Characterization of a cytokinesis defective (cyd1) mutant of Arabidopsis
Journal of Experimental Botany
Characterization of a cytokinesis defective (cyd1) mutant of Arabidopsis
Ming Yang 0
Jeanette A. Nadeau 0
Liming Zhao 0
Fred D. Sack 0
0 Department of Plant Biology, Ohio State University , 1735 Neil Avenue, Columbus, OH 43210-1293 , USA
Although several mutations and genes affecting plant cytokinesis have been identified, mutant screens are not yet saturated and knowledge about gene function is still limited. A novel Arabidopsis mutation, cytokinesis defective1 (cyd1), was identified by partial or missing cell walls in stomata. Stomata with incomplete or no cytokinesis still differentiate and some contain swellings of the outer wall not found in the wild type. The incomplete walls are correctly placed opposite stomatal wall thickenings suggesting that the mutation interferes with the execution of cytokinesis rather than with the placement of the division site. Cytokinesis defects are also detectable in other cell types throughout the plant, defects which include cell wall protrusions, two or more nuclei in one cell, and reduced cell number. The extent of cytokinetic partitioning correlates with nuclear number in abnormal stomata. Many cyd1 epidermal cells, stomata and pollen are larger, and trichomes have more branches. cyd1 is partially lethal with poor seed set and some defective ovules, but many plants are fertile despite abnormalities in vegetative and reproductive development such as missing, reduced, fused or misshapen leaves and floral organs. cyd1 appears to be the only cytokinesis mutant described where defects are known to occur in both mature vegetative and reproductive organs. Thus, the CYD1 gene product appears to be necessary for the execution of cytokinesis throughout the shoot. The examination of stomata by microscopy may be a useful screen for the directed isolation of additional cytokinesis mutations that are not embryo or seedling lethal.
Cytokinesis; stomata; cell division
Cell division is a fundamental process involving division
of the nucleus (karyokinesis) and cytoplasm (cytokinesis).
This process functions at all levels of development, such
as in producing new cells for growth and reproduction,
and in allocating separate cell fates in asymmetric
Cytokinesis in plants encompasses events unique among
eukaryotes, including the functioning of a plant-specific
cytoskeletal array, the phragmoplast, and the centrifugal
development of the cell plate which forms the new cell
wall (Mineyuki and Gunning, 1990; Samuels et al., 1995;
Staehelin and Hepler, 1996; Assaad et al., 1997; Heese
et al., 1998 ). Stages of cytokinesis include the transport
and fusion of exocytic Golgi vesicles, the assembly and
subsequent fusion of a network of tubules, the outward
growth of the cell plate, and the fusion of the cell plate
with the parental cell wall at the division site.
A few proteins are known to be associated with
cytokinesis in plants ( Heese et al., 1998). Phragmoplastin, a
dynamin-like protein, has been shown to localize to the
cell plate and may be involved in vesicle traYcking (Gu
and Verma, 1996). A syntaxin-like protein was identified
by the knolle mutation in Arabidopsis in which cytokinesis
is defective perhaps through impaired vesicle fusion in
the nascent cell plate ( Lauber et al., 1997; Lukowitz et al.,
1996 ). Many other proteins must participate, but little is
known about the genetic and cellular regulation of
cytokinesis in plants.
Several plant mutations have been described that aVect
cytokinesis and are likely candidates for genes encoding
other components of the cytokinetic machinery. These
include the cyd (cytokinesis defective) mutant of pea, and
the knolle, keule and cyt1 mutants of Arabidopsis (Liu
et al. 1995; Lukowitz et al., 1996; Assaad et al., 1996;
Nickle and Meinke, 1998). These were identified indirectly
by screening for abnormalities in the morphology or
vitality of the embryo which proved to exhibit
perturbations in cytokinesis. In other cytokinesis mutations in
Arabidopsis the phenotype seems restricted to specific
domains of the mature plant. For example, the tso1
mutation aVects the inflorescence meristem but not
vegetative tissues ( Liu et al., 1997), and the multinucleic (mun)
mutations may aVect only root tissues ( Hauser et al.,
1997). Cytokinesis mutants in pollen development have
also been described in Arabidopsis (stud, tetraspore) and
alfalfa (McCoy and Smith, 1983; Hu¨ lskamp et al., 1997;
Spielman et al., 1997).
Here a new locus is reported that aVects cytokinesis in
Arabidopsis, cytokinesis defective1 (cyd1), that was
isolated in a microscopy-based screen for stomatal mutants
( Yang and Sack, 1995 ). Because the division of
Arabidopsis guard mother cells is stereotyped and
symmetric (Larkin et al., 1997 ), cytokinesis defects were readily
detected in stomata. Although cytokinesis is also
disrupted throughout the shoot and appears to result in
defective organogenesis, this mutation is not fully lethal.
Materials and methods
Plant material and culture
M2 seeds of Arabidopsis thaliana Columbia ecotype (ethyl
methane sulphonate mutagenesis; Lehle Seeds, Round Rock,
Texas, USA) were used for screening as in Yang and Sack
(1995). The seeds were also homozygous for the glabrous1 ( gl1)
mutation (Larkin et al., 1997). Seeds were sown on 1% nutrient
agar plus 2% sucrose. Seedlings were grown at 22 °C under
continuous 50–100 mmol m−2 s−1 white light.
