Plant Cell Physiol. 46(11): 1779–1786 (2005)

doi:10.1093/pcp/pci190, available online at www.pcp.oupjournals.org

JSPP © 2005

Differential Distribution of Proteins Expressed in Companion Cells in the Sieve Element-Companion Cell Complex of Rice Plants

Akari Fukuda 1, 4, *, Syu Fujimaki 1, 5, Tomoko Mori 1, 6, Nobuo Suzui 1, 5, Keiki Ishiyama 2, 7, Toshihiko

Hayakawa 2, Tomoyuki Yamaya 2, Toru Fujiwara 1, 3, 8, Tadakatsu Yoneyama 1 and Hiroaki Hayashi 1, 9

1 Department of Applied Biological Chemistry, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Yayoi, Bunkyo-ku,

Tokyo, 113-8657 Japan 2 Department of Applied Plant Science, Graduate School of Agricultural Science, Tohoku University, Aoba-ku, Sendai, 981-8555 Japan 3 PRESTO, JST, Honcho, Kawaguchi, Saitama, 332-0012 Japan

;

Sieve tubes are comprised of sieve elements, enucle-

ated cells that are incapable of RNA and protein synthesis.

The proteins in sieve elements are supplied from the neigh-

boring companion cells through plasmodesmata. In rice

plants, it was unclear whether or not all proteins produced

in companion cells had the same distribution pattern in the

sieve element-companion cell complex. In this study, the

distribution pattern of four proteins, β-glucuronidase

(GUS), green fluorescent protein (GFP), thioredoxin h

(TRXh) and glutathione S-transferase (GST) were ana-

lyzed. The foreign proteins GUS and GFP were expressed

in transgenic rice plants under the control of the TRXh

gene promoter (PTRXh), a companion cell-specific pro-

moter. Analysis of leaf cross-sections of PTRXh-GUS and

PTRXh-GFP plants indicated high accumulation of GUS

and GFP, respectively, in companion cells rather than in

sieve elements. GUS and GFP were also detected in phloem

sap collected from leaf sheaths of the transgenic rice plants,

suggesting these proteins could enter sieve elements. Rela-

tive amounts of GFP and endogenous phloem proteins,

TRXh and GST, in phloem sap and total leaf extracts were

compared. Compared to TRXh and GST, GFP content was

higher in total leaf extracts, but lower in phloem sap, sug-

gesting that GFP accumulated mainly in companion cells

rather than in sieve elements. On the other hand, TRXh

and GST appeared to accumulate in sieve elements rather

than in companion cells. These results indicate the evi-

dence for differential distribution of proteins between sieve

elements and companion cells in rice plants.

Keywords: Companion cell — Oryza sativa — Phloem —

Rice — Sieve element.

Abbreviations: GFP, green fluorescent protein; GST, glutathione

S-transferase; GUS, β-glucuronidase; PTRXh, thioredoxin h pro-

moter; TRXh, thioredoxin h.

Introduction

In higher plants, phloem is the major route of long dis-

tance transport of metabolites and signals. Sieve elements, the

individual cells that make up sieve tubes, are highly special-

ized for assimilate translocation, and lose most of their intra-

cellular organelles, nuclei, vacuoles, Golgi bodies and most

ribosomes during the course of differentiation (Cronshaw

1981). More than a hundred proteins were detected in phloem

exudates of Cucurbita maxima, Triticum aestivum, Ricinus

communis and Oryza sativa (Eschrich and Heyser 1975, Fisher

et al. 1992, Nakamura et al. 1993, Sakuth et al. 1993). These

phloem proteins were considered to maintain the physiological

functions of sieve tubes, such as sieve plate occlusion, signal

transduction, redox regulation, and phloem loading, among

others (Hayashi et al. 2000).

Since the enucleated sieve elements are unable to synthe-

size proteins, the proteins in sieve tubes are thought to be

supplied from the neighboring companion cells through plas-

modesmata. Indeed, a high frequency of plasmodesmata

between companion cells and sieve elements has been reported

(Chonan et al. 1981, Lucas et al. 1993). Moreover, mRNAs

encoding phloem proteins, such as C. maxima PP1, PP2 and

CmPP16, and O. sativa thioredoxin h (TRXh), have been

detected in companion cells by in situ hybridization analysis

(Bostwick et al. 1992, Clark et al. 1997, Dannenhoffer et al.

1997, Xoconostle-Cázares et al. 1999, Ishiwatari et al. 2000),

suggesting the possible synthesis of phloem proteins in

companion cells. In addition, some phloem proteins, such as C.

maxima PP2 and CmPP16, R. communis glutaredoxin and

cystatin, and O. sativa TRXh, were reported to increase the size

exclusion limit of plasmodesmata and modify their own cell-to-

cell movement after being microinjected into mesophyll cells

(Balachandran et al. 1997, Ishiwatari et al. 1998, Xoconostle-

Cázares et al. 1999). Furthermore, analysis of O. sativa TRXh

4 Present address: Department of Paddy Farming, National Agricultural Research Center for Tohoku Region, Yotsuya, Daisen, Akita, 014-0102 Japan.

5 Present address: Takasaki Radiation Chemistry Research Establishment, Japan Atomic Energy Research Institute, Gunma, 370-1207 Japan.

6 Present address: Fuji Photo Film Co. Senzui, Asaka, Saitama, 351-8585 Japan.7 Present address: RIKEN Yokohama Institute, Suehiro-cho, Tsurumi-ku, Yokohama City, Kanagawa, 230-0045 Japan.8 Present address: Biotechnology Research Center, The University of Tokyo, Yayoi, Bunkyo-ku, Tokyo, 113-8657 Japan.9 Present address: 694 Futago, Aki, Higashikunisaki, Oita, 873-0356 Japan.* Corresponding author: E-mail, akfukuda@affrc.go.jp; Fax, +81-187-66-2639.

