![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 17, Issue 12, 5153-5162, December 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Institute of Physiology, University of Zurich, CH-8057 Zurich, Switzerland
Submitted August 2, 2006;
Revised September 13, 2006;
Accepted September 27, 2006
Monitoring Editor: Vivek Malhotra
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1,4-galactosyltransferase 1 (galT) and 
2,6-sialyltransferase 1 (siaT), two trans-Golgi glycosyltransferases, with respect to their different pathways in monensin-treated cells. Upon addition of monensin galT dissociates from siaT and the GA and accumulates in swollen vesicles derived from the trans-Golgi network (TGN), as shown by colocalization with TGN46, a specific TGN marker. We analyzed various chimeric constructs of galT and siaT by confocal fluorescence microscopy and time-lapse videomicroscopy as well as Optiprep density gradient fractionation. We show that the first 13 amino acids of the cytoplasmic tail of galT are necessary for its localization to swollen vesicles induced by monensin. We also show that the monensin sensitivity resulting from the cytoplasmic tail can be conferred to siaT, which leads to the rapid accumulation of the galT–siaT chimera in swollen vesicles upon monensin treatment. On the basis of these data, we suggest that cycling between the trans-Golgi cisterna and the trans-Golgi network of galT is signal mediated. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
), interest focused on the molecular organization of the glycosylation machinery. The key mediators of glycosylation are the glycosyltransferases, which are type 2 membrane proteins (Paulson and Colley, 1989). Those involved in N-glycosylation of proteins (Rabouille et al., 1995) appear to be arranged from the cis to the trans face of the GA like an assembly line, as proposed two decades ago (Berger, 1985). One of the best studied Golgi-associated glycosyltransferases is 1,4-galactosyltransferase 1 (galT), which traditionally has been used as Golgi marker enzyme in fractionation studies (Bretz et al., 1980 and references cited therein). Early work on immunolocalization of galT determined its steady state localization to the trans side of the GA and its exclusion from the trans-most cisterna, which can be considered as part of the TGN (Roth and Berger, 1982; Griffiths and Simons, 1986); this confinement to the trans side (Rabouille et al., 1995; Kweon et al., 2004) has to our knowledge not been called into question. However, intracellular trafficking of galT and its retention specifically to trans located cisternae is insufficiently understood. The enzyme follows the secretory pathway as determined by pulse-chase experiments (Strous and Berger, 1982) and is somehow retained in the trans cisternae, whereas a small fraction of galT moves, in defined cell types, to the plasma membrane (Hathaway et al., 2003) and possibly leaves the cell by shedding (for review see Strous, 1986). The site and mechanisms of cleavage of galT are unknown. More recently, an additional pathway of trafficking has been proposed: GFP-tagged forms of galT appear to recycle to the endoplasmic reticulum (ER; Sciaky et al., 1997; Storrie, 2005). This retrograde transport was postulated to be mediated by tubules carrying galT-GFP back to the ER, a process much enhanced by brefeldin A (BFA; Lippincott-Schwartz et al., 1990). In no instances galT-GFP was found in structures near or at the plasma membrane. Therefore, post-Golgi trafficking of this enzyme remains mysterious because at least two different routes (transport to the cell surface and secretion by shedding as well as recycling to the ER) have been described. Post-Golgi trafficking intermediates containing galT have indeed not been characterized.
Golgi disturbing agents such as BFA and monensin have been useful to dissect some of the mechanisms involved in transport through the GA (for review see Dinter and Berger, 1998). Monensin, a cationophore exchanging luminal H+ for Na+, leads to the formation of swollen vesicles mainly in post-Golgi/endosomal compartments by Na+-driven water influx (Zhang et al., 1993; for review see Mollenhauer et al., 1990). Surprisingly, galT was found by immunoelectron microscopy to decorate the membranes of these swollen vesicles (Strous et al., 1985), an effect interpreted as swelling of trans-Golgi cisternae along the lines of the then widely accepted view of monensin being a Golgi transport disruptor (Tartakoff, 1983). However, 2,6-sialyltransferase 1 (siaT), a trans-Golgi enzyme that colocalizes with galT (Taatjes et al., 1987; Kweon et al., 2004), dissociates from galT in monensin-treated cells (Berger et al., 1993), leading to the notion of displacement of galT to TGN-derived swollen vesicles, where galT colocalizes with the cation insensitive mannose-6-phosphate receptor (Berger et al., 2001).
The present work was carried out to demonstrate monensin-induced dissociation of galT from siaT by live microscopy and its colocalization with the authentic TGN marker TGN46 by double immunofluorescence to swollen vesicles and to quantify the morphological data by subcellular fractionation. Moreover, we show that this displacement is mediated by 13 N-terminal amino acids of galT and is necessary and in the context of the entire cytosolic tail sufficient to mediate anterograde transport of glycosyltransferases from the trans-Golgi cisterna to the TGN.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The mAb to galT was galT#2/36/118 (Berger et al., 1986), the polyclonal antibodies to galT were described in Watzele et al. (1991) and to siaT in Berger et al. (1993).
