Developmental Cell

Article

Callose Biosynthesis Regulates SymplasticTrafficking during Root DevelopmentAnne Vaten,1 Jan Dettmer,1 Shuang Wu,2,10 York-Dieter Stierhof,3,10 Shunsuke Miyashima,1,10 Shri Ram Yadav,1

Christina J. Roberts,4 Ana Campilho,1 Vincent Bulone,5 Raffael Lichtenberger,1 Satu Lehesranta,1 Ari Pekka Mahonen,1,6

Jae-Yean Kim,7 Eija Jokitalo,1 Norbert Sauer,8 Ben Scheres,6 Keiji Nakajima,9 Annelie Carlsbecker,1,4,11

Kimberly L. Gallagher,2,11 and Yka Helariutta1,*1Institute of Biotechnology/Department of Bio and Environmental Sciences, University of Helsinki, FIN-00014, Finland2Department of Biology, University of Pennsylvania, Philadelphia, PA 19104, USA3Microscopy Unit, Center for Plant Molecular Biology, University of Tubingen, 72076 Tubingen, Germany4Department of Physiological Botany, Evolutionary Biology Centre, Uppsala University, SE-752 36 Uppsala, Sweden5Division of Glycoscience, School of Biotechnology, Royal Institute of Technology, SE–106 91 Stockholm, Sweden6Department of Molecular Genetics, University of Utrecht, 3584 CH Utrecht, The Netherlands7Division of Applied Life Science (BK21 and WCU Program), PMBBRC and EB-NCRC, Gyeongsang National University,

Jinju 660–701, Republic of Korea8Friedrich-Alexander-University Erlangen-Nuremberg, Molecular Plant Physiology, Erlangen D-91058, Germany9Graduate School of Biological Sciences, Nara Institute of Science and Technology, 8916-5 Takayama, Ikoma, Nara 630-0192 Japan10These authors contributed equally to this work11These authors contributed equally to this work

*Correspondence: yrjo.helariutta@helsinki.fi

DOI 10.1016/j.devcel.2011.10.006

SUMMARY

Plant cells are connected through plasmodesmata(PD), membrane-lined channels that allow symplas-tic movement of molecules between cells. However,little is known about the role of PD-mediatedsignaling during plant morphogenesis. Here, wedescribe an Arabidopsis gene, CALS3/GSL12.Gain-of-function mutations in CALS3 result inincreased accumulation of callose (b-1,3-glucan) atthe PD, a decrease in PD aperture, defects in rootdevelopment, and reduced intercellular trafficking.Enhancement of CALS3 expression during phloemdevelopment suppressed loss-of-functionmutationsin the phloem abundant callose synthase, CALS7indicating that CALS3 is a bona fide callose synthase.CALS3 alleles allowed us to spatially and temporallycontrol the PD aperture between plant tissues. Usingthis tool, we are able to show that movement of thetranscription factor SHORT-ROOTandmicroRNA165between the stele and the endodermis is PD depen-dent. Taken together, we conclude that regulatedcallose biosynthesis at PD is essential for cellsignaling.

INTRODUCTION

During the last two decades, an impressive array of molecular

signals have been identified that control plant development by

moving between cells. For example, the cellular pattern of the

Arabidopsis root is maintained by a group of pluripotent cells in

the root meristem that contact the quiescent center cells (QC)

(Dolan et al., 1993; van den Berg et al., 1997) (Figure 1D). These

1144 Developmental Cell 21, 1144–1155, December 13, 2011 ª2011

cells communicate with the surrounding meristematic cells by

various mobile signals, including the phytohormone auxin and

mobile proteins/peptides (Sabatini et al., 1999; Stahl et al.,

2009; Schlereth et al., 2010; Matsuzaki et al., 2010). In addition,

mobile signals move from the stele to the QC to maintain stem

cell fate; the SHORT-ROOT (SHR) transcription factor (Nakajima

et al., 2001) moves from the stele into the neighboring QC and

endodermis where it serves as a cell fate determinant. SHR is

also required in the endodermis to turn on microRNA165/6

(miRNA165/6 or miR165/6) (Carlsbecker et al., 2010). Movement

of the miR165/6 species into the stele is required for the down-

regulation of their targets, the class III homeodomain leucine

zipper (HD-ZIP III) transcripts including PHABULOSA (PHB)

(McConnell et al., 2001; Emery et al., 2003; Prigge et al., 2005),

which in turn is necessary for the radial patterning of the xylem

tissue and the surrounding pericycle (Carlsbecker et al., 2010;

Miyashima et al., 2011). In addition to these known signals, there

appear to be many highly regulated cell proliferation and speci-

fication events that rely on the movement of signals whose

molecular context is not yet known.

