High-Resolution Whole-Mount Imaging of Three-DimensionalTissue Organization and Gene Expression Enables the Studyof Phloem Development and Structure in Arabidopsis W

Elisabeth Truernit,a Helene Bauby,a,1 Bertrand Dubreucq,b Olivier Grandjean,c John Runions,d,2

Julien Barthelemy,a and Jean-Christophe Palauquia,3

a Laboratoire de Biologie Cellulaire, Institut Jean-Pierre Bourgin, Unite de Recherche 501, Institut National de la Recherche

Agronomique, 78026 Versailles cedex, Franceb Laboratoire de Biologie des Semences, Institut Jean-Pierre Bourgin, Unite Mixte de Recherche 204, Institut National de la

Recherche Agronomique/AgroParistech, 78026 Versailles cedex, Francec Laboratoire Commun de Cytologie, Institut Jean-Pierre Bourgin, Unite de Recherche 254, Institut National de la Recherche

Agronomique, 78026 Versailles cedex, Franced Department of Plant Sciences, University of Cambridge, Cambridge CB2 3EA, United Kingdom

Currently, examination of the cellular structure of plant organs and the gene expression therein largely relies on the

production of tissue sections. Here, we present a staining technique that can be used to image entire plant organs using

confocal laser scanning microscopy. This technique produces high-resolution images that allow three-dimensional

reconstruction of the cellular organization of plant organs. Importantly, three-dimensional domains of gene expression

can be analyzed with single-cell precision. We used this technique for a detailed examination of phloem cells in the wild type

and mutants. We were also able to recognize phloem sieve elements and their differentiation state in any tissue type and

visualize the structure of sieve plates. We show that in the altered phloem development mutant, a hybrid cell type with

phloem and xylem characteristics develops from initially normally differentiated protophloem cells. The simplicity of sieve

element data collection allows for the statistical analysis of structural parameters of sieve plates, essential for the

calculation of phloem conductivity. Taken together, this technique significantly improves the speed and accuracy of the

investigation of plant growth and development.

INTRODUCTION

During plant development, cells interpret positional cues and

differentiate accordingly. As a result, multicellular plants are

composed of various cell types with specific functions and

characteristic gene expression profiles. To understand the

mechanisms that underlie plant growth and development, it is

essential to visualize gene expression and the organization of

plant organs on a cellular level.

For the study of plant tissue organization, microscopy obser-

vation of plant tissue sections has been used for centuries.

However, the histological sectioning of plant tissue is labor

intensive, and it is difficult to obtain a three-dimensional (3D)

picture of cellular arrangements and gene expression patterns

from tissue sections. Optical sectioning of plant tissue using

confocal laser scanning microscopy (CLSM) overcomes these

problems. It does not require physical tissue sectioning, and it

allows the collection of a series of z axis images and the

subsequent 3D reconstruction of the sample using specialized

computer software.

To visualize living plant tissue using CLSM, fluorescent dyes

are routinely used. One of the limitations of confocal microscopy

using living plant material is the difficulty of observing cellular

details in deeper tissue layers ($50 to 100 mm; Haseloff, 2003).

Most plant cells, especially in the aerial parts of the plant,

produce a variety of substances that protect them from excess

light radiation and that therefore prevent laser light and fluores-

cent emission from getting through the sample (Moreno et al.,

2006). In addition, cellular contents and cell walls cause spherical

aberration and light scattering (Haseloff, 2003). As a conse-

quence, confocal imaging of living plant tissue only works well for

thin and semitransparent organs, such as Arabidopsis thaliana

roots (Helariutta et al., 2000; Birnbaum et al., 2003; Kurup et al.,

2005; Laplaze et al., 2005; Stadler et al., 2005), and for the

observation of external tissue layers, such as the epidermis and

subepidermal cells (Grandjean et al., 2004; Tian et al., 2004).

To allow visualization of internal tissue layers of more complex

organs, several advances have been made in the last years.

Optical projection tomography (OPT) allows optical sectioning

and 3D reconstruction of plant organs of up to 15 mm thickness

(Sharpe et al., 2002; Sharpe, 2003; Lee et al., 2006). Usually,

1 Current address: Institut Cochin, Universite Paris Descartes, UniteMixte de Recherche 8104, Centre National de la Recherche Scientifique,Unite 567, INSERM, 75014 Paris, France.2 Current address: School of Life Sciences, Oxford Brookes University,Oxford OX3 0BP, UK.3 Address correspondence to palauqui@versailles.inra.fr.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: Jean-ChristophePalauqui (palauqui@versailles.inra.fr).W Online version contains Web-only data.www.plantcell.org/cgi/doi/10.1105/tpc.107.056069

The Plant Cell, Vol. 20: 1494–1503, June 2008, www.plantcell.org ª 2008 American Society of Plant Biologists

tissue samples need to be fixed and cleared, but, in case of

semitransparent structures, OPT can also be used for imaging of

live tissues (Lee et al., 2006). However, the resolution of OPT is

limited to 5 3 5 3 5 mm; therefore, smaller cells or intracellular

details are not resolvable.

The combination of confocal microscopy of fixed plant mate-

rial with clearing agents of high refractive indices is a way to

achieve imaging of deeper tissue layers at higher resolution

(0.3 3 0.3 3 0.5 mm). Aniline blue can be used as a stain of cell

contents of Arabidopsis embryos (Bougourd et al., 2000) and

very young (#2 d old) seedlings (Bauby et al., 2007). Aniline blue–

stained samples can be cleared with chloral hydrate to allow

high-resolution confocal microscopy up to a depth of >200 mm

(Bougourd et al., 2000). More recently, we have used a pseudo-

Schiff propidium iodide staining technique for the staining of cell

walls of fixed plant material. The technique is based on the

covalent labeling of cell wall material with fluorophors and the

subsequent clearing of the tissue with chloral hydrate. Fixed

plant tissue is treated with periodic acid, which leads to the

formation of aldehyde groups in the carbohydrates of cell walls.

