Interaction of Antiparallel Microtubules in the Phragmoplast IsMediated by the Microtubule-Associated Protein MAP65-3in Arabidopsis W

Chin-Min Kimmy Ho,a Takashi Hotta,a Fengli Guo,a,1 Robert W. Roberson,b Yuh-Ru Julie Lee,a and Bo Liua,2

a Department of Plant Biology, University of California, Davis, CA 95616b School of Life Sciences, Arizona State University, Tempe, AZ 85287

In plant cells, microtubules (MTs) in the cytokinetic apparatus phragmoplast exhibit an antiparallel array and transport

Golgi-derived vesicles toward MT plus ends located at or near the division site. By transmission electron microscopy, we

observed that certain antiparallel phragmoplast MTs overlapped and were bridged by electron-dense materials in

Arabidopsis thaliana. Robust MT polymerization, reported by fluorescently tagged End Binding1c (EB1c), took place in

the phragmoplast midline. The engagement of antiparallel MTs in the central spindle and phragmoplast was largely

abolished in mutant cells lacking the MT-associated protein, MAP65-3. We found that endogenous MAP65-3 was selectively

detected on the middle segments of the central spindle MTs at late anaphase. When MTs exhibited a bipolar appearance

with their plus ends placed in the middle, MAP65-3 exclusively decorated the phragmoplast midline. A bacterially expressed

MAP65-3 protein was able to establish the interdigitation of MTs in vitro. MAP65-3 interacted with antiparallel microtubules

before motor Kinesin-12 did during the establishment of the phragmoplast MT array. Thus, MAP65-3 selectively cross-

linked interdigitating MTs (IMTs) to allow antiparallel MTs to be closely engaged in the phragmoplast. Although the presence

of IMTs was not essential for vesicle trafficking, they were required for the phragmoplast-specific motors Kinesin-12 and

Phragmoplast-Associated Kinesin-Related Protein2 to interact with MT plus ends. In conclusion, we suggest that the

phragmoplast contains IMTs and highly dynamic noninterdigitating MTs, which work in concert to bring about cytokinesis in

plant cells.

Introduction

In plant cells, cytokinesis is brought about by the phragmoplast,

which assembles the cell plate upon the completion of nuclear

division. In the phragmoplast, microtubules (MTs) are organized

in an antiparallel array and are aligned perpendicularly to the

division plane. The antiparallel MTs have their plus ends located

at or near the cell division site and their minus ends facing

two reforming daughter nuclei. The organization of such a bipolar

MT array allows directional transport of Golgi-derived vesicles

toward the division site (Staehelin and Hepler, 1996; Jurgens,

2005). The array is a result of polymerization/depolymerization

and reorganization of MTs with mixed polarities in the central

spindle (Zhang et al., 1990). Whereas the overall organization of

phragmoplastMTs iswell appreciated, themethodbywhich they

are organized into such an elegant array while remaining highly

dynamic remains unknown.

In Haemanthus endosperm cells, short cross-bridges were

detected between antiparallel MTs from opposite sides of

the phragmoplast (Hepler and Jackson, 1968; Euteneuer and

McIntosh, 1980). In themature phragmoplast, an electron opaque

zone of;0.3 mmwas observed by transmission electron micros-

copy (TEM), where an overlap of MTs from both sides of the

developing cell plate can be seen (Hepler and Jackson, 1968).

Such aMT interdigitation phenomenon has also been observed in

the phragmoplast in the moss Physcomitrella patens (Hiwatashi

et al., 2008). In dividing somatic cells of Arabidopsis thaliana

prepared by rapid freezing/freeze substitution, however, the ma-

jority of phragmoplast MTs do not seem to overlap when

observed by electron tomography (Austin et al., 2005). It was

thus suggested that a cell plate assembly matrix (CPAM) was

responsible for engaging the antiparallel MTs in the phragmo-

plast. This observation contradicts the conventional view of

how antiparallel MTs are held together in the phragmoplast.

However, it is in agreement with immunofluorescence images

of antitubulin and images of green fluorescent protein (GFP)–

marked MTs in somatic cells, because the corresponding im-

ages show a dark line in the middle of the phragmoplast (Liu

et al., 1995; Yoneda et al., 2005).

Phragmoplast MTs undergo rapid turnover (Hush et al., 1994).

Using antibody immunofluorescence, it has been shown that

the MT-nucleating factor g-tubulin preferentially decorates MTs

toward the distal ends, facing the daughter nuclei (Liu et al.,

1994; Liu et al., 1995). This suggests thatMTs are nucleated from

the distal ends and polymerized toward the division site. How-

ever, in permeabilized tobacco (Nicotiana tabacum) cells and

1Current address: Stowers Institute for Medical Research, Kansas City,MO 64110.2 Address correspondence to bliu@ucdavis.edu.The author responsible for distribution of materials integral to thefindings presented in this article in accordance with the policy describedin the Instructions for Authors (www.plantcell.org) is: Bo Liu (bliu@ucdavis.edu).WOnline version contains Web-only data.www.plantcell.org/cgi/doi/10.1105/tpc.110.078204

The Plant Cell, Vol. 23: 2909–2923, August 2011, www.plantcell.org ã 2011 American Society of Plant Biologists. All rights reserved.

Haemanthus endosperm cells, exogenous tubulins were incor-

porated into the phragmoplast overwhelmingly in its midline

(Vantard et al., 1990; Asada et al., 1991). These studies also

showed that newly polymerized MTs overlapped and then slid

outward. The presence of the MT plus end-tracking protein End

Binding1 (EB1) in the phragmoplastmidline also favors the notion

that MT nucleation takes place in the phragmoplast midline

(Chan et al., 2005; Bisgrove et al., 2008; Komaki et al., 2010).

Thus, further examination with high spatial and temporal reso-

lution is necessary to characterize MT nucleation events in live

phragmoplasts.

In the past few years, the list of proteins known to decorate the

phragmoplast midline has grown (Otegui et al., 2005). Among

them, the most likely MT cross-linking factors are proteins in the

MT-associated protein65 (MAP65)/Anaphase Spindle Elonga-

tion1/Protein Regulator of Cytokinesis1 family and members of

the Kinesin-12 subfamily (Jiang and Sonobe, 1993; Lee and Liu,

2000; Smertenko et al., 2000; Pan et al., 2004; Smertenko et al.,

2008). MAP65 proteins can form a homodimer to promote

antiparallel MT bundling in the phragmoplast and in other MT

arrays during cell division (Smertenko et al., 2004; Gaillard et al.,

2008). Thus, they can bridge overlapping MTs in the phragmo-

plast. Similar bridging effects have also been proposed and

observed in fungal and animal cells (Zhu et al., 2006; Kapitein

et al., 2008). Among the nine MAP65 proteins in A. thaliana,

MAP65-3 plays a primary role in cytokinesis, as its loss leads to a

wider dark central gap in the phragmoplast by tubulin immuno-

fluorescence and frequent failures in cytokinesis (Muller et al.,

2004; Caillaud et al., 2008). However, MAP65-3 is not required

for the bipolar appearance of phragmoplast MTs. Nevertheless,

this result suggests that MT segments near their plus ends are

likely bundled by MAP65-3 in the phragmoplast, and such a

bundling phenomenon is critical for cell plate formation.

In contrast to MAP65, Kinesin-12 motors seem to play a role in

keeping MT plus ends anchored at the division site, because, in

their absence, MTs cross over the phragmoplast midline and

become continuous bundles (Lee et al., 2007). Consequently,

proteins like the syntaxin KNOLLE, normally localizing to the

phragmoplast midline, become mis-localized in the Kinesin-12

mutant because of the disorganization of MT plus ends.

We sought to establish how the plus ends of antiparallel MTs

are engaged with each other in the phragmoplast and how the

activities of proteins involved in cytokinesis are integrated at

the site of cell division. The results reported here show that a

subpopulation of spindle midzone and phragmoplast MTs are

bundled byMAP65-3 in an antiparallel manner. Meanwhile, other

MTs in the region do not overlap. We conclude that the integra-

tion of MT plus ends by MAP65-3 is required for kinesin motors

to interact with the plus ends, but is not essential for vesicle

trafficking during cell plate assembly.

Results

MT Interdigitation in the Phragmoplast Detected by TEM

To investigate MT organization in the phragmoplast midzone,

root meristematic cells of A. thaliana were prepared by high-

pressure rapid freezing and freeze substitution for TEM obser-

vation. At an early stage of cytokinesis, when Golgi-derived

vesicles began to accumulate at the division site, long MTs were

frequently detected in the phragmoplast formed between two

reforming daughter nuclei (Figure 1A). Although many MTs were

terminated in regions where vesicles accumulated, others

crossed the midline and overlapped (Figures 1A and 1B). Such

a MT-overlapping phenomenon was also found at the later

stages of cytokinesis, when a complex tubular–vesicular network

was already established in the middle of the phragmoplast

(Figure 1C). MTs were found to pass over the phragmoplast

midline (Figure 1C). In cells exhibiting a fenestrated early cell

plate, MTs were seen to cross the open region left by membrane

sacs (Figures 1D and 1E). Again, MTs from one side were found

to overlap with those coming from the opposite side of the cell

(Figures 1D and 1E). Electron-dense materials were detected

between the overlapping MTs as if they were bridges (arrow, Fig-

ure 1E). We found that a subpopulation of antiparallel MTs pen-

etrated the vesicle-dense region and sometimes overlapped with

each other in the phragmoplast at different stages of cytokinesis,

whereas many MTs ended before reaching the cell plate.

