key: cord-0978268-q0rsfzgw
authors: LAVI, EHUD; WANG, QIAN; WEISS, SUSAN R.; GONATAS, NICHOLAS K.
title: Syncytia Formation Induced by Coronavirus Infection Is Associated with Fragmentation and Rearrangement of the Golgi Apparatus
date: 1996-07-15
journal: Virology
DOI: 10.1006/viro.1996.0382
sha: a014febfeeb03575a31fdc1bdedcf236898dea50
doc_id: 978268
cord_uid: q0rsfzgw

Abstract Coronavirus mouse hepatitis virus (MHV) possesses a membrane glycoprotein (M) which is targeted to the Golgi apparatus (GA). We used immunocytochemistry with an organelle-specific antiserum to investigate the morphologic changes of the GA during infection of L2 murine fibroblasts with MHV-A59. Twenty-four hours after infection the GA was fragmented and translocated in the center of syncytia, while the microtubular network was also rearranged displaying radiating elements toward the center of syncytia. Two fusion-defective mutants, which contain an identical amino acid substitution in the cleavage signal sequence of the spike glycoprotein (S), induced fragmentation of the GA. However, the GA migrated only partially to the centers of syncytia during infection with these mutants. Revertant viruses, in which the above mutation was corrected, had fusion properties and GA staining similar to wtMHV-A59. Experiments with brefeldin A (BFA), which induces redistribution of the GA into the rough endoplasmic reticulum (RER), revealed that an intact GA for a period of 4–16 hr postinfection, is required for coronavirus replication and syncytia formation. Thus, during MHV infection, syncytia formation is associated with fragmentation of the GA, followed by a previously undescribed phenomenon of migration of the organelle into the centers of syncytia. The fragmentation of the GA, however, may occur without the formation of syncytia. Therefore, two distinct mechanisms may be responsible for the fragmentation of the GA and its subsequent migration to the center of syncytia.

undergo endocytosis into the trans-Golgi-network (TGN), while shiga toxin is internalized into the GA and the RER Polypeptides synthesized in the rough endoplasmic (Sandvig et al., 1991; . Pretreatment of cells reticulum (RER) and destined for plasma membranes, with BFA protects them against the lethal effects of cerlysosomes, and secretion are transported through the tain toxins, implying that the toxin's entry into TGN and Golgi apparatus for posttranslational modifications and the GA is necessary for translocation into their cytosolic targeting (Farquhar and Palade, 1981; Mellman and Sitargets (Yoshida et al., 1991) . In the human disease mons, 1992; Rothman and Orci, 1992) . The GA is a dyamyotrophic lateral sclerosis (ALS), the GA of spinal cord namic organelle which undergoes morphologic and funcmotor neurons is fragmented into numerous small eletional modifications under physiologic and pathologic ments which resemble the dispersion of the organelle conditions. In mitotic Hela cells the GA disperses in early induced by agents depolymerizing microtubules (Moureprophase and reaggregates in telophase (Robbins and latos et al., 1990, 1993; Gonatas et al., 1992) . A similar Gonatas, 1964) . During mitosis, the Golgi apparatus fragfragmentation of the GA has been observed in motor ments into numerous small groups of vesicles which neurons of transgenic mice expressing a mutant Cu,Zn have been referred to as the mitotic form of the organelle superoxide dismutase (Mourelatos et al., 1996) . (Lucocq et al., 1987) . In interphase cells treated with Migration and rearrangements of the GA have been drugs which depolymerize microtubules, the GA fragreported during syncytia formation and fusion of Vero ments into small randomly distributed elements (Robbins cells infected with Sindbis virus (Ho et al., 1990) . Within and Gonatas, 1964; Turner and Tartakoff, 1989) . In many 3-5 hr after infection, individual elements of the GA, cells the secretion blocker Brefeldin A (BFA) reversibly which are associated initially with separate microtubuleredistributes membranes and enzymes of the GA back organizing centers in perinuclear areas of fused cells, into the RER, but does not inhibit endocytosis (Doms et congregate in the center of syncytia and form an exal., 1989; Lippincott-Schwartz et al., 1989; Johnston et al., tended network of nondisrupted intact Golgi complexes 1994). Ricin, cholera toxin, and wheat germ agglutinin (Ho et al., 1990) . In contrast to Sindbis infections, the Golgi apparatus is fragmented in cells infected with herpes simplex virus 1 (Campadelli et al., 1993) .