To detect abnormalities in vegetative organ development,
cyd1 seeds were sown on agar and then grown under
four diVerent cultural conditions: (1) 30 mmol m−2s−1 light
intensity and 22 °C, (2) 100 mmol m−2 s−1 and 22 °C, (3)
30 mmol m−2 s−1 and 30 °C, and (4) 100 mmol m−2 s−1 and
30 °C. For each cultural condition the phenotypes of 108–122
seeds or seedlings were evaluated. Although there was a slight
influence of culture condition on abnormality frequency ( Yang,
1996), plants from all four conditions were pooled for collective
analysis. Class 3 and 4 cyd1 seedlings (see Results) were
transferred to a soil mix (peat, perlite, and vermiculite) for
further examination. These plants were maintained at their
original temperatures (22 °C or 30 °C ) with a light intensity of
50 mmol m−2 s−1 near the top of the pots. A total of 75–99
flowers from 20–22 plants for each of the four diVerent growing
conditions was examined under a dissecting microscope. For
crosses and mapping, etc, plants were grown in pots as above.
Genetic analysis and mapping
cyd1 was backcrossed to its wild-type (except for gl1) Columbia
parent twice and then to wild-type (GL1) Columbia for the
third backcross. For mapping, cyd1 (Col ) was crossed to
Landsberg erecta (Ler) and the segregating F2 plants were used
for linkage analysis. Genomic DNA was prepared from
individual F2 plants and SSLPs analysed (Dellaporta et al.,
1983; Bell and Ecker, 1994). Primer pairs were obtained from
Research Genetics ( Huntsville, AL, USA).
Quantitative characterization of phenotypes
The numbers and densities of stomata and non-stomatal
epidermal cells ( ECs) in the abaxial epidermis of mature
cotyledons from 18-d-old pot-grown plants were determined
( Yang and Sack, 1995). Guard cell length (along the ventral
wall ) and width, and stomatal pore length were measured from
the abaxial epidermis of cotyledons of 18–21-d-old agar-grown
plants, using an ocular reticule at a magnification of 400×.
Pollen size was determined as the product of the polar and
equatorial diameters (Altmann et al., 1994). Pollen grains from
newly opened flowers were measured within 10–20 min after
wetting, since dimensions change as the pollen hydrates (n=60
wild type and 120 cyd1 grains). All quantitative diVerences
reported were statistically significant at the 0.05 level (t-test).
Cells were observed with diVerential interference contrast and
fluorescence optics ( Zeiss IM35 microscope). Whole cotyledons
or roots and separated mesophyll cells were stained
with 10 mg ml−1 4∞,6-diamidino-2-phenylindole dihydrochloride
(DAPI; Melaragno et al., 1993). To determine nuclear number
in individual cells, mesophyll cells were separated according to
Pyke and Leech (Pyke and Leech, 1991). Ovules and embryos
were cleared as described ( Yadegari et al., 1994).
Roots of both normal and abnormal morphology were
examined for possible cell wall protrusions and for multiple or
abnormal nuclei using the propidium iodide method (van den
Berg et al., 1995). Plants were cultivated on sterile nutrient
agar for 10–15 d, and then immersed in 0.5 mg ml−1 propidium
iodide for 2–15 min. Whole roots were destained briefly in
water and examined using a Nikon Optiphot fluorescence
microscope attached to a BioRad MRC-600 confocal laser
scanning system. Over 3000 cells of the epidermis, cortex and
meristem were examined for each genotype (cyd1 and wild-type
Columbia) from many primary and lateral roots.
Material for light microscopy was also fixed in 3%
glutaraldehyde, 1.3% formaldehyde and 1% acrolein in 0.05 M phosphate
buVer at pH 6.8 for 10 h at room temperature. The tissue was
embedded in Spurr’s epoxy resin, sectioned at 1.5–4 mm, stained
with 0.05% toluidine blue at 58 °F, and photographed using
Kodak Ektachrome 100 film. For electron microscopy, leaf
pieces were fixed in 3% glutaraldehyde in 75 mM potassium/
sodium phosphate buVer, post-fixed in 2% osmium tetroxide,
dehydrated in an acetone series, and infiltrated with Spurr’s
epoxy resin. Sections were stained with uranyl acetate and lead
acetate and examined with a CM-12 Phillips electron microscope
at 60 kV.
Cytokinesis is incomplete or absent in many cyd1 stomata
The cyd1 mutant was identified by the presence of
abnormal stomata that were larger or that had incomplete
cytokinesis. Normally during stomatal development, the
guard mother cell divides symmetrically and the new cell
wall formed (ventral wall ) then develops a pore along its
mid-length. The cytokinesis defects found in cyd1 stomata
were grouped into categories for cytological analysis and
quantification ( Fig. 1). Stomata which lack any
cytokinesis defect were scored as ‘Type 1’ stomata ( Fig. 2A) and
these comprised 63% of all stomata sampled. Type 2
Fig. 1. The frequencies of occurrence of four diVerent stomatal types
in the abaxial epidermis of cyd1 cotyledons (n=20; 18-d-old,
Fig. 2. Abnormal guard mother cell cytokinesis in cyd1. Abaxial
epidermis from 15-d-old leaves. DiVerential interference contrast optics.