Analysis of MT Plus Ends in the Phragmoplast

Published data using fixed samples suggest that antiparallel MTs

in the phragmoplast are separated by the CPAM (Austin et al.,

2005). Previously, the EB1c protein was detected in the phrag-

moplast by immunofluorescence, and its signal was pronounced

near themidline (Bisgrove et al., 2008). To examine the activity of

MT plus ends in the phragmoplast of live cells, we observed cells

expressing fluorescently labeled EB1c using a transgenic line in

which an EB1c-33FLAG-GFP fusion protein was expressed

under the control of the native EB1c promoter. The transgene

completely suppressed the oryzalin hypersensitive phenotype of

the eb1c mutant (see Supplemental Figure 1 online). In cells

undergoing cytokinesis, EB1c-33FLAG-GFP signals were de-

tected in the phragmoplast (Figure 2). The signal was widely

present in the phragmoplast at the early stages of cytokinesis,

and numerous EB1c-33FLAG-GFP comets were found to move

toward the division site (see Supplemental Movie 1 online). At the

later stages of cytokinesis, EB1c-33FLAG-GFP became grad-

ually restricted to the midline of the phragmoplast, where the cell

plate was going to form (Figure 2). If antiparallel MTs were

spaced by the CPAM, we would expect to see a persistent dark

line in the middle of the developing phragmoplast. In all cytoki-

netic cells examined, no dark line was detected at any stage of

the phragmoplast. This finding suggests that it is unlikely that the

developing cell plate prevents MT polymerization from taking

place at the division site, and that the plus ends of polymerizing,

stabilized, or paused MTs likely overlap in the middle region of

the phragmoplast.

MAP65-3 Is Required for the Engagement of Antiparallel

MTs in the Phragmoplast

An earlier study showed that the ple mutation in the gene

encoding MAP65-3 causes the widening of the dark gap in the

middle of the phragmoplast when MTs were revealed by anti-

tubulin immunofluorescence (Muller et al., 2004). A plausible

2910 The Plant Cell

explanation could be that the antiparallel MTsmight interact with

each other at their plus ends in the absence of MAP65-3. To

investigate how MT plus ends behaved during the development

of the phragmoplast, the interface of antiparallel MTs was

examined in the dyc283 mutant, which contains an inactivating

insertion allelic to ple (Caillaud et al., 2008). After sister chromo-

somes were completely segregated in the wild-type cells, MTs in

the central spindle became prominent and aligned in parallel

without an obvious break in themiddle (Figure 3A; 100%, n = 37).

This phenomenon suggested that the MTs initiated from the

opposite regions near segregated chromosomes were bundled

together and tightly engaged. In a dyc283mutant cell at a similar

stage, however, MTs became splayed away from the segregated

chromosomes (Figure 3B; 91%, n = 66). The two populations of

MTs did not show any sign of interaction in thismutant cell. Later,

parallel bundles of phragmoplast MTs were aligned perpendic-

ularly to the division plane, and left a sharp dark line in themiddle

of the wild-type phragmoplast, as shown by antitubulin immu-

nofluorescence (Figure 3C; 100%, n = 155). In mutant cells at

similar stages, bundles of phragmoplastMTs tilted away from the

right angle relative to the division plane (Figure 3D; 100%, n =

280). Such a MT alignment pattern in dyc283 cells indicated that

the antiparallel MTs in the phragmoplast became disengaged.

Endogenous MAP65-3 Highlights Overlapping MTs during

Late Anaphase and Cytokinesis

Because of the altered pattern of antiparallel MTs in the dyc283

mutant cells, we wanted to dissect how endogenous MAP65-3

interacted with MTs during the anaphase–telophase transition.

Antibodies prepared against a truncated fusion protein ex-

pressed in bacteria were shown to recognize a single band

Figure 1. Interaction between Antiparallel MTs in the Phragmoplast

Observed by TEM.

(A) Phragmoplast at an early stage of cytokinesis. Most MTs do not pass

through the division site enriched with vesicles (arrow). MTs crossing the

phragmoplast midline are indicated by arrowheads. Two reforming

daughter nuclei are marked by asterisks.

(B) A region of the early phragmoplast is enlarged to show MTs crossing

the midline (arrowheads).

(C) A mature phragmoplast. The tubular–vesicular network is greatly

elaborated at the division site (arrow). Examples of MTs crossing the

midline are indicated by arrowheads.

(D) Clustered MTs between the membranous sacks (asterisk). MTs from

two sides of the forming cell plate are closely engaged.

(E) Interdigitation of MT clusters in regions flanked by membranous sack

(asterisks, *). Electron-dense materials (arrow) are visible in the MT-

overlapping region.

Bar in (A) = 0.5 mm; bar in (B) to (E) = 0.2 mm.

Figure 2. EB1c Is Enriched in the Phragmoplast Midline.

Time-lapsed images show the changes in fluorescence of the EB1c-

33FLAG-GFP fusion in two cytokinetic cells. At the 0-s time point, the

cell on the right has EB1c signals across the entire region of the

phragmoplast (asterisk), whereas other more discrete comets can be

found between the reforming daughter nuclei and the phragmoplast

(arrows). The EB1c signal becomes more concentrated at the phragmo-

plast midline at later stages of division (arrowheads). This site coincides

with that of the forming cell plate (arrowheads) as seen in the brightfield

DIC image on the bottom right. Bar = 5 mm.

Phragmoplast Microtubule Organization 2911

at ;85 kD by immunoblotting against the wild-type extract

(see Supplemental Figure 2 online). The dyc283 mutation was-

caused by an insertion in the fifth exon of the gene encoding

MAP65-3, and thus likely was null (Caillaud et al., 2008). Similar

amounts of proteins extracted from the dyc283 seedlings, as

indicated by the reference antiactin bands, were probed with the

anti-MAP65-3 antibodies. No signal was detected at the position

corresponding to the At MAP65-3 band, confirming that the

dyc283 plant did not produce the protein (see Supplemental

Figure 2 online).

When themonospecific anti-MAP65-3 antibodies were used in

immunofluorescent localization, mitotic cells after completing

anaphase exhibited pronounced signals in the middle region of

the central spindle and the phragmoplast midline (Figure 4).

When the antibodies were applied to dyc283 cells, no noticeable

signal was detected in the phragmoplast (see Supplemental

Figure 3 online), indicating that they did not cross-react with

other MAP65 isoforms. When cells progressed to the late stages

of anaphase, interzonal MTs started to coalesce between seg-

regating sister chromosomes (Figures 4A to 4C). Punctate anti–

MAP65-3 signals were detected near the midzone MTs in the

central spindle (Figure 4A). No noticeable anti–MAP65-3 signal

was detected with kinetochore fibers containing parallel MTs.

When kinetochore fibers were completely depolymerized,

central spindle MTs became more prominently bundled (Figure

4E). MAP65-3 was associated with segments of interzonal MTs

across the middle region of the central spindle (Figure 4D). It

was also detected toward the distal ends of MTs emanating

from spindle poles (Figure 4F), albeit with much lower signal

intensity compared with the signal along bundled MTs in the

central spindle. Upon further coalescence of interzonal MTs,

the central spindle had two clear sets of MTs, and the dark

Figure 3. Failed Engagement of Antiparallel MTs in dyc283 Cells.

MTs are revealed by antitubulin immunofluorescence at late anaphase/

telophase ([A] and [B]) and cytokinesis ([C] and [D]) in the wild-type ([A]

and [C]) and mutant ([B] and [D]) cells.

(A) In a wild-type cell at late anaphase/telophase, MTs in the central

spindle are highly bundled and aligned in parallel (arrows).

(B) In a dyc283 cell, MTs nucleated from the regions of sister chromatin

masses are splayed in fan-like orientations (arrowheads), and the anti-

parallel MTs are not engaged.

(C) In a wild-type cytokinetic cell, phragmoplast MTs are arranged in

parallel bundles. The mirror-imaged antiparallel bundles leave a very

narrow dark line in the middle (arrows).

(D) In a dyc283 phragmoplast, antiparallel MT bundles exhibit a fan-like

pattern, and are not engaged with each other (arrowheads).

Bar = 5 mm.

Figure 4. Localization of Endogenous MAP65-3 from Late Anaphase to

Cytokinesis by Immunofluorescence.

In merged images, MAP65-3 is pseudocolored in green, MTs in red, and

DNA in blue.

(A) to (C) At late anaphase, the midzone MTs are still not as prominent as

the shortening kinetochore fibers. MAP65-3 is concentrated in the

middle of the central spindle.

(D) to (F) When sister chromosomes reach spindle poles, midzone MTs

become highly bundled in parallel. MAP65-3 accumulated in the middle

segments of the midzone MTs. Arrows indicate MAP65-3 association

with the distal ends of MTs emanated from the spindle pole.

(G) to (I) At telophase, when the antitubulin dark line (arrowheads)

becomes visible in the middle of the developing phragmoplast, the

localization of MAP65-3 becomes much more restricted to the midline of

the phragmoplast.

(J) to (L) At late cytokinesis, the phragmoplast becomes a ring surround-

ing the forming cell plate. MAP65-3 remains associated with the phrag-

moplast midline, where antiparallel MTs are present.

Bar = 5 mm.

2912 The Plant Cell

antitubulin line started to appear (arrowheads, Figure 4H). At

this stage, MAP65-3 became restricted to the middle region of

the central spindle and seemed to connect the two sets of

bundled MTs (Figures 4G to 4I). Later, MTs radiating from

spindle poles disappeared, and the central spindle was re-

placed by the phragmoplast. Throughout the development of

the phragmoplast, MAP65-3 was always associated with the

midline of the phragmoplast, wherever the MTs from the op-

posite sides of the division plane met (Figures 4J to 4L). Such a

localization pattern persisted until the complete depolymeriza-

tion of phragmoplast MTs.

We analyzed the distribution patterns of MT bundles and

MAP65-3 in the central spindle and the phragmoplast based on

quantitative measurement of the fluorescent intensities. When

the antitubulin intensity was measured along MT bundles in the

central spindle, it was found that the signal dropped toward

themiddle segment of suchMT bundles (Figures 5A and 5B). The

reduction in antitubulin signal was coincident with the rise of anti-

MAP65-3 signal (Figures 5A and 5B). The reversed changes of

antitubulin and anti-MAP65-3 signals remained when MAP65-3

became restricted toa narrow region in the central spindle (Figures

5C and 5D). At this stage, as well as during cytokinesis, the anti-

tubulin fluorescence away from the midline became striking. In

comparison, the signal in themidlinewas so low that it was almost

invisible in images exposed according to the fluorescent intensity

in the rest of the phragmoplast. Ourmeasurement showed that the

intensity of the background fluorescencewas less than half of that

of the lowest point of antitubulin fluorescence. Thus, MTs did

cross the midline of the phragmoplast.

