Reassembly of Golgi stacks from mitotic Golgi fragments in a cell-free system

Rat liver Golgi stacks were incubated with mitotic cytosol for 30 min at 37 degrees C to generate mitotic Golgi fragments comprising vesicles, tubules, and cisternal remnants. These were isolated and further incubated with rat liver cytosol for 60 min. The earliest intermediate observed by electron microscopy was a single, curved cisterna with tubular networks fused to the cisternal rims. Elongation of this cisterna was accompanied by stacking and further growth at the cisternal rims. Stacks also fused laterally so that the typical end product was a highly curved stack of 2-3 cisternae mostly enclosing an electron-lucent space. Reassembly occurred in the presence of nocodazole or cytochalasin B but not at 4 degrees C or in the absence of energy supplied in the form of ATP and GTP. Pretreatment of the mitotic fragments and cytosol with N-ethylmaleimide (NEM) also prevented reassembly. GTP gamma S and A1F prevented reassembly when added during fragmentation but not when added to the reassembly mixture. In fact, GTP gamma S stimulated reassembly such that all cisternae were stacked at the end of the incubation and comprised 40% of the total membrane. In contrast, microcystin inhibited stacking so that only single cisternae accumulated. Together these results provide a detailed picture of the reassembly process and open up the study of the architecture of the Golgi apparatus to a combined morphological and biochemical analysis.     

Characterization of azadirachtin binding to Sf9 nuclei in vitro

[22,23-(3)H(2)]dihydroazadirachtin was incorporated by Sf9 cells in culture and was bound specifically to the nuclear fraction. The observed association constant of the binding of the radioligand to a purified nuclear fraction was determined to be 0.037 +/- 0.008 min(-1) using a one-phase exponential association equation, and binding appeared to be to a single population of sites. The binding was essentially irreversible, and the dissociation constant was estimated to be 0.00065 +/- 0.00013 min(-1). An association rate constant of 7.3 x 10(6) M(-1) min(-1) was calculated from these data. Binding was saturable, and the receptor number and affinity were determined as B(max) = 23.87 +/- 1.15 pmol/mg protein, K(d) = 18.1 +/- 2.1 nM. The order of potency of semisynthetic azadirachtin analogues for competition for the binding site was as follows (IC(50) in parentheses): azadirachtin (1.55 x 10(-8) M) > dihydroazadirachtin (3.16 x 10(-8) M) > dansyl dihydroazadirachtin (7.40 x 10(-8) M) > DNP-azadirachtin (7.50 x 10(-8) M) > biotin dihydroazadirachtin (1.27 x 10(-7) M) > 11-methoxy 22,23-dihydroazadirachtin (6.67 x 10(-7) M). [Originally published in Volume 34, Archives of Insect Biochemistry and Physiology, 34:461-473 (1997).] Copyright 1997 Wiley-Liss, Inc.     

Characterization of azadirachtin binding to Sf9 nuclei in vitro

[22,23-³H₂]dihydroazadirachtin was incorporated by Sf9 cells in culture and was bound specifically to the nuclear fraction. The observed association constant of the binding of the radioligand to a purified nuclear fraction was determined to be 0.037 ± 0.008 min ¹ using a one-phase exponential association equation, and binding appeared to be to a single population of sites. The binding was essentially irreversible, and the dissociation constant was estimated to be 0.00065 ± 0.00013 min ¹. An association rate constant of 7.3 × 10⁶ M ¹ min ¹ was calculated from these data. Binding was saturable, and the receptor number and affinity were determined as B_max = 23.87 ± 1.15 pmol/mg protein, K_d = 18.1 ± 2.1 nM. The order of potency of semisynthetic azadirachtin analogues for competition for the binding site was as follows (IC₃₀ in parentheses): azadirachtin (1.55 × 10⁻⁸ M) > dihydroazadirachtin (3.16 × 10⁻⁸ M) > dansyl dihydroazadirachtin (7.40 × 10⁻⁸ M) > DNP-azadirachtin (7.50 × 10⁻⁸ M) > biotin dihydroazadirachtin (1.27 × 10⁻⁷ M) ≫ 11-methoxy 22,23-dihydroazadirachtin (6.67 × 10⁻⁷ M). Arch. Insect Biochem. Physiol. 34:461–473, 1997. © 1997 Wiley-Liss, Inc.     

