Secretory membrane traffic between the ER and the Golgi isdependent on the function of two coat protein complexes,COPI and COPII. Genetic data from yeast has shown that thecomponents of the COPII coat are required for proteintransport between the ER and Golgi apparatus (Kaiser andSchekman, 1990; Novick et al., 1980). The in vitroreconstitution of ER-to-Golgi transport with purified COPIIcomponents lead to the hypothesis that COPII vesiclesmediate vectorial ER-to-Golgi transport (Barlowe et al., 1994;Schekman and Orci, 1996) because they can directly delivertheir cargo content to Golgi membranes (Barlowe et al., 1994;Bednarek et al., 1995). In addition to its role in promotingvesicle formation, COPII is implicated in the selection ofcargo into transport vesicles (Aridor et al., 1998; Kuehn et al.,1998; Pagano et al., 1999; Roberg et al., 1999). In mammaliancells, ER-to-Golgi transport is believed to involve COPII-coated vesicles and tubules, as well as COPI-coated tubulo-vesicular transport complexes (TCs; Scales et al., 1997;Presley et al., 1997). COPI function has been shown to berequired for the movement of TCs to the Golgi apparatus, as
well as for the segregation of anterograde and retrograde cargowithin these TCs (Shima et al., 1999). However, the dynamicrelationship between COPII- and COPI-coated TCs in ER-to-Golgi transport has remained unclear. Are COPI-coated TCsalso COPII-coated? Do COPII-coated vesicles mediatetransport to the Golgi directly, or indirectly into COPI-coatedtransport complexes, or both?
To address this question, we tagged SEC24p with spectralvariants of green fluorescent protein and visualised thedynamics of COPII labelling in living mammalian cells. Weshow that COPII accumulates on distinct sites (0.5 to 1 µmin size) that are associated directly with the ER membraneand are relatively immobile compared to COPI TCs. We alsoshow that cargo molecules, visualised using GFP-taggedversions of ts-O45-G (a temperature sensitive mutant of thevesicular stomatitis glycoprotein; Scales et al., 1997), emergefrom these COPII sites and move, independently of COPII,to the Golgi apparatus. These data are consistent with a modelin which COPII mediates selection of cargo at theendoplasmic reticulum, while COPI is involved in formationof ER-to-Golgi transport complexes and subsequent cargosorting.
Transport of proteins between the endoplasmic reticulumand Golgi apparatus is mediated by two distinct membranecoat complexes, COPI and COPII. Genetic, biochemicaland morphological data have accumulated into a modelwhich suggests a sequential mode of action with COPIImediating the selection of cargo and formation of transportvesicles at the ER membrane for ER-to-Golgi transportand COPI mediating recycling of the transport machineryfrom post-ER membranes.
To test this transport model directly in vivo, and to studythe precise temporal sequence of COPI and COPII actionin ER-to-Golgi transport, we have used time lapsemicroscopy of living cells to visualise simultaneously thedynamics of COPII and COPI, as well as COPII and GFPtagged secretory markers in living cells. The majority ofCOPII labelling appears tightly associated with ERmembranes that move only within a limited area (less than
2 µm). Secretory cargo segregates from these sites and isthen transported to the Golgi apparatus without anyapparent association with COPII. COPI-coated transportcomplexes are seen to form adjacent to the COPII sites onthe ER before segregating and moving directionallytowards the Golgi apparatus. COPII is not present on thesetransport complexes and remains associated with the ER.These data demonstrate for the first time directly in vivothat ER-to-Golgi transport is organised in two stepscharacterised by a sequential mode of action of COPII andCOPI.
Movies available on-line:http://www.biologists.com/JCS/movies/jcs1368.html
COPI-coated ER-to-Golgi transport complexes segregate from COPII in close
proximity to ER exit sites
David J. Stephens 1, Nathalie Lin-Marq 2, Alessandra Pagano 2, Rainer Pepperkok 1 and Jean-Pierre Paccaud 2,*1Department of Cell Biology and Biophysics, EMBL Heidelberg, Meyerhofstrasse 1, D-69117 Heidelberg, Germany2Department of Morphology, University of Geneva School of Medicine, 1, rue Michel Servet, 1211 Geneva 4, Switzerland*Author for correspondence (e-mail: [email protected])
Generation of expression constructsThe GFP- and YFP-SEC24p were constructed by cloning thehSEC24D cDNA (from pBSKSII-hSEC24D cDNA; Nagase et al.,1998) in frame in either the pEGFP-C1 or pEYFP-C1 vectors,respectively (Clontech, Heidelberg). Throughout this manuscript GFP,CFP and YFP refer to the enhanced versions of green, cyan and yellowspectral variants of green fluorescent protein (Clontech, Heidelberg).The integrity of each construct was confirmed by DNA sequencing.Microinjection was performed as described (Pepperkok et al., 1993).
