A Novel Plant Kinesin-Related Protein Specifically Associates with the Phragmoplast Organelles

The Plant Cell, Vol. 13, 2427–2439, November 2001, www.plantcell.org © 2001 American Society of Plant Biologists

A Novel Plant Kinesin-Related Protein Specifically Associates with the Phragmoplast Organelles

Y.-R. Julie Lee, Hoa M. Giang, and Bo Liu

1

Section of Plant Biology, University of California, Davis, California 95616-8537

In higher plants, the formation of the cell plate during cytokinesis requires coordinated microtubule (MT) reorganizationand vesicle transport in the phragmoplast. MT-based kinesin motors are important players in both processes. To un-derstand the mechanisms underlying plant cytokinesis, we have identified AtPAKRP2 (for

Arabidopsis

thaliana

phrag-moplast-associated kinesin-related protein 2). AtPAKRP2 is an ungrouped N-terminal motor kinesin. It first appeared ina punctate pattern among interzonal MTs during late anaphase. When the phragmoplast MT array appeared in a mirrorpair, AtPAKRP2 became more concentrated near the division site, and additional signal could be detected elsewhere inthe phragmoplast. In contrast, the previously identified AtPAKRP1 protein is associated specifically with bundles ofMTs in the phragmoplast at or near their plus ends. Localization of the tobacco homolog(s) of AtPAKRP2 was alteredby treatment of brefeldin A in BY-2 cells. We discuss the possibility that AtPAKRP1 plays a role in establishing and/ormaintaining the phragmoplast MT array, and AtPAKRP2 may contribute to the transport of Golgi-derived vesicles in thephragmoplast.

INTRODUCTION

In higher plants, cytokinesis involves the formation of the cellplate, which depends on the cytokinetic apparatus of thephragmoplast. The phragmoplast has a framework of micro-tubules (MTs) that are oriented perpendicular to the divisionplane by having their plus ends at or near the division site.

Phragmoplast MTs are highly dynamic, as revealed by flu-orescent analog histochemistry and the green fluorescentprotein–tagging approach (Zhang et al., 1993; Granger andCyr, 2000; Hasezawa et al., 2000). Derived from interzonalMTs of the anaphase spindle, phragmoplast MTs are ar-ranged in the pattern of a solid cylinder before the cell plateis formed. Concomitant with the centrifugal buildup of thecell plate, phragmoplast MTs depolymerize in the region inwhich the cell plate has been formed. Remarkably, at thesame time, new MTs form at the periphery of the phragmo-plast. During the whole process of phragmoplast develop-ment, MT polymerization takes place at the MT plus ends,and new MT segments are constantly translocated awayfrom the division site (Vantard et al., 1990; Asada et al.,1991). It has been shown that stabilization of MTs with taxolprevents the centrifugal growth of the phragmoplast MT ar-ray (Yasuhara et al., 1993). The disassembly of phragmo-plast MTs requires the arrival of Golgi-derived vesicles,

because disruption of the Golgi apparatus with brefeldin Aprevents the depolymerization event in the center of thephragmoplast (Yasuhara and Shibaoka, 2000).

During cell plate formation, Golgi-derived vesicles aretransported rapidly along the MTs toward their plus ends.The vesicles contain xyloglucans and other componentsthat contribute to the cell plate (Samuels et al., 1995; Oteguiand Staehelin, 2000; Sonobe et al., 2000; Verma, 2001).Vesicle transport along MTs depends on MT-based motorproteins, the dyneins and the kinesins (Goldstein and Philp,1999). Although a number of kinesin-related proteins (KRPs)have been reported in higher plants (Reddy, 2001), the mo-tor(s) responsible for vesicle transport in the phragmoplasthas not yet been identified.

Motor proteins play critical roles in multiple processesduring cell division (Sharp et al., 2000; Wittmann et al.,2001). Kinesin and KRPs are members of the kinesin super-family (Kim and Endow, 2000). In higher plants, severalKRPs have been localized to the phragmoplast (Liu and Lee,2001). Plus end–directed KRPs in the Block-In-Mitosis Cprotein (BIMC) subfamily are required for MT translocation inthe phragmoplast, likely by a sliding filament mechanism(Asada et al., 1997; Sharp et al., 1999). One such KRP fromcarrot especially concentrates in the midline of the phrag-moplast MTs (Barroso et al., 2000). Minus end–directedKRPs from the C-terminal motor KRP subfamily decoratephragmoplast MTs as well, but their roles have not been de-termined (Liu and Palevitz, 1996; Liu et al., 1996; Mitsui etal., 1996; Bowser and Reddy, 1997; Smirnova et al., 1998).

1

To whom correspondence should be addressed. E-mail [email protected]; fax 530-752-5410.Article, publication date, and citation information can be found atwww.aspb.org/cgi/doi/10.1105/tpc.010225.

2428 The Plant Cell

Among them, Kinesin-like Calmodulin-Binding Protein(KCBP) appears to be inactivated, possibly by Ca

2

�

/cal-modulin binding in the phragmoplast, although it appears tobe associated with the phragmoplast MTs (Vos et al., 2000).

In a previous study, we identified AtPAKRP1 (for

Arabi-dopsis

thaliana

phragmoplast-associated kinesin-relatedprotein 1), an N-terminal motor KRP that does not resembleKRPs from other organisms (Lee and Liu, 2000). It does notassociate with the preprophase band or the mitotic spindle.During late anaphase, however, it appears along interzonalMTs. Later, it decorates phragmoplast MTs at or near theirplus ends. AtPAKRP1 likely plays a role in establishing and/or maintaining the phragmoplast MT array.

One of the central questions about phragmoplast opera-tion is what the force generator(s) for unidirectional vesicletransport is during cell plate formation. Here, we report aphragmoplast-specific KRP, AtPAKRP2 (Arabidopsis phragmo-plast-associated kinesin-related protein 2), which is namedbased on its intracellular localization pattern. AtPAKRP2 hasdistinct structure and localization patterns that differ fromthose of AtPAKRP1. On the basis of pharmacological data,we suggest that AtPAKRP2 is a candidate motor for trans-porting Golgi-derived vesicles toward the division site.

RESULTS

Isolation of AtPAKRP2 cDNA

To identify KRPs in Arabidopsis, the expressed sequencetag clone OAO358 from green shoots (GenBank accession

number Z34049) was obtained because it overlapped withthe AtFCA1.2 locus, which was predicted to encode a KRP(GenBank accession number Z97336). After being se-quenced, however, the OAO358 clone did not show a full-length open reading frame. To determine the 5

�

end of thecoding sequence, 5

�

rapid amplification of cDNA ends(RACE) experiments were performed. We predicted that thefull-length open reading frame would encode a polypeptideof 869 amino acids, which has been named AtPAKRP2(Figure1). The deduced polypeptide has a calculated molec-ular mass of 97 kD and a pI of 5.8.