It is largely unknown how proteins and miRNA signals move

between cells. In principle two routes are possible in plants.

The first is through the process of exo- and endocytosis. In this

scheme cargo molecules are targeted to secretory vesicles

that fuse with the plasma membrane (PM) and release their

contents into the extracellular space (An et al., 2006; Zarsky

et al., 2009). These molecules are then endocytosed by neigh-

boring cells. The second is through direct symplastic transport

of cytoplasmic molecules via plasmodesmata (PD) (Erwee and

Goodwin, 1985; Botha et al., 1993; Ehlers and Kollmann, 2001)

(Figure 1C). PD are membrane lined channels that connect

neighboring cells creating a cytoplasmic sleeve through which

small molecules can diffuse and continuity between the endo-

plasmic reticulum and plasma membranes of neighboring cells.

It has been postulated that PD evolved to facilitate communica-

tion between plant cells otherwise separated by rigid cell walls

Elsevier Inc.

Figure 1. The cals3-d Mutations Impair Phloem Unloading

(A–D) Schematic presentation of the Arabidopsis root highlighting phloem (A). Companion cells (CC) and sieve elements (SE) and their symplastic connections;

the SE form a cell file connected by sieve plates which are enriched with callose (marked with blue) (B). PD containing central desmotubule (DT) and callose

(marked with blue) in its neck region (C). Currently known non-cell-autonomous signaling pathways in the root stele and the surrounding tissues (D). Red arrows in

B and C indicate non-cell-autonomous signals.

(E) Seedling phenotype of 8-day-old wild-type (C24), cals3-1d, -2d, and -3d.

(F–H) pSUC2::GFP in wild-type C24 (F), cals3-1d (G), cals3-1d /+ (H).

(I) pSUC2::GFP-sporamin in wild-type C24.

(J and K) pSUC2::GFP in the cals3-1d transformed with the genomic insert containing At5g13000 locus (J), and in wild-type C24 transformed with the genomic

insert containing the At5g13000 locus with the 1d and 2d mutations (K).

(L) Schematic view of the intron-exon structure of the CALS3 and sites of the mutations/T-DNA insertions.

Scale bars in (E), 1 cm; in (F)–(K), 50 mm; dashed boxes; postphloem unloading domain. See also Figure S1.

Developmental Cell

Callose Biosynthesis Regulates Cell Signaling

(Zambryski and Crawford, 2000). Plant viruses use PD to spread

between cells and colonize their host (Wolf et al., 1989; Citovsky

et al., 1990; Waigmann et al., 1994; Benitez-Alfonso et al., 2010).

Some endogenous plant proteins have also been shown to traffic

via PD (Aoki et al., 2002; Lucas and Lee, 2004). Regulation of PD

permeability is developmentally controlled (Kim et al., 2002);

however, studies on the effects of PD misregulation on plant

development are largely lacking. PD have the ability to regulate

the size of their cytoplasmic sleeve and hence to fine tune inter-

cellular trafficking (Sivaguru et al., 2000; Stadler et al., 2005;

Benitez-Alfonso et al., 2009; Simpson et al., 2009). How this

regulation is orchestrated is not well understood, although

callose deposition has been shown to affect cell-to-cell

signaling. Callose degradation enhances cell-to-cell movement

(Bucher et al., 2001; Rinne et al., 2005). In contrast, impaired traf-

ficking results from increased callose accumulation at the PD

(Sivaguru et al., 2000; Simpson et al., 2009) or loss of enzymes

such as b-1,3-glucanases (Iglesias and Meins, 2000; Levy

et al., 2007), which degrade callose. While these results indicate

Developmenta

an important role for callose in regulating the PD aperture or size

exclusion limit until now the regulatory mechanism based on cal-

lose biosynthesis has remained poorly characterized (Guseman

et al., 2010).

Here, we report that the regulated activity of a member of

the glycosyltransferase 48 family, CALLOSE SYNTHASE

3/GLUCAN SYNTHASE-LIKE12 (CALS3/GSL12, At5g13000),

determines PD aperture and regulates molecular trafficking by

callose biosynthesis in a cell wall domain surrounding the PD.