1779

Distribution of proteins in phloem tissues1780

mutants indicated that certain structural motifs are necessary

for cell-to-cell movement of this protein in Nicotiana tabacum

mesophyll cells (Ishiwatari et al. 1998). These results sug-

gested the possibility of plasmodesmata carriers, which recog-

nize specific structural features of phloem proteins, increase the

size exclusion limit, and transport phloem proteins through plas-

modesmata (Kragler et al. 1998). The plasmodesmata carrier

hypothesis implies that macromolecules may be loaded into

sieve elements from companion cells by a selective process.

On the other hand, two non-phloem proteins, snowdrop

lectin and green fluorescent protein (GFP) were reported to

traffic into sieve elements from companion cells by a non-

selective mechanism in transgenic N. tabacum plants. Snow-

drop lectin was detected in the honeydew of aphids feeding on

transgenic N. tabacum expressing the snowdrop lectin gene

under the companion cell-specific O. sativa sucrose synthase1

promoter. This result suggested that snowdrop lectin could

enter the sieve tubes from the companion cells and then be

sucked out by the aphids (Shi et al. 1994). GFP was detected in

wild-type N. tabacum shoots grafted onto transgenic plants

expressing the GFP gene from the companion cell-specific

AtSUC2 promoter, suggesting that GFP could enter the sieve

tubes and be transported into the grafted shoots (Imlau et al.

1999).

It has been revealed that the proteins expressed in

companion cells are able to enter the sieve elements. However,

it is unclear whether or not all proteins have the same distribu-

tion pattern in the sieve element-companion cell complex. In

the present study, we generated transgenic rice plants express-

ing foreign proteins, either β-glucuronidase (GUS) or GFP

from the companion cell-specific TRXh promoter (PTRXh)

(Ishiwatari et al. 1995, Ishiwatari et al. 1998), and asked if the

abundance of these proteins, GUS and GFP, and two endog-

enous phloem proteins, TRXh and GST, in phloem sap was dif-

ferent from that in companion cells.

Results

Histochemical assay for GUS activity in the leaves of PTRXh-

GUS transgenic rice plants

The localization of GUS protein was examined histo-

chemically in transgenic rice plants expressing the GUS gene

from the TRXh promoter (PTRXh-GUS) (Fig. 1A). Two inde-

pendent lines were analyzed and there were no observable dif-

ferences between them. The results for one of the lines are

described below. In leaf cross-sections, GUS activity was

detected in both large and small vascular bundle tissues, but not

in mesophyll or epidermal cells (Fig. 2A). Strong GUS activity

was detected only in phloem tissues (Fig. 2B). Weak GUS

activity was also detected in xylem parenchyma cells and some

bundle sheath cells (Fig. 2B). TRXh promoter induced weak

expression in these cells (Ishiwatari et al. 2000). At higher

magnification, companion cells and sieve elements were easily

distinguishable in the large vascular bundle (Fig. 2C; CC and

SE, respectively). The companion cells in particular showed

strong GUS activity (Fig. 2C). Although the sieve elements

also showed a significant level of GUS activity, the intensity of

blue staining was much lower than that in companion cells

(Fig. 2C).

Detection of GFP fluorescence in the leaves of PTRXh-GFP

transgenic rice plants

GFP fluorescence was observed by fluorescence micro-

scopy in transgenic rice expressing the GFP gene from the

TRXh promoter (PTRXh-GFP) (Fig. 1B). Four independent

PTRXh-GFP T0 plants were examined and found to have the

same pattern of fluorescence. The results for one of the plants

are described below. In leaf cross sections, we detected red

autofluorescence from chlorophyll in mesophyll cells and yel-

low autofluorescence in sclerencyma cells in both PTRXh-GFP

(Fig. 2D, E) and wild-type plants (Fig. 2F). In PTRXh-GFP

plants, green fluorescence was detected in large and small vas-

cular bundle tissues (Fig. 2D). Strong green fluorescence was

detected in the phloem tissue (Fig. 2D), and at least five cells

were observed to have especially strong green fluorescence

(Fig. 2E). These cells were identified as companion cells by

their size and position in the phloem region (Chonan et al.

1981). Weak green fluorescence was also detected in xylem

parenchyma cells of PTRXh-GFP plants. The TRXh promoter

induced weak GFP expression in xylem parenchyma cells

(Ishiwatari et al. 2000). Wild type plants showed no green fluo-

rescence in the leaves (Fig. 2F).

Fig. 1 Structure of the chimeric genes used for rice transformation. (A) GUS expression vector. (B) GFP expression vector. P35S, cauliflower

mosaic virus 35SRNA promoter; PTRXh, thioredoxin h promoter; GFP, green fluorescent protein gene; GUS, β -glucuronidase gene; TNOS,

nopaline synthase terminator; PNOS, nopaline synthase promoter; HPH, hygromycin phosphotransferase gene; NPT, neomycin phospho-

transferase gene; RB, TDNA right border; LB, TDNA left border.