The deduced AtPAKRP2 sequence showed a number ofdifferences from the amino acid sequence of the AtFCA1.2locus annotated by the Arabidopsis Genome Initiative, es-pecially toward the C terminus. This difference was causedby the discrepancy of intron/exon prediction by the Arabi-dopsis Genome Initiative and the cDNA sequence deter-mined in the present study. The N-terminal part of AtPAKRP2(amino acids 32 to 375) resembled the kinesin motor do-main, although it had some unusual features (Figure 1). Typ-ically, the motor domain of kinesin/KRPs contains an ATPbinding site, which includes a highly conserved peptide ofIF/VAYGQTGA/SGKS/T. The corresponding sequence inAtPAKRP2, however, was IMMYGPTGAGKS. Nevertheless,the AtPAKRP2 motor domain clearly had well-conservedmotifs, such as SSRSH, LVDMAGSE, and HVPFRDSKL,which have been found commonly in the MT binding site.

To date, identified KRPs are grouped into nine differentsubfamilies plus some ungrouped ones (Kim and Endow,2000). To analyze the relationship between AtPAKRP2 andother KRPs in the kinesin superfamily, phylogenetic analysiswas performed by aligning its motor sequence and the mo-

Figure 1. AtPARKP2 Is a Divergent KRP.

(A) The amino acid sequence was deduced from AtPAKRP2 cDNA clones. The N-terminal region similar to KRP motor domains is shaded. Thehighly conserved peptides in the motor domain are highlighted with black background.(B) Phylogenetic tree built from a sequence alignment of KRP motor domains.

A Novel Phragmoplast-Associated Kinesin 2429

tor sequences of 17 representative kinesin/KRPs, includingmost if not all reported plant KRPs. First, AtPAKRP2 wasvery divergent from any of these plant KRPs (Figure 1B).Second, AtPAKRP2 did not fall into any of the existing sub-families. Therefore, we considered AtPAKRP2 an ungroupedor orphan KRP, like many others such as the

Saccharomy-ces

cerevisiae

Smy1p protein (http://www.blocks.fhcrc.org/~kinesin/BE6c_ungrouped.html).

Many kinesin/KRPs bear coiled-coil domains required foroligomerization. To further analyze whether AtPAKRP2 mightoligomerize, its polypeptide sequence was analyzed for theprobability of the formation of coiled coils. Coiled coilscould be formed near the central region of the polypeptide(amino acids 390 to 517; P

�

0.5) (Figure 2A). The C-termi-nal region after the coiled coils does not have significantsimilarity to any available sequence in GenBank, nor is anystructural motif found in this so-called tail domain. A typicaltripartite structure was predicted for AtPAKRP2 (Figure 2B).It would not be surprising if AtPAKRP2 formed a dimer in itsnative form, given such a tripartite structure.

AtPAKRP2 Binds to MTs in an ATP-Dependent Manner

Because AtPAKRP2 is very divergent from other KRPs, andthe ATP binding site of AtPAKRP2 had three amino acidsubstitutions compared with the majority of kinesin super-family members, we wondered whether it could still behavelike a motor protein. Most members of the kinesin super-family bind to MTs in an ATP-dependent manner. A glutathi-one

S

-transferase (GST) fusion protein (GST-AtPAKRP2-M)containing the AtPAKRP2 motor domain (amino acids 1 to435) was expressed in

Escherichia coli

and affinity purified(Figure 2B). This fusion protein was subjected to MT cosedi-mentation experiments as described previously (Lee andLiu, 2000). In the absence of exogenous nucleotide, GST-AtPAKRP2-M largely cosedimented with the MT pellet (Fig-ure 2C). The presence of ATP rendered at least 50% of thefusion protein in the supernatant instead of in the MT pellet(Figure 2C). When ATP was replaced with the nonhydrolyz-able ATP analog adenylylimidodiphosphate (AMPPNP), thefusion protein appeared almost exclusively in the MT pellet(Figure 2C). Apparently, the fusion protein itself did not sed-iment (Figure 2C). In a control experiment, the fusion tagGST alone did not show any sign of cosedimentation withMTs (data not shown). Therefore, we concluded thatAtPAKRP2 behaved like a typical kinesin/KRP in vitro.

Generation of Anti-AtPAKRP2 Antibodies

Antibodies against two independent regions of AtPAKRP2were generated. Affinity-purified anti-PAKRP2-N

�

and anti-PAKRP2-T antibodies from four different animals gaveidentical intracellular localization patterns in dividing Arabi-dopsis cells (see below). Therefore, we concluded that both

Figure 2. AtPAKRP2 Structure and Its ATP-Dependent MT BindingActivity.

(A) Coiled-coil domains of AtPAKRP2 were predicted by the Lupusalgorithm. The x axis represents the amino acid residue number, andthe y axis represents the probability of the formation of coiled coils.A P value �0.5 indicated that the region likely forms a coiled coil.(B) Schemes of AtPARKP2 and its protein constructs used for MTbinding and antibody preparation. The relative positions of the mo-tor, coiled coil, and tail are indicated.(C) The MT binding property of the GST-AtPAKRP2-M fusion proteinwas demonstrated by cosedimentation with MTs. The resulting su-pernatant (S) and pellet (P) were analyzed by SDS-PAGE and visual-ized by Coomassie blue staining. (�), the GST-AtPARKP2-M fusionprotein without addition of MTs; MTs, the fusion protein incubatedwith MTs alone; MTs � ATP, the fusion protein incubated with MTsand 2 �M ATP; MTs � AMPPNP, the fusion protein incubated withMTs and 2 �M AMPPNP.

2430 The Plant Cell

anti-PAKRP2-N

�

and anti-PAKRP2-T antibodies recognizedthe native AtPAKRP2 protein.

Immunoblotting experiments were performed to detectthe AtPAKRP2 polypeptide from protein extracts of etiolatedArabidopsis seedlings and young inflorescence buds. Wecould not detect the protein in extracts from etiolated seed-lings. But in protein extracts of young inflorescence buds, a

single 88-kD band was detected with anti-PAKRP2-N

�

anti-bodies (Figure 3A). Unfortunately, the anti-PAKRP2-T anti-bodies did not work in immunoblotting, possibly because ofthe availability of the corresponding epitope(s) in our proteinpreparations. Because of the difference in the apparent mo-lecular mass and the calculated 97 kD, we wanted to verifywhether this 88-kD band was in fact the AtPAKRP2 protein.We preincubated the antibodies with the AtPAKRP2-N

�

fu-sion polypeptide before they were applied to the proteinblots. The preincubation prevented the 88-kD band from be-ing revealed (Figure 3A). These results led us to believe thatour antibodies had recognized the native AtPAKRP2 pro-tein.

AtPAKRP2 Is Concentrated in the Phragmoplast

To gain insights into the function of AtPAKRP2 in plant cells,we used purified anti-AtPAKRP2-N

�

and anti-AtPAKRP2-Tantibodies to perform immunolocalization experiments incells from three different species. Although both antibodiesgave identical localization signals in Arabidopsis and

Bras-sica

oleracea

cells, only anti-AtPAKRP2-N

�

antibodies workedin tobacco BY-2 cells.