Recombinant DNA
All basic DNA procedures were as described (Sambrook et al., 1998). The PCR procedure of Ho et al. (1989) was used to generate the mutants and the galT–siaT fusion chimeras as described (Rohrer and Kornfeld, 2001). The final PCR products were subcloned into pcDNA3.1(–) (Invitrogen) as described (Rohrer et al., 1995). In brief, galT-GFP and siaT-GFP were produced by PCR using pUC18-galT (Watzele and Berger, 1990) and pIC20H-siaT (Berger et al., 1993) as templates with respective oligos (list of oligos used can be found in Supplementary Material) containing restriction sites NotI and BglII. The PCR products were subsequently digested for 2 h at 37°C using NotI and BglII. BglII and HindIII restriction sites were introduced into GFP using the same method, and the resulting PCR product was digested. The vector pcDNA 3.1(–) was cut using NotI/HindIII. Subsequently the fragments were analyzed and purified by agarose gel electrophoresis. The three fragments (galT or siaT, GFP, and vector) were extracted from the agarose gel using the QIAEX II kit form Qiagen (Valencia, CA) and were assembled in an overnight ligation at 15°C.
To create chimeras of galT and siaT, two separate PCR fragments of the two parts to be fused together were generated using galT-GFP or siaT-GFP as templates. The fragments were analyzed and purified by agarose gel electrophoresis. In a second PCR these purified fragments were merged together using outer oligos (Ho et al., 1989). The resulting fragment was subcloned into pcDNA 3.1(–), as described for galT-GFP and siaT-GFP above. The tail mutants were created similarly, using galT-GFP as template and overlapping oligos containing the mutations (Ho et al., 1989).
Cell Culture and Transfection
HepG2 cells were grown in DMEM supplemented with 10% FCS to 20–30% confluency in six-well plates before transfection with 1.4 µg of linearized plasmid DNA combined with 2.2 µg PEI according to the protocol from Polyplus-Transfection (Illkirch, France). Selection for resistance to neomycin (G418) was carried out using 1 mg/ml G418 as the final concentration. Resistant colonies were picked individually and screened for the expression of the various GFP-chimeras by fluorescence analysis, or several colonies were pooled to give a mixture of cells expressing the construct at various levels.
Confocal Immunofluorescence Microscopy
Confocal immunofluorescence microscopy was carried out as described (Rohrer and Kornfeld, 2001). HepG2 cells were grown to 30–50% confluency on coverslips in DMEM + 10% FCS. The cells were then incubated for 30 min in DMEM + 10% FCS containing 2 µM monensin before fixation with 3% paraformaldehyde. After fixation, the cells were permeabilized using 0.1% saponin in PBS for 20 min, incubated with suitably diluted primary antibody in 0.1% saponin in PBS for 30 min, then washed in 0.1% saponin in PBS three times and incubated with secondary antibodies in 0.1% saponin in PBS, and finally washed three times in PBS. The coverslips were mounted on glass slides with ProLong Antifade (Invitrogen) for viewing with a Leica TCS SP2 AOBS (Leica, Wetzlar, Germany) confocal laser-scanning microscope. The resulting stacks of images were exported as Tiff files and analyzed with the Imaris program (Bitplane).
Live Confocal Immunofluorescence Microscopy and Deconvolution
Live confocal immunofluorescence microscopy was carried out on a Leica SP2 microscope equipped with an incubation chamber (Life Imaging Services, Reinach, Switzerland). They were recorded at an xy-resolution of 512 pixels and 4–6 z-slices. The 4D stacks were processed using AutoDeblur (Media Cybernetics, Silver Spring, MD) and compiled using the Imaris Software (Bitplane).
Optiprep Density Gradient
Optiprep density gradients were performed as described in the manufacturer's protocol (Axis Shield, Roskilde, Denmark; protocol S36). Briefly, HepG2 cells were grown to 70–80% confluency on 10-cm dishes and incubated for 30 min in DMEM + 10% FCS containing 2 µM monensin before harvest by a rubber policeman and collection in 2 ml homogenization buffer (0.25 M sucrose, 1 mM EDTA, 10 mM HEPES-NaOH, pH 7.4). Cells were then broken using 15 strokes in a ball bearing homogenizer (Balch and Rothman, 1985) with a clearance of 16 µm, followed by removal of nuclei by low-speed centrifugation (800 x g). The homogenate (0.8 ml) was then loaded on top of a step gradient containing 2.5–30% Optiprep (final volume of 9.2 ml: 2.5%, 0.8 ml; 5%, 1.3 ml; 7.5%, 1.3 ml; 10%, 1.3 ml; 12.5%, 0.8 ml; 15%, 1.3 ml; 17.5%, 0.8 ml; 20%, 0.8 ml; and 30%, 0.8 ml) and centrifuged for 2.5 h at 200,000 x gav in the Sorvall TH641 swing-out rotor. Ten 1-ml fractions were collected from the bottom by pinching a hole in the tube and letting the gradient slowly drop into the collecting tubes. The resulting fractions were analyzed by Western blotting techniques using mAb against GFP and secondary antibody anti-mouse conjugated with horse-radish peroxidase.