CALS3 is expressed in the stele and in the root meristem during

plant development. It was identified based on three allelic gain-

of-function mutations. CALS3 can suppress loss-of-function

mutations in another related locus, CALLOSE SYNTHASE

7/GLUCAN SYNTHASE-LIKE7 (CALS7/GSL7, At1g06490),

which is required for callose biosynthesis during phloem devel-

opment (Xie et al., 2011; Barratt et al., 2011). CALS3 localizes

to the PD, consistent with its biosynthetic role in the regulation

of PD aperture. Using conditional gain-of-function mutations in

CALS3 we were able to spatially and temporally reduce the PD

l Cell 21, 1144–1155, December 13, 2011 ª2011 Elsevier Inc. 1145

Figure 2. The cals3-d Mutations Lead to Enhanced Callose

Biosynthesis

(A and B) Subcellular localization of the GFP-CALS3 protein in the cotyledon

cell before (A), and after (B) plasmolysis. Scale bar, 25 mm.

(C) Localization of the pCALS3::GFP in wild-type root tip.

(D–G) Aniline blue stained primary root of the wild-type C24 (D), young lateral

root of wild-type C24 (E), and cals3-2d (F), and primary root of the transgenic

line with pCALS3::icals3m after 17 hr induction (G). Panel d is inset from the (D).

Red arrowhead, phloem sieve plate; white arrowhead, punctate staining of the

SE lateral walls; white arrows, plasmodesmal localization; open arrows,

localization at the plasmamembrane; white asterisks, endodermis; red aster-

isks, ectopic callose deposition; scale bars, 50 mm. See also Figure S2.

Developmental Cell

Callose Biosynthesis Regulates Cell Signaling

aperture and thereby demonstrate that both the SHR protein and

miR165 move symplastically. Our results indicate that spatial

and temporal control of callose biosynthesis regulates the

passage of signaling molecules through the PD during plant

development.

RESULTS

Gain-of-Function Mutations in CALS3 EncodingPD-Localized Callose Synthase Lead to DecreasedMolecular TraffickingWe performed two independent genetic screens to identify

mutants with altered stele development (Carlsbecker et al.,

2010). In wild-type plants, small proteins (such as GFP) driven

by the companion cell (CC)-specific SUC2 promoter traffic

through the phloem and freely diffuse throughout the entire

root tip (Figures 1A, 1B, and 1F) (Imlau et al., 1999). We identified

three allelic semidominant mutants that we call cals3-1d, �2d,

and -3d and collectively refer to as cals3-d (Figure 1E; see

Figure S1 available online). These plants were identified in part

based upon an aberrant pattern of pSUC2::GFP localization.

The cals3-d showed a weak, fragmented pattern of pSUC2::GFP

in the upper part of the root and no GFP signal in the elongation

and meristematic zones of root tip (Figure 1G) indicating defec-

tive establishment of phloem identity and decreased diffusion of

GFP in cals3-d. All of the cals3-d had shorter roots thanwild-type

(Figure 1E); we considered cals3-2d as the most severe allele

as it had the shortest root (Figure 1E). In heterozygous cals3-d

mutants where root length was normal, GFP (27 kDa) moved

only in the postphloem domain of the root (Stadler et al., 2005)

(Figures 1A and 1H) and did not diffuse throughout the root tip.

This unloading pattern is similar to that of the larger

pSUC2::GFP-sporamin (47 kDa) in wild-type (Figure 1I). These

results suggest that the PD of heterozygous cals3-d mutants

have a decreased size exclusion limit of the PD in the post-

phloem domain.

To analyze the cals3-dmutations at amolecular level, wemap-

ped them to the At5g13000 locus (Figure 1L), which encodes

a putative callose synthase (Hong et al., 2001), CALLOSE

SYNTHASE3 (CALS3)/GLUCAN SYNTHASE-LIKE12 (GSL12).

We cloned the corresponding cDNA that codes for a putative

protein with 1947 amino acid residues (GenBank accession

number HM594867, Figure 1L). Consistent with the semidomi-

nant inheritance, we were able to partially suppress the homozy-

gous cals3-1d and -2d phenotypes with a 13 kb genomic

fragment containing the wild-type CALS3 gene (Figure 1J).

Furthermore, we added the 1d and 2d mutations to the afore-

mentioned fragment (to give cals3m). Transgenic plants contain-

ing the cals3m fragment displayed partial phenocopy of cals3-d

phenotype (Figure 1K). The cals3-1d, cals3-2d, and cals3-3d

mutations cause, respectively, R84K, R1926K, and P189L amino

acid changes in the deduced amino acid sequence (see Supple-

mental Experimental Procedures). All three amino acid changes

are in the predicted cytosolic face of CALS3, which suggests

a possible shared mechanism in modulating CALS3 function.