Bar in (A)=10 mm for all micrographs. (A) cyd1 Type 1 stomate with
normal cytokinesis and pore. (B) Type 2 stomate with defective
cytokinesis lacking any ventral wall or pore. Chloroplasts are typical
of normal stomata. (C ) Type 3 stomate with two wall protrusions at
opposite ends of the cell with no detectible pore. (D) Type 4 stomate
with incomplete wall that has a pore. (E, F ) Two diVerent planes of
focus of the same defective Type 2 stomate showing a swelling on the
outer tangential wall (arrow in F ).
stomata ( 21%) entirely lack a ventral wall and a pore
( Fig. 2B). Type 3 stomata (4%) have one or two ventral
wall protrusions at the ends of the cell, but the wall is
incomplete and no pore was detected using light
microscopy ( Fig. 2C ). Type 4 stomata (12%) have a pore that
is attached to one wall protrusion; if a second protrusion
is present on the other side, it does not extend to the
pore ( Fig. 2D). Overall, 37% of all stomata showed some
type of defective cytokinesis. While data for the frequency
of occurrence of cytokinesis defects in stomata were
obtained from examination of cotyledons, these defects
were also regularly observed in stomata located in all
organs of the plant, including rosette leaves, cauline leaves
and sepals. Abnormal stomata with incomplete
cytokinesis were never seen in wild-type plants.
Types 2 and 3 stomata develop an irregular papillate
thickening of the outer cell wall ( Fig. 2E, F ) that is often
elliptical to circular in the paradermal plane. In Type 2
stomata it is usually centred along the mid-length of the
cell where the outer ledges of the stomatal pore would
normally form. In Type 3 stomata, the swelling is
associated with an incomplete protrusion of the ventral wall,
closer to one end of the cell.
Incomplete walls are present in other cell types
To assess whether the defect was confined to the division
of guard mother cells, other cell types were examined for
incomplete cell walls. Wall protrusions were found in
mature epidermal cells ( ECs; Fig. 5). These protrusions
usually extended from only one side of the cell and were
of varying lengths. Examination of whole mounts
typically revealed about three ECs with incomplete walls per
cyd1 cotyledon. Protrusions were also seen in ECs in
rosette leaves, cauline leaves and sepals, but not in
Wall protrusions were also found in developing and
mature leaf mesophyll cells ( Fig. 3A, B) and in the ground
tissue of mature hypocotyls, stems and petioles ( Fig. 3C ).
Stubs could also be found in successive serial sections
through the shoot apical meristem ( Fig. 3D, E ) and more
commonly, in meristematic cells of leaf primordia
( Fig. 3F ). Thus, incomplete cytokinesis occurs
throughout the shoot, and some defects in cytokinesis that occur
early in development are not resolved during cell
Roots were also examined for cell wall stubs using
diVerential interference contrast microscopy of cleared
roots, by optical sectioning using confocal scanning laser
microscopy, and in longitudinal sections of fixed and
embedded tissue. Most cyd1 roots exhibited wild-type
tissue organization ( Fig. 3I ) and growth rates. No wall
protrusions were detected in any roots, even in those few
cyd1 roots that displayed irregular cell files, cell size and
architecture ( Fig. 3J ). These abnormal roots were always
found in plants exhibiting the most severe defects in
overall seedling organogenesis.
cyd1 has fewer and larger cells
Since defects in cytokinesis would be expected to aVect cell
number, the numbers of cells of diVerent types in the
epidermis were quantified in fully expanded cotyledons.
cyd1 cotyledons have fewer cells per unit area and reduced
absolute numbers of stomata and epidermal cells ( Fig. 4).
The reduction in cell number is much greater than the
number of cells containing wall protrusions. The majority
of cyd1 ECs are slightly larger than the wild-type, but since
there are fewer cells, cyd1 leaf area is somewhat reduced.
cyd1 guard cells and stomatal pores are, on average,
147% and 156% longer than those of the wild-type. cyd1
stomata also have a much broader range of sizes than
the wild type ( Yang, 1996 ). The ratio of ECs to stomata
is 2.6±0.1 (±SE ) for the wild-type and 3.6±0.2 for cyd1.
cyd1 pollen grains are about 25% larger than those of the
wild type ( 754±15 versus 608±7 mm2) and also exhibit
a greater range of sizes.
Fig. 4. Number and density of stomata and epidermal cells in mature
wild-type and cyd1 cotyledons. The cyd1 abaxial epidermis contains
fewer stomata (of all types) and epidermal cells than the wild type on
both an absolute number and density basis. Same plants as in Fig. 1.
The cyd1 mutation was crossed into a GL1 (Columbia)
background so that the eVect on trichomes could be
assessed. The number of trichome branches in cyd1 was
higher than in the wild-type in the adaxial epidermis of
the first leaf. Most ( 96%) wild-type trichomes had three
branches, whereas only 3% and 1% had four and two
branches, respectively, and none had five branches. In
contrast, 65% of cyd1 trichomes had four branches, 26%
had three, 1% had two, and 8% had five branches. Thus,
cyd1 has fewer and larger epidermal cells and stomata,
larger pollen, and trichomes with more branches than the
Two or more nuclei in the same cell
Nuclear number and morphology were studied using
DAPI staining in the cotyledon epidermis, in isolated leaf
mesophyll cells, and in roots. While most cells lack cell
wall protrusions and contain only one nucleus per cell,
each cotyledon usually contains a few ECs without wall
protrusions that had two or three nuclei per cell. ECs
with cell wall protrusions had either one nucleus, which
was frequently abnormally large, or two nuclei. When
two nuclei were present, their position in the cell was
variable. Nuclei were often found in contact, or connected
by irregular extensions of DAPI-staining material
( Fig. 5A, B). Double nuclei were also observed in some
abnormal stomata ( Fig. 5C, D) and in some mesophyll
cells, but not in cortical or epidermal cells of cyd1 roots
or in any wild-type cells.