SDS-PAGE and Immunoblotting
Proteins were separated on 10% SDS-polyacrylamide minigels (Bio-Rad, Hercules, CA) by using the Laemmli system (Laemmli, 1970). After electrophoresis the proteins were transferred from the gels onto nitrocellulose membranes according to the method of Towbin et al. (1979). The immunoblotting was performed as previously described (Rohrer et al., 1995). The chemoluminescence was recorded by using a quantitative densitometer from Raytest (Straubenhardt, Germany), and signals were quantified by calculating the total fluorescence of each band and correcting it by subtracting the background fluorescence using the Aida Software (Straubenhardt, Germany).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Subsequently, this monensin-induced dissociation was reproduced in CHO cells transfected with both galT and siaT (Berger et al., 2001), indicating a general rather than a cell-specific phenomenon for HepG2 cells. To further investigate the dynamics of this dissociation, time-lapse videomicroscopy was carried out. Both enzymes exhibit a similar domain structure consisting of a short cytoplasmic tail, a transmembrane domain, a stem region, and a catalytic domain (Paulson and Colley, 1989). This latter domain was replaced in both enzymes with the fluorescent protein GFP or RFP. The resulting constructs were stably transfected into HepG2 cells, and to avoid potential problems by overexpressing the constructs, we isolated individual colonies with modest expression levels. Using a mAb specific for galT (Berger et al., 1986) or a polyclonal antibody specific for siaT (Berger et al., 1993), we could show by indirect confocal immunofluorescence that the two chimeras almost perfectly colocalized with their respective endogenous counterparts in the GA, as indicated by the predominant yellow staining in the perinuclear region (Figure 1, A and C). Furthermore, we could demonstrate that galT-GFP is dispersed to the same swollen vesicles upon treatment with 2 µM monensin for 30 min as endogenous galT (Figure 1B), whereas siaT-GFP is not affected by monensin treatment and remains in the GA together with endogenous siaT (Figure 1D). To directly compare the dynamics of the two chimeras in presence of monensin, we transiently expressed siaT-RFP in HepG2 cells stably transfected with galT-GFP. The cells were cultured in glass-bottom dishes where they can readily be examined by live videomicroscopy on an inverted microscope equipped with an incubation chamber. For live experiments the medium was replaced by PBS supplemented by 10% FCS. Both chimeras were colocalized before monensin treatment and also immediately after the addition of 2 µM monensin (Figure 2A, 0 min) in the perinuclear region, representing the GA. However, upon treatment of double-labeled cells with monensin, a rapid separation of the two chimeras was observed. Although the localization siaT-GFP within the GA remained almost unchanged over the entire time course of 30 min, galT-GFP was found in swollen vesicles, which appeared early in the Golgi region, and then separated from the Golgi-localized siaT-RFP (Figure 2A, Supplementary Movie S1). There are two remarkable features of these swollen vesicles: first they appear within minutes after treatment (Figure 2A, 3 min) and second, their size reached a diameter of up to 2 µm, which is rather gigantic compared with other vesicles such as COPII-coated vesicles that are 50–60 nm in size (Figure 2A; Barlowe, 1995). In addition, to identify the Golgi scaffold we used the marker giantin, a vesicle-tethering protein localized to the cytoplasmic side of the GA (Linstedt and Hauri, 1993). Under control conditions, where the cells are grown in regular growth media supplemented with FCS before fixation, galT-GFP (Figure 2B) and siaT-GFP (Figure 2D) both showed substantial costaining with giantin. However, upon treatment with 2 µM monensin for 30 min, siaT-GFP remained colocalized with giantin (Figure 2E), whereas galT-GFP dissociated from giantin (Figure 2C) and is found in swollen vesicles. These results show that monensin treatment did not affect the scaffold of the Golgi in general but rather the localization of individual components, i.e., galT-GFP.
|
|
; Cherfils and Melancon, 2005) BFA forces genuine Golgi membranes to fuse with the ER, thereby transporting their content to the ER (Lippincott-Schwartz et al., 1989; Strous et al., 1990). In contrast, the TGN proteins do not redistribute to the ER but rather form a tubular network in an early step and later are in close proximity to the microtubule organization center (Lippincott-Schwartz et al., 1991). In control experiments without monensin pretreatment, the cells were monitored by time-lapse videomicroscopy after the addition of BFA at a concentration of 10 µM. Both galT-GFP (Figure 3A and Supplementary Movie S2) and siaT-GFP (Figure 3C and Supplementary Movie S4) were relocated from the GA to the ER. Pretreatment of the cells expressing galT-GFP with 2 µM monensin for 15 min before addition of BFA led to the initial formation of swollen vesicles containing galT-GFP (Figure 3B, 0 min). The subsequent addition of 10 µM BFA had no effect on these swollen vesicles as they could be observed over the entire time course (Figure 3B, Supplementary Movie S3). In contrast, cells expressing siaT-GFP were insensitive to the pretreatment with 2 µM monensin for 15 min (Figure 3D, 0 min), but were redistributed to the ER upon addition of BFA (Figure 3D, Supplementary Movie S5). These data support our contention that monensin treatment forms swollen vesicles in which galT is trapped and remained incompetent for its return to the GA and BFA-induced backflow to the ER.
|
) we investigated the presence of TGN46, a marker of the TGN excluded from the GA (Banting and Ponnambalam, 1997; Prescott et al., 1997), in the swollen vesicles by confocal immunofluorescence microscopy. As expected for proteins localized to compartments that are in such close proximity as the trans-Golgi cisterna and the TGN, we could show that both galT-GFP and siaT-GFP staining overlap partially with TGN46 staining in the Golgi area of untreated cells, with TGN46 being also present in peripheral structures (Figure 4, A and C), which are likely endosomes. The formation of TGN46-positive swollen vesicles is clearly visible upon treatment with monensin (Figure 4, B and D), which also contain galT-GFP (Figure 4B) but are devoid of siaT-GFP (Figure 4D).