Distribution of proteins in phloem tissues 1781

Detection of GUS in the phloem sap of PTRXh-GUS transgenic

rice plants

The presence of GUS protein in the phloem sap of

PTRXh-GUS plants was detected by immunoblotting. Phloem

sap was collected from severed stylets of brown planthoppers

by YAG laser (Kawabe et al. 1980). Protein extracts were pre-

pared from whole leaf and phloem sap of wild-type plants and

T1 individuals from two independent PTRXh-GUS transgenic

lines that showed GUS activity in the phloem region by histo-

chemical assay (Fig. 2A, B, C). Silver staining of the gel

showed that almost the same amount of protein was loaded in

each lane (Fig. 3B). Immunoblot analysis showed that the

phloem sap and leaf extract of PTRXh-GUS plants contained

bands of about 90 kDa (black arrowhead) and about 68 kDa

(black diamond), respectively (lane 1 and lane 3 in Fig. 3A).

Because phloem sap and leaf extract of wild-type plants did not

have these bands, they were presumed to be GUS protein.

Since the predicted molecular mass of GUS is 68 kDa, the band

of 90 kDa in phloem sap might be modified GUS. Immunoblot

analysis also showed smaller bands of 33 kDa and 13 kDa in

phloem sap of transgenic plants (white arrowheads). Because

wild type plants did not have these smaller bands, they might

originate in GUS. Modified band and smaller bands were

detected in the phloem sap of T1 plants of two independent

pTRXh-GUS lines used in this study (data not shown).

Detection of GFP in the phloem sap of PTRXh-GFP transgenic

rice plants

Phloem sap collected from severed stylets of brown plan-

thoppers and leaf extract of PTRXh-GFP plants were analyzed

by immunoblotting using GFP-specific monoclonal antibody.

We examined four independent PTRXh-GFP T0 individuals

that had shown GFP fluorescence in companion cells in the flu-

orescence microscopic analysis (Fig. 2D, E). In wild-type

plants, no band was detected in leaf sheath or phloem sap by

the GFP antibody (Fig. 4A). GFP was detected in both leaf

extracts and phloem sap of the four independent PTRXh-GFP

lines (Fig. 4B, C). The line which contained the highest amount

of GFP in leaves showed the highest concentration of GFP in

phloem sap (Fig. 4B, C).

Fig. 2 Localization of GUS activity

and GFP fluorescence in leaf blades of

PTRXh-GUS and PTRXh-GFP rice

plants, respectively. (A) Transverse sec-

tion of a leaf blade of a T1 PTRXh-GUS

plant stained for GUS activity. (B) and

(C) Magnification of the vascular bundle

region of (A). (D) Transverse section of a

leaf blade of a T0 PTRXh-GFP plant. (E)

Magnification of the vascular bundle

region of (D). (F) Fluorescence of a

transverse section of a leaf blade of a

wild-type plant. CC, companion cell; SE,

sieve element; Ph, phloem; Xy, xylem

vessel; XP, xylem parenchyma cell; BS,

bundle sheath cell; SC, sclerenchyma

cell. Scale bar = 100 µm in (A), (D) and

(F), 25 µm in (B) and (E), and 10 µm in

(C).

Distribution of proteins in phloem tissues1782

Comparison of the contents of GFP, TRXh, and GST in phloem

sap and leaves

We used immunoblot analysis to compare the amount of

GFP with that of endogenous phloem proteins, TRXh and glu-

tathione S-transferase (GST), in phloem sap collected from

severed stylets of brown planthoppers and leaf sheath extract.

The protein concentrations were calculated from the band

intensities of the recombinant proteins. The intensity of bands

detected in the immunoblot analysis was measured using NIH-

image software (National Institute of Health, MD, USA). A

line derived from one PTRXh-GFP T0 plant that showed GFP

fluorescence in companion cells (Fig. 2D, E) was used for the

analysis.

Fig. 5 shows a representative result from the immunoblot

analysis. Though non-specific bands were detected in GFP and

GST analysis, they were not used for a calculation of their

amounts (Fig. 5). The reason why non-specific bands of GFP in

leaf-sheath extract were not detected in Fig. 4B might be that

Fig. 3 Detection of GUS protein in phloem sap and leaf extract of

PTRXh-GUS rice plants. (A) Immunoblot analysis of GUS protein in

phloem sap and leaf extract; lane 1, 20 µl of phloem sap from a T1

PTRXh-GUS plant; lane 2, 20 µl of phloem sap from a wild-type

plant; lane 3, 3 µg of soluble leaf extract protein from a T1 PTRXh-

GUS plant; lane 4, 3 µg of soluble leaf extract protein from a wild-

type plant. Lane 1 contains GUS protein with a molecular mass of

about 90 kDa (black arrowhead) and degradation products of GUS

protein (white arrowhead). Lane 3 contains GUS protein with a molec-

ular mass of about 68 kDa (black diamond). (B) Silver-stained SDS-

PAGE gel of the same fractions used in (A).

Fig. 4 Detection of GFP in phloem sap and leaf extract of PTRXh-

GFP rice plants. (A) Immunoblot analysis of GFP protein in phloem

sap and leaf extract; lane P, 10 ng of recombinant GFP as positive con-

trol; lane 1, the transgenic line of T0 generation; lane wt, wild type

plant. Five µg of leaf soluble proteins and 2 µl of phloem sap were

analyzed. (B) Immunoblot analysis of GFP in phloem sap. Lane P,

10 ng of recombinant GFP as positive control; lanes 1, 2, 3 and 4,

independent transgenic lines of T0 generation. Three µl of phloem sap

from PTRXh-GFP lines were analyzed. (C) Immunoblot analysis of

GFP in leaf extract. Lane P, 10 ng of recombinant GFP as positive

control; lanes 1, 2, 3, and 4, independent transgenic lines of T0 genera-

tion. Five µg of leaf soluble proteins from PTRXh-GFP lines were

analyzed.

Distribution of proteins in phloem tissues 1783

the reaction time with the substrate diaminobenzidine was

shorter than that in Fig. 5A. For GFP, stronger bands were

detected in leaf sheath extract than in phloem sap (Fig. 5A).