Characterisation of GFP-SEC24p and time lapsemicroscopyVero cells (ATCC CCL81) were grown as described (Pepperkok et al.,1993) and were plated on to live cell dishes (MatTek Corp., MA,USA) or sterile glass coverslips 24 hours prior to microinjection. Cellswere transfected with plasmids encoding hSEC24D using Fugene(Roche, Mannheim) according to the manufacturer’s instructions.After 48 hours, cells were fixed with 3.5% paraformaldehyde,permeabilised with 0.1% Triton X-100 in PBS and processed forimmunofluorescence using an anti-sec13p polyclonal antibody (Tanget al., 1997).
Co-immunoprecipitation of GFP-SEC24Dp was performed ondetergent extracts (1% CHAPS, 20 mM Hepes, pH 7.2, 150 mM NaCl,1 mM PMSF, with the addition of a cocktail of protease inhibitors;Complete™, Roche, Mannheim) of transiently transfected cells usingeither the anti-GFP mAb B34 (Eurogentech, Switzerland) or anti-sec24p. Briefly, 100 µg of cell extract was immunoprecipitated with0.5 µg of antibody, the precipitate collected on Protein-G Sepharose(Amersham-Pharmacia Biotech, Freiburg, Germany), washed andseparated by SDS-PAGE. After transfer to nitrocellulose, the blot wasprobed with anti-SEC23A antibodies.
Transport of ts-045-G to the plasma membrane was measured aspreviously described (Scales et al., 1997). All imaging was performedusing the equipment contained in the Advanced Light MicroscopyFacility at EMBL, Heidelberg. Briefly, cells were imaged using atemperature controlled Olympus/TILL Photonics time lapsemicroscope (Olympus IX70 microscope with TILL Imago CCDcamera and Polychrome II multiwavelength illumination system allcontrolled by TILL visION software version 3.3) at 37°C inprewarmed MEM lacking Phenol Red supplemented with 30 mMHEPES and 100 µg.ml−1 cycloheximide. Custom filter sets (Chroma,VT, USA) were used to separate CFP/YFP or GFP/Cy3 emissionspectra. GFP- or YFP-SEC24Dp was expressed for 16 hoursfollowing microinjection into cells. pCFP-ts-045-G was similarlyexpressed but was incubated at 39.5°C during the expression period;COPI was visualised by microinjection of Cy3-labelled monoclonal
antibody, CM1A10, as described (Shima et al., 1999). The ERnetwork was labelled by microinjection and subsequent expression ofpCFP-ER, a commercially available marker of the ER network(Clontech, Heidelberg). pCFP-ER encodes a fusion protein consistingof enhanced cyan fluorescent protein (CFP); the ER targetingsequence of calreticulin cloned at the 5′ end; and the sequenceencoding the ER retrieval sequence, KDEL, cloned at the 3′ end. Totalcycle times for sequential acquisition of the two channels was 1second.
Confocal images were taken using a Zeiss LSM510 confocal laserscanning microscope. For photobleaching experiments, the designated
D. J. Stephens and others
Fig. 1. (A) Immunolocalisation of endogenous SEC23Ap or SEC13pin GFP-SEC24Dp transfected cells by confocal microscopy. Threecolocalising structures are marked with arrowheads. Bar, 5 µm.(B) Coimmunoprecipitation of GFP-SEC24Dp with othercomponents of the COPII complex. GFP-SEC24Dp wasimmunoprecipitated from detergent extracts of transiently transfectedcells using the antibodies indicated below. The immunoprecipitatewas subsequently immunoblotted with antibodies directed againstSEC23A. T=total (no immunoprecipitation), 1= mock transfected(anti-GFP immunoprecipitated), 2= GFP-SEC24Dp transfected (anti-GFP immunoprecipitated), 3= GFP-SEC24Dp transfected (anti-Sec24C immunoprecipitated as a positive control). (C) Expression ofGFP-SEC24Dp has no effect on the transport kinetics of ts-045-G-GFP. Cells transfected with GFP-SEC24Dp were subsequentlyinfected with ts-045-G. After incubation for the indicated times at32°C, cells were fixed with paraformaldehyde and surface ts-045-Gquantitated by immunofluorescence microscopy.
2179In vivo dynamics of COPII
area of the cell was bleached for 3 seconds at 100% laser power (488nm line) and recovery of fluorescence into the bleached area wasmonitored at 7.3 second intervals by scanning at 0.2% laser power.QuickTime movies were prepared using IPlab version 3.2 and AppleMovieplayer 2.1, still images were assembled using Adobe Photoshopversion 5.0.
Data analysisQuantification of movement was performed with NIH Image 1.62using a macro developed by Jens Rietdorf (Advanced LightMicroscopy Facility, EMBL Heidelberg). The x-y-co-ordinates of 30to 85 clusters per cell were recorded over 30 to 120 successive images,taken at less than 2 seconds intervals. 395 COPI and 396 COPIIstructures were quantified on 8 different cells. The distance travelledby a given cluster was defined as the distance between the location ofa cluster at time t=0 and time t=n. ‘Moving structures’ were arbitrarilydefined as those that travelled over more than 2 µm from their startingpoint during an entire observation period, or that disappeared duringthis period by leaving the plane of focus.
The fluorescence intensity of individual GFP-SEC24Dp labelledstructures during fluorescence recovery after photobleachingexperiments was quantitated using Zeiss LSM510 software. An areaapproximately double the size of the particle of interest was selectedsuch that the chosen structure did not move outside of this area duringthe experiment.
Online supplemental materialQuickTime movies of the data in Figs 2A,B, 3A, 4A,B, 5B,C, 6A,Bare available on the internet at http://www.biologists.com/JCS/movies/jcs1368.html
RESULTS
Vero cells were microinjected with a plasmid encoding GFP-SEC24Dp, incubated at 37°C for 24 hours and the distributionof GFP-SEC24Dp determined. Immunostaining of fixed cellswith an anti-Sec13p antibody (Tang et al., 1997) showed co-localisation of endogenous sec13p or endogenous sec23p withGFP-SEC24Dp in more than 90% of GFP-SEC24Dp labelledstructures (Fig. 1A). Immunoprecipitation with anti-GFPfollowed by western blotting with anti-sec23p showed thatGFP-SEC24Dp was integrated into the COPII complex (Fig.1B). Furthermore, expression of GFP-SEC24Dp had no effecton the kinetics of delivery of the secretory marker ts-O45-G tothe cell surface (Fig. 1C). Identical results were obtained withYFP-SEC24Dp. Together these data suggest that GFP- andYFP-tagged SEC24Dp behave identically to endogenousSEC24Dp.
The dynamics of GFP-SEC24Dp labelled structures wereanalysed in living cells by time lapse microscopy. The majorityof the GFP-SEC24Dp labelled structures underwentmovements within a limited area (less than 2 µm in diameter)
during time periods of up to 15 minutes. Only occasionallywere longer range (>2 µm) movements observed (Fig. 2A;Table 1) which were randomly directed and appeared to bemicrotubule-dependent, since they were not observed aftertreatment of cells with the microtubule disrupting agentnocodazole (Fig. 2B; Table 1). Simultaneous visualisation ofYFP-SEC24Dp and a CFP-tagged ER marker (see Materialsand Methods) in living cells revealed that the YFP-SEC24Dplabelled structures remained intimately associated with the ERmembrane for an extended period of time (more than oneminute of continuous observation, Fig. 3). It should be notedhowever that COPII structures were never seen to dissociatefrom, nor to be located separate from the ER membrane. Mostoften the non-directed, short range movements coincided with
Table 1. Quantitation of COPII (GFP-SEC24p) and COPI(Cy3-CM1A10) movement in Vero cells
COPIICOPII +5 µM nocodazole COPI
Mean (std dev.) Mean (std dev.) Mean (std dev.)
Mean range (µm) 1.7 (1.1) 0.9 (0.52) 3.6 (1.0)Mean speed (µm s−1) 1.0 (0.6) 0.5 (0.0) 1.7 (0.4)Average total 14.7 (5.6) 6.1 (1.5) 17.5 (7.0)displacement (µm)
Fig. 2. (A) Time lapse imaging of GFP-SEC24Dp expressed in Verocells. The image shown is the first frame of the associatedQuickTime movie. Overlayed on to this image in red, yellow, blueand cyan, are the trajectories followed by five randomly chosenstructures during 10 minutes of imaging. Bar, 10 µm. (B) Time lapseimaging of GFP-SEC24Dp expressed in Vero cells following 30minutes incubation in the presence of 5 µM nocodazole. The imageshown is the first frame of the associated QuickTime movie.Overlayed on to this image in red, yellow, blue and cyan, are thetrajectories followed by six randomly chosen structures during 10minutes of imaging. Bar, 5 µm.
the movements of the ER network itself (Fig. 3). Theseobservations are consistent with the hypothesis that these stableCOPII structures constitute the ER exit sites into whichsecretory cargo is segregated from ER residents by COPII.
A direct prediction from the fact that COPII drives vesiclesbudding at the ER is that this coat complex should turn overat these membranes with dynamics comparable to thoseexpected for vesicles to form and bud off the membranes. Toaddress this question we analysed the membrane association/dissociation of COPII by recovery after photobleaching. GFP-SEC24Dp specific fluorescence rapidly recovered atphotobleached ER exit sites with an average half-time of 38seconds (Fig. 4A and B) consistent with the idea that COPIIvesicles bud from ER exit sites. This half-time was determinedby measuring the fluorescence intensity of individual GFP-SEC24Dp labelled structures which were still faintly visiblefollowing the photobleaching. The recovery of fluorescencewithin these structures suggests that COPII can recycledirectly on to these same sites on the membrane. This isentirely consistent with the longevity of these structures inliving cells (for example see Fig. 2A). Furthermore, duringlonger periods of observation (up to 20 minutes), GFP-SEC24Dp labelled structures appearing spontaneously withinthe cell were not observed suggesting that they are notroutinely formed de novo.
To analyse the relationship between the SEC24p labelledstructures and secretory cargo transport from the ER to theGolgi complex, we visualised simultaneously YFP-SEC24Dpwith a CFP-tagged version of the well characterised secretorytransport marker ts-045-G (ts-O45-G-CFP). This glycoproteinaccumulates in the ER at the non-permissive temperature of39.5°C and upon shifting the temperature to 32°C, it exits theER as a relatively synchronous pulse. A GFP-tagged versionof ts-O45-G has been shown to be transported in large TCs(approx. ≈1 µm in size) which move in a microtubule-dependent manner into the Golgi complex, from where themarker is subsequently delivered to the plasma membrane(Presley et al., 1997; Scales et al., 1997).
Some of the ts-O45-G-CFP transport complexes observedat the permissive temperature during the period of analysis (5minutes) co-localised with YFP-SEC24Dp when theyappeared in the field of view for the first time (Fig. 5A,arrow). At later time-points the YFP-SEC24p and ts-O45-G-CFP were seen to dissociate from one another after which thelatter was transported towards the Golgi complex (Fig. 5Aand B). However, the majority of the ts-O45-G-CFPcontaining transport complexes moving to the Golgi werenever seen to be coated with YFP-SEC24Dp during the time-frame of analysis (Fig. 5C, arrowhead). They presumablyrepresent ts-O45-G-CFP containing transport complexes thathad formed and segregated from YFP-SEC24p positive ERexit sites outside of the time frame and/or focal plane duringimaging.
The dynamics of COPII-coated membranes describedabove is in sharp contrast with the dynamics described forCOPI-coated ER-to-Golgi transport complexes. COPI-coatedTCs are highly mobile and show directed movements fromthe cell periphery towards the Golgi complex (Shima et al.,1999). These COPI-coated TCs have been shown toparticipate in the transport of ts-O45-G from the cellperiphery to the Golgi complex (Shima et al., 1999). Todetermine the sequence of events for COPII and COPImembrane association, we imaged both coat complexessimultaneously in living cells. Cells expressing GFP-SEC24Dp were microinjected with Cy3-labelled CM1A10 tovisualise COPI (Shima et al., 1999). Approximately 25% ofthe SEC24Dp positive structures co-localised with COPI atany time during the observation period. COPI TCs were seento be transported to the Golgi apparatus after segregationfrom SEC24Dp positive sites (Fig. 6A). COPI TCs werefrequently observed moving towards the Golgi without beingassociated at any time with COPII (Fig. 6B), suggesting thatthey had segregated from SEC24Dp positive sites outside ofthe time frame of observation or outside the focal planeduring imaging. This behaviour directly correlates with thatof ts-O45-G-CFP (Fig. 5).
D. J. Stephens and others
Fig. 3.Time lapse imaging of YFP-SEC24Dp and CFP-ER expressingcells. (A) YFP-SEC24Dp(pseudocoloured red) positivestructures are associated with theendoplasmic reticulum (labelled byexpression of CFP-ER, pseudocolouredgreen) and their movements areprimarily due to movement of the ERnetwork itself. The three images in thetop panel are taken from the associatedQuickTime movie at time points of 0seconds, 20 seconds and 40 seconds.Arrowheads mark the starting positionsof two YFP-SEC24Dp structures.(B) Enlarged region of the centre of A(also shown as a QuickTime movie).Note that there is considerablemovement from these starting points(arrowhead) during the course of theexperiment but that the YFP-SEC24Dpstructures remain associated with theER membrane at all times. Bar, 5 µm.
2181In vivo dynamics of COPII
Fig. 4. (A) Fluorescence recovery afterphotobleaching of GFP-SEC24Dp structures.The marked (square) area was photobleached for3 seconds using 100% laser power of the 488nm line a Zeiss LSM510 confocal microscope att=0. Fluorescence recovery was subsequentlymonitored by acquiring images every 7.3seconds at 0.2% laser power. The arrowheadshighlight three such structures which are seen tobe partially bleached with subsequentfluorescence recovery. The arrows mark twostructures outside of the bleached area of whichthe fluorescence intensity remains essentiallyunchanged during the experiment. See alsoassociated QuickTime movie. (B) Shown is anenlarged version of the bleached area from the
first three panels of A. The arrowheads highlight the same three structures that are marked in A. These structures are incompletely bleachedduring the experiment and show subsequent recovery of fluorescence at the first time point taken (+7.3 seconds). This demonstrates thatpreexisting ER exit sites are replenished with COPII from the cytosolic pool. (C) The mean fluorescence intensity of three randomly selectedindividual structures within the bleached area (but from different cells) was quantitated and plotted against time.
C
2182
DISCUSSION
Time lapse microscopy has previously been used to visualisethe transport of secretory membrane markers to the Golgi
apparatus as well as the dynamics and organisation of COPI-coated transport complexes (Scales et al., 1997; Presley et al.,1997; Shima et al., 1999). Here we have extended this approachto characterise for the first time in vivo the dynamics of COPII
D. J. Stephens and others
Fig. 5. Dual visualisation ofYFP-SEC24Dp and CFP-ts-045-G in microinjected Verocells. (A) COPII (red) and ts-045-G (green) initiallycolocalise but ts-045v-Gtranslocates to the Golgi alone.The two panels show that startand end points of a time lapseexperiment shown in greaterdetail in B. The arrow shows theinitial position of the TC ofinterest, the black asteriskshows the position of the Golgi.Bar, 5 µm. (B) The arrow at t=0marks the starting position ofthe TC, the second arrow (t=8seconds) marks the position atwhich COPII and ts-045-Gsegregate. The arrowhead tracesthe trajectory of a structureinitially labelled with bothproteins from which ts-045-Gsegregates after 8 seconds ofimaging and continues to theGolgi apparatus. Bar, 5 µm. Seealso the associated QuickTimemovie. C. CFP-ts-045-G (green)structures are frequentlyvisualised moving towards theGolgi apparatus but do notcontain YFP-SEC24Dp (red).The arrow marks the startingpoint of the TC, the arrowheadtraces the trajectory of a CFP-ts-045-G TC moving to the Golgiwhich is not associated withYFP-SEC24Dp during thecourse of imaging. The timepoint of each frame taken fromthe associated QuickTime movieis shown in seconds. Bar, 5 µm.
2183In vivo dynamics of COPII
coat complex in ER-to-Golgi transport and its temporal andspatial relationship to COPI. Our data show that COPIIlocalises to long lived, relatively immobile structures tightlyassociated with ER membranes. Recovery after photobleachingexperiments show that COPII turns over on these membranewith a half-life of approximately 38 seconds. Secretory cargomoves through these ‘ER exit sites’ and segregates into COPI-coated transport complexes, which are long range carriersmoving cargo into the Golgi complex in a directed fashion.Although COPII and COPI show an overlapping distributionat ER exit sites, the COPI-coated TCs moving towards theGolgi do not contain any detectable amounts of COPII.