is characterized by retention of the viral envelope glyco-1991). The primary structure of MG-160 revealed significant homology with a chicken fibroblast growth factor protein M (previously known as E1) within the GA (Tooze et al., 1988; Klumperman et al., 1994; Krijnse-Locker et receptor and a ligand to E-selectin (Gonatas et al., 1995; Steegmaier et al., 1995) . In chicken, MG-160 appears al., 1994) and budding of virions from internal membranes (Holmes and Behnke, 1981; Tooze et al., 1988) . The close early in development and the gene coding the protein, named GLG1, has been assigned to chromosome 16 interaction between coronavirus particles with the GA provides an unusual opportunity to study morphologic (Stieber et al., 1995; Mourelatos et al., 1995) . and functional properties of the GA and various aspects of virus-cell interactions. In this study, we used organ-Immunohistochemistry elle-specific antibodies, immunohistochemistry, and transmission electron microscopy to examine the fate of Cells, grown on poly-D-lysine-treated coverslips, were the GA during cell fusion and syncytia formation in mouse fixed with 2% paraformaldehyde for 20 min at room tem-L-2 cells infected with MHV-A59 and fusion defective perature, washed three times in PBS, then incubated for mutants. Fragmentation of the GA into small immuno-30 min in 0.05% Saponin/10% goat serum (GS) in PBS stained elements occurred prior to their migration into and washed three times in 10% GS in PBS. Cells were the centers of syncytia. Experiments with BFA revealed then incubated with primary antibody (1:1000 dilution in that the period of 4-16 hr postinfection is critical for PBS of immunoaffinity-purified rabbit anti-MG-160 anticoronavirus replication and syncytia formation. The rebodies or with a supernatant from the anti MG-160 hysults of studies with fusion defective mutants suggest bridoma (10A8) overnight at room temperature. Cultures that different mechanisms are responsible for the initial were then washed, incubated with a biotinylated goat fragmentation of the GA and its subsequent migration to anti-rabbit IgG antibody, incubated with the avidin-biotin the centers of syncytia. complex (ABC), and stained with diaminobenzidine tetrahydrochloride (DAB) (5 mg DAB/10 ml Tris-saline con-MATERIALS AND METHODS taining 10 mM Imidazole and 0.03% H 2 O 2 ), according to standard methods (Graham and Karnovsky, 1966; Gues-Cell cultures don et al., 1979) . L-2 cells (murine fibroblasts), originally obtained from American Type Culture Collection (ATCC; Rockville, MD), Electron microscopy were used to maintain viral growth cultures, preparation of viral stocks, and viral plaque assays. In some experi-Cells grown on Thermanox plastic (EM Sciences, Fort ments 17Cl-1 murine fibroblasts were used. The cultures Washington, PA), were fixed overnight at 4Њ with 2.5% were maintained in DMEM/heat-inactivated 10% fetal boglutaraldehyde / 1% paraformaldehyde in 0.1% cacodylvine serum/1% Pen/Strep/4500 mg/liter D-glucose/L-gluate buffer, pH 7.4, / 0.002% CaCl 2 . Subsequently cultures tamine.

were postfixed in 1% osmium tetroxide / 1.5% potassium ferrocyanide, dehydrated in ethanol, and embedded in Viruses and infections

Araldite. Sections (8-10 nm thick) were stained with lead and uranyl salts and viewed in a transmission electron Three times plaque purified MHV-A59 stock virus was microscope (JEOL 100 CX) at 80 kv (Karnovsky, 1971; used as previously described (Lavi and Weiss, 1989) . Stieber et al., 1987) . Fusion defective mutants and revertants were prepared as previously described (Gombold et al., 1993; Hingley et al., 1994) . Infections of cultures were done by incuba-Viral infectivity assay tion of virus with cells for 1 hr at a multiplicity of infection (m.o.i.) of 1, followed by washing of cells with fresh me-Viral titers were determined by duplicate plaque dium three times. assays of several 10-fold dilution of samples in L-2 cell grown in six-well plates (Lavi et al., 1984) .