|
20% in the lighter fractions (7–10) in untreated cells (Table 1). In cells treated with monensin before fractionation, galT-GFP accumulated in fractions 7–10 to 59%, which is an increase by 39% (Table 1), whereas no significant shift was observed in siaT-GFP–transfected cells (Table 1). The results for TGN46 were comparable to those of galT-GFP: 26% of TGN46 were found in fractions 7–10 of untreated cells and in cells treated with monensin TGN46 accumulated to 52% in the light fractions (7–10; Table 1).
|
|
|
|
). Both iso-forms have been localized to the trans Golgi cisterna (Russo et al., 1992), with a small fraction of the long form present at the plasma membrane (Shur et al., 1998). To further investigate the involvement of the cytoplasmic tail in monensin sensitivity of galT, we compared the two start variants of galT. Although the amount of the short form expressed in cells stably transfected with galT-GFP has not been determined, all the green fluorescence in cells stably transfected with galTctshort-GFP can be attributed to the short form. The short form entirely colocalized with endogenous galT in the GA (Figure 8B). GalTctshort-GFP remains in the GA in monensin-treated cells (Figure 8C), showing that the sensitivity to monensin is highly reduced in galTctshort-GFP.
|
), which might affect its trafficking behavior (Hathaway et al., 2003) and/or monensin sensitivity. We analyzed all three potential phosphorylation sites (Ser9 and Ser11 as pair or Ser18 alone) by mutating them either to alanine to prevent any phosphorylation or to aspartic acid to mimic the negative charge conferred by phosphorylation (Casanova et al., 1990). All mutant constructs were stably transfected into HepG2 cells and analyzed by double immunofluorescence with giantin as a Golgi marker. Preventing phosphorylation (galT-S9,11A-GFP, and galT-S18A-GFP) did not have any effect on the localization of the mutants, and monensin sensitivity was preserved in all of the mutants (Supplementary Figure S4). These results demonstrated that phosphorylation is not required for monensin sensitivity. Changing the serine residues to aspartic acid residues (galT-S9,11D-GFP, and galT-S18D-GFP) did not affect the localization in the GA or the monensin sensitivity of the mutants (Supplementary Figure S5). This indicates that monensin sensitivity of galT is not affected by phosphorylation or alternatively, that aspartic acid is not capable of mimicking phosphorylation of the cytoplasmic tail of galT.
Transfer of the Cytoplasmic Tail of galT Renders siaT Sensitive to Monensin
To demonstrate that the cytoplasmic tail alone is not only necessary but also sufficient to confer monensin sensitivity, we added the complete cytoplasmic tail of galT to the 5' end of siaT-GFP and SGG-GFP, respectively, to obtain the constructs GSSS-GFP and GSGG-GFP. In these constructs we also removed the starting methionine of the cytoplasmic tail of siaT to prevent alternative translation initiation. Both chimeras were found in the GA in stably transfected HepG2 cells, as indicated by a colabeling with giantin by confocal immunofluorescence (Figure 9, A and C). On treatment with 2 µM monensin both chimeras were separated from giantin into swollen vesicles (Figure 9, B and D). This clearly demonstrated that the cytoplasmic domain of galT is necessary and sufficient to target a protein into swollen vesicles induced by treatment with monensin.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Kweon et al., 2004): a dramatic shift of galT to swollen vesicles could be observed within minutes upon addition of monensin, whereas siaT remained Golgi-associated (Figure 2). The separation of these two enzymes in the presence of monensin was confirmed by quantitative subcellular fractionation. Furthermore, the origin of the swollen vesicles was shown to be derived from the TGN by colocalization of galT with the established marker protein TGN46 (Prescott et al., 1997) upon treatment with monensin. The 13 N-terminal amino acids of galT were identified as necessary and the entire cytoplasmic tail as sufficient to confer monensin sensitivity, suggesting the presence of sorting signals that mediate a continuous cycling of galT from its major localization in the trans-Golgi stack to the TGN and back.
Mechanism of Monensin Effect on the Secretory Pathway
In contrast to our findings, monensin was previously suggested to impose a trafficking block between the medial and the trans cisternae of the Golgi (Griffiths et al., 1983; Quinn et al., 1983). In fact, a medial Golgi block was always difficult to reconcile with the presence of galT, an established trans-Golgi marker, in monensin-induced swollen vesicles, as first described by Strous (1986). Additional reports made clear that Golgi proteins can be separated into two groups: one group that is affected by treatment with monensin such as galT (Berger et al., 1993), 1,3-fucosyltransferase 6 (Borsig et al., 1999), Golgi phosphoprotein of 130 kDa (GPP130), and Golgi protein-73 (GP73; Puri et al., 2002) and redistributes to swollen vesicles, and another group that includes proteins such as glucose-6-phosphatase (Griffiths et al., 1983), siaT (Berger et al., 1993), mannosidase II, N-acetylgalactosaminyltransferase 2, and giantin (Berger et al., 2001), which show little disturbance in presence of monensin. Such a spatial separation of the enzymes upon monensin treatment into different compartments obviously leads to a heterogeneous phenotype of impaired glycosylation. For review see Dinter and Berger (1998).