Microtubule independent vesiculation of Golgi membranes and the reassembly of vesicles into Golgi stacks

We have recently shown that ilimaquinone (IQ) causes the breakdown of Golgi membranes into small vesicles (VGMs for vesiculated Golgi membranes) and inhibits vesicular protein transport between successive Golgi cisternae (Takizawa et al., 1993). While other intracellular organelles, intermediate filaments, and actin filaments are not affected, we have found that cytoplasmic microtubules are depolymerized by IQ treatment of NRK cells. We provide evidence that IQ breaks down Golgi membranes regardless of the state of cytoplasmic microtubules. This is evident from our findings that Golgi membranes break down with IQ treatment in the presence of taxol stabilized microtubules. Moreover, in cells where the microtubules are first depolymerized by microtubule disrupting agents which cause the Golgi stacks to separate from one another and scatter throughout the cytoplasm, treatment with IQ causes further breakdown of these Golgi stacks into VGMs. Thus, IQ breaks down Golgi membranes independently of its effect on cytoplasmic microtubules. Upon removal of IQ from NRK cells, both microtubules and Golgi membranes reassemble. The reassembly of Golgi membranes, however, takes place in two sequential steps: the first is a microtubule independent process in which the VGMs fuse together to form stacks of Golgi cisternae. This step is followed by a microtubule-dependent process by which the Golgi stacks are carried to their perinuclear location in the cell. In addition, we have found that IQ has no effect on the structural organization of Golgi membranes at 16 degrees C. However, VGMs generated by IQ are capable of fusing and assembling into stacks of Golgi cisternae at 16 degrees C. This is in contrast to the cells recovering from BFA treatment where, after removal of BFA at 16 degrees C, resident Golgi enzymes fail to exit the ER, a process presumed to require the formation of vesicles. We propose that at 16 degrees C there may be general inhibition in the process of vesicle formation, whereas the process of vesicle fusion is not affected.     

The binding of asialo-glycoprotein to isolated Golgi apparatus

Membranes of the Golgi apparatus isolated from rat liver were capable of binding ¹²⁵I-asialo-fetuin in a manner similar to the binding to liver plasma membranes. Although the binding capacity of the Golgi membranes was less than that of plasma membranes, binding was dependent on Ca⁺⁺ ions and inhibited by α-lactalbumin in both cases. Specific activities of galactosyl and sialyl transferases were about 20 times greater in Golgi than in plasma membranes isolated from the same livers. This dramatic reciprocal relationship between enzyme levels and binding capacities of the two membranous fractions argues against either of these enzymes being the actual binding site.     

Golgi dispersal during microtubule disruption: regeneration of Golgi stacks at peripheral endoplasmic reticulum exit sites.

Microtubule disruption has dramatic effects on the normal centrosomal localization of the Golgi complex, with Golgi elements remaining as competent functional units but undergoing a reversible "fragmentation" and dispersal throughout the cytoplasm. In this study we have analyzed this process using digital fluorescence image processing microscopy combined with biochemical and ultrastructural approaches. After microtubule depolymerization, Golgi membrane components were found to redistribute to a distinct number of peripheral sites that were not randomly distributed, but corresponded to sites of protein exit from the ER. Whereas Golgi enzymes redistributed gradually over several hours to these peripheral sites, ERGIC-53 (a protein which constitutively cycles between the ER and Golgi) redistributed rapidly (within 15 minutes) to these sites after first moving through the ER. Prior to this redistribution, Golgi enzyme processing of proteins exported from the ER was inhibited and only returned to normal levels after Golgi enzymes redistributed to peripheral ER exit sites where Golgi stacks were regenerated. Experiments examining the effects of microtubule disruption on the membrane pathways connecting the ER and Golgi suggested their potential role in the dispersal process. Whereas clustering of peripheral pre-Golgi elements into the centrosomal region failed to occur after microtubule disruption, Golgi-to-ER membrane recycling was only slightly inhibited. Moreover, conditions that impeded Golgi-to-ER recycling completely blocked Golgi fragmentation. Based on these findings we propose that a slow but constitutive flux of Golgi resident proteins through the same ER/Golgi cycling pathways as ERGIC-53 underlies Golgi Dispersal upon microtubule depolymerization. Both ERGIC-53 and Golgi proteins would accumulate at peripheral ER exit sites due to failure of membranes at these sites to cluster into the centrosomal region. Regeneration of Golgi stacks at these peripheral sites would re-establish secretory flow from the ER into the Golgi complex and result in Golgi dispersal.     

