Journal of Cell Science 107, 2249-2257 (1994) 2249 Printed in Great Britain © The Company of Biologists Limited 1994 Isolation of polypeptides with microtubule-translocating activity from phragmoplasts of tobacco BY-2 cells Tetsuhiro Asada* and Hiroh Shibaoka Department of Biology, Faculty of Science, Osaka University, Osaka 560, Japan *Author for correspondence SUMMARY As part of our efforts to understand the molecular basis of caused by the fraction that contained the 125 kDa and 120 the microtubule-associated motility that is involved in kDa polypeptides as main components was 1.28 µm/minute cytokinesis in higher plant cells, an attempt was made to in the presence of ATP and 0.50 µm/minute in the presence identify proteins with the ability to translocate micro- of GTP. This fraction contained some microtubule-associ- tubules in an extract from isolated phragmoplasts. Homog- ated polypeptides in addition to the 125 kDa and 120 kDa enization of isolated phragmoplasts in a solution that polypeptides, but a fraction that contained only these addi- contained MgATP, MgGTP and a high concentration of tional polypeptides did not cause any translocation of NaCl resulted in the release from phragmoplasts of factors microtubules. Thus, it appeared that the 125 kDa and 120 with ATPase and GTPase activity that were stimulated by kDa polypeptides were responsible for translocation of microtubules. A protein fraction with microtubule- microtubules. These polypeptides with plus-end-directed dependent ATPase and GTPase activity caused minus-end- motor activity may play an important role in formation of headed gliding of microtubules in the presence of ATP or the cell plate and in the organization of the phragmoplast. GTP. Polypeptides with microtubule-translocating activity cosedimented with microtubules that had been assembled Abbreviations: PIPES, 1,4-piperazinebis(ethanesulfonic acid); TAME, in vitro from brain tubulin and were dissociated from sed- p-tosyl-arginine methyl ester; DTT, dithiothreitol; PMSF, phenyl- methylsulfonyl fluoride; AMPPNP, 5′-adenylylimidodiphosphate; imented microtubules by addition of ATP or GTP. After GMPPNP, 5′-guanylylimidodiphosphate; SDS-PAGE, sodium cosedimentation and dissociation procedures, a 125 kDa dodecylsulfate-polyacrylamide gel electrophoresis. polypeptide and a 120 kDa polypeptide were recovered in a fraction that supported minus-end-headed gliding of Key words: tobacco BY-2 cell, phragmoplast, cytokinesis, microtubules. The rate of microtubule gliding that was microtubule motor protein INTRODUCTION of the cell plate, several lines of evidence support the hypoth- esis that the mechanisms involve the microtubule-associated When cells of higher plants divide, microtubules function not motility systems. During cytokinesis in endosperm cells of only in mitosis but also in cytokinesis (Gunning, 1982; Baskin Haemanthus, organelles in the phragmoplasts move towards and Cande, 1990). At the late stage of mitosis, microtubules and away from the equatorial plane, suggesting that motility between separating anaphase chromosomes increase in number systems for the transport of organelles are associated with the to form the cytokinetic apparatus, known as the phragmoplast phragmoplast (Bajer et al., 1987). Electron microscopic obser- (Zhang et al., 1990). The phragmoplast consists of two oppo- vations indicating the close association of vesicles with micro- sitely directed sets of microtubules with their plus ends inter- tubules of the phragmoplast support the hypothesis that digitating on the equatorial plane (Euteneur and McIntosh, vesicles that include cell-plate material are transported to the 1980; Hepler and Jackson, 1968), where a number of Golgi- equatorial plane along tracks laid down by phragmoplast derived vesicles accumulate and fuse to form an immature microtubules (Kakimoto and Shibaoka, 1988). The involve- cross-wall, the cell plate (Hepler, 1982). The disruption of ment of microtubule-associated motility systems in the organ- microtubules after the separation of chromosomes results in ization of the phragmoplast was suggested by our previous binuclear cells (Palevitz and Hepler, 1974; Hardham and study (Asada et al., 1991) in which we demonstrated that Gunning, 1980). Thus, the phragmoplast microtubules appear phragmoplast microtubules were translocated away from the to be essential for formation of the cell plate. equatorial plane at the expense of energy from GTP or ATP. Although little is known about the molecular mechanisms Although these movements of organelles or microtubules in the responsible for the organization of phragmoplast microtubules phragmoplast can be accounted for by the activities of motor and the accumulation of the vesicles at the plane of formation proteins that generate forces that act to cause microtubule- 2250 T. Asada and H. Shibaoka related movement, such motor proteins have not been identi- (>80%) of which were in telophase, were pelleted, washed twice with fied. a 0.5 M solution of mannitol, and suspended in a glycerination buffer