To determine the subcellular localization of CALS3 we consti-

tutively expressed a GFP-CALS3 fusion (that partially sup-

pressed root elongation in cals3-1d, data not shown) in

Arabidopsis. Fluorescence was detected preferentially at cell

1146 Developmental Cell 21, 1144–1155, December 13, 2011 ª2011

wall punctae, presumably at PD, based on the previous reports

with similar pattern (e.g., Thomas et al., 2008; Simpson et al.,

2009) and in the cell wall and at the PMof plasmolyzed cotyledon

epidermal cells (Figures 2A and 2B). In order to study whether

cals3-1d mutations change the subcellular localization of the

CALS3, we generated 35S::GFP-cals3-1d transgenic plants

(Figure S2A). The localization of the GFP-CALS3-1D protein

was indistinguishable from wild-type CALS3.

Next, we studied CALS3 transcript accumulation in wild-type

shoots and roots using RT-PCR. We found that CALS3 mRNA

is equally abundant both in shoot and root (Figure S2B).

CALS3 is predicted to be broadly expressed in the root (Brady

et al., 2007). In order to define the domain of CALS3 expression

more closely, we performed in situ localization of CALS3 mRNA

and created a transcriptional reporter containing the 1.5 kb

Elsevier Inc.

Figure 3. Tissue-Specific Induction of Callose Biosynthesis by

cals3m

(A–C) Aniline blue stained wild-type (Col-0) root after 0 hr induction (A),

transgenic line with p6xUAS::icals3m, J0571 after a 4 hr induction (B) and with

pCRE1::icals3m after a 6 hr induction (C).

(D–F) Immunofluorescence labeling of callose in ultrathin resin sections

obtained from cryofixed and freeze-substituted root tips. Wild-type (Col-0)

root after 0 hr induction (D), transgenic line with p6xUAS::icals3m, J0571 after

a 5 hr induction (E) and with pCRE1::icals3m after a 8 hr induction (F). The

boxed area is three times enlarged. DNA is stained with DAPI. Resolution in z

direction in ultrathin resin section presented in (D)–(F) is in the range of few

nanometers. The signal intensity in the images of the transgenic lines was

reduced to prevent signal saturation in the induced regions. Signal intensity in

wild-type and in noninduced regions of the transgenic lines did not differ.

e, endodermis; p, pericycle; c, cortex; red asterisks, ectopic callose deposi-

tion. Scale bars, 50 mm (A–C), 20 mm (D–F). See also Figure S3.

Developmental Cell

Callose Biosynthesis Regulates Cell Signaling

sequence upstream of theCALS3 translation initiation site (Dong

et al., 2008) fused to GFP andGUS.With bothmethodswe found

thatCALS3 is expressed in the root meristem (Figure 2C; Figures

S2C and S2D). The transcriptional reporter highlighted strong

expression domains throughout the seedling in the stele (espe-

cially in the phloem) and in the QC and cells surrounding it.

In addition, we studied CALS3 transcript levels in the roots of

cals3-d mutants using RT-PCR. We found that cals3-d mutants

accumulate similar amounts of CALS3 transcripts to wild-type

(Figure S2E). Taken together, these results indicate that cals3-d

mutations neither affect the transcript levels of the CALS3 gene

nor alter the subcellular localization of the CALS3 protein.

The cals3-d Mutations Lead to ReversibleOverproduction of CalloseIn order to approach the effect of the cals3-d mutations on

callose biosynthesis, we examined callose levels in the root

meristem and stele based upon aniline blue staining (Ingram

et al., 2011). A strong signal was detected in the phloem tissue

of wild-type seedlings (Figure 2D). The sieve plates, callose en-

riched structures connecting conducting phloem sieve elements

(SE) (Figure 1B), were clearly identified with a strong, even stain-

ing pattern (Figure 2D). The lateral walls of protophloem SEs

were also intensely stained, but unlike sieve plates these walls

displayed punctate staining that could be interpreted as being

located at PD. In addition to the phloem, the transverse walls

of most root cells displayed strong staining. In order to under-

stand the effect of the cals3-d mutations on callose accumula-

tion, we analyzed the callose staining in the primary roots of

the cals3-dmutants. However, we observed no significant differ-

ences in the pattern of aniline blue staining between the primary

roots of mutant and wild-type seedlings. Therefore, we next

observed the early stages of lateral root development. Young

lateral roots of cals3-d typically displayed slightly stronger stain-

ing than wild-type (Figures 2E and 2F). Mature lateral roots did

not show increased callose accumulation and eventually

displayed a staining pattern similar to primary roots. Hence,

the pattern of increased callose deposition in cals3-d lateral

roots gradually disappeared with age.