MAP65-3 Alone Is Sufficient to Cross-Link MTs In Vitro

It was suggested in the CPAM model that MAP65 could be

among the CPAM-associated proteins that interact with MT plus

ends (Austin et al., 2005). If MAP65-3 were associated with

CPAM, we would expect it to form an end-on interaction with MT

plus ends. To test this possibility, a bacterially expressed 63His-

GFP-33FLAG-MAP65-3 fusion protein was prepared, and the

full-length fusion protein and adegradation productwere purified

by successive affinity chromatography and gel filtration (Figure

6A). The tendency of this MAP65 family of proteins to dimerize

might be the reason that the full-length and truncated proteins

copurified with each other. When they were tested for MT

binding, both the intact (arrow, Figure 6B) and the degradation

product (arrowhead) exhibited strong binding to bovine brain

MTs as demonstrated by the MT cosedimentation experiment

(Figure 6B). In the absence of MTs, at the maximum concentra-

tion of 0.3 mM MAP65-3, the fusion protein remained in the

supernatant. To test whether MT cosedimentation of the fusion

protein was accompanied by MT bundling, 2.7 mM fluorescently

labeled MTs were observed inthe presence of 0.0125–0.1 mM

63His-GFP-33FLAG-MAP65-3. As a negative control, 0.1 mM

glutathione S-transferase (GST)-GFP-33FLAG was incubated

with MTs, and this did not cause any obvious MT bundling effect

(Figure 6C). MTs were seen as discrete single filaments in cases

of MTs alone or plus GST-GFP-33FLAG (Figures 6C and 6D). In

the presence of 63His-GFP-33FLAG-MAP65-3, MT bundles

appeared (Figure 6C). Concomitant with the greater amount

of 63His-GFP-33FLAG-MAP65-3, more MT bundles were

observed. In the presence of 0.1 mM 63His-GFP-33FLAG-

MAP65-3, thick MT bundles were induced (Figure 6C). We also

quantified bundles of MTs formed at different concentrations of

63His-GFP-33FLAG-MAP65-3. When the fusion protein was

added at 0.0125 mM,;11.6%MTs appeared in bundles (Figure

6D). Upon addition of 0.1mMof 63His-GFP-33FLAG-MAP65-3,

over 41.2% were in bundles (Figure 6D).

We also tested whether MAP65-3 selectively bundled anti-

parallel MTs in vitro. Polarity-marked MTs were prepared

before the addition of 63His-GFP-33FLAG-MAP65-3. All bun-

dled MTs were decorated by the fusion protein. We selected

bundled MTs that had new segments grown out of the MT

seeds, so that the polarities of both MTs were clearly dis-

cerned. Among 24 bundled MT doublets, 21 had the polarities

of both MTs clearly discerned, and all showed a clear antipar-

allel arrangement, as shown in Figure 7. The antiparallel MTs

were bundled toward their plus ends, which faced inwardly

(Figures 7C to 7F). TheMT-overlapping regions were decorated

by 63His-GFP-33FLAG-MAP65-3 (Figures 7A to 7D). In the

meantime, the fusion protein was not detected along single and

Figure 5. Analysis of Antiparallel MT Bundles and Associated MAP65-3.

Merged images have MTs in red and MAP65-3 in green ([A] and [C]). The

regions subjected to intensity measurement are highlighted in white

boxes. In the intensity plots ([B] and [D]), the x axis represents the

distance in pixels along the boxed region, and the y axis represents

intensity in gray levels.

(A) and (B) In a late anaphase cell, the intensity of antitubulin immuno-

fluorescence drops toward the middle of MT bundles, which coincides

with the rise in MAP65-3 signal. Avg, average.

(C) and (D) When MAP65-3 is restricted to the midline of the phragmo-

plast, the intensity of MTs drops sharply in the middle, but is still

significantly higher than that of the background. Again, the drop in the

antitubulin signal coincides with the rise in that of anti-MAP65-3.

Bar = 5 mm.

Phragmoplast Microtubule Organization 2913

free MTs (Figure 7D). These results suggest that MAP65-3 is

sufficient to cross-linkMTs in an antiparallel fashion in vitro, and

consequently MT bundles can be induced.

MAP65-3 Is Enriched near the Plus Ends of Antiparallel MTs

Because MAP65-3 was detected at the bridging region of

antiparallel MTs after late anaphase, we asked whether the

MAP65-3–MT interaction only took place in the presence of

antiparallel MTs in vivo. The temperature-sensitive Radially

Swollen7 (rsw7) mutation at a locus encoding a Kinesin-5 motor

causes bipolar spindles to collapse into monopolar ones at

restrictive temperatures (Bannigan et al., 2007). In collapsed

monopolar spindles in mutant cells, MAP65-3 was detected

atthe distal plus ends ofMTs radiating away from the single pole

(arrowheads, Figure 8). Occasionally, fragments ofMTs attached

to monopolar spindles exhibited a bipolar appearance, as indi-

cated by a dark line by antitubulin immunofluorescence (arrow,

Figure 8). In such fragments, the minus ends of the antiparallel

MTs were located toward the distal ends when revealed by anti-

g-tubulin staining (see Supplemental Figure 4 online). A more

pronounced MAP65-3 signal was detected in association with

MTs in the middle region of the antiparallel MTs when compared

with the signal associated with distal ends of emanating MTs

(Figure 8). Therefore, the result indicates that MAP65-3 specif-

ically recognizes MT segments near the plus end at later stages

of mitosis, and the association becomes greatly enhanced when

encountering the plus ends of antiparallel MTs.

Accumulation of MT Plus Ends in the Phragmoplast Midline

Is Dependent on MAP65-3

Because the loss of MAP65-3 leads to the widening of the

dark line that is visualized by antitubulin immunofluorescence

(Muller et al., 2004), we asked whether the phenotype was

caused by defects in MT plus end distribution. The functional

Figure 6. Preparation of the 63His-GFP-33FLAG-MAP65-3 Fusion Protein and MT Cosedimentation and Bundling Tests.

(A) Purified 63His-GFP-33FLAG-MAP65-3 and GST-GFP-33FLAG proteins. The GFP-MAP65 lane has two major bands of full-length 63His-GFP-

33FLAG-MAP65-3 and a degradation product. The GFP lane has a single major band of GST-GFP-33FLAG. The molecular weight markers (M; in kD)

are indicated on the left.

(B) Cosedimentation of 63His-GFP-33FLAG-MAP65-3 with MTs. In the absence of MTs, the fusion protein appears in the supernatant. At 0.075–0.3

mM, both the full-length fusion protein (arrow) and themajor degradation product (arrowhead) appear in the pellet fractions with MTs. The tubulin band is

marked by the asterisk. The molecular weight markers (in kD) are indicated on the left.

(C)MT bundling caused by the addition of 63His-GFP-33FLAG-MAP65-3. MTs by themselves appear in discrete single filaments. The addition of GST-

GFP-33FLAG does not make a difference. When 63His-GFP-33FLAG-MAP65-3 is added, greater bundling can be seen when more fusion protein is

added.

(D) Quantification of MT bundling induced by 63His-GFP-33FLAG-MAP65-3. At 0.075 to 0.3 mM, more than 10 to 40% of MTs appear in bundles.

2914 The Plant Cell

EB1c-33FLAG-GFP was expressed in the dyc283mutant back-

ground, and its localization was compared with that of control

cells expressing the identical fusion protein. In control cells,

similarly towhat has been shown in live cells, the EB1c-33FLAG-

GFP signal was particularly abundant in the phragmoplast mid-

line while also present elsewhere in both mature and a late

phragmoplasts (Figures 9A to 9F). In the dyc283 mutant cells,

however, EB1c-33FLAG-GFP no longer accumulated in the

widened midlines in phragmoplasts of similar stages (Figures 9G

to 9L). Instead, the signal becamemore or less evenly distributed

in the phragmoplast region and elsewhere in the cell. Hence, we

conclude that MAP65-3 is required for the rich presence of MT

plus ends in the middle region of the phragmoplast.

MAP65-3 Is Required for MT Plus End Localization of

Kinesin-12 and PAKRP2 in the Phragmoplast

Our previous work showed that the MT motor Kinesin-12

exhibits exclusive localization at MT plus ends in the phragmo-

plast (Lee and Liu, 2000; Pan et al., 2004; Lee et al., 2007). We

asked whether it colocalized with MAP65-3. A transgenic line

was established to express a MAP65-3-FLAG fusion protein

under its native promoter. The seedling growth retardation and

Figure 7. MAP65-3 Preferentially Bundles Antiparallel MTs In Vitro.

Two examples are shown.

(A) and (D) Fluorescent images of MT seeds MTs (first row), Rhodamine

X–labeled MTs (second row), MAP65-3 (third row), and merged images

(fourth row) with MT seeds in blue, newly grown out MTs in red, and

63His-GFP-33FLAG-MAP65-3 in green.

(B) and (E) Fluorescence intensity scans across the areas of bundled

MTs in (A) and (D) with the starting point marked by an arrow and the

ending point by an arrowhead, respectively. The x axis represents

distance in mm, and the y axis represents gray level in an 8-bit scale.

(C) and (F) The diagrams depict the positions of MT seeds, elongated

segments of MTs, and At MAP65-3. The keys for the diagrams are shown

on the bottom.

Bar = 5 mm.Figure 8. Antiparallel MT Bundles Enrich MAP65-3 In Vivo.

Amonopolar spindlewas generated in the rsw7mutant cell at the restrictive

temperature. The merged image has MTs in red, MAP65-3 in green, and

DNA in blue. MAP65-3 can be seen in the distal ends of radiating MTs of an

aster-like array (arrowheads). Occasionally, MTs are arranged in an anti-

parallel fashion, and the fortuitous phragmoplast-likeMTs have theirmiddle

line (arrows) decorated by MAP65-3 with greater signal strength than the

protein appeared at the MT distal ends of the asters. Bar = 5 mm.