Examination of sectioned tissues confirmed the
presence of multiple nuclei in some leaf epidermal and
mesophyll cells as well as in cells of developing leaf primordia
Fig. 5. Nuclei in cyd1 cells. Fluorescence micrographs (A, C ) of
DAPIstained abaxial epidermis of 14–15-d-old whole-mounted cotyledons
from agar-grown plants. Corresponding bright-field images of same
regions shown in (B) and (D). Bar in (A)=10 mm for all micrographs.
(A, B) Elongate, two-lobed nucleus (arrows) in a single epidermal cell
with an undulate cell wall protrusion (arrowhead in B). (C, D) Two
nuclei (arrows) in a single-celled, Type 2 stomate.
( Fig. 3F ). Multiple nuclei were not observed in cells of
sectioned root tissue.
Nuclear number correlates with the extent of cytokinetic
partitioning in stomata
Because division of the guard mother cell is stereotyped
and even stomata without a pore or wall protrusions are
easily recognized, a complete failure of cytokinesis is
readily apparent. It is more diYcult to identify an EC
which failed to undergo cytokinesis because the plane of
EC division is not predictable and EC size is variable.
Thus, a possible relationship between defective cytokinesis
and nuclear number was evaluated quantitatively in
stomata ( Fig. 6). All stomata of normal morphology
( Type 1) contained one nucleus per guard cell. In contrast
all Type 4 stomata contained two nuclei. Types 2 and 3
stomata were intermediate in nuclear number. These data
indicate that the binucleate phenotype is positively
correlated with the extent of cell partitioning, so that the greater
Fig. 6. Percentage of Types 2–4 cyd1 stomata with one or two nuclei
obtained using DAPI fluorescence of cotyledons. The proportion of
cells containing two nuclei increases with greater cell partitioning,
e.g. all Type 4 stomata had two nuclei; n=25–70.
the degree of ventral wall formation in cyd1 stomata with
incomplete cytokinesis, the higher the percentage of
cyd1 wall protrusions are correctly positioned in stomata
The ultrastructure of mature Type 1 cyd1 stomata is
roughly comparable to those of the wild-type, for
example, in chloroplast diVerentiation and positioning
( Fig. 7A). In cyd1 stomata with defective cytokinesis,
electron microscopy confirms the presence of two nuclei
in some single cells, and of nuclei with abnormal
morphology and position ( Fig. 7B, C ).
Electron microscopy also reveals that some short wall
protrusions form small pore openings, although only one
pore has been found per cell, even in cells with two wall
stubs ( Fig. 7C ). cyd1 wall protrusions are composed of
material that closely resembles that of the parent wall in
staining intensity and texture ( Fig. 7B, C ). These
protrusions appear to be located where the ventral wall would
normally attach to the parent wall during division of the
guard mother cell. In wild-type Arabidopsis guard mother
cells, the division site becomes thickened at opposing
ends of the cell prior to division, and this thickening can
still be distinguished once cytokinesis is complete ( Zhao
and Sack, 1999 ). In cyd1 stomata, protrusions are also
associated with wall thickenings, and when two
protrusions are present, the underlying wall thickenings are
located opposite each other as in the wild-type. cyd1
stomatal wall stubs never appear to be associated with
thinner portions of the wall. Thus, the placement of the
new cell wall appears normal in cyd1 stomata with
cyd1 is a nuclear recessive and partially lethal mutation
cyd1 was backcrossed to wild-type plants for genetic
analysis. cyd1 is a recessive mutation since all F1 plants
(n=17) from two independent crosses had a wild-type
cellular phenotype. After selfing of F1 plants, F2 plants
(n=247) displayed a segregation ratio (wild type:cyd1)
of 4.451 instead of 351, suggesting that cyd1 is a partially
lethal nuclear mutation. Similar segregation ratios were
obtained regardless of whether cyd1 was the pollen donor
or acceptor in the original backcross. The cyd1 phenotype
was stable after three successive backcrosses. The
cyd1 gene was mapped to the interval between nga225
and nga249 in the top arm of chromosome 5 using simple
sequence length polymorphisms (SSLPs), a position that
distinguishes it from other mapped cytokinesis mutants
(Lukowitz et al., 1996; Assaad et al., 1996; Liu et al.,
1997; H u¨lskamp et al., 1997; Nickle and Meinke, 1998 ).
Abnormalities in cyd1 vegetative development
To evaluate possible eVects of the cyd1 mutation on
organogenesis, individual plants were analysed
throughFig. 7. Transmission electron micrographs of cyd1 stomata. Wall
thickenings (arrowheads) underlie the point of attachment of the ventral
wall in both the wild-type and in cyd1 stomata: n, nucleus; p, stomatal
pore; s, wall stub. Bars=2 mm. (A).cyd1 stomate with complete
cytokinesis. (B) Binucleate, Type 4 stomate with incomplete cytokinesis.
One wall protrusion is present and it has developed a pore. (C ) Stomate
with two wall protrusions, one of which contains a small pore. Either
two nuclei or two lobes of the same nucleus are present to the left of
the pore, and are abnormally located at the end of the stomate.
out their development, first as seedlings grown on agar
for 8–10 d and then after transfer from agar to pots. All
cyd1 plants showed at least some alterations in vegetative
growth relative to wild-type seedlings ( Fig. 8A–E ), but
the severity of the phenotype was highly variable. The
same range of abnormalities was seen under four diVerent
cultural conditions (see Materials and methods). Of 447
seeds sown, about 25% did not germinate. 19% of all
seeds germinated and developed four or fewer abnormal
leaves and an abnormal root, and then the development
of these seedlings arrested. For example, some of these
seedlings had only one cotyledon and no apparent root,
or they had only limited root development ( Fig. 8B). 20%
of all seeds developed into mature, fertile plants, but
contained gross deformities such as missing, reduced,
fused or abnormally lobed cotyledons or leaves, or the
production of unidentifiable, teratological structures. The
remaining 36% of seeds produced mature plants that at
a gross level appeared normal. But almost all of these
plants contained abnormalities such as unequal sizes of
young leaves especially those of the first pair, or leaves
that were irregular and asymmetric in shape with larger
marginal teeth ( Fig. 8C, E ).