The Nature of galT-containing Swollen Vesicles
The strongest evidence of the present study that the swollen vesicles induced by monensin originate from the TGN is, on one hand, their fast appearance after the addition of monensin from Golgi (-like) structures (Figure 2), indicating that they are not derived from endosomes or ER and, on the other hand, the colocalization of galT containing swollen vesicles with TGN46 (Figure 4), a marker protein of the TGN (Prescott et al., 1997). Additional evidence that the swollen vesicles are TGN derived is provided by showing that in monensin-treated cells, galT-GFP was insensitive to BFA and remained in swollen vesicles (Figure 3). As shown by Berger et al. (2001), galT trapped in these vesicles does not return to the GA when monensin is replaced by chloroquine, an agent which blocks exit from an acidic post-Golgi compartment (as shown for TGN38; Chapman and Munro, 1994). The swollen vesicles may still release a small fraction of galT to an anterograde pathway leading to shedding, because monensin only slows but does not block post-Golgi trafficking of galT (Strous et al., 1985). That swollen vesicles originate from the TGN was first observed in a very elegant study of Zhang et al. (1993) conducted in sycamore maple suspension cells, showing that swollen vesicles were formed specifically from the TGN upon treatment with monensin and that the organization of the remaining Golgi stacks stayed intact as analyzed by electron microscopy.
Distinction between TGN and trans-Golgi Cisternae
Because morphological assignments based on single labeling experiments are not precise enough to differentiate trans-Golgi and TGN, siaT was often referred to as a TGN-localized transferase (see for example, Gleeson et al., 2004). If the TGN is operationally defined as the part of the secretory pathway that is not subject to BFA-induced retrograde flow, our data clearly associate siaT to the trans-Golgi cisternae codistributed with galT (Figure 3), which was also demonstrated by previous fractionation studies (Strous et al., 1993). Indeed, qualitatively we observed no difference in reactivity to BFA for both galT and siaT (Figure 3). Importantly, the TGN can also be delineated by the presence of specific markers such as TGN38/46, which revealed a different reaction to BFA, i.e., a collapse of TGN membranes around the microtubule organizing center (MTOC; Reaves and Banting, 1992). Transition of membrane or cargo proteins from the proximal to the distal part of the BFA divide is poorly understood and may involve vesicular transfer, fusion of trans-Golgi subdomains with the TGN, or maturation of (parts of) trans-Golgi cisternae to become the TGN.
Evidence for Golgi-sorting Signals Mediating Golgi-to-TGN Transition
In many instances, molecular determinants were localized within proteins mediating either their vesicular transfer or transition via a microdomain (for recent reviews see Aridor and Traub, 2002; Bonifacino and Traub, 2003; Robinson, 2004). Therefore, we investigated whether exchanging the domains (cytoplasmic tail, transmembrane domain, and stalk region) between galT-GFP and siaT-GFP would provide evidence for such determinants. Indeed, the cytoplasmic tail of galT was found necessary for monensin sensitivity: replacing the cytoplasmic tail of galT by the cytoplasmic tail of siaT (SGG-GFP) made galT insensitive to monensin; conversely replacing the cytosolic tail of siaT by the cytosolic tail of galT (GSS) conferred monensin sensitivity to siaT (Figure 7). Furthermore, by transferring the cytoplasmic tail of galT onto full-length siaT-GFP (GSSS-GFP), we could show that the monensin-induced redistribution to swollen vesicles is a dominant effect (Figure 9), suggesting a positive sorting signal that is sufficient to mediate the transport from the trans-Golgi to the TGN. A more detailed analysis of the short and the long iso-form of galT, which are the result of an alternative start methionine within the cytosolic tail, demonstrated that only the first 13 amino acids are required for monensin sensitivity.
Current Concepts of Trafficking of galT
GalT's primary localization is the GA where, in steady state, more than 90% of total cell-associated enzyme is accumulated (Rhee et al., 2005). Over time a small fraction is shed from the cell in a soluble and enzymatically active form (Strous and Berger, 1982). Of the internal pool, the fraction of galT that is not localized to the GA appears to be cycling through proximal compartments of the secretory pathway (Zaal et al., 1999; Rhee et al., 2005). In addition, the two iso-forms of galT, resulting from alternative translation initiation, display a difference in post-Golgi trafficking. Although the long isoform of galT has been shown to move to the plasma membrane, the short isoform has not been found at the plasma membrane (Hathaway et al., 2003). Our data support the finding that the two isoforms exhibit different intracellular trafficking properties and that the short isoform may be a true Golgi resident. Shur and associates (Hathaway et al., 2003) showed that phosphorylation of the long isoform of galT promotes its retention in the Golgi. However, in our hands phosphorylation mutants of the long isoform do not influence monensin sensitivity. This also indicates that phosphorylation is not critical for the transition from trans-Golgi to the TGN because the phosphorylation mutants are also found in monensin-induced swollen vesicles derived from the TGN.