MG-160 is a conserved sialoglycoprotein of the medial Brefeldin A (BFA) treatment cisternae of the GA. The preparations of the anti-MG-160 monoclonal antibody (10A8), and the immunoaffinity BFA (Sigma) stock solution (5 mg/ml in ethanol) was diluted in PBS and applied to cultures at a concentration purified anti-MG-160 polyclonal antibodies, were described in previous publications (Gonatas et al., 1989;  of 5 mg/ml for periods of 1 hr or longer, as specifically indicated in the text (Fig. 3) . In control experiments cul- Croul et al., 1988 Croul et al., , 1990 . The preparation of the monoclonal antibody 2H1, a RER marker of a 60-to 65-kDa tures were incubated for the same time periods with the same dilution of ethanol used to dissolve BFA. protein, has been previously described (Chen et al.,

and cyncytia formation when infected with MHV-A59. Following infection, cells were fixed and examined by immu-Changes of the GA during MHV infection and nohistochemistry with identical methods used for L2 formation of syncytia cells. The changes in the GA observed in these cells were identical to those seen in L2-infected cells. Morphological changes of the GA during coronavirus infection of L2 cells (4, 8, 16, 24 , and 48 hr) were examined by immunocytochemistry using organelle-specific Changes of the GA following infection of L2 cells with antibodies (Croul et al., 1988 (Croul et al., , 1990 Gonatas et al., 1989) .

fusion-defective mutants and a fusion-negative strain Typical infection of L2 cells with MHV-A59 caused syncyof MHV tia formation of the entire monolayer within 24 hr and complete cytolysis within 48 hr. At 16 hr postinfection, To investigate whether the fragmentation of the GA is dependent on cell fusion, we examined the organelle in cell borders were indistinct and the nuclei were aggregated; by 24 hr postinfection, the syncytia acquired their L2 cells infected with two fusion defective mutants of MHV-A59. The C12 and B11 mutants, used in these ex-typical morphology consisting of a ring of nuclei surrounding a cytoplasmic center devoid of nuclei. Between periments, were isolated from primary cultures of glial cells infected with MHV-A59. These mutants are fusion-24 and 48 hr postinfection, cells underwent pyknosis, karyorhexis, and then died and detached from the culture delayed and defective but not fusion-negative and produce reduced number of syncytia formations. Thus at dish.

