Microvesiculation and Disease 277

The cis-acting signals that target proteinsto exosomes and microvesiclesJr-Ming Yang and Stephen J. Gould1

Department of Biological Chemistry, Johns Hopkins University, Baltimore, MD 21205, U.S.A.

AbstractProteins bud from cells in small single-membraned vesicles (50250 nm) that have the same topology asthe cell. Known variously as exosomes and microvesicles (EMVs), these extracellular organelles are enrichedfor specic proteins, lipids, carbohydrates and nucleic acids. EMV biogenesis plays critical roles in proteinquality control and cell polarity, and, once released, EMVs can transmit signals andmolecules to neighbouringcells via a non-viral pathway of intercellular vesicle trafc. In the present paper, we discuss the cis-actingtargeting signals that target proteins to EMVs and mediate protein budding from the cell.

BackgroundSecreted vesicles play roles in numerous physiologicalprocesses, including calcification of the extracellular matrix,germ cell development, erythroid differentiation and immunesignalling [14]. Secreted vesicles also contribute to humandiseases, including cancer, viral infections and neurode-generative disorders. This array of biological functionsstarts with the biogenesis of secreted vesicles, a processthat has been apparent for several decades, but remainspoorly understood. The first observations of vesicle secretioncoincided with some of the earliest electron micrographsof animal cells [5,6], which showed the release of smallsingle-membraned vesicles 100 nm in diameter. However,the notion that cells might produce biochemically importantvesicles only arose slowly, over the course of several decades.Key observations along this path were the recognition thathost cell molecules retained functionality when released onthe surface of retrovirus particles [711], and, even moreimportantly, the discovery of calcifying matrix vesiclesthat bud from chondrocytes [12], a vesicular clotting factorreleased by platelets [13], and budding of antigen receptorsby lymphocytes [14].

Although these early reports focused on vesicles thatwere secreted by specialized cell types or under specializedconditions, it is nowapparent that vesicle secretion is a generalproperty of animal cells. The most common term appliedto secreted vesicles is exosome, defined first by Tramset al. in 1981 [15] as cell-derived vesicles that may serve aphysiologic function. Trams et al. [15] also demonstratedthat animal cells secrete two size classes of vesicles. Thesmaller vesicles are in the range 50250 nm in diameterand can be easily separated from the larger class of vesicles

Key words: exosome, extracellular vesicle, HIV, microvesicle, plasma membrane, retrovirus,

secreted vesicle.

Abbreviations used: Alix, ALG-2 (apoptosis-linked gene 2)-interacting protein X; ARRDC1,

arrestin domain-containing protein-1; EMV, exosome/microvesicle; ERM, ezrin/radixin/moesin;

ESCRT, endosomal sorting complex required for transport; MFG-E8, milk fat globuleepidermal

growth factor factor 8 protein; miRNA, microRNA; PE, phosphatidylethanolamine; RNP, ribonucle-

oprotein.1To whom correspondence should be addressed (email sgould@jhmi.edu).

(0.52 m or more) by differential centrifugation: largervesicles sediment at 10000 g (after 3060 min), whereasthe smaller class of secreted vesicles pellet preferentially athigher force, 100000 g (after 6090 min) [16,17]. In fact,it has been proposed that these two classes of vesiclesbe named differently, with the term microvesicle used todescribe the larger vesicles that pellet at 10000 g, andexosome used to describe the smaller class of secretedvesicles that pellet at 100000 g [18]. Other investigatorshave proposed a pair of biogenic definitions,with vesicles thatbud from the plasma membranes being called microvesicles,and exosome referring only to extracellular vesicles thatarise by (i) budding into endosomes, and (ii) fusion ofvesicle-containing endosomes with the plasmamembrane [3].However, there are two substantial flaws with this latterapproach: first, there is as yet no evidence for a significantmechanistic difference between vesicle budding at the plasmaand endosome membranes; and, secondly, the definition isempirically unworkable because it is not possible to separatevesicles of similar physical and molecular properties basedon the membrane from which they arose. In this context, itis difficult to know exactly which terms should be appliedto which vesicles. In lieu of a better option, our grouprefers to the smaller class of secreted vesicle by the hybridterm exosome/microvesicle, or EMV [1922], even thoughit conforms to the empirical (i.e. centrifugal sedimentation-based [18]) definition of an exosome.

EMV biogenesis: cis-acting EMV targetingsignalsEMVbiogenesis involves (i) the trafficking of cargomoleculesto sites of outward vesicle budding (outward = away fromthe cytoplasm), (ii) enrichment of cargo molecules in nascentvesicles, (iii) scission of cargo-containing vesicles from the cellmembrane, and (iv) the release of vesicles into the extracellularmilieu. As with all organelle biogenesis pathways, the cis-acting signals that target proteins to EMVs are likely toprovide important insights into the mechanisms of EMVbiogenesis and protein budding. One of the earliest studies

Biochem. Soc. Trans. (2013) 41, 277282; doi:10.1042/BST20120275 CThe Authors Journal compilation C2013 Biochemical SocietyBio

chem

ical

Soc

iety

Tra

nsac

tions

w

ww

.bio

chem

soct

rans

.org

278 Biochemical Society Transactions (2013) Volume 41, part 1

on the cis-acting signals sufficient to target proteins to EMVswas by Fang et al. [23], which demonstrated that plasmamembrane binding and higher-order oligomerization aresufficient to target proteins to sites of vesicle budding andinto EMVs. This conclusion was based on two separate, butcomplementary, observations: first, that antibody-inducedhigher-order oligomerization of cell-surface proteins led totheir secretion from the cell in EMVs; and, secondly, thataddition of a plasmamembrane anchorwas sufficient to targeta highly oligomeric cytoplasmic protein to EMVs [23]. Shenet al. [22] extended these observations by showing that (i)plasma membrane anchors can target additional unrelatedcytoplasmic proteins into EMVs, (ii) EMV targeting canbe induced by a wide array of plasma membrane anchors[a myristoylation tag, a prenylation/palmitoylation tag,a PtdIns(4,5)P2-binding domain, a PtdIns(3,4,5)P3-bindingdomain and a type-1 integral plasma membrane protein], and(iii) these plasma membrane anchors can be placed at eitherthe N-terminal or the C-terminal end of the cargo proteinsand still target them to EMVs [22].

Shen et al. [22] also reported the somewhat surprisingfinding that endosomemembrane anchors are unable to targetproteins into EMVs, even though the exact same test proteinswere readily targeted to EMVs by plasmamembrane anchors.These findings run counter to the widespread presumptionthat the endosome is a major site of vesicle biogenesis.However, they are highly similar to what has been foundfor retrovirus budding, a process that is driven by Gag, themajor structural protein of retroviruses. Specifically, it hasbeen demonstrated that (i) retroviral Gag proteins possessplasmamembrane anchors, not endosomemembrane anchors[24], (ii) rational targeting of Gag to the plasma membranewith heterologous plasma membrane anchors continues tosupport its budding and infectivity [25], and (iii) rationaltargeting of Gag to endosome membranes cannot supportvirus budding from the cell [25].

Do EMV cargoes have the predictedproperties?The hypothesis that plasma-membrane-binding and higher-order oligomerization target proteins to EMVs makes anumber of predictions about the nature of EMV cargoproteins. The most direct prediction is that known EMVcargoes form higher-order oligomeric complexes in or on theplasma membrane. This appears to be the case. Tetraspaninsrepresent some of the most widely used exosome markerproteins (e.g. CD9, CD63, CD81 and CD82 [26]), areall plasma membrane proteins, and are commonly foundin large heterogeneous complexes (e.g. the tetraspaninweb [27]). Another example are the ERM (ezrin/radixin/moesin) proteins and other adaptors [e.g. EBP50 (ERM-binding protein 50)] that cross-link actin filaments toselected plasma membrane proteins, generating large hetero-oligomeric complexes [28] that enter the EMV proteinsorting pathway [2931]. The Gag proteins of endogenousand exogenous retroviruses represent yet another group of

EMV cargo proteins that possess a plasma membrane anchorand multiple oligomerization domains [32,33]. Furthermore,detailed studies of the HIV Gag protein have demonstratedthat these properties, i.e. higher-order oligomerizationand plasma membrane binding, are its primary buddinginformation [23]. In fact, the mechanistic similarity betweenretrovirus budding and EMV biogenesis is so strong thatone can target virtually any plasma membrane protein to thesurface of HIV particles merely by inducing its higher-orderoligomerization [23].

The hypothesis that proteins are targeted to EMVs on thebasis of higher-order oligomerization and plasma membranebinding also predicts that a proteins trafficking to EMVs canbe regulated, by altering its affinity for the plasma membraneand/or degree of oligomerization. Several lines of evidencesatisfy this prediction, including the differential traffickingof TfR (transferrin receptor) to EMVs during reticulocytematuration [3436], the antibody-induced budding of antigenreceptors from B-cells and T-cells [14,23], the aggregation-induced budding of the prion protein [37,38], and the rapidand reversible trafficking of uropod proteins to sites of EMVbudding during the polarization of human leucocytes [21].

A third prediction of our operating hypothesis is thatEMV cargo proteins will, in general, lack shared amino acidsequence motifs or features. This prediction is based onthe fact that a wide array of sequences can confer plasmamembrane binding and/or higher-order oligomerization. Forexample, plasma membrane binding can be mediated bya diverse array of sequence motifs and structures, fromacylation sites (e.g. a glycine residue at position 2 of a protein,which can confer its N-myristoylation) to phospholipid-binding domains such as those that bind PtdIns(4,5)P2 orPtdIns(3,4,5)P3 [22]. As for protein oligomerization, it toois mediated by unrelated sequences that range from thesimple (e.g. a leucine zipper), moderately complex (e.g.oligomerization interfaces on large globular enzymes) tohighly complex [e.g. the multiple protein and RNA contactsin large RNP (ribonucleoprotein) complexes] amino acidsequences. This predictionmatches empirical observations, asEMVs contain proteins as diverse as tetraspanins, integrins,MHC proteins, chaperones, RNPs, cytoskeletal proteins andretroviral Gag proteins.

ESCRTs and ESCRT-binding sites in proteinbuddingPerhaps the most intriguing group of EMV cargoes arethe ESCRTs (endosomal sorting complexes required fortransport) [3941]. These protein complexes (ESCRT-0,ESCRT-I, ESCRT-II and ESCRT-III) are best knownfor their roles in the trafficking of ubiquitylated plasmamembrane proteins to lysosomes in a pathway that involvestheir (i) endocytosis, (ii) trafficking to the limiting membraneof endosomes, (iii) budding into the endosome lumen,(iv) delivery to the lysosome lumen, and (v) lysosomaldestruction.However, ESCRTproteins are also found at highlevels at the plasma membrane [42] and are characterized

CThe Authors Journal compilation C2013 Biochemical Society

Microvesiculation and Disease 279

by their presence in membrane-bound highly oligomericprotein complexes. Given that these properties are sufficientto target proteins to EMVs, it is not surprising that manyESCRT proteins are enriched in secreted vesicles [4345].The targeting of ESCRT proteins to EMVs can also be seen inthe constitutive budding of oligomeric membrane-associatedESCRT-III proteins that lack their autoinhibitoryC-terminaldomain [46]. Not surprisingly, a proteins association withESCRT components can, in some instances, trigger itsbudding from the cell. For example, the budding ofARRDC1 (arrestin domain-containing protein-1) [47], andof syntenin [48], appears to be mediated by their Tsg101(tumour susceptibility gene 101)-binding motif [P(T/S)AP]and Alix [ALG-2 (apoptosis-linked gene 2)-interactingprotein X]-binding motif (YPXL) respectively. One wayto interpret these findings is that the binding of ARRDC1and/or syntenin with their ESCRT ligands converts theminto plasma-membrane-localized higher-order oligomericcomplexes that enter the EMV protein sorting pathway.

Another way to interpret these observations is that theESCRT-binding motifs induce the budding of proteins suchas ARRDC1 and syntenin by recruiting a catalytic activityof the ESCRT machinery. In vivo, loss of ESCRT functioncan inhibit several topologically similar processes, includingthe biogenesis of most (but not all) MVBs (multivesicularbodies), cytokinesis and the budding of several envelopedviruses, particularlyHIV andother retroviruses [40,41,49,50].Furthermore, in vitro studies have established that theaddition of constitutively oligomeric ESCRT-III proteinsto phospholipid membranes results in their budding intosmall vesicles [51]. These and other observations have ledto the hypothesis that a major role of ESCRTs is to catalysethe scission of nascent vesicles from parent membranes.However, other lines of evidence raise doubts about whetherthis mechanistic model applies to the role of ESCRTs inEMV biogenesis. First, protein ubiquitylation, the primarysignal for engagement with the ESCRT machinery, does nottarget membrane proteins to EMVs [23]. Secondly, inhibitionof ESCRT function fails to impair EMV biogenesis [23], aresult that has been replicated subsequently [52]. Thirdly,the fact that ESCRT-III proteins causes budding of vesiclesfrom synthetic membranes in vitro demonstrates that vesiclebudding is not only a cargo-dependent process, but is alsoa cargo-driven process. The emerging complexity of the rolethat ESCRTs play in EMV biogenesis is underscored by theobservation that VPS4 (vacuolar protein sorting 4), an AAA(ATPase associated with various cellular activities) proteinthat dissociates ESCRT-III protein complexes, also binds toother EMV cargo proteins such as CD63 and CD81 [19],even though these proteins bud from cells in an ESCRT-independent manner [23].

Other cargoes, other signalsAlthough higher-order oligomerization and plasma mem-brane binding might be the primary signals for targetingproteins into nascent vesicles, it is unlikely that all EMV

cargoes will possess these properties. For instance, virtuallyany cytoplasmic protein could be brought into nascentEMVs, piggyback style, merely by binding to an EMVcargo protein. In addition to potentially explaining theimport of some ESCRT-associated proteins such as Alix orARRDC1, a piggyback mechanism might also explain thevesicular secretion of B-crystallin [53,54] and chaperones[55,56], proteins that are known to bind misfolded andaggregated plasma membrane proteins. As for the existenceof completely distinct mechanisms for targeting proteinsto secreted vesicles, precedent alone suggests that these arelikely to exist, as multiple signals are able to target proteinsto all other organelles. In fact, the vesicular budding ofMFG-E8 (milk fat globuleepidermal growth factor factor 8protein)/lactadherin appears to involve a distinct mechanism:MFG-E8/lactadherin is co-translationally translocated intothe endoplasmic reticulum lumen, secreted by the classic exo-cytic pathway, and is recruited to the outer surface of EMVsvia its phosphatidylserine-binding C2 domains [57,58].

EMVs are also enriched for specific nucleic acids,particularly miRNAs (microRNAs) [59,60]. Given thatRNAs typically exist in complex with proteins, their selectiveenrichment in EMVs is likely to be another case of piggybacktrafficking, wherein the selective trafficking of one or moreRNA-binding proteins to EMVs results in the selectivebudding of their cognate RNAs. Numerous RNPs havebeen reported in EMVs, includingRNPA2B1 (heterogeneousRNPA2/B1) [61] and RBM3 (RNA-bindingmotif protein 3)[62]. However, the most abundant RNA-binding proteins inEMVsare retroviralGagproteins expressed fromendogenousretroviruses [63]. Gag proteins bind a wide array of RNAspecies, are capable of incorporating long RNAs of up to10 kb in length into a secreted vesicle, and might representthe mechanistic basis for EMV-mediated intercellular trafficof mRNAs. The vesicular secretion of miRNAs mightalso be mediated by Gag proteins or other RNA-bindingproteins that are secreted from the cell in EMVs, suchas RNPA2B1 [61] or the RISC (RNA-induced silencingcomplex) components GW182 and Argonaut [64].

Covalent linkage to EMV cargo proteins is anotherpotential mode of EMV protein targeting. For example,the isopeptide linkage of ubiquitin to highly oligomericplasma membrane proteins is likely to explain its presencein secreted vesicles [65]. Another example is the enrichmentof specific carbohydrate adducts on surface glycoproteinsand glycolipids, particularly high mannose, polylactosa-mine and -2,6-sialic acid residues [66]. The discovery ofshared carbohydrate residues on the surface of EMVs isintriguing, for it suggests that either (i) this carbohydratemodification is selectively added to EMVs cargos, or (ii)addition of this carbohydrate moiety somehow targetsglycoproteins and glycolipids to sites of EMV budding.Either way, it suggests that a better understanding of EMV-selective carbohydrate modifications will lead to a betterunderstanding of EMV biogenesis. Interestingly, the samecarbohydrates that are enriched on the surface of EMVs arealso enriched on the surface of HIV particles [67].

CThe Authors Journal compilation C2013 Biochemical Society

280 Biochemical Society Transactions (2013) Volume 41, part 1

The biogenesis of EMVs also involves the selectivetargeting of specific lipids to sites of vesicle budding andtheir enrichment in nascent vesicles. EMVs are typicallyenriched for cholesterol and sphingolipids, and ceramide haseven been implicated as a catalyst of vesicle budding [52,68].The selective trafficking of lipids to EMV buddinghas also been observed for N-modified forms of PE (phos-phatidylethanolamine), including 1,2-dioleoyl-sn-glycero-3-PE-N-(lissamine rhodamine B sulfonyl), 1,2-dioleoyl-sn-glycero-3-PE-N-(carboxyfluorescein) and 1,2-dioleoyl-sn-glycero-3-PE-N-(biotinyl) [33,69]. Specifically, whenthese lipids are incorporated into the outer leaflet of theplasma membrane of living cells, they are rapidly traffickedto sites of EMV biogenesis and secreted from the cell invesicles. Interestingly, depletion of cholesterol by additionof methyl--cyclodextrin to the culture medium causes therelease of these N-modified forms of PE from sites of vesiclebudding and their redistribution to the plasma membranein general, as well as to endomembrane structures [33].However, cholesterol depletion does not alter the enrichmentof EMV cargo proteins at sites of vesicle budding, andremoval of methyl--cyclodextrin from the medium leadsto the re-accumulation of N-modified forms of PE at sitesof vesicle budding [33]. The most parsimonious explanationfor these varied results is that the sites of vesicle budding aredetermined by the trafficking of EMV cargo proteins, theenrichment of these cargoes generates a unique membranedomain that attracts specific lipid molecules, and that alteringthe levels of these lipids can enhance or inhibit the formationof secreted vesicles. It is currently unclear whether thesephenomena are determined solely by the intrinsic biophysicalproperties of EMV lipids and cargo proteins [70], or whetheractive sorting mechanisms are involved.

It should also be noted that EMVs contain numerousmolecules that are not selectively enriched in EMVs. TheseEMV components arrive at sites of budding merely bydiffusion within the cytoplasm and/or plasma membrane.As such, the mere detection of a protein or lipid in EMVsdoes not mean that it will possess a specific sorting signal. Italso suggests that the bulk proteomic and lipidomic studiesof purified EMVs are likely to overestimate the number ofEMV-targeted proteins, since additional tests are requiredto determine whether a given cargo is actually enriched insecreted vesicles, or whether it is present in EMVs due todiffusion alone.

Future directionsThe identification of cis-acting signals that target proteinsto EMVs represents the first step in understanding EMVbiogenesis. Next, it will be important to understand how thecell recognizes proteins that contain these signals, trafficksthem to sites of EMV budding, and secretes them from thecell in vesicles. In the case of the ESCRT-binding sequencemotifs that appear to promote the budding of at least someEMV cargoes, much can be deduced from the existingknowledge about these well-studied motifs and the ESCRT

machinery, which should lead to relatively rapid progresson this front. A greater challenge lies in elucidating howhigher-order oligomerization and plasma membrane bindinggenerate a discrete biochemical signal, how this signal mightbe recognized by the cell, and how signal-containing proteinsand/or protein complexes are targeted to EMVs.

Funding

This work was supported by Johns Hopkins University and the

National Institutes of Health [grant number R01 DK45787].

References1 Anderson, H.C., Garimella, R. and Tague, S.E. (2005) The role of matrix

vesicles in growth plate development and biomineralization. Front.Biosci. 10, 822837

2 van der Pol, E., Boing, A.N., Harrison, P., Sturk, A. and Nieuwland, R.(2012) Classication, functions, and clinical relevance of extracellularvesicles. Pharmacol. Rev. 64, 676705

3 Simons, M. and Raposo, G. (2009) Exosomes: vesicular carriers forintercellular communication. Curr. Opin. Cell Biol. 21, 575581

4 Keller, S., Sanderson, M.P., Stoeck, A. and Altevogt, P. (2006) Exosomes:from biogenesis and secretion to biological function. Immunol. Lett. 107,102108

5 Claude, A., Porter, K.R. and Pickels, E.G. (1947) Electron microscope studyof chicken tumor cells. Cancer Res. 7, 421430

6 Porter, K.R. and Thompson, H.P. (1948) A particulate body associatedwith epithelial cells cultured from mammary carcinomas of mice of amilkfactor strain. J. Exp. Med. 88, 1524

7 de The, G., Becker, C. and Beard, J.W. (1964) Virus of avianmyeloblastosis (BAI strain A). XXV. Ultracytochemical study of virus andmyeloblast phosphatase activity. J. Natl. Cancer Inst. 32, 201235

8 Azocar, J. and Essex, M. (1979) Incorporation of HLA antigens into theenvelope of RNA tumor viruses grown in human cells. Cancer Res. 39,33883391

9 Azocar, J., Essex, M. and Yunis, E.J. (1983) Cell culture density modulatesthe incorporation of HLA antigens by enveloped viruses. Hum. Immunol.7, 5965

10 Bubbers, J.E., Chen, S. and Lilly, F. (1978) Nonrandom inclusion of H-2Kand H-2D antigens in Friend virus particles from mice of various strains.J. Exp. Med. 147, 340351

11 Bubbers, J.E. and Lilly, F. (1977) Selective incorporation of H-2 antigenicdeterminants into Friend virus particles. Nature 266, 458459

12 Anderson, H.C. (1969) Vesicles associated with calcication in the matrixof epiphyseal cartilage. J. Cell Biol. 41, 5972

13 Wolf, P. (1967) The nature and signicance of platelet products inhuman plasma. Br. J. Haematol. 13, 269288

14 Karnovsky, M.J., Unanue, E.R. and Leventhal, M. (1972) Ligand-inducedmovement of lymphocyte membrane macromolecules. II. Mapping ofsurface moieties. J. Exp. Med. 136, 907930

15 Trams, E.G., Lauter, C.J., Salem, Jr, N. and Heine, U. (1981) Exfoliation ofmembrane ecto-enzymes in the form of micro-vesicles. Biochim.Biophys. Acta 645, 6370

16 Redman, C.W., Tannetta, D.S., Dragovic, R.A., Gardiner, C., Southcombe,J.H., Collett, G.P. and Sargent, I.L. (2012) Does size matter? Placentaldebris and the pathophysiology of pre-eclampsia. Placenta 33 (Suppl.),S48S54

17 Gercel-Taylor, C., Atay, S., Tullis, R.H., Kesimer, M. and Taylor, D.D. (2012)Nanoparticle analysis of circulating cell-derived vesicles in ovariancancer patients. Anal. Biochem. 428, 4453

18 Thery, C., Ostrowski, M. and Segura, E. (2009) Membrane vesicles asconveyors of immune responses. Nat. Rev. Immunol. 9, 581593

19 Gan, X. and Gould, S.J. (2011) Identication of an inhibitory buddingsignal that blocks the release of HIV particles and exosome/microvesicleproteins. Mol. Biol. Cell 22, 817830

20 Gan, X. and Gould, S.J. (2012) HIV Pol inhibits HIV budding and mediatesthe severe budding defect of Gag-Pol. PLoS ONE 7, e29421

21 Shen, B., Fang, Y., Wu, N. and Gould, S.J. (2011) Biogenesis of theposterior pole is mediated by the exosome/microvesicle protein-sortingpathway. J. Biol. Chem. 286, 4416244176

CThe Authors Journal compilation C2013 Biochemical Society

Microvesiculation and Disease 281

22 Shen, B., Wu, N., Yang, J.M. and Gould, S.J. (2011) Protein targeting toexosomes/microvesicles by plasma membrane anchors. J. Biol. Chem.286, 1438314395

23 Fang, Y., Wu, N., Gan, X., Yan, W., Morrell, J.C. and Gould, S.J. (2007)Higher-order oligomerization targets plasma membrane proteins and HIVgag to exosomes. PLoS Biol. 5, 12671283

24 Balasubramaniam, M. and Freed, E.O. (2011) New insights into HIVassembly and trafcking. Physiology 26, 236251

25 Jouvenet, N., Neil, S.J., Bess, C., Johnson, M.C., Virgen, C.A., Simon, S.M.and Bieniasz, P.D. (2006) Plasma membrane is the site of productiveHIV-1 particle assembly. PLoS Biol. 4, e435

26 Escola, J.M., Kleijmeer, M.J., Stoorvogel, W., Grifth, J.M., Yoshie, O. andGeuze, H.J. (1998) Selective enrichment of tetraspan proteins on theinternal vesicles of multivesicular endosomes and on exosomes secretedby human B-lymphocytes. J. Biol. Chem. 273, 2012120127

27 Charrin, S., le Naour, F., Silvie, O., Milhiet, P.E., Boucheix, C. andRubinstein, E. (2009) Lateral organization of membrane proteins:tetraspanins spin their web. Biochem. J. 420, 133154

28 Fehon, R.G., McClatchey, A.I. and Bretscher, A. (2010) Organizing the cellcortex: the role of ERM proteins. Nat. Rev. Mol. Cell Biol. 11, 276287

29 Hegmans, J.P., Bard, M.P., Hemmes, A., Luider, T.M., Kleijmeer, M.J.,Prins, J.B., Zitvogel, L., Burgers, S.A., Hoogsteden, H.C. and Lambrecht,B.N. (2004) Proteomic analysis of exosomes secreted by humanmesothelioma cells. Am. J. Pathol. 164, 18071815

30 Gatti, J.L., Metayer, S., Belghazi, M., Dacheux, F. and Dacheux, J.L. (2005)Identication, proteomic proling, and origin of ram epididymal uidexosome-like vesicles. Biol. Reprod. 72, 14521465

31 Staubach, S., Razawi, H. and Hanisch, F.G. (2009) Proteomics ofMUC1-containing lipid rafts from plasma membranes and exosomes ofhuman breast carcinoma cells MCF-7. Proteomics 9, 28202835

32 Freed, E.O. and Martin, M.O. (2006) HIVs and their replication. In FieldsVirology (Knipe, D.M. and Howley, P.M., eds), pp. 21072185, LippincottWilliams & WIlkins, Philadelphia

33 Booth, A.M., Fang, Y., Fallon, J.K., Yang, J.M., Hildreth, J.E. and Gould, S.J.(2006) Exosomes and HIV Gag bud from endosome-like domains of the Tcell plasma membrane. J. Cell Biol. 172, 923935

34 Pan, B.T. and Johnstone, R.M. (1983) Fate of the transferrin receptorduring maturation of sheep reticulocytes in vitro: selectiveexternalization of the receptor. Cell 33, 967978

35 Harding, C., Heuser, J. and Stahl, P. (1984) Endocytosis and intracellularprocessing of transferrin and colloidal goldtransferrin in ratreticulocytes: demonstration of a pathway for receptor shedding. Eur. J.Cell Biol. 35, 256263

36 Geminard, C., Nault, F., Johnstone, R.M. and Vidal, M. (2001)Characteristics of the interaction between Hsc70 and the transferrinreceptor in exosomes released during reticulocyte maturation. J. Biol.Chem. 276, 99109916

37 Fevrier, B., Vilette, D., Archer, F., Loew, D., Faigle, W., Vidal, M., Laude, H.and Raposo, G. (2004) Cells release prions in association with exosomes.Proc. Natl. Acad. Sci. U.S.A. 101, 96839688

38 Coleman, B.M., Hanssen, E., Lawson, V.A. and Hill, A.F. (2012)Prion-infected cells regulate the release of exosomes with distinctultrastructural features. FASEB J. 26, 41604173

39 Henne, W.M., Buchkovich, N.J. and Emr, S.D. (2011) The ESCRT pathway.Dev. Cell 21, 7791

40 Hurley, J.H. (2010) The ESCRT complexes. Crit. Rev. Biochem. Mol. Biol.45, 463487

41 Hurley, J.H., Boura, E., Carlson, L.A. and Rozycki, B. (2010) Membranebudding. Cell 143, 875887

42 Welsch, S., Habermann, A., Jager, S., Muller, B., Krijnse-Locker, J. andKrausslich, H.G. (2006) Ultrastructural analysis of ESCRT proteins suggestsa role for endosome-associated tubular-vesicular membranes in ESCRTfunction. Trafc 7, 15511566

43 Taylor, D.D. and Gercel-Taylor, C. (2005) Tumour-derived exosomes andtheir role in cancer-associated T-cell signalling defects. Br. J. Cancer 92,305311

44 Thery, C., Regnault, A., Garin, J., Wolfers, J., Zitvogel, L.,Ricciardi-Castagnoli, P., Raposo, G. and Amigorena, S. (1999) Molecularcharacterization of dendritic cell-derived exosomes: selectiveaccumulation of the heat shock protein hsc73. J. Cell Biol. 147, 599610

45 Tauro, B.J., Greening, D.W., Mathias, R.A., Ji, H., Mathivanan, S., Scott,A.M. and Simpson, R.J. (2012) Comparison of ultracentrifugation, densitygradient separation, and immunoafnity capture methods for isolatinghuman colon cancer cell line LIM1863-derived exosomes. Methods 56,293304

46 Bodon, G., Chassefeyre, R., Pernet-Gallay, K., Martinelli, N., Effantin, G.,Hulsik, D.L., Belly, A., Goldberg, Y., Chatellard-Causse, C., Blot, B. et al.(2011) Charged multivesicular body protein 2B (CHMP2B) of theendosomal sorting complex required for transport-III (ESCRT-III)polymerizes into helical structures deforming the plasma membrane. J.Biol. Chem. 286, 4027640286

47 Nabhan, J.F., Hu, R., Oh, R.S., Cohen, S.N. and Lu, Q. (2012) Formationand release of arrestin domain-containing protein 1-mediatedmicrovesicles (ARMMs) at plasma membrane by recruitment of TSG101protein. Proc. Natl. Acad. Sci. U.S.A. 109, 41464151

48 Baietti, M.F., Zhang, Z., Mortier, E., Melchior, A., Degeest, G., Geeraerts,A., Ivarsson, Y., Depoortere, F., Coomans, C., Vermeiren, E. et al. (2012)SyndecansynteninALIX regulates the biogenesis of exosomes. Nat. CellBiol. 14, 677685

49 Martin-Serrano, J. and Neil, S.J. (2011) Host factors involved in retroviralbudding and release. Nat. Rev. Microbiol. 9, 519531

50 Bieniasz, P.D. (2009) The cell biology of HIV-1 virion genesis. Cell HostMicrobe 5, 550558

51 Wollert, T., Wunder, C., Lippincott-Schwartz, J. and Hurley, J.H. (2009)Membrane scission by the ESCRT-III complex. Nature 458, 172177

52 Trajkovic, K., Hsu, C., Chiantia, S., Rajendran, L., Wenzel, D., Wieland, F.,Schwille, P., Brugger, B. and Simons, M. (2008) Ceramide triggersbudding of exosome vesicles into multivesicular endosomes. Science319, 12441247

53 Sreekumar, P.G., Kannan, R., Kitamura, M., Spee, C., Barron, E., Ryan, S.J.and Hinton, D.R. (2010) B crystallin is apically secreted withinexosomes by polarized human retinal pigment epithelium and providesneuroprotection to adjacent cells. PLoS ONE 5, e12578

54 Gangalum, R.K., Atanasov, I.C., Zhou, Z.H. and Bhat, S.P. (2011)B-crystallin is found in detergent-resistant membrane microdomainsand is secreted via exosomes from human retinal pigment epithelialcells. J. Biol. Chem. 286, 32613269

55 Mathew, A., Bell, A. and Johnstone, R.M. (1995) Hsp-70 is closelyassociated with the transferrin receptor in exosomes from maturingreticulocytes. Biochem. J. 308, 823830

56 Clayton, A., Turkes, A., Navabi, H., Mason, M.D. and Tabi, Z. (2005)Induction of heat shock proteins in B-cell exosomes. J. Cell Sci. 118,36313638

57 Oshima, K., Aoki, N., Kato, T., Kitajima, K. and Matsuda, T. (2002)Secretion of a peripheral membrane protein, MFG-E8, as a complex withmembrane vesicles. Eur. J. Biochem. 269, 12091218

58 Zeelenberg, I.S., Ostrowski, M., Krumeich, S., Bobrie, A., Jancic, C.,Boissonnas, A., Delcayre, A., Le Pecq, J.B., Combadiere, B., Amigorena, S.and Thery, C. (2008) Targeting tumor antigens to secreted membranevesicles in vivo induces efcient antitumor immune responses. CancerRes. 68, 12281235

59 Valadi, H., Ekstrom, K., Bossios, A., Sjostrand, M., Lee, J.J. and Lotvall, J.O.(2007) Exosome-mediated transfer of mRNAs and microRNAs is a novelmechanism of genetic exchange between cells. Nat. Cell Biol. 9,654659

60 Skog, J., Wurdinger, T., van Rijn, S., Meijer, D.H., Gainche, L.,Sena-Esteves, M., Curry, Jr, W.T., Carter, B.S., Krichevsky, A.M. andBreakeeld, X.O. (2008) Glioblastoma microvesicles transport RNA andproteins that promote tumour growth and provide diagnosticbiomarkers. Nat. Cell Biol. 10, 14701476

61 Hong, B.S., Cho, J.H., Kim, H., Choi, E.J., Rho, S., Kim, J., Kim, J.H., Choi,D.S., Kim, Y.K., Hwang, D. and Gho, Y.S. (2009) Colorectal cancercell-derived microvesicles are enriched in cell cycle-related mRNAs thatpromote proliferation of endothelial cells. BMC Genomics 10, 556

62 Alcazar, O., Hawkridge, A.M., Collier, T.S., Cousins, S.W., Bhattacharya,S.K., Muddiman, D.C. and Marin-Castano, M.E. (2009) Proteomicscharacterization of cell membrane blebs in human retinal pigmentepithelium cells. Mol. Cell. Proteomics 8, 22012211

63 Thery, C., Boussac, M., Veron, P., Ricciardi-Castagnoli, P., Raposo, G.,Garin, J. and Amigorena, S. (2001) Proteomic analysis of dendriticcell-derived exosomes: a secreted subcellular compartment distinct fromapoptotic vesicles. J. Immunol. 166, 73097318

64 Pegtel, D.M., van de Garde, M.D. and Middeldorp, J.M. (2011) ViralmiRNAs exploiting the endosomalexosomal pathway for intercellularcross-talk and immune evasion. Biochim. Biophys. Acta 1809, 715721

65 Welton, J.L., Khanna, S., Giles, P.J., Brennan, P., Brewis, I.A., Staffurth, J.,Mason, M.D. and Clayton, A. (2010) Proteomics analysis of bladdercancer exosomes. Mol. Cell. Proteomics 9, 13241338

66 Batista, B.S., Eng, W.S., Pilobello, K.T., Hendricks-Munoz, K.D. and Mahal,L.K. (2011) Identication of a conserved glycan signature formicrovesicles. J. Proteome Res. 10, 46244633

CThe Authors Journal compilation C2013 Biochemical Society

282 Biochemical Society Transactions (2013) Volume 41, part 1

67 Krishnamoorthy, L., Bess, Jr, J.W., Preston, A.B., Nagashima, K. andMahal, L.K. (2009) HIV-1 and microvesicles from T cells share acommon glycome, arguing for a common origin. Nat. Chem. Biol. 5,244250

68 Wubbolts, R., Leckie, R.S., Veenhuizen, P.T., Schwarzmann, G., Mobius,W., Hoernschemeyer, J., Slot, J.W., Geuze, H.J. and Stoorvogel, W. (2003)Proteomic and biochemical analyses of human B cell-derived exosomes:potential implications for their function and multivesicular bodyformation. J. Biol. Chem. 278, 1096310972

69 Willem, J., ter Beest, M., Scherphof, G. and Hoekstra, D. (1990) Anon-exchangeable uorescent phospholipid analog as a membranetrafc marker of the endocytic pathway. Eur. J. Cell Biol. 53, 173184

70 Stachowiak, J.C., Schmid, E.M., Ryan, C.J., Ann, H.S., Sasaki, D.Y., Sherman,M.B., Geissler, P.L., Fletcher, D.A. and Hayden, C.C. (2012) Membranebending by proteinprotein crowding. Nat. Cell Biol. 14, 944949

Received 26 October 2012doi:10.1042/BST20120275

CThe Authors Journal compilation C2013 Biochemical Society