When the purified antibodies were used for immunofluo-rescence in Arabidopsis root tip cells, they labeled the cen-tral region of telophase cells (Figure 3B). To verify thespecificity of the purified antibodies for immunolocalization,the anti-AtPAKRP2-N

�

antibodies were subjected to prein-cubation with the AtPAKRP2-N

�

polypeptide before theywere applied to the cells. The preincubation completelyabolished the localization signal (Figure 3C). Identical resultswere obtained when the anti-AtPAKRP2-T antibodies wereused (data not shown).

Because the AtPAKRP2 protein obviously was present inthe dividing cells, we wondered whether its localizationchanged dynamically during the cell division cycle by usingArabidopsis root tip cells that were dividing actively (Figures4 and 5). Dual localizations of AtPAKRP2 and tubulin wereperformed. During interphase, AtPAKRP2 was presentthroughout the cytoplasm and the nucleus, with a concen-tration of signals at or near the periphery of the nucleus (Fig-ures 4A to 4D). During the progression of mitosis, no signalwas detected in the MT preprophase band and in MTs onthe nuclear envelope (data not shown). In metaphase cells,judged by their spindle MTs, very little if any anti-AtPAKRP2signal could be detected with MT bundles by fluorescencemicroscopy (Figures 4E to 4H). The same was true for earlyanaphase cells (data not shown). When sister chromatidswere separated completely during anaphase, AtPAKRP2appeared among the interzonal MTs between segregatedchromatids (Figures 4I to 4L). Unlike clear bundles of MTs inthe spindle midzone, AtPAKRP2 appeared in a punctatepattern among these MT bundles.

After the completion of anaphase, MTs coalesced andwere organized into an early phragmoplast array with a dark

Figure 3. Immunodetection of the AtPAKRP2 Protein in Arabidopsis.

(A) Immunoblots containing proteins extracted from inflorescencebuds were stained with affinity-purified anti-AtPAKRP2-N� antibod-ies (lane 1) or anti-AtPAKRP2-N� antibodies incubated with GST-AtPAKRP2-N� proteins before immunoblot staining (lane 2). Molecu-lar mass markers (kD) are shown at left.(B) Intracellular localization of AtPAKRP2 in an Arabidopsis root tipcell at cytokinesis. AtPAKRP2 is shown in green, and DNA is shownin blue.(C) Anti–AtPAKRP2-N� staining was absent in an Arabidopsis roottip cell at telophase when the antibodies were incubated with theGST-AtPAKRP2-N� protein before immunofluorescence staining.(D) MTs (red) and DNA (blue) in the same cell as in (C). Bar � 5 �mfor (B) to (D).


midline between the two mirror halves, as judged by anti-tubulin staining (Figure 5B). At this stage, AtPAKRP2 ap-peared very concentrated at the midline of the phragmo-plast, and more punctate AtPAKRP2 signal could bedetected weakening away from this midline (Figures 5A to5D). Although the phragmoplast MT array expanded towardthe cell periphery, AtPAKRP2 remained in a punctate pat-tern among phragmoplast MTs, with a peak of fluorescenceintensity near the midline (Figures 5E to 5H). The expansionof AtPAKRP2 signal was concomitant with the expansion ofthe MT array. MTs depolymerized from the central region ofthe phragmoplast, as indicated by the anti-tubulin fluores-cence signal (Figure 5J) upon the formation of daughter nu-clei. At this stage, AtPAKRP2 remained distributed among

phragmoplast MTs by being pronounced toward the mid-line, and the anti-AtPAKRP2 signal was more obvious wheremore MTs were present toward the periphery of the phrag-moplast (Figures 5I and 5L). In the area in which MTs hadbeen completely depolymerized, there was little if anyAtPAKRP2 signal (Figure 5I). Therefore, the results sug-gested that AtPAKRP2 localization correlated with phrag-moplast MT organization.

Because AtPAKRP2 behaved like a typical kinesin motorlocalized among phragmoplast MTs, we tested whether thelocalization was dependent on MTs. The MT depolymeriza-tion agent amiprophos methyl (APM) was used to depoly-merize MTs in root cells. After Arabidopsis seedlings weretreated with 100

�

M APM for 2 hr, no filamentous structure

Figure 4. AtPAKRP2 Was Localized among Interzonal MTs.

Triple localizations of AtPAKRP2 with anti-AtPAKRP2-N� ([A], [E], and [I]), of MTs with anti–�-tubulin ([B], [F], and [J]), and of DNA with DAPI([C], [G], and [K]) are shown in Arabidopsis root tip cells at interphase ([A] to [D]), metaphase ([E] to [H]), and late anaphase ([I] to [L]). Thepseudocolored composite images ([D], [H], and [L]) show AtPAKRP2 in green, �-tubulin in red, and DNA in blue. Bar in (L) � 5 �m for (A) to (L).

2432 The Plant Cell

was detected in the root cells by anti-tubulin immunofluo-rescence, indicating that MTs had been depolymerizedcompletely (Figure 6B). In this APM-treated cell with reform-ing daughter nuclei indicated by 4

�

,6-diamidino-2-phenylin-dole (DAPI) staining (Figure 6C), AtPAKRP2 no longer haddistinct localization between the nuclei; rather, it becamedistributed diffusely across the cell (Figure 6A). In the controlcells treated with the identical amount of isopropanol usedto dissolve APM, intact MTs and phragmoplast-localizedAtPAKRP2 were observed (Figures 6D to 6F). Therefore, weconcluded that the localization of the AtPAKRP2 protein inthese root cells was dependent on the integrity of MTs.

Distinct Localization Patterns of AtPAKRP2and AtPAKRP1

In a previous study, another Arabidopsis motor, AtPAKRP1,was shown to localize in the midline of the phragmoplastMTs (Lee and Liu, 2000). We wanted to determine whetherAtPAKRP2 colocalized with AtPAKRP1 in the phragmoplast.Dual localizations of AtPAKRP2 and AtPAKRP1 were per-formed in Arabidopsis cells at different stages of cell divi-

sion (Figure 7). At an early stage of cytokinesis, as judged bythe appearance of the reforming daughter nuclei (Figure 7C),AtPAKRP2 was detected as particles with different sizes be-tween two daughter nuclei (Figure 7A). Compared with thepunctate localization of AtPAKRP2 in the phragmoplast,AtPAKRP1 was much more discrete and concentrated inthe midline (Figures 7A and 7B). At a later stage of cytokine-sis, the punctate localization pattern of AtPAKRP2 was stillclear in the phragmoplast (Figure 7D). AtPAKRP1, however,remained exclusively at the equatorial plane (Figure 7E). Werealized the limitation of the resolution by immunofluores-cence and did not intend to exclude the possibility of partialcolocalization of these two proteins. It was concluded thatthe two proteins might have distinct functions in the phrag-moplast because of the difference in localizations.

AtPAKRP2 Is Enriched in the InsolubleCytosolic Fraction

The punctate localization pattern of AtPAKRP2 suggestedthat this motor might be associated with membranous com-partment(s) such as Golgi-derived vesicles in the phrag-

Figure 5. Association of AtPAKRP2 with the Phragmoplast.