All cyd1 leaves, regardless of the severity of the overall
Fig. 8. Vegetative (B–E) and reproductive (F–H ) abnormalities in cyd1
plants. Plants were grown on agar (A–C ) or in pots (D–H ) for 8–10 d.
(A) 10-d-old wild-type seedling. Bar=10 mm. (B) Abnormal cyd1
seedling with a single fan-shaped organ, perhaps a cotyledon, and a
root. Bar=3 mm. (C ) Leaf lobing or possible fusion between two leaves
(arrow). Bar=5 mm. (D) Clearings of cyd1 (left) and wild-type (right)
cotyledons illustrating wavy venation and irregular margin in cyd1.
Bar=1 mm. (E) Leaves of cyd1 ( left) and wild-type (right) showing
more pronounced dentation in cyd1. Bar=1 cm. (F ) Two pairs of
laterally fused stamens from the same flower with diVerent degrees of
fusion in each pair. Bar=0.4 mm. (G) A stamen fused with a carpel
along the entire length of the filament. Bar=1 mm. (H ) Abnormal
cyd1 ovule with tracheids (arrow) that developed in place of the
gametophyte. Bar=30 mm.
plant phenotype, exhibited a roughened and undulate
blade. cyd1 cotyledons and leaves have wavy and distorted
veins, but the overall pattern is comparable to the wild
type ( Fig. 8D). The pattern of epidermal cell expansion
is also disrupted so that the long axis of many epidermal
cells parallels the veins. However, the overall anatomical
organization and diVerentiation of tissue layers in cyd1
leaves is comparable to the wild type ( Fig. 3G, H ).
Embryo morphology was evaluated in cleared green
seeds. A range of phenotypes was found in cyd1 seeds
from the same silique, from embyros that exhibited severe
defects in organization to those that were normal
morphologically. Defects appeared as early as the pre-globular
stage. Thus, cyd1 plants often have missing, reduced,
fused or abnormally-shaped leaves, cotyledons, and roots,
defects which are rarely seen in wild-type plants.
Abnormalities in cyd1 reproductive development
A sample of 354 flowers from 84 cyd1 plants were
classified and scored for defective development and 55%
of all flowers contained abnormalities. The most frequent
defects were, in order, (1) flowers with five (rather than
six) stamens, (2) lateral fusion of two sepals, (3 ) fusion
of a stamen and a carpel ( Fig. 8G ), (4) fusion of a
stamen and a petal, ( 5) fusion of two stamens ( Fig. 8F ),
and (6) flowers with one reduced and five normal stamens.
Other types of abnormalities were found at much lower
frequencies such as fusion between more than two organs
in a whorl including between three or four stamens or
between three sepals. Stamens were the organ most likely
to show a reduction in number or size, but the minimum
number of stamens observed was four. Thus, the most
frequent defects in cyd1 floral organogenesis were a loss
of an organ, fusion of organs within a whorl and fusion
of organs in adjacent whorls.
In addition, each carpel contained at least some
defective ovules in cyd1 plants. Abnormalities included ovules
with altered patterns of cell division, and mature ovules
that were distorted and composed of irregular cell files.
Ovule morphology ranged from wild-type to severely
distorted. In some cases tracheids developed where a
gametophyte would normally be found ( Fig. 8H ).
The seed set of self-fertilized cyd1 plants is poor, for
example, in one experiment, cyd1 siliques had only
11.2±1.1 (mean±SE ) relatively normal-looking seeds,
compared to 34±0.8 in the wild type. Unlike the wild
type, cyd1 siliques also contained some collapsed seeds
(4.0±0.6 seeds per silique). Seed set was also poor when
cyd1 plants were used as the female parent in backcrosses
with wild-type plants although it was better when cyd1
was the pollen donor. This suggests that ovule or female
gametophyte production in cyd1 is sometimes defective.
Thus, almost all reproductive and vegetative organs of
cyd1 showed morphological abnormalities, but there was
great variation in which organ was aVected and how it
was aVected between individual plants.
Stomatal division site placement and differentiation are
The cytokinesis defective1 mutant of Arabidopsis identifies
a novel locus that appears to be necessary for the
execution of cytokinesis and for the normal development of
the vegetative and reproductive organs of the shoot.
cyd1 and other plant cytokinesis mutations
cyd1 shares some of the cellular defects described in other
cytokinesis-defective mutants in plants such as the
presence of several nuclei in one cell, larger and fewer cells,
and incomplete cell walls (Assaad et al., 1996; Kitada
et al., 1983; Liu et al., 1995, 1997; Lukowitz et al., 1996;
Hauser et al., 1997; Nickle and Meinke, 1998). cyd1 is
also similar to many of these mutations in that defective
cytokinesis is correlated with the abnormal development
of embryos and organs.