The immunostained GA of uninfected cells formed con-any given time point postinfection with the mutant viruses, only a small percentage of L2 cells form syncytia, tiguous coarsely granular focal or ring-like perinuclear profiles (Figs. 1A and 1D). Following infection of L2 cells as compared to cells infected with MHV-A59 (Gombold et al., 1993) . These fusion-defective mutants have a histi-with MHV-A59, the coarsely granular stain of the GA became fine and smaller individual elements of the GA dine to aspartic acid mutation (H716D) within the cleavage signal of the spike (S) glycoprotein. Cleavage of S were discernible. At 16 hr postinfection, the majority of the syncytia consisted of clustered nuclei within a cyto-is necessary for efficient cell fusion during MHV-A59 infection (Gombold et al., 1993) . Infection with these mutant plasmic mass, lacking distinguishable cell borders; in those syncytia, the immunostained GA appeared as viruses does not affect the efficiency of viral replication or titers of infectious virus (Gombold et al., 1993) . Immu-strands of finely granular elements forming a honeycomb-like network with interspersed nuclei (Figs. 1B and nostaining of the GA in L2 cultures infected with the fusion defective mutants B11 and C12 showed fragmen-1E). However, before full syncytia formation, in cells surrounding the forming syncytia the distribution of the GA tation of the GA similar to that found in cells infected with MHV-A59. However, the translocation and rear-was perinuclear, similar to that of controls but with finer granular GA elements (compare Figs. 1E and 1D) . This rangement of the GA during infection with these two mutant viruses was distinctly different from the changes phenomenon probably represented an early stage of fragmentation and rearrangement of the GA. At 24 hr of the organelle seen during infections with the wild-type MHV-A59. Specifically, cells infected with either B11 or postinfection, the process of fragmentation of the GA and its relocation in the centers of syncytia, as well as the C12 mutants displayed both central and peripheral aggregates of finely granular GA (Figs. 1H and 1I). The rearrangement of the nuclei in a ring formation within the syncytia, was complete (Figs. 1C and 1F) . The honey-granular stain of the GA at the periphery of syncytia, and specifically around the peripherally located nuclei, was comb morphology due to the interspersed nuclei was replaced by a typical central aggregate of finely granular observed only in cultures infected with these two mutant viruses. This pattern of GA distribution was not seen in or fragmented GA, surrounded by a rim of nuclei (compare Figs. 1F with 1E). Furthermore, the nuclei in the any of the stages of MHV-A59 viral replication (compare Figs. 1I and 1F ). In order to rule out the possibility that syncytium were not surrounded by any residual GA. Thus, in the fully developed syncytium, all elements of the frag-the fusion-defective variants showed only delay in translocation of the GA into the center of cyncytia, immunocy-mented GA had moved into the center, while the more peripherally arranged nuclei formed a ring devoid of adja-tochemical analysis was performed at 36 and 48 hr after viral infection. These experiments showed that the typical cent elements of the GA (Fig. 1F) . Staining of syncytia with rabbit anti-MHV polyclonal antibodies revealed an complete central translocation of the GA, similar to that seen in A59-infected cultures, never occurred in cultures abundance of viral proteins in the center of syncytia (not shown).

infected with the two fusion defective mutants, even prior to or at the stage of complete cell death. In order to determine whether the fragmentation and rearrangement of the GA was cell-type dependent, an-To further explore the link between fragmentation of the GA and fusion we then infected L2 cells with MHV-other cell line, 17Cl-1 murine fibroblasts, was infected with MHV-A59. This cell line supports both cell fusion 2, a fusion-negative strain of MHV. L2 cells are suscepti- The network of microtubules is rearranged during MHV-A59 infection pletely destroyed after 48 hr with titers similar to MHV-A59. However, there was no cell fusion or syncytia forma-Since the GA is associated with microtubules of tion in these cultures. Infection with MHV-2 produced interphase cells (Robbins and Gonatas, 1964; Turner dispersion and fragmentation of the GA in individual and Tartakoff, 1989) , we investigated whether microtucells, some of which appeared to be balooned after 24 bules are affected during coronavirus infection. Spehr (Fig. 1G) . These observations are consistent with the cifically, we investigated whether the fragmentation of conclusion that the fragmentation of the GA during infecthe GA within the centers of the syncytia is associated tion with MHV is independent of cell fusion. However, with a similar change of the microtubules. The immuthe translocation of the GA in the center of the syncytia nocytochemical staining of the GA at 4, 8, 16, 24, and is probably linked to cell fusion since fusion-defective 48 hr after MHV-A59 infection was compared with the mutants were also defective in their ability to induce immunostaining of microtubules with antibodies translocation of the GA to the center of the syncytia.

