There is little consensus on the nature of the storage compartment of the glucose transporter GLUT4, in non-stimulated cells of muscle and fat. More specifically, it is not known whether GLUT4 is localized to unique, specialized intracellular storage vesicles, or to vesicles that are part of the constitutive endosomal-lysosomal pathway. To address this question, we have investigated the localization of the endogenous GLUT4 in non-stimulated skeletal myotubes from the cell line C2, by immunofluorescence and immunoelectron microscopy. We have used a panel of antibodies to markers of the Golgi complex (α mannosidase II and giantin), of the trans-Golgi network (TGN38), of lysosomes (lgp110), and of early and late endosomes (transferrin receptor and mannose-6-phosphate receptor, respectively), to define the position of their subcellular compartments. By immunofluorescence, GLUT4 appears concentrated in the core of the myotubes. It is primarily found around the nuclei, in a pattern suggesting an association with the Golgi complex, which is further supported by colocalization with giantin and by immunogold electron microscopy. GLUT4 appears to be in the trans-most cisternae of the Golgi complex and in vesicles just beyond, i.e. in the structures that constitute the trans-Golgi network (TGN). In myotubes treated with brefeldin A, the immunofluorescence pattern of GLUT4 is modified, but it differs from both Golgi complex markers and TGN38. Instead, it resembles the pattern of the transferrin receptor, which forms long tubules. In untreated cells, double staining for GLUT4 and transferrin receptor by immunofluorescence shows similar but distinct patterns. Immunoelectron microscopy localizes transferrin receptor, detected by immunoperoxidase, to large vesicles, presumably endosomes, very close to the GLUT4-containing tubulovesicular elements. In brefeldin A-treated cells, a network of tubules of ∼70 nm diameter, studded with varicosities, stains for both GLUT4 and transferrin receptor, suggesting that brefeldin A has caused fusion of the transferrin receptor and GLUT4-containing compartments. The results suggest that GLUT4 storage vesicles constitute a specialized compartment that is either a subset of the TGN, or is very closely linked to it. The link between GLUT4 vesicles and transferrin receptor containing endosomes, as revealed by brefeldin A, may be important for GLUT4 translocation in response to muscle stimulation.

Stimulation by insulin results, in muscle and adipose tissue, in a large, acute increase in glucose uptake, which is mediated by the facilitative glucose transporter GLUT4, one member in a family of altogether six glucose transporter proteins (Bell et al., 1993). Upon insulin stimulation, GLUT4 appears to translocate from an intracellular storage compartment to the plasma membrane. This mechanism is supported by subcellular fractionation analysis of both adipose cells and muscle (for reviews see Birnbaum, 1992; Holman and Cushman, 1994) and by the detailed quantitative immunogold analysis of GLUT4 localization in rat adipose tissue (Slot et al., 1991). In skeletal muscle, stimulation by contractions also induces translocation of GLUT4 to the surface of the cell (Douen et al., 1989; Ploug et al., 1993).

Of the many questions that remain unanswered, to do with localization and trafficking of GLUT4 (James et al., 1994), one of the most debated is whether GLUT4 storage vesicles, in unstimulated cells, represent a separate, specific class of vesicles. Several other membrane proteins appear to be translocated to the plasma membrane following insulin stimulation, suggesting that they may be in the same vesicles. The transferrin receptor (TfR) is one of these proteins (Davis et al., 1986; Tanner and Lienhard, 1987), and its well-described endocytosis via coated vesicles and endosomes and rapid recycling to the plasma membrane (Klausner et al., 1983; Dautry-Varsat et al., 1983; Yamashiro et al., 1984) present many features in common with the trafficking of GLUT4. Several studies have addressed the possible colocalization of GLUT4 and TfR, without reaching a consensus: there have been reports that the two colocalize totally (Tanner and Lienhard, 1989), partially (Hudson et al., 1992, 1993; Laurie et al., 1993; Martin et al., 1994), or not at all (Herman et al., 1994; Aledo et al., 1995).

The lack of consensus may result from differences in the various cell types used in these studies, several of which were carried out on non-insulin-responsive cells transfected to express GLUT4, and others on different types of fat cells. Despite the predominant role of muscle in the pathophysiology of non-insulin-dependent diabetes (De Fronzo, 1992), little of this work has been carried out on muscle cells. In fact, it was believed for years that only the L6 muscle cell line expresses and translocates GLUT4 and that widely used muscle cell lines such as C2 and BC3H1 do not express detectable amounts of GLUT4 (Sargeant et al., 1993). Studies of GLUT4 localization in muscle, in vivo and in vitro, may help to explain the features of glucose transport that are common to insulin-responsive tissues, and those that are specific to muscle cells, whose geometry and subcellular architecture differ considerably from those of fat cells.

The goal of the present work is to identify the GLUT4 storage compartment in C2 myotubes and to determine its relationship to the subcellular organelles involved in the endosomal-lysosomal pathway. Using immunofluorescence and immunogold electron microscopy, we observe that the majority of GLUT4 is stored in vesicles that are intimately associated with the Golgi complex and do not contain TfR. We use the fungal metabolite brefeldin A (BfA) to further analyze these vesicles and highlight their functional connections to other organelles. We find that GLUT4 is dramatically redistributed in BfA-treated cells. Surprisingly, however, its pattern is now distinct from that of several markers of the Golgi complex and, instead, it colocalizes with the TfR. These results suggest that GLUT4 is stored in a specific subcompartment of the TGN, and has a functional, dynamic connection with the TfR-containing endosomes.

Antibodies and reagents

Primary antibodies are listed in Table 1. Anti-mouse transferrin receptor (TfR) was a gift from Dr Jayne Lesley (Salk Institute); anti-lgp110 from Dr Bruce Granger (Montana State University); anti-TGN38 from Dr Paul Luzio (Medical Research Council, Cambridge); anti-mannose-6-phosphate receptor/type II insulin-like growth factor receptor (M6PR) from Dr Peter Nissley (National Cancer Institute, NIH); anti-α-mannosidase II from Dr Kelley Moremen (University of Georgia); and anti-giantin from Dr Adam Linstedt (Carnegie Mellon Institute). Fluorescein- and rhodamine-conjugated secondary antibodies were purchased from Organon-Teknika (Malvern, PA) and biotinylated secondary antibodies and fluorescein- and Texas red-conjugated avidin were from Vector Laboratories (Burlingame, CA). Nanogold-conjugated anti-rabbit Fab fragments and silver enhancement kits (HQ) were from Nanoprobes Inc. (Stony Brook, NY). Brefeldin A was obtained from Epicentre Technologies (Madison, WI).

Table 1.

Antibodies used in this work

Antibodies used in this work
Antibodies used in this work

Cell cultures

C2 cultures (Yaffe and Saxel, 1977) were grown and maintained as described (Ralston and Hall, 1989), except that fusion medium contained 4% horse serum. Once in fusion medium, half of the medium was replaced daily. Myoblasts were plated on glass coverslips coated with a layer of carbon, followed by 0.1% gelatin (Daniels, 1990). Before fixation for staining, the medium was replaced with serum-free medium for 2 hours. In some experiments, BfA (1 mg/ml in ethanol) was added to the cultures to a final concentration of 10 µg/ml, 30 minutes before fixation.