These aldehyde groups can then react covalently with fluores-

cent pseudo-Schiff reagents, such as propidium iodide, resulting

in samples with highly fluorescent cell walls that are well suited

for confocal microscopy (Haseloff, 2003; Moreno et al., 2006;

Truernit et al., 2006). This method works well for the staining of

Arabidopsis embryos and roots but was not suitable for the

staining of other plant organs or developmental stages.

To study gene expression within plant tissue, reporter genes

such as green fluorescent protein (Haseloff et al., 1997) or

b-glucuronidase (GUS) are used in plants. Green fluorescent

protein expression studies can only be made using live tissue

samples with the constraints described above. GUS expression

studies can be combined with the fixing and clearing of plant

material, but tissue sectioning is often necessary for a detailed

cellular analysis.

The acquisition of vascular tissue was a prerequisite for the

development of land plants. During phloem development, pro-

tophloem sieve element cells develop prior to metaphloem

sieve elements, which are the main solute conducting cells in

the vasculature of differentiated plant organs (reviewed in

Bauby et al., 2007). Both protophloem and metaphloem cells

are long and narrow cells stretched in parallel to the longitudinal

axis of plant organs. Perforated cell walls at their apical and

basal ends, the sieve plates, allow for solute conductivity (Esau,

1969). As the vasculature is the most internal tissue of plant

organs, it is especially difficult to access with classical micros-

copy techniques. Moreover, due to the characteristic shape of

phloem cells, it is hard to obtain physical sections through the

whole length of phloem cells. With the aim to study early phloem

development, we have systematically developed and adapted

the pseudo-Schiff propidium iodide (PS-PI) staining technique

for the visualization of internal tissue structures of all Arabidop-

sis organs at all stages of development. We show that the

modified PS-PI (mPS-PI) staining technique is a highly valuable

tool for the study of the cellular organization of plant tissue. The

method is fast and simple, and the images obtained are of high

resolution and of a quality suitable for 3D reconstruction of

cellular arrangements within a plant organ. Moreover, we show

that the mPS-PI staining method can be combined with GUS

marker gene expression analysis. This makes it possible to

obtain a 3D view of gene expression patterns at the cellular

level. Protophloem cells and their differentiation state can be

easily identified in mPS-PI–stained tissue samples; moreover,

even the visualization of sieve plates was achieved. We dem-

onstrate the usefulness of the mPS-PI technique for the char-

acterization of phloem development and for phloem mutant

analysis.

RESULTS

The mPS-PI Staining Technique Allows the Study of

Arabidopsis Cellular Architecture at All Stages of

Plant Development

The PS-PI staining technique described previously (Haseloff,

2003; Moreno et al., 2006; Truernit et al., 2006) was suitable for

the staining of Arabidopsis embryos and roots. Aboveground

organs of seedlings, however, were not satisfactorily stained,

suggesting poor stain penetration. The differences in staining

efficiency were also visible with the eye: Embryos and roots

appeared pink after the staining procedure, but the aboveground

organs looked white (see Supplemental Figure 1 online). We

therefore systematically tested and improved the PS-PI staining

technique for the staining of all organ types at all stages of plant

development. A treatment with ethanol was necessary to in-

crease staining efficiency for most aboveground organs. The

length of this treatment had to be adapted to the plant organ. For

good staining of ovules and developing embryos in intact ovules,

an alternative method that incorporated sodium hydroxide,

sodium dodecyl sulfate, and sodium hypochlorite treatment

was more effective (see Methods).

Figure 1 and corresponding movies (see Supplemental Movies

1 to 4 online) show optical sections of several Arabidopsis tissue

types stained with the mPS-PI staining method. CLSM shows

that the staining method labels cell walls and starch in plastids

(examples of plastid staining can be seen in Figure 1A in petioles

and in Figure 1H in columella cells). All tissue types are well

stained, and different cell types are easily distinguishable, inde-

pendent of their size and their vacuolation state. The cellular

structure of leaf primordia, developing leaves, roots, and lateral

root primordia was clearly visible (Figures 1A, 1B, 1H, and 1I).

Small cells in meristematic tissues, such as the floral meristem

(Figure 1C), could also be resolved. Ovules could be stained and

imaged without the need to dissect them out of the silique

(Figures 1E and 1F). Even developing embryos inside intact seed

coats were stained (Figure 1G). When scanning toward deeper

tissue layers, the signal intensity decreased only slightly (see

Supplemental Movies 1 to 4 online). Therefore, in all cases, the

working distance of the objective, and not the quality of the

staining, limited the depth of image collection.

The method gave reproducible results, and stained and

mounted tissue could be stored for several months in the dark

without significant loss of stain intensity. Furthermore, very little

photo bleaching occurred while laser scanning a sample. Taken

together, the mPS-PI staining method is highly suitable for the

Plant Tissue Organization and Gene Expression 1495

analysis of the cellular structures of all Arabidopsis organ types

during all stages of development.

The mPS-PI Staining Method Can Be Combined with Gene

Expression Studies

From a collection of promoter-trap lines (Bechtold et al., 1993),

we recently isolated some GUS gene expressing marker lines

(PD1 to PD5) that display GUS activity in protophloem cells at

different stages of their development (Bauby et al., 2007). To

investigate if we could analyze gene expression in mPS-PI–

stained samples, we used line PD2, which shows GUS expres-

sion in immature protophloem cells (Bauby et al., 2007). GUS

activity staining of this line was combined with mPS-PI staining.

To monitor GUS expression in the mPS-PI–stained sample, the

reflection mode of the confocal microscope was used. This

Figure 1. mPS-PI–Stained Arabidopsis Organs at Different Developmental Stages.

(A) Leaf primordium.

(B) Mature leaf with vascular strand.

(C) Flower bud.

(D) Anther with pollen grains.

