Immunofluorescence staining of cultured human umbilical vein endothelial cells has shown the presence of von Willebrand protein in the perinuclear region, in small rodlike structures through the cytoplasm, and on filaments of the extracellular matrix. Nonendothelial cells showed no staining with anti-von Willebrand protein antiserum. At the light microscope level, immunoperoxidase treatment of endothelial cells revealed the same pattern and antibody specificity as the fluorescence staining. Thin sections of the peroxidase-stained cells showed decorated filaments close to the substratum and also specific deposits in the endoplasmic reticulum and WeibeI-Palade bodies. Control antisera against other selected proteins in endothelial cells failed to stain the WeibeI-Palade bodies. These data suggest that the WeibeI-Palade bodies of endothelial cells are storage and/or processing organelles for von Willebrand protein.Von Willebrand protein is a large glycoprotein of complex multimeric structure (1, 2) that mediates attachment ofplatelets to the subendothelium after vascular injury (3). It is synthesized by megakaryocytes (4), which assure its presence in platelets in granulelike storage compartments (5, 6). After activation, platelets bind both von Willebrand protein released from internal storage sites and protein recruited from plasma to their surface membrane (7). Endothelial cells also synthesize von Willebrand protein (8) and the low concentration present in plasma and the subendothelium is probably derived from this source. We studied the distribution of von Willebrand protein in endothelial cells, in an attempt to detect and identify a storage compartment which would allow rapid release of this protein upon appropriate stimulus or physiologic demand. Using indirect immunofluorescence and immunoelectron microscopy, we determined that von Willebrand protein is concentrated in Weibel-Palade bodies. These are membrane-bound, elongate vesicles of 0.1 x 2-3 #In size that contain regularly spaced tubular structures aligned parallel to the longitudinal axis (9). Our data suggest that Weibel-Palade bodies serve as storage and/or processing vesicles for this protein and provide the first demonstration of unique function for these endothelial cell-specific organelles. MATERIALS AND METHODS Cells and Culture ConditionsEndothelial cells were obtained from human umbilical vein using a modification of the method described by Gimbrone et al.(10). Mild proteolytic digestion was carried out with 5 mg/ml of pronase (Calbiochem-Behring,

No abstract

No abstract

No abstract

No abstract

The ionic and nucleotide requirements for the in vitro polymerization of microtubules from purified brain tubulin have been characterized by viscometry. Protein was purified by successive cycles of a temperature dependent assembly-diassembly scheme. Maximal polymerization occurred at a concentration of 0.1 M Pipes (piperazine-N,N'-bis(2-ethanesulfonic acid)); increasing ionic strength by addition of NaCl to samples prepared in lower buffer concentrations did not result in an equivalent level of polymerization. Both Na-+ and K-+ inhibited microtubule formation at levels greater than 240 mM, withmaximal assembly occurring at physiological concentrations of 150 mM. Maximal extent of assembly occurred at pH 6.8 and optimal rate at pH 6.6. Inhibition of polymerization was half-maximal at added calcium concentrations of 1.0 mM and magnesium concentrations of 10.0 mM. EGTA (ethylene glycol bis(beta-aminoethyl ether)tetraacetic acid), which chelates Ca-2+, had no effect on polymerization over a concentration range of 0.01-10.0 mM. In contrast, EDTA (ethylenediaminetetraacetic acid), which chelates both Mg-2+ and Ca-2+, inhibited assemble half-maximally at 0.25 mM and totally at 2.0 mM. As determined from experiments using Mg-2+-EDTA buffers, magnesium was required for polymerization. Magnesium promoted the maximal extent of assembly at substoichiometric levels relative to tubulin, but was maximal for both rate and extent at stoichiometric concentrations. Elemental analyses indicated that approximately 1 mol of magnesium was tightly bound/mol of tubulin dimer. Viscosity development was dependent upon hydrolyzable nucleoside triphosphate, and stoichiometric levels of GTP were sufficient for maximal polymerization. The effect of magnesium in increasing the rate of GTP-dependent polymerization suggests that a Mg-2+-GTP complex is the substrate required for a step in assembly.

No abstract

Abstract. MAP 4 is a ubiquitous microtubule-associated protein thought to play a role in the polymerization and stability of microtubules in interphase and mitotic cells. We have analyzed the behavior of protein domains of MAP 4 in vivo using chimeras constructed from these polypeptides and the green fluorescent protein (GFP). GFP-MAP 4 localizes to microtubules; this is confirmed by colocalization of GFP-MAP 4 with microtubules that have incorporated microinjected rhodamine-tubulin, and by loss of localized fluorescence after treatment of cells with anti-microtubule agents. Different subdomains of MAP 4 have distinct effects on microtubule organization and dynamics. The entire basic domain of MAP 4 reorganizes microtubules into bundles and stabilizes these arrays against depolymerization with nocodazole. Within the basic domain, the PGGG repeats, which are conserved with MAP 2 and tau, have a weak affinity for microtubules and are dispensable for microtubule binding, whereas the MAP 4-unique PSP region can function independently in binding. The projection domain shows no microtubule localization, but does modulate the association of various binding subdomains with microtubules. The acidic carboxy terminus of MAP 4 strongly affects the microtubule binding characteristics of the other domains, despite constituting less than 6% of the protein. These data show that MAP 4 association with microtubules is modulated by sequences both within and outside the basic domain. Further, our work demonstrates that GFP chimeras will allow an in vivo analysis of the effects of MAPs and their variants on microtubule dynamics in real time.