Journal of Experimental Botany, Vol. 59, No. 6, pp. 1419–1430, 2008
doi:10.1093/jxb/ern055 Advance Access publication 28 March, 2008
This paper is available online free of all access charges (see http://jxb.oxfordjournals.org/open_access.html for further details)
RESEARCH PAPER
Difference in light-induced increase in ploidy level and cell
size between adaxial and abaxial epidermal pavement cells
of Phaseolus vulgaris primary leaves
Isao Kinoshita1,*, Akiko Sanbe2,† and E-iti Yokomura2,‡
1
Department of Molecular and Cell Biology, Forestry and Forest Products Research Institute, Matsunosato,
Tsukuba, Ibaraki 305-8687, Japan
2
Department of Biology, Faculty of Science, Nara Women’s University, Nara 630-8506, Japan
Received 26 October 2007; Revised 1 February 2008; Accepted 4 February 2008
Abstract
Changes in nuclear DNA content and cell size of
adaxial and abaxial epidermal pavement cells were
investigated using bright light-induced leaf expansion
of Phaseolus vulgaris plants. In primary leaves of bean
plants grown under high (sunlight) or moderate (ML;
photon flux density, 163 mmol m22 s21) light, most
adaxial epidermal pavement cells had a nucleus with
the 4C amount of DNA, whereas most abaxial pavement cells had a 2C nucleus. In contrast, plants grown
under low intensity white light (LL; 15 mmol m22 s21)
for 13 d, when cell proliferation of epidermal pavement
cells had already finished, had a 2C nuclear DNA
content in most adaxial pavement cells. When these
LL-grown plants were transferred to ML, the increase
in irradiance raised the frequency of 4C nuclei in
adaxial but not in abaxial pavement cells within 4 d.
On the other hand, the size of abaxial pavement cells
increased by 53% within 4 d of transfer to ML and
remained unchanged thereafter, whereas adaxial pavement cells continuously enlarged for 12 d. This
suggests that the increase in adaxial cell size after
4 d is supported by the nuclear DNA doubling. The
different responses between adaxial and abaxial epidermal cells were not induced by the different light
intensity at both surfaces. It was shown that adaxial
epidermal cells have a different property than abaxial
ones.
Key words: Cell enlargement, endopolyploidization, epidermal
pavement cells, incident light intensity, leaf expansion, nuclear
DNA content, Phaseolus vulgaris.
Introduction
The adaxial (upper) side and abaxial (lower) side of
a dicotyledonous leaf are morphologically asymmetrical.
Corresponding with this asymmetry, the upper side of the
leaf is specialized for the efficient capture of sunlight,
whereas the lower side is specialized for gas exchange.
Previously, it was shown that adaxial and abaxial
epidermal pavement cells of bean (Phaseolus vulgaris L.)
primary leaves differ from each other in nuclear DNA
content (Kinoshita et al., 1991). In primary leaves of bean
plants grown in moderate intensity (163 lmol m2 s1)
white light, most adaxial epidermal pavement cells had
a nucleus with the 4C amount of DNA, whereas most
abaxial epidermal pavement cells had a nucleus with the
2C amount of DNA. In the present study, the influence of
incident light intensity on the DNA content of adaxial and
abaxial epidermal pavement cells and mesophyll cells was
investigated.
The Phaseolus primary leaf is useful material to
investigate leaf cell enlargement (Dale, 1988; Van
Volkenburgh, 1999). Both cell division and cell enlargement contribute to leaf growth. Cell proliferation occurs in
the early phase of dicotyledonous leaf development, and
then the contribution of cell enlargement increases in the
* To whom correspondence should be addressed. E-mail: [email protected]
y
Present address: Development Laboratory 5, Lion Corporation, 13-12 Hirai 7-chome, Edogawa-ku, Tokyo 132-0035, Japan.
z
Deceased in June 2006.
ª 2008 The Author(s).
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which
permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
1420 Kinoshita et al.
late phase (Van Volkenburgh and Cleland, 1979; Beemster et al., 2005). Phaseolus plants grown in low intensity
red light (4 lmol m2 s1) for 10 d (cell proliferation in
the primary leaves had already completed) showed greatly
elongated stems, but the primary leaves, although green,
remained unexpanded (Dale, 1988). Bringing these primary leaves into white light (250–400 lmol m2 s1) led
to rapid leaf enlargement without cell division (Van
Volkenburgh and Cleland, 1979, 1980). Using this
system, Van Volkenburgh et al. (1990) showed that
maximum expansion of bean leaves required at least 100
lmol m2 s1 white light. Light-stimulated growth of
green leaves has the typical characteristics of a high
irradiance response. This leaf enlargement (cell enlargement) is regulated by non-photosynthetic photosystems
including phytochrome (Van Volkenburgh and Cleland,
1990; Van Volkenburgh et al., 1990). When light
stimulates leaf enlargement, the proton pump in the
plasma membrane is stimulated and, as a result of this
event, proton efflux from the cells increases (Stahlberg
and Van Volkenburgh, 1999). Acidification of the
apoplast loosens the cell wall, making it more responsive
to stress imposed by turgor (Rayle and Cleland, 1992).
This is the early response in light-stimulated leaf cell
enlargement. However, it has not been investigated
whether nuclear DNA content in epidermal and mesophyll
cells changes under these conditions. In the present study,
these changes were investigated using primary leaves of
bean plants grown under a condition similar to that
described by Van Volkenburgh and Cleland (1979), but
low intensity white light was used instead of low intensity
red light. Bean plants were grown under the low light
condition (LL, 15 lmol m2 s1). At 10–13 d, when
cell proliferation in the primary leaves had finished, the
plants were transferred to the moderate light condition
(ML, 163 lmol m2 s1).