Microtubule-associated motor proteins, such as dynein, (50 mM K-PIPES, pH 7.0, 1 mM MgSO4, 10 mM EGTA, 60% (v/v) kinesin and kinesin-like proteins, have been reported to be glycerol, 2 mM DTT, 1 mM PMSF, 5 µg/ml leupeptin, 3 µg/ml µ required for a variety of cellular processes, such as flagellar pepstatin A, 50 g/ml TAME). Cells in the glycerination buffer were kept on ice for 5 minutes and then washed three times with PME movement, the transport of organelles and mitosis (Gibbons, buffer (75 mM K-PIPES, pH 7.0, 1 mM MgSO4, 1 mM EGTA, 1 mM 1981; Sawin and Scholey, 1991; Paschal and Vallee, 1987; DTT, 0.2 mM PMSF, 50 µg/ml TAME). During the repeated washing, Vale et al., 1985). Recent studies on microtubule motor plasma membranes of protoplasts were ruptured and complexes proteins in higher plant cells have focused on kinesin and have composed of a phragmoplast and two daughter nuclei were released relied on the use of antibodies against bovine brain kinesin, or (see Fig. 1). Such released complexes will be called isolated phrag- strategies involving the polymerase chain reaction (PCR). Cai moplast complexes for convenience. The isolated phragmoplasts were et al. (1993) identified a 100 kDa polypeptide that reacted with extracted in four different ways to yield four different kinds of extract. a monoclonal antibody against kinesin in pollen tubes of The complexes were pelleted by centrifugation, mixed with an equal Nicotiana tabacum and they purified this protein. The purified volume of: (1) PME buffer; (2) PME buffer that contained 2.5 mM MgATP and 2.5 mM MgGTP; (3) PME buffer that contained 0.3 M 100 kDa polypeptide had microtubule-stimulated ATPase NaCl; or (4) PME buffer that contained 2.5 mM MgATP, 2.5 mM activity and bound to microtubules in an ATP-sensitive MgGTP and 0.3 M NaCl. Each suspension was then homogenized manner, as does kinesin itself. Another study, employing PCR, with glass homogenizer on ice. Homogenates were centrifuged for 1 resulted in the identification of a gene, katA, encoding a hour at 132,000 g at 4°C. The four kinds of extract obtained by these kinesin-related polypeptide of 90 kDa in Arabidopsis thaliana procedures were named: (1) control extract; (2) ATP/GTP extract; (3) (Mitsui et al., 1993). In spite of these studies, no direct NaCl extract; and (4) ATP/GTP/NaCl extract, respectively. Each evidence has been presented to demonstrate that these kinesin- extract was dialyzed against PME buffer for 6 hours at 4°C and the related or kinesin-like polypeptides have microtubule-translo- resultant precipitates were removed by centrifugation. The concen- cating activity. Without the isolation and characterization of trations of protein in the dialyzed extracts were 0.52, 0.73, 1.18 and 1.35 mg/ml, respectively. The dialysates were kept on ice until proteins with the microtubule-translocating activity, we cannot they were assayed for their ATPase or GTPase activities. The expect to have a full understanding of the molecular mecha- ATP/GTP/NaCl extract (number 4 above) was further fractionated. nisms of microtubule-associated motility, which may be The extract was desalted by gel filtration with a column of Sephadex important in both cell division and cell morphogenesis in G-25 (Pharmacia LKB, Uppsala, Sweden) and the proteins were higher plant cells (Bajer et al., 1987; Williamson, 1991). applied to a column of DEAE-Sephacel (Pharmacia) that had been As part of our effort to understand the molecular basis of the pre-equilibrated with PME buffer that contained 0.1 mM ATP and 0.1 microtubule-associated motility that is involved in cytokinesis mM GTP. The column was washed with the same buffer and adsorbed of higher plant cells, we attempted to identify proteins that are proteins were eluted stepwise with PME buffers that contained 80, associated with the phragmoplast and have the ability to 160 and 400 mM NaCl. The flow-through fraction and eluted fractions were dialyzed against PME buffer, and were assayed for ATPase and translocate microtubules. A well established method for syn- GTPase activities and for microtubule-translocating activity. The con- chronization of the cell cycle (Kakimoto and Shibaoka, 1988) centrations of protein in the flow-through fraction, and in the fractions enabled us to isolate motor proteins directly from phragmo- eluted by 80, 160, 400 mM NaCl were 0.19, 0.38, 0.23 and 0.59 plasts that had been isolated from tobacco BY-2 cells in mg/ml, respectively. The concentration of protein was determined by telophase. In this report, we describe the isolation of polypep- the method of Bradford (1976) in each case. tides with microtubule-translocating activity from isolated phragmoplasts. Isolation of microtubule-binding polypeptides Tubulin was prepared from bovine brains by a two-cycle assembly and disassembly procedure that was followed by purification on a MATERIALS AND METHODS phosphocellulose column (Shelanski et al., 1973). The brain tubulin obtained was stored at −70°C until use. A 35 ml sample of the ATP/GTP/NaCl extract was applied to a column of DEAE-Sephacel.