To further characterize the nature of the cals3-dmutations and

their effect on callose accumulation, we engineered lines that

express ectopically CALS3 upon induction (Zuo et al., 2000). In

these lines, the expression of the wild-type CALS3, the cals3-1d

version of CALS3 or the CALS3 carrying both the cals3-1d and

cals3-2d mutations (icals3m) were induced in a tissue-specific

manner. Induced wild-type CALS3 and noninduced pCALS3::

icals3m showed a similar pattern of callose accumulation as

induced or noninduced wild-type (Figures S2F and S2G), but

lead to only a modest increase in callose accumulation (Figures

S2H and S2J). Upon induction, expression of pCALS3::icals3m

resulted in high levels of callose accumulation in the cell walls

of cells in which the promoter was active (Figure 2G; Figure S2I).

Induced expression of the pCALS3::icals3-1d mutant version

increased levels of callose formation above wild-type

(Figure S2K), but not as much as pCALS3::icals3m expression

indicating that the cals3-1d and cals3-2d mutations caused an

additive effect. Taken together, this suggests that the cals3-d

mutations disturb regulated callose synthesis, resulting in

elevated callose accumulation. Consistent with these observa-

Developmenta

tions, the level of callose synthase activity, measured in vitro,

was 10%–50% higher in tissue extracts prepared from the roots

of the pCALS3::icals3m plants induced for 12 hr compared with

noninduced plants (Table S1).

Controlled Expression of cals3m Leads toOverproduction of PD-Localized Callose and DecreasedPD ApertureNext, we examined the effect of the cals3-d mutations specifi-

cally on the deposition of callose at PD. We expressed icals3m

under the stele-specific CRE1 promoter (Bonke et al., 2003)

and monitored callose accumulation in the stele cells. We used

J0571; p6xUAS::icals3m (J0571, http://www.plantsci.cam.ac.

uk/Haseloff/) as negative control line for stele expression as

J0571 is ground tissue-specific driver. Aniline blue staining and

immunofluorescence labeling of callose showed in wild-type

an even pattern of callose accumulation throughout the root tip

(Figures 3A and 3D). Upon induction, modest levels of callose

were observed in the stele cells of the ground tissue-specific

driver line, J0571; p6xUAS::icals3m (Figures 3B and 3E). In

contrast, stele cell walls displayed strong punctate callose

l Cell 21, 1144–1155, December 13, 2011 ª2011 Elsevier Inc. 1147

Figure 4. Stele-Specific Induction of cals3m Leads to Highly

Elevated Callose Levels at the Zone Surrounding PD Thereby

Decreasing PD Aperture

(A and B) Immunogold labeling (ultrathin resin sections) of callose in wild-type

PD (A), and in transgenic line with pCRE1::icals3m after an 8 hr induction (B)

(Cryofixed and freeze-substituted sample).

(C and D) Ultrastructural analysis of the PD channel in wild-type (C) and in

transgenic line with pCRE1::icals3m after 8 hr induction (D) (Cryofixed and

freeze-substituted sample).

(E) Quantitative analysis of gold labeled callose surrounding PD in stele cells of

the transgenic lines with pCRE1::icals3m after 8 hr induction (n = 21) and with

J0571, p6xUAS::icals3m (n = 21) after 5 hr induction. Data are shown as

mean ± SEM. A 100 nm zone and a 200 nm zones around PD were analyzed.

(F) Quantitative analysis of the PD aperture in the stele of the wild-type Col-0

(n = 47) and pCRE1::icals3m (n = 119) after 24 hr induction.

Data are shown as mean ± SEM. Scale bar, 200 nm; arrowheads, PD channel;

asterisks, swollen cell wall microdomain; CW, cell wall. See also Figure S4.

Developmental Cell

Callose Biosynthesis Regulates Cell Signaling

deposition in pCRE1::icals3m line (Figures 3C and 3F). Taken

together, callose accumulation by icals3m can be controlled

tissue specifically.

To more specifically localize callose in the stele cells of these

two transgenic lines, we performed immunogold labeling on

resin embedded sections of cryofixed root tips. Upon 5 hr induc-

tion in J0571; p6xUAS::icals3m most of the gold particles in the

stele cells were localized in clusters in the cell wall (Figure 4A).

