Kinesin is the motor for microtubule-mediated Golgi-to-ER membrane traffic [published errata appear in J Cell Biol 1995 Mar;128(5):following 988 and 1995 May;129(3):893]

The distribution and dynamics of both the ER and Golgi complex in animal cells are known to be dependent on microtubules; in many cell types the ER extends toward the plus ends of microtubules at the cell periphery and the Golgi clusters at the minus ends of microtubules near the centrosome. In this study we provide evidence that the microtubule motor, kinesin, is present on membranes cycling between the ER and Golgi and powers peripherally directed movements of membrane within this system. Immunolocalization of kinesin at both the light and electron microscopy levels in NRK cells using the H1 monoclonal antibody to kinesin heavy chain, revealed kinesin to be associated with all membranes of the ER/Golgi system. At steady-state at 37 degrees C, however, kinesin was most concentrated on peripherally distributed, pre- Golgi structures containing beta COP and vesicular stomatitis virus glycoprotein newly released from the ER. Upon temperature reduction or nocodazole treatment, kinesin's distribution shifted onto the Golgi, while with brefeldin A (BFA)-treatment, kinesin could be found in both Golgi-derived tubules and in the ER. This suggested that kinesin associates with membranes that constitutively cycle between the ER and Golgi. Kinesin's role on these membranes was examined by microinjecting kinesin antibody. Golgi-to-ER but not ER-to-Golgi membrane transport was found to be inhibited by the microinjected anti-kinesin, suggesting kinesin powers the microtubule plus end-directed recycling of membrane to the ER, and remains inactive on pre-Golgi intermediates that move toward the Golgi complex.     

Effect of the microtubule disrupting agents,colchicine and vinblastine,on seminiferous tubule structure in the rat

Injections of colchicine or vinblastine were given intratesticularly and rats sacrificed 6 and 12 hr later. Colchicine and vinblastine produced identical morphological patterns of response in the seminiferous tubules resulting in arrest of germcell mitoses and meioses and a rapid depletion of the microtubules normally found within the Sertoli cell. Sloughing of cells into the lumen of seminiferous tubules was the most prominent feature noted. Germ cells and portions of the apical Sertoli cells were frequently sloughed together where they remained in close association. Usually germ cells and associated Sertoli cell fragments were cleaved from the wall of the seminiferous tubule at a level between dissimilar generations of germ cells, e.g. between spermatocytes and spermatids. This selective sloughing probably occurred as the result of the support normally provided by intercellular bridges which link clones of like germ cell types. Sequential steps in the process leading to sloughing of Sertoli-germ cell associations could be inferred from observations made in plastic 1 μm sections. Cell sloughing at 12 hr post-injection was generally more extensive. It was frequently noted that germ cells and the apical portions of Sertoli cells had been extruded to the level of the most adluminal tight junctions forming the blood-testis barrier. It was concluded that disruption of Sertoli microtubules was responsible for sloughing of Sertoli fragments and associated germ cells, and that the cytoskeletal support of the Sertoli cell was, at least in part, dependent upon the integrity of Sertoli microtubules. The Sertoli cell could not round-up after loss of its cytoskeletal support, due to the numerous attachment devices known to link it with various apically positioned germ cells. Thus, the cell was severed at some point along its delicate apical processes, as the consequence of forces produced by the ‘rounding-up’ process. Long-term sacrifice after vinblastine or colchicine treatment allowed the Sertoli cells to regain microtubules and long processes but not their typical configuration. Spermatogenesis remained severely impaired.     

The microtubule-dependent formation of a tubulovesicular network with characteristics of the ER from cultured cell extracts

The formation of a dynamic tubulovesicular membrane network that resembles the endoplasmic reticulum (ER) has been observed in extracts of cultured chick embryo fibroblasts (CEF cells) using video-enhanced differential interference contrast microscopy. Initially, membranes in the CEF extracts appeared amorphous and aggregated, but with time, membrane tubules moved out along stationary microtubules. The membrane tubules formed new branches on intersecting microtubules and fused with other branches to form a network of interconnected polygons. The tubulovesicular network was solubilized by detergent and took on a beaded morphology in a hypotonic buffer. Formation of the tubulovesicular network required ATP and microtubules. The network did not contain elements of the plasma membrane, Golgi apparatus, or mitochondria but could be labeled with ER markers. We suggest that the tubulovesicular network contains components from the ER and is formed by membrane associated motors moving upon microtubules in a process we call microtubule-dependent tethering.     