Immunofluorescence

Cells were rinsed in phosphate-buffered saline (PBS) and fixed for 15 minutes with 4% formaldehyde from a 16% stock (Electron Microscopy Sciences, Ft. Washington, PA) diluted in calcium- and magnesium-free PBS. After blocking for 40 minutes in PBS containing 3% goat serum, 1% bovine serum albumin and 0.02% saponin (all from Sigma, St Louis, MO), cells were stained for 2 hours with the primary antibody in the same solution, rinsed three times with PBS containing 0.02% saponin, stained for 1 hour with the secondary antibody, washed again, and mounted in Vectashield (Vector Laboratories). For biotinylated secondary antibodies, there was an additional incubation for 20 minutes with Texas red or fluorescein-conjugated avidin in 20 mM Tris-HCl, 150 mM NaCl, 0.02% saponin at pH 8.5. Staining was observed with a Zeiss Axioskop fluorescence microscope or with a Leitz DMRD, both fitted with narrow bandpass filters. Confocal images were obtained on a Zeiss LSM 410 confocal microscope fitted with 488 nm and 568 nm krypton-argon lines, using a ×100 (NA 1.4) lens and zoom 2.5. The images were printed in Photoshop 3.0 on a Macintosh.

The results presented here are representative of several experiments. Staining with single markers was repeated at least three times; TfR-M6PR double staining, twice; and TfR-GLUT4 double staining, four times.

Immunogold electron microscopy

Cultures were rinsed with warm PBS and fixed with 4% formaldehyde at 37°C for 30 minutes, then at room temperature for 30 minutes, and finally at 4°C overnight. For cultures treated with BfA, the reagent was also added to the fixative. Permeabilization and staining were as described for immunofluorescence, except that the secondary antibody for GLUT4 was anti-rabbit Fab fragments conjugated to a 1.4 nm gold cluster and for TfR a biotinylated anti-rat IgG followed by ABC peroxidase (Vector Laboratories). Peroxidase was developed for 10 to 30 minutes with diaminobenzidine (DAB; 0.5 mg/ml) and H2O2 (0.5 µl of 30% stock per ml of DAB). Cells were postfixed with 1% glutaraldehyde for 15 minutes and silver-enhanced according to the manufacturer’s recommendations. Samples were osmicated with 0.2% OsO4, en bloc mordanted with uranyl acetate, dehydrated in graded ethanol and embedded in Epon. Glass coverslips were separated from the resin block by brief immersion in liquid nitrogen, and fragments of the blocks were sectioned en face. Sections of 70 and 200 nm, unstained, were observed with a Jeol 1200 microscope at 60 kV. The results presented here are representative of four separate experiments with double staining and additional experiments with single staining. In two experiments, duplicate coverslips were silver-enhanced for different times. Longer silver enhancement produces larger grains that are easier to visualize but tend to increase background and to hide underlying membranes. Controls with omission of the first or of the secondary antibody were included in each experiment.

Immunofluorescence localization of endogenous GLUT4 in muscle cultures

Fig. 1 shows permeabilized cultures of the mouse muscle cell line C2 stained with the anti-GLUT4 antibody P-1 at four developmental stages and observed by immunofluorescence.

Fig. 1.

Immunofluorescence localization of GLUT4 in C2 myoblasts (a), and in myotubes in fusion medium for 1 (b), 3 (c) and 6 days (d). Cultures were fixed and permeabilized as described in Materials and Methods and stained with the anti-GLUT4 antibody followed by fluorescein-conjugated anti-rabbit. Notice in (a) that most myoblasts show only a faint punctate staining (small arrow). Only occasional myoblasts (large arrow) show a stronger staining. The arrowheads in b-d point to the edge of the cell as seen in phase contrast. Most of the staining is restricted to the central core of the myotubes. Bar, 20 µm.

Fig. 1.

Immunofluorescence localization of GLUT4 in C2 myoblasts (a), and in myotubes in fusion medium for 1 (b), 3 (c) and 6 days (d). Cultures were fixed and permeabilized as described in Materials and Methods and stained with the anti-GLUT4 antibody followed by fluorescein-conjugated anti-rabbit. Notice in (a) that most myoblasts show only a faint punctate staining (small arrow). Only occasional myoblasts (large arrow) show a stronger staining. The arrowheads in b-d point to the edge of the cell as seen in phase contrast. Most of the staining is restricted to the central core of the myotubes. Bar, 20 µm.

Myoblasts (Fig. 1a, arrow) have low levels of GLUT4 (confirmed by immunoblotting, results not shown). Some show a staining of large aggregates, often at both poles of the nucleus, whereas most (small arrows) only show a fine punctate staining. In myotubes, the staining encircles the nuclei with an interrupted ring, and forms additional aggregates, often aligned along the axes of the myotubes. The position of these aggregates is not fixed in relation to the nuclei and their number and intensity varies along the myotubes, thus creating some variability in the pattern. As myotubes mature (from 1 to 6 days in fusion medium, Fig. 1b-d), the basic pattern remains unchanged. Most of the staining is concentrated in the central core of the myotubes. Only in the most mature cultures (Fig. 1d) can a fine punctate staining be detected towards and occasionally at the edge of the cell, which is marked by small arrowheads. Thus, GLUT4, in myotubes, appears to be efficiently excluded from the cell membrane, as has also been shown in other cells.

Although we only show results obtained with C2 myotubes, we also find GLUT4 in Sol 8 myotubes, in differentiated cultures of the non-fusing muscle cell line BC3H1, and in rat primary myotubes. The intensity and pattern of staining are roughly similar in all and comparable to what we observe in L6 myotubes. Thus GLUT4 is expressed as part of the normal myogenic development in all muscle cells.

GLUT4 is stored in the Golgi complex area

In an attempt to identify the GLUT4 storage compartment, we used a panel of antibodies, described in Table 1, that recognize markers of the Golgi complex, of the trans-Golgi network (TGN), of early and late endosomes, and of lysosomes. Since we have observed that TfR and GLUT4 are upregulated during development of C2 cultures, whereas α mannosidase II, M6PR and TGN38 are downregulated, all the experiments presented are with myotubes that have been in fusion medium for 3 to 4 days, by which time all markers can readily be detected. The staining patterns are presented in Fig. 2, with each panel showing a single representative myotube. GLUT4 (GT4) and lgp110 panels show the same myotube double-stained with the two antibodies. All other antibodies were applied individually. To some degree, all markers show some concentration around the nuclei but the GLUT4 resembles mannosidase (man) the most in pattern and location. Both form interrupted rings around the nuclei (small arrowheads) and present aggregates between the nuclei. In both cases, the staining is concentrated in the central core of the myotubes. Staining for TGN38 also shows large aggregates around and between nuclei, but superimposed on a finer punctate staining that covers the whole myotube. In contrast, staining for M6PR and lgp 110 gives a finer punctate staining, more scattered throughout the cell. The differences between distribution of GLUT4 and lgp110 are clear from the examination of the myotube shown here: there is little if any overlap in the two patterns and lgp110 staining is extensive in areas near the plasma membrane where there is no GLUT4 (large arrowhead in lgp110 panel). TfR is the only marker that shows a distinct staining of the plasma membrane (arrow). In the photograph shown here, focus was on the fraction of TfR that appears in the perinuclear area. The pattern appears finer, more continuous than that of GLUT4.

Fig. 2.