However, cyd1 maps to a position that diVers from
that of published cytokinesis mutations in Arabidopsis,
i.e. keule, knolle, tso1, stud, tetraspore, and cyt1 ( Assaad
et al., 1996; Lukowitz et al., 1996; Liu et al., 1997;
Hu¨ lskamp et al., 1997; Spielman et al., 1997; Nickle and
The cyd1 phenotype also diVers in cytology and/or in
the locations of the cytokinesis defects. Most other
cytokinesis mutants have more nuclei and wall protrusions in
one cell than cyd1 and may have branched protrusions
(Liu et al., 1995) which are absent in cyd1. Similarly,
cyd1 lacks the irregular wall thickenings found in cyt1
(Nickle and Meinke, 1998). In cyd1, wall protrusions are
present both in meristematic and in mature vacuolated
cells. In contrast, in other mutants, incomplete walls have
been found only in embryonic, meristematic and dividing
cells (Assaad, et al., 1996; Lukowitz, et al., 1996; Nickle
and Meinke, 1998), or only in mature vacuolated cells
(Liu et al., 1995), but not in both categories of cells.
Finally, cyd1 is the only plant cytokinesis mutant
described where defects are known to occur throughout
the vegetative and reproductive shoot. In some mutants
the defect is confined to a specific set of organs or cell
types such as flowers (tso1), roots (mun), or to specific
stages of pollen development, for example, stud and
tetraspore (McCoy and Smith, 1983; Hauser et al., 1997;
Hu¨ lskamp et al., 1997; Liu et al., 1997; Spielman et al.,
1997). Other mutants are either embryo or seedling lethal
(cyt1, knolle, keule) or the embryos can be rescued in
culture but do not progress to maturity (pea cyd; Liu
et al., 1995; Assaad et al., 1996; Lukowitz et al., 1996;
Nickle and Meinke, 1998).
The CYD1 gene product appears to be necessary for the
execution of cytokinesis rather than for its placement, at
least in stomata. In Arabidopsis guard mother cells, the
division site—the eventual location of fusion of the cell
plate with the parental cell wall—is marked by wall
thickenings at opposite ends of the cell ( Zhao and Sack,
1999 ). Incomplete walls in cyd1 stomata are attached to
division site thickenings indicating that the site of cell
plate fusion is normal. This contrasts with the fass or ton
mutant in which the placement rather than the execution
of cytokinesis is defective ( Torres-Ruiz and J u¨rgens, 1994;
Traas et al., 1995).
Cytokinesis defects in stomata have been described in
wild-type plants that were treated with cytoskeletal
inhibitors and with caVeine (Galatis, 1977; Galatis and
Apostolakos, 1991; Terryn et al., 1993 ). However, cyd1
appears to be the first mutation documented to display
Data from both cyd1 and from inhibitor treatment
show that aspects of stomatal diVerentiation do not
require cytokinesis. For example, stomata without any
dividing wall can still be recognized by cell shape and
chloroplast diVerentiation. Pore diVerentiation requires
the presence of at least a part of a ventral wall because
cyd1 stomata without any dividing wall do not form a
pore. In cells with short stubs, the pore forms at the end
of the stub, suggesting that the site of pore targeting is
correct, i.e. as close as possible to the middle of the cell.
Pore number also seems correctly regulated, because even
cells with two stubs have only one pore. Thus, pore
diVerentiation requires the presence of some new cell
wall, but not full cytokinesis, a conclusion reinforced by
the observation that pores form naturally on incomplete
walls in wild-type stomata of several genera (Sack, 1987).
cyd1 stomata with short stubs or with no ventral wall
have an abnormal, papillate swelling on the outer wall.
Because the location of this swelling might be a default
target for the fusion of exocytic Golgi vesicles, further
analysis could provide an insight into the mechanisms
that regulate the site of vesicle targeting.
Nature of cytokinesis defect
The CYD1 gene product appears necessary for the
execution of cytokinesis. CaVeine, which also causes wall stubs,
may guide vesicle fusion to the cell plate, or block plate
stabilization ( Hepler and Bonsignore, 1990; Liu et al.,
1995; Valets and Hepler, 1997). The presence of wall
protrusions, rather than central ‘islands’, in both
cytokinesis mutants and in caVeine-treated plants could imply
that cell plates are more stable where they join the parent
wall than in the cell centre. Evidence suggests that ‘wall
maturation and insertion factors’ originate from the
parent wall and spread into the new wall to stabilize it
(Mineyuki and Gunning, 1990). Thus, the CYD1 gene
product might function in stabilizing the first-formed
regions of the cell plate.
Other aspects of the cyd1 phenotype suggest that
cellular processes in addition to cytokinesis might be aVected.
For example, the finding that the extent of cytokinetic
partitioning in stomata correlates with nuclear number
could indicate that nuclear division, or post-divisional
nuclear reassembly or fusion are aVected as well. Thus,
it cannot be ruled out that CYD1 functions only indirectly
in cytokinesis and directly in mitosis (e.g. as in Mackay
et al., 1998 ) or in karyokinesis. Obviously, knowledge of
the identity and the sub-cellular localization of the CYD1
gene product could prove helpful in analysing gene
Increase in cell size and trichome branching
Larger stomata, pollen and epidermal cells are common
indicators of polyploidy, and trichome branch number
correlates with the degree of endoreduplication in
Arabidopsis (Altmann et al., 1994; Hu¨ lskamp et al., 1994;
Melaragno et al., 1993 ). While cyd1 exhibits these
phenotypes, cyd1 is not simply a polyploid plant. In 4n and 6n
Arabidopsis, all stomata and pollen are larger, and all
stomata have a normal morphology, whereas in cyd1,
almost half of stomata and pollen grains are the same
size as in the wild type ( Yang, 1996). Moreover, cyd1
appears to be a recessive, single-locus mutation.