against alpha and beta tubulin. While the GA appeared fragmented early during infection and syncytia forma-Ultrastructural changes of the GA in MHV-infected tion, fragmentation and disintegration of microtubules cells occurred late (48 hr), when cells die. The kinetics of To further investigate the morphologic aspects of the the microtubule changes after infection with MHV-A59 fragmented GA in syncytia of L2 cells infected with MHVwas depicted by immunofluorescence on L-2 cells A59, an electron microscopic examination was perafter infection with MHV-A59 (m.o.i. Å 1 PFU/cell) and formed in cells 24 hr after infection and at a multiplicity after staining with anti-tubulin antibodies and FITCof infection (m.o.i.) of 1 plaque forming unit per cell. Areas conjugated secondary antibodies (Fig. 3) . In uninof syncytia with a typical peripheral rim of nuclei were fected cells and at 4 hr postinfection the microtubules selected from semithin (0.5-1 mm) sections. In uninwere distributed throughout the entire cytoplasm of fected cells the GA was seen in a perinuclear location the individual cells. At 16 hr postinfection the syncytia and consisted of several groups of stacked cisternae were beginning to form and the microtubules were surrounded by numerous coated and uncoated vesicles still distributed within the entire cytoplasm. At 24 hr (Fig. 2B) . In contrast to this typical morphology of the GA, postinfection, when the GA was fragmented and transin infected cells the stacks of the cisternae were marklocated to the centers of syncytia, the microtubules edly diminished in size and were replaced by numerous were rearranged in a characteristic pattern (Fig. 3) . tubulovesicular structures, some of which containing vi-Specifically, at the periphery of the syncytia, an inrus particles ( Fig. 2A) . Furthermore, in infected cells the tense stain for tubulins suggested that the nuclei were region containing remnants of GA cisternae and the surrounded by a rich network of microtubules. At the abundant tubulovesicular structures was rich in intermediate zone, between the periphery and the centransversing microtubules ( Fig. 2A) .

ter of syncytia, the microtubules formed a radiating network. At the center of the syncytia, the immunostain The distribution of the RER is not affected by MHVfor alpha and beta tubulin was less intense and amor-A59 infection phous (Fig. 3) . These changes suggest that during coronavirus infection the microtubules undergo rear-The distribution of the RER in MHV-A59-infected and rangement and perhaps provide guidance for the control L2 cells was investigated by immunocytochemtranslocation of the fragmented elements of the GA istry with the organelle-specific monoclonal antibody into the center of the syncytia. 2H1 . There was no detectable difference between the infected and uninfected cells in the