However, the TRXh and GST bands were stronger in phloem

sap than in leaf sheath extract (Fig. 5B, C). Fig. 6 shows the

GFP, TRXh, and GST contents in leaf sheath and phloem sap.

Because of the small amount of phloem sap for the analysis,

GST in phloem sap showed a wide range of standard devia-

tion. In leaf sheath, the GFP content was higher than that of

TRXh or GST. In phloem sap, on the other hand, GFP content

was lower than that of TRXh or GST. These results suggested

that, compared to TRXh and GST, the GFP content of sieve

tubes was low, and that GFP was mainly present in other cells,

especially companion cells. In wild type plants, the contents of

TRXh and GST in leaf sheath (per g FW) were lower than in

phloem sap (per liter). TRXh showed a concentration of 5.7 µg

g–1 in leaf sheath FW and 9.4 µg l–1 in phloem sap in wild type

plants. GST showed a concentration of 2.9 µg g–1 in leaf sheath

FW and 3.4 µg l–1 in phloem sap in wild-type plants.

Because of the absence of positive control protein, GUS

contents could not be calculated. But the intensity of the GUS

band in leaf sheath (per gFW) was 1.4 times higher than in

phloem sap (per liter). Only the major bands (indicated by the

black arrowhead and black diamond in Fig. 3A) were used for

the GUS intensity analysis.

Discussion

Immunoblot analysis revealed both GUS and GFP in the

phloem sap of transgenic plants expressing these foreign pro-

teins from the TRXh promoter (Fig. 3, 4). Moreover, histo-

chemical analysis of PTRXh-GUS plants showed that sieve

elements, as well as companion cells, had GUS activity (Fig.

2A, B, C). Since enucleated sieve elements are thought to be

incapable of protein synthesis, and the TRXh promoter has been

shown to drive transcription in companion cells (Ishiwatari et

al. 1998), the results suggested that GUS and GFP are synthe-

sized in companion cells and transported into sieve elements

through plasmodesmata. Based on experiments using trans-

genic plants, snowdrop lectin and GFP have been reported to

traffic into sieve elements in N. tabacum (Shi et al. 1994, Imlau

et al. 1999). In the present study, GUS, which has a larger

molecular mass (68 kDa) than snowdrop lectin (12 kDa) or

GFP (27 kDa), was shown to enter sieve elements. Previous

microinjection experiments demonstrated that 10 kDa, but not

40 kDa, fluorescent dextran could move through plasmodes-

mata between sieve elements and companion cells of Vicia faba

(Kempers and van Bel 1997). In O. sativa, direct introduction

experiments produced similar results: that 3 kDa dextran could

move through the plasmodesmata between sieve elements and

companion cells, but 42 kDa dextran could not (Fujimaki et al.

2000). Dextrans of 3, 10 and 40 kDa have Stokes’ radii of

approximately 1.2, 2.0 and 4.3 nm, respectively (Fisher and

Cash-Clark 2000), suggesting that the size exclusion limit of

the sieve element-companion cell boundary is at least 1.2 nm in

V. faba and 2.0 nm in O. sativa. In the present study, GUS, with

a Stokes’ radius of 3.3 nm (Fisher and Cash-Clark 2000), could

enter the sieve elements. Some phloem proteins, like C.

maxima PP2 and CmPP16, R. communis glutaredoxin and

cystatin, and O. sativa TRXh, were reported to increase the size

exclusion limit of mesophyll plasmodesmata and facilitate their

own cell-to-cell movement after microinjection (Balachandran

et al. 1997, Ishiwatari et al. 1998, Xoconostle-Cázares et al.

1999). In contrast, it was reported that GUS protein alone did

not show cell-to-cell movement after microinjection into tri-

chomes of Nicotiana clevelandii (Waigmann and Zambryski

Fig. 5 Immunoblot analysis of GFP, TRXh, and GST in the phloem

sap and leaf sheath extracts from rice plants. Five µg of leaf sheath

soluble proteins and 2 µl of the phloem sap were analyzed. M indicates

molecular mass markers. (A) Immunoblot analysis of GFP in a T0

PTRXh-GFP plant. Three different amounts of recombinant GFP (15,

10 and 5 ng) were loaded as a positive control. GFP (27 kDa) bands

were detected in the lanes containing recombinant protein, leaf sheath

soluble proteins, and phloem sap. Non-specific bands of higher molec-

ular mass were also detected in the lanes containing leaf sheath solu-

ble proteins. Only the GFP bands (indicated by the arrow) were used

for the intensity analysis. (B) Immunoblot analysis of TRXh in a T0

PTRXh-GFP plant. Recombinant TRXh was loaded at 15, 10, and 5 ng

per lane as a positive control. TRXh (13 kDa) bands were detected in

the lanes containing recombinant protein, leaf sheath soluble proteins,

and phloem sap. (C) Immunoblot analysis of GST in a vector control

plant. Recombinant GST was loaded at 30, 20 and 10 ng per lane as a

positive control. As the recombinant proteins contained a His-tag, they

had a higher molecular mass than the GST in leaf sheath and phloem

sap. GST (31 kDa) bands in leaf sheath and phloem sap are indicated

by an arrow. Two non-specific bands were also detected in phloem sap.

Only the GST bands (indicated by the arrow) were used for the inten-

sity analysis.