(E) Silique with developing ovules.

(F) Developing ovule in silique.

(G) Developing embryos inside their seed coats.

(H) Primary root.

(I) Lateral root primordium.

All images are optical sections taken with a confocal microscope. Bars ¼ 20 mm.

1496 The Plant Cell

makes it possible to visualize the crystals that result from GUS

activity assays using 5-bromo-4-chloro-3-indolyl-b-D-glucuronic

acid as a substrate (Pautot et al., 2001).

Figure 2A shows the cotyledon of a mature embryo of marker

line PD2. The long and narrow GUS-stained immature proto-

phloem cells can be recognized easily within the mPS-PI–stained

sample (Figure 2C). Moreover, by taking stacks of confocal

images at steps of 0.1 to 0.2 mm, optical sections along the z axis

of the sample can be reconstructed. The small GUS express-

ing protophloem cells can also be identified in these digital

z-sections (Figure 2B). Also, with this technique, it is possible to

follow cells or tissue types through stacks of z-sections, elimi-

nating the need for exact sections through the tissue under

investigation.

3D Reconstruction of mPS-PI–Stained Samples

Having a 3D view of plant tissue and gene expression is essential

to understand cell–cell interactions. The suitability of the mPS-PI

staining technique for the use with 3D reconstruction software

was therefore important to test. Stacks of images from Arabi-

dopsis embryos were taken along their z axis at steps of 0.2 mm,

and the OsiriX software was used to reconstruct a 3D image of

the embryonic organs (Rosset et al., 2004). Figures 2D, 2E, and

Supplemental Movies 5 and 6 online show that we were able to

obtain high-quality 3D images from mPS-PI–stained samples.

GUS marker gene activity was also easily visible in these 3D

reconstructions (shown for the PD2 GUS marker in Figure 2E).

Virtual sections through the samples revealed internal tissue

organization and gene expression patterns in cellular detail.

Characteristics of Protophloem Cells in

mPS-PI–Stained Samples

After germination, protophloem precursor cells differentiate, and

this process is associated with a thickening of the protophloem

cell walls (Esau, 1969; Busse and Evert, 1999; Bauby et al., 2007).

To investigate if we could recognize protophloem cells and their

differentiation state in mPS-PI–stained samples, we analyzed

roots of promoter-trap line PD1, which shows GUS expression in

protophloem cells from the onset of their differentiation (Bauby

et al., 2007). In the root, two protophloem cell files start to

differentiate basipetally ;2 d after germination. Figure 3A shows

one of these protophloem cell files. PD1 marker gene expression

marks the start of the differentiation process (arrow) at a distance

from the meristem. Basipetal differentiation is accompanied by

cell wall thickening that is clearly visible in the mPS-PI–stained

samples. To confirm our observation, we measured the signal

intensity coming from the transverse cell walls of the developing

Figure 2. GUS Marker Gene (PD2) Expression and 3D Reconstruction.

PD2 marker gene expressing marks specified protophloem cells. Shown

are CLSM images showing propidium iodide fluorescence (white) and

GUS reflection (blue). Bars ¼ 50 mm.

(A) Overview of a cotyledon of PD2 marker line. Digital z-sections

through the cotyledon are shown on top and on the right.

(B) Magnification of digital z-section showing GUS expression in vascu-

lar bundle.

(C) Magnification of protophloem strand showing elongated proto-

phloem cells with GUS expression.

(D) 3D reconstruction of an Arabidopsis embryonic root and hypocotyl

using OsiriX software. Virtual sectioning reveals part of the vasculature

and the cellular structure of the epidermis.

(E) 3D view of PD2 GUS marker gene expression in the cotyledons of an

Arabidopsis embryo. PD2 GUS marker gene expression is seen in

immature protophloem cells throughout the vasculature.

Plant Tissue Organization and Gene Expression 1497

protophloem cell file and compared it to a neighboring cell file.

Signal intensity increased significantly specifically in the proto-

phloem cell file as protophloem cells differentiated (Figures 3C

and 3D). Protophloem cell elongation and cell wall thickening

were linked processes (Figure 3E).

We also looked at other organ types, such as developing leaf

primordia (Figure 3F), mature leaves (Figure 3G), or floral stems

(Figure 3H). In all cases, the protophloem cells could be identi-

fied. Therefore, the mPS-PI staining technique allowed us to

recognize protophloem cells and their differentiation state in all

tissue types, even without the use of marker genes. Moreover,

cellular features, such as cell length or fluorescence intensity of

the mPS-PI–stained cell walls, could easily be quantified.

Visualization of Sieve Plates Suitable for

High-Throughput Analysis

Sieve plates are part of sieve elements. They are mostly found in

transverse and occasionally also in longitudinal orientation.

Sieve plates are perforated by sieve pores, which are essential

for the fluid conducting function of the phloem. In younger sieve

elements, a thin lining of callose is associated with the sieve

pores, while in older sieve elements, callose can be found as

more or less massive deposit that can also cover the whole area

of the sieve plate (Esau, 1969).

The study of sieve plates is extremely difficult and time

consuming. The rare occurrence of this structure within a plant

tissue necessitates extensive tissue sectioning to obtain sec-

tions that contain visible sieve plates. Moreover, the small size of

the sieve pores (<0.5 mM) (Esau, 1969; Thompson and Holbrook,

2003), which is at the resolution limit of light microscopy, mostly

requires the use of electron microscopy. Transverse sieve plates

were seen in longitudinal optical CLSM sections through mPS-

PI–stained phloem cells (see Supplemental Figure 2 online). Due

to the lower z axis resolution when scanning through tissue

samples, however, we were unable to obtain satisfactory images

of transverse sieve plate surfaces when using 3D reconstruction.

To visualize sieve plates in a more convenient way, transverse

Figure 3. Features of Differentiating and Differentiated Protophloem Cells after Germination in mPS-PI–Stained Organs.