Triple localizations of AtPAKRP2 with anti-AtPAKRP2-N� ([A], [E], and [I]), of MTs with anti-�-tubulin ([B], [F], and [J]), and of DNA with DAPI([C], [G], and [K]) are shown in Arabidopsis cells during telophase/cytokinesis. Composite images ([D], [H], and [L]) are pseudocolored withAtPAKRP2 shown in green, �-tubulin shown in red, and DNA shown in blue. Bar in (L) � 5 �m for (A) to (L).


moplast. To determine whether AtPAKRP2 was in factassociated with the insoluble cytosolic fraction, protein ex-tracts of Arabidopsis inflorescence buds were subjected tofractionation (see Methods); the resulting supernatant wasdesignated the soluble cytosolic fraction, and the pellet wasdesignated the fraction of endomembranes and other insol-uble cytosolic compartments. When equal amounts of pro-tein from the two fractions were loaded on an SDS-PAGE gelfollowed by immunoblotting with anti-AtPAKRP2-N

�

, it ap-peared that AtPAKRP2 was enriched greatly in the insolublefraction (Figure 8A). Our results suggest that AtPAKRP2could be a motor protein that associates with vesicles thatare enriched in such preparations. When anti-AtPAKRP1 an-tibodies were applied, AtPAKRP1 protein was found mainlyin the supernatant fraction, suggesting that AtPAKRP1 wasunlikely to be associated with the endomembranes (data notshown).

AtPAKRP2 Localization Is Altered by Brefeldin A

We further tested whether AtPAKRP2 associated with theGolgi-derived vesicles because such vesicles were com-monly found in the phragmoplast. To disrupt the distributionof these vesicles, we used brefeldin A (BFA), which disruptsthe Golgi apparatus and inhibits cell plate formation inBY-2 cells (Yasuhara and Shibaoka, 2000). Although BFAcould prevent MT depolymerization in the phragmoplast, itdid not change the overall organization of phragmoplastMTs (Figure 8C) (Yasuhara and Shibaoka, 2000). BecauseBFA was dissolved in methanol, control cells were treatedwith an amount of methanol equal to that used to dissolveBFA. In the phragmoplasts of control cells, the AtPAKRP2homolog was concentrated near the equatorial plane (Figure8E), which was similar to what was observed in cells nottreated with methanol (Figure 8H). In BFA-treated cells,however, much reduced anti-AtPAKRP2 signal could bedetected in the same area (Figure 8B). Such a result was un-likely to be caused by an overall reduction of fluorescenceintensity because diffuse signal was present in the cyto-plasm (Figure 8B). To quantify the distortion of AtPAKRP2localization caused by BFA treatment, we measured the rel-ative fluorescence intensities in both the BFA-treated cell andcontrol cells (Figures 8D, 8G, and 8J). Although a clear peakof AtPAKRP2 signal could be detected near the equatorialplane in the control cells, such a peak was much less obviousin the same area of the BFA-treated cell (Figures 8D, 8G, and8J). Moreover, the overall level of fluorescence intensity of

Figure 6.

Requirement of MTs for AtPAKRP2 Localization.

(A)

After APM treatment, AtPAKRP2 appeared in a diffuse pattern inthe cytoplasm in an Arabidopsis root tip cell after the completion ofmitosis.

(B)

MTs were depolymerized completely in this cell, as indicated byanti–

�

-tubulin.

(C)

Daughter nuclei stained with DAPI.

(D)

to

(F)

A control root tip cell at a similar stage as in the cell

(A)

to

(C)

, (shown by DAPI staining of the nuclei in

[F]

) was treated with

isopropanol alone. AtPAKRP2 localization was not disturbed

(D)

, norwere phragmoplast MTs

(E)

.Bar in (F) � 5 �m for (A) to (F).

2434 The Plant Cell

AtPAKRP2 across the cell, or the basal level, increased signif-icantly in the BFA-treated cell (Figure 8D). We concluded thatas a result of the reduced amount of these vesicles amongphragmoplast MTs after BFA treatment, AtPAKRP2 was notable to localize among the phragmoplast MTs properly andwas distributed randomly in the cytoplasm.

AtPAKRP2 and KNOLLE Show DistinctLocalization Patterns

The KNOLLE protein is a cytokinesis-specific syntaxin thatlocalizes to the developing cell plate and functions in vesicletargeting during cell plate formation (Lauber et al., 1997).Because no vesicle motor had been found in the phragmo-plast, we tested whether AtPAKRP2 was involved in thetransport of KNOLLE-bearing vesicle(s) to the cell plate.Dual localizations of AtPAKRP2 and KNOLLE were per-formed in B. oleracea because of its large cells comparedwith those of Arabidopsis and because of its evolutionaryrelationship with Arabidopsis. Our antibodies cross-reactedwith the B. oleracea homolog of AtPAKRP2 and gave similarlocalization in B. oleracea cells and Arabidopsis cells (Figure9A). In a cell undergoing cytokinesis, as judged by the ap-pearance of the daughter nuclei (Figure 9C), the B. oleraceaAtPAKRP2 homolog was in a punctate pattern in the phrag-moplast, being more pronounced toward the division site(Figure 9A). In this cell, KNOLLE was in a sharp line clearlymarking the cell plate (Figure 9B). We concluded, therefore,that the KNOLLE protein was not present in the punctatestructure(s) bearing the AtPAKRP2 homolog.

DISCUSSION

We have identified AtPAKRP2 cDNA encoding a novel N-ter-minal motor KRP in Arabidopsis and determined the intra-cellular localization of this AtPAKRP2 protein. AtPAKRP2and its homologs in tobacco and B. oleracea localize specif-ically to the phragmoplast during cell division. In addition toAtPAKRP1 (Lee and Liu, 2000), AtPAKRP2 is the secondplant KRP that associates only with the phragmoplast MTarray. Unlike AtPAKRP1, which is localized exclusively at ornear the plus end of phragmoplast MT bundles, however,AtPAKRP2 appears in a punctate pattern with more pro-nounced distribution toward the division site. On the basisof biochemical and pharmacological data, AtPAKRP2likely associates with possibly Golgi-derived vesicles in thephragmoplast. Thus, this protein is a potential motor for ves-icle transport during cell plate formation.

Although AtPAKRP2 has a tripartite structure with acoiled-coil domain flanked by the motor and tail domains, itdoes not fall into any of the established kinesin/KRP sub-families. Phylogenetic analysis based on the alignment ofthe motor domains of AtPAKRP2 and kinesin/KRPs from dif-

Figure 7. Distinct Localization Patterns of AtPAKRP2 andAtPAKRP1.