These clusters were often detected in the vicinity of PD, presum-

ably in the apoplastic space. Whereas after 8 hr of pCRE1::

icals3m induction, dramatically more gold particles were

observed in cell walls between stele cells compared with the

J0571; p6xUAS::icals3m (Figure 4B; Figure S4A). The general

pattern of callose distribution at the PM of the stele cells was

not significantly changed (Figures S4A–S4F), but strong callose

deposition was often localized in swollen microdomains

surrounding the PD (Figure 4B; Figures S4B–S4D). In the

pCRE1::icals3m roots 80.4% of the PD were located in swollen

microdomains with enhanced gold particle deposition (n = 92),

whereas in the J0571; p6xUAS::icals3m roots we did not

observed callose enriched swollen microdomains in the stele

cells (n = 71). Furthermore, three times more gold particles

were located in the 200 nm zone surrounding PD of the stele in

pCRE1::icals3m compared to J0571; p6xUAS::icals3m, indi-

cating that overexpression of cals3-d leads to increased produc-

tion of callose, which was observed at PD (Figure 4E). This was

also supported by our observation of a partial rescue of root

development in cals3-1d seedlings when specifically degrading

callose at PD, by driving a b-1,3-glucanase (At5g42100) under

the control of the CALS3 promoter (Figure S3A).

Next, we measured the PD aperture in the stele following

pCRE1::icals3m induction. The aperture was significantly

decreased upon induction (Figures 4C, 4D, and 4F). After 24 hr

of induction the PD aperture was 24% narrower in pCRE1::

icals3m line than in wild-type. Furthermore, when we expressed

icals3m under the phloem-specific APL promoter we observed

phloem-specific callose accumulation (Figure S3B) accompa-

nied by impaired unloading of pSUC2::GFP to the root tip (Fig-

ure S3C). In addition, the phloem differentiation zone was

expanded toward the root tip accompanied by reduced cell divi-

sion zone (Figures S3D and S3E). Taken together, these results

suggest that the cals3-dmutations lead to increased production

of callose and as a result a reduced PD aperture that impairs

intercellular trafficking.

CALS3 Can Substitute for Phloem-Specific CALS7

Disruption of CALS3 function by a T-DNA insertion in the gene

had no effect on callose accumulation or root development

(cals3-4, cals3-5, and cals3-6) (Figure 1L), consistent with the

high degree of genetic redundancy in the GT48 gene family

(Brady et al., 2007). Hence, we looked for other phloem abundant

members in this family. We noticed that CALS7 (At1g06490) is

predicted to be phloem specific (Brady et al., 2007) as recently

confirmed by two other studies (Xie et al., 2011; Barratt et al.,

2011). Analysis of CALS7mRNA localization by in situ hybridiza-

tion indicated that it is strongly and specifically expressed in

developing protophloem SE cells (Figure 5A). We next observed

expression of a transcriptional reporter containing the 3 kb

genomic region upstream of the CALS7 translation initiation

1148 Developmental Cell 21, 1144–1155, December 13, 2011 ª2011

site fused to YFP. This revealed specific expression in phloem

SEs throughout the length of the root (Figure 5B).

To determine the role of CALS7 in phloem-specific callose

synthesis, we analyzed a T-DNA insertion line of CALS7,

cals7-1. As recently reported by Xie et al. (2011) and Barratt

et al. (2011), the phenotype of these plants was distinguishable

from wild-type only in that their apical growth was eventually

compromised. The mutant line displayed a callose pattern

similar to wild-type in tissues outside the phloem, but had

dramatically less callose in the phloem SEs (Figures 5C and

5E). Here, very little callose was detected at the presumptive

PD of the lateral walls of the protophloem SEs and at the sieve

plates of the SEs.

To explore whether CALS3 can suppress the callose deposi-

tion defect of cals7-1 during phloem development, we intro-

duced pAPL::iCALS3 to cals7-1. We were able to partially

restore the callose accumulation pattern in the cals7-1 phloem,

thus demonstrating a role for CALS3 as a callose synthase

Elsevier Inc.

Figure 5. CALS3 Suppresses Decreased Callose

Levels in cals7-1 Phloem

(A) In situ hybridization with CALS7 mRNA-specific probe

on cross sections of the root meristem.

(B) Expression of the pCALS7::YFP in the differentiated

stele.

(C) Aniline blue stained cals7-1 after 0 hr induction.

(D) Analysis of the callose deposition during phloem

development in two independent pAPL::iCALS3 lines in

the cals7-1 background (T2 generation, Line 1: n = 50, Line

2: n = 47) after 18 hr induction. p = 0.0003 by Pearson’s

Chi-square test. **p < 0.01.

(E–G), Aniline blue stained phloem cells in the cals7-1 (E),

transgenic cals7-1 line with pAPL::iCALS3 (F), and trans-

genic cals7-1 line with pAPL::icals3m (G) after 18 hr

induction. We observed a specific effect of estradiol in

slightly promoting callose formation in cals7-1 back-

ground. To ignore this effect, we used specifically here

one-tenth of the estradiol concentration used in other

induction experiments.