Comparison of GLUT4 staining to that of markers of various subcellular compartments in C2 myotubes. Cultures of C2 myotubes (3 days in fusion medium) were stained with antibodies to GLUT4, to α mannosidase II (man) as marker of the Golgi complex, to TGN38 as marker of the trans-Golgi network, to the transferrin receptor (TfR) as marker of the early endosomes, to the mannose-6-phosphate receptor (M6PR) as marker of the late endosomes and to lgp110 as marker of the lysosomes, followed by the appropriate fluorescein-or rhodamine-conjugated secondary antibody. Notice that most markers are concentrated in the perinuclear area (white arrowheads) and that only TfR gives a plasma membrane staining (arrow). GLUT4 and lgp110 staining were performed simultaneously and the same myotube is photographed under fluorescein (GT4) and rhodamine (lgp110) optics. Large arrowheads point to areas of strong staining in non-perinuclear. Bar, 20 µm.

Fig. 2.

Comparison of GLUT4 staining to that of markers of various subcellular compartments in C2 myotubes. Cultures of C2 myotubes (3 days in fusion medium) were stained with antibodies to GLUT4, to α mannosidase II (man) as marker of the Golgi complex, to TGN38 as marker of the trans-Golgi network, to the transferrin receptor (TfR) as marker of the early endosomes, to the mannose-6-phosphate receptor (M6PR) as marker of the late endosomes and to lgp110 as marker of the lysosomes, followed by the appropriate fluorescein-or rhodamine-conjugated secondary antibody. Notice that most markers are concentrated in the perinuclear area (white arrowheads) and that only TfR gives a plasma membrane staining (arrow). GLUT4 and lgp110 staining were performed simultaneously and the same myotube is photographed under fluorescein (GT4) and rhodamine (lgp110) optics. Large arrowheads point to areas of strong staining in non-perinuclear. Bar, 20 µm.

In order to further explore the link between GLUT4 and the Golgi complex, we double-stained C2 myotubes for GLUT4 and for giantin, a protein believed to be involved in the formation of crossbridges between cisternae of the Golgi complex (Linstedt and Hauri,1993). There is an excellent correlation between the two stainings (Fig. 3), but their simultaneous observation at high magnification (Fig. 4) reveals that they do not totally overlap.

Fig. 3.

Comparison between GLUT4 (G) and giantin (N) staining of C2 myotubes. Cultures were double-stained with rabbit anti-GLUT4 and mouse anti-giantin, followed by biotinylated anti-rabbit and Texas red-conjugated avidin, and fluorescein-conjugated anti-mouse. The same myotube is photographed with Texas red (G) and fluorescein (N) filters. At this level of resolution, there is a very good correspondence between the two stainings, as emphasized by arrowheads. Bar, 20 µm.

Fig. 3.

Comparison between GLUT4 (G) and giantin (N) staining of C2 myotubes. Cultures were double-stained with rabbit anti-GLUT4 and mouse anti-giantin, followed by biotinylated anti-rabbit and Texas red-conjugated avidin, and fluorescein-conjugated anti-mouse. The same myotube is photographed with Texas red (G) and fluorescein (N) filters. At this level of resolution, there is a very good correspondence between the two stainings, as emphasized by arrowheads. Bar, 20 µm.

Fig. 4.

Myotubes double-stained for GLUT4 and giantin as described in the legend to Fig. 3, but observed in the confocal microscope at high magnification. Although some of the staining appears yellow, indicating superposition of giantin and GLUT4, the colocalization is not complete. Many separate elements show layers of green, yellow and red (arrowheads). The insert shows part of the staining around a single nucleus. The GLUT4 staining (red) always overlaps partially with the giantin staining (green) and extends further away (in the trans direction) from the nucleus. Bars, 2.5 µm.

Fig. 4.

Myotubes double-stained for GLUT4 and giantin as described in the legend to Fig. 3, but observed in the confocal microscope at high magnification. Although some of the staining appears yellow, indicating superposition of giantin and GLUT4, the colocalization is not complete. Many separate elements show layers of green, yellow and red (arrowheads). The insert shows part of the staining around a single nucleus. The GLUT4 staining (red) always overlaps partially with the giantin staining (green) and extends further away (in the trans direction) from the nucleus. Bars, 2.5 µm.

Immunogold electron microscopy (EM) localization (Fig. 5) shows that most grains indicating the presence of GLUT4 are associated with Golgi complexes. The grains are concentrated on one side of the Golgi complex which, based on the distal location in relation to the nuclear membrane (Fig. 5b), must be the trans side (Tassin et al., 1985b). They are present in the last cisterna(e) of the Golgi stacks (arrowheads) and in vesicles beyond the cisternae. Such a pattern suggests that GLUT4 is concentrated in the TGN, which is generally defined as including the trans-most cisternae of the Golgi complex, and vesicles originating from these cisternae (Mellman and Simons, 1992; Ladinsky et al., 1994; Clermont et al., 1995). GLUT4 grains are also observed scattered in the cytoplasm, but their concentration there is close to background.

Fig. 5.

Electron micrograph showing immunogold staining for GLUT4 (see Materials and Methods). (a) In an area of the cell that is not near a nucleus, two ribbons of Golgi cisternae (arrows) appear twisted. The GLUT4 grains (arrowheads) are clearly on one side of the ribbon, and follow the course of the cisternae on the left, but appear to be mostly in vesicles on the right. (b) Three sets of Golgi cisternae, two of which are close to a nucleus (N), show heavy labeling of the trans-most cisternae (arrowhead) or of vesicles further beyond. Bars, 500 nm.

Fig. 5.

Electron micrograph showing immunogold staining for GLUT4 (see Materials and Methods). (a) In an area of the cell that is not near a nucleus, two ribbons of Golgi cisternae (arrows) appear twisted. The GLUT4 grains (arrowheads) are clearly on one side of the ribbon, and follow the course of the cisternae on the left, but appear to be mostly in vesicles on the right. (b) Three sets of Golgi cisternae, two of which are close to a nucleus (N), show heavy labeling of the trans-most cisternae (arrowhead) or of vesicles further beyond. Bars, 500 nm.

When C2 myotubes are stained for GLUT4 after 4 hours of treatment with 20 µg/ml of the protein synthesis inhibitor cycloheximide, there is no evident decrease in perinuclear immunofluorescent staining intensity compared to untreated cultures (not shown). Thus, localization to the Golgi complex is unlikely to result from post-translational transit of GLUT4, which is N-glycosylated (Mitsumoto and Klip, 1992).

GLUT4 is stored in a specific, distinct compartment

To confirm the TGN as the GLUT4 storage compartment, we decided to take advantage of the effects of the fungal metabolite BfA (reviewed by Klausner et al., 1992). Although the effects of BfA vary between cell lines, in most it affects the Golgi complex, the endosomes and the TGN differentially (see Discussion). If this is the case in C2 myotubes as well, BfA should provide some additional information regarding the GLUT4 storage compartment. C2 myotubes were treated for 30 minutes with 10 µM BfA, and stained with the panel of antibodies described previously. All staining patterns (Fig. 6), including that of GLUT4, are dramatically altered, except that for the lysosomal marker lgp110. Mannosidase staining, on the nuclear membrane and in a diffuse pattern throughout the cytoplasm, resembles the pattern of ER markers in myotubes, as expected for markers of the medial Golgi (Doms et al., 1989; Lippincott-Schwartz et al., 1989). Giantin staining is similar to mannosidase (not shown). TGN38, in contrast, presents a strong punctate pattern, also observed with another protein of the TGN, furin (Bosshart et al., 1994; not shown). But the GLUT4 pattern, unexpectedly, is quite distinct from that of mannosidase, giantin, and TGN38; it consists of dots and fine threads that actually resemble TfR and M6PR stainings (see arrowheads). The same pattern was observed after shorter (15 minutes) or longer (2 hours) BfA treatments (not shown).