Phragmoplast Microtubule Organization 2915

cytokinetic defects caused by the dyc283 mutation were com-

pletely suppressed upon the expression of this fusion (see

Supplemental Figure 5 online). This line allowed us to detect

both MAP65-3 and Kinesin-12A in the same cells (Figure 10). In

a late anaphase cell, MAP65-3 was found to decorate under-

lying MT bundles in the central spindle; however, the Kinesin-

12A signal was barely detectable at the same location, but largely

diffused in the cytoplasm (Figures 10A to 10C). This result sug-

gests that MAP65-3 likely precedes Kinesin-12A to interact with

midzone MTs in the central spindle. In a cell undergoing cytoki-

nesis, both proteins exhibited exclusive decoration of the midline

of the phragmoplast (Figures 10D and 10E). However, the two

proteins did not completely overlap as visualized in the merged

image, in which there were areas where only MAP65-3 was seen

(arrows, Figure 10F). However, wherever Kinesin-12A was de-

tected, MAP65-3 was also present.

The loss of engagement of the plus ends of antiparallel MTs in

the dyc283 phragmoplast prompted us to examine whether

the localization of phragmoplast-specific kinesins was affected

in the mutant cells. Compared with control cells, in which

Kinesin-12A exclusively decorated the phragmoplast midline

(Figure 11A, a to c; 100%, n = 16), it was no longer concentrated

in the phragmoplast of the dyc283mutant cells. Instead, the anti–

Kinesin-12A signal became largely dispersed in the cytoplasm

(Figure 11A, d to f; 79%, n = 115).

Phragmoplast-Associated Kinesin-Related Protein2 (PAKRP2)

has a wider localization pattern in the phragmoplast than

Kinesin-12 (Figure 11B, a to c.) (Lee et al., 2001). Nevertheless,

its signal was more pronounced toward the phragmoplast

midline than elsewhere in the phragmoplast (Figure 11B, a to c;

100%, n = 55). In the dyc283 mutant cells, however, PAKRP2

was no longer concentrated in the phragmoplast midline (Figure

11B, d to f). Instead, it evenly decorated phragmoplast MTs in a

punctate manner, leaving behind a dark line in the middle (Figure

11B, d to f; 94%, n = 72). Thus, these results suggest that

MAP65-3 is required for the interaction of these kinesin motors

with MT plus ends in Arabidopsis.

MAP65-3 Is Not Essential for Cell Plate–Destined

Membrane Trafficking

Various membrane trafficking markers have been shown to

specifically decorate the phragmoplast midline (Bednarek and

Figure 9. Localization of EB1c-33FLAG-GFP in Control and dyc283

Cells by Immunofluorescence.

The merged images have EB1c-33FLAG-GFP in green, MTs in red, and

DNA in blue.

(A) to (F) In a control cell containing a mature phragmoplast ([A] to [C])

and another one at late cytokinesis ([D] to [F]), EB1c-33FLAG-GFP is

abundantly present along antiparallel MTs. Its signal is particularly

pronounced in the phragmoplast midline, which is shown in dark by

tubulin immunofluorescence.

(G) to (L) In dyc283 cells at similar stages as the control cells, the EB1c-

33FLAG-GFP signal no longer accumulated in the phragmoplast mid-

line.

Bar = 5 mm.

Figure 10. Dual Localization of MAP65-3 and Kinesin-12A by Immuno-

fluorescence with Anti-FLAG for the MAP65-3-FLAG Fusion Protein and

Anti-Kinesin-12.

The merged images have Kinesin-12A in green, MAP65-3 in red, and

DNA in blue.

(A) to (C) In a late anaphase/telophase cell, MAP65-3 is already asso-

ciated with bundled MTs, but Kinesin-12A is barely visible.

(D) to (F) In the phragmoplast, both signals are concentrated in the

midline. However, the MAP65-3 signal does not completely overlap with

that of Kinesin-12A. Kinesin-12A is absent from some of the regions

containing MAP65-3 signal (arrows).

Bars = 5 mm.

2916 The Plant Cell

Falbel, 2002; Jurgens, 2005).Wewonderedwhether the changes

in MT organization caused by the loss of MAP65-3 would alter

membrane trafficking patterns. The membrane fusion SNARE

protein KNOLLE appeared at the phragmoplast midline as

shown before (Figures 12A to 12C; 100%, n = 27) (Lauber

et al., 1997). In the dyc283 mutant cells undergoing cytokinesis,

KNOLLE again predominantly appeared in the cell division site,

and the signal often appeared in thicker lines than in the control

cells (Figures 12D to 12F; 100%, n = 37). Similarly, the dynamin

Dynamin-Related Protein 1A (DRP1A) was found to be concen-

trated in the phragmoplast midline in the dyc283 cells as in the

control cells (see Supplemental Figure 6 online). Thus, these

results indicate that the membrane trafficking required for cell

plate formation is not significantly affected by the absence of

MAP65-3, even though the MT plus ends become disengaged

under such conditions.

DISCUSSION

Our results show that in the phragmoplast, antiparallel MTs often

have their plus ends engaged with each other, and active MT

polymerization takes place at the cell division site. Our results also

suggest that MAP65-3 functions in cross-linking antiparallel MTs

near their plus ends during telophase and cytokinesis. Direct

evidence was provided here indicating that only overlapping MTs

interactwithMAP65-3.AlthoughotherMAP65 isoformshavebeen

implicated in cross-linking antiparallel MTs in the phragmoplast

(Van Damme et al., 2004b; Smertenko et al., 2008), MAP65-3

apparently plays a more critical role than other isoforms. The MT

cross-linking activity of MAP65-3 is required for the interaction of

kinesins that are important for cytokinesis with MT plus ends, and

is consequently critical for cytokinesis.

MTOrganization in the Central Spindle and

the Phragmoplast

In the central spindle at late anaphase and telophase, long

segments of antiparallel MTs overlap toward the midzone and

Figure 11. Requirement of MAP65-3 for the Localization of Kinesin-12

and PAKRP2 at MT Plus Ends in the Phragmoplast.

(A) Localization of Kinesin-12A in the wild-type and dyc283 mutant cells.

The merged images have Kinesin-12A in green, MTs in red, and DNA in

blue. Kinesin-12A exclusively decorates the midline of the phragmoplast

MTs in a wild-type control cell ([a] to [c]). Kinesin-12A is no longer detected

in the phragmoplast in dyc283 cells ([d] to [f]).

(B) Localization of PAKRP2 in the wild-type and dyc283 mutant cells. The

merged images have PAKRP2 in green, MTs in red, and DNA in blue.

PAKRP2 is associated with phragmoplast MTs with greater emphasis near

themidline of the phragmoplast in a wild-type control cell ([a] to [c]). Puctate

PAKRP2 signal can be seen among phragmoplast MTs in a dyc283 cell ([d]

to [f]). The PAKRP2 signal appears along phragmoplast MTs, and does not

accumulate at the division site in a dyc283 cell ([g] to [i]).

Bars = 5 mm. Figure 12. Accumulation of the Syntaxin KNOLLE in the Phragmoplast

Midline in the Absence of MAP65-3.

Triple labeling of KNOLLE ([A] and [D]), MTs ([B] and [E]), and DNA in

cytokinetic cells of the wild-type control ([A] to [C]) and dyc283 mutant

([D] to [F]). The merged images ([C] and [F]) have KNOLLE pseudocol-

ored in green, MTs in red, and DNA in blue. KNOLLE can be detected in

the midline (arrows) of phragmoplast MTs in both the wild-type (A) and

dyc283 mutant (D) cells. Bar = 5 mm.

Phragmoplast Microtubule Organization 2917

are cross-linked by factors such as MAP65-3. Through an un-

known mechanism, the overlapping region of these interdigitat-

ing MTs (IMTs) is gradually reduced while the cell progresses

toward cytokinesis. Although MAP65-3 becomes restricted to a

narrow region when a typical phragmoplast MT array is estab-

lished, theMAP65-3 signal is believed to reflect the presence of a

narrow MT-overlapping region. The decrease in MT density

seems to be coincident with the localization of MAP65-3, which

highlights the cross-linked antiparallel MTs.

In the Haemanthus endosperm cells as well as the moss

protenema cells, antiparallel MTs clearly interdigitate or overlap

near their plus ends in the phragmoplast (Hepler and Jackson,

1968; Hiwatashi et al., 2008). In root meristematic cells in

Arabidopsis, however, electron tomographic results showed

that antiparallel MTs were largely separated by membranous

sacks of CPAM enclosing large quantities of carbohydrates

(Austin et al., 2005). This discrepancy had left us with the question

of whether MT organization patterns are different in phragmo-

plasts of different cells. Based on our biochemical, TEM, and

fluorescent imaging data, we provide direct evidence that a

subpopulation of antiparallel MTs indeed interdigitates in the

phragmoplast. The interdigitation is brought about by the MT–

cross-linking factor MAP65-3. Our data do not contradict the

finding that the majority of antiparallel MTs do not interdigitate in

the phragmoplast.

How dowe explain the appearance of a dark line in themiddle

of the phragmoplast in both antitubulin immunofluorescence

and GFP-tubulin images? The dark line is likely caused by an

imaging artifact due to the sharp difference in fluorescence

intensity between the midline and the rest of the phragmoplast.

The linescan of fluorescent intensity showed that the MT signal

intensity in the midline was significantly higher than the back-

ground signal. However, when the neighboring fluorescent

signal was overwhelmingly strong because of the abundance

of MTs, themidline seemed dark due to significantly fewer MTs.

A similar example can be found in the study of the Kinesin-14

memberArabidopsis thalianaKinesin 5 (ATK5) in prometaphase

spindles (Ambrose and Cyr, 2007). Upon the breakdown of the

nuclear envelope, MTs, highlighted by an MBD (MT binding

domain)-DsRed, were organized into a spindle configuration

with the DsRed signal largely seen at the spindle periphery and

spindle pole region. However, the yellow fluorescent protein

(YFP)-ATK5 signal occupied themiddle region of the spindle, as

the fusion protein decorated MT bundles. Although YFP-ATK5

seemed to be present in areas devoid of MTs, the linear

appearance of the signal suggests that this labeling was most

likely due to the presence of underlying MTs in the area. Hence,

we conclude that MTs are present in the region of the phrag-

moplast midline.