Larger cells might result from selected progenitor cells
becoming polyploid. Alternatively, increased size could
be a consequence of slower or fewer cell cycles. For
example, overexpression of a dominant negative mutant
form of Cdc2 kinase, a cell cycle regulator, slowed cell
division and produced larger cells without increasing
DNA content ( Hemerly et al., 1995).
Cytokinesis and development
Because other cytokinesis mutants are embryo/seedling
lethal or phenotypes only occur in specific organs, the
cyd1 mutation provides a novel opportunity to analyse
the relationships between cytokinesis defects and shoot
organogenesis. Although this cyd1 allele is fully penetrant,
the extent to which diVerent cyd1 plants exhibit organ
abnormalities is variable. This could be explained if the
cytokinesis defect aVected all cells randomly and if the
consequences of a defect depended on how essential that
cell was for development. A lack of division in
strategically located cells in primordia and during organogenesis
could produce the organ abnormalities described. The
observation that cytokinesis defects can be observed in
developing primordia and meristems supports this
hypothesis. Defects during embryonic development may be of
more consequence and more deleterious than those that
occur after germination.
Altered leaf morphogenesis in cyd1 might be related to
reduced cell number. Cell deficits may restrict local
expansion and cause diVerential stress–strain relationships
resulting in vascular distortions, exaggerated dentation, a
roughened leaf surface, and leaf asymmetry. Although
veins are wavy, the anatomy of cyd1 leaves is essentially
normal. This contrasts with the knolle and keule mutants
where non-epidermal cells diVerentiate in the embryonic
epidermis (Assaad et al., 1996; Lukowitz et al., 1996).
The cyd1 mutant of Arabidopsis identifies a new locus
aVecting cytokinesis. A fraction of cells throughout the
shoot exhibits phenotypes shared by other cytokinesis
mutants in plants such as multiple nuclei, wall
protrusions, and larger and fewer cells. Cytokinesis defects in
cyd1 stomata were readily detected because guard mother
cell divisions are stereotyped and because diVerentiation
continues in the absence of cytokinesis. Since many cyd1
plants are fertile and produce seeds, this mutant provides
the opportunity to evaluate the relationships between a
cytokinesis defect and organ development. Based on
phenotype, the CYD1 gene product would be expected to
encode a novel component of the cytokinetic machinery.
This research was supported, in part, by NSF grant
IBN-9505687. Technical assistance from Connie Lee and Rachel
Heath are gratefully acknowledged.
Altmann T , Damm B , Frommer WB , Martin T , Morris PC , Scheizer D , Willmitzer L , Schmidt R . 1994 . Easy determination of ploidy level in Arabidopsis thaliana, plants by means of pollen size measurement . Plant Cell Reports 13 , 652 - 656 .
Assaad FF , Mayer U , Lukowitz W , Ju¨rgens G. 1997 . Cytokinesis in somatic plant cells . Plant Physiology and Biochemistry 35 , 177 - 184 .
Assaad FF , Mayer U , Wanner G , Ju¨rgens G. 1996 . The KEULE gene is involved in cytokinesis in Arabidopsis . Molecular and General Genetics 253 , 267 - 277 .
Bell CJ , Ecker JR . 1994 . Assignment of 30 microsatellite loci to the linkage map of Arabidopsis . Genomics 19 , 137 - 144 .
Dellaporta SL , Wood J , Hicks JB . 1983 . A plant DNA minipreparation: version II . Plant Molecular Biology Reporter 1 , 19 - 21 .
Galatis B. 1977 . DiVerentiation of stomatal meristemoids and guard mother cells into guard-like cells in Vigna sinensis leaves after colchicine treatment . Planta 136 , 103 - 114 .
Galatis B , Apostolakos P. 1991 . Microtubule organization and morphogenesis of stomata in caVeine-aVected seedlings of Zea mays . Protoplasma 165 , 11 - 26 .
Gu X , Verma DPS . 1996 . Phragmoplastin, a dynamin-like protein associated with cell plate formation in plants . EMBO Journal 15 , 695 - 704 .
Hauser M-T , Dorner M , Fuchs E , Ovecka M , Baluska F , Benfey P , Gloessl J. 1997 . Genetic evidence for postembryonic and organ-specific regulation of cytokinesis in roots of Arabidopsis thaliana . 8th International Meeting for Arabidopsis Research, Abstract 46.
Heese M , Mayer U , Ju¨rgens G. 1998 . Cytokinesis in flowering plants: cellular process and developmental integration . Current Biology 1 , 486 - 491 .
Hemerly A , de Almeida Engler J , Bergounioux C , Van Montagu M , Engler G , Inze´ D , Ferriera P. 1995 . Dominant negative mutants of the Cdc2 kinase uncouple cell division from iterative plant development . EMBO Journal 14 , 3925 - 3936 .
Hepler PK , Bonsignore CL . 1990 . CaVeine inhibition of cytokinesis: ultrastructure of cell plate formation/degradation . Protoplasma 157 , 182 - 192 .
Hu¨lskamp M , Mise´ra S, Ju¨rgens G. 1994 . Genetic dissection of trichome cell development in Arabidopsis . Cell 76 , 555 - 566 .
Hu¨lskamp M , Parehk NS , Grini P , Schneitz K , Zimmerman I , Lolle SJ , Pruitt RE . 1997 . The STUD gene is required for male-specific cytokinesis after Telophase II of meiosis in Arabidopsis thaliana . Developmental Biology 187 , 114 - 124 .