Distribution of proteins in phloem tissues1784

1995), suggesting that GUS protein is unable to mediate its

own cell-to-cell transport through plasmodesmata, at least in

trichomes. To explain the cell-to-cell movement of proteins

larger than the size exclusion limit of plasmodesmata, protein

unfolding is proposed. Large proteins like GUS might be

unfolded prior to moving through plasmodesmata and then

refolded in the sieve elements. Indeed, homologues of two

chaperones, rubisco-subunit-binding protein and cyclophilin,

have been detected in sieve-tube exudates from T. aestivum, O.

sativa, Yucca filamentosa, C. maxima, Robinia pseudoacacia

and Tilia platyphyllos (Schobert et al. 1998). Moreover, a small

heat-shock protein has been detected in phloem sap from O.

sativa (Fukuda et al. 2004b).

Although GUS can enter sieve elements, histochemical

analysis of PTRXh-GUS plants revealed that the companion

cells had stronger GUS activity than the sieve elements (Fig.

2C). Furthermore, in leaf sections of PTRXh-GFP plants,

strong GFP fluorescence was detected only in companion cells

(Fig. 2E). The weaker GUS and GFP signals in sieve elements

of the transgenic plants suggested that these proteins were less

abundant in sieve elements than in companion cells. This distri-

bution pattern differs from those of several phloem proteins of

rice, GST (Fukuda et al. 2004a) and small heat-shock protein

(Fukuda et al. 2004b). GST and small heat-shock protein were

reported to be mainly in sieve elements and only weakly

detected in companion cells (Fukuda et al. 2004a, Fukuda et al.

2004b). Immunoblot analysis of PTRXh-GFP plants clearly

showed that, compared to TRXh and GST, GFP was less abun-

dant in the phloem sap even though it was more abundant in

leaf sheath extract (Fig. 6). Moreover, intensity of the GUS

band in leaf sheath (per g FW) was higher than in phloem sap

(per liter) in immunoblot analysis, although TRXh or GST con-

tents were higher in phloem sap. Because strong GFP and GUS

fluorescence was restricted to companion cells (Fig. 2C, E), the

GFP and GUS in leaf sheath extract were considered to origi-

nate mainly from companion cells. The abundance of GFP and

GUS in leaf sheath extract also suggested that GFP and GUS

accumulated in companion cells rather than in sieve elements.

On the other hand, TRXh and GST appeared to accumulate in

sieve elements. These results revealed that proteins expressed

in companion cells had differential distribution pattern in the

sieve element-companion cell complex of rice plants. In other

plants, several phloem proteins, C. maxima PP1 (Clark et al.

1997), CmPP16 (Xoconostle-Cázares et al. 1999), SUT1 of N.

tabacum, Solanum tuberosum, and Lycopersicon esculentum

(Kühn et al. 1997) were reported to be mainly in sieve ele-

ments. PP1 was reported to be mainly in sieve elements and

only weakly detected in companion cells (Clark et al. 1997).

CmPP16 (Xoconostle-Cázares et al. 1999) and SUT1 (Kühn et

al. 1997) were detected only in sieve elements. On the other

hand, C. maxima PP2 and trypsin inhibitor were mainly in

companion cells and weakly detected in sieve elements (Dan-

nenhoffer et al. 1997, Dannenhoffer et al. 2001). One of the

reasons for differential distribution proteins might be effi-

ciency of protein transport from companion cells to sieve ele-

ments. TRXh and GST might be efficiently transported from

companion cells to sieve elements. In contrast, the rate of GUS

and GFP trafficking might be low, and these proteins might

accumulate in companion cells. Because GFP (27 kDa) and

GST (31 kDa) had a similar molecular mass, the distribution

pattern in sieve element-companion cells was not determined

only by molecular mass. Microinjection experiments on

N. tabacum mesophyll cells revealed that TRXh increased the

size exclusion limit of plasmodesmata and modified their own

cell-to-cell movement, and that certain structural motifs of

TRXh are necessary for cell-to-cell movement of this protein

(Ishiwatari et al. 1998). The efficient transport of proteins from

companion cells to sieve elements might need specific struc-

tural motifs, which are recognized by plasmodesmata carriers,

and modify cell-to-cell movements of proteins. Another reason

for the differential distribution pattern of proteins in the sieve

element-companion cell complex might be the degradation

speed of proteins in sieve elements. The degradation speed of

GUS and GFP in sieve elements might be faster than those of

Fig. 6 Amount of GFP, TRXh and GST in

leaf sheath and phloem sap. (A) Proteins in

leaf sheath (µg g–1 FW). (B) Proteins in

phloem sap (µg ml–1 phloem sap). The GFP

bars represent the average of four experi-

ments with standard deviations. The TRXh

bars represent the average of seven experi-

ments with standard deviations. The GST col-

umns represent the average of seven

experiments with standard deviations.

Distribution of proteins in phloem tissues 1785

TRXh and GST, causing the differential distribution pattern of

these proteins in the sieve element-companion cell complex.

The analysis of the phloem sap from PTRXh-GUS plants sug-

gested the modification of GUS protein in phloem sap (Fig. 3A),

although it was not clear whether the modification of GUS was

correlated with the degradation of this protein or not. Further

analysis of the transport efficiency and degradation speed of

proteins is necessary for the understanding of the nature of

phloem.

Materials and Methods

Construction of GUS expression vector and GFP expression vector for

rice transformation

The TRXh promoter (between –1025 and –16 bp from the trans-

lation start site ATG) and GUS gene were transcriptionally fused as

follows. The promoter region of the TRXh gene (Genbank accession

No. D26547, Ishiwatari et al. 1997) was amplified by PCR using two

primers, 5′-CTAGTGGATCCATCTAAAATGGGAATAGAG-3′ and

5′-CTCGGATCCGAATTCCTCGGCGACGAGATC-3′. The ampli-

fied TRXh promoter fragment (PTRXh) was subcloned into the Hin-

dIII-EcoRI sites of pIG121Hm (pBI121 containing a hygromycin-

resistance gene; Ohta et al. 1990), replacing the cauliflower mosaic

virus 35S RNA promoter. The resulting plasmid was termed PTRXh-

GUS (Fig. 1).