In summary these data suggest that ER-to-Golgi transport inmammalian cells can be viewed as a two step process. Firstcargo is concentrated and packaged by COPII at distinct ERexit sites. Second, coincident with the binding of COPI tothese, or adjacent, membranes, secretorycargo segregates from them into COPI-coated transport complexes which thenmove to the Golgi complex leaving theCOPII coat behind. This is consistentwith functional studies demonstratingthat COPII and COPI operatesequentially in ER-to-Golgi transport(Aridor et al., 1995; Pepperkok et al.,1998). Our data provide the first direct invivo evidence of this sequential mode ofaction.
Our data show that COPI-coated TCformation occurs in direct proximity toER cargo exit sites. This is consistentwith studies at the ultrastructural levelanalysing the distribution of COPI- andCOPII-coated membranes in cells.These studies have described atransitional zone of the ER,characterised by buds and vesicularprofiles coated by COPII (Orci et al.,1991; Kuge et al., 1994; Paccaud et al.,1996; Tang et al., 1997), whereas theintermediate compartment (identical tothe TCs described here and previously;Presley et al., 1997; Scales et al., 1997)is mainly decorated with COPI proteins(Oprins et al., 1993; Klumperman et al.,1998; Martinez-Menarguez et al., 1999).Interestingly, these two membranestructures are tightly juxtaposed, bothwhen located at the cell periphery or inthe close vicinity of the Golgi apparatus(Martinez-Menarguez et al., 1999).
We have not obtained any evidencefor small COPII-coated vesiclestransporting secretory cargo from the ERto the Golgi as suggested previously(Barlowe et al., 1994). Thus, our data arenot consistent with a model for COPII asa vectorial carrier of cargo, in this casets-045-G, from the ER to the Golgias is frequently proposed (Barlowe,1998; Kaiser and Ferro-Novick, 1998;
Martinez-Menarguez et al., 1999). Although we cannotexclude the possibility that our detection methods are notsufficiently sensitive to detect such small COPII-coatedvesicles, we consider it unlikely that they would contributesignificantly to the transport of ts-045-G to the Golgiapparatus. Considerable evidence now exists showing thatlarge pleiomorphic TCs are responsible for transport of ts-045-G and other cargo to the Golgi apparatus (Presley et al., 1997;Scales et al., 1997; Shima et al., 1999). Therefore, we considerit more likely that COPII vesicles are predominantly involvedin short distance transport at the level of the ER, close to ERexit sites.
At least two possibilities exist describing the role of COPIIvesicles in this process. A model integrating our current datawith existing results is shown in Fig. 7. ER cargo isconcentrated and packaged into COPII vesicles at distinct,
Fig. 6.Dual visualisation of COPI and COPII complexes. COPII was labelled by expressionof GFP-SEC24Dp, COPI by microinjection of expressing cells with Cy3-labelled CM1A10which labels but does not interfere with the function of the COPI complex (Shima et al.,1999). A. COPI (red) and COPII (green) are seen to localise very close to one another initially(arrowhead, t=0), after which COPI alone is transported to the Golgi apparatus. The arrowmarks the starting point of the TC, arrowheads trace the trajectories of the two markers. Bar,5 µm. See associated QuickTime movie. B. COPI TCs (red) are often seen trafficking to theGolgi without being associated with COPII (green) at any point during imaging. Bar, 5 µm.This is most clearly shown in the associated QuickTime movie.
2184
stable ER exit sites (Fig. 7, step I). Once they pinch off fromthe ER, COPII vesicles can fuse either with a pre-existing TC(Fig. 7, model A, step II), or with each other to form a TC denovo (Fig. 7, model B, step II). Both models are consistentwith the localisation of components of membrane fusionmachinery (v- and t-SNARES) to ER exit sites or adjacentmembranes similar to those described here (Hay et al., 1998;Chao et al., 1999). In either model, the cargo content is nowtopologically separated from the ER as well as from the Golgicomplex. In a second step, the segregation of cargo fromCOPII- to COPI-coated TCs occurs (Fig. 7, step II). TheseTCs subsequently move to the Golgi complex in a COPIdependent (Shima et al., 1999) but COPII independentmanner.