Fig. 6.

Localization of markers of various subcellular compartments in BfA-treated C2 myotubes. The staining was identical to that described in Fig. 2, but cultures had been treated for 30 minutes with 10 µg/ml brefeldin A before fixation. All panels show different myotubes. Arrowheads emphasize thread-like staining of GLUT4 and of endosomal markers. Bar, 20 µm.

Fig. 6.

Localization of markers of various subcellular compartments in BfA-treated C2 myotubes. The staining was identical to that described in Fig. 2, but cultures had been treated for 30 minutes with 10 µg/ml brefeldin A before fixation. All panels show different myotubes. Arrowheads emphasize thread-like staining of GLUT4 and of endosomal markers. Bar, 20 µm.

Thus, if GLUT4 is stored in the TGN, it must be in a subcompartment that does not contain TGN38. Alternately, it may be in a membrane compartment geographically close to the TGN.

A connection between GLUT4 and TfR compartments is revealed by BfA

In order to clarify the relationship between GLUT4 and TfR storage vesicles in these myotubes, the cultures were doublestained for the two markers. In cultures without BfA, TfR and GLUT4 immunofluorescent stainings appear distinct (Fig. 7, top two panels), as suggested previously (Fig. 2). In BfAtreated cultures (Fig. 7, bottom two panels), TfR staining shows striking anastomosing tubules, generally aligned with the axis of the myotubes (see also Fig. 5). The GLUT4 pattern appears similar, though the tubules are generally shorter and often fuzzier. The longer of the GLUT4 tubules appear colocalized with the TfR staining (arrowheads). Colocalization is more difficult to assess for the shorter tubules. As a control, BfA-treated cultures were double-stained for TfR and M6PR (Fig. 8). In this case, extensive colocalization was observed after BfA treatment.

Fig. 7.

Simultaneous localization of TfR (T) and GLUT4 (G) in C2 myotubes, without (top two panels) or with BfA treatment (bottom two panels). Cultures similar to those described in Figs 5 and 6 were double stained with rabbit anti-GLUT4 and rat anti-TfR, followed by biotinylated anti-rabbit IgG and fluorescein-conjugated anti-rat IgG and by Texas red-conjugated avidin. Arrows point to features that stand out in one staining only, whereas arrowheads point to features that can be recognized in both. Bars, 20 µm.

Fig. 7.

Simultaneous localization of TfR (T) and GLUT4 (G) in C2 myotubes, without (top two panels) or with BfA treatment (bottom two panels). Cultures similar to those described in Figs 5 and 6 were double stained with rabbit anti-GLUT4 and rat anti-TfR, followed by biotinylated anti-rabbit IgG and fluorescein-conjugated anti-rat IgG and by Texas red-conjugated avidin. Arrows point to features that stand out in one staining only, whereas arrowheads point to features that can be recognized in both. Bars, 20 µm.

Fig. 8.

Simultaneous localization of TfR (T) and M6PR (M) in C2 myotubes treated with BfA. Cultures similar to those described in Fig. 7 were double-stained with rat anti-TfR and rabbit anti-M6PR, followed by fluorescein-conjugated anti-rat IgG (for TfR) and biotinylated anti-rabbit IgG, followed by Texas red avidin (for M6PR). Arrowheads point to features that can be recognized in both panels. Bar, 20 µm.

Fig. 8.

Simultaneous localization of TfR (T) and M6PR (M) in C2 myotubes treated with BfA. Cultures similar to those described in Fig. 7 were double-stained with rat anti-TfR and rabbit anti-M6PR, followed by fluorescein-conjugated anti-rat IgG (for TfR) and biotinylated anti-rabbit IgG, followed by Texas red avidin (for M6PR). Arrowheads point to features that can be recognized in both panels. Bar, 20 µm.

C2 myotubes double-stained for GLUT4 and TfR were then observed by electron microscopy. GLUT4 was detected with immunogold, and TfR with immunoperoxidase. In untreated myotubes (Fig. 9a-c), GLUT4 and TfR are often observed close to one another, but in separate membrane systems: TfR in large multivesicular bodies, and GLUT4 in tubulo-vesicular structures that are often next to recognizable Golgi stacks. TfR, as expected, is also found along the plasma membrane, in coated pits and in coated and uncoated vesicles (Fig. 9a). In this part of the cell, GLUT4 staining is found as single grains, at a density that is similar to that of the background, or in small aggregates of 2-4 grains, which do not colocalize with immunoperoxidase (not shown).

Fig. 9.

Electron micrographs showing TfR and GLUT4 staining in C2 myotubes. (a) Staining for TfR only, observed in a thin (∼70 nm) section. Arrows point to coated pits and the arrowhead to what appears to be a vesicle near a labeled section of plasma membrane. (b) A thicker section (∼200 nm), with immunogold grains indicating GLUT4 (arrows) in tubulovesicular structures very close to a large multivesicular endosome (E), labeled with immunoperoxidase (arrowhead), indicating TfR. (c) Another area, again showing a TfR-containing endosome (E, arrowhead), close to but distinct from immunogold labeling of GLUT4 (arrows), in a Golgi area (G) and in non-characteristic tubulo-vesicular elements close to a nucleus (N). Bars, 500 nm.

Fig. 9.

Electron micrographs showing TfR and GLUT4 staining in C2 myotubes. (a) Staining for TfR only, observed in a thin (∼70 nm) section. Arrows point to coated pits and the arrowhead to what appears to be a vesicle near a labeled section of plasma membrane. (b) A thicker section (∼200 nm), with immunogold grains indicating GLUT4 (arrows) in tubulovesicular structures very close to a large multivesicular endosome (E), labeled with immunoperoxidase (arrowhead), indicating TfR. (c) Another area, again showing a TfR-containing endosome (E, arrowhead), close to but distinct from immunogold labeling of GLUT4 (arrows), in a Golgi area (G) and in non-characteristic tubulo-vesicular elements close to a nucleus (N). Bars, 500 nm.

In BfA-treated cultures (Fig. 10), the patterns of both GLUT4 and TfR change strikingly. No large concentration of GLUT4 silver grains is observed near the nuclei or anywhere, and the grains are more scattered. Often they can be seen aligned on thin tubules of diameter ∼70 nm (Fig. 10a-d). TfR is found on the tubules as well, and on varicosities along the tubules. Following BfA treatment, the previously separate TfR and GLUT4 compartments now appear to have fused to form a network of tubules and still recognizable vesicles (the varicosities). In this network, diffusion of TfR and GLUT4 may not be completely free since they are not entirely superimposed (see Fig. 10c, for example).

Fig. 10.

Electron micrographs showing TfR and GLUT4 staining in C2 myotubes treated with BfA. (a-c) Thick (∼200 nm) sections; (d) thin (∼70 nm) section. Immunoperoxidase staining (arrows) and immunogold (arrowheads) are mostly on a network of tubules and varicosities. Bars, 500 nm.

Fig. 10.

Electron micrographs showing TfR and GLUT4 staining in C2 myotubes treated with BfA. (a-c) Thick (∼200 nm) sections; (d) thin (∼70 nm) section. Immunoperoxidase staining (arrows) and immunogold (arrowheads) are mostly on a network of tubules and varicosities. Bars, 500 nm.