Reconstitution of the Golgi apparatus after microinjection of rat liver Golgi fragments into Xenopus oocytes

We have studied the reconstitution of the Golgi apparatus in vivo using an heterologous membrane transplant system. Endogenous glycopeptides of rat hepatic Golgi fragments were radiolabeled in vitro with [3H]sialic acid using detergent-free conditions. The Golgi fragments consisting of dispersed vesicles and tubules with intraluminal lipoprotein-like particles were then microinjected into Xenopus oocytes and their fate studied by light (LM) and electron microscope (EM) radioautography. 3 h after microinjection, radiolabel was observed by LM radioautography over yolk platelet-free cytoplasmic regions near the injection site. EM radioautography revealed label over Golgi stacked saccules containing the hepatic marker of intraluminal lipoprotein-like particles. At 14 h after injection, LM radioautographs revealed label in the superficial cortex of the oocytes between the yolk platelets and at the oocyte surface. EM radioautography identified the labeled structures as the stacked saccules of the Golgi apparatus, the oocyte cortical granules, and the plasmalemma, indicating that a proportion of microinjected material was transferred to the surface via the secretion pathway of the oocyte. The efficiency of transport was low, however, as biochemical studies failed to show extensive secretion of radiolabel into the extracellular medium by 14 h with approximately half the microinjected radiolabeled constituents degraded. Vinblastine (50 microM) administered to oocytes led to the formation of tubulin paracrystals. Although microinjected Golgi fragments were able to effect the formation of stacked saccules in vinblastine-treated oocytes, negligible transfer of heterologous material to the oocyte surface could be detected by radioautography. The data demonstrate that dispersed fragments of the rat liver Golgi complex (i.e., unstacked vesicles and tubules) reconstitute into stacked saccules when microinjected into Xenopus cytoplasm. After the formation of stacked saccules, reconstituted Golgi fragments transport constituents into a portion of the exocytic pathway of the host cell by a microtubule-regulated process.     

Interaction of the Golgi membranes isolated from rabbit liver with microtubules in vitro.

We have developed a reconstituted model system to study the interaction of the Golgi membranes isolated from rabbit liver with taxol-stabilized bovine-brain microtubules without microtubule-associated proteins (MAPs). The Golgi membranes are associated with microtubules. The sheets of vesicles and the membranous tubules are observed along microtubules by direct visualization using differential-interference-contrast, dark field, or fluorescence microscopy. The monoclonal antibody against Golgi membranes suggests that the Golgi membranes, but not the contaminating vesicles, are interacting with microtubules. The degree of association is assayed quantitatively using rhodamine-labeled microtubules after separation of the complex from unbound microtubules by centrifugation upon sucrose gradient. The association is inhibited by crude MAPs, purified MAP2, or 1.0 mM ATP. However, the association neither requires the cytosol from rat liver or bovine brain nor N-ethylmaleimide, brefeldin A, or GTP-gamma-S. The association is mediated by trypsin-sensitive peripheral protein(s) on the Golgi membranes.     

Tubulation of Golgi membranes in vivo and in vitro in the absence of brefeldin A

Recent in vivo studies with the fungal metabolite, brefeldin A (BFA), have shown that in the absence of vesicle formation, membranes of the Golgi complex and the trans-Golgi network (TGN) are nevertheless able to extend long tubules which fuse with selected target organelles. We report here that the ability to form tubules (> 7 microns long) could be reproduced in vitro by treatment of isolated, intact Golgi membranes with BFA under certain conditions. Surprisingly, an even more impressive degree of tubulation could be achieved by incubating Golgi stacks with an ATP-reduced cytosolic fraction, without any BFA at all. Similarly, tubulation of Golgi membranes in vivo occurred after treatment of cells with intermediate levels of NaN3 and 2-deoxyglucose. The formation of tubules in vitro, either by BFA treatment or low-ATP cytosol, correlated precisely with a loss of the vesicle-associated coat protein beta-COP from Golgi membranes. After removal of BFA or addition of ATP, membrane tubules served as substrates for the rebinding of beta-COP and for the formation of vesicles in vitro. These results provide support for the idea that a reciprocal relationship exists between tubulation and vesiculation (Klausner, R. D., J. G. Donaldson, and J. Lippincott-Schwartz. 1992. J. Cell Biol. 116:1071- 1080). Moreover, they show that tubulation is an inherent property of Golgi membranes, since it occurs without the aid of microtubules or BFA treatment. Finally the results indicate the presence of cytosolic factors, independent of vesicle-associated coat proteins, that mediate the budding/tubulation of Golgi membranes.     