It has been known since the early 20th century that
a higher ploidy nucleus is often associated with an increase
in cell size (Sugimoto-Shirasu and Roberts, 2003). Increase
in cell size is often paralleled by an increase in ploidy level
brought about by endoreduplication, which is successive
rounds of DNA replication without mitosis (Gendreau
et al., 1997, 1998). Although many studies have been done
on the molecular mechanism that induces endoreduplication
(Grafi and Larkins, 1995; Gendreau et al., 1997, 1998;
Larkins et al., 2001; Imai et al., 2006; Yoshizumi et al.,
2006), the biological significance of endopolyploidization
is not well understood.
Many environmental and internal signals (e.g. light,
temperature, plant growth regulators) induce endoreduplication in a variety of plant cells (Nagl, 1978). In leaf
epidermal cells, environmental factors also affect the
nuclear DNA content. Schlayer (1971) showed that the
DNA content per cell increased 2–4-fold in epidermal
cells of sugar beet cotyledons on applying stronger light
or adding more moisture to the soil. Griffiths et al. (1994)
showed that high temperature reduced the endopolyploidy
level in adaxial epidermal bulliform cells of Lolium
multiflorum. Cavallini et al. (1995) reported that endoreduplication occurred in the dark in epidermal bulliform
cells of Triticum durum and that lower endopolyploidy
levels were induced by light treatment. In the last two
reports, light had a reductive effect on endoreduplication
cycles. A similar effect of light on endoreduplication was
reported in Arabidopsis hypocotyls (Gendreau et al.,
1997, 1998).
In Arabidopsis leaves, the timing of endoreduplication
cycles coincides with that of post-meristematic cell
enlargement (Beemster et al., 2005). It has been shown
that shading reduced the number of endoreduplication
cycles, but increased cell area in Arabidopsis leaves
(Cookson et al., 2006). However, many studies on the
effect of environmental and internal signals on endoreduplication have considered the ploidy level in an organ as
a whole without making a distinction between the
different tissues. Moreover, it is unknown how endopolyploidization and cell enlargement are orchestrated in lightinduced leaf expansion. In this study, the changes in cell
size of adaxial and abaxial epidermal pavement cells that
occurred when bean plants were transferred from LL to
ML were investigated, and correlated with the change in
nuclear DNA content of adaxial pavement cells.
Materials and methods
Plant material
Bean seeds (Phaseolus vulgaris L. cv. Yamashiro-kurosando-saito)
were purchased from Takii Co., Kyoto, Japan. About 30 selected
seeds were surface-sterilized with NaOCl solution (1% Cl) for 10
min, then soaked in running tap water for 30 min. Next, the seeds
were kept in about 3000 cm3 of water (about 3 cm depth) at 25 C
overnight. The seeds were then germinated on wet filter paper in
a plastic tray at 25 C in the dark. After 24 h, three of the
germinated seeds were transplanted to a plastic beaker (diameter
10 cm, height 11 cm) containing 500 cm3 of vermiculite wetted with
250 cm3 of Hoagland’s solution. In a few experiments, seedlings
were grown in a glass room at 15–30 C under sunlight. Bean
plants were grown from 12 May 2000 (2-d-old) to 28 May 2000
(18-d-old). The intensity of sunlight was about 1800 lmol m2 s1
at midday. In most experiments, beakers were placed in a growth
chamber at 25 C under a light regimen of 16 h light and 8 h dark.
Two light intensities (moderate, ML: 163 lmol m2 s1 and low,
LL: 15 lmol m2 s1) were used in the growth chamber
experiments. The light source was white fluorescent lamps (Toshiba
FLR40S-W/M/36). To produce the LL condition, the number of
fluorescent lamps was reduced and the light was shaded with filter
paper (No. 2, Advantec, Tokyo, Japan). Deionized water was added
to maintain about 50 cm2 of gravitational water in a plastic beaker.
Primary leaves completed unfolding when the plants were 7-d-old.
At this stage, seedlings were selected for uniformity.
Cytofluorometry
The most serious obstacle to getting reliable nuclear DNA content
data of leaf mesophyll cells is high and irregular background
Difference between adaxial and abaxial epidermal cells 1421
fluorescence emitted by chloroplast DNA (Fig. 1A). To reduce this
background level, mesophyll protoplasts, flattened between the
microscope slide and coverslip (Kinoshita et al., 1991) (Fig. 1B)
were used. Pavement cells in the peeled epidermal tissue are very
good material for cytofluorometric measurement of nuclear DNA
content because of their low background level (Fig. 1C). In the
present study, peeled epidermal tissues and flattened mesophyll
protoplasts were stained with 4#,6-diamidino-2-phenylindole
(DAPI) and the intensity of fluorescence emitted from the nuclei
was measured.
Epidermal tissue was peeled from the central half of the leaf
between the tip and the base. At least five epidermal fragments from
the adaxial or abaxial surface of 2–3 leaves were placed on
a gelatin-coated microscope slide with their external face in contact
with the gelatin, and stained with DAPI as described by Kinoshita
et al. (1991). Mesophyll protoplasts were isolated from the leaves,
flattened between the microscope slide and cover slip, and then stained
with DAPI as described by Kinoshita et al. (1991). Experiments were
repeated several times and the data were accumulated.