Arabidopsis root tip cells at early cytokinesis ([A] to [C]) and late cy-tokinesis ([D] to [F]) were stained with rat anti–AtPAKRP2-T ([A] and[D]) and rabbit anti–AtPAKRP1-C ([B] and [E]) antibodies. Daughternuclei are shown by DAPI staining ([C] and [F]). AtPAKRP1 ap-peared much more discretely at the equatorial plane than didAtPAKRP2. Bar in (F) � 5 �m for (A) to (F).


ferent subfamilies suggested that AtPAKRP2 is very diver-gent from various KRPs among eukaryotes. Therefore, itshould be considered an outgroup KRP. Such a divergence,however, did not affect its ATP-dependent MT binding ac-tivity. There is little doubt that it behaves as a motor proteinin plant cells. AtPAKRP2 does not contain a clear consen-sus sequence in the neck domain that is present amongplus end–directed kinesin/KRP motors (Vale and Fletterick,1997), nor does it have a C-terminal motor KRP-specificneck region. Because the neck sequence of C-terminal mo-tor KRPs has been suggested to direct KRPs to move to-ward the MT minus ends (Endow, 1999), we believe thatAtPAKRP2 cannot be a minus end–directed motor. Becauseof the nature of the intrinsic plus end–directed activity of themotor domain, AtPAKRP2 most likely is a plus end–directedmotor. Unfortunately, under our experimental conditions,GST-AtPAKRP2-M failed to demonstrate motility in vitro.This might be because the GST fusion protein was not ro-bust enough or the conditions we tested were not favorablefor the motility assays.

The previously identified AtPAKRP1 protein also is asso-ciated exclusively with the phragmoplast (Lee and Liu,2000). Although both AtPAKRP1 and AtPAKRP2 are N-ter-minal motor KRPs, they have very different structural fea-tures. AtPAKRP1 has a motor domain that is much less

Figure 8. Association of AtPAKRP2 with the Insoluble CytosolicFraction and Disturbance of AtPAKRP2 Localization by BFA.

(A) Inflorescence buds of Arabidopsis were used for subcellular frac-tionation. Equal amounts of protein from total protein extract (lane 1),the cytosolic fraction (lane 2), and the insoluble cytosolic fraction (lane 3)were subjected to immunoblotting with anti–AtPAKRP2-N� antibodies.The arrow indicates a single band recognized by the anti–AtPAKRP2-N�

antibodies, which is enriched in the soluble cytosolic fraction.(B) to (D) A tobacco BY-2 cell was treated with BFA in methanol.(E) to (G) A tobacco BY-2 cell was treated with methanol alone.(H) to (J) A tobacco BY-2 cell was fixed and stained without anytreatment.The cells were double stained with anti–AtPAKRP2-N� ([B], [E], and[H]) and anti–�-tubulin ([C], [F], and [I]) antibodies. (D), (G), and (J)show relative intensities of immunofluorescence of AtPAKRP2 (thickgray lines) and tubulin (thin black lines). The x axis represents the po-sition of the data point, which was aligned to the micrographs, andthe y axis represents the relative fluorescence intensity with the valuegiven arbitrarily by ImagePro software. Bar in (I) � 10 �m for (B) to (I).

Figure 9. Distinct Localization Patterns of AtPAKRP2 and KNOLLEin the Phragmoplast.

Triple localizations were performed with anti–AtPAKRP2-T (A), anti-KNOLLE (B), and DAPI (C) in B. oleracea cells undergoing cytokine-sis as revealed by the appearance of the daughter nuclei (C). Thecomposite image (D) was pseudocolored, with AtPAKRP2 shown ingreen, KNOLLE shown in red, and DNA shown in blue. Bar in (D) � 5�m for (A) to (D).

2436 The Plant Cell

divergent from other known kinesins/KRPs than is AtPAKRP2.AtPAKRP1 also has an extremely long nonmotor domainand does not have a typical tripartite structure, as doesAtPAKRP2 (Lee and Liu, 2000). AtPAKRP1 has coiled coilsmore scattered in the nonmotor domain. AtPAKRP2, how-ever, has coiled coils immediately after the motor domain.The significance of the nonmotor domains of both motorshas yet to be characterized.

Our results suggest that AtPAKRP1 and AtPAKRP2 donot colocalize in the phragmoplast. Their distinct localiza-tions could be distinguished from the very beginning stagewhen they started to appear in the spindle midzone at lateanaphase. AtPAKRP1 clearly associated with MT bundles,whereas AtPAKRP2 was not restricted to the MT bundles. Inlater stages of phragmoplast development, AtPAKRP1 isconspicuous in the midline of the phragmoplast, coincidingwith the plus ends of MT bundles in a narrow area, as re-vealed by immunofluorescence. As a result of the limitationsof our fixation and staining procedures, it is possible that wemissed some AtPAKRP1 proteins in the phragmoplast thatwere not concentrated along MT bundles. But identical pro-cedures allowed us to clearly detect AtPAKRP2 in thephragmoplast, which was not restricted to MT bundles. It isconceivable that phragmoplast operation requires both mo-tors that are responsible for MT reorganization and motorsthat are responsible for vesicle transport. Obviously, motorsbearing two distinct roles will have different tail domains.Elucidation of the functions of the nonmotor domains ofboth motors will shed light on the mechanisms of plant cy-tokinesis. For example, it would be interesting to determinewhether either tail domain could bind MTs in the absence ofthe motor domain in vivo or in vitro.

Treatment with BFA suggested that AtPAKRP2 associ-ated with Golgi vesicles that were destined to the phragmo-plast, although careful examination with immunogoldlabeling is needed to reveal the AtPAKRP2 localization atthe ultrastructural level. When the formation of these vesi-cles was hampered by BFA as it disrupted the Golgi apparatus,fewer vesicles were expected in the phragmoplast. We sug-gest that binding to the vesicle might be necessary forAtPAKRP2 to associate with the phragmoplast MTs. There-fore, when fewer vesicles were produced, AtPAKRP2 wasmislocalized. Conversely, AtPAKRP2 was required for Golgivesicles to travel along phragmoplast MTs becauseAtPAKRP2 is a MT-based motor. It was not surprising thatAtPAKRP2 localization to the phragmoplast was dependenton the integrity of phragmoplast MTs.

Because KNOLLE is a syntaxin-like protein that plays anessential role in vesicle docking during cytokinesis in plantcells (Lauber et al., 1997), it is reasonable to wonderwhether AtPAKRP2 is the motor protein that transportsKNOLLE-bearing vesicles. However, our results from doublelocalizations of homologs of AtPAKRP2 and KNOLLE in B.oleracea cells suggested that they had different localizationpatterns. Thus, it is possible that other motor protein(s)might transport KNOLLE-bearing vesicles.

It is worth noting that dynamic activities of vesicles are re-quired in animal cytokinesis as well (Straight and Field,2000). The ras-like GTPase Rabs play critical roles in vesicletargeting (Deacon and Gelfand, 2001). In a search for inter-actors with Rab6, the kinesin-like Rab6-KIFL has been iden-tified (Echard et al., 1998). Rab6-KIFL localizes to thespindle midzone and the midbody in mammalian cells (Hill etal., 2000). Introduction of anti–Rab6-KIFL antibodies into di-viding cells blocks cytokinesis (Hill et al., 2000). Rab6-KIFLprobably is involved in transporting Golgi-derived vesiclesduring cytokinesis in animal cells. Such speculation is fos-tered by the evidence that Rab6-KIFL also associates withthe Golgi apparatus (Echard et al., 1998). Thus, cytokinesisin animals requires the active transport of Golgi vesicles byMT-based motor(s), as in plant cells.

A number of KRPs have been implicated in cytokinesis inhigher plants (Liu and Lee, 2001). The most studied ones aremembers of the BIMC subfamily, which form homotetra-mers in their native forms (Kashina et al., 1996; Asada et al.,1997; Barroso et al., 2000). Among them, the tobaccoTKRP125 protein and the carrot DcKRP120-2 protein areactivated to render the sliding activity of antiparallel MTs toestablish the phragmoplast MT array (Asada et al., 1997;Barroso et al., 2000). The activity of BIMC-like KRPs likely isbalanced by C-terminal motor KRPs, most likely KatA/B/C,which also localize to phragmoplast MTs (Liu and Palevitz,1996; Liu et al., 1996; Mitsui et al., 1996). Moreover, down-regulation of KCBP is required for the organization ofphragmoplast MTs (Vos et al., 2000). In addition to thesemotors, AtPAKRP1 is believed to play a role in the mainte-nance of the mirror-shaped phragmoplast MT array afterthe sliding activity is complete (Liu and Lee, 2001). It isconceivable that stabilization of the phragmoplast MT ar-ray allows vesicles to be transported to the right place atthe division site, possibly by AtPAKRP2. In a simplifiedview of cytokinesis, we believe there are collaborative effortsof different motors that result in the smooth progression ofcell plate formation.

Recently, the tobacco mitogen-activated protein kinasekinase kinase NPK1 has been implicated in a signalingpathway that regulates cytokinesis in tobacco cells(Machida et al., 1998). Upon overexpression of a dominantnegative construct or a kinase-negative mutant, cytokine-sis can be blocked and multinucleate cells can be formed(Nishihama et al., 2001). It appears that this kinase is re-quired for the expansion of the phragmoplast MT array to-ward the cell periphery during cell plate formation(Nishihama et al., 2001). NPK1 localizes with spindle MTsat metaphase, at the equatorial plane during late ana-phase, and in the midzone of the phragmoplast (Nishihamaet al., 2001). More interestingly, it has been mentioned thata novel tobacco KRP, NACK1, acts as an activator ofNPK1 and localizes to the division site as well (Machida etal., 1998; Nishihama et al., 2001). It will be interesting tocompare NACK1 and AtPAKRP1/2 once the Nack1 se-quence is revealed.


METHODS

Isolation and Characterization of AtPAKRP2 cDNAs

The Arabidopsis thaliana expressed sequence tag clone OAO358was provided by Dr. Alain Lecharny (Laboratoire de Biologie du De-veloppement des Plantes, Universite de Paris Sud, France). 5� RACE(rapid amplification of cDNA ends; Clontech, Palo Alto, CA) was ap-plied to obtain the 5� sequence of AtPAKRP2 cDNA. Trizol (LifeTechnologies, Rockville, MD) was used to isolate total RNA from in-florescence buds of Arabidopsis Columbia according to the manu-facturer’s protocol. Two AtPAKRP2-specific primers were designedfor 5� RACE: one primer (5�-GTGTTAAAAGAGATCTTAAAGCAGC-3�)was used for reverse transcription, and the other nested primer (5�-ATCTTCCCCCAACAGTTGGCACATC-3�) was used for subsequentpolymerase chain reaction (PCR). DNA samples were sequenced in acommercial laboratory (Davis Sequencing, Davis, CA). The GCG pro-gram (Genetics Computer Group, Madison, WI) was used to analyzeDNA sequences. Phylogenetic analysis of kinesin/KRP motor do-mains was performed with the heuristic search method of PAUP in-cluded in GCG. A maximum parsimony with random stepwiseaddition was used. In addition to the AtPAKRP2 reported here, thephylogenetic analysis included Arabidopsis AtKatA (GenBank ac-cession number D11371), AtKatB (D21137), AtKatC (D21138), At-KatD (AF080249), AtKCBP (L40358), and AtPAKRP1 (AF193767);Aspergillus nidulans AnBIMC (M32075); Cricetulus griseus CgMCAK(U11790); Daucus carota DcKRP120-2 (AF283505); Drosophila mela-nogaster DmNcd (X52814) and DmKHC (M24441); Gallus gallusGgChrkin (U18309); Homo sapiens HsMKLP1 (X67155); Mus muscu-lus MmKIF1A (D29951); Nicotiana tabacum TKRP125 (D83711); Sac-charomyces cerevisiae ScSmy1p (M69021); and Strongylocentrotuspurpuratus SpKRP95 (U00996). The phylogenetic tree shown in Fig-ure 1B was one of two optimal trees and was rooted arbitrarily usingScSmy1p as an outgroup KRP. The CoilScan program of GCG wasused to predict the coiled-coil domains of a protein.

Production of Recombinant Fusion Proteins

To construct plasmids encoding fusion proteins of glutathioneS-transferase (GST) and various regions of AtPAKRP2, new restric-tion sites were introduced into AtPAKRP2 cDNA by PCR. Four plas-mids were made for this study: (1) pGEX-AtPAKRP2-M contained theDNA fragment encoding amino acid residues 1 to 435, which wasamplified by PCR with the primers 5�-TCTTGAATTCTGGCACCGACA-CCATC-3� and 5�-GTGTCTCGAGAGATCTTAAAGCAGC-3�; (2) pGEX-AtPAKRP2-N contained the DNA fragment encoding amino acid res-idues 1 to 202, which was amplified with the primers 5�-GCACGAATT-CCGACACCATCTTCTTC-3� and 5�-AGTACTCGAGTGCTTGCT-CCTTTAG-3�; (3) pGEX-AtPAKRP2-N� contained the DNA fragmentencoding amino acid residues 1 to 127, which was amplified with theprimers 5�-GCACGAATTCCGACACCATCTTCTTC-3� and 5�-TAG-GCTCGAGCATCATAATCGTGCA-3�; and (4) pGEX-AtPAKRP2-Tcontained the DNA fragment encoding amino acid residues 516 to 869,which was amplified with the primers 5�-CGCTCCATGGGAGGAAGT-GTTGATG-3� and 5�-AATACTCGAGACAAGACATAGCGAC-3�. ThePCR fragments for the first three plasmids were digested with EcoRI andXhoI (sites underlined in the primers) and ligated to pGEX-KG EcoRI-XhoI (Guan and Dixon, 1991). The PCR fragment for the last plasmid wasdigested with NcoI and XhoI and ligated to pGEX-KG NcoI-XhoI.

All pGEX constructs were expressed in the E. coli strainBL21(DE3)pLysS (Novagen, Madison, WI). GST fusion proteins wereaffinity purified with glutathione–Sepharose as described by themanufacturer (Pierce Chemical Co., Rockford, IL), except for the pro-tein expressed by pGEX-AtPAKRP2-N�. This protein was not soluble,so it was purified from inclusion bodies with B-PER bacterial proteinextraction reagent as described by the manufacturer (Pierce).

Microtubule Sedimentation Assay

This assay was performed essentially as described previously (Leeand Liu, 2000). In brief, purified GST-AtPAKRP2-M fusion protein (5�g) was incubated with 25 �g of microtubules (MTs), 10 �M taxol,and 2 �M ATP (or adenylylimidodiphosphate, AMPPNP) in BRB80buffer (80 mM Pipes, 1 mM MgCl2, and 1 mM EGTA, pH 6.8). After in-cubation at room temperature, the reaction was loaded on top of asucrose cushion (15% sucrose and 10 �M taxol in BRB80) and thensubjected to centrifugation at 25,000g. The supernatant above thesucrose cushion and the MT pellet were analyzed by SDS-PAGE.Proteins of the supernatant and pellet were visualized by staining thegels with Coomassie Brilliant Blue R250 (Sigma, St. Louis, MO).

Antibody Preparation and Immunoblotting

The GST fusion proteins GST-AtPAKRP2-N and GST-AtPAKRP2-Twere used as antigens to raise polyclonal antibodies in rabbits andrats, respectively, at commercial facilities (Strategic BioSolutions,Romona, CA; and Antibodies, Inc., Davis, CA). Sera were passedthrough a column that had the GST protein covalently linked to thematrix (GST Orientation Kit; Pierce) to deplete antibodies againstGST. Depletion of anti-GST antibodies was confirmed by immuno-blotting against the GST protein. Because the PAKRP2-N polypep-tide covered approximately half of the motor domain, the anti–PAKRP2-N sera might contain antibodies against conserved kinesin/KRP epitopes. To eliminate such antibodies, PAKRP2-N� polypep-tide was used for the purpose of antibody purification because it wasupstream of the ATP binding site and few conserved residues werefound in this region. Antibodies specific to AtPAKRP2 were affinitypurified with GST-AtPAKRP2-N� or GST-AtPAKRP2-T by a blot puri-fication method (Olmsted, 1981). To test the specificity of antibodies,control experiments were performed by incubating antibodies withblots containing their corresponding antigens before immunostaining.

Total protein extraction from plant tissues and immunoblottingwas performed as described previously (Liu et al., 1996). To preparesubcellular fractions, Arabidopsis inflorescence buds were ground inthe extraction buffer containing 30 mM Pipes, 10 mM EGTA, 6 mMMgCl2, 1 mM phenylmethylsulfonyl fluoride, 2 mM DTT, and 250 mMsucrose, pH 7.0, with a mortar and pestle. The homogenate was cen-trifuged at 10,000g for 10 min at 4C to remove large cellular debris.This low-speed supernatant was centrifuged at 100,000g for 30 minat 4C. The resulting supernatant (the cytosolic fraction) and the pel-let (the insoluble cytosolic fraction) then were used for immunoblot-ting analysis.

Fluorescence Microscopy and Image Processing

Slides containing Arabidopsis root tip cells of 2-day-old etiolatedseedlings or tobacco BY-2 suspension cells were prepared and pro-cessed for immunofluorescence as described previously (Lee and

2438 The Plant Cell

Liu, 2000). In addition to the rabbit anti-AtPAKRP2-N� and rat anti-AtPAKRP2-T antibodies reported here, antibodies used in this studyincluded rabbit anti-AtPAKRP1-C (Lee and Liu, 2000), mouse anti-�-tubulin (DM1A; Sigma), fluorescein isothiocyanate–conjugated goatanti-rabbit IgG (Sigma), Texas Red X–conjugated goat anti-rabbit IgG(Molecular Probes, Eugene, OR), Texas Red X–conjugated goat anti-mouse IgG (Molecular Probes), fluorescein isothiocyanate–conjugatedgoat anti-rat IgG (Sigma), and rabbit anti-KNOLLE (Rose Biotech,Winchendon, MA). Images were acquired by conventional epifluores-cence microscopy as described previously (Lee and Liu, 2000).

Pharmacological Studies

To depolymerize MTs, Arabidopsis 2-day-old etiolated seedlings wereincubated in 5 mL of germination medium (half-strength Murashigeand Skoog [1962] salts [Life Technologies] and 5 mM Mes, pH 5.7)containing 100 �M APM (a gift from Dr. C. Gregg of Bayer Corp, Kan-sas CIty, MO) for 2 hr. The same amount of isopropanol, used to dis-solve APM, was added in the control seedlings. BFA treatment wasperformed in tobacco BY-2 suspension cells 4 days after subculture.Ten-milliliter cells were treated with 20 �M BFA for 2 hr. The solventmethanol was added in the control sample. The measurement offluorescence intensity was performed using the Line Profile mea-surement function in the ImagePro software package (Media Cyber-netics, Silver Spring, MD). Data then were exported to MicrosoftExcel software to generate Figures 8D, 8G, and 8J.

Accession Number

The GenBank accession number for AtPAKRP2 is AF320248.

ACKNOWLEDGMENTS

We thank Dr. Alain Lecharny (Universite de Paris Sud) for theOAO358 clone, Dr. Carl Gregg (Agricultural Division of Bayer Corp.)for APM, and Mary Lai for critical comments on the manuscript. Thiswork was supported by a grant from the U.S. Department of Agricul-ture (NRI/CSREES 99-35304-8142).

Received June 4, 2001; accepted August 29, 2001.

REFERENCES

Asada, T., Sonobe, S., and Shibaoka, H. (1991). Microtubule trans-location in the cytokinetic apparatus of cultured tobacco cells.Nature 350, 238–241.

Asada, T., Kuriyama, R., and Shibaoka, H. (1997). TKRP125, akinesin-related protein involved in the centrosome-independentorganization of the cytokinetic apparatus in tobacco BY-2 cells. J.Cell Sci. 110, 179–189.

Barroso, C., Chan, J., Allan, V., Doonan, J., Hussey, P., andLloyd, C. (2000). Two kinesin-related proteins associated with thecold-stable cytoskeleton of carrot cells: Characterization of anovel kinesin, DcKRP120-2. Plant J. 24, 859–868.

Bowser, J., and Reddy, A.S.N. (1997). Localization of a kinesin-likecalmodulin-binding protein in dividing cells of Arabidopsis andtobacco. Plant J. 12, 1429–1437.

Deacon, S.W., and Gelfand, V.I. (2001). Of yeast, mice, and men:Rab proteins and organelle transport. J. Cell Biol. 152, F21–F24.

Echard, A., Jollivet, F., Martinez, O., Lacapere, J.J., Rousselet,A., Janoueix-Lerosey, I., and Goud, B. (1998). Interaction of aGolgi-associated kinesin-like protein with Rab6. Science 279,580–585.

Endow, S.A. (1999). Determinants of molecular motor directionality.Nat. Cell Biol. 1, E163–E167.

Goldstein, L.S.B., and Philp, A.V. (1999). The road less traveled:Emerging principles of kinesin motor utilization. Annu. Rev. CellDev. Biol. 15, 141–183.

Granger, C.L., and Cyr, R.J. (2000). Microtubule reorganization intobacco BY-2 cells stably expressing GFP-MBD. Planta 210,502–509.

Guan, K.L., and Dixon, J.E. (1991). Eukaryotic proteins expressedin Escherichia coli: An improved thrombin cleavage and purifica-tion procedure of fusion proteins with glutathione S-transferase.Anal. Biochem. 192, 262–267.

Hasezawa, S., Ueda, K., and Kumagai, F. (2000). Time-sequenceobservations of microtubule dynamics throughout mitosis in livingcell suspensions of stable transgenic Arabidopsis: Direct evi-dence for the origin of cortical microtubules at M/G(1) interface.Plant Cell Physiol. 41, 244–250.

Hill, E., Clarke, N., and Barr, F.A. (2000). The Rab6-binding kinesin,Rab6-KIFL, is required for cytokinesis. EMBO J. 19, 5711–5719.

Kashina, A.S., Baskin, R.J., Cole, D.G., Wedaman, K.P., Saxton,W.M., and Scholey, J.M. (1996). A bipolar kinesin. Nature 379,270–272.

Kim, A.J., and Endow, S.A. (2000). A kinesin family tree. J. Cell Sci.113, 3681–3682.

Lauber, M.H., Waizenegger, I., Steinmann, T., Schwarz, H.,Mayer, U., Hwang, I., Lukowitz, W., and Jurgens, G. (1997). TheArabidopsis KNOLLE protein is a cytokinesis-specific syntaxin. J.Cell Biol. 139, 1485–1493.

Lee, Y.R.J., and Liu, B. (2000). Identification of a phragmoplast-associated kinesin-related protein in higher plants. Curr. Biol. 10,797–800.

Liu, B., and Lee, Y.R.J. (2001). Kinesin-related proteins in plantcytokinesis. J. Plant Growth Regul. 20, 141–150.

Liu, B., and Palevitz, B.A. (1996). Localization of a kinesin-like pro-tein in generative cells of tobacco. Protoplasma 195, 78–89.

Liu, B., Cyr, R.J., and Palevitz, B.A. (1996). A kinesin-like protein,KatAp, in the cells of Arabidopsis and other plants. Plant Cell 8,119–132.

Machida, Y., Nakashima, M., Morikiyo, K., Banno, H., Ishikawa,M., Soyano, T., and Nishihama, R. (1998). MAPKKK-related pro-tein kinase NPK1: Regulation of the M phase of plant cell cycle. J.Plant Res. 111, 243–246.

Mitsui, H., Hasezawa, S., Nagata, T., and Takahashi, H. (1996).Cell cycle–dependent accumulation of a kinesin-like protein,KatB/C, in synchronized tobacco BY-2 cells. Plant Mol. Biol. 30,177–181.


Murashige, T., and Skoog, F. (1962). A revised medium for rapidgrowth and bioassays with tobacco tissue culture. Physiol. Plant.15, 473–497.

Nishihama, R., Ishikawa, M., Araki, S., Soyano, T., Asada, T., andMachida, Y. (2001). The NPK1 mitogen-activated protein kinasekinase kinase is a regulator of cell-plate formation in plant cytoki-nesis. Genes Dev. 15, 352–363.

Olmsted, J.B. (1981). Affinity purification of antibodies from diazo-tized paper blots of heterogeneous protein samples. J. Biol.Chem. 256, 11955–11957.

Otegui, M., and Staehelin, L.A. (2000). Syncytial-type cell plates: Anovel kind of cell plate involved in endosperm cellularization ofArabidopsis. Plant Cell 12, 933–947.

Reddy, A.S.N. (2001). Molecular motors and their functions inplants. Int. Rev. Cytol. 204, 97–178.

Samuels, A.L., Giddings, T.H., and Staehelin, L.A. (1995). Cytoki-nesis in tobacco BY-2 and root tip cells: A new model of cell plateformation in higher plants. J. Cell Biol. 130, 1345–1357.

Sharp, D.J., McDonald, K.L., Brown, H.M., Matthies, H.J., Walczak,C., Vale, R.D., Mitchison, T.J., and Scholey, J.M. (1999). Thebipolar kinesin, KLP61F, cross-links microtubules within interpolarmicrotubule bundles of Drosophila embryonic mitotic spindles. J.Cell Biol. 144, 125–138.

Sharp, D.J., Rogers, G.C., and Scholey, J.M. (2000). Microtubulemotors in mitosis. Nature 407, 41–47.

Smirnova, E.A., Reddy, A.S.N., Bowser, J., and Bajer, A.S. (1998).Minus end-directed kinesin-like motor protein, KCBP, localizes toanaphase spindle poles in Haemanthus endosperm. Cell Motil.Cytoskeleton 41, 271–280.

Sonobe, S., Nakayama, N., Shimmen, T., and Sone, Y. (2000).

Intracellular distribution of subcellular organelles revealed by anti-body against xyloglucan during cell cycle in tobacco BY-2 cells.Protoplasma 213, 218–227.

Straight, A.F., and Field, C.M. (2000). Microtubules, membranesand cytokinesis. Curr. Biol. 10, R760–R770.

Vale, R.D., and Fletterick, R.J. (1997). The design plan of kinesinmotors. Annu. Rev. Cell Dev. Biol. 13, 745–777.

Vantard, M., Levilliers, N., Hill, A.M., Adoutte, A., and Lambert,A.M. (1990). Incorporation of paramecium axonemal tubulin intohigher plant cells reveals functional sites of microtubule assembly.Proc. Natl. Acad. Sci. USA 87, 8825–8829.

Verma, D.P.S. (2001). Cytokinesis and building of the cell plate inplants. Annu. Rev. Plant Physiol. 52, 751–784.

Vos, J.W., Safadi, F., Reddy, A.S.N., and Hepler, P.K. (2000). Thekinesin-like calmodulin binding protein is differentially involved incell division. Plant Cell 12, 979–990.

Wittmann, T., Hyman, A., and Desai, A. (2001). The spindle: Adynamic assembly of microtubules and motors. Nat. Cell Biol. 3,E28–E34.

Yasuhara, H., and Shibaoka, H. (2000). Inhibition of cell-plate for-mation by brefeldin A inhibited the depolymerization of microtu-bules in the central region of the phragmoplast. Plant Cell Physiol.41, 300–310.

Yasuhara, H., Sonobe, S., and Shibaoka, H. (1993). Effects of taxolon the development of the cell plate and of the phragmoplast intobacco BY-2 cells. Plant Cell Physiol. 34, 21–29.

Zhang, D., Wadsworth, P., and Hepler, P.K. (1993). Dynamics ofmicrofilaments are similar, but distinct from, microtubules duringcytokinesis in living, dividing plant cells. Cell Motil. Cytoskeleton24, 151–155.

DOI 10.1105/tpc.010225 2001;13;2427-2439Plant Cell

Y.-R. Julie Lee, Hoa M. Giang and Bo LiuA Novel Plant Kinesin-Related Protein Specifically Associates with the Phragmoplast Organelles

This information is current as of April 10, 2019

References /content/13/11/2427.full.html#ref-list-1

This article cites 36 articles, 13 of which can be accessed free at:

Permissions https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists

https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

http://www.plantcell.org/cgi/alerts/ctmain

http://www.plantcell.org/cgi/alerts/ctmain

http://www.aspb.org/publications/subscriptions.cfm

Documents

A Novel Plant Kinesin-Related Protein Specifically Associates with the Phragmoplast Organelles