GLUT4 vesicles are associated with microtubules

To determine if microtubules are involved in the effects of BfA on GLUT4 redistribution, cultures were treated with 1 µg/ml nocodazole for 2 hours. Staining with an antibody against α-tubulin confirmed that microtubules had disappeared, except for some very short fragments radiating from the nuclear membrane (not shown). The cultures appeared healthy. In nocodazole-treated cultures stained for GLUT4 (Fig. 11), the perinuclear staining can still be observed but it appears fragmented. Similarly, large aggregates between nuclei appear mostly fragmented and dispersed. When BfA is added to nocodazole-treated cultures (Fig. 11), large flat aggregates form, but no tubules can be found. Nocodazole also prevents the tubularization of TfR staining (not shown). Thus, GLUT4-containing vesicles must be associated with microtubules, along which the tubules observed after BfA treatment extend.

Fig. 11.

Microtubule depolymerization blocks BfA effects on GLUT4. C2 myotubes were treated with 1 µg/ml nocodazole (N) for 2 hours and then stained for GLUT4. In some cultures, BfA (10 µg/ml) was added for the last 30 minutes. (a) Control culture; (b) culture treated with nocodazole alone; (c) culture treated with nocodazole and BfA. Arrowheads point to fragmented pattern in nocodazole-treated cultures and to large flat aggregates in BfA and nocodazole-treated cultures. Bar, 20 µm.

Fig. 11.

Microtubule depolymerization blocks BfA effects on GLUT4. C2 myotubes were treated with 1 µg/ml nocodazole (N) for 2 hours and then stained for GLUT4. In some cultures, BfA (10 µg/ml) was added for the last 30 minutes. (a) Control culture; (b) culture treated with nocodazole alone; (c) culture treated with nocodazole and BfA. Arrowheads point to fragmented pattern in nocodazole-treated cultures and to large flat aggregates in BfA and nocodazole-treated cultures. Bar, 20 µm.

In this work, we have addressed two often debated questions that are key issues in understanding glucose transport: the nature and the specificity of the GLUT4 storage site. Because of the importance of muscle for insulin-stimulated glucose disposal (De Fronzo, 1992), we have choosen to examine GLUT4 in cultured myotubes, which express GLUT4 as part of their normal development, and respond to insulin stimulation (Ramlal et al., 1988; Galante et al., 1995). We conclude that GLUT4, in unstimulated muscle cells, is segregated from the transferrin receptor and from several available markers of the endosomal-lysosomal pathway and stored in a specific tubulo-vesicular compartment structurally resembling the TGN. But although GLUT4 and TfR are not colocalized, we find that there is a functional link between the two compartments, most likely via microtubules.

When the possible colocalization of GLUT4 and markers of early and late endosomes, Golgi complex, TGN and lysosomes is assessed by immunofluorescence (Fig. 2), the GLUT4 pattern differs from that of TfR but strikingly resembles that of Golgi complex markers. Double staining of myotubes for GLUT4 and giantin (Fig. 3) confirms a close association between GLUT4 and the Golgi complex, which we also observe in rat muscle fibers in vivo (T. Ploug et al., unpublished). Immunogold EM localization of GLUT4 (Fig. 5) shows staining of the trans-most Golgi cisterna(e) and of vesicles beyond, which together structurally define the TGN (Mellman and Simons, 1992; Ladinsky et al., 1994; Clermont et al., 1995). TGN localization of GLUT4 would explain that GLUT4 and giantin only partially overlap at high resolution (Fig. 4), since giantin is believed to be in the cisternal part of the Golgi complex (Linstedt and Hauri, 1993). The ability of confocal microscopy to resolve different compartments of the Golgi complex has been established by others (Antony et al., 1992; Nilsson et al., 1993).

In order to further explore the connection between GLUT4 and the TGN, we decided to take advantage of the effects of the fungal metabolite brefeldin A. Although the best known effects of BfA are its redistribution of Golgi complex markers to the ER (Doms et al., 1989; Lippincott-Schwartz et al., 1989), BfA has been observed to affect TGN, endosomes and lysosomes as well (Lippincott-Schwartz et al., 1991; Wood et al., 1991), fusing them into long tubules. We reasoned that the differential effect of BfA on the different compartments of the Golgi complex should provide additional evidence for localization of GLUT4 to one of them.

The effects of BfA are not universal, however, since some cells show resistance or a different reaction to the drug (Sandvig et al., 1991). There has not been any previous morphological study of the effects of BfA on organelles in muscle cells. It was thus necessary to first establish how C2 myotubes respond to BfA treatment. Except for lgp110, all markers examined (Fig. 6), including GLUT4, were dramatically redistributed in BfA-treated myotubes. Resistance of lysosomes to BfA has been noticed in some other cells (Prydz et al., 1992). For several markers, the changes were as expected from observations in other cells: redistribution of mannosidase to the ER, and tubularization of TfR and of M6PR endosomes. However, some features can be attributed to the unique geometry of multinucleated myotubes: even after long treatments with BfA, the staining for TfR or M6PR does not ‘collapse’ to the centrosome, as observed in mononucleated cells (LippincottSchwartz et al., 1991), most likely because of the inactivation of the centrosomes during myogenesis (Tassin et al., 1985a,b; Ralston, 1993). In contrast, the dispersion of TGN38 by BfA differs from the tubularization observed in fibroblast-like cells, but is similar to what is observed in neocortical neurons (Lowenstein et al., 1994). As to GLUT4, it is induced by BfA to form tubular structures resembling those of TfR but totally distinct from the markers of the Golgi proper or the TGN. This result was surprising in view of the strong EM evidence that GLUT4 is associated with the Golgi complex but it is consistent with a report (Martin et al., 1994) that there is very little, if any, colocalization of GLUT4 and TGN38 in subcellular fractions of 3T3-L1 adipocytes.

The simplest interpretation of our results is that GLUT4 is not in the TGN, since its pattern diverges from that of TGN38, but is geographically close to it. This interpretation is difficult to sustain in view of the EM pattern of GLUT4. However, the TGN is a complex structure, whose geometry and relative importance vary between cells (Clermont et al., 1995). A second interpretation is that GLUT4 is in a TGN subcompartment that does not contain TGN38. In favor of this interpretation is the observation that p200, another marker of the TGN, colocalizes only partially with TGN38 (Narula and Stow, 1995). In NRK cells, BfA causes tubularization of the TGN38 pattern (Lippincott-Schwartz et al., 1991) but dispersion of p200 (Narula et al., 1992). p200 itself may not be involved in transport of proteins to the cell surface (Ikonen et al., 1996), but there may still be other unknown markers of TGN subcompartments. Finally, we cannot exclude at this point the possibility that the vesicles labeled with TGN38 do not represent, in myotubes, the functional TGN, where sorting of proteins takes place before vesicles bud off to various destinations. TGN38 is downregulated during myogenesis, and is undetectable in muscle fibers in vivo (Jasmin et al., 1995). Therefore, GLUT4 might label the ‘real’ TGN, and its tubular pattern and connection with the TfR endosomes in BfA-treated cells would be easily explained. Clearly, more work will be required to solve this interesting question, which has implications beyond the immediate goals of the present paper.