The fluorescence intensity of each nucleus was measured by
cytofluorometry. Measurements were made under an epifluorescence microscope, a Nikon Optiphot (Nikon, Tokyo, Japan)
equipped with a microphotometer Nikon P1 (Nikon, Tokyo, Japan).
UV excitation (excitation filter, EX 330–380) was used. Emissions
shorter than 470 nm were cut by an IF470 filter. Fluorescence
intensity was measured earlier than 10 s after beginning excitation.
Values were converted to arbitrary units by dividing them by the
mean value of 20 measurements of chick red blood cells on the
same slide.
Detection of DNA synthesis
5-bromo-2#-deoxyuridine (BrdU), an analogue of thymidine, was
used. A primary leaf attached to a bean plant was submerged in
50 cm3 of BrdU solution (10 lM) in a 9 cm Petri dish. The Petri
dish was then placed in a plastic bag, and the open end of the bag was
sealed by wrapping cellophane tape around the petiole, after which the
outside of the plastic bag was completely covered with aluminium foil
because BrdU solution is light-sensitive (Darzynkiewicz and Juan,
1997). The primary leaves were incubated in BrdU solution for 19 h.
Epidermal tissues were peeled from the adaxial and abaxial
surfaces, then guided by a dissecting microscope spread on the
surface of a water droplet placed on a gelatin-coated microscope
Fig. 1. Fluorescence micrographs of nuclei and their backgrounds. (A–
C) Isolated mesophyll cell (A), flattened mesophyll protoplast (B), and
pavement cells of peeled abaxial epidermal tissue (C) were stained with
DAPI and viewed by fluorescence microscopy. Bar, 10 lm. Mesophyll
protoplasts were prepared using an enzyme solution (pH 5–7) that
contained 0.3 M mannitol, 0.25% Macerozyme R-10 (Yakult Honsha
Co., Ltd., Tokyo, Japan), and Cellulase ‘Onozuka’ RS (Yakult Honsha
Co., Ltd., Tokyo, Japan). To prepare isolated mesophyll cells, Cellulase
‘Onozuka’ RS was omitted from the enzyme solution.
slide. The external face of the epidermis was pasted to the
microscope slide by air-drying and then fixed with 100% methanol.
This procedure was the same as the microscope slide preparation for
cytofluorometry.
BrdU incorporated into the nuclear DNA was made visible by the
method of Levi et al. (1987). Slides were hydrolysed in 1.5 N HCl
for 30 min at 25 C, washed thoroughly in TRIS buffer [10 mM tris
(hydroxy-methyl)-aminomethane, 10 mM EDTA-2Na, 100 mM
NaCl, pH 7.2 containing 0.5% (w/v) Triton X-100], and dried.
Next, the samples were treated with anti-BrdU mouse monoclonal
antibody (Becton Dickinson Immunocytometry Systems, San Jose,
CA, USA) and goat anti-mouse IgG conjugated with FITC (Becton
Dickinson Immunocytometry Systems, San Jose, CA, USA), after
which they were counterstained for 15 min with DAPI, as described
by Kinoshita et al. (1991), and observed under an epifluorescence
microscope with appropriate filter combinations.
Leaf area and cell area determination
At least six primary leaves were harvested. They were photocopied,
and the copied leaf shapes were cut out and weighed. Based on
a paper weight of 10310 cm (100 cm2), the weights of the copied
leaf shapes were converted to leaf area.
Epidermal tissues were peeled from adaxial and abaxial surfaces
of the leaves and pasted on to a microscope slide as described
above. Next, photographs were taken under a microscope with
a digital camera. For each sample, areas of 200 pavement cells were
measured with the computer software Image-Pro Express (Media
Cybernetics, Inc., Silver Spring, MD, USA). At least six leaves
from separate plants were used in each measurement.
Statistical analysis
The data of leaf area and epidermal cell area were analysed,
respectively, by the t test and the Mann–Whitney test.
Results
Nuclear DNA content distribution in primary leaves of
sunlight-grown bean plants
Using primary leaves of bean plants grown in ML, it has
been shown that the nuclear DNA contents of adaxial and
abaxial epidermal pavement cells differ from each other
(Kinoshita et al., 1991). However, the nuclear DNA
content of bean plants grown in sunlight has not been
measured, although the intensity of sunlight is much
higher. Figure 2 shows the nuclear DNA content distributions in adaxial and abaxial epidermal pavement cells.
This indicates that most adaxial pavement cells have a 4C
DNA nucleus and that most abaxial pavement cells have
a 2C nucleus. This is similar to our previous reported
findings about the plants grown under ML (Kinoshita
et al., 1991). This indicates that ML is sufficient to
produce the difference in nuclear DNA content between
adaxial and abaxial pavement cells.
Induction of rapid leaf expansion
When bean plants were grown under the LL condition,
stems elongated, and LL-grown plants were more than
twice the height of ML-grown ones at 13 d. When 13-d-old
1422 Kinoshita et al.
Fig. 2. Nuclear DNA content distributions in epidermal cells of
sunlight-grown plants. (A, B) Epidermal tissues were peeled from
adaxial (A) and abaxial (B) surfaces of primary leaves of 18-d-old bean
plants grown under sunlight, and nuclear DNA contents were measured
by cytofluorometry. The numbers of nuclei examined were 100 for (A)
and 100 for (B).
LL-grown plants (Fig. 3A) were transferred to ML, cells in
the primary leaves enlarged rapidly (Fig. 3B). Both the
thickness of the leaves and the thickness of epidermal cells
increased markedly during the ML treatment (Fig. 3B).