The effect of Brefeldin A on coronavirus infection immunostaining of the RER. In both cases the fine granular staining of the RER was evenly distributed Since coronavirus infection is associated with the processing of viral proteins through the GA, including within the entire cytoplasm including the centers of syncytia (not shown). within the organelle, we investigated the effect of BFA 24 hr later. Uninfected cells, however, had completely recovered from the BFA effect 16 -24 hr later. These on the morphology and kinetics of coronavirus infection. As summarized in Fig. 4 , BFA was introduced at findings suggest that virus infection accentuates and prolongs the BFA effect (Fig. 4) . various time points during infection of L2 cell with MHV-A59. In uninfected L2 cells incubated for 1 hr When cells were treated continuously with BFA during infection there was dispersion of the GA, lack of syncytia with 5 mg/ml of BFA, immunostaining with anti-MG-160 showed a diffuse cytoplasmic pattern consistent formation, and no detectable viral titers (Fig. 4) . If BFA treatment began at 4 or 6 hr after viral inoculation, the with the known redistribution of MG-160 and other Golgi markers within the RER (Doms et al., 1989; Lip- effect of this treatment on infection was similar to the effect observed after continuous BFA treatment (i.e., re-pincott- Schwartz et al., 1989; Johnston et al., 1994) . Following 1 hr of BFA treatment at the beginning of duced syncytia formation and no viral titers). However, when the BFA treatment began 12 hr or more after viral infection cells had diffuse staining of the GA when observed 8 hr postinfection as did uninfected cells inoculation, there were relatively minimal effects on syncytia formation, and viral titers were 1-2 logs lower as following a similar treatment with BFA. However, at 16 hr postinfection (and 1 hr of treatment with BFA at the compared to viral titers without BFA treatment (which usually fluctuate about 1-1.5 logs in various experi-beginning of infection) there was incomplete recovery from the BFA effect. In uninfected cells treated in a ments). When BFA was introduced 20 min or an hour before similar fashion with BFA, the recovery was complete. Therefore, when cells were infected with MHV-A59 inoculation and continued through the entire infection or 1, 3, or 5 hr after inoculation, there was no significant prior to or at the same time of treatment with BFA, the effect of the drug on the GA was not abolished 16 or effect on the ability of the virus to form syncytia 24 hr by cisternae derived from the intermediate compartment between the ER and the Golgi stacks, thus acquiring two membranes in one step (Sodeik et al., 1993 (Sodeik et al., , 1995 . The second wrapping cisternae in vaccinia virus assembly is derived from the trans Golgi network (Schmelz et al., 1994) . In CMV, ultrastructural as well as biochemical studies suggested that short-term exposure of infected cultures to BFA during the late infectious cycle primarily prevented Golgi-dependent processes, e.g., envelopment of naked cytoplasmic nucleocapsids in the trans-Golgi network (TGN) and normal processing of glycoprotein B (Eggers et al., 1992) . In Uukuniemi virus, a member of the bunyaviridae, immunofluorescent staining indicated that G1 glycoprotein expressed alone localized to the GA. G2 expressed alone was associated with the RER (Melin et al., 1995) . Coronavirus MHV M glycoprotein (previously known as E1) is targeted to the GA and contains a retention signal for the GA (Machamer et al., 1990; Swift and Machamer, 1991; Armstrong and Patel, 1991) . Previous reports have shown that the GA undergoes rearrangement (Ho et al., 1990) and fragmentation (Campadelli et al., 1993) in viral infections. The data presented here detail the changes in the morphology of the GA in cells infected by a virus which induces the formation of syncytia. In coronavirus infection, the virus displays Cultures were incubated with virus (") for 1 hr and exposed to BFA (ٖ) for various periods. At the end of 24-hr incubation the cultured cells complex and close interactions with the GA. In coronaviwere stained for MG-160 by immuohistochemistry and the supernatants rus-induced syncytia formation there is a unique translowere titered for virus by plaque assay. Titers labeled 0 indicate levels cation and aggregation of the GA into the center of the below 50 PFU/ml, which is the lowest level of detection in this assay. syncytia which is not accompanied by similar changes of the RER and cytoskeleton (Fig. 1F) . The observed fragmentation of the GA is not related to the formation of postinfection and viral titers at that time were only 1-1.5 syncytia as demonstrated by the experiments using a logs lower than without treatment. Thus BFA did not fusion-negative strain of MHV and fusion-defective mublock the endocytosis and the initial processing of the tants in which fragmentation of the GA also occurs. Fragvirus into cells. BFA reduced viral replication only by mentation and translocation of the GA may be important 1-2 logs when introduced at 12-24 or 16-24 hr after in the life cycle of coronavirus replication. Further studies infection. The results of the exposure to BFA at the beginare necessary to determine whether there are viruses ning of infection for 4 hr and for various periods at the that replicate within cells without causing alteration of end of infection (Fig. 4) indicated that the BFA effect on GA morphology. the GA during 4-16 hr postinoculation was associated Central translocation and aggregation of the GA may with reduced viral replication. If BFA was introduced after be unique to coronaviruses as it was not associated with or before this stage (4-16 hr), there was no effect on other fusion and syncytia forming viruses such as Sindbis viral replication, assembly, and maturation. These results virus infection (Ho et al., 1990) or herpes simplex virus suggest that the interval of 4-16 hr postinfection is the infection (Campadelli et al., 1993) . Since the aggregation most significant period requiring an intact GA for viral of the fragmented GA in the centers of syncytia has not replication and syncytia formation. This conclusion is been previously reported, we investigated the GA in cells based on infection at an m.o.i. of 1 PFU/cell which may infected with LaCrosse bunyavirus, which is known to be an asynchronous infection. In a more synchronous induce the formation of syncytia (Gonzalez-Scarano et infection the period of BFA effect may terminate earlier al., 1984) . Immunostaining of the GA in BHK cells infected than 16 hr postinfection.

with LaCrosse virus for 24 hr showed network formation and fragmentation of the GA, without the aggregation of DISCUSSION the GA into centralized zones and without the formation of amorphous centers of syncytia which is characteristic The GA plays an important role in the life cycle of many viruses such as vaccinia, CMV, bunyaviruses, and of MHV infection (unpublished observations). Thus the translocation of the GA to the center of the syncytia may coronaviruses. Vaccinia virus DNA becomes enwrapped