AModular Model of the Phragmoplast MT Array

Based on the findings presented here, we postulate a model

depicting the organization of antiparallel MTs in both the central

spindle and the phragmoplast (Figure 13). In this modular model,

both IMTs and noninterdigitating MTs (nIMTs) are represented,

but nIMTs outnumber IMTs. IMTs and nIMTs are integrated

together to form modules. Each module has the IMTs in the

center surrounded by nIMTs. Earlier modules are established

during telophase, when the central spindle contains conspicu-

ous MT bundles. These MT bundles have lower MT density

toward the middle.

Such amodularmodel echoes themini-phragmoplast concept

that was introduced when describing clusters of antiparallel MTs

during endosperm cellularization (Otegui and Staehelin, 2000).

Wewould like to refer to eachmodular cluster of IMTs and nIMTs

as a mini-phragmoplast MT array (Figure 13). In this array,

MAP65-3 as well as Kinesin-12 only associate with IMTs toward

their plus ends. Because of the cross-linking effect, IMTs are

predicted to be highly stable, and their plus ends remain at the

phragmoplast midline. Among the MAP65 members that deco-

rate phragmoplast MTs, MAP65-3 has the most restricted as-

sociation with the phragmoplast midline (Van Damme et al.,

2004a; Van Damme et al., 2004b). Compared with other MAP65

members, MAP65-3 exhibits the slowest turnover rate (Smertenko

et al., 2008), suggesting stable association with IMTs. Both IMTs

and nIMTs undergo robust MT polymerization, as demonstrated by

the EB1c activity.

During cytokinesis, the phragmoplast MT array expands

centrifugally toward the cell periphery. The older parts of the

array, toward the center, gradually disappear, while the new

ones appear as if the MTs of the new modules assemble at the

same rate as the MTs of the old modules disassemble. How do

the mini-phragmoplast modules propagate? This is believed to

be accomplished by the robust polymerization of nIMTs and

their interactions with cross-linkers, like MAP65-3. ATK5 may

be one of the factors that promotes rapid MT polymerization at

the advancing edge of the phragmoplast, as suggested by its

preferential localization to this region (Ambrose et al., 2005).

Once MAP65-3 binds to the plus-end regions of these MTs, it

would seek to interact with antiparallel MTs initiated from the

opposite side of the phragmoplast. One may also consider that

newly polymerized antiparallel MTs could be captured by

MAP65-3 already associated with extant ones. Based on our

in vitro results, we hypothesize that, once bound to antiparallel

MTs, MAP65-3 undergoes self-oligomerization by recruiting

more of its dimers. The antiparallel MTs cross-linked by

MAP65-3 are then stabilized and become IMTs. Concomitantly,

nIMTs continue to undergo rapid polymerization and depoly-

merization. It is worth noting that in Haemanthus endosperm

cells, accessory phragmoplasts can be formed when the cells are

incubated at elevated temperatures (Bajer et al., 1993). Pro-

nounced MT elongation took place under these conditions. One

could interpret the formation of accessory phragmoplasts asbeing

due to the establishment of an antiparallel array resulting from

fortuitous interactions between MTs that will be stabilized upon

interactions with proteins like MAP65-3.

During cellularization in the endosperm, the mini-phragmoplast

contains clusters that each consist of ;10 overlapping MTs

(Otegui and Staehelin, 2000). Surprisingly, nonoverlapping MTs

were not obvious in this case. Such a featuremay be unique to the

formation of a syncytial-type cell plate. In somatic cells, however,

the nIMTs become dominant. Whether this difference in MT

organization between endosperm cellularization and somatic cy-

tokinesis reflects the difference in cell plate formation remains to

be tested.

2918 The Plant Cell

Functions of Other MAP65 Family Proteins in

the Phragmoplast

Although MAP65-3 plays a critical role in phragmoplast MT

organization, other MAP65 members also are detected in the

Arabidopsis phragmoplast by GFP tagging and immunofluo-

rescence (Smertenko et al., 2000; Van Damme et al., 2004a).

However, their localization patterns differ from each other.

Unlike MAP65-3, for example, a MAP65-1-GFP fusion protein

did not appear in the midzone but along phragmoplast MTs

elsewhere (Gaillard et al., 2008). Because of their common

cross-linking and bundling activity (Smertenko et al., 2004;Mao

et al., 2005; Gaillard et al., 2008; Smertenko et al., 2008), it is

intriguing to determine whether MAP65 isoforms exhibit func-

tional redundancy in the phragmoplast. However, mutations in

genes encoding MAP65-1 and -2 only cause obvious pheno-

types in elongating cells (Lucas et al., 2011). In tobacco, the

MAP65-1 protein, which is highly similar to the Arabidopsis

MAP65-1, plays a critical role in MT turnover in the phragmo-

plast (Sasabe et al., 2006). Its phosphorylation by aMAP kinase

downregulates its MT-bundling activity and consequently pro-

motes the destabilization of MTs and expansion of the phragmo-

plast. In A. thaliana, MAP65-1, -2, and -3 all have been shown to

serve as putative substrates of a similar kinase at the phragmo-

plast midzone (Sasabe et al., 2011). Moreover, mutations in the

genes encoding MAP65-1 or -2 would enhance the cytokinetic

phenotype caused by the loss of MAP65-3, suggesting that there

is functional redundancy among the three homologous proteins

(Pesquet and Lloyd, 2011). However, it is unclear how phragmo-

plast MTs are altered in these double mutants when compared

with single mutants like dyc283. It would be interesting to test

whether ectopic expression of MAP65-1 or -2 could suppress the

cytokinetic phenotypes in dyc283 cells.

Maintenance of the Bipolar Configuration of

Phragmoplast MTs

The localization of MAP65-3 could imply that it might be critical

for maintaining the bipolar arrangement of antiparallel MTs in

the phragmoplast. However, genetic data indicate that MAP65-3

is not required for either establishing or maintaining the bipolar

pattern. Its absence causes antiparallel MTs to be disengaged

with each other, as demonstrated by a wide gap in the phragmo-

plastmidline in the plemutant (Muller et al., 2004). The loss of such

close engagement seriously compromises the effectiveness of cell

plate formation, but doesnot completely abolish the process. How

do highly dynamic MTs remain in a bipolar configuration in the

phragmoplast? The bipolarity may be brought about by the uni-

directional MT polymerization toward the division site in the

phragmoplast. It still remains to be determined how the phrag-

moplast prevents MTs from being polymerized toward the op-

posite direction.

When exogenous tubulins were supplied to permeabilized

cells containing phragmoplasts, the added tubulin subunits

predominantly localized to the phragmoplast midline, presum-

ably onto the plus ends of IMTs (Vantard et al., 1990; Asada

et al., 1991). These observations seem to be contradictory to

the scenario of robust polymerization taking place for both IMTs

and nIMTs. We assumed that the result was due to the loss of

nIMTs caused by harsh experimental conditions, because both

experiments required a permeabilization step to allow tubulins

Figure 13. A Model Depicting MT Organization in the Central Spindle at Late Anaphase or Early Telophase and the Phragmoplast during Cytokinesis.

The keys of the diagrams are included in the box. Two populations of MTs, IMTs and nIMTs, are present at both stages. MAP65-3 and Kinesin-12 only

associate with IMTs near their plus ends. EB1 decorates polymerizingMT plus ends of both IMTs and nIMTs. Plus (+) andminus (2) ends of MTs are illustrated.

Phragmoplast Microtubule Organization 2919

to freely enter the cells. IMTs likely exhibit higher stability than

nIMTs, because of being stabilized by MAP65-3.

Whenmore tubulin dimerswere added to the plus ends of IMTs,

theseMTs tended to grow across the phragmoplast midline. If the

plus ends were not properly positioned, cytokinesis failed due to

irregular deposition of cell plate materials (Lee et al., 2007). The

function of the Kinesin-12 motor is to keep the plus ends posi-

tioned at the division site. Although the initial sliding of central

spindleMTs depends on Kinesin-5, additional segments added to

IMTs are pushed apart by Kinesin-12 (Lee et al., 2007).

Other MT-interacting factors, like EB1 and Microtubule Or-

ganization1, may not directly contribute to the establishment of

the bipolar phragmoplast MT array (Eleftheriou et al., 2005;

Kawamura et al., 2006). However, they contribute to other

processes, like the regulation of polymerization and depoly-

merization, as well as the stability of the MTs. These and other

MT-interacting factors become localized in the phragmoplast

when they interact with MTs there. Before interacting with MTs,

they would diffuse in the cytoplasm.

Interaction of Proteins atMTPlus Ends in the Phragmoplast

Our results show that stable interactions with MT plus ends in the

phragmoplast require the IMTs and nIMTs probably undergo rapid

turnover. Motor proteins, like Kinesin-12 and PAKRP2, exhibit

conspicuous localization to the MT plus ends in the phragmoplast

(Lee and Liu, 2000; Lee et al., 2001; Pan et al., 2004). When IMTs

were lost, their plus end association was compromised. If the motor

accumulated at the plus end by active movement, it would become

evenly distributed along MTs when the IMTs were lost, as seen in

PAKRP2 here. However, if a motor was directly bound to the plus

endswithoutwalkingalongMTs, its localizationwouldbecompletely

lost in the absence of IMTs, as seen in Kinesin-12. It is unknown

whetherMAP65-3directly interactswithKinesin-12atMTplus ends.

In summary, our results suggest thatMAP65-3 decorates IMTs

but not nIMTs in the phragmoplast. Successful cytokinesis

requires the activity of both IMTs and nIMTs.

METHODS

Plant Materials, Growth Conditions, and Transformation

Arabidopsis thaliana materials used in this study include the wild-type

(Columbia-0) plants, and the dyc283 (Caillaud et al., 2008), rsw7 (Bannigan

et al., 2007), and eb1c (Bisgrove et al., 2008) mutants. Plant growth con-

ditions and agrobacterium-mediated transformation are as described

previously (Kong et al., 2010).