(A) PD1 GUS expression marks differentiating and differentiated protophloem cells in a 5-d-old mPS-PI–stained root (arrow).

(B) Differentiating and differentiated protophloem cells can be recognized on the basis of their characteristic shape and their thickened cell walls in the

same sample without the need for marker gene expression.

(C) and (D) Signal quantification of cell wall fluorescence in protophloem cell file (green line) and neighboring cell file (red line). The graph (D) shows

relative signal intensity (I) along the green and red lines shown in (C), with 0 being black and 255 being saturated white (y axis, signal intensity; x axis,

micrometers along quantification line). Signal intensity increases specifically in the cell walls of differentiating protophloem cells.

(E) Graph showing protophloem cell length (y axis) plotted against the distance from the first cell showing increased fluorescence of its basipetal cell

wall in mPS-PI–stained sample (x axis). The zero position on the x axis corresponds to position 1 in (C) and (D). Lengths of cells of protophloem cell files

of three roots were measured, and their position relative to the first cell showing cell wall thickening was determined. Each color represents one phloem

cell file.

(F) PD1 GUS expression in protophloem cells of a leaf primordium (arrow). Thickened cell walls of protophloem cells can be recognized. Differentiation

starts from the base of the primordium.

(G) Protophloem cell file in mature leaf.

(H) Protophloem cell file in floral stalk.

Bars ¼ 50 mm.

1498 The Plant Cell

Arabidopsis (Wassilewskija [Ws]) floral stem sections of 60 to

100 mm were cut and stained. The use of relatively thick sections

increased the probability of finding several sieve plates in one

tissue sample. As CLSM was used, there was no need to cut

exactly at the level of a sieve plate. Figure 4 shows that we were

able to image transverse and longitudinal sieve plates at a very

good resolution and sieve pores were clearly visible. We found

that sieve plates in 2-d-old stems (4 to 5 cm long) were signif-

icantly (P ¼ 0.0001) smaller than sieve plates in stems 1 week

after bolting: On average, the first sieve plates had a surface of 20

mm2 (n¼ 10) (Figure 4B), while sieve plates at later stages (n¼ 11,

stems 10 to 15 cm long) had an average surface of 31 mm2 (Figure

4F; see Supplemental Figure 3 online). Sieve pore area also

increased significantly (P ¼ 0.002) from 0.17 to 0.21 mm2 (see

Supplemental Figure 3 online). At the same time, sieve pore

density remained relatively constant (see Supplemental Figure 3

online). In 1-week-old stems, we occasionally found sieve plates

that were clogged with a structure that most likely represented

callose (Figure 4H). To confirm this, we performed a double

labeling with aniline blue, which is known to stain callose (Stadler

et al., 1995). Callose deposits could be found on those sieve

plates (see Supplemental Figure 4 online).

Early Phloem Development in the woodenleg and altered

phloem development Mutants

To date, only a few mutants with defects in phloem development

have been described. The woodenleg (wol ) mutant lacks phloem

tissue in roots and the lower part of the hypocotyl (Scheres et al.,

1995). It has been demonstrated that WOL is required for the

asymmetric division of cells in the stele tissue and that the lack of

phloem in wol mutants is the indirect consequence of the

reduced number of cells in the vascular cylinder (Scheres et al.,

1995; Mahonen et al., 2000). So far, the mutant phenotype of wol

has mainly been studied in transverse sections. Therefore, it is

not clear how the transition of phloem cells to nonphloem cells

occurs in wol mutants.

We used the mPS-PI staining technique to look more carefully

at this transition in stacks of longitudinal optical sections of wol

mutants. While in the wild type two continuous files of differen-

tiated protophloem cells could be seen throughout the root

hypocotyl axis (Figures 5C and 5D), only a few differentiated

protophloem cells in the upper part of the hypocotyl were seen in

wol (Figures 5A and 5B; see Supplemental Movie 7 online). These

cells were continuous with a file of cells that was not differen-

tiated as protophloem. The end of protophloem cell differentia-

tion was abrupt. We did not find cells that showed partial

protophloem cell differentiation. We also could not detect any

protophloem cell specification in the lower part of the root

hypocotyl axis when using our PD marker lines (data not shown).

In most cases, protophloem cell differentiation in the two proto-

phloem cell files did not stop at the same time, consistent with

our observation that the number of cells in the vascular cylinder

of wol mutants decreased gradually.

The altered phloem development (apl ) mutant shows defects

that are more specific to phloem cell specification. In this mu-

tant, cells with characteristics of tracheary elements of xylem

were found in the position normally occupied by phloem sieve

Figure 4. Visualization of Arabidopsis Sieve Plates.

CLSM images taken from 60- to 100-mm transverse sections through Arabidopsis Ws stems.

(A) to (D) Two-day-old stems.

(E) to (H) One-week-old stems.

(A) and (E) Overviews of the stem sections.

(B) and (F) Vascular bundles with transverse sieve plates.

(C) and (G) Longitudinal sieve plates in the same sample.

(D) and (H) 3D reconstruction of sieve plates using OsiriX software. A callose plug is visible in (H).

Bars ¼ 100 mm in (A) and (E) and 5 mm in (B) to (D) and (F) to (H).

Plant Tissue Organization and Gene Expression 1499

elements (Bonke et al., 2003). Metaphloem and protophloem cell

markers were shown to be absent in the apl mutant background.

Expression of APL in protoxylem cells prevented those cells from

developing xylem characteristics. These results suggested that

APL promotes phloem differentiation and is required for the

inhibition of xylem differentiation in phloem poles.

To find out how cells at the position of protophloem cells

developed in the apl background, we analyzed a segregating apl

population (apl homozygous plants are seedling lethal) from the

mature embryo stage to 4 d after germination using the mPS-PI

technique (n ¼ 50 for each stage). Until 2 d after germination, all

plants from the segregating apl/APL population looked like wild-

type plants. Moreover, all plants (50 of 50) showed continuous

cell files of elongated protophloem cells with brightly fluorescent

cell walls (Figures 5E and 5F). This suggests that neither the

number of cells nor the position of phloem cell files or the first

steps of protophloem differentiation were compromised in this

mutant. This is in agreement with our observation that early

protophloem PD markers were expressed in the apl background

(data not shown). Three days after germination, plants homozy-

gous for the apl mutation could be identified due to their shorter

primary root (12 of 50). All of those plants now showed defects in

protophloem differentiation: Cells with the appearance of proto-

phloem cells developed spiralled cell wall thickenings reminis-

cent of the cell wall modifications found in xylem tracheary

elements (Figure 5J). In the root, this secondary modification

appeared at the same time in xylem and phloem cell files,

suggesting that both cell types underwent the same develop-

mental program at this stage. Taking into account the initially

normal development of the cells in protophloem cell position in

apl mutants, we were curious to know if those cells showed other

features of phloem cells. Surprisingly, protophloem cells of 7-d-

old apl mutants displayed sieve plates (Figures 5K and 5L),

suggesting that although this mutant seems to have nonfunc-

tional phloem, it still shows some phloem-specific structures.

DISCUSSION

We present a plant tissue staining technique that can be used in

combination with CLSM to visualize the cellular structure of plant

organs at high resolution. Initially, this technique was used for the

staining of embryonic tissue (Haseloff, 2003). To use it for other

developmental stages and tissue types, some crucial modifica-

tions to the staining procedure had to be introduced. A treatment

of plant tissue with hot ethanol is conventionally used to isolate

cuticle components. In our case, this treatment significantly

increased stain penetration into aboveground organs, suggest-

ing that the cuticle was interfering with this process. Treatment

with sodium hydroxide and sodium dodecyl sulfate had the same

effect for the staining of ovules. Subsequent sodium hypochlorite

treatment most likely bleached tannins that accumulate in the

seed integuments (Lepiniec et al., 2006).

With the mPS-PI staining technique, internal tissue layers of all

organ types at all developmental stages from embryogenesis to

seed set were well stained. Cellular organization, even of the long

and narrow vascular cells in the center of the plant, could be

discerned with great detail in optical sections through image

stacks taken along the z axis of the samples. Moreover, the

Figure 5. Phloem Development in wol and apl Mutants.

(A) Optical section through one protophloem cell file in the root-

hypocotyl axis of a 36-h-old wol seedling (Col-0 background). Differen-

tiated protophloem cells can only be seen in the upper part of the

hypocotyl.

(C) and (E) Corresponding section through the root-hypocotyl axis of an

Arabidopsis wild-type (Col-0) (C) seedling and apl (E) seedling. Differ-

entiated protophloem cells form a continuous file.

(B), (D), and (F) Higher magnifications of the insets in (A), (C), and (E),

respectively. Arrows show the protophloem cells.

(G) Xylem cell file of an Arabidopsis root 3-d after germination.

(H) Protophloem cell file of the same root.

(I) Xylem cell file in an apl root 3 d after germination.

(J) Protophloem cell file of the same root showing xylem characteristics.

(K) Longitudinal sieve plate (arrow) in the hypocotyl of an apl plant.

(L) Transverse sieve plate in the same sample.

PP, protophloem cell; X, xylem cell. Bars ¼ 50 mm in (A) to (F) and 10 mm

in (G) to (L).

1500 The Plant Cell

staining method can be combined with GUS staining to analyze

gene expression on the cellular level. The 3D reconstruction of

image stacks is also possible, giving a detailed 3D view of tissue

structure. Here, we have concentrated our analysis on Arabi-

dopsis organs and tissue types, but the technique can be used

equally for the staining of other plant species, such as tomato

(Solanum lycopersicum) or Brachypodium (see Supplemental

Movie 8 online).

Using the mPS-PI staining technique together with CLSM has

many obvious advantages when compared with conventional

histological tissue sectioning. It delivers a view of cellular ar-

rangements and gene expression in three dimensions without

the need for labor-intensive tissue sectioning. Whole organs can

be scanned and optical sections can be made through the

sample at any desired position. The production of a stack of

images suitable for 3D reconstruction through the cotyledon of a

mature embryo, for example, did not take us more than 5 min,

and this time is likely to decrease with the improvement of

confocal microscopy. Analysis of 3D gene expression and tissue

organization will be useful for the study of asymmetric organs, of

bilateral symmetry, or of gene expression patterns with axial

preferences.

So far, phloem development has been mainly studied using

histological sections in combination with light or electron mi-

croscopy (Esau, 1969; Busse and Evert, 1999; Helariutta et al.,

2000; Bonke et al., 2003). Most likely because of the difficulty of

producing longitudinal sections through the long and narrow

phloem cells, many studies of phloem development have been

limited to the use of transverse sections (Scheres et al., 1995;

Helariutta et al., 2000; Bonke et al., 2003). Confocal microscopy

eliminates the need for the production of physical sections

through the long and narrow phloem cells. Moreover, it allows

following the continuity of vascular strands through a series of

z-stack images. Thus, even if the specimen is not completely flat,

vascular cells can still be followed along their longitudinal axis.

Differentiating and mature protophloem cells in different plant

organs can be unequivocally recognized on the basis of their

characteristic shape and their thickened cell walls.

Moreover, also for the analysis of transverse sections, our

method shows clear advantages. Confocal microscopy elimi-

nates the need to produce sections at exactly the point where the

tissue should be analyzed, and this is especially useful for the

study of sieve plate structures. To study sieve plates with CLSM,

recent advances have been made using plants transgenic for a

membrane-anchored protein fused to a yellow fluorescent pro-

tein expressed under the control of a companion cell–specific

promoter (Thompson and Wolniak, 2008). The method presented

here will be a valuable alternative because it is not limited to the

use of transgenic plants. It will allow for a detailed and rapid study

of sieve plate structure throughout plant development, in condi-

tions of physiological constraints, and in different genetic back-

grounds. Due to the relatively easy and fast sample collection,

statistical analysis of sieve plate and sieve pore parameters,

such as number, shape, or size, will now be possible. Knowing

these parameters will be important for the calculation of sieve

tube conductivity. For example, we show that sieve plates in

young stems are significantly smaller and have smaller sieve

pores than those of older stems. This means that the conductivity

of the sieve tubes of older stems will be higher than those of

stems shortly after bolting (Thompson and Wolniak, 2008).

We also used the mPS-PI technique to analyze mutant phloem

phenotypes. In longitudinal optical sections through the vascu-

lature of wol mutants (Scheres et al., 1995; Mahonen et al., 2000),

we show that the developed protophloem cell file in the upper

part of the hypocotyl of wol mutants is continuous with cells that

do not show any protophloem cell specification or differentiation.

During plant development, radial and longitudinal signals need to

be integrated to obtain a continuous vascular network. While

radial signals most likely determine the position of the vascula-

ture in the center of the plant, longitudinal signals ensure the

continuity of the vascular network. In agreement with this, the

differentiation of already specified protophloem cells in young

seedlings is a gradual process that starts from distinct locations

in the plant (Bauby et al., 2007). In the root hypocotyl axis of wol

mutants, radial signals seem to be dominant over longitudinal

signals. Although differentiation of the two protophloem cell files

in the upper part of the hypocotyl is normal, it cannot continue

toward the lower part of the hypocotyl, since protophloem cells

do not seem to be specified in this part of the seedling.

The apl mutant has previously been described as being

defective in proto- and metaphloem development (Bonke et al.,

2003). We confirm this observation; however, we show that the

early steps of protophloem differentiation seem to occur nor-

mally in this mutant. Until 2 d after germination, cells in the

position of protophloem cells display the characteristic shape

and cell wall thickening of protophloem cells. Only after that time

do they start to develop xylem characteristics, while at the same

time sieve plates can be found in this cell type. Consequently,

these cells can be described as hybrid cells with both phloem

and xylem characteristics. Therefore, APL may be required for

later steps of sieve element differentiation but may not be

necessary for the first steps of this process. While the proto-

phloem cell marker J0701 was not expressed in apl (Bonke et al.,

2003), we could identify early PD marker gene expression in apl

plants. One explanation for this may be the timing of the onset of

expression of those markers, with J0701 being expressed later

than some of our early PD marker genes. The other explanation

may be that not all genes that are normally expressed during

protophloem differentiation are expressed in the hybrid cell type

found in apl plants.

Our data show that the use of the mPS-PI staining technique

will open exciting new avenues for the study of phloem devel-

opment. Currently, only a few mutants impaired in phloem

development have been identified. This might be due to the

difficulty of imaging phloem cells and also due to the paucity of

criteria that can be applied to the search for new phloem

mutants. More subtle deviations from normal phloem cellular

structure might not be obvious using conventional imaging

techniques. Moreover, to date, it is difficult to assess vascular

continuity, as it is difficult to obtain longitudinal histological

sections through a row of narrow phloem cells. The technique

described here can be used for a detailed high-throughput

screen for mutants impaired in vascular development.

While we focus our attention on vascular development, the

technique presented here will facilitate and improve the detailed

study of plant anatomy, and it will increase the number of

Plant Tissue Organization and Gene Expression 1501

questions we will be able to ask about almost any aspect of plant

development. Because of its speed, the screening of a large

number of samples will be made possible. Therefore, it will also

be easier to perform statistical analysis of cellular architecture

and gene expression patterns of plants grown under different

conditions or of plants with different genetic backgrounds.

Because of the possibility of visualizing 3D tissue structures,

the relative position of cells in a tissue context and their volumes

or sizes can be studied and evaluated statistically. Taken to-

gether, the method will significantly improve the speed and

accuracy with which plant developmental processes can be

studied.

METHODS

Plant Material and Growth Conditions

If not stated otherwise, Arabidopsis thaliana ecotype Ws was used as the

wild type. The phloem mutants wol and apl (Scheres et al., 1995) were a

gift from Yka Helariutta. The phloem marker lines used were described

previously (Bauby et al., 2007).

Plants were grown on Murashige and Skoog medium supplemented

with 1% sucrose in growth chambers (208C, 70% humidity, 16 h light/8 h

dark).

GUS Staining

Tissue was immersed in GUS staining solution (100 mM sodium phos-

phate buffer, pH 7.2, 10 mM sodium EDTA, 0.1% Triton X-100, and 1 mg/

mL 5-bromo-4-chloro-3-indolyl-D-glucuronic acid [Duchefa], to which

potassium ferrocyanide and potassium ferricyanide to a final concentra-

tion of 2.5 mM were freshly added). The staining solution was infiltrated

into the tissue by subjecting samples to a vacuum for 2 3 2 min. Samples

were incubated at 378C for defined times, depending on the marker used

(Bauby et al., 2007), and then rinsed with water. PS-PI staining followed.

mPS-PI Staining

Whole seedlings or plant organs were fixed in fixative (50% methanol and

10% acetic acid) at 48C for at least 12 h. Tissue could also be stored in the

fixative for up to 1 month. The tissue was then transferred to 80% ethanol

and incubated at 808C for 1 to 5 min, depending on tissue type (for

example leaves, 1 min; floral stalks, 5 min). Tissue was transferred back to

fixative and incubated for another hour. Next, tissue was rinsed with water

and incubated in 1% periodic acid at room temperature for 40 min. The

tissue was rinsed again with water and incubated in Schiff reagent with

propidium iodide (100 mM sodium metabisulphite and 0.15 N HCl;

propidium iodide to a final concentration of 100 mg/mL was freshly

added) for 1 to 2 h or until plants were visibly stained. The samples were

then transferred onto microscope slides and covered with a chloral

hydrate solution (4 g chloral hydrate, 1 mL glycerol, and 2 mL water).