Although GLUT4 staining in the Golgi complex region has been reported for brown adipose cells (Slot et al., 1991) and transfected L6 myoblasts (Haney et al., 1995), a close association has not been noticed. In EM sections, labeling is mostly of vesicles and the associated cisternae are not always visible (see Fig. 9, for example); and in light microscopy, the compact shape of the Golgi complex in mononucleated cells makes colocalization difficult to assess. In contrast, the unique architecture of the Golgi complex in myotubes, especially its fragmented appearance, facilitates the interpretation of colocalization experiments such as the giantin-GLUT4 staining (Fig. 3). In addition, the EM pre-embedding staining approach followed here provides a high efficiency of labeling and large sections to observe.

The demonstration that GLUT4 is redistributed following BfA treatment has not been made previously, although several studies have examined whether BfA perturbs glucose transport. BfA inhibited glucose transport in adipocytes in one study (Lachaal et al., 1994), but not in two others (Chakrabarti et al., 1994; Bao et al., 1995) or in L6 myotubes (Hundal et al., 1994). The density of GLUT4-containing fractions did not seem affected by BfA either (Hundal et al., 1994), leading to the interpretation that BfA has no effect on GLUT4 distribution. We show this not to be the case. A partial explanation of these apparently contradictory results is that BfA may change the distribution of a protein without affecting its recycling, as shown for TfR (Klausner et al., 1992).

We observe segregation of GLUT4 and TfR, both by immunofluorescence (Figs 2 and 7) and by immunogold EM (Fig. 9). In the central part of the myotubes, TfR is observed in large multivesicular bodies (Fig. 9b-c), presumably endosomes (van Deurs et al., 1993). GLUT4 EM immunogold staining is on tubulovesicular structures, most of them next to identifiable Golgi complexes. These tubulovesicular structures are often very close to the endosomes, sometimes to the point of appearing connected to them, but the staining for GLUT4 and TfR does not overlap. These results are especially important since it is the first time that GLUT4 and TfR have been observed simultaneously in whole cells at the EM level. Most previous studies of colocalization were based on fractionation studies, and the only immunofluorescence study (Hudson et al., 1992) was on transfected non-insulin-responsive cells. Although fractionation studies have undoubtedly been important in unraveling several aspects of glucose transport, it is not always possible to determine where in the intact cells the fractions originate from. This is especially true when traditional markers are not present in these fractions, as is the case for GLUT4 vesicles, which were recently suggested to be free vesicles (Kandror et al., 1995). The present work suggests, in contrast, that these vesicles may originate from the TGN. The distinction is important for our understanding of GLUT4 trafficking.

Finally, the demonstration, by BfA, of an underlying connection between the separate TfR and GLUT4 compartments provides a basis for reconciliation of the apparently contradictory observations of the coordinate translocation of TfR and GLUT4 following insulin stimulation, and of their segregation in different compartments.

We thank all those, mentioned in the text, who generously provided the antibodies. The EM work greatly benefited from the expert assistance from Dr Tao Cheng, Ms Virginia Tanner and Ms Pat Zerfas at the NINDS EM Facility and the confocal imaging from Dr Carolyn Smith’s help at the NINDS Light Imaging Facility. We are grateful to Ms Christine Winters for help with the tissue culture. We thank Drs T. S. Reese and S. W. Cushman for support through this work and Drs B. van Deurs and Julie Donaldson for reading of the manuscript. T. Ploug was supported by the Danish National Research Foundation and by a fellowship from the Weimann Foundation, Denmark, and we acknowledge a NATO Collaborative Research Grant.