The Golgi mitotic checkpoint is controlled by BARS-dependent fission of the Golgi ribbon into separate stacks in G2

The Golgi ribbon is a complex structure of many stacks interconnected by tubules that undergo fragmentation during mitosis through a multistage process that allows correct Golgi inheritance. The fissioning protein CtBP1-S/BARS (BARS) is essential for this, and is itself required for mitotic entry: a block in Golgi fragmentation results in cell-cycle arrest in G2, defining the 'Golgi mitotic checkpoint'. Here, we clarify the precise stage of Golgi fragmentation required for mitotic entry and the role of BARS in this process. Thus, during G2, the Golgi ribbon is converted into isolated stacks by fission of interstack connecting tubules. This requires BARS and is sufficient for G2/M transition. Cells without a Golgi ribbon are independent of BARS for Golgi fragmentation and mitotic entrance. Remarkably, fibroblasts from BARS-knockout embryos have their Golgi complex divided into isolated stacks at all cell-cycle stages, bypassing the need for BARS for Golgi fragmentation. This identifies the precise stage of Golgi fragmentation and the role of BARS in the Golgi mitotic checkpoint, setting the stage for molecular analysis of this process.     

Stop-and-go movements of plant Golgi stacks are mediated by the acto-myosin system

The Golgi apparatus in plant cells consists of a large number of independent Golgi stack/trans-Golgi network/Golgi matrix units that appear to be randomly distributed throughout the cytoplasm. To study the dynamic behavior of these Golgi units in living plant cells, we have cloned a cDNA from soybean (Glycine max), GmMan1, encoding the resident Golgi protein alpha-1,2 mannosidase I. The predicted protein of approximately 65 kD shows similarity of general structure and sequence (45% identity) to class I animal and fungal alpha-1,2 mannosidases. Expression of a GmMan1::green fluorescent protein fusion construct in tobacco (Nicotiana tabacum) Bright Yellow 2 suspension-cultured cells revealed the presence of several hundred to thousands of fluorescent spots. Immuno-electron microscopy demonstrates that these spots correspond to individual Golgi stacks and that the fusion protein is largely confined to the cis-side of the stacks. In living cells, the stacks carry out stop-and-go movements, oscillating rapidly between directed movement and random "wiggling." Directed movement (maximal velocity 4.2 microm/s) is related to cytoplasmic streaming, occurs along straight trajectories, and is dependent upon intact actin microfilaments and myosin motors, since treatment with cytochalasin D or butanedione monoxime blocks the streaming motion. In contrast, microtubule-disrupting drugs appear to have a small but reproducible stimulatory effect on streaming behavior. We present a model that postulates that the stop-and-go motion of Golgi-trans-Golgi network units is regulated by "stop signals" produced by endoplasmic reticulum export sites and locally expanding cell wall domains to optimize endoplasmic reticulum to Golgi and Golgi to cell wall trafficking.     