Taking all these findings together, we propose a model (Figure 10), in which galT is localized to the Golgi by its transmembrane domain as shown previously by several groups (reviewed by Colley, 1997). Although the short isoform of galT is resident to the trans cisterna, the long isoform contains a forward signal mediating its transport from the trans-Golgi cisterna to the TGN and an additional signal for its return to the trans-Golgi, permitting this form to constantly cycle between the two compartments. On monensin treatment, retrograde transport from the trans-Golgi network back to the trans-Golgi cisterna, but not anterograde transport (Strous et al., 1985), is blocked and galT is found in TGN-derived swollen vesicles together with other TGN proteins such as TGN46 or cation-independent mannose 6-phosphate receptor. Although phosphorylation appears to have no influence on the cycling of galT between trans-Golgi cisterna and TGN, it might well be that it is involved in a regulated exit from the TGN to the plasma membrane, as suggested by the group of Shur (Youakim et al., 1994). The difference in behavior of galT and siaT upon monensin treatment could also explain the different findings regarding Golgi inheritance during mitosis. Although galT-GFP was found within the ER during mitosis (Altan-Bonnet et al., 2006), siaT-FKBP could not be trapped in the ER during mitosis (Pecot and Malhotra, 2004). Whether this difference goes along with the different sensitivity to monensin may be clarified by further work by using both proteins under both experimental conditions.
|
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
The online version of this contains supplemental material at MBC Online (http://www.molbiolcell.org). 
This article was published online ahead of print in MBC in Press (http://www.molbiolcell.org/cgi/doi/10.1091/mbc.E06-08-0665) on October 4, 2006.
Address correspondence to: Jack Rohrer (rohrer.jack{at}access.unizh.ch)
Abbreviations used: galT, 1,4-galactosyltransferase 1; siaT, 2,6-sialyltransferase 1; GA, Golgi apparatus; TGN, trans-Golgi network; ER, endoplasmic reticulum; BFA, brefeldin A; ECL, enhanced chemiluminescence.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Aridor, M. and Traub, L. M. (2002). Cargo selection in vesicular transport: the making and breaking of a coat. Traffic 3, 537–546.[CrossRef][Medline]
Balch, W. E. and Rothman, J. E. (1985). Characterization of protein transport between successive compartments of the Golgi apparatus: asymmetric properties of donor and acceptor activities in a cell-free system. Arch. Biochem. Biophys 240, 413–425.[CrossRef][Medline]
Banting, G. and Ponnambalam, S. (1997). TGN38 and its orthologues: roles in post-TGN vesicle formation and maintenance of TGN morphology. Biochim. Biophys. Acta 1355, 209–217.[Medline]
Barlowe, C. (1995). COPII: a membrane coat that forms endoplasmic reticulum-derived vesicles. FEBS Lett 369, 93–96.[CrossRef][Medline]
Berger, E. G. (1985). How Golgi-associated glycosylation works. Cell Biol. Int. Rep 9, 407–417.[CrossRef][Medline]
Berger, E. G., Aegerter, E., Mandel, T., Hauri, H. P. (1986). Monoclonal antibodies to soluble, human milk galactosyltransferase (lactose synthase A protein). Carbohydr. Res 149, 23–33.[CrossRef][Medline]
Berger, E. G., Berger, B., Höchli, M., Dinter, A. (2001). Colocalization of beta 1,4 galactosyltransferase with mannose 6-phosphate receptor in monensin-induced TGN-derived structures. Histochem. Cell Biol 115, 157–168.[CrossRef][Medline]
Berger, E. G., Grimm, K., Bächi, T., Bosshart, H., Kleene, R., Watzele, M. (1993). Double immunofluorescent staining of alpha 2,6 sialyltransferase and beta 1,4 galactosyltransferase in monensin-treated cells: evidence for different Golgi compartments? J. Cell. Biochem 52, 275–288.[CrossRef][Medline]
Bonifacino, J. S. and Traub, L. M. (2003). Signals for sorting of transmembrane proteins to endosomes and lysosomes. Annu. Rev. Biochem 72, 395–447.[CrossRef][Medline]
Borsig, L., Imbach, T., Höchli, M., Berger, E. G. (1999). alpha 1,3 Fucosyltransferase VI is expressed in HepG2 cells and codistributed with beta1,4galactosyltransferase I in the Golgi apparatus and monensin-induced swollen vesicles. Glycobiology 9, 1273–1280.
Bretz, R., Bretz, H., Palade, G. E. (1980). Distribution of terminal glycosyltransferases in hepatic Golgi fractions. J. Cell Biol 84, 87–101.
Casanova, J. E., Breitfeld, P. P., Ross, S. A., Mostov, K. E. (1990). Phosphorylation of the polymeric immunoglobulin receptor required for its efficient transcytosis. Science 248, 742–745.
Chapman, R. E. and Munro, S. (1994). Retrieval of TGN proteins from the cell surface requires endosomal acidification. EMBO J 13, 2305–2312.[Medline]
Cherfils, J. and Melancon, P. (2005). On the action of brefeldin A on Sec7-stimulated membrane-recruitment and GDP/GTP exchange of Arf proteins. Biochem. Soc. Trans 33, 635–638.[CrossRef][Medline]
Colley, K. J. (1997). Golgi localization of glycosyltransferases: more questions than answers. Glycobiology 7, 1–13.[Medline]
Dinter, A. and Berger, E. G. (1998). Golgi-disturbing agents. Histochem. Cell Biol 109, 571–590.[CrossRef][Medline]
Donaldson, J. G., Finazzi, D., Klausner, R. D. (1992). Brefeldin A inhibits Golgi membrane-catalysed exchange of guanine nucleotide onto ARF protein. Nature 360, 350–352.[CrossRef][Medline]
Gleeson, P. A., Lock, J. G., Luke, M. R., Stow, J. L. (2004). Domains of the TGN: coats, tethers and G proteins. Traffic 5, 315–326.[CrossRef][Medline]
Griffiths, G., Quinn, P., Warren, G. (1983). Dissection of the Golgi complex. I. Monensin inhibits the transport of viral membrane proteins from medial to trans Golgi cisternae in baby hamster kidney cells infected with Semliki Forest virus. J. Cell Biol 96, 835–850.