Arrowheads, protophloem SE; arrows, xylem axis; red

arrows, callose deposition in the phloem; scale bars,

50 mm.

Developmental Cell

Callose Biosynthesis Regulates Cell Signaling

(Figures 5D, 5E, and 5F). Furthermore, the introduction of

pAPL::icals3m to cals7-1 resulted in massive accumulation of

callose in the phloem of the mutant (Figure 5G), in line with our

observations of the effect of icals3m in wild-type background

(Figure S3B).

SHR and miR165 Move via the Symplastic PathwayIn addition to impaired formation of the phloem, xylem identity

was unstable in the cals3-d mutants. In wild-type, protoxylem

cells are distinguished by their position in the stele and their

ring-like cell wall thickenings (open arrowheads in Figure 6A).

Protoxylem cells are spatially distinct from the central meta-

xylem, which has more complete cell wall thickenings (arrow-

heads in Figure 6A). The cals3-d mutants displayed a loss of

protoxylem cells, which was accompanied by ectopic differenti-

ation ofmetaxylem in the protoxylem position (Figure 6B; Figures

S5A and S5B). In addition, cell files with characteristics of both

protoxylem andmetaxylem cell typeswere observed (Figure 6C).

Consistent with changes in xylem identities, cals3-d mutants

display a weak and fragmented expression pattern of the

protoxylemmarker, pAHP6::GFP (Mahonen et al., 2006) (Figures

S5C and S5D).

The xylem phenotype of the cals3-dmutants strikingly resem-

bled that of the miR165/6 resistant mutations in the PHB locus

(Carlsbecker et al., 2010). PHB is normally expressed in the

central domain of the stele, such as metaxylem and posttran-

scriptionally controlled by miR-mediated degradation in proto-

xylem. Hence, mutations in the miR165/6 target site results in

a PHB transcript that is resistant to degradation by miR165/6

and expansion of the PHB mRNA domain into the protoxylem.

It has been recently shown that the miR165/6 species are tran-

scribed outside of the stele in the neighboring endodermis,

and that miR165/6 moves from the endodermis into the stele

Developmenta

(Carlsbecker et al., 2010; Miyashima et al., 2011). The transcrip-

tion of themiRNA165/6 species in the endodermis requires SHR,

which itself moves there from the stele (Carlsbecker et al., 2010).

To determine whether the xylem phenotype of cals3-1d is PHB

dependent, we introduced the loss-of-function phb-13 allele

into the cals3-1d background and observed a suppression of

the protoxylem defect (Figures 6B, 6D, and 6E). Next, we asked

if SHR movement from the stele to the endodermis is affected in

cals3-d. We found that in roots expressing pSHR::SHR:GFP the

GFP signal in the endodermis relative to the stele was decreased

by 50% and 59% in cals3-1d and cals3-2d lines, respectively

(Figure S5E) indicating a strong decrease in SHR movement.

Consistent with this we often observed defects in ground tissue

patterning (Figures S5A and S5B).

If the defective formation of protoxylem, we see in the cals3-d

mutants is the result of loss of symplastical signaling between

the stele and the endodermis, either because of a loss of SHR

movement or a loss of movement of the miRNA, then blocking

signaling between these two tissues should phenocopy cals3-d.

To this end, we analyzed the effect of icals3m under a ground

tissue-specific enhancer (J0571; p6xUAS::icals3m) (Figures 3B

and 3E). After a 6 hr induction, we detected elevated levels of

PHB transcript (Figure 6F) and after a 12 hr induction we

observed an expansion of PHB expression domain inside the

stele (Figures 6G and 6H). Ectopic metaxylem formation was

observed after 48 hr of induction (Figures 6I–6K). Taken together,

these results suggest that ground tissue-specific induction of

icals3m disrupts PD-mediated intercellular signaling between

the endodermis and the stele eventually leading to loss of post-

transcriptional regulation of PHB associated by loss of proto-

xylem identity. This mechanism is likely behind the xylem

differentiation defects observed in the cals3-d mutants (Figures

6B and 6C; Figure S5B).

l Cell 21, 1144–1155, December 13, 2011 ª2011 Elsevier Inc. 1149

Figure 6. Induced Callose Biosynthesis at PD

Modulates Symplastic Transport of SHR

(A–E)Basic fuchsin-stainedxylemofwild-type (A),cals3-1d

(B), cals3-3d (C), phb13 cals3-1d, (D) and phb13 (E).