We thank Drs Kai Simons, B. L. Tang and Wanjin Hong forantibodies, Patrick Keller and Derek Toomre for the ts-O45-G-CFPvariant. Jeremy Simpson and Andreas Girod for help, advice and
critical reading of the manuscript, Jens Rietdorf for invaluableassistance with microscopy and Martin Lowe for critical reading ofthe manuscript. We also thank Eppendorf, Improvision, Olympus andT.I.L.L. Photonics for support of the Advanced Light MicroscopyFacility at EMBL Heidelberg with equipment. This work wassupported in part by grants from the Schmidheiny Foundation, theHelmut Horten Stiffung and the de Reuter Foundation (to J.-P.Paccaud). David Stephens is funded by a Long Term Fellowship fromthe European Molecular Biology Organisation.
REFERENCES
Aridor, M., Bannykh, S. I., Rowe, T. and Balch, W. E. (1995). Sequentialcoupling between COPII and COPI vesicle coats in endoplasmic reticulumto Golgi transport. J. Cell Biol. 131, 875-893.
Aridor, M., Weissman, J., Bannykh, S., Nuoffer, C. and Balch, W. E.(1998). Cargo selection by the COPII budding machinery during exportfrom the ER. J. Cell Biol. 141, 61-70.
Barlowe, C., Orci, L., Yeung, T., Hosobuchi, M., Hamamoto, S., Salama,N., Rexach, M. F., Ravazzola, M., Amherdt, M. and Schekman, R.(1994). COPII: a membrane coat formed by Sec proteins that drive vesiclebudding from the endoplasmic reticulum. Cell 77, 895-907.
Barlowe, C. (1998). COPII and selective export from the endoplasmicreticulum. Biochim. Biophys. Acta1404, 67-76.
Bednarek, S. Y., Ravazzola, M., Hosobuchi, M., Amherdt, M., Perrelet, A.,Schekman, R. and Orci, L. (1995). COPI- and COPII-coated vesicles buddirectly from the endoplasmic reticulum in yeast. Cell 83, 1183-1196.
Chao, D. S., Hay, J. C., Winnick, S., Prekeris, R., Klumperman, J. andScheller, R. H. (1999). SNARE membrane trafficking dynamics in vivo. J.Cell Biol. 144, 869-881.
Hay, J. C., Klumperman, J., Oorschot, V., Steegmaier, M., Kuo, C. S. andScheller, R. H. (1998). Localization, dynamics, and protein interactionsreveal distinct roles for ER and Golgi SNAREs. J. Cell Biol. 141, 1489-1502.
Kaiser, C. A. and Ferro-Novick, S.(1998). Transport from the endoplasmicreticulum to the Golgi. Curr. Opin. Cell Biol. 10, 477-482.
Kaiser, C. A. and Schekman, R. (1990). Distinct sets of SEC genes governtransport vesicle formation and fusion early in the secretory pathway. Cell61, 723-733.
Klumperman, J., Schweizer, A., Clausen, H., Tang, B. L., Hong, W.,Oorschot, V. and Hauri, H. P. (1998). The recycling pathway of proteinERGIC-53 and dynamics of the ER-Golgi intermediate compartment. J. CellSci. 111, 3411-3425.
Kuehn, M. J. and Schekman, R. (1997). COPII and secretory cargo captureinto transport vesicles. Curr. Opin. Cell Biol. 9, 477-483.
Kuehn, M. J., Herrmann, J. M. and Schekman, R. (1998). COPII-cargointeractions direct protein sorting into ER-derived transport vesicles. Nature391, 187-190.
Kuge, O., Dascher, C., Orci, L., Rowe, T., Amherdt, M., Plutner, H.,Ravazzola, M., Tanigawa, G., Rothman, J. E. and Balch, W. E. (1994).Sar1 promotes vesicle budding from the endoplasmic reticulum but notGolgi compartments. J. Cell Biol. 125, 51-65.
Martinez-Menarguez, J. A., Geuze, H. J., Slot, J. W. and Klumperman, J.(1999). Vesicular tubular clusters between the ER and Golgi mediateconcentration of soluble secretory proteins by exclusion from COPI-coatedvesicles. Cell 98, 81-90.