Slides were kept overnight at room temperature in a closed environment

to prevent drying out. Next, excess chloral hydrate was removed and

several drops of Hoyer’s solution (30 g gum arabic, 200 g chloral hydrate,

20 g glycerol, and 50 mL water) were placed over the tissue, and a cover

slip was placed on top. Slides were left undisturbed for a minimum of 3 d

to allow the mounting solution to set. For the staining of roots and

emerged lateral roots, the ethanol step was omitted. For staining of ovules

and seeds, siliques were harvested and slit open on one side. Tissue was

fixed as described above and then subjected to an overnight treatment of

1% SDS and 0.2 N NaOH at room temperature. Siliques were rinsed in

water, incubated in 25% bleach solution (2.5% active Cl�) for 1 to 5 min,

rinsed again, and then transferred to 1% periodic acid. The samples were

then further processed as described above.

Imaging of Sieve Plates and Aniline Blue Staining

Stem and hypocotyl cross sections (60 to 100 mm thick) were obtained

with a vibratome (Leica VT1000S). Samples were then subjected to PS-PI

staining as described above. Callose staining was performed as de-

scribed by Stadler et al. (1995).

Confocal Microscopy

A Leica TCS-SP2-AOBS spectral confocal laser scanning microscope

(Leica Microsystems) was used. The excitation wavelength for PS-PI–

stained samples was 488 nm, and emission was collected at 520 to 720

nm. GUS staining was imaged with the AOBS reflection mode of the

confocal microscope. The excitation wavelength was 488 nm, and the

reflection signal was collected between 485 and 491 nm. Callose fluo-

rescence was collected between 480 to 515 nm using a 405-nm laser.

Data Processing

Data were processed for some two-dimensional orthogonal sections, 3D

rendering, and movie exports using the open source software Osirix

(Rosset et al., 2004; http://homepage.mac.com/rossetantoine/osirix/) on

a quadxeon 2.66-Ghz, 2-GB RAM Apple Mac pro workstation. RGB

stacks of confocal images were imported as DICOm files into Osirix prior

to surface rendering.

For the production of optical sections, for signal quantification, and for

cell length measurements, we used the Leica Confocal Software version

2.61.

Supplemental Data

The following materials can be found in the online version of this article.

Supplemental Figure 1. Treatment with Hot Ethanol Increases Stain

Penetration into Aboveground Organs.

Supplemental Figure 2. Sieve Plate in Longitudinal Optical Section.

Supplemental Figure 3. Parameters of Sieve Plates in 2-d-Old and

1-Week-Old Stems.

Supplemental Figure 4. Sieve Plate with Callose.

Supplemental Movie 1. Z-Scan through Arabidopsis Leaf from

Abaxial to Adaxial Epidermis.

Supplemental Movie 2. Z-Scan through Arabidopsis Inflorescence

Meristem.

Supplemental Movie 3. Z-Scan through Arabidopsis Silique with

Developing Ovules.

Supplemental Movie 4. Z-Scan through Arabidopsis Ovule with

Developing Embryo at Late Globular Stage.

Supplemental Movie 5. 3D Reconstruction of an Arabidopsis Em-

bryonic Root Using OsiriX Software.

Supplemental Movie 6. 3D Reconstruction of Cotyledon of PD2

Marker Line with GUS Expression in Immature Protophloem Cells

Using OsiriX Software.

Supplemental Movie 7. Z-Scan through Root Hypocotyl Axis of the

wol Mutant Showing the Abrupt End of Protophloem Differentiation.

Supplemental Movie 8. Z-Scan through Brachypodium Shoot Mer-

istem with Leaf Primordia.

1502 The Plant Cell

ACKNOWLEDGMENT

We thank ‘‘La Region Ile de France’’ for the cofinancing of the confocal

microscope and Yka Helariutta for wol and apl seeds. We also thank

Gregory Mouille for helpful discussions and Celine Savattero for help in

the lab as a summer student. We thank Patrick Laufs for critical reading

of the manuscript. E.T. was funded by a Marie Curie Intra-European

Fellowship.

Received October 3, 2007; revised May 6, 2008; accepted May 14, 2008;

published June 3, 2008.

REFERENCES

Bauby, H., Divol, F., Truernit, E., Grandjean, O., and Palauqui, J.C.

(2007). Protophloem differentiation in early Arabidopsis thaliana de-

velopment. Plant Cell Physiol. 48: 97–109.

Bechtold, N., Ellis, J., and Pelletier, G. (1993). In planta Agrobacterium

mediated gene transfer by infiltration of adult Arabidopsis thaliana

plants. C. R. Acad. Sci. III, Sci. Vie Life Sci. 316: 1194–1199.

Birnbaum, K., Shasha, D.E., Wang, J.Y., Jung, J.W., Lambert, G.M.,

Galbraith, D.W., and Benfey, P.N. (2003). A gene expression map of

the Arabidopsis root. Science 302: 1956–1960.

Bonke, M., Thitamadee, S., Mahonen, A.P., Hauser, M.T., and

Helariutta, Y. (2003). APL regulates vascular tissue identity in

Arabidopsis. Nature 426: 181–186.

Bougourd, S., Marrison, J., and Haseloff, J. (2000). Technical ad-

vance: An aniline blue staining procedure for confocal microscopy

and 3D imaging of normal and perturbed cellular phenotypes in

mature Arabidopsis embryos. Plant J. 24: 543–550.

Busse, J.S., and Evert, R.F. (1999). Pattern of differentiation of the first

vascular elements in the embryo and seedling of Arabidopsis thaliana.

Int. J. Plant Sci. 160: 1–13.

Esau, K. (1969). The Phloem. (Berlin: Gebruder Borntrager).

Grandjean, O., Vernoux, T., Laufs, P., Belcram, K., Mizukami, Y., and

Traas, J. (2004). In vivo analysis of cell division, cell growth, and

differentiation at the shoot apical meristem in Arabidopsis. Plant Cell

16: 74–87.

Haseloff, J. (2003). Old botanical techniques for new microscopes.

Biotechniques 34: 1174–1178, 1180, 1182.

Haseloff, J., Siemering, K.R., Prasher, D.C., and Hodge, S. (1997).

Removal of a cryptic intron and subcellular localization of green

fluorescent protein are required to mark transgenic Arabidopsis plants

brightly. Proc. Natl. Acad. Sci. USA 94: 2122–2127.

Helariutta, Y., Fukaki, H., Wysocka-Diller, J., Nakajima, K., Jung, J.,

Sena, G., Hauser, M.T., and Benfey, P.N. (2000). The SHORT-ROOT

gene controls radial patterning of the Arabidopsis root through radial

signaling. Cell 101: 555–567.

Kurup, S., Runions, J., Kohler, U., Laplaze, L., Hodge, S., and

Haseloff, J. (2005). Marking cell lineages in living tissues. Plant J. 42:

444–453.

Laplaze, L., Parizot, B., Baker, A., Ricaud, L., Martiniere, A., Auguy,

F., Franche, C., Nussaume, L., Bogusz, D., and Haseloff, J. (2005).

GAL4-GFP enhancer trap lines for genetic manipulation of lateral root

development in Arabidopsis thaliana. J. Exp. Bot. 56: 2433–2442.

Lee, K., Avondo, J., Morrison, H., Blot, L., Stark, M., Sharpe, J.,

Bangham, A., and Coen, E. (2006). Visualizing plant development

and gene expression in three dimensions using optical projection

tomography. Plant Cell 18: 2145–2156.

Lepiniec, L., Debeaujon, I., Routaboul, J.M., Baudry, A., Pourcel, L.,

Nesi, N., and Caboche, M. (2006). Genetics and biochemistry of

seed flavonoids. Annu. Rev. Plant Biol. 57: 405–430.

Mahonen, A.P., Bonke, M., Kauppinen, L., Riikonen, M., Benfey,

P.N., and Helariutta, Y. (2000). A novel two-component hybrid

molecule regulates vascular morphogenesis of the Arabidopsis root.

Genes Dev. 14: 2938–2943.

Moreno, N., Bougourd, S., Haseloff, J., and Feijo, J. (2006). Imaging

plant cells. In Handbook of Biological Confocal Microscopy, J.

Pawley, ed (New York: SpringerScience and Business Media), pp.

769–787.

Pautot, V., Dockx, J., Hamant, O., Kronenberger, J., Grandjean, O.,

Jublot, D., and Traas, J. (2001). KNAT2: Evidence for a link between

knotted-like genes and carpel development. Plant Cell 13: 1719–1734.

Rosset, A., Spadola, L., and Ratib, O. (2004). OsiriX: An Open-Source

Software for Navigating in Multidimensional DICOM Images. J. Digit.

Imaging 17: 205–216.

Scheres, B., Di Laurenzio, L., Willemsen, V., Hauser, M.T., Janmaat,

K., Weisbeek, P., and Benfey, P.N. (1995). Mutations affecting the

radial organisation of the Arabidopsis root display specific defects

throughout the embryonic axis. Development 121: 53–62.

Sharpe, J. (2003). Optical projection tomography as a new tool for

studying embryo anatomy. J. Anat. 202: 175–181.

Sharpe, J., Ahlgren, U., Perry, P., Hill, B., Ross, A., Hecksher-

Sorensen, J., Baldock, R., and Davidson, D. (2002). Optical pro-

jection tomography as a tool for 3D microscopy and gene expression

studies. Science 296: 541–545.

Stadler, R., Brandner, J., Schulz, A., Gahrtz, M., and Sauer, N. (1995).

Phloem loading by the PmSUC2 sucrose carrier from Plantago major

occurs into companion cells. Plant Cell 7: 1545–1554.

Stadler, R., Wright, K.M., Lauterbach, C., Amon, G., Gahrtz, M.,

Feuerstein, A., Oparka, K.J., and Sauer, N. (2005). Expression of

GFP-fusions in Arabidopsis companion cells reveals non-specific

protein trafficking into sieve elements and identifies a novel post-

phloem domain in roots. Plant J. 41: 319–331.

Thompson, M.V., and Holbrook, N.M. (2003). Application of a single-

solute non-steady-state phloem model to the study of long-distance

assimilate transport. J. Theor. Biol. 220: 419–455.

Thompson, M.V., and Wolniak, S.M. (2008). A plasma membrane-

anchored fluorescent protein fusion illuminates sieve element plasma

membranes in Arabidopsis and tobacco. Plant Physiol. 146: 1599–

1610.

Tian, G.W., et al. (2004). High-throughput fluorescent tagging of full-

length Arabidopsis gene products in planta. Plant Physiol. 135: 25–38.

Truernit, E., Siemering, K.R., Hodge, S., Grbic, V., and Haseloff, J.

(2006). A map of KNAT gene expression in the Arabidopsis root. Plant

Mol. Biol. 60: 1–20.

Plant Tissue Organization and Gene Expression 1503

DOI 10.1105/tpc.107.056069; originally published online June 3, 2008; 2008;20;1494-1503Plant Cell

Barthélémy and Jean-Christophe PalauquiElisabeth Truernit, Hélène Bauby, Bertrand Dubreucq, Olivier Grandjean, John Runions, Julien

ArabidopsisExpression Enables the Study of Phloem Development and Structure in High-Resolution Whole-Mount Imaging of Three-Dimensional Tissue Organization and Gene

 This information is current as of July 11, 2020

 

Supplemental Data /content/suppl/2008/05/27/tpc.107.056069.DC1.html

References /content/20/6/1494.full.html#ref-list-1

This article cites 26 articles, 11 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