(F) PHB and CALS3 transcript levels in transgenic line with

J0571, p6xUAS::icals3m after 6 and 24 hr induction. Data

are shown as mean ± SEM.

(G and H) In situ hybridization with PHB mRNA-specific

probe on cross-sections of J0571, p6xUAS::icals3m

without induction (G), and after 12 hr induction (H).

Asterisks, endodermis.

(I and J) Basic fuchsin-stained xylem of transgenic line

with p6xUAS::icals3m, J0571 without induction (I), and

after 5 day germination on inductive conditions (J).

(K) Characterization of the xylem phenotype in J0571 line

(n = 18), in transgenic line with J0571, p6xUAS::icals3m

after 0 hr (n = 44) and 48 hr induction (n = 44). mx,

metaxylem; px, protoxylem.

(L–N) Localization of the pSHR::SHR:GFP in the trans-

genic line with pCRE1::icals3m without induction (L), after

7 (M) and 22 (N) hours of induction.

Open arrowhead, protoxylem; arrowheads, metaxylem;

yellow arrow, transition from metaxylem to protoxylem;

white asterisks, endodermis; yellow asterisks, pericycle;

Scale bars, 50 mm. See also Figure S5.

Developmental Cell

Callose Biosynthesis Regulates Cell Signaling

We next focused on the reciprocal signaling mediated by SHR

and miR165/6 between the stele and the endodermis required

for the proper control of PHBmRNA domain. We asked whether

movement of only SHR or both SHR and miR165/6 could be in-

hibited using the icals3m system. We first examined SHR move-

ment in plants carrying icals3m driven by the stele-specificCRE1

promoter. Induction of this construct resulted in stele-specific

deposition of callose (Figures 3C, 3F, and 4E). By 7 hr after

induction, the pSHR::SHR:GFP signal in the endodermis was

significantly decreased, and by 22 hr it was barely detectable

(Figures 6L–6N). In contrast, the intensity of the pSHR::SHR:GFP

signal in the stele was unaltered. These results indicate that SHR

mobility from the stele to the endodermis is blocked by

enhanced accumulation of callose in the stele cells.

1150 Developmental Cell 21, 1144–1155, December 13, 2011 ª2011 Elsevier Inc.

To test whether the movement of the

miR165/6 species is dependent on symplastic

trafficking, we analyzed a line with SHR inde-

pendent regulation of MIR165a expression

(p6xUAS::icals3m introduced into shr-2 with

J0571; p6xUAS:MIR165a) (Carlsbecker et al.,

2010) (Figure S6). The spatial distribution of

mature miR165 in the root was analyzed in situ

after 24 hr of induction using highly sensitive

locked nucleic acid (LNA) probes. Without

induction, miR165 was detected throughout in

the stele (evident as dark brown staining in Fig-

ure 7A). After induction, the strong signal was

detected only in the ground tissue cell layer,

which is the source for the miR165, as well as

the site of induced callose production (Fig-

ure 7B; Figure S6). In contrast, the stele and

the epidermis displayed significantly reduced

levels of miR165 as indicated by the lighter

brown staining pattern in the corresponding

tissues. In addition, induction also resulted in a dramatic expan-

sion of PHB expression into all the cells of the stele (Figures 7C

and 7D).

To further evaluate this, we created a ‘‘miRNA-sensor’’ system

that uses a modified version of the MIR165A gene, MIR165Amu

(designated MIR165Amu, Miyashima et al., 2011) and a nuclear

localized YFP with the target sequence for miR165mu in its

30 untranslated region (hereafter called nlsYFP_165mu_tgt). In

this system, the nlsYFP_165mu_tgt was under the control of

the SPR1 promoter (Nakajima et al., 2004), which is active in

all cell types of Arabidopsis root meristem and MIR165Amu

was placed under the UAS (UAS::MIR165Amu). As nlsYFP_

165mu_tgt is expressed through the meristem, the domain of

miR165Amu activity can be deduced by examining the loss of

Figure 7. Symplastic Trafficking Is Required for

miRNA Mobility and for Meristem Maintenance

(A–D) In situ hybridization with a miR165-specific LNA

probe (A and B) and PHB mRNA-specific probe (C and D)

on cross sections of shr-2 with p6xUAS::miR165a,

p6xUAS::icals3m, J0571 after 0 hr (A and C) and 24 hr

(B and D) induction. Asterisks, ground tissue.

(E and F) pSPR1::nlsYFP_165mu_tgt localization in

transgenic line with UAS::MIR165Amu/J0571; p6xUAS::

icