main topic

Wien Med Wochenschr (2016) 166:236–241DOI 10.1007/s10354-016-0453-2

Endoplasmosis and exoplasmosis: the evolutionaryprinciples underlying endocytosis, exocytosis, and vesiculartransport

Johannes A. Schmid

Published online: 11 May 2016© The Author(s) 2016. This article is available at SpringerLink with Open Access

Summary Eukaryotic cells are characterized by amulticompartmental structure with a variety of or-ganelles. Vesicular transport between these com-partments requires membrane fusion events. Basedon a membrane topology view, we conclude thatthere are two basic mechanisms of membrane fusion,namely where the membranes first come in contactwith the cis-side (the plasmatic phase of the lipidbilayer) or with the trans-side (the extra-plasmaticface). We propose to designate trans-membrane fu-sion processes as “endoplasmosis” as they lead touptake of a compartment into the plasmatic phase.Vice versa we suggest the term “exoplasmosis” (asalready suggested in a 1964 publication) for cis-mem-brane fusion events, where the interior of a vesicle isreleased to an extraplasmatic environment (the extra-cellular space or the lumen of a compartment). Thisconcept is supported by the fact that all cis- and alltrans-membrane fusions, respectively, exhibit notice-able similarities implying that they evolved from twofunctionally different mechanisms.

Keywords Lipid Bilayers · Membrane Fusion · Eukary-otic Cells · Extracellular Space · Membranes · Endocy-tosis · Exocytosis

J. A. Schmid (�)Center for Physiology and Pharmacology, Department ofVascular Biology and Thrombosis Research, MedicalUniversity Vienna, Schwarzspanierstraße 17, 1090 Vienna,Austriae-mail: Johannes.Schmid@meduniwien.ac.at

Endoplasmose und Exoplasmose: Entwicklungs-prinzipien der Endo- und Exozytose sowie des ve-sikulären Transports

Zusammenfassung Eukaryotische Zellen besitzeneine Multi-Kompartiment-Struktur mit unterschiedli-chen Organellen. Vesikulärer Transport zwischen denKompartimenten erfordert Membranfusionsprozes-se. Bei Betrachtung aus Sicht der Membrantopologiekann man schließen, dass es 2 grundsätzliche Mecha-nismen der Membranfusion gibt: einerseits solche,bei denen die Membranen zuerst mit der cis-Seite(der plasmatischen Seite der Membrandoppelschicht)in Kontakt treten, und andererseits solche, bei denensich die trans-Seiten zuerst berühren (die extraplas-matischen Seiten). Der Autor schlägt die Bezeichnung„Endoplasmose“ für alle trans-Membranfusionen vor,da sie zu einer Aufnahme eines Kompartiments indie plasmatische Phase führen, und vice versa dieBezeichnung „Exoplasmose“ (wie bereits 1964 ineiner Publikation vorgeschlagen) für alle cis-Mem-branfusionen, wo das Innere eines Vesikels in einenextraplasmatischen Raum entladen wird (den Extra-zellulärraum oder das Lumen eines Kompartiments).Dieses Konzept wird durch die Tatsache unterstützt,dass alle cis- bzw. alle trans-Membranfusionen be-merkenswerte Ähnlichkeiten zeigen, was darauf hin-deutet, dass sie aus 2 funktionell unterschiedlichenMechanismen entstanden sind.

Schlüsselwörter Lipiddoppelschicht · Membranfu-sion · Eukaryotische Zellen · Extrazellulärraum ·Membrane · Endozytose · Exozytose

List of abbreviationsARF ADP-ribosylation factorCOP coat protein complex

236 Endoplasmosis and exoplasmosis, the 2 basic membrane fusions K

main topic

Fig. 1 General schemeofthecompartmental struc-tureof eukaryotic cells andvesicular transport pro-cesses. Examplesof differ-ent intracellular organellesareoutlined. Plasmaticdomainsare in red andplas-maticmembranesurfacesin red thick lines, nonplas-maticspheresare inblueandextraplasmaticmembranesurfaces inblack lines. En-doplasmosis (thebuddingofvesicles into aplasmatic en-vironment, due to fusionofextraplasmaticmembranesurfaces) is indicatedbyred arrows; exoplasmosis(releaseof vesicle contentsinto nonplasmatic domainsafter cis-membrane fusion)is outlinedbyblue arrows

Fig. 2 Generalmechanismof trans- andcis-membrane fusionevents. Theextraplasmatic sidesofmembranes (trans-sides) areshown inblue, theplasmatic faces in red (cis-sides). Theupper partshows thebuddingof a vesicle into theplasmatic phase (endo-plasmosis), characterizedbytheassemblyofcoatcomplexesandthesegregationoffluidphasecargoor receptor–ligandcomplexesbyafissionprocess,which isactuallydrivenbyatrans-membrane fusionevent. The lower partshowsthe fusionofavesiclewitha tar-getmembrane (exoplasmosis). RabGTPases, their effectorproteins,SNAREproteinsandaccessory factors (suchasNSForSNAPs)are involved in thesecis-membrane fusionprocesses

ER endoplasmic reticulumGTPase guanosine triphosphataseNSF N-ethylmaleimide sensitive factorSNAP soluble NSF attachment proteinSNARE SNAP receptor

Introduction

While Darwin developed his theory of evolution forthe macrobiological world, it is becoming increasinglyevident that the general concept of evolution also ap-plies to the cellular and subcellular world. In gen-eral, a central characteristic of all known organisms is

K Endoplasmosis and exoplasmosis, the 2 basic membrane fusions 237

main topic

Fig. 3 Structural similaritiesbetweendifferent coat com-plexes: Componentsof theclathrin,COPI andCOPII coatsare shown ina structural schematic view. Coat structuresareshown ingreen; adaptor complexes inblueandGTPases in red.(Adapted from [6, 16])

their compartmentalisation, their delimitation fromthe surrounding environment. This compartmentali-sation is a prerequisite for the generation and mainte-nance of electrochemical gradients that are essentialfor the energy processes of life. In contrast to prokary-otes, which consist in principle of only one “reac-tion compartment”, eukaryotes are characterised bya multicompartmental structure of various organelleswithin one common compartment. One of the cru-cial questions in that respect is how the intracellu-lar organelles of eukaryotes evolved. The cytoplasmicmembrane, which functions as the ultimate borderof a cell, separates a cytoplasmic from an extra-cy-toplasmic phase, or in other terms a plasmatic froma nonplasmatic sphere. Due to the multicompart-mental organisation of eukaryotes, the cell contains

Fig. 4 Structural conservationbetweendistinct vesicular coat complexesoccurring at thecell surface, ERorGolgimembranes.(Reprintedwithpermission fromMacmillanPublishersLtd [27])

internal nonplasmatic domains, namely all the com-partments that are enclosed by a single lipid bilayer,like the endoplasmic reticulum (ER), the Golgi, endo-somes or lysosomes.

Organelles that are enclosed by two lipid bilayerslike chloroplasts or mitochondria exhibit a nonplas-matic intermembrane space, but their internal lumenis defined as plasmatic sphere comparable to the cy-toplasm ([2, 3]; Fig. 1). This view is in agreement withthe commonly accepted hypothesis that they weretaken up from the extracellular environment by an-cient pre-eukaryotes as endosymbionts [4–7]. Anotherorganelle enclosed by two membranes is the nucleus,which is thought to have evolved from the ER [8] andwhich contains a plasmatic lumen that is linked to thecytosol via nuclear pore complexes that are permeablefor ions and small proteins.

An evolutionary perspective of vesiculartransport

An important question in the evolution of eukary-otic cells is how the compartments arose, which areenclosed by a single membrane. The common expla-nation is that they were originally formed by invagina-tion and internalization of the cytoplasmic membrane[9]. Interestingly, these first endomembranes mighthave had rather secretory functions than featuresof current endocytic compartments given that cer-tain GTPases of the secretory pathway seem to haveevolved before those of the endocytic pathway [10].Regardless of how these first endomembranes formedand whether they were more endocytic or more secre-tory in nature, they can be regarded as “internalizedextracellular space”. From this point of view, it is

238 Endoplasmosis and exoplasmosis, the 2 basic membrane fusions K

main topic

Fig. 5 AlignmentofhumanRabGTPasesandallocationto theirmain intracellularorganelles: homologousaminoacidsareshownindifferent shadesof red. Themain localizationsof theGTPaseswereobtained from reference [29]

a provoking but intriguing aspect to regard all non-plasmatic compartments within the cell as spheres,which are in some way functionally outside of the cell.The ER for instance can be regarded as first outstationof the cell in the course of secretion, which might haveevolved as an internalised and specialised part of thecytoplasmic membrane [9]. Molecules that are co- orposttranslationally transported into the lumen of theER are therefore equivalent to proteins secreted fromprokaryotes. Strikingly, the ER and other organellesare not synthesised de novo in developing cells, buthanded over from parental cells (“omnis membrana emembrana” [9]), supporting the notion that the com-partmental organisation of eukaryotes is not encodedin the genome, but directly inherited. This notionalso suggests that the eukaryotic ancestor cells musthave had biological membranes [11]. The structureof eukaryotic cells with plasmatic and nonplasmaticspheres implies that the lipid bilayers, which form themembrane of organelles or the boundary between thecytoplasm and the extracellular environment, exhibita plasmatic (or cis-) and an extraplasmatic (or trans-)face [3]. Endocytosis and exocytosis, as well as themembrane traffic within the cell require fusion andbudding events that are all based on membrane fu-sion, where either the plasmatic or the extraplasmaticfaces of two lipid bilayers first come in contact. Theseprocesses can be designated as cis- or trans-mem-brane fusion events, respectively. As a consequenceof this view, we propose to designate all trans-mem-brane fusion processes (where the extraplasmaticmembrane surfaces first come in contact) with theterm “endoplasmosis”, meaning an uptake of a vesicleor membrane compartment into a plasmatic phase.Vice versa, all cis-membrane fusion processes canbe termed as “exoplasmosis” meaning that a com-partment fuses with a membrane with the two cis-sides coming into contact first, resulting in the re-lease of the cargo into a nonplasmatic phase (Figs. 1and 2). It has to be stated that the term “exoplas-mosis” was used in an early publication on leukocytedegranulation in 1964 [1], but has not been appliedsystematically since then. Using these terms has theadvantage of emphasizing conceptual similarities be-

tween different processes of endocytosis, exocytosisand vesicular transport as they stress the two underly-ing basic principles from a cell topology perspective.Endoplasmosis then subsumes processes such as thefirst step of endocytosis, or the budding and forma-tion of carrier vesicles, for instance from ER, Golgistacks or endosomes. Exoplasmosis would be the su-perordinate term for fusion of secretory vesicles withthe plasma membrane, the fusion of carrier vesicleswith target membranes and all other cis-membranefusion events, where vesicles functionally leave theplasmatic sphere.

This concept is in line with the fact that thereare many similarities between different forms of en-doplasmosis or exoplasmosis, meaning trans- andcis-membrane fusion events, respectively. A com-mon feature of endoplasmosis is the requirement forcoat proteins at the plasmatic face of the membrane(Fig. 2). For endocytic internalisation, these coats canconsist of clathrin [12] or caveolin [13, 14]. Further-more, clathrin-coated areas can also be detected onthe trans-Golgi network [12] and under certain cir-cumstances also on sorting endosomes [15]. Along theexocytic route, the budding and formation of carriervesicles from ER en route to the intermediate com-partment requires COP-II (coat protein complex II)containing coats [16]. Finally, recycling of vesicles tothe ER, as well as transport within the Golgi is depen-dent on COP-I [17–20]. Some of these coat proteinsthat are characteristic for the exocytic pathway (COP-Iand ARF [ADP-ribosylation factor]) were also detectedon endosomal membranes [21].

Thus, all these examples of trans-membrane fu-sion events, which can be subsumed as endoplasmo-sis events, are characterised by the formation of coatstructures at the cytosolic face of the membrane [22,23], which seems to be important for generating a cur-vature of the membrane preceding the budding pro-cess [24–26]. This view of a general principle under-lying all these trans-membrane fusions events is sup-ported by the fact that coat subunits from functionallydistinct coats such as clathrin, COPI and COPII coatsexhibit noticeable structural similarities with β-pro-peller and α-solenoid elements in a specific common

K Endoplasmosis and exoplasmosis, the 2 basic membrane fusions 239

main topic

Fig. 6 Phylogenetic analysisofQa-typeSNAREproteins:Up-per part: Humansyntaxins specific for variousorganelles arecompared; lower part: the respectiveQaproteins fromS. cere-visiae are shown

arrangement (Fig. 3; reviewed in [6]). Furthermore,the proteins involved bind to each other in a way thatleads to a very similar molecular architecture of thecoat complex on the membrane as illustrated in Fig. 4(reviewed in [27]) pointing at a common evolutionaryorigin of trans-membrane fusion events.

On the other hand, cis-membrane fusion processesfrom clearly different organelles also show strikingmechanistic similarities as well as sequence homolo-gies of the proteins that are involved therein. Thisholds true for fusion of carrier vesicles with targetmembranes in Golgi transport, fusion of secretorygranules with the cytoplasmic membrane or endo-some fusion, which are all dependent on certain fu-sion factors such as NSF (N-ethylmaleimide sensitivefactor), SNAPs (soluble NSF attachment proteins),SNAREs (SNAP receptors) and Rab family GTPases(guanosine triphosphatases) (Fig. 2; [28–39]). Theevolutionary coherence of all these distinct mem-brane fusion processes is well demonstrated by thestrong conservation and high degree of homology ofRab GTPases, which are important regulators of cis-membrane fusions at completely different organellesand locations within the cell (Fig. 5). Significanthomologies are also found for other fusion factorssuch as SNARE proteins—and again their functionalrole seems to be the same for all the different cis-membrane fusion processes at distinct organelleswithin the cell. Phylogenetic analysis of the Qa fam-ily of human SNARE proteins (syntaxins) for variousendomembranes reveals clear links, with syntaxinslocated at endosomes being closely related to syn-taxins of the trans-Golgi network and slightly moredistant to those of secretory vesicles. A very similarphylogenetic tree is observed for the Qa proteins ofyeast (Fig. 6). However, in the latter case the proteinsare more related to each other suggesting that higherorder organisms developed a higher diversity of thesefusion factors. While these sequence analyses clearlypoint at a common evolutionary origin of the differentcis-membrane fusion processes, it is also clear that incurrent organisms and cells a high level of specificityis observed for cis-membrane fusion events of dis-tinct intracellular compartments. This specificity is

crucial for maintaining the functional integrity of theorganelles, an ordered progress of vesicular transportand the “identity” of the membrane compartments.The available data indicate a model, in which all cis-membrane fusions go back to an ancient process of“exoplasmosis”, which was then altered by evolution-ary diversification forming specific organelles andwell-controlled fusion processes.

Conclusion

Based on the similarities between different membranefusion processes, we suggest that there are two gener-ally distinct mechanisms of membrane fusion, namelyendoplasmosis (trans-membrane fusion) and exoplas-mosis (cis-membrane fusion), which are the mecha-nistic ancestors of all the cellular membrane fusionevents. These two basic principles of membrane fu-sion might be the origin for the evolution of the eu-karyotic endomembrane system, which developed itscomplexity by diversification of the components, fi-nally defining the identities of intracellular compart-ments and regulating the membrane traffic betweenthem [40–42].

Acknowledgements Open access funding provided byMedi-cal University of Vienna.

Compliance with ethical guidelines

Conflict of interest J.A. Schmid states that there are no com-peting interests.

Ethical standards The accompanying manuscript does notinclude studies on humans or animals.

Open Access This article is distributed under the terms ofthe Creative Commons Attribution 4.0 International License(http://creativecommons.org/licenses/by/4.0/), which per-mits unrestricted use, distribution, and reproduction in anymedium, provided you give appropriate credit to the origi-nal author(s) and the source, provide a link to the CreativeCommons license, and indicate if changes were made.

References

1. Zucker-FranklinD,Hirsch JG.ElectronMicroscopeStudiesOn The Degranulation Of Rabbit Peritoneal LeukocytesDuringPhagocytosis. JExpMed. 1964;120:569–576.

2. BlobelG. Intracellular protein topogenesis. ProcNatl AcadSciUSA.1980;77(3):1496–1500.

3. Duve C de, Hardy NO. A Guided Tour Of The LivingCell, vol. 1: Scientific American Library. Quart Rev Biol.1985;60(4):497.

4. Cavalier-Smith T. The phagotrophic origin of eukaryotesand phylogenetic classification of Protozoa. Int J Syst EvolMicrobiol. 2002;52(2):297–354.

5. GrayMW,DoolittleWF.Has the endosymbiont hypothesisbeenproven?MicrobiolRev. 1982;46(1):1–42.

6. GurkanC,KoulovAV,BalchWE.Anevolutionaryperspectiveon eukaryotic membrane trafficking. Adv Exp Med Biol.2007;607:73–83.

240 Endoplasmosis and exoplasmosis, the 2 basic membrane fusions K

main topic

7. Sagan L. On the origin of mitosing cells. J Theor Biol.1967;14(3):255–274.

8. Jekely G. Glimpsing over the event horizon: evolution ofnuclearporesandenvelope. CellCycle. 2005;4(2):297–299.

9. DuveCde. Theoriginof eukaryotes: a reappraisal. NatRevGenet. 2007;8(5):395–403.

10. JékelyG.SmallGTPasesandtheevolutionof theeukaryoticcell. Bioessays. 2003;25(11):1129–1138.

11. Jekely G. Did the last common ancestor have a biologicalmembrane? BiolDirect. 2006;1(1):35.

12. Pearse BM, RobinsonMS. Clathrin, adaptors, and sorting.AnnuRevCellBiol. 1990;6:151–171.

13. RothbergKG,Heuser JE,DonzellWC, Ying YS, Glenney JR,Anderson RG. Caveolin, a protein component of caveolaemembranecoats. Cell. 1992;68(4):673–682.

14. Lebbink MN, Jiménez N, Vocking K, Hekking LH, VerkleijAJ, Post JA. Spiral Coating of the Endothelial CaveolarMembranes as Revealed by Electron Tomography andTemplateMatching. Traffic. 2010;11(1):138–150.

15. Stoorvogel W, Oorschot V, Geuze HJ. A novel class ofclathrin-coated vesicles budding from endosomes. J CellBiol. 1996;132(1-2):21–33.

16. GurkanC, StaggSM,LaPointeP, BalchWE.TheCOPII cage:unifying principles of vesicle coat assembly. Nat RevMolCellBiol. 2006;7(10):727–738.

17. AridorM, BalchWE. Principles of selective transport: coatcomplexesholdthekey. TrendsCellBiol. 1996;6(8):315–320.

18. Bethune J, Wieland F, Moelleken J. COPI-mediated trans-port. JMembrBiol. 2006;211(2):65–79.

19. Harter C. COP-coated vesicles in intracellular proteintransport. FEBSLett. 1995;369(1):89–92.

20. Wieland F, Hartert C. Mechanisms of vesicle formation:Insights from the COP system. Curr Opin Cell Biol.1999;11(4):440–446.