The TRXh promoter (between –1025 and –3 bp from the transla-

tion start site ATG) and GFP gene were transcriptionally fused as

follows. The promoter region of the TRXh gene (Genbank accession

No. D26547, Ishiwatari et al. 1997) was amplified by PCR using two

primers, 5′-CTAGTGGATCCATCTAAAATGGGAATAGAG-3′ and

5′-CGGAATTCGGCCGCCATGGCTCTCTCTCCTCC-3′. The ampli-

fied TRXh promoter fragment (PTRXh) was digested with BamHI and

NcoI, and subcloned into the BamHI-NcoI sites of the 35Somega-

sGFP (S65T) plasmid (Chiu et al. 1996), containing the GFP gene and

nopaline synthase terminator (TNOS). The fusion gene PTRXh-GFP-

TNOS was digested with HindIII and EcoRI, and subcloned into the

same sites of pBIN19 (Bevan 1984). The fragment containing PTRXh-

GFP-TNOS was excised using HindIII and SpeI, and inserted into the

corresponding sites of the pTF338 vector, which contained the hygro-

mycin phosphotransferase gene in place of the kanamycin selectable

marker of pBIN19 (Bevan 1984) (Fig. 1).

Transformation of rice plants

The transgenic rice plants were generated by the Agrobacterium

tumefaciens-mediated method (Hiei et al. 1994). Rice (O. sativa L.)

cultivar Sasanishiki and A. tumefaciens strain EHA101 were used to

produce GUS-expressing plants (Goto et al. 1997). Rice cultivar Tsu-

kinohikari and A. tumefaciens strain C58RifR (pGV2260) (Deblaere et

al. 1985) were used to produce GFP-expressing plants. The regener-

ated plants (T0 generation) were grown on sterilized soil in a tempera-

ture controlled (30/25°C, day/night) greenhouse. The progeny seeds of

primary transformed rice plants (T1 generation) were selected on 0.8%

(w/v) agar plates containing 50 mg l–1 hygromycin, Murashige and

Skoog salts and vitamins (Murashige and Skoog 1962), and 30 g l–1

sucrose (pH 5.8). Transgenic plants of the T0 generation and the

hygromycin-resistant T1 generation were used for the experiments.

Histochemical assay for GUS activity

Sections of rice leaves were cut to about 3 mm in length with a

razor blade and fixed in fixation buffer (0.3% formaldehyde, 10 mM

MES (pH 5.6), 0.3 M mannitol) for 45 min. The sections were stained

in staining buffer (50 mM sodium phosphate buffer (pH 7.0), 0.5 mM

potassium ferricyanide, 0.5 mM potassium ferrocyanide, 1 mM 5-

bromo-4-chloro-3-indolyl-β-D-glucuronide) for 24 h at 37°C and

washed in 50 mM sodium phosphate buffer (pH 7.0). The sections

were then refixed in 5% formaldehyde. After dehydration through a

graded ethanol series, the samples were embedded in Technovit 7100

(Heraus Kulzer Co., Wehrheim, Germany) and sliced into 10 µm sec-

tions using a rotary microtome (LR-85, YAMATO KOHKI Industrial

Co., Saitama, Japan).

Microscopic detection of GFP fluorescence of leaf blade

For observation of transverse sections, the leaves were mounted

in 5% (w/v) agar (Nacalai Tesque Co., Kyoto, Japan) and sliced with a

razor blade by hand. GFP fluorescence of these sections was moni-

tored with a video camera attached to a fluorescence microscope sys-

tem: BX50WI, BX-FLA (OLYMPUS Co., Tokyo, Japan). The light

filter cube U-MWIB was used to obtain the microscopic images (460–

490 nm wavelength light path filter for excitation, longer than 515 nm

wavelength light filter for detection).

Collection of phloem sap

Rice phloem sap was collected from stage five to nine leaves of

rice through the cut ends of brown planthopper’s stylets (Kawabe et al.

1980). All procedures for collecting phloem sap were carried out at

25°C and 60% relative humidity under room light conditions (20 µmol

s–1 m–2). Phloem sap samples were stored at –20°C until analysis.

Extraction of leaf proteins

The leaves were homogenized in liquid nitrogen and mixed with

an equal (w/v) amount of extraction buffer (25 mM Tris-HCl (pH 6.8),

5% β-mercaptoethanol, 5 mM p-amidinophenyl methanesulfonyl fluo-

ride and 25 mg ml–1 leupeptin). The homogenate was centrifuged

twice at 12,000×g for 30 min and the supernatant was stored at –20°C

until analysis. Protein Assay Reagent (BIO-RAD Co., CA, USA) was

used for the determination of total protein content.

Production of recombinant proteins

The recombinant GFP used for a positive control in the immuno-

blot analysis was produced in Escherichia coli. A fragment containing

the open reading frame encoding GFP was excised from 35Somega-

sGFP (S65T) (Chiu et al. 1996) with NcoI and EcoRI, and subcloned

into the same sites of pET30b (Novagen Co., Darmstadt, Germany).

The resulting plasmid was used to transform E. coli strain BL21DE

competent cells (Novagen Co.) according to the manufacturer’s

instructions. His-tagged recombinant GFP was produced in trans-

formed E. coli and purified using a His-bind buffer kit following the

manufacturer’s instructions (Novagen Co.). The His-tag was then

cleaved off with enterokinase (Novagen Co.). Recombinant TRXh was

produced as described by Ishiwatari et al. (1995), and recombinant

GST was produced as described by Fukuda et al. (2004a).