For observation of the EB1c-33FLAG-GFP and MAP65-3-FLAG fu-

sions, seeds were germinated on 0.8% phytagel (Sigma-Aldrich) agar

plates containing half-strength Murashige and Skoog medium (Sigma-

Aldrich) with 0.5 g/L MES (Acros) (pH 5.8) after stratification in the dark at

48C. Oryzalin (Sigma-Aldrich) was added to the germination medium at

100 nM to test the hypersensitivity of the eb1c mutant to the drug

(Bisgrove et al., 2008).

Establishment of Lines Expressing MAP65-3-FLAG and

EB1c-33FLAG-GFP

A MAP65-3-FLAG fusion protein was expressed under the native At

MAP65-3 promoter. To do so, the 1.3-kb promoter region was amplified

by PCR using the Phusion High-Fidelity DNA polymerase (New England

Biolabs) and primers AtMAP65 1: 59-CAC CAC ACT CTT CCC TAC ACA

AAA CCG C-39 and AtMAP65 2: 59-GAA GAA TCG GAT CTT TTT GAA

CAC TTG CCA T-39. The MAP65-3-coding sequence was amplified

using the full-length cDNA clone RAFL09-53-C05 (Seki et al., 1998; Seki

et al., 2002) as the template and the primers AtMAP65 3: 59-ATG GCA

AGTGTTCAAAAAGATCCGATTCTTC-39 andAtMAP65 8: 59-AACCAA

ACG ACA TTC AGA CTG TAG CAT G-39. The resulting two DNA frag-

ments were fused by an additional PCR reaction using the primers

AtMAP65 1 and AtMAP65 8. The fused product was cloned into the

Gateway pENTR/D-TOPO vector (Invitrogen), followed by cloning into the

destination vector pGWB10 (Nakagawa et al., 2007) by an LR recombi-

nation reaction according to the manufacturer’s instructions, rendering a

FLAG tag in the C terminus after expression. The resulting plasmid was

transformed into the dyc283 mutant to test for genetic suppression/

complementation.

To detect EB1c in live cells, an EB1c-33FLAG-GFP fusion protein was

expressed under the control of its native promoter. Briefly, a 2399-bp

genomic fragment encompassing a 665-bp promoter region plus the

EB1c-coding sequence was amplified by PCR with the Phusion poly-

merase using the primers EB1c-F2: 59-CAC CGC GGC CGC TTG GAC

AGG TTA ATG GGC TTG TG-39 and EB1c-R2: 59-GTC ATC TAG AGC

AGG TCA AGA GAG GAG ATG AAC C-39 (the NotI and XbaI sites are

underlined). The PCR product was cloned into the p33FLAG-CMV-14

vector (Sigma-Aldrich) at the NotI and XbaI sites. The At EB1c genomic

sequence together with the 33FLAG-coding sequence was subse-

quently amplified together by the primers EB1c-F2 and FLAG33-R1:

59-CTT GTC ATC GTC TCC TTG TAG TC-39. The resulting fragment was

cloned into pENTR/D-TOPO and then into the pGWB4 vector (Nakagawa

et al., 2007) by an LR recombination reaction. The resulting plasmid was

transformed into eb1c mutant plants. Transformants were tested for

genetic suppression against hypersensitivity to 100 nM oryzalin. The

same plasmid was transformed into the dyc283 mutant plants.

Fusion Protein and Antibody Production

To produce a GST-MAP65-3 fusion protein as an antigen, the HindIII

fragment of the cDNA encoding amino acid 161-487 was cleaved from the

RAFL09-53-C05 plasmid and cloned into the pGEX-KG vector (Guan and

Dixon, 1991) at theHindIII site. TheGST-MAP65-3161-487 fusion proteinwas

expressed in BL21 (DE3) pLysS cells (EMD Biosciences Novagen) and

purified using immobilized glutathione resin (Pierce Chemical Co.) before

being used for antibody production at the Comparative Pathology Labo-

ratory on the University of California-Davis campus. Specific antibodies

against MAP65-3 were purified using columns of GST and GST-MAP65-

3161-487 proteins, which had been immobilized on agarose using the

AminoLink Plus Kit (Pierce Chemical Co.), according to the manufacturer’s

instructions. Briefly, anti-GST antibodies were depleted from the anti-

serum by the GST column. The flow-through was applied to the GST-

MAP65-3161-487 column, and subsequently specific antibodies were

eluted from the column with 100 mM glycine (pH 2.5). Purified antibodies

were immediately neutralized with 1/10 volume of 1 M Tris HCl (pH 8.0).

To produce a full-length MAP65-3 fusion protein, the MAP65-3–coding

sequence was amplified from the cDNA clone RAFL09-53-C05 with the

primers At MAP65-3 5XbaI: 59-CAC CTC TAG AGC AAG TGT TCA AAA

AGA TCC GAT TC-39 and AtMAP65-3 3BHI: 59-GAT CGG ATC CTC AAA

CCA AAC GAC ATT CAG ACT GT-39 (the XbaI and BamHI sites are

underlined). The PCR product was cloned into the p33FLAG-CMV-7

vector (Sigma-Aldrich) at the XbaI and BamHI sites. The 33FLAG-

MAP65-3 fragment was then amplified by the primers N-33FLAG-f1:

59-CAC CGA CTA CAA AGA CCA TGA CGG TGA TTA TAA-39and

AtMAP65-3 3BHI. The resulting fragment was cloned into pENTR/D-

TOPO, and subsequently into the pGWB6 vector (Nakagawa et al.,

2007) by an LR recombination reaction. The coding sequence of the

2920 The Plant Cell

GFP-33FLAG-MAP65-3 fusion was amplified by PCR using primers

GFP5KpnI: 59-CAC CGG TAC CAT GGT GAG CAA GGG CGA GGA GCT

G-39 and AtMAP65-3 3GXhoI: 59-CAT CTCGAGCAACCAAACGACATT

CAG ACT GTA GCA TG-39. After digestion with KpnI and XhoI, the

fragment was cloned into the pET30a vector at the corresponding sites.

The ligation product was then transformed into Rosetta2 competent cells

(Novagen). The expressed 63His-GFP-FLAG-MAP65-3 fusion protein

was purified using Ni Sepharose 6 Fast Flow according to the manufac-

turer’s instructions (GEHealthcare Life Sciences). The protein was further

purified by gel filtration using HiPrep Sephacryl S-300 (GEHealthcare Life

Sciences) according to a published protocol (Tao and Scholey, 2010). To

prepare a protein to be used as a negative control in the in vitro MT

binding assay, a GST-GFP-33FLAG fusion construct was made by

digesting the GST-GFP-33FLAG-MAP65-3 fusion construct with BamHI

and XbaI. The 0.7-kb fragment containing the GFP-33FLAG-coding

sequence was cloned into the pGEX-KG vector at the BamHI and XbaI

sites. Expressed fusion protein was purified as described above.

MT Cosedimentation and In Vitro MT Binding Assays

Purified 63His-GFP-33FLAG-MAP65-3 and GST-GFP-33FLAG proteins

were dialyzed against BRB80 (80mMPipes, 1mMMgCl2, 1mMEGTA, pH

6.8) and centrifuged at 100,000 3 g for 15 min at 48C to remove aggrega-

tions before being used for MT cosedimentation and in vitro MT binding

assays. A MT cosedimentation assay was performed as described previ-

ously (Lee and Liu, 2000). The binding reaction was performed in a 38-mL

reaction volume containing 2.7 mMMTs and 0.075, 0.15, or 0.3 mM63His-

GFP-33FLAG-MAP65-3 in the presence of 10 mM taxol in BRB80. As

negative control experiments, the same assay was performed in the

absence of either MTs or 63His-GFP-33FLAG-MAP65-3.

For in vitro MT binding assays, fluorescent MTs were polymerized from

Rhodamine X–labeled bovine brain tubulin (Cytoskeleton) and stabilized

in 20mM taxol in BRB80 supplementedwith 1mMGTP and 10%glycerol.

The binding reaction was performed in a 20-mL reaction volume contain-

ing 0.0125, 0.025, 0.05, or 0.1 mM 63His-GFP-33FLAG-MAP65-3 with

MTs polymerized from 0.06 mM Rhodamine X–labeled bovine brain

tubulin (Cytoskeleton) in the same buffer. The mixture was immediately

loaded onto a glass slide and observed under an Axiovert 200 inverted

fluorescentmicroscopewith a 403Plan-Apo objective (Carl Zeiss). All the

images were taken within 10 min of 63His-GFP-33FLAG-MAP65-3

being mixed with MTs. As a negative control, 0.0125, 0.025, 0.05, or

0.1 mM of the GST-GFP-33FLAG fusion protein was incubated with an

identical amount of MTs as above.

The number of bundled MTs was counted by analyzing the fluorescent

intensity of the Rhodamine X in the filaments in each image. Bundled MTs

weredefinedas filamentswhosefluorescence intensitywas at least twice as

high as that of a single MT. The MT bundling index was defined as the

numberofbundledMTsdividedby thenumberof all theMTfilaments in each

image. For each mixture, the bundling index was calculated with at least

200–300 filaments. The average of three independent tests was plotted.

For the in vitro MT binding/bundling assay using polarity-marked MTs,

polarity-marked MTs were prepared by incubating Rhodamine X–labeled

bovine brain tubulin with short MT seeds polymerized from HiLyte 647-

labeled porcine brain tubulin (Cytoskeleton) for 1 h at 378C and stabilized

in 20mM taxol in BRB80 supplementedwith 1mMGTP and 10%glycerol.

The binding/bundling reaction was performed in a 20-mL reaction volume

containing 0.025 mM His-GFP3FLAG-MAP65-3, and polarity-marked

MTs were polymerized from 0.06 mM tubulin in the same buffer. Obser-

vations were performed as described above.