However, the thickness of epidermal cells was not measured
as it was difficult to make accurate transverse sections
suitable for quantitative analysis. Leaf area had increased by
58% 5 d after transfer to ML (Fig. 3C). Leaves of the plants
exposed to ML for 5 d were significantly larger than those
of LL-grown plants of the same age (P <0.001) (Fig. 3C).
However, leaves of 18-d-old LL-grown plants were not
significantly larger than those of 13-d-old LL-grown plants
(P >0.05) (Fig. 3C). This shows that the increase in the leaf
area of LL-grown plants had almost reached a plateau at
13 d. However, when 13-d-old plants were transferred to
ML, leaf area increased rapidly.
Changes in nuclear DNA content distribution after
transfer to ML
When bean plants were grown under the LL condition,
most adaxial epidermal pavement cells had a 2C DNA
Fig. 3. Rapid enlargement of bean primary leaves after transfer to ML.
(A, B) Transverse section of primary leaf of a 13-d-old LL-grown plant
(A) and that of a plant aged 5 d after 13-d-old LL-grown plants were
transferred to ML (B). These sections were cut with a Plant Microtome
MTH-1 (NK System, Osaka, Japan). Bar, 50 lm. (C) Changes in leaf
area of bean primary leaves. Squares, LL-grown plants; circles, plants
transferred from LL to ML at 13 d. Solid lines and dashed lines indicate
LL and ML, respectively. Vertical bars through the points indicate the
magnitude of the SE. The SE of the data point without bar is smaller
than the width of the symbol.
nucleus, and the 4C peak area did not increase even when
plants reached 21-d-old (Fig. 4A–D). Moreover, abaxial
pavement cells and mesophyll cells showed the same
nuclear DNA content distributions as did those cells of
ML-grown plants (Kinoshita et al., 1991); most of those
cells had a 2C DNA nucleus (Fig. 4E–G).
Next, 13-d-old plants grown under the LL condition
were transferred to ML, and the nuclear DNA contents of
their adaxial pavement cells were measured at 2, 4, 6, and
8 d after exposure (Fig. 5A–D). Two days after transfer to
the ML condition, the 4C peak area had increased slightly
and became higher at 4 d after transfer. Subsequently, the
proportion of 4C nuclei did not change significantly. On
the other hand, nuclear DNA content distribution in the
abaxial epidermal pavement cells was not affected by
exposure to ML (Fig. 5E, F). This was also the case with
mesophyll cells (Fig. 5G). These results indicate that
abaxial pavement cells and mesophyll cells had been
differentiated from adaxial pavement cells in their reactivity
to ML. Similar results were obtained when 12-d-old
Difference between adaxial and abaxial epidermal cells 1423
Fig. 4. Nuclear DNA content distributions in LL-grown bean primary leaves. (A–D) Adaxial epidermal pavement cells at (A) 11 d; (B) 13 d; (C) 17
d; (D) 21 d. The numbers of nuclei examined were 200 for (A), 310 for (B), 190 for (C), and 282 for (D). (E, F) Abaxial epidermal pavement cells at
(E) 13 d; (F) age 21 d. The numbers of nuclei examined were 200 for (E) and 100 for (F). (G) Mesophyll cells at 13 d; 220 nuclei were examined.
LL-grown plants were transferred to ML, but nuclear
DNA content of most adaxial pavement cells doubled
within two days (data not shown). Also, similar results
were obtained when low intensity red light was used
instead of LL, which is the same condition as that used by
Van Volkenburgh and Cleland (1979) (data not shown).
When 21-d-old LL-grown plants were transferred to
ML, nuclear DNA content distributions of the leaves did
not change (Fig. 6).
Investigation of DNA synthesis
In nuclei whose DNA content increased from 2C to 4C,
one round of DNA synthesis must have occurred. DNA
synthesis in epidermal pavement cells was therefore
investigated by a non-radioactive labelling method that
used BrdU. BrdU was incorporated into replicating DNA
in bean primary leaves and bound with anti-BrdU mouse
monoclonal antibody. The antibody was then bound with
anti-mouse goat IgG conjugated with FITC for visualization under a fluorescence microscope. Nuclei made visible
with FITC fluorescence under blue excitation (Fig. 7A, C)
had progressed through the S-phase. All the nuclei present
were made visible with DAPI fluorescence under UV
excitation (Fig. 7B, D), but only a small number were
labelled with BrdU (arrows, Fig. 7B, D). Both the total
and labelled nuclei were counted in the same field by
1424 Kinoshita et al.
Fig. 5. Changes in nuclear DNA content distributions in primary leaves, after 13-d-old LL-grown plants were transferred to ML. (A–D) Nuclear
DNA content distributions in adaxial pavement cells of the leaves aged 2 d (A), 4 d (B), 6 d (C), and 8 d (D) after transfer to ML. The numbers of
nuclei examined were 410 for (A), 250 for (B), 124 for (C), and 401 for (D). (E, F) Nuclear DNA content distributions in abaxial pavement cells of
the leaves aged 4 d (E) and 8 d (F) after transfer to ML. The numbers of nuclei examined were 100 for (E) and 200 for (F). (G) Nuclear DNA content
distribution in mesophyll cells of the leaves aged 8 d after transfer to ML; 100 nuclei were examined.
fluorescence microscopy, respectively, under UV and blue
excitation. The ratios of labelled nuclei were calculated
(Fig. 8). No DNA synthesis occurred in guard cells and
trichomes (Fig. 7).