Interaction of early secretory pathway and Golgi membranes with microtubules and microtubule motors

This review summarizes the data describing the role of cellular microtubules in transportation of membrane vesicles — transport containers for secreted proteins or lipids. Most events of early vesicular transport in animal cells (from the endoplasmic reticulum to the Golgi apparatus and in the opposite recycling direction) are mediated by microtubules and microtubule motor proteins. Data on the role of dynein and kinesin in early vesicle transport remain controversial, probably because of the differentiated role of these proteins in the movements of vesicles or membrane tubules with various cargos and at different stages of secretion and retrograde transport. Microtubules and dynein motor protein are essential for maintaining a compact structure of the Golgi apparatus; moreover, there is a set of proteins that are essential for Golgi compactness. Dispersion of ribbon-like Golgi often occurs under physiological conditions in interphase cells. Golgi is localized in the leading part of crawling cultured fibroblasts, which also depends on microtubules and dynein. The Golgi apparatus creates its own system of microtubules by attracting γ-tubulin and some microtubule-associated proteins to membranes. Molecular mechanisms of binding microtubule-associated and motor proteins to membranes are very diverse, suggesting the possibility of regulation of Golgi interaction with microtubules during cell differentiation. To illustrate some statements, we present our own data showing that the cluster of vesicles induced by expression of constitutively active GTPase Sar1a[H79G] in cells is dispersed throughout the cell after microtubule disruption. Movement of vesicles in cells containing the intermediate compartment protein ERGIC53/LMANI was inhibited by inhibiting dynein. Inhibiting protein kinase LOSK/SLK prevented orientation of Golgi to the leading part of crawling cells, but the activity of dynein was not inhibited according to data on the movement of ERGIC53/LMANI-marked vesicles.     

The Golgi apparatus in honeybee photoreceptor cells: structural organization and spatial relationship to microtubules and actin filaments

The architecture of the Golgi complex in honeybee photoreceptors has been analyzed by electron-microscopic techniques. The Golgi apparatus consists of several hundred individual stacks of cisternae dispersed throughout the soma of the photoreceptor cell. Two distinct subpopulations of Golgi stacks are distinguishable by their topographic features: (1) a dense row of Golgi stacks is aligned along the palisade-like cisternae of smooth endoplasmic reticulum backing the photoreceptive microvilli; (2) other Golgi stacks are scattered in the remainder of the cell body. The spatial relationship of Golgi stacks to microtubules and actin filaments has also been determined. Electron-microscopic examination of high-pressure-frozen freeze-substituted retinae reveals that Golgi stacks backing the submicrovillar endoplasmic reticulum reside in a cell area without microtubules, whereas the second subpopulation of Golgi stacks is embedded amidst microtubules. Labeling studies with several actin-specific probes, viz., rhodamine phalloidin, monoclonal anti-actin antibodies, and myosin fragments, provide evidence for a juxtaposition of the submicrovillar Golgi stacks to actin filaments. The Golgi membranes are thus ideally positioned to facilitate the transport of Golgi-derived material toward the microvilli along actin filaments.     

CYTOPLASMIC MICROTUBULES : I. Hydra

Small cytoplasmic tubules are present in the interstitial cells and cnidoblasts of hydra. They are referred to here as "microtubules." These tubular elements have an outside diameter of 180 A and an inside diameter of 80 A. By difference, the membranous wall is estimated to be 50 A thick. The maximum length of the microtubules cannot be determined from thin sections but is known to exceed 1.5 µ. In the interstitial cells the microtubules are found in the intercellular bridges, free in the cytoplasm and in association with the centrioles. In the cnidoblast they form a framework around the developing nematocyst and in late stages are related to the cnidocil forming a tight skein in the basal part of the cell. Especially in this cell, confluence of microtubules with small spherical vesicles of the Golgi complex has been observed. It is proposed that these tubules function in the transport of water, ions, or small molecules.     

Possible role of centrioles in stabilizing cytoplasmic microtubules

Small fragments of the peripheral cytoplasm were obtained from cytochalasine B-treated mouse embryo fibroblasts and studied for distribution of microtubules by indirect immunofluorescence. Microtubules were demonstrated to progressively depolymerize in these fragments which did not contain any tubules after 6 hours of incubation in the growth medium. This effect was specific for microtubules, since the distribution of intermediate filaments remained unchanged during incubation. The fragments remained viable during incubation, inasmuch no changes were detectable in the membrane potential of the mitochondria stained with rhodamine 123. Progressive destruction of microtubules in the tiny cell fragments is likely to be related to the lack of centrioles in such fragments.     