Immunoblot analysis

Proteins were separated by SDS-PAGE (Nakamura et al. 1993)

and transferred to polyvinylidene difluoride membranes. Immuno-

blotting was performed to detect GUS, GFP, TRXh, and GST. GUS

was detected with rabbit anti-β-glucuronidase IgG (H+L) fraction

(Molecular Probes Co., OR, USA) and goat anti-rabbit IgG (H+L)

horseradish peroxidase conjugate (BIO-RAD Co., CA, USA) as the

secondary antibody. GFP was detected with mouse monoclonal anti-

GFP IgG (Zymed Co., CA, USA) and anti-mouse IgG horseradish per-

oxidase-linked whole antibody from sheep (Amersham Co., Bucking-

hamshire, UK) as the secondary antibody. TRXh was detected with

rabbit IgGs against recombinant rice TRXh and goat anti-rabbit IgG

Distribution of proteins in phloem tissues1786

(H+L) horseradish peroxidase conjugate (BIO-RAD Co.) as the sec-

ondary antibody. GST was detected with rabbit IgGs against recombin-

ant GST as described by Fukuda et al. (2004a). Diaminobenzidine was

used as the substrate for detection. The intensity of bands detected in

the immunoblot analysis was measured using NIH-image software

(National Institute of Health, MD, USA).

References

Balachandran, S., Xiang, Y., Schobert, C., Thompson, G.A. and Lucas, W.J.

(1997) Phloem sap proteins from Cucurbita maxima and Ricinus communis

have the capacity to traffic cell to cell through plasmodesmata. Proc. Natl

Acad. Sci. USA 94: 14150–14155.

Bevan, M.W. (1984) Binary Agrobacterium vectors for plant transformation.

Nucleic Acids Res. 12: 8711–8721.

Bostwick, D.E., Dannenhoffer, J.M., Skaggs, M.I., Lister, R.M., Larkins, B.A.

and Thompson, G.A. (1992) Pumpkin phloem lectin genes are specifically

expressed in companion cells. Plant Cell 4: 1539–1548.

Chiu, W., Niwa, Y., Zeng, W., Hirano, T., Kobayashi, H. and Sheen, J. (1996)

Engineered GFP as a vital reporter in plants. Curr. Biol. 6: 325–330.

Chonan, N., Kaneko, M., Kawahara, H. and Matsuda, T. (1981) Ultrastructure

of the large vascular bundles in the leaves of rice plants. Jpn. J. Crop Sci. 50:

323–331.

Clark, A.M., Jacobson, K.R., Bostwick, D.E., Dannenhoffer, J.M., Skaggs, M.I.

and Thompson, G.A. (1997) Molecular characterization of a phloem-specific

gene encoding the filament protein, phloem protein 1 (PP1), from Cucurbita

maxima. Plant J. 12: 49–61.

Cronshaw, J. (1981) Phloem structure and function. Annu. Rev. Plant Physiol.

32: 465–484.

Dannenhoffer, J.M., Schulz, A., Skaggs, M.I., Bostwick, D.E. and Thompson,

G.A. (1997) Expression of the phloem lectin is developmentally linked to

vascular differentiation in cucurbits. Planta 201: 405–414.

Dannenhoffer, J.M., Suhr, R.C. and Thompson, G.A. (2001) Phloem-specific

expression of the pumpkin fruit trypsin inhibitor. Planta 212: 155–162.

Deblaere, R., Bytebier, B., De Greve, H., Deboeck, F., Van Montagu, M. and

Leemans, J. (1985) Efficient octopine Ti plasmid-derived vectors for Agro-

bacterium-mediated gene transfer to plants. Nucleic Acids Res. 13: 4777–

4788.

Eschrich, W. and Heyser, W. (1975) Biochemistry of phloem constituents. In

Transport in Plants I. Phloem Transport. Edited by. Zimmermann, M.H. and

Milburn, J.A. pp. 101–136. Springer-Verlag, Berlin, Heidelberg, New York.

Fisher, D.B. and Cash-Clark, C.E. (2000) Sieve tube unloading and post-phloem

transport of fluorescent tracers and proteins injected into sieve tubes via sev-

ered aphid stylets. Plant Physiol. 123: 125–137.

Fisher, D.B., Wu, Y. and Ku, M.S.B. (1992) Turnover of soluble proteins in the

wheat sieve tube. Plant Physiol. 100: 1433–1441.

Fujimaki, S., Fujiwara, T. and Hayashi, H. (2000) A new method for direct

introduction of chemicals into a single sieve tube of intact rice plants. Plant

Cell Physiol. 41: 124–128.

Fukuda, A., Okada, Y., Suzui, N., Fujiwara, T., Yoneyama, T. and Hayashi, H.

(2004a) Cloning and characterization of the gene for a phloem-specific glu-

tathione S-transferase from rice leaves. Physiol. Plant. 120: 595–602.

Fukuda, A., Okada, Y., Suzui, N., Fujiwara, T., Yoneyama, T. and Hayashi, H.

(2004b) Cloning of the phloem-specific small heat-shock protein from leaves

of rice plants. Soil Sci. Plant Nutr. 50: 1255–1262.

Goto, S., Ishii, Y., Hayakawa, T. and Yamaya, T. (1997) Agrobacterium-medi-

ated transformation of Sasanishiki, a leading cultivar of rice (Oryza sativa L)

in northern Japan. In Plant Nutrition – For Sustainable Food Production and

Environment. Edited by Ando, T., Fujita, K., Mae, T., Matsumoto, H., Mori,

S. and Sekiya J. pp. 839–840. Kluwer Academic Publishers, Dordrecht, The

Netherlands.