Cell division continued during the early phase of leaf
development in leaves of dicotyledonous plants, division
ceasing earlier in the tip than in the base (Granier and
Tardieu, 1998, 1999; Tardieu and Granier, 2000). Epider-
mal tissue, peeled from the central half of the leaf between
the tip and base, was used. To determine the age at which
the cell cycle progression of adaxial and abaxial pavement
cells stops, DNA synthesis was assayed in nuclei of
pavement cells of primary leaves of bean plants grown
under the ML or LL condition. In the adaxial epidermis of
10-d-old and 12-d-old ML-grown plants, respectively,
1.03% and 0.1% of the nuclei synthesized DNA in the
Difference between adaxial and abaxial epidermal cells 1425
Fig. 6. Changes in nuclear DNA content distributions in primary leaves, after 21-d-old LL-grown plants were transferred to ML. (A–C) Nuclear
DNA content distributions in adaxial pavement cells of the leaves aged 2 d (A), 4 d (B), and 6 d (C) after transfer to ML. The numbers of nuclei
examined were100 for (A), 113 for (B), and 100 for (C). (D) Nuclear DNA content distribution in abaxial pavement cells of the leaves aged 8 d after
transfer to ML; 100 nuclei were examined.
period of BrdU uptake (19 h), whereas no synthesis was
detected in nuclei in the abaxial epidermis (Fig. 8).
However, in the primary leaves of LL-grown plants of the
same ages, nuclear DNA synthesis was not detected in the
adaxial or abaxial epidermal pavement cells. When plants
were 13-d-old, no nuclear DNA synthesis was detected in
either type of epidermal pavement cell, irrespective of the
growth irradiance used (Fig. 8). This shows that mitotic
cell cycle progression ceased before 10 d of age in both
adaxial and abaxial epidermal pavement cells of LLgrown plants, as it did in abaxial pavement cells of MLgrown plants. In the case of adaxial pavement cells of
ML-grown plants, DNA synthesis finished between 12
d and 13 d. This result agrees with the report of Van
Volkenburgh and Cleland (1979) showing that primary
leaves of bean plants grown in low intensity red light
complete cell proliferation before 10 d.
To investigate the effect of elevated light intensity on
nuclear DNA synthesis in leaf epidermal cells, 13-d-old
LL-grown plants were transferred to ML. These LL-
grown plants were exposed to ML for 1–3 d, after which
their primary leaves were submerged in BrdU solution for
19 h to take it up. After exposure to ML for 1 d, 2.59% of
the nuclei in adaxial pavement cells, but no nuclei in
abaxial pavement cells, synthesized DNA (Fig. 8). After
exposure to ML for 3 d also, 2.16% of the nuclei in the
adaxial pavement cells and 0.06% (almost zero) of those
in the abaxial pavement cells synthesized DNA. Similar
results were obtained when both 10-d-old and 12-d-old
LL-grown plants were transferred to the ML condition
(data not shown). The frequencies of the nuclei that
synthesized DNA during 19 h of BrdU uptake ranged
from about 2% to 3% in adaxial pavement cells (Fig. 8)
clearly showing that ML induced nuclear DNA synthesis
in these cells, although the efficiency of BrdU uptake
might be low.
In every experiment done in this study, the light source
was positioned higher than the plants. Therefore, light
intensity at the upper surface of the leaves was higher than
that at the lower surface. Was this the cause of the
1426 Kinoshita et al.
different responses between adaxial and abaxial epidermal
pavement cells? To answer this question, primary leaves
of 13-d-old LL-grown plants were fixed on a glass plate
with the abaxial surface facing fluorescent lamps and kept
Fig. 7. Detection of nuclei that have synthesized DNA. (A–D) Peeled
adaxial epidermal tissues viewed by blue (A, C) and UV (B, D)
fluorescence microscopy. (A) and (B), (C) and (D) are respective fields.
Ten-day-old (A, B) and 13-d-old (C, D) LL-grown plants were exposed
to ML for 1 d, after which their primary leaves were submerged in
BrdU solution. Arrows indicate nuclei that have synthesized DNA. Bar,
100 lm.
for 1 d, after which the leaves were submerged in BrdU
solution. Although the abaxial surface received higher
irradiance than the adaxial one, DNA synthesis frequencies of the nuclei were 2.09% for the adaxial and 0% for
the abaxial epidermal pavement cells (Fig. 8). These
results indicate that adaxial pavement cells have a different
property than abaxial ones.
Cell enlargement after transfer to ML
Photographs of peeled epidermal tissues (Fig. 9A–F) show
that adaxial pavement cells are much larger than abaxial
ones at 12 d after transfer to ML (Fig. 9C, F). To compare
cell sizes quantitatively, areas of pavement cells were
measured. In 13-d-old LL-grown plants, median cell area
of abaxial pavement cells was 673 lm2. The size
increased by 53% in 4 d after transfer to ML (1027 lm2)
and remained unchanged thereafter (951 lm2 at 12 d after
transfer to ML) (Fig. 9H). The size at 4 d after transfer to
ML was significantly different from the size of abaxial
cells in 13-d-old LL-grown plants (P¼0.008). However,
abaxial cell size was not significantly different between
4 d and 12 d after transfer (P¼0.912). In contrast to
abaxial cells, adaxial pavement cells continuously enlarged for 12 d after transfer to ML (Fig. 9G). The median
size of adaxial cells in 13-d-old LL-grown plants was
958 lm2, the size was 1817 lm2 at 4 d after transfer and
2294 lm2 at 12 d after transfer. The size at 4 d after
transfer was significantly different from the size of adaxial
cells in 13-d-old LL-grown plants (P <0.001). The size of
adaxial cells at 12 d after transfer was also significantly
different from the size of adaxial cells at 4 d after transfer
(P¼0.005). These results show that both nuclear DNA
doubling (Fig. 5) and continuous cell enlargement occurred
in adaxial pavement cells after transfer to ML. At 12 d after
transfer to ML, adaxial pavement cells were 2.4 times the
size of abaxial ones, although this ratio had been 1.4 times
Fig. 8. Percentages of nuclei labelled with BrdU. (A, B) Adaxial (A) and abaxial (B) epidermal pavement cells. More than 1000 nuclei were
investigated in each experiment. Asterisks show the experiments in which the abaxial surface faced fluorescent lamps in ML.
Difference between adaxial and abaxial epidermal cells 1427
Fig. 9. Enlargement of epidermal pavement cells after transfer to ML. (A–F) Micrographs of peeled adaxial (A–C) and abaxial (D–F) epidermal
tissues of 13-d-old LL-grown leaves (A, D), and those of leaves aged 4 d (B, E) and 12 d (D, F) after 13-d-old LL-grown plants were transferred to
ML. Bar, 50 lm. (G, H) Changes in size of adaxial (G) and abaxial (H) pavement cells after transfer to ML. Box plots of pavement cell size. The box
includes observations from the 25th to the 75th percentile, and the horizontal line within the box represents the median value. Lines outside the box
represent the 10th and 90th percentiles, and the circles represent outlying values. In each sample, 200 cells were measured. (I) Photograph of
a primary leaf of a plant aged 14 d after 13-d-old LL-grown plants were transferred to ML.
in 13-d-old LL-grown plants. This seems to be a cause of
bulging (Fig. 9I) of the adaxial side of the primary leaves.
The leaves had been flat before transfer to ML.
Discussion
When bean plants grown in low intensity red light were
transferred to bright white light, leaf expansion began
within 10–20 min (Dale, 1988). This leaf expansion is
brought about by activation of the proton pump in the
plasma membrane and loosening of acidified cell wall
(Van Volkenburgh and Cleland, 1980; Cleland et al.,
1983; Staal et al. 1994; Linnemeyer et al., 1990;
Stahlberg and Van Volkenburgh, 1999). In contrast to this
short-term response, it is shown here that the nuclear
DNA content of adaxial epidermal pavement cells of 13d-old LL-grown plants increased 4 d after transfer to ML
(Fig. 5). This is a long-term response. This increase in the
proportion of 4C nuclei is paralleled by cell enlargement
(Fig. 9). However, when 21-d-old LL-grown plants were
transferred to ML, the increase in the proportion of 4C
nuclei did not occur (Fig. 6). This indicates that the bright
white light-induced nuclear DNA doubling does not occur
after the beginning of leaf senescence.
1428 Kinoshita et al.
According to the results obtained in this study, changes
of ploidy level and cell size after transfer to ML are
schematically shown in Fig. 10 for adaxial and abaxial
pavement cells of 13-d-old LL-grown plants. In abaxial
pavement cells, ploidy level remained unchanged during
the experimental period, whereas cell size increased in the
early phase. This cell enlargement is brought about by
activation of the plasma membrane proton pump and
loosening of acidified wall. However, cell enlargement
finished 4 d after transfer to ML and cell size remained
unchanged thereafter. On the other hand, ploidy level
increased to 4C in most adaxial pavement cells within
4 d and cell enlargement continued after the increase in
ploidy level, suggesting that cell enlargement occurring
after 4 d is supported by nuclear polyploidization.
When 13-d-old LL-grown plants were transferred to
ML, leaf area increased by 58% in 5 d (Fig. 3) and cell
area of abaxial and adaxial pavement cells increased,
respectively, by 53% and 89% in 4 d (Fig. 9G, H). The
reason why the rate of increase of adaxial pavement cell
area exceeded that of leaf area is probably that the leaf
surfaces were not flat. These results suggest that few cell
divisions occurred in the epidermal pavement cells after
transfer to ML.
In our previous work (Kinoshita et al., 1991), it was
shown that benzyladenine treatment did not induce cell
division but induced an increase in nuclear DNA content
from 4C to 8C in most adaxial epidermal pavement cells
of 13-d-old ML-grown bean primary leaves. This indicates that these cells had switched from mitotic cell cycle
to endocycle when cell proliferation had finished. Therefore, the ML-induced increase in nuclear DNA content
from 2C to 4C in adaxial epidermal cells of the leaves of
LL-grown plants (Fig. 5A-D) is shown to be an
endopolyploidization.
Fig. 10. A schematic illustration of changes in ploidy level and cell size
after transfer to ML.
It is known that there is proportionality between nuclear
DNA level and cell size in epidermal pavement cells of
Arabidopsis leaves (Melaragno et al., 1993). This
suggests that cell ploidy has some impact on cell size
determination (Sugimoto-Shirasu and Roberts, 2003).