Role of Microtubules in the Organization of the Golgi Complex

The Golgi complex of mammalian cells is composed of cisternal stacks that function in processing and sorting of membrane and luminal proteins during transport from the site of synthesis in the endoplasmic reticulum to lysosomes, secretory vacuoles, and the cell surface. Even though exceptions are found, the Golgi stacks are usually arranged as an interconnected network in the region around the centrosome, the major organizing center for cytoplasmic microtubules. A close relation thus exists between Golgi elements and microtubules (especially the stable subpopulation enriched in detyrosinated and acetylated tubulin). After drug-induced disruption of microtubules, the Golgi stacks are disconnected from each other, partly broken up, dispersed in the cytoplasm, and redistributed to endoplasmic reticulum exit sites. Despite this, intracellular protein traffic is only moderately disturbed. Following removal of the drugs, scattered Golgi elements move along reassembling microtubules back to the centrosomal region and reunite into a continuous system. The microtubule-dependent motor proteins cytoplasmic dynein and kinesin bind to Golgi membranes and have been implicated in vesicular transport to and from the Golgi complex. Microinjection of dynein heavy chain antibodies causes dispersal of the Golgi complex, and the Golgi complex of cells lacking cytoplasmic dynein is likewise spread throughout the cytoplasm. In a similar manner, kinesin antibodies have been found to inhibit Golgi-to-endoplasmic reticulum transport in brefeldin A-treated cells and scattering of Golgi elements along remaining microtubules in cells exposed to a low concentration of nocodazole. The molecular mechanisms in the interaction between microtubules and membranes are, however, incompletely understood. During mitosis, the Golgi complex is extensively reorganized in order to ensure an equal partitioning of this single-copy organelle between the daughter cells. Mitosis-promoting factor, a complex of cdc2 kinase and cyclin B, is a key regulator of this and other events in the induction of cell division. Cytoplasmic microtubules depolymerize in prophase and as a result thereof, the Golgi stacks become smaller, disengage from each other, and take up a perinuclear distribution. The mitotic spindle is thereafter put together, aligns the chromosomes in the metaphase plate, and eventually pulls the sister chromatids apart in anaphase. In parallel, the Golgi stacks are broken down into clusters of vesicles and tubules and movement of protein along the exocytic and endocytic pathways is inhibited. Using a cell-free system, it has been established that the fragmentation of the Golgi stacks is due to a continued budding of transport vesicles and a concomitant inhibition of the fusion of the vesicles with their target membranes. In telophase and after cytokinesis, a Golgi complex made up of interconnected cisternal stacks is recreated in each daughter cell and intracellular protein traffic is resumed. This restoration of a normal interphase morphology and function is dependent on reassembly of a radiating array of cytoplasmic microtubules along which vesicles can be carried and on reactivation of the machinery for membrane fusion.     

Fine structure of the Golgi complex during mitosis of cartilaginous cells in vitro

Chondrocytes were isolated enzymatically from guinea-pig epiphyses and grown in vitro. The fate of the Golgi complex during mitosis in relation to changes in the cytoplasmic microtubules was then studied by transmission electron microscopy. Interphase cells were observed to be polarized, with the Golgi complex occupying a well-defined juxtanuclear area of the cell's cytoplasmic pole. During prophase the cytoplasmic microtubules were largely lost, the nucleus moved to the center of the cell and the Golgi complex dissolved into single dictyosomes spread diffusely throughout the cytoplasm. The distribution of other organelles also changed to a more random pattern. In telophase, i.e. after the completion of nuclear division, the mitotic spindle decomposed and cytoplasmic microtubules reappeared. Furthermore, the organization of the Golgi complex and other organelles returned to that characteristic of interphase cells. Previous studies on cells treated with colchicine have indicated that the polarized distribution of cell organelles is dependent on the presence of intact cytoplasmic micro-tubules. It is suggested that the disappearance of such tubules observed here to be coupled with the disorganization of cell interphase structure fulfills the double function of providing free tubulin units from which to build the mitotic spindle and ensuring an approximately equal distribution of dictyosomes and other organelles to the daughter cells during cytokinesis.     