Aledo
,
J. C.
,
Darakhshan
,
F.
and
Hundal
,
H. S.
(
1995
).
Rab4, but not the transferrin receptor, is colocalized with GLUT4 in an insulin-sensitive intracellular compartment in rat skeletal muscle
.
Biochem. Biophys. Res. Commun
.
215
,
321
328
.
Antony
,
C.
,
Cibert
,
C.
,
Géraud
,
G.
,
Santa Maria
,
A.
,
Maro
,
B.
,
Mayau
,
V.
and
Goud
,
B.
(
1992
).
The small GTP-binding protein rab6p is distributed from medial Golgi to the trans-Golgi network as determined by a confocal microscopic approach
.
J. Cell Sci
.
103
,
785
796
.
Bao
,
S.
,
Smith
,
R. M.
,
Jarett
,
L.
and
Garvey
,
W. T.
(
1995
).
The effects of brefeldin A on the glucose transport system in rat adipocytes. Implications regarding the intracellular locus of insulin-sensitive Glut4
.
J. Biol. Chem
.
270
,
30199
30204
.
Bell
,
G. I.
,
Burant
,
C. F.
,
Takeda
,
J.
and
Gould
,
G. W.
(
1993
).
Structure and function of mammalian facilitative sugar transporters
.
J. Biol. Chem
.
268
,
19161
19164
.
Birnbaum
,
M. J.
(
1992
).
The insulin-sensitive glucose transporter
.
Int. Rev. Cytol
.
137A
,
239
297
.
Bosshart
,
H.
,
Humphrey
,
J.
,
Deignan
,
E.
,
Davidson
,
J.
,
Drazba
,
J.
,
Yuan
,
L.
,
Oorschot
,
V.
,
Peters
,
P. J.
and
Bonifacino
,
J. S.
(
1994
).
The cytoplasmic domain mediates localization of furin to the trans-Golgi network en route to the endosomal/lysosomal system
.
J. Cell Biol
.
126
,
11571172
.
Chakrabarti
,
R.
,
Buxton
,
J.
,
Joly
,
M.
and
Corvera
,
S.
(
1994
).
Insulinsensitive association of GLUT4 with endocytic clathrin-coated vesicles revealed with the use of brefeldin A
.
J. Biol. Chem
.
269
,
7926
7933
.
Clermont
,
Y.
,
Rambourg
,
A.
and
Hermo
,
L.
(
1995
).
Trans-Golgi network (TGN) of different cell types: three-dimensional structural characteristics and variability
.
Anat. Rec
.
242
,
289
301
.
Daniels
,
M.
(
1990
).
Localization of actin, beta-spectrin, 43×103 Mr and 58×103 Mr proteins to receptor-enriched domains of newly formed acetylcholine receptor aggregates in isolated myotube membranes
.
J.Cell Sci
.
97
,
615
627
.
Dautry-Varsat
,
A.
,
Ciechanover
,
A.
and
Lodish
,
H. F.
(
1983
).
pH and the recycling of transferrin during receptor-mediated endocytosis
.
Proc. Nat. Acad. Sci. USA
80
,
2258
2262
.
Davis
,
R. J.
,
Corvera
,
S.
and
Czech
,
M. P.
(
1986
).
Insulin stimulates cellular iron uptake and causes the redistribution of intracellular transferrin receptors to the plasma membrane
.
J. Biol. Chem
.
261
,
8708
8711
.
De Fronzo
,
R. A.
(
1992
).
Pathogenesis of NIDDM: a balanced overview
.
Diabetes Care
15
,
318
368
.
Doms
,
R. W.
,
Russ
,
G.
and
Yewdell
,
J. W.
(
1989
).
Brefeldin A redistributes resident and itinerant Golgi proteins to the endoplasmic reticulum
.
J. Cell Biol
.
109
,
61
72
.
Douen
,
A. G.
,
Ramlal
,
T.
,
Klip
,
A.
,
Young
,
D. A.
,
Cartee
,
G. D.
and
Holloszy
,
J. O.
(
1989
).
Exercise-induced increase in glucose transporters in plasma membranes of rat skeletal muscle
.
Endocrinology
124
,
449
54
.
Galante
,
P.
,
Mosthaf
,
L.
,
Kellerer
,
M.
,
Berti
,
L.
,
Tippmer
,
S.
,
Bossenmaier
,
B.
,
Fujiwara
,
T.
,
Okuno
,
A.
,
Horikoshi
,
H.
and
Häring
H. U.
(
1995
).
Acute hyperglycemia provides an insulin-independent inducer for GLUT4 translocation in C2C12 myotubes and rat skeletal muscle
.
Diabetes
44
,
64653
.
Granger
,
B. L.
,
Green
,
S. A.
,
Gabel
,
C. A.
,
Howe
,
C. L.
,
Mellman
,
I.
and
Helenius
,
A.
(
1990
).
Characterization and cloning of lgp110, a lysosomal membrane glycoprotein from mouse and rat cells
.
J. Biol. Chem
.
265
,
1203643
.
Haney
,
P. M.
,
Aach Levy
,
M.
,
Strube
,
M. S.
and
Mueckler
,
M.
(
1995
).
Insulin-sensitive targeting of the GLUT4 glucose transporter in L6 myoblasts is conferred by its COOH-terminal cytoplasmic tail
.
J. Cell Biol
.
129
,
641658
.
Herman
,
G. A.
,
Bonzelius
,
F.
,
Cieutat
,
A.-M.
and
Kelly
,
R. B.
(
1994
).
Evidence for a class of intracellular storage vesicles, identified by expression of the glucose transporter GLUT4
.
Proc. Nat. Acad. Sci. USA
91
,
1275012754
.
Holman
,
G. D.
and
Cushman
,
S. W.
(
1994
).
Subcellular localization and trafficking of the GLUT4 glucose transporter isoform in insulin-responsive cells
.
BioEssays
16
,
753
9
.
Hudson
,
A. W.
,
Ruiz
,
M.
and
Birnbaum
,
M. J.
(
1992
).
Isoform-specific subcellular targeting of glucose transporter in mouse fibroblast
.
J. Cell Biol
.
116
,
785
797
.
Hudson
,
A. W.
,
Fingar
,
D. C.
,
Seidner
,
G. A.
,
Griffiths
,
G.
,
Burke
,
B.
and
Birnbaum
,
M. J.
(
1993
).
Targeting of the insulin-responsive glucose transporter (GLUT4) to the regulated secretory pathway in PC12 cells
J. Cell Biol
.
122
,
579
-
588
(with erratum in J. Cell Biol. 1993 122, following 1143).
Hundal
,
H. S.
,
Bilan
,
P. J.
,
Tsakiridis
,
T.
,
Marette
,
A.
and
Klip
,
A.
(
1994
).
Structural disruption of the trans-Golgi network does not interfere with the acute stimulation of glucose and amino acid uptake by insulin-like growth factor I in muscle cells
.
Biochem. J
.
297
,
289
295
.
Ikonen
,
I.
,
Parton
,
R. G.
,
Lafont
,
F.
and
Simons
,
K.
(
1996
).
Analysis of the role of p200-containing vesicles in post-Golgi traffic
.
Mol. Biol. Cell
.
7
,
961974
.
James
,
D. E.
,
Piper
,
R. C.
and
Slot
,
J. W.
(
1994
).
Insulin stimulation of GLUT4 translocation: a model for regulated recycling
.
Trends Cell Biol
.
4
,
120
126
.
Jasmin
,
B. J.
,
Antony
,
C.
,
Changeux
,
J.-P.
and
Cartaud
,
J.
(
1995
).
Nervedependent plasticity of the Golgi complex in skeletal muscle fibres: compartmentalization within the subneural sarcoplasm
.
Eur. J. Neurosci
.
7
,
470
479
.
Kandror
,
K. V.
,
Coderre
,
L.
,
Pushkin
,
A. V.
and
Pilch
,
P. F.
(
1995
).
Comparison of glucose-transporter-containing vesicles from rat fat and muscle tissues: evidence for a unique endosomal compartment
.
Biochem. J
.
307
,
383
90
.
Kiess
,
W.
,
Haskell
,
J. F.
,
Lee
,
L.
,
Greenstein
,
L. A.
,
Miller
,
B. E.
,
Aarons
,
A. L.
,
Rechler
,
M. M.
and
Nissley
,
S. P.
(
1987
).
An antibody that blocks insulin-like growth factor (IGF) binding to the type II IGF receptor is neither an agonist nor an inhibitor of IGF-stimulated biologic responses in L6 myoblasts
.
J. Biol. Chem
.
262
,
12745
12751
.
Klausner
,
R. D.
,
Ashwell
,
G.
,
Van Renswoude
,
J.
,
Harford
,
J. B.
and
Bridges
,
K. R.
(
1983
).
Binding of apotransferrin to K562 cells: explanation of the transferrin cycle
.
Proc. Nat. Acad. Sci. USA
80
,
2263
2266
.
Klausner
,
R. D.
,
Donaldson
,
J. G.
and
Lippincott-Schwartz
,
S. J.
(
1992
).
Brefeldin A: insights into the control of membrane traffic and organelle structure
.
J. Cell Biol
.
116
,
1071
80
.
Lachaal
,
M.
,
Moronski
,
C.
,
Liu
,
H.
and
Jung
,
C. Y.
(
1994
).
Brefeldin A inhibits insulin-induced glucose transport stimulation and GLUT4 recruitment in rat adipocytes
.
J. Biol. Chem
.
269
,
23689
23693
.
Ladinsky
,
M. S.
,
Kremer
,
J. R.
,
Furcinitti