21. Whitney JA, Gomez M, Sheff D, Kreis TE, Mellman I.Cytoplasmiccoatproteins involved inendosomefunction.Cell. 1995;83(5):703–713.

22. Faini M, Beck R, Wieland FT, Briggs JAG. Vesicle coats:structure, function, and general principles of assembly.TrendsCellBiol. 2013;23(6):279–288.

23. Jackson LP, Kümmel D, Reinisch KM, Owen DJ. Struc-tures and mechanisms of vesicle coat components andmultisubunit tethering complexes. Curr Opin Cell Biol.2012;24(4):475–483.

24. Jackson LP. Structure and mechanism of COPI vesiclebiogenesis. CurrOpinCellBiol. 2014;29:67–73.

25. Derganc J, Antonny B, Copic A. Membrane bending:the power of protein imbalance. Trends Biochem Sci.2013;38(11):576–584.

26. Stachowiak JC, Brodsky FM,Miller EA. A cost-benefit anal-ysis of the physical mechanisms of membrane curvature.NatCellBiol. 2013;15(9):1019–1027.

27. HarrisonSC,KirchhausenT. Structural biology: Conserva-tioninvesiclecoats. Nature. 2010;466(7310):1048–1049.

28. GurkanC, LappH,AloryC, SuAI,Hogenesch JB, BalchWE.Large-Scale Profiling of RabGTPase Trafficking Networks:TheMembrome.MolBiolCell. 2005;16(8):3847–3864.

29. StenmarkH.RabGTPases as coordinators of vesicle traffic.NatRevMolCellBiol. 2009;10(8):513–525.

30. Denesvre C, Malhotra V. Membrane fusion in organellebiogenesis. CurrOpinCellBiol. 1996;8(4):519–523.

31. Kienle N, Kloepper TH, Fasshauer D. Differences in theSNAREevolutionof fungiandmetazoa.BiochemSocTrans.2009;37(4):787–791.

32. Rossi V, Picco R, Vacca M, D’Esposito M, D’Urso M, GalliT, Filippini F. VAMP subfamilies identified by specificR-SNAREmotifs. BiolCell. 2004;96(4):251–256.

33. RothmanJE.Mechanismsofintracellularproteintransport.Nature. 1994;372(6501):55–63.

34. RothmanJE.Theproteinmachineryofvesiclebuddingandfusion. ProteinSci. 1996;5(2):185–194.

35. SudhofTC,Rothman JE.Membrane fusion: grapplingwithSNAREandSMproteins. Science. 2009;323(5913):474–477.

36. McNew JA, Sondermann H, Lee T, Stern M, Brandizzi F.GTP-dependentmembranefusion. AnnuRevCellDevBiol.2013;29:529–550.

37. Rizo J, SüdhofTC.Themembrane fusionenigma: SNAREs,Sec1/Munc18 proteins, and their accomplices--guilty ascharged? AnnuRevCellDevBiol. 2012;28:279–308.

38. Risselada HJ, Grubmüller H. How SNARE moleculesmediatemembrane fusion: recent insights frommolecularsimulations. CurrOpinStructBiol. 2012;22(2):187–196.

39. JenaBP.Roleof SNAREs inmembrane fusion. AdvExpMedBiol. 2011;713:13–32.

40. DacksJB,FieldMC.Evolutionof theeukaryoticmembrane-trafficking system: origin, tempo and mode. J Cell Sci.2007;120(17):2977–2985.

41. Dacks JB,PedenAA,FieldMC.Evolutionofspecificity in theeukaryoticendomembranesystem. Int JBiochemCellBiol.2009;41(2):330–340.

42. Dacks JB, Poon PP, Field MC. Phylogeny of endocyticcomponents yields insight into the process of nonen-dosymbiotic organelle evolution. ProcNatl AcadSciUSA.2008;105(2):588–593.

K Endoplasmosis and exoplasmosis, the 2 basic membrane fusions 241