Ion regulation of homotypicvacuole fusion in Saccharomyces cerevisiae, J. Biol. Chem. 280 (2005) 16754–16762.

[94] J.C. Hay, Calcium: a fundamental regulator of intracellular membrane fusion?EMBO Rep. 8 (2007) 236–240.

[95] C. Peters, P.D. Andrews, M.J. Stark, S. Cesaro-Tadic, A. Glatz, A. Podtelejnikov, M.Mann, A. Mayer, Control of the terminal step of intracellular membrane fusion byprotein phosphatase 1, Science 285 (1999) 1084–1087.

[96] A.J. Merz, W.T. Wickner, Resolution of organelle docking and fusion kinetics in acell free assay, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 11548–11553.

[97] Y. Jun, W. Wickner, Assays of vacuole fusion resolve the stages of docking, lipidmixing, and content mixing, Proc. Natl. Acad. Sci. U. S. A. 104 (2007)13010–13015.

[98] D.C. Ko, M.D. Gordon, J.Y. Jin, M.P. Scott, Dynamic movements of organellescontaining Niemann–Pick C1 protein: NPC1 involvement in late endocyticevents, Mol. Biol. Cell 12 (2001) 601–614.

[99] B.M. Collins, The structure and function of the Retromer protein complex, Traffic9 (2008) 1811–1822.

[100] J.S. Bonifacino, J.H. Hurley, Retromer, Curr. Opin. Cell Biol. 20 (2008) 427–436.[101] S.A. Slaugenhaupt, J.S. Acierno Jr., L.A. Helbling, C. Bove, E. Goldin, G. Bach, R.

Schiffmann, J.F. Gusella, Mapping of the mucolipidosis type IV gene tochromosome 19p and definition of founder haplotypes, Am. J. Hum. Genet. 65(1999) 773–778.

[102] R. Bargal, N. Avidan, E. Ben-Asher, Z. Olender, M. Zeigler, A. Frumkin, A. RaasRothschild, G. Glusman, D. Lancet, G. Bach, Identification of the gene causingmucolipidosis type IV, Nat. Genet. 26 (2000) 118–123.

[103] M.T. Bassi, M. Manzoni, E. Monti, M.T. Pizzo, A. Ballabio, G. Borsani, Cloning of thegene encoding a novel integralmembrane protein,mucolipidin and identificationof the two major founder mutations causing mucolipidosis type IV, Am. J. Hum.Genet. 67 (2000) 1110–1120.

[104] M. Sun, E. Goldin, S. Stahl, J.L. Falardeau, J.C. Kennedy, J.S. Acierno Jr., C. Bove, C.R.Kaneski, J. Nagle, M.C. Bromley, M. Colman, R. Schiffmann, S.A. Slaugenhaupt,Mucolipidosis type IV is caused by mutations in a gene encoding a noveltransient receptor potential channel, Hum. Mol. Genet. 9 (2000) 2471–2478.

[105] S. Treusch, S. Knuth, S.A. Slaugenhaupt, E. Goldin, B.D. Grant, H. Fares, Caenor-habditis elegans functional orthologue of human protein h-mucolipin-1 isrequired for lysosome biogenesis, Proc. Natl. Acad. Sci. U. S. A. 101 (2004)4483–4488.

[106] R.C. Piper, J.P. Luzio, CUPpling calcium to lysosomal biogenesis, Trends Cell Biol.14 (2004) 471–473.

624 P.R. Pryor, J.P. Luzio / Biochimica et Biophysica Acta 1793 (2009) 615–624

[107] E.G. Thompson, L. Schaheen, H. Dang, H. Fares, Lysosomal trafficking functions ofmucolipin-1 in murine macrophages, BMC Cell Biol. 8 (2007) 54.

[108] M.T. Miedel, Y. Rbaibi, C.J. Guerriero, G. Colletti, K.M.Weixel, O.A.Weisz, K. Kiselyov,Membrane traffic and turnover in TRP-ML1-deficient cells: a revised model formucolipidosis type IV pathogenesis, J. Exp. Med. 205 (2008) 1477–1490.