Griffiths, G. and Simons, K. (1986). The trans Golgi network: sorting at the exit site of the Golgi complex. Science 234, 438–443.
Hathaway, H. J., Evans, S. C., Dubois, D. H., Foote, C. I., Elder, B. H., Shur, B. D. (2003). Mutational analysis of the cytoplasmic domain of beta 1,4-galactosyltransferase I: influence of phosphorylation on cell surface expression. J. Cell Sci 116, 4319–4330.
Ho, S. N., Hunt, H. D., Horton, R. M., Pullen, J. K., Pease, L. R. (1989). Site-directed mutagenesis by overlap extension using the polymerase chain reaction. Gene 77, 51–59.[CrossRef][Medline]
Kweon, H.-S., et al. (2004). Golgi enzymes are enriched in perforated zones of Golgi cisternae but are depleted in COPI vesicles. Mol. Biol. Cell 15, 4710–4724.
Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685.[CrossRef][Medline]
Linstedt, A. D. 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.[Abstract]
Lippincott-Schwartz, J., Donaldson, J. G., Schweizer, A., Berger, E. G., Hauri, H. P., Yuan, L. C., Klausner, R. D. (1990). Microtubule-dependent retrograde transport of proteins into the ER in the presence of brefeldin A suggests an ER recycling pathway. Cell 60, 821–836.[CrossRef][Medline]
Lippincott-Schwartz, J., Yuan, L., Tipper, C., Amherdt, M., Orci, L., 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.[CrossRef][Medline]
Lippincott-Schwartz, J., Yuan, L. C., Bonifacino, J. S., 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, 801–813.[CrossRef][Medline]
Marxer, A., Stieger, B., Quaroni, A., Kashgarian, M., Hauri, H. P. (1989). (Na++ K+)-ATPase and plasma membrane polarity of intestinal epithelial cells: presence of a brush border antigen in the distal large intestine that is immunologically related to beta subunit. J. Cell Biol 109, 1057–1069.
Mollenhauer, H. H., Morre, D. J., Rowe, L. D. (1990). Alteration of intracellular traffic by monensin; mechanism, specificity and relationship to toxicity. Biochim. Biophys. Acta 1031, 225–246.[Medline]
Neutra, M. and Leblond, C. P. (1966). Synthesis of the carbohydrate of mucus in the Golgi complex as shown by electron microscope radioautography of goblet cells from rats injected with glucose-H3. J. Cell Biol 30, 119–136.
Paulson, J. and Colley, K. (1989). Glycosyltransferases. Structure, localization, and control of cell type- specific glycosylation. J. Biol. Chem 264, 17615–17618.
Pecot, M. Y. and Malhotra, V. (2004). Golgi membranes remain segregated from the endoplasmic reticulum during mitosis in mammalian cells. Cell 116, 99–107.[CrossRef][Medline]
Prescott, A. R., Lucocq, J. M., James, J., Lister, J. M., Ponnambalam, S. (1997). Distinct compartmentalization of TGN46 and beta 1,4-galactosyltransferase in HeLa cells. Eur. J. Cell Biol 72, 238–246.[Medline]
Puri, S., Bachert, C., Fimmel, C. J., Linstedt, A. D. (2002). Cycling of early Golgi proteins via the cell surface and endosomes upon lumenal pH disruption. Traffic 3, 641–653.[CrossRef][Medline]
Quinn, P., Griffiths, G., Warren, G. (1983). Dissection of the Golgi complex. II. Density separation of specific Golgi functions in virally infected cells treated with monensin. J. Cell Biol 96, 851–856.
Rabouille, C., Hui, N., Hunte, F., Kieckbusch, R., Berger, E. G., Warren, G., Nilsson, T. (1995). Mapping the distribution of Golgi enzymes involved in the construction of complex oligosaccharides. J. Cell Sci 108, 1617–1627.[Abstract]
Reaves, B. and Banting, G. (1992). Perturbation of the morphology of the trans-Golgi network following Brefeldin A treatment: redistribution of a TGN-specific integral membrane protein, TGN38. J. Cell Biol 116, 85–94.
Rhee, S. W., Starr, T., Forsten-Williams, K., Storrie, B. (2005). The steady-state distribution of glycosyltransferases between the Golgi apparatus and the endoplasmic reticulum is approximately 90:10. Traffic 6, 978–990.[CrossRef][Medline]
Robinson, M. S. (2004). Adaptable adaptors for coated vesicles. Trends Cell Biol 14, 167–174.[CrossRef][Medline]
Rohrer, J. and Kornfeld, R. (2001). Lysosomal hydrolase mannose 6-phosphate uncovering enzyme resides in the trans-Golgi network. Mol. Biol. Cell 12, 1623–1631.