HOW MICROTUBULE PATTERNS ARE GENERATED : The Relative Importance of Nucleation and Bridging of Microtubules in the Formation of the Axoneme of Raphidiophrys

The axonemes of Raphidiophrys converge near the center of the cell in an electron-opaque material, the centroplast. In order to establish whether this material acts not only to nucleate the microtubules which form the axonemes but also to give the axoneme its characteristic pattern, the microtubules were disassembled with low temperature and stages in their reformation were studied. It was shown that even though the microtubules appear to be nucleated from the centroplast, pattern formation first appeared at a distance from the centroplast. Thus, the axonemal pattern could not be attributed to any prepattern in the centroplast. Rather, the pattern appears to arise by specific interactions between tubules brought about by bridges. It was concluded that each tubule could bind to a maximum of four other tubules and that once one bridge attached to a tubule it specified the binding positions of the others, thus giving the characteristic axonemal pattern of Raphidiophrys.     

THE DEVELOPMENT OF THE SECONDARY WALL OF THE XYLEM IN ACER PSEUDOPLATANUS

The development of the spirally thickened xylem element from a cambium initial of sycamore Acer pseudoplatanus has been traced by means of electron microscopy. The narrow elongated cambial initial undergoes considerable expansion in all dimensions. The cytoplasm at this stage is distributed in a thin skin between the cell wall and a large vacuole. No correlation has been observed between the distribution of any organelle and the pattern of the eventual thickenings. After the sites of thickening deposition have become apparent, the most conspicuous feature of the cell is the proliferation of Golgi bodies and vesicles. It is suggested that the material of the developing thickenings stems from direct apposition of the material in the Golgi vesicles. After glutaraldehyde fixation, microtubules (200 to 220 A in diameter) are seen to be sited in specific relation to the thickenings, the orientation of the tubules mirroring that of the fibrils seen in the thickenings. Possible reasons for absence of an observable pattern in the expanded but relatively undifferentiated cell are given, and the possible roles of the Golgi apparatus and microtubules in the thickening production are discussed     

Change of microtubular arrangement around centrioles in rat incisor odontoblasts during cell differentiation

Three stages during cell differentiation of rat incisor odontoblasts were classified, and change of microtubular arrangement around centrioles in the odontoblasts was examined with three-dimensional analyses using serial ultrathin sections. In the undifferentiated odontoblasts, microtubules were observed to radiate from the pericentriolar area, whereas, in the differentiating odontoblasts, some microtubules became poorly related to the centrioles. In the differentiated odontoblasts, arrangement of most microtubules appeared to have a poor relationship to the centrioles. Throughout the differentiation of the odontoblasts, one of the centriolar pair was ciliated, and Golgi apparatus was invariably observed near the centrioles. The present study suggests that a pericentriolar area, or a centrosome, could function as a microtubule-organizing center (MTOC) in the undifferentiated odontoblasts, but their function might be attenuated during cell differentiation.     

Ultrastructural effects of colchicine on perikarya and initial segment of rat retinal ganglion cells.

Our ultrastructural study was focused on the perikaryal region and initial segment of the axon of rat retinal ganglion cells in controls and after intraocular injections of colchicine. In control rats that region contained, among other organelles, elements of the Golgi complex and, close to them, short isolated microtubules oriented preferentially toward the axon where they funnel and aggregate in bundles. One day after sufficient doses of colchicine to inhibit axoplasmic transport (2-20 micrograms) these cytoplasmic microtubules were absent, whereas some axonal microtubules were still present but reduced in number. In addition, colchicine induced an altered distribution of organelles, leaving empty spaces in the periphery and most organelles concentrated in the perinuclear region, especially around Golgi elements where numerous vesicles and tubules accumulate at the trans face of Golgi elements. These results suggest that the vesicles that leave the Golgi and have been directed towards axoplasmic transport may need the cytoplasmic microtubules located between Golgi elements and the axonal initial segments to reach the axon.     

THE FINE STRUCTURE OF ACANTHAMOEBA CASTELLANII : I. The Trophozoite

The fine structure of the trophozoite of Acanthamoeba castellanii (Neff strain) has been studied. Locomotor pseudopods, spikelike "acanthopodia," and microprojections from the cell surface are all formed by hyaline cytoplasm, which excludes formed elements of the cell and contains a fine fibrillar material. Golgi complex, smooth and rough forms of endoplasmic reticulum, digestive vacuoles, mitochondria, and the water-expulsion vesicle (contractile vacuole) are described. A canicular system opening into the water-expulsion vesicle contains tubules about 600 A in diameter that are lined with a filamentous material. The tubules are continuous with unlined vesicles or ampullae of larger diameter. Centrioles were not observed, but cytoplasmic microtubules radiate from a dense material similar to centriolar satellites and are frequently centered in the Golgi complex. Cytoplasmic reserve materials include both lipid and glycogen, each of which amounts to about 10% of the dry weight.     