[109] N.P. Kinnear, F.X. Boittin, J.M. Thomas, A. Galione, A.M. Evans, Lysosome–sarcoplasmic reticulum junctions. A trigger zone for calcium signaling bynicotinic acid adenine dinucleotide phosphate and endothelin-1, J. Biol. Chem.279 (2004) 54319–54326.

[110] R.D. Bagshaw, D.J. Mahuran, J.W. Callahan, A proteomic analysis of lysosomalintegral membrane proteins reveals the diverse composition of the organelle,Mol. Cell. Proteomics 4 (2005) 133–143.

[111] F. Zhang, P.L. Li, Reconstitution and characterization of a nicotinic acid adeninedinucleotide phosphate (NAADP)-sensitive Ca2+ release channel from liverlysosomes of rats, J. Biol. Chem. 282 (2007) 25259–25269.

[112] T.J. Jentsch, M. Poet, J.C. Fuhrmann, A.A. Zdebik, Physiological functions of CLC Cl− channels gleaned from human genetic disease and mouse models, Annu. Rev.Physiol. 67 (2005) 779–807.

[113] T.J. Jentsch, Chloride and the endosomal–lysosomal pathway: emerging roles ofCLC chloride transporters, J. Physiol. 578 (2007) 633–640.

[114] M. Yoshikawa, S. Uchida, J. Ezaki, T. Rai, A. Hayama, K. Kobayashi, Y. Kida, M. Noda,M. Koike, Y. Uchiyama, F. Marumo, E. Kominami, S. Sasaki, CLC-3 deficiency leadsto phenotypes similar to human neuronal ceroid lipofuscinosis, Genes Cells 7(2002) 597–605.

[115] M. Poet, U. Kornak, M. Schweizer, A.A. Zdebik, O. Scheel, S. Hoelter, W. Wurst, A.Schmitt, J.C. Fuhrmann, R. Planells-Cases, S.E. Mole, C.A. Hubner, T.J. Jentsch,

Lysosomal storage disease upon disruption of the neuronal chloride transportprotein ClC-6, Proc. Natl. Acad. Sci. U. S. A. 103 (2006) 13854–13859.

[116] D. Kasper, R. Planells-Cases, J.C. Fuhrmann, O. Scheel, O. Zeitz, K. Ruether, A. Schmitt,M. Poet, R. Steinfeld, M. Schweizer, U. Kornak, T.J. Jentsch, Loss of the chloridechannel ClC-7 leads to lysosomal storage disease and neurodegeneration, EMBO J.24 (2005) 1079–1091.

[117] A.R. Graves, P.K. Curran, C.L. Smith, J.A. Mindell, The Cl−/H+ antiporter ClC-7is the primary chloride permeation pathway in lysosomes, Nature 453 (2008)788–792.

[118] K. Bowers, B.P. Levi, F.I. Patel, T.H. Stevens, The sodium/proton exchanger Nhx1pis required for endosomal protein trafficking in the yeast Saccharomycescerevisiae, Mol. Biol. Cell 11 (2000) 4277–4294.

[119] N. Nakamura, S. Tanaka, Y. Teko, K. Mitsui, H. Kanazawa, Four Na+/H+exchanger isoforms are distributed to Golgi and post Golgi compartments andare involved in organelle pH regulation, J. Biol. Chem. 280 (2005) 1561–1572.

[120] P. Liberali, P. Ramo, L. Pelkmans, Protein kinases: starting a molecular-systemsview of endocytosis, Annu. Rev. Cell Dev. Biol. 24 (2008) 501–523.

[121] G. Di Paolo, P. De Camilli, Phosphoinositides in cell regulation and membranedynamics, Nature 443 (2006) 651–657.

[122] J.A. Philips, M.C. Porto, H. Wang, E.J. Rubin, N. Perrimon, ESCRT factors restrictmycobacterial growth, Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 3070–3075.

[123] T.E. Rusten, T. Vaccari, K. Lindmo, L.M. Rodahl, I.P. Nezis, C. Sem-Jacobsen, F.Wendler, J.P. Vincent, A. Brech, D. Bilder, H. Stenmark, ESCRTs and Fab1 regulatedistinct steps of autophagy, Curr. Biol. 17 (2007) 1817–1825.

[124] D.C. Rubinsztein, The roles of intracellular protein-degradation pathways inneurodegeneration, Nature 443 (2006) 780–786.