Redistribution of Golgi stacks and other organelles during mitosis and cytokinesis in plant cells

We have followed the redistribution of Golgi stacks during mitosis and cytokinesis in living tobacco BY-2 suspension culture cells by means of a green fluorescent protein-tagged soybean alpha-1,2 mannosidase, and correlated the findings to cytoskeletal rearrangements and to the redistribution of endoplasmic reticulum, mitochondria, and plastids. In preparation for cell division, when the general streaming of Golgi stacks stops, about one-third of the peripheral Golgi stacks redistributes to the perinuclear cytoplasm, the phragmosome, thereby reversing the ratio of interior to cortical Golgi from 2:3 to 3:2. During metaphase, approximately 20% of all Golgi stacks aggregate in the immediate vicinity of the mitotic spindle and a similar number becomes concentrated in an equatorial region under the plasma membrane. This latter localization, the "Golgi belt," accurately predicts the future site of cell division, and thus forms a novel marker for this region after the disassembly of the preprophase band. During telophase and cytokinesis, many Golgi stacks redistribute around the phragmoplast where the cell plate is formed. At the end of cytokinesis, the daughter cells have very similar Golgi stack densities. The sites of preferential Golgi stack localization are specific for this organelle and largely exclude mitochondria and plastids, although some mitochondria can approach the phragmoplast. This segregation of organelles is first observed in metaphase and persists until completion of cytokinesis. Maintenance of the distinct localizations does not depend on intact actin filaments or microtubules, although the mitotic spindle appears to play a major role in organizing the organelle distribution patterns. The redistribution of Golgi stacks during mitosis and cytokinesis is consistent with the hypothesis that Golgi stacks are repositioned to ensure equal partitioning between daughter cells as well as rapid cell plate assembly.     

COP-coated vesicles are involved in the mitotic fragmentation of Golgi stacks in a cell-free system

Rat liver Golgi stacks fragmented when incubated with mitotic but not interphase cytosol in a process dependent on time, temperature, energy (added in the form of ATP) and cdc2 kinase. The cross-sectional length of Golgi stacks fell in the presence of mitotic cytosol by approximately 50% over 30 min without a corresponding decrease in the number of cisternae in the stack. The loss of membrane from stacked and single cisternae occurred with a half-time of approximately 20 min, and was matched by the appearance of both small (50-100 nm in diameter) and large (100-200 nm in diameter) vesicular profiles. Small vesicular profiles constituted more than 50% of the total membrane after 60 min of incubation and they were shown to be vesicles or very short tubules by serial sectioning. In the presence of GTP gamma S all of the small vesicles were COP-coated and both the extent and the rate at which they formed were sufficient to account for the production of small vesicles during mitotic incubation. The involvement of the COP-mediated budding mechanism was confirmed by immunodepletion of one of the subunits of COP coats (the coatomer) from mitotic cytosol. Vesicles were no longer formed but highly fenestrated networks appeared, an effect reversed by the readdition of purified coatomer. Together these experiments provide strong support for our hypothesis that the observed vesiculation of the Golgi apparatus during mitosis in animal cells is caused by continued budding of COP-coated transport vesicles but an inhibition of their fusion with their target membranes.     

Characterization of clofibrate-induced retrograde Golgi membrane movement to the endoplasmic reticulum: clofibrate distinguishes the Golgi from the trans Golgi network

Clofibrate-induced retrograde Golgi membrane movement was blocked or retarded when NRK cells were treated with sodium azide/2-deoxyglucose, nocodazole, taxol, and destruxin B, indicating that it depends on energy, and the dynamic state of microtubules, and being acidic or vacuolar-type ATPase function. PDMP and phospholipase A2 inhibitors also blocked it. These characteristics are similar to those of brefeldin A (BFA) and nordihydroguaiaretic acid (NDGA), inducers of retrograde Golgi membrane movement. However, clofibrate was distinguished from BFA in that BFA action was insensitive to phospholipase A2 inhibitors and from NDGA in that NDGA stabilized microtubules against nocodazole and its action was almost insensitive to taxol. The trans Golgi network (TGN) was resistant to clofibrate, while BFA and NDGA dispersed it. To our knowledge, clofibrate is the first drug to show such different effects on the Golgi and TGN and, therefore, is expected to be a useful tool to distinguish their architecture and/or membrane dynamics.     

Characterization of Clofibrate-induced Retrograde Golgi Membrane Movement to the Endoplasmic Reticulum: Clofibrate Distinguishes the Golgi from the Trans Golgi Network

Clofibrate-induced retrograde Golgi membrane movement was blocked or retarded when NRK cells were treated with sodium azide/2-deoxyglucose, nocodazole, taxol, and destruxin B, indicating that it depends on energy, and the dynamic state of microtubules, and being acidic or vacuolar-type ATPase function. PDMP and phospholipase A₂ inhibitors also blocked it. These characteristics are similar to those of brefeldin A (BFA) and nordihydroguaiaretic acid (NDGA), inducers of retrograde Golgi membrane movement. However, clofibrate was distinguished from BFA in that BFA action was insensitive to phospholipase A₂ inhibitors and from NDGA in that NDGA stabilized microtubules against nocodazole and its action was almost insensitive to taxol. The trans Golgi network (TGN) was resistant to clofibrate, while BFA and NDGA dispersed it. To our knowledge, clofibrate is the first drug to show such different effects on the Golgi and TGN and, therefore, is expected to be a useful tool to distinguish their architecture and/or membrane dynamics.     

Reconstitution of vesiculated Golgi membranes into stacks of cisternae: requirement of NSF in stack formation

We have developed an in vitro system to study the biochemical events in the fusion of ilimaquinone (IQ) induced vesiculated Golgi membranes (VGMs) into stacks of cisternae. The Golgi complex in intact normal rat kidney cells (NRK) is vesiculated by treatment with IQ. The cells are washed to remove the drug and then permeabilized by a rapid freeze-thaw procedure. VGMs of 60 nm average diameter assemble into stacks of Golgi cisternae by a process that is temperature dependent, requires ATP and a high speed supernatant from cell extract (cytosol), as revealed by immunofluorescence and electron microscopy. The newly assembled stacks are functionally active in vesicular protein transport and contain processing enzymes that carry out Golgi specific modifications of glycoproteins. The fusion of VGMs requires NSF, a protein known to promote fusion of transport vesicles with the target membrane in the exocytic and endocytic pathways. Immunoelectron microscopy using Golgi specific anti-mannosidase II antibody reveals that VGMs undergo sequential changes in their morphology, whereby they first fuse to form larger vesicles of 200-300-nm average diameter which subsequently extend into tubular elements and finally assemble into stacks of cisternae.     

Subcompartment localization of the side chain xyloglucan-synthesizing enzymes within Golgi stacks of tobacco suspension-cultured cells

Xyloglucan is the dominant hemicellulosic polysaccharide of the primary cell wall of dicotyledonous plants that plays a key role in plant development. It is well established that xyloglucan is assembled within Golgi stacks and transported in Golgi-derived vesicles to the cell wall. It is also known that the biosynthesis of xyloglucan requires the action of glycosyltransferases including α-1,6-xylosyltransferase, β-1,2-galactosyltransferase and α-1,2-fucosyltransferase activities responsible for the addition of xylose, galactose and fucose residues to the side chains. There is, however, a lack of knowledge on how these enzymes are distributed within subcompartments of Golgi stacks. We have undertaken a study aiming at mapping these glycosyltransferases within Golgi stacks using immunogold-electron microscopy. To this end, we generated transgenic lines of tobacco (Nicotiana tabacum) BY-2 suspension-cultured cells expressing either the α-1,6-xylosyltransferase, AtXT1, the β-1,2-galactosyltransferase, AtMUR3, or the α-1,2-fucosyltransferase AtFUT1 of Arabidopsis thaliana fused to green-fluorescent protein (GFP). Localization of the fusion proteins within the endomembrane system was assessed using confocal microscopy. Additionally, tobacco cells were high pressure-frozen/freeze-substituted and subjected to quantitative immunogold labelling using anti-GFP antibodies to determine the localization patterns of the enzymes within subtypes of Golgi cisternae. The data demonstrate that: (i) all fusion proteins, AtXT1-GFP, AtMUR3-GFP and AtFUT1-GFP are specifically targeted to the Golgi apparatus; and (ii) AtXT1-GFP is mainly located in the cis and medial cisternae, AtMUR3-GFP is predominantly associated with medial cisternae and AtFUT1-GFP mostly detected over trans cisternae suggesting that initiation of xyloglucan side chains occurs in early Golgi compartments in tobacco cells.     

Characterization of ribosome binding on Rous sarcoma virus RNA in vitro.

We determined the sites at which ribosomes form initiation complexes on Rous sarcoma virus RNA in order to determine how initiation of Pr76gag synthesis at the fourth AUG codon from the 5' end of Rous sarcoma virus strain SR-A RNA occurs. Ribosomes bind almost exclusively at the 5'-proximal AUG codon when chloride is present as the major anion added to the translational system. However, when chloride is replaced with acetate, ribosomes bind at the two 5'-proximal AUG codons, as well as at the initiation site for Pr76gag. We confirmed that the 5'-proximal AUG codon is part of a functional initiation site by identifying the seven-amino acid peptide encoded there. Our results suggest that (i) translation in vitro of Rous sarcoma virus virion RNA results in the synthesis of at least two polypeptides; (ii) the pattern of ribosome binding observed for Rous sarcoma virus RNA can be accounted for by the modified scanning hypothesis; and (iii) the interaction between 40S ribosomal subunits or 80S ribosomal complexes is stronger at the 5'-proximal AUG codon than at sites farther downstream, including the initiation site for the major viral proteins.     

Characterization of caldesmon binding to myosin

Caldesmon inhibits the binding of skeletal muscle subfragment-1 (S-1).ATP to actin but enhances the binding of smooth muscle heavy meromyosin (HMM).ATP to actin. This effect results from the direct binding of caldesmon to myosin in the order of affinity: smooth muscle HMM greater than skeletal muscle HMM greater than smooth muscle S-1 greater than skeletal muscle S-1 (Hemric, M. E., and Chalovich, J. M. (1988) J. Biol. Chem. 263, 1878-1885). We now show that the difference between skeletal muscle HMM and S-1 is due to the presence of the S-2 region in HMM and is unrelated to light chain composition or to two-headed versus single-headed binding. Differences between the binding of smooth and skeletal muscle myosin subfragments to actin do not result from the lack of light chain 2 in skeletal muscle S-1. In the presence of ATP, caldesmon binds to smooth muscle myosin filaments with a stoichiometry of 1:1 (K = 1 x 10(6) M-1). Similar results were obtained for the binding of caldesmon to smooth muscle rod as well as the binding of the purified myosin-binding fragment of caldesmon to smooth muscle myosin. The binding of caldesmon to intact myosin is ATP sensitive. The interaction of caldesmon with myosin is apparently specific and sensitive to the structure of both proteins.     

Characterization of monoclonal antibodies to Acanthamoeba myosin-I that cross-react with both myosin-II and low molecular mass nuclear proteins

We characterized nine monoclonal antibodies that bind to the heavy chain of Acanthamoeba myosin-IA. Eight of these antibodies bind to myosin-IB and eight cross-react with Acanthamoeba myosin-II. All but one of the antibodies bind to a 30-kD chymotryptic peptide of myosin-IA that derives from the COOH terminus of the molecule, and to tryptic peptides as small as 17 kD, hence these epitopes are clustered closely together on the heavy chain. None of the antibodies prevent heavy chain phosphorylation by myosin-I heavy chain kinase. One antibody inhibits the K+-EDTA ATPase activity and three antibodies inhibit the actin- activated Mg++-ATPase activity of myosin-I under the set of conditions that we tested. When fluorescent antibody staining of both whole cells and isolated nuclei is done, several of these monoclonal antibodies react strongly with nuclei. These antibodies also stain the cytoplasmic matrix, especially the cortex near the plasma membrane. All nine of the monoclonal antibodies bind to polypeptides of 30-34 kD that are highly enriched in nuclei isolated from Acanthamoeba. There is no myosin-I in the isolated nuclei, so the 30-34-kD polypeptides, not myosin-I, are responsible for the nuclear staining.