Immunoblotting Experiments

To extract proteins for an immunoblotting assay, developing flower buds

werecollected from thewild-type anddyc283plants and frozen andground

in liquid nitrogenwith amortar and pestle. A trichloroacetic acidprecipitation

method was used to extract proteins as described previously (Lee et al.,

2001). Extracted proteins were separated on a 7.5% SDS-PAGE gel and

processed for immunoblotting analysis. Purified anti-MAP65-3 antibodies

and the 3H11 antiactin monoclonal antibody (Andersland et al., 1994) were

used to probe the blots. The antibodies were diluted at 1:1000 in Hikari

reagent (Nacalai Tesque). Secondary antibodies were HRP-conjugated

anti-rabbit and anti-mouse antibodies (Bio-Rad). Signals were detected by

the chemiluminescent method of exposing an X-ray film on the blots.

TEM Examination of the Dividing Cells of Roots

Seedlings grown on the agar plate were rapidly frozen using an HPM 010

high-pressure freezing machine (Bal-Tec Products). Frozen samples were

freeze-substituted according to a published protocol (van de Meene et al.,

2006). Roots were embedded in Spurr’s resin (Ted Pella) and sectioned

using a microtome (Leica). After staining with uranium acetate and lead

citrate, root sections were observed under a 410 LS TEM (Philips).

Fluorescent and Confocal Microscopy

Root tip cells were processed for indirect immunofluorescence staining as

described previously (Lee and Liu, 2000). Besides anti-MAP65-3 anti-

bodies, other primary antibodies include DM1A anti–a-tubulin (Sigma-

Aldrich), sheep antitubulin (Cytoskeleton), anti–PAKRP1-C for Kinesin-12

(Lee and Liu, 2000), anti-PAKRP2 (Lee et al., 2001), anti-KNOLLE (Rose

Biotechnology), anti-FLAG (Shanghai Genomics), and anti-DRP1A (Kang

et al., 2001). Secondary antibodies were Texas Red-conjugated donkey

anti-mouse IgG (Molecular Probes) and FITC-conjugated donkey anti-

rabbit IgG (Rockland Immunochemicals).

Slides were observed under an Eclipse E600 microscope equipped

with epifluorescence optics (Nikon) and ET filter sets (Chroma Technol-

ogy Corp.). Images were captured by an Orca CCD camera (Hamamatsu

Photonics) driven by the MetaMorph software package (Molecular De-

vices) before being assembled by the Photoshop software (Adobe).

Z-stacks were taken under a DeltaVision Real Time Deconvolution

Microscope (Applied Precision), and images were processed using the

associated API softWoRx Explorer Suite.

The EB1c-33FLAG-GFP signal was visualized in the root tip cells of

3-to 4-day-old seedlings with their roots mounted in 0.3 M mannitol.

Time-lapsed imageswere taken using an FV1000 confocal laser scanning

microscope (Olympus USA) with a PLAPON 603 oil objective lens and

using the 488-nm argon laser for GFP excitation. The GFP signal was

collected at the 500–550 nm range. Acquired images were processed

using MetaMorph (Molecular Devices) and Photoshop (Adobe).

Accession Numbers

The Arabidopsis Information Resource (TAIR) locus identifier for the

genes mentioned in this study are At5g51600 encoding At MAP65-3 and

At5g67270 for At EB1c. The cDNA clone RAFL09-53-C05 contains the

full-length coding sequence of At MAP65-3.

Supplemental Data

The following materials are available in the online version of this article.

Supplemental Figure 1. Genetic Suppression/Complementation of the

Eb1c Mutation by Expressing an EB1c-33FLAG-GFP Fusion Protein.

Supplemental Figure 2. Detection of the Endogenous MAP65-3

Protein by Immunoblotting.

Supplemental Figure 3. Negative Control for the Anti-MAP65-3

Immunofluorescence.

Phragmoplast Microtubule Organization 2921

Supplemental Figure 4. Antiparallel MT Bundles in an rsw7Mutant Cell.

Supplemental Figure 5. Suppression of the dyc283 Mutation by

Expressing MAP65-3-FLAG.

Supplemental Figure 6. Localization of the Dynamin DRP1A in the

Wild-Type and dyc283 Cells.

Supplemental Movie 1. Time-Lapse Movie of the EB1c-33FLAG-

GFP Fusion in Two Cytokinetic Cells.

Supplemental Movie Legends.

ACKNOWLEDGMENTS

We thank members of the Liu laboratory for critical input in the project.

Our gratitude goes to Bruno Favery at Institut National de la Recherche

Agronomique in France for the dyc283 mutant, Tobias Baskin at

the University of Massachusetts, Amherst for the rsw7 mutant, Kazuo

Shinozaki and Motoaki Seki at the Plant Science Center of the RIKEN

Yokohama Institute in Japan for providing the cDNA clone, Tsuyoshi

Nakagawa at Shimane University in Japan for the pGWB vectors,

and Steven Backues and Sebastian Bednarek at the University of

Wisconsin-Madison for the DRP1A antibody. We especially want to thank

Li Tao and Jon Scholey for their advice on protein biochemistry and the

use of their fast protein liquid chromatography system and John Jordan at

Olympus USA for his assistance with the FV1000 confocal microscope.

This report is based on work supported by the National Science Foun-

dation under the grant MCB-0920454. Any opinions, findings, and con-

clusions or recommendations expressed in this material are those of the

authors and do not necessarily reflect the views of the National Science

Foundation. T.H. was a Katherine Esau postdoctoral fellow.

AUTHOR CONTRIBUTIONS

C.-M.K.H and B.L. designed the research. C.-M.K.H, T.H., F.G., R.W.R.,

and Y.-R.J.L performed the research and analyzed data. C.-M.K.H, T.H.,

and B.L. wrote the article together. B.L. revised the article.

Received July 23, 2010; revised June 28, 2011; accepted August 4, 2011;

published August 26, 2011.

References

Ambrose, J.C., and Cyr, R. (2007). The kinesin ATK5 functions in early

spindle assembly in Arabidopsis. Plant Cell 19: 226–236.

Ambrose, J.C., Li, W., Marcus, A., Ma, H., and Cyr, R. (2005). A

minus-end-directed kinesin with plus-end tracking protein activity is

involved in spindle morphogenesis. Mol. Biol. Cell 16: 1584–1592.

Andersland, J.M., Fisher, D.D., Wymer, C.L., Cyr, R.J., and

Parthasarathy, M.V. (1994). Characterization of a monoclonal an-

tibody prepared against plant actin. Cell Motil. Cytoskeleton 29:

339–344.

Asada, T., Sonobe, S., and Shibaoka, H. (1991). Microtubule translo-

cation in the cytokinetic apparatus of cultured tobacco cells. Nature

350: 238–241.

Austin II, J.R., Seguı-Simarro, J.M., and Staehelin, L.A. (2005).

Quantitative analysis of changes in spatial distribution and plus-end

geometry of microtubules involved in plant-cell cytokinesis. J. Cell

Sci. 118: 3895–3903.

Bajer, A.S., Smirnova, E.A., and Mole-Bajer, J. (1993). Microtubule-

converging centers-implications for microtubule dynamics in higher

plants. In Chromosome Segregation and Aneuploidy, B.K. Vig, ed

(Berlin: Springer-Verlag), pp. 225–239.

Bannigan, A., Scheible, W.-R., Lukowitz, W., Fagerstrom, C.,

Wadsworth, P., Somerville, C., and Baskin, T.I. (2007). A conserved

role for kinesin-5 in plant mitosis. J. Cell Sci. 120: 2819–2827.

Bednarek, S.Y., and Falbel, T.G. (2002). Membrane trafficking during

plant cytokinesis. Traffic 3: 621–629.

Bisgrove, S.R., Lee, Y.-R.J., Liu, B., Peters, N.T., and Kropf, D.L.

(2008). The microtubule plus-end binding protein EB1 functions in root

responses to touch and gravity signals in Arabidopsis. Plant Cell 20:

396–410.

Caillaud, M.-C., Lecomte, P., Jammes, F., Quentin, M., Pagnotta, S.,

Andrio, E., de Almeida Engler, J., Marfaing, N., Gounon, P., Abad,

P., and Favery, B. (2008). MAP65-3 microtubule-associated protein is

essential for nematode-induced giant cell ontogenesis in Arabidopsis.

Plant Cell 19: 423–437.

Chan, J., Calder, G., Fox, S., and Lloyd, C. (2005). Localization of the

microtubule end binding protein EB1 reveals alternative pathways of

spindle development in Arabidopsis suspension cells. Plant Cell 17:

1737–1748.

Eleftheriou, E.P., Baskin, T.I., and Hepler, P.K. (2005). Aberrant cell

plate formation in the Arabidopsis thaliana microtubule organization

1 mutant. Plant Cell Physiol. 46: 671–675.

Euteneuer, U., and McIntosh, J.R. (1980). Polarity of midbody and

phragmoplast microtubules. J. Cell Biol. 87: 509–515.

Gaillard, J., Neumann, E., Van Damme, D., Stoppin-Mellet, V., Ebel, C.,

Barbier, E., Geelen, D., and Vantard, M. (2008). Two microtubule-

associated proteins of Arabidopsis MAP65s promote antiparallel mi-

crotubule bundling. Mol. Biol. Cell 19: 4534–4544.

Guan, K.L., and Dixon, J.E. (1991). Eukaryotic proteins expressed in

Escherichia coli: An improved thrombin cleavage and purification

procedure of fusion proteins with glutathione S-transferase. Anal.

Biochem. 192: 262–267.

Hepler, P.K., and Jackson, W.T. (1968). Microtubules and early stages

of cell-plate formation in the endosperm of Haemanthus katherinae

Baker. J. Cell Biol. 38: 437–446.

Hiwatashi, Y., Obara, M., Sato, Y., Fujita, T., Murata, T., and Hasebe,

M. (2008). Kinesins are indispensable for interdigitation of phragmo-

plast microtubules in the moss Physcomitrella patens. Plant Cell 20:

3094–3106.

Hush, J.M., Wadsworth, P., Callaham, D.A., and Hepler, P.K. (1994).

Quantification of microtubule dynamics in living plant cells using

fluorescence redistribution after photobleaching. J. Cell Sci. 107:

775–784.

Jiang, C.-J., and Sonobe, S. (1993). Identification and preliminary

characterization of a 65 kDa higher-plant microtubule-associated

protein. J. Cell Sci. 105: 891–901.