,
P. S.
,
McIntosh
,
J. R.
and
Howell
,
K. E.
(
1994
).
HVEM Tomography of the trans-Golgi network: structural insights and identification of a lace-like vesicle coat
.
J. Cell Biol
.
127
,
29
38
.
Laurie
,
S. M.
,
Cain
,
C. C.
,
Lienhard
,
G. E.
and
Castle
,
J. D.
(
1993
).
The glucose transporter Glut4 and secretory carrier membrane proteins(SCAMPs) colocalize in rat adipocytes and partially segregate during insulin stimulation
.
J. Biol.Chem
.
268
,
19110
19117
.
Lesley
,
J.
,
Shulte
,
R.
and
Woods
,
J.
(
1989
).
Modulation of transferrin receptor expression and function by anti-transferrin receptor antibodies and antibody fragments
.
Exp. Cell Res
.
182
,
215
233
.
Linstedt
,
A.
and
Hauri
,
H.-P.
(
1993
).
Giantin, a novel conserved Golgi membrane protein containing a cytoplasmic domain of at least 350 kDa
.
Mol. Biol. Cell
4
,
679
693
.
Lippincott-Schwartz
,
J.
,
Yuan
,
L. C.
,
Bonifacino
,
J. S.
and
Klausner
,
R. D.
(
1989
).
Rapid redistribution of Golgi proteins into the ER in cells treated with brefeldin A: evidence for membrane cycling from Golgi to ER
.
Cell
56
,
801813
.
Lippincott-Schwartz
,
J.
,
Yuan
,
L. C.
,
Tipper
,
C.
,
Amherdt
,
M.
,
Orci
,
L.
and
Klausner
,
R. D.
(
1991
).
Brefeldin A’s effects on endosomes, lysosomes and the TGN suggest a general mechanism for regulating organelle structure and membrane traffic
.
Cell
67
,
601
616
.
Lowenstein
,
J. R.
,
Morrison
,
E. E.
,
Bain
,
D.
,
Shering
,
A. F.
,
Banting
,
S. G.
,
Douglas
,
P.
and
Castro
,
M. G.
(
1994
).
Polarized distribution of the transGolgi network marker TGN38 during the in vitro development of neocortical neurons: effects of nocodazole and brefeldin A
.
Eur. J. Neurosci
.
6
,
14531465
.
Luzio
,
J. P.
,
Brake
,
B.
,
Banting
,
G.
,
Howell
,
K.
,
Braghetta
,
P.
and
Stanley
,
K. K.
(
1990
).
Identification, sequencing and expression of an integral membrane protein of the trans-Golgi network (TGN38)
.
Biochem. J
.
270
,
97102
.
Martin
,
S.
,
Reaves
,
B.
,
Banting
,
G.
and
Gould
,
G. W.
(
1994
).
Analysis of the co-localization of the insulin-responsive glucose transporter (GLUT4) and the trans-Golgi network marker TGN38 within 3T3-L1 adipocytes
.
J. Biochem
.
300
,
743
749
.
Mellman
,
I.
and
Simons
,
K.
(
1992
).
The Golgi complex: in vitro veritas?
Cell
68
,
829
840
.
Mitsumoto
,
Y.
and
Klip
,
A.
(
1992
).
Developmental regulation of the subcellular distribution and glycosylation of GLUT1 and GLUT4 glucose transporters during myogenesis of L6 muscle cells
.
J. Biol. Chem
.
267
,
49574962
.
Moremen
,
K.
and
Touster
,
O.
(
1985
).
Biosynthesis and modification of Golgi mannosidase II in HeLa and 3T3 cells
.
J. Biol. Chem
.
260
,
6654
6662
.
Narula
,
N.
,
McMorrow
,
I.
,
Plopper
,
G.
,
Doherty
,
J.
,
Matlin
,
K.S.
,
Burke
,
B.
and
Stow
,
J. L.
(
1992
).
Identification of a 200-kD, Brefeldin-sensitive protein on Golgi membranes
.
J. Cell Biol
.
117
,
27
38
.
Narula
,
N.
and
Stow
,
J. L.
(
1995
).
Distinct coated vesicles labeled for p200 bud from trans-Golgi network membranes
.
Proc. Nat. Acad. Sci. USA
92
,
2874
2878
.
Nilsson
,
T.
,
Pypaert
,
M.
,
Hoe
,
M. H.
,
Slusarewicz
,
P.
,
Berger
,
E. G.
and
Warren
,
G.
(
1993
).
Overlapping distribution of two glycosyltransferases in the Golgi apparatus of HeLa cells
.
J. Cell Biol
.
120
,
5
13
.
Ploug
,
T.
,
Stallknecht
,
B. M.
,
Pedersen
,
O.
,
Kahn
,
B. B.
,
Ohkuwa
,
T.
,
Vinten
,
J.
and
Galbo
,
H.
(
1990
).
Effect of endurance training on glucose transport capacity and glucose transporter expression in rat skeletal muscle
.
Am. J. Physiol
.
259
,
E778
786
.
Ploug
,
T.
,
Wojtaszewski
,
J.
,
Kristiansen
,
S.
,
Hespel
,
P.
,
Galbo
,
H.
and
Richter
,
E. A.
(
1993
).
Glucose transport and transporters in muscle giant vesicles: differential effects of insulin and contractions
.
Am. J. Physiol
.
264
,
E270
278
.
Prydz
,
K.
,
Hansen
,
S. H.
,
Sandvig
,
K.
and
van Deurs
,
B.
(
1992
).
Effects of brefeldin A on endocytosis, transcytosis and transport to the Golgi complex in polarized MDCK cells
.
J. Cell Biol
.
119
,
259
272
.
Ralston
,
E.
(
1993
).
Changes in architecture of the Golgi complex and other subcellular organelles during myogenesis
.
J. Cell Biol
.
120
,
399
409
.
Ralston
,
E.
and
Hall
,
Z. W.
(
1989
).
Transfer of a protein encoded by a single nucleus to nearby nuclei in multinucleated myotubes
.
Science
244
,
10661069
.
Ramlal
,
T.
,
Sarabia
,
V.
,
Bilan
,
P. J.
and
Klip
,
A.
(
1988
).
Insulin-mediated translocation of glucose transporters from intracellular membranes to plasma membrane: sole mechanism of stimulation of glucose transport in L6 muscle cells
.
Biochem. Biophys. Res. Commun
.
157
,
1329
1335
.
Sandvig
,
K.
,
Prydz
,
K.
,
Hansen
,
S. H.
and
van Deurs
,
B.
(
1991
).
Ricin transport in Brefeldin A-treated cells: correlation between Golgi structure and toxic effect
.
J. Cell Biol
.
115
,
971
981
.
Sargeant
,
R.
,
Mitsumoto
,
Y.
,
Sarabia
,
V.
,
Shillabeer
,
G.
and
Klip
,
A.
(
1993
).
Hormonal regulation of glucose transporters in muscle cells in culture
.
J. Endocrinol. Invest
.
16
,
147
162
.
Slot
,
J. W.
,
Geuze
,
H. J.
,
Gigengack
,
S.
,
Lienhard
,
G. E.
and
James
,
D. E.
(
1991
).
Immuno-localization of the insulin regulatable glucose transporter in brown adipose tissue of the rat
.
J. Cell Biol
.
113
,
123
135
.
Tanner
,
L. I.
and
Lienhard
,
G. E.
(
1987
).
Insulin elicits a redistribution of transferrin receptors in 3T3-L1 adipocytes through an increase in the rate constant for receptor externalization
.
J. Biol. Chem
.
262
,
8975
8980
.
Tanner
,
L. I.
and
Lienhard
,
G. E.
(
1989
).
Localization of transferrin receptors and insulin-like growth factor II receptors in vesicles from 3T3-L1 adipocytes that contain intracellular glucose transporters
.
J. Cell Biol
.
108
,
1537
1545
.
Tassin
,
A. M.
,
Maro
,
B.
and
Bornens
,
M.
(
1985a
).
Fate of microtubule organizing centers during in vitro myogenesis
.
J. Cell Biol
.
100
,
35
46
.
Tassin
,
A. M.
,
Paintrand
,
M.
,
Berger
,
E. G.
and
Bornens
,
M.
(
1985b
).
The Golgi apparatus remains associated with microtubule organizing centers during myogenesis
.
J. Cell Biol
.
101
,
630
638
.
van Deurs
,
B.
,
Holm
,
P. K.
,
Kayser
,
L.
,
Sandvig
,
K.
and
Hansen
,
S. H.
(
1993
)
Multivesicular bodies in HEp-2 cells are maturing endosomes
.
Eur. J. Cell Biol
.
61
,
208
224
.
Wood
,
S. A.
,
Park
,
J. E.
and
Brown
,
W. J.
(
1991
).
Brefeldin A causes a microtubule-mediated fusion of the trans-Golgi network and early endosomes
.
Cell
67
,
591
600
.
Yaffe
,
D.
and
Saxel
,
O.
(
1977
).
Serial passaging and differentiation of myogenic cells isolated from dystrophic mouse muscle
.
Nature
270
,
725727
.
Yamashiro
,
D. J.
,
Tycko
,
B.
,
Fluss
,
S. R.
and
Maxfield
,
F. R.
(
1984
).
Segregation of transferrin to a mildly acidic (pH 6.5) para-Golgi compartment in the recycling pathway
.
J. Cell Biol
.
37
,
789
800
.