Kitada K , Kurata N , Satoh H , Omura H. 1983 . Genetic control of meiosis in rice , Oryza sativa L. I. Classification of meiotic mutants induced by MNU and their cytogenetical characteristics . Japanese Journal of Genetics 58 , 231 - 240 .
Larkin JC , Marks MD , Nadeau J , Sack FD . 1997 . Epidermal cell fate and patterning in leaves . The Plant Cell 9 , 1109 - 1120 .
Lauber MH , Waizenegger I , Steinmann T , Schwarz H , Mayer U , Hwang I , Lukowitz W , Ju¨rgens G. 1997 . The Arabidopsis KNOLLE protein is a cytokinesis-specific syntaxin . Journal of Cell Biology 139 , 1485 - 1493 .
Liu C-M , Johnson S , Wang TL . 1995 . cyd, a mutant of pea that alters embryo morphology is defective in cytokinesis . Developmental Genetics 16 , 321 - 331 .
Liu Z , Running MP , Meyerowitz EM . 1997 . TSO1 functions in cell division during Arabidopsis flower development . Development 124 , 665 - 672 .
Lukowitz W , Mayer U , Ju¨rgens G. 1996 . Cytokinesis in the Arabidopsis embryo involves the syntaxin-related KNOLLE gene product . Cell 84 , 61 - 71 .
Mackay AM , Ainsztein AM , Eckley DM , Earnshaw WC . 1998 . A dominant mutant of the inner centromere protein ( INCENP), a chromosome protein, disrupts prometaphase congression and cytokinesis . Journal of Cell Biology 140 , 991 - 1002 .
McCoy TJ , Smith LY . 1983 . Genetics, cytology, and crossing behavior of an alfalfa (Medicago sativa) mutant resulting in failure of the post-meiotic cytokinesis . Canadian Journal of Genetics and Cytology 25 , 390 - 397 .
Mineyuki Y , Gunning BES . 1990 . A role for preprophase bands of microtubules in maturation of new cell walls, and a general proposal on the function of preprophase band sites in cell division in higher plants . Journal of Cell Science 97 , 527 - 537 .
Melaragno JE , Mehrotra BM , Coleman AW . 1993 . Relationship between endopolyploidy and cell size in epidermal tissue of Arabidopsis . The Plant Cell 5, 1661 - 1668 .
Nickle TC , Meinke DW . 1998 . A cytokinesis-defective mutant of Arabidopsis (cyt1) characterized by embryonic lethality, incomplete cell walls, and excessive callose accumulation . The Plant Journal 15 , 321 - 332 .
Pyke KA , Leech RM . 1991 . Rapid image analysis screening procedure for identifying chloroplast number mutants in mesophyll cells of Arabidopsis thaliana (L.) Heynh. Plant Physiology 96 , 1193 - 1195 .
Sack FD . 1987 . The development and structure of stomata . In: Zeiger E, Farquhar GD , Cowan IR, eds. Stomatal function. Stanford: Stanford University Press, 59 - 89 .
Samuels AL , Giddings TH Jr., Staehelin LA . 1995 . Cytokinesis in tobacco BY-2 and root tip cells: a new model of cell plate formation in higher plants . Journal of Cell Biology 130 , 1345 - 1357 .
Spielman M , Preuss D , Li F-L , Browne WE , Scott RJ , Dickinson HG . 1997 . TETRASPORE is required for male cytokinesis in Arabidopsis thaliana . Development 124 , 2645 - 2657 .
Staehelin LA , Hepler PK . 1996 . Cytokinesis in higher plants . Cell 84 , 821 - 824 .
Terryn N , Arias MB , Engler G , Tire ´ C, Villarroel R , Van Montagu M , Inze´ D. 1993 . rha1, a gene encoding a small GTP binding protein from Arabidopsis, is expressed primarily in developing guard cells . The Plant Cell 5 , 1761 - 1769 .
Torres-Ruiz RA , Ju¨rgens G. 1994 . Mutations in the FASS gene uncouple pattern formation and morphogenesis in Arabidopsis development . Development 120 , 2967 - 2978 .
Traas J , Bellini C , Nacry P , Kronenberger J , Bouchez D , Caboche M. 1995 . Normal diVerentiation patterns in plants lacking microtubular preprophase bands . Nature 375 , 676 - 677 .
Valets AHD , Hepler PK . 1997 . CaVeine inhibition of cytokinesis: eVect on the phragmoplast cytoskeleton in living Tradescantia stamen hair cells . Protoplasma 196 , 155 - 166 .
van den Berg C , Willemsen V , Hage W , Weisbeek P , Scheres B. 1995 . Cell fate in the Arabidopsis root meristem determined by directional signaling . Nature 378 , 62 - 65 .
Yadegari R , de Paiva GR , Laux T , Koltunow AM , Apuya N , Zimmerman JL , Fischer RL , Harada JJ , Goldberg RB . 1994 . Cell diVerentiation and morphogenesis are uncoupled in Arabidopsis embryos . The Plant Cell 6 , 1713 - 1729 .
Yang M. 1996 . Isolation and characterization of stomatal and cytokinesis defective mutants in Arabidopsis . PhD dissertation , Ohio State University, Columbus, Ohio, USA.
Yang M , Sack FD . 1995 . The too many mouths and four lips mutations aVect stomatal production in Arabidopsis . The Plant Cell 7, 2227 - 2239 .
Zhao L , Sack FD . 1999 . Ultrastructure of stomatal development in Arabidopsis leaves . American Journal of Botany 86 , 929 - 939 .