Rohrer, J., Schweizer, A., Johnson, K. F., Kornfeld, S. (1995). A determinant in the cytoplasmic tail of the cation-dependent mannose 6-phosphate receptor prevents trafficking to lysosomes. J. Cell Biol 130, 1297–1306.
Roth, J. and Berger, E. G. (1982). Immunocytochemical localization of galactosyltransferase in HeLa cells: codistribution with thiamine pyrophosphatase in trans-Golgi cisternae. J. Cell Biol 93, 223–229.
Russo, R. N., Shaper, N. L., Taatjes, D. J., Shaper, J. H. (1992). Beta 1,4-galactosyltransferase: a short NH2-terminal fragment that includes the cytoplasmic and transmembrane domain is sufficient for Golgi retention. J. Biol. Chem 267, 9241–9247.
Sambrook, J., Fritsch, E. F., Maniatis, T. (1998). In: Molecular Cloning: A Laboratory Manual, Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
Sciaky, N., Presley, J., Smith, C., Zaal, K. J., Cole, N., Moreira, J. E., Terasaki, M., Siggia, E., Lippincott-Schwartz, J. (1997). Golgi tubule traffic and the effects of brefeldin A visualized in living cells. J. Cell Biol 139, 1137–1155.
Shur, B. D., Evans, S., Lu, Q. (1998). Cell surface galactosyltransferase: current issues. Glycoconj. J 15, 537–548.[CrossRef][Medline]
Storrie, B. (2005). Maintenance of Golgi apparatus structure in the face of continuous protein recycling to the endoplasmic reticulum: making ends meet. In: A Survey of Cell Biology, ed. W. J. Kwang. Academic Press: New York, 69–94.
Strous, G., Berger, E. G., van Kerkhof, P., Bosshart, H., Berger, B., Geuze, H. J. (1990). Brefeldin A induces a microtubule-dependent fusion of galactosyltransferase-containing vesicles with the rough endoplasmic reticulum. Biol. Cell 71, 25–31.
Strous, G. J. (1986). Golgi and secreted galactosyltransferase. CRC Crit. Rev. Biochem 21, 2119–151.[Medline]
Strous, G. J. and Berger, E. G. (1982). Biosynthesis, intracellular transport, and release of the Golgi enzyme galactosyltransferase (lactose synthetase A protein) in HeLa cells. J. Biol. Chem 257, 7623–7628.
Strous, G. J., Kerkhof, P., Fallon, R. J., Schwartz, A. L. (1987). Golgi galactosyltransferase contains serine-linked phosphate. Eur. J. Biochem 169, 307–311.[Medline]
Strous, G. J., Van Kerkhof, P., Berger, E. G. (1988). In vitro biosynthesis of two human galactosyltransferase polypeptides. Biochem. Biophys. Res. Commun 151, 314–319.[Medline]
Strous, G. J., van Kerkhof, P., van Meer, G., Rijnboutt, S., Stoorvogel, W. (1993). Differential effects of brefeldin A on transport of secretory and lysosomal proteins. J. Biol. Chem 268, 2341–2347.
Strous, G. J., Van Kerkhof, P., Willemsen, R., Slot, J. W., Geuze, H. J. (1985). Effect of monensin on the metabolism, localization, and biosynthesis of N- and O-linked oligosaccharides of galactosyltransferase. Eur. J. Cell Biol 36, 256–262.[Medline]
Taatjes, D. J., Roth, J., Weinstein, J., Paulson, J. C., Shaper, N. L., Shaper, J. H. (1987). Codistribution of galactosyl- and sialyltransferase: reorganization of trans Golgi apparatus elements in hepatocytes in intact liver and cell culture. Eur. J. Cell Biol 44, 187–194.[Medline]
Tartakoff, A. M. (1983). Perturbation of vesicular traffic with the carboxylic ionophore monensin. Cell 32, 1026–1028.[CrossRef][Medline]
Towbin, H., Staehelin, T., Gordon, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76, 4350–4354.
Watzele, G., Bachofner, R., Berger, E. G. (1991). Immunocytochemical localization of the Golgi apparatus using protein-specific antibodies to galactosyltransferase. Eur. J. Cell Biol 56, 451–458.[Medline]
Watzele, G. and Berger, E. G. (1990). Near identity of HeLa cell galactosyltransferase with the human placental enzyme. Nucleic Acids Res 18, 7174.
Youakim, A., Dubois, D. H., Shur, B. D. (1994). Localization of the long form of beta-1,4-galactosyltransferase to the plasma membrane and Golgi complex of 3T3 and F9 cells by immunofluorescence confocal microscopy. Proc. Natl. Acad. Sci. USA 91, 10913–10917.
Zaal, K. J., et al. (1999). Golgi membranes are absorbed into and reemerge from the ER during mitosis. Cell 99, 589–601.[CrossRef][Medline]
Zhang, G. F., Driouich, A., Staehelin, L. A. (1993). Effect of monensin on plant Golgi: re-examination of the monensin-induced changes in cisternal architecture and functional activities of the Golgi apparatus of sycamore suspension-cultured cells. J. Cell Sci 104, 819–831.[Medline]
This article has been cited by other articles:
|
|
 |