Evolution of a coronavirus during persistent infection in vitro

Use of ferrocyanide-reduced osmium tetroxide cessed through the Golgi apparatus after infection with herpes simin electron microscopy

Coronavirus M proteins accumulate in the 60-65 kD membrane polypeptide of the rough endoplasmic reticulum Golgi complex beyond the site of virion budding

An anti-organelle antibody in pathology. The chromatolytic reaction studied with a Characterization of the budding compartment of mouse hepatitis virus: evidence that transport from the RER to the Golgi complex Rothman

Experimental demyelination produced by the A59 strain of ture and toxic effect

Assembly of vaccinia virus: The second lar Aspects of Neurotropic Viral Infections

Assembly of vaccinia virus: Incorporation of p14 treated with Brefeldin A: evidence for membrane cycling from Golgi and p32 into the membrane of the intracellular mature virus

form of the Golgi apparatus in HeLa cells

Assembly of vaccinia virus: role of the intermediate compartment

between the endoplasmic reticulum and the Golgi stacks

Proteolytic cleavage Golgi complex

Ronnholm, R., and fusion activity

virus contains a signal for localization to the Golgi complex

The Golgi complex: In vitro veritas?

Fragmentation of the Golgi apparatus of motor anti-organelle antibody

neurons in amyotrophic lateral sclerosis revealed by organelle-spe-Stieber

MG-160, A membrane protein of the Golgi apparatus

Assignment of the GLG1 gene for MG-160, a fibroblast ligand for E-Selectin, is found only in the Golgi apparatus and apgrowth factor and E-Selectin binding membrane sialoglycoprotein of pears early in chicken embryo development

354-355. membrane-spanning domain of coronavirus E1 protein

Fusion formation by the uncleaved spike protein of in transgenic mice expressing mutant Cu,Zn superoxide dismutase murine coronavirus JHMV variant cl-2

Acetyl-galactosamine to the E1 glycoprotein of mouse hepatitis virus-Mourelatos

Histochemical and ultrastrucand membrane traffic in Golgi complex organization

The ultrastructure of a mammaof the Golgi apparatus by brefeldin A inhibits the cytotoxicity of ricin, modeccin, and Pseudomonas toxin

monoclonal antibody against the Golgi apparatus. Am. J. Pathol. 133, be related to a unique interaction between coronavirus 355-362. proteins with the membranes of the GA. Croul, S. E., Mezitis, S. G. E., Stieber, A., Chen, Y., Gonatas, J. O., Goud, Since mutant viruses that are defective in their ability B., and Gonatas, N. K. (1990) . Immunohistochemical visualization of to cause efficient fusion are also inefficient in translocatthe Golgi apparatus in several species, including human, and tissue ing the GA to the center of the syncytia, the translocation with antiserum against MG-160, a sialoglycoprotein of rat Golgi apparatus. J. Histochem. Cytochem. 38, [957] [958] [959] [960] [961] [962] [963] of the GA appears to be linked to the ability of the virus Doms, R. W., Russ, G., and Yewdell, J. W. (1989) . Brefeldin A redistribto cause fusion. Although the fusion property has been utes resident and itinerant Golgi proteins to the endoplasmic reticuassociated with the S gene and the site encoding cleav- lum. J. Cell Biol. 109, [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] age of the S protein (Gombold et al., 1993) , other reports Eggers, M., Bogner, E., Agricola, B., Kern, H. F., and Radsak, K. (1992). showed that cleavage of the S gene in coronaviruses is Inhibition of human cytomegalovirus maturation by brefeldin A. J. Gen. Virol. 73, 2679 -2692 not an absolute requirement for fusion (Stauber et al., Farquhar, M. G., and Palade, G. E. (1981) . The Golgi apparatus (com-1993; Taguchi et al., 1993) . However, fragmentation of plex)- useful in future studies of aspects of both viral replication Gonzalez-Scarano, F., Pobjecky, N., and Nathanson, N. (1984) . LaCross and the biology of the GA.Bunyavirus can mediate pH-dependent fusion from without. Virology 132, 222-225.