However, phenomena contradicting this relationship are
often observed. For example, Beemster et al. (2002)
showed that root cells from different ecotypes of Arabidopsis are of various sizes but that there was no
correlation between mature cell size and endoreduplication. In this study, nuclear DNA content in abaxial
pavement cells of LL-grown plants did not change after
transfer to ML (Fig. 5E, F), although the cell area of those
cells increased by 53% (Fig. 9H). Consequently, it has
been claimed that there is a positive correlation between
nuclear DNA content and amount of cytoplasm and that
some control mechanism ensures that the amount of
cytoplasm a cell can make and sustain is proportional to
the amount of DNA in its nucleus (Sugimoto-Shirasu and
Roberts, 2003). Both cell growth (i.e. increase in
cytoplasmic macromolecular mass) and cell expansion
(i.e. increase in cell volume through vacuolation) contribute independently to an increase in cell size in plants
(Sugimoto-Shirasu and Roberts, 2003). Endoreduplication
could allow greater cell growth, but cell expansion
(vacuolation) occurs independently of endoreduplication.
Furthermore, the correlation between ploidy level and cell
growth is explained as follows: an increase in nuclear
DNA content could induce an increase in protein
synthesis activity of the cells through activation of
ribosome regeneration and gene expression (Nagl, 1976;
Barow and Meister, 2003; Sugimoto-Shirasu and Roberts,
2003). In this study, it is shown that nuclear DNA
doubling only occurred in adaxial pavement cells. This
suggests that not only vacuolation but also cell growth
contributed to the enlargement of adaxial epidermal cells
(Fig. 9).
Another possible significance of the increase in nuclear
DNA content in adaxial epidermal pavement cells could
be protection of cells from irradiation damage (Nagl,
1978), because solar radiation is absorbed first by the
adaxial epidermal cells in field-grown plants.
It is well known that endopolyploidy levels differ
largely among plant species. For example, Arabidopsis
leaf cells have 2C, 4C, 8C, and 16C nuclei (Galbraith
et al., 1991). In woody plants, in contrast, the mean
number of endoreduplication cycles per nucleus (cycle
value) is very low (Barow and Meister, 2003), namely,
endoreduplication rarely occurs in woody plants. In the
leaves of poplar (Populus sieboldii Miq.), all adaxial and
abaxial epidermal pavement cells had a 2C nucleus,
irrespective of light intensity during their growth
(I Kinoshita, unpublished data).
Nagl (1976) demonstrated that there is a negative
correlation between genome size and endopolyploidy
Difference between adaxial and abaxial epidermal cells 1429
level. To explain the significance of this phenomenon, it
has been suggested that endopolyploidization occurs in
the cells of species with a small genome in order to supply
a minimum amount of nuclear DNA to maintain the
regulatory and functional state (Nagl, 1976; De Rocher
et al., 1990; Galbraith et al., 1991). On the other hand,
Barow and Meister (2003) conjectured that endopolyploidization is a means to accelerate the growth of the
plant species in niches, which require and support fast
development. Genome sizes (2C) of P. vulgaris and
Arabidopsis thaliana are 1.58 pg and 0.43 pg, respectively
(Barow and Meister, 2003). The mean numbers of
endoreduplication cycles per nucleus are 0.31 and 1.66
for lower leaves of P. vulgaris and A. thaliana, respectively (Barow and Meister, 2003). This agrees with
Nagl’s hypothesis (Nagl, 1976). In this study, the
following was shown: although the average endopolyploidization level of bean primary leaves is low, endopolyploidization occurred specifically in adaxial pavement
cells, which rapidly grew for a long time under moderate
irradiance conditions.
References
Barow M, Meister A. 2003. Endopolyploidy in seed plants is
differently correlated to systematics, organ, life strategy and
genome size. Plant, Cell and Environment 26, 571–584.
Beemster GTS, De Veyler L, Vercruysse S, West G,
Rombaut D, Van Hummelen P, Galichet A, Gruissem W,
Inzé D, Vuylsteke M. 2005. Genome-wide analysis of gene
expression profiles associated with cell cycle transitions in
growing organs of Arabidopsis. Plant Physiology 138, 734–743.
Beemster GTS, De Vusser K, De Tavernier E, De Bock K,
Inzé D. 2002. Variation in growth rate between Arabidopsis
ecotypes is correlated with cell division and A-type cyclindependent kinase activity. Plant Physiology 129, 854–864.
Cavallini A, Baroncelli S, Lercari B, Cionini G, Rocca M,
D’Amato F. 1995. Effect of light and gibberellic acid on
chromosome endoreduplication in leaf epidermis of Triticum
durum Desf. Protoplasma 186, 57–62.
Cleland RE, Cosgrove D, Tepfer M. 1983. Characterization of the
in vitro acid extension curves of Avena coleoptiles. Planta 170,
379–385.
Cookson SJ, Radziejwoski A, Granier C. 2006. Cell and leaf size
plasticity in Arabidopsis: what is the role of endoreduplication?
Plant, Cell and Environment 29, 1273–1283.
Dale JE. 1988. The control of leaf expansion. Annual Review of
Plant Physiology and Plant Molecular Biology 39, 267–295.
Darzynkiewicz Z, Juan G. 1997. Analysis of DNA content and
BrdU incorporation. In: Robinson JP, Darzynkiewicz Z, Hyun W,
Orfao A, Rabinovitch PS, eds. Current protocols in cytometry.
New York: John Wiley & Sons, Inc, 7.7.1–7.7.9.
De Rocher EJ, Harkins KR, Galbraith DW, Bohnert HJ. 1990.
Developmentally regulated systemic endopolyploidy in succulents with small genomes. Science 250, 99–101.
Galbraith DW, Harkins KR, Knapp S. 1991. Systemic endopolyploidy in Arabidopsis thaliana. Plant Physiology 96, 985–989.
Gendreau E, Höfte H, Grandjean O, Brown S, Traas J. 1998.