The fine structure and phosphatase cytochemistry of the Golgi complex and associated structures in the Sertoli cells of Syrian hamsters

The Golgi complex in the Sertoli cell of the Syrian hamster is well developed and consists of stacks of cisternae and associated vesicles. The inner- and outermost cisternae of the Golgi stacks are usually moderately dilated and exhibit numerous fenestrations. The middle portions of the intermediate cisternae are greatly flattened and not fenestrated, but toward the periphery these cisternae gradually become dilated and show a few fenestrations. On the inner aspect of the Golgi stacks the following structures are seen frequently: (1) one or two series of linearly arrayed circular profiles some of which are interconnected by tubules; (2) networks of anastomosing tubules with circular or oval meshes (800 to 1200 A in diameter); and/or (3) irregularly disposed tubules. The circular profiles and tubules are approximately 450 A in diameter. Acid phosphatase activity was localized in these anastomosing tubules when the tissues were incubated for more than one hour in a modified Gomori's medium (Barka and Anderson, 1963). Strong thiamine pyrophosphatase activity was demonstrated in the inner one to three cisternae of the Golgi stacks but not in the associated tubules. The system of the Golgi associated tubules is morphologically and histochemically distinct from the Golgi stacks and is probably equivalent to the Golgi-endoplasmic reticulum-lysosome system (GERL) in other cell types. The three dimensional aspects of the GERL-equivalent system are discussed.     

Cytoplasmic dynein participates in the centrosomal localization of the Golgi complex

The localization of the Golgi complex depends upon the integrity of the microtubule apparatus. At interphase, the Golgi has a restricted pericentriolar localization. During mitosis, it fragments into small vesicles that are dispersed throughout the cytoplasm until telophase, when they again coalesce near the centrosome. These observations have suggested that the Golgi complex utilizes a dynein-like motor to mediate its transport from the cell periphery towards the minus ends of microtubules, located at the centrosome. We utilized semi-intact cells to study the interaction of the Golgi complex with the microtubule apparatus. We show here that Golgi complexes can enter semi-intact cells and associate stably with cytoplasmic constituents. Stable association, termed here "Golgi capture," requires ATP hydrolysis and intact microtubules, and occurs maximally at physiological temperature in the presence of added cytosolic proteins. Once translocated into the semi-intact cell cytoplasm, exogenous Golgi complexes display a distribution similar to endogenous Golgi complexes, near the microtubule-organizing center. The process of Golgi capture requires cytoplasmic tubulin, and is abolished if cytoplasmic dynein is immunodepleted from the cytosol. Cytoplasmic dynein, prepared from CHO cell cytosol, restores Golgi capture activity to reactions carried out with dynein immuno-depleted cytosol. These results indicate that cytoplasmic dynein can interact with isolated Golgi complexes, and participate in their accumulation near the centrosomes of semi-intact, recipient cells. Thus, cytoplasmic dynein appears to play a role in determining the subcellular localization of the Golgi complex.     

Tension in tubulovesicular networks of Golgi and endoplasmic reticulum membranes

The endoplasmic reticulum (ER) and Golgi have robust bidirectional traffic between them and yet form distinct membrane compartments. Membrane tubules are pulled from large aggregates of ER or Golgi by microtubule motors to form ER tubulovesicular networks or Golgi tubules both in vivo and in vitro. The physical properties of membranes are critical for membrane traffic and organelle morphology. For example, tension applied to membranes can create tethers, drive membrane flow, and set the diameter of the tubules. Here, we formed ER and Golgi membrane networks in vitro and used optical tweezers to measure directly, for the first time, the membrane tensions of these organelles to clarify the possible role of tension in membrane flow. We report that higher forces are needed to form tethers from ER (18.6 +/- 2.8 pN) than from Golgi (11.4 +/- 1.4 pN) membrane tubules in vitro. Since ER tubules are smaller in diameter than Golgi tubules, it follows that Golgi networks have a lower tension than ER. The lower tension of the ER could be an explanation of how Golgi tubules can be rapidly drawn into the ER by tension-driven flow after fusion, as is observed in vivo.