Hayashi, H., Fukuda, A., Suzui, N. and Fujimaki, S. (2000) Proteins in the sieve

element-companion cell complexes: their detection, localization and possible

functions. Aust. J. Plant Physiol. 27: 489–496.

Hiei, Y., Ohta, S., Komari, T. and Kumashiro, T. (1994) Efficient transforma-

tion of rice (Oryza sativa L.) mediated by Agrobacterium and sequence anal-

ysis of the boundaries of the T-DNA. Plant J. 6: 271–282.

Imlau, A., Truernit, E. and Sauer, N. (1999) Cell-to-cell and long-distance traf-

ficking of the green fluorescent protein in the phloem and symplastic unload-

ing of the protein into sink tissues. Plant Cell 11: 309–322.

Ishiwatari, Y., Fujiwara, T., McFarland, K.C., Nemoto, K., Hayashi, H., Chino,

M. and Lucas, W.J. (1998) Rice phloem thioredoxin h has the capacity to

mediate its own cell-to-cell transport through plasmodesmata. Planta 205:

12–22.

Ishiwatari, Y., Hayashi, H., Fujiwara, T. and Chino, M. (1997) Isolation of

genomic DNA fragment of rice phloem thioredoxin h. In Plant Nutrition –

For Sustainable Food Production and Environment. Edited by Ando, T.,

Fujita, K., Mae, T., Matsumoto, H., Mori, S. and Sekiya J. pp. 809–810.

Kluwer Academic Publishers, Dordrecht, The Netherlands.

Ishiwatari, Y., Honda, C., Kawashima, I., Nakamura, S., Hirano, H., Mori, S.,

Fujiwara, T., Hayashi, H. and Chino, M. (1995) Thioredoxin h is one of the

major proteins in rice phloem sap. Planta 195: 456–463.

Ishiwatari, Y., Nemoto, K., Fujiwara, T., Chino, M. and Hayashi, H. (2000) In

situ hybridization study of the rice phloem thioredoxin h mRNA accumula-

tion – possible involvement in the differentiation of vascular tissues. Physiol.

Plant. 109: 90–96.

Kawabe, S., Fukumorita, T. and Chino, M. (1980) Collection of rice phloem sap

from stylets of homopterous insects severed by YAG laser. Plant Cell

Physiol. 21: 1319–1327.

Kempers, R. and van Bel, A.J.E. (1997) Symplasmic connections between sieve

element and companion cell in the stem phloem of Vicia faba L. have a

molecular exclusion limit of at least 10 kDa. Planta 201: 195–201.

Kragler, F., Monzer, J., Shash, K., Xoconostle-Cázares, B. and Lucas, W.J.

(1998) Cell-to-cell transport of proteins: requirement for unfolding and char-

acterization of binding to a putative plasmodesmal receptor. Plant J. 15: 367–

381.

Kühn, C., Franceschi, V.R., Schulz, A., Lemoine, R. and Frömmer, W.B. (1997)

Macromolecular trafficking indicated by localization and turn over of sucrose

transporters in enucleate sieve elements. Science 275: 1298–1300.

Lucas, W.J., Ding, B. and van der Schoot, C. (1993) Plasmodesmata and the

supercellular nature of plants. New Phytol. 125: 435–476.

Murashige, T. and Skoog, F. (1962) A revised medium for rapid growth and bio-

assays with tobacco tissue cultures. Physiol. Plant. 15: 473–497.

Nakamura, S., Hayashi, H., Mori, S. and Chino, M. (1993) Protein phosphoryl-

ation in the sieve tubes of rice plants. Plant Cell Physiol. 34: 927–933

Ohta, S., Mita, S., Hattori, T. and Nakamura, K. (1990) Construction and

expression in tobacco of a β-glucuronidase (GUS) reporter gene containing

an intron within the coding sequence. Plant Cell Physiol. 31: 805–813.

Sakuth, T., Schobert, C., Pecsvaradi, A., Eichholz, A., Komor, E. and Orlich, G.

(1993) Specific proteins in the sieve-tube exudate of Ricinus communis L.

seedlings: separation, characterization and in-vivo labelling. Planta 191:

207–213.

Schobert, C., Baker, L., Szederkényi, J., Großmann, P., Komor, E., Hayashi, H.,

Chino, M. and Lucas, W.J. (1998) Identification of immunologically related

proteins in sieve-tube exudates collected from monocotyledonous and dicoty-

ledonous plants. Planta 206: 245–252.

Shi, Y., Wang, M.B., Powell, K.S., Van-Damme, E., Hilder, V.A., Gatehouse,

A.M.R., Boulter, D. and Gatehouse, J.A. (1994) Use of the rice sucrose

synthase-1 promoter to direct phloem-specific expression of β-glucuronidase

and snowdrop lectin genes in transgenic tobacco plants. J. Exp. Bot. 45: 623–

631.

Waigmann, E. and Zambryski, P. (1995) Tobacco mosaic virus movement pro-

tein-mediated protein transport between trichome cells. Plant Cell 7: 2069–

2079.

Xoconostle-Cázares, B., Xiang, Y., Ruiz-Medrano, R., Wang, H.-L., Monzer, J.,

Yoo, B.-C., McFarland, K.C., Franceschi, V.R. and Lucas, W.J. (1999) Plant

paralog to viral movement protein that potentiates transport of mRNA into

the phloem. Science 283: 94–98.

(Received May 13, 2005; Accepted August 15, 2005)