Phytochrome controls the number of endoreduplication cycles in the
Arabidopsis thaliana hypocotyl. The Plant Journal 13, 221–230.
Gendreau E, Traas J, Desnos T, Grandjean O, Caboche M,
Höfte H. 1997. Cellular basis of hypocotyl growth in Arabidopsis thaliana. Plant Physiology 114, 295–305.
Grafi G, Larkins BA. 1995. Endoreduplication in maize endosperm: involvement of M phase-promoting factor inhibition and
induction of S phase-related kinases. Science 269, 1262–1264.
Granier C, Tardieu F. 1998. Spatial and temporal analyses of
expansion and cell cycle in sunflower leaves. A common pattern
of development for all zones of a leaf and different leaves of
a plant. Plant Physiology 116, 991–1001.
Granier C, Tardieu F. 1999. Water deficit and spatial pattern
of leaf development. Variability in responses can be simulated
using a simple model of leaf development. Plant Physiology 119,
609–619.
Griffiths PD, Ougham HJ, Jones RN. 1994. Genotypic and
environmental effects on endopolyploidy in the epidermal tissues
of Lolium perenne L. and Lolium multiflorum Lam. New
Phytologist 128, 339–345.
Imai KK, Ohashi Y, Tsuge T, Yoshizumi T, Matsui M, Oka A,
Aoyama T. 2006. The A-type cyclin CYCA2;3 is a key regulator
of ploidy levels in Arabidopsis endoreduplication. The Plant Cell
18, 382–396.
Kinoshita I, Sanbe A, Yokomura E. 1991. Increases in nuclear
DNA content without mitosis in benzyladenine-treated primary
leaves of intact and decapitated bean plants. Journal of
Experimental Botany 42, 667–672.
Larkins BA, Dilkes BP, Dante RA, Coelho CM, Woo Y, Liu Y.
2001. Investigating the hows and whys of DNA endoreduplication. Journal of Experimental Botany 52, 183–192.
Levi M, Sparvoli E, Sgorbati S, Chiatante D. 1987. Rapid
immunofluorescent determination of cells in the S phase in pea
root meristems: an alternative to autoradiography. Physiologia
Plantarum 71, 68–72.
Linnemeyer P, Van Volkenburgh E, Cleland RE. 1990. Characterization and effect of light on the plasma membrane H+-ATPase
of bean leaves. Plant Physiology 94, 1671–1676.
Melaragno JE, Mehrotra B, Coleman AW. 1993. Relationship
between endopolyploidy and cell size in epidermal tissue of
Arabidopsis. The Plant Cell 5, 1661–1668.
Nagl W. 1976. DNA endoreduplication and polyteny understood as
evolutionary strategies. Nature 261, 614–615.
Nagl W. 1978. Endopolyploidy and polyteny in differentiation and
evolution. Towards an understanding of quantitative and qualitative variation of nuclear DNA in ontogeny and phylogeny.
Amsterdam: Elsevier/North-Holland Biochemical Press.
Rayle DL, Cleland RE. 1992. The acid growth theory of auxininduced cell elongation is alive and well. Plant Physiology 99,
1271–1274.
Schlayer G. 1971. Modifikationen des DNS-Gehalts in Zuckerrübenzellen. Planta 98, 294–299.
Staal M, Elzenga JTM, Van Elk AG, Prins HBA, Van
Volkenburgh E. 1994. Red and blue light-stimulated proton
efflux by epidermal leaf cells of the Argenteum mutant of Pisum
sativum. Journal of Experimental Botany 45, 1213–1218.
Stahlberg R, Van Volkenburgh E. 1999. The effect of light on
membrane potential, apoplastic pH and cell expansion in leaves
of Pisum sativum L. var Argenteum. Role of the plasmamembrane H+-ATPase and photosynthesis. Planta 208,
188–195.
Sugimoto-Shirasu K, Roberts K. 2003. ‘Big it up’: endoreduplication and cell-size control in plants. Current Opinion in Plant
Biology 6, 544–553.
Tardieu F, Granier C. 2000. Quantitative analysis of cell division
in leaves: methods, developmental patterns and effects of
environmental conditions. Plant Molecular Biology 43, 555–567.
1430 Kinoshita et al.
Van Volkenburgh E. 1999. Leaf expansion: an integrating plant
behaviour. Plant, Cell and Environment 22, 1463–1473.
Van Volkenburgh E, Cleland RE. 1979. Separation of cell
enlargement and division in bean leaves. Planta 146, 245–247.
Van Volkenburgh E, Cleland RE. 1980. Proton excretion and cell
expansion in bean leaves. Planta 148, 273–278.
Van Volkenburgh E, Cleland RE. 1990. Light-stimulated cell
expansion in bean (Phaseolus vulgaris L.) leaves. I. Growth can
occur without photosynthesis. Planta 182, 72–76.
Van Volkenburgh E, Cleland RE, Watanabe M. 1990. Lightstimulated cell expansion in bean (Phaseolus vulgaris L.) leaves.
II. Quantity and quality of light required. Planta 182, 77–80.
Yoshizumi T, Tsumoto Y, Takiguchi T, Nagata N,
Yamamoto YY, Kawashima M, Ichikawa T, Nakazawa M,
Yamamoto N, Matsui M. 2006. INCREASED LEVEL OF
POLYPLOIDY1, a conserved repressor of CYCLINA2 transcription, controls endoreduplication in Arabidopsis. The Plant Cell
18, 2452–2468.