Nagase, T., Ishikawa, K., Suyama, M., Kikuno, R., Miyajima, N., Tanaka,A., Kotani, H. Nomura, N. and Ohara, O. (1998). Prediction of the codingsequences of unidentified human genes. XI. The complete sequences of 100new cDNA clones from brain which code for large proteins in vitro. DNARes. 5, 277-286.
Novick, P., Field, C. and Schekman, R. (1980). Identification of 23complementation groups required for post-translational events in the yeastsecretory pathway. Cell 21, 205-215.
Oprins, A., Duden, R., Kreis, T. E., Geuze, H. J. and Slot, J. W. (1993).Beta-COP localizes mainly to the cis-Golgi side in exocrine pancreas. J CellBiol. 121, 49-59.
Orci, L., Ravazzola, M., Meda, P., Holcomb, C., Moore, H. P., Hicke, L.and Schekman, R. (1991). Mammalian Sec23p homologue is restricted tothe endoplasmic reticulum transitional cytoplasm. Proc. Nat. Acad. Sci. USA88, 8611-8615.
Paccaud, J. P., Reith, W., Carpentier, J. L., Ravazzola, M., Amherdt, M.,
D. J. Stephens and others
Fig. 7. Model A: COPII-coated vesicles bud from the ER and fusewith a pre-existing transport complex (step I). Model B: alternatively,COPII-coated vesicles could homotypically fuse to form a transportcomplex de novo (step I). In both cases, association of COPI (step II)initiates segregation of cargo from COPII to COPI-coated structures.These TCs subsequently move to the Golgi apparatus in a COPI-dependent but COPII-independent manner (step III). COPII coatmolecules are shown recycling directly back to the ER membraneconsistent with our photobleaching studies.
2185In vivo dynamics of COPII
Schekman, R. and Orci, L. (1996). Cloning and functional characterizationof mammalian homologues of the COPII component Sec23. Mol. Biol. Cell7, 1535-1546.
Pagano, A., Letourneur, F., Garcia-Estefania, D., Carpentier, J. L., Orci,L. and Paccaud, J. P. (1999). SEC24 proteins and sorting at theendoplasmic reticulum. J. Biol. Chem. 274, 7833-7840.
Pepperkok, R., Scheel, J., Horstmann, H., Hauri, H. P., Griffiths, G. andKreis, T. E. (1993). Beta-COP is essential for biosynthetic membranetransport from the endoplasmic reticulum to the Golgi complex in vivo. Cell74, 71-82.
Pepperkok, R., Lowe, M., Burke, B. and Kreis, T. E. (1998). Three distinctsteps in transport of vesicular stomatitis virus glycoprotein from the ER tothe cell surface in vivo with differential sensitivities to GTP gamma S. J.Cell Sci. 111, 1877-1888.
Presley, J. F., Cole, N. B., Schroer, T. A., Hirschberg, K., Zaal, K. J. andLippincott-Schwartz, J. (1997). ER-to-Golgi transport visualized in livingcells. Nature389, 81-85.
Roberg, K. J., Crotwell, M., Espenshade, P., Gimeno, R. and Kaiser, C. A.(1999). LST1 is a SEC24 homologue used for selective export of the plasmamembrane ATPase from the endoplasmic reticulum. J. Cell Biol. 145, 659-672.
Scales, S. J., Pepperkok, R. and Kreis, T. E. (1997). Visualization of ER-to-Golgi transport in living cells reveals a sequential mode of action for COPIIand COPI. Cell 90, 1137-1148.
Schekman, R. and Orci, L. (1996). Coat proteins and vesicle budding.Science271, 1526-1533.
Shima, D. T., Scales, S. J., Kreis, T. E. and Pepperkok, R. (1999).Segregation of COPI-rich and anterograde-cargo-rich domains inendoplasmic-reticulum-to-Golgi transport complexes. Curr. Biol. 9, 821-824.
Tang, B. L., Peter, F., Krijnse-Locker, J., Low, S. H., Griffiths, G. andHong, W. (1997). The mammalian homolog of yeast Sec13p is enriched inthe intermediate compartment and is essential for protein transport from theendoplasmic reticulum to the Golgi apparatus. Mol. Cell. Biol. 17, 256-266.
![[Irving M. Copi, Carl Cohen] Introduction to Logic(BookFi.org)](https://static.documents.pub/doc/80x56/55cf9a7f550346d033a205d7/irving-m-copi-carl-cohen-introduction-to-logicbookfiorg.jpg)