Jurgens, G. (2005). Cytokinesis in higher plants. Annu. Rev. Plant Biol.

56: 281–299.

Kang, B.H., Busse, J.S., Dickey, C., Rancour, D.M., and Bednarek,

S.Y. (2001). The arabidopsis cell plate-associated dynamin-like pro-

tein, ADL1Ap, is required for multiple stages of plant growth and

development. Plant Physiol. 126: 47–68.

Kapitein, L.C., Janson, M.E., van den Wildenberg, S.M., Hoogenraad, C.

C., Schmidt, C.F., and Peterman, E.J.G. (2008). Microtubule-driven

multimerization recruits ase1p onto overlapping microtubules. Curr. Biol.

18: 1713–1717.

Kawamura, E., Himmelspach, R., Rashbrooke, M.C., Whittington, A.

T., Gale, K.R., Collings, D.A., and Wasteneys, G.O. (2006). MICRO-

TUBULE ORGANIZATION 1 regulates structure and function of mi-

crotubule arrays during mitosis and cytokinesis in the Arabidopsis

root. Plant Physiol. 140: 102–114.

Komaki, S., Abe, T., Coutuer, S., Inze, D., Russinova, E., and

2922 The Plant Cell

Hashimoto, T. (2010). Nuclear-localized subtype of end-binding 1

protein regulates spindle organization in Arabidopsis. J. Cell Sci. 123:

451–459.

Kong, Z., Hotta, T., Lee, Y.R., Horio, T., and Liu, B. (2010). The

g-tubulin complex protein GCP4 is required for organizing functional

microtubule arrays in Arabidopsis thaliana. Plant Cell 22: 191–204.

Lauber, M.H., Waizenegger, I., Steinmann, T., Schwarz, H., Mayer,

U., Hwang, I., Lukowitz, W., and Jurgens, G. (1997). The Arabidop-

sis KNOLLE protein is a cytokinesis-specific syntaxin. J. Cell Biol.

139: 1485–1493.

Lee, Y.-R.J., Giang, H.M., and Liu, B. (2001). A novel plant kinesin-

related protein specifically associates with the phragmoplast organ-

elles. Plant Cell 13: 2427–2439.

Lee, Y.R.J., and Liu, B. (2000). Identification of a phragmoplast-

associated kinesin-related protein in higher plants. Curr. Biol. 10:

797–800.

Lee, Y.R.J., Li, Y., and Liu, B. (2007). Two Arabidopsis phragmoplast-

associated kinesins play a critical role in cytokinesis during male

gametogenesis. Plant Cell 19: 2595–2605.

Liu, B., Joshi, H.C., and Palevitz, B.A. (1995). Experimental manipu-

lation of g-tubulin distribution in Arabidopsis using anti-microtubule

drugs. Cell Motil. Cytoskeleton 31: 113–129.

Liu, B., Joshi, H.C., Wilson, T.J., Silflow, C.D., Palevitz, B.A., and

Snustad, D.P. (1994). g-Tubulin in Arabidopsis: Gene sequence,

immunoblot, and immunofluorescence studies. Plant Cell 6: 303–314.

Lucas, J.R., Courtney, S., Hassfurder, M., Dhingra, S., Bryant, A.,

and Shaw, S.L. (2011). Microtubule-associated proteins MAP65–1

and MAP65–2 positively regulate axial cell growth in etiolated Arabi-

dopsis hypocotyls. Plant Cell 23: 1889–1903.

Mao, G.J., Chan, J., Calder, G., Doonan, J.H., and Lloyd, C.W. (2005).

Modulated targeting of GFP-AtMAP65-1 to central spindle microtu-

bules during division. Plant J. 43: 469–478.

Muller, S., Smertenko, A., Wagner, V., Heinrich, M., Hussey, P.J.,

and Hauser, M.T. (2004). The plant microtubule-associated protein

AtMAP65-3/PLE is essential for cytokinetic phragmoplast function.

Curr. Biol. 14: 412–417.

Nakagawa, T., Kurose, T., Hino, T., Tanaka, K., Kawamukai, M.,

Niwa, Y., Toyooka, K., Matsuoka, K., Jinbo, T., and Kimura, T.

(2007). Development of series of gateway binary vectors, pGWBs, for

realizing efficient construction of fusion genes for plant transforma-

tion. J. Biosci. Bioeng. 104: 34–41.

Otegui, M., and Staehelin, L.A. (2000). Syncytial-type cell plates: A

novel kind of cell plate involved in endosperm cellularization of

Arabidopsis. Plant Cell 12: 933–947.

Otegui, M.S., Verbrugghe, K.J., and Skop, A.R. (2005). Midbodies and

phragmoplasts: Analogous structures involved in cytokinesis. Trends

Cell Biol. 15: 404–413.

Pan, R., Lee, Y.R.J., and Liu, B. (2004). Localization of two homologous

Arabidopsis kinesin-related proteins in the phragmoplast. Planta 220:

156–164.

Pesquet, E., and Lloyd, C. (2011). Microtubules, MAPs and xylem

formation. In The Plant Cytoskeleton, B. Liu, ed (New York: Springer),

pp. 277–306.

Sasabe, M., Kosetsu, K., Hidaka, M., Murase, A., and Machida, Y.

(2011). Arabidopsis thaliana MAP65-1 and MAP65-2 function redun-

dantly with MAP65-3/PLEIADE in cytokinesis downstream of MPK4.

Plant Signal. Behav. 6: 743–747.

Sasabe, M., Soyano, T., Takahashi, Y., Sonobe, S., Igarashi, H., Itoh,

T.J., Hidaka, M., and Machida, Y. (2006). Phosphorylation of

NtMAP65-1 by a MAP kinase down-regulates its activity of microtu-

bule bundling and stimulates progression of cytokinesis of tobacco

cells. Genes Dev. 20: 1004–1014.

Seki, M., Carninci, P., Nishiyama, Y., Hayashizaki, Y., and Shinozaki,

K. (1998). High-efficiency cloning of Arabidopsis full-length cDNA by

biotinylated CAP trapper. Plant J. 15: 707–720.

Seki, M., et al. (2002). Functional annotation of a full-length Arabidopsis

cDNA collection. Science 296: 141–145.

Smertenko, A., Saleh, N., Igarashi, H., Mori, H., Hauser-Hahn, I.,

Jiang, C.-J., Sonobe, S., Lloyd, C.W., and Hussey, P.J. (2000). A

new class of microtubule-associated proteins in plants. Nat. Cell Biol.

2: 750–753.

Smertenko, A.P., Kaloriti, D., Chang, H.Y., Fiserova, J., Opatrny, Z.,

and Hussey, P.J. (2008). The C-terminal variable region specifies the

dynamic properties of Arabidopsis microtubule-associated protein

MAP65 isotypes. Plant Cell 20: 3346–3358.

Smertenko, A.P., Chang, H.Y., Wagner, V., Kaloriti, D., Fenyk, S.,

Sonobe, S., Lloyd, C., Hauser, M.T., and Hussey, P.J. (2004). The

Arabidopsis microtubule-associated protein AtMAP65-1: Molecular

analysis of its microtubule bundling activity. Plant Cell 16: 2035–2047.

Staehelin, L.A., and Hepler, P.K. (1996). Cytokinesis in higher plants.

Cell 84: 821–824.

Tao, L., and Scholey, J.M. (2010). Purification and assay of mitotic

motors. Methods 51: 233–241.

Van Damme, D., Bouget, F.Y., Van Poucke, K., Inze, D., and

Geelen, D. (2004a). Molecular dissection of plant cytokinesis and

phragmoplast structure: A survey of GFP-tagged proteins. Plant J.

40: 386–398.

Van Damme, D., Van Poucke, K., Boutant, E., Ritzenthaler, C., Inze,

D., and Geelen, D. (2004b). In vivo dynamics and differential micro-

tubule-binding activities of MAP65 proteins. Plant Physiol. 136: 3956–

3967.

van de Meene, A.M., Hohmann-Marriott, M.F., Vermaas, W.F., and

Roberson, R.W. (2006). The three-dimensional structure of the

cyanobacterium Synechocystis sp. PCC 6803. Arch. Microbiol. 184:

259–270.

Vantard, M., Levilliers, N., Hill, A.M., Adoutte, A., and Lambert, A.M.

(1990). Incorporation of Paramecium axonemal tubulin into higher

plant cells reveals functional sites of microtubule assembly. Proc.

Natl. Acad. Sci. USA 87: 8825–8829.

Yoneda, A., Akatsuka, M., Hoshino, H., Kumagai, F., and Hasezawa,

S. (2005). Decision of spindle poles and division plane by double

preprophase bands in a BY-2 cell line expressing GFP-tubulin. Plant

Cell Physiol. 46: 531–538.

Zhang, D.H., Wadsworth, P., and Hepler, P.K. (1990). Microtubule

dynamics in living dividing plant cells: Confocal imaging of micro-

injected fluorescent brain tubulin. Proc. Natl. Acad. Sci. USA 87:

8820–8824.

Zhu, C.J., Lau, E., Schwarzenbacher, R., Bossy-Wetzel, E., and Jiang,

W. (2006). Spatiotemporal control of spindle midzone formation by

PRC1 in human cells. Proc. Natl. Acad. Sci. USA 103: 6196–6201.

Phragmoplast Microtubule Organization 2923

DOI 10.1105/tpc.110.078204; originally published online August 26, 2011; 2011;23;2909-2923Plant Cell

Chin-Min Kimmy Ho, Takashi Hotta, Fengli Guo, Robert W. Roberson, Yuh-Ru Julie Lee and Bo LiuArabidopsisMicrotubule-Associated Protein MAP65-3 in

Interaction of Antiparallel Microtubules in the Phragmoplast Is Mediated by the

 This information is current as of April 10, 2019

 

Supplemental Data /content/suppl/2011/08/18/tpc.110.078204.DC2.html /content/suppl/2011/08/10/tpc.110.078204.DC1.html

References /content/23/8/2909.full.html#ref-list-1

This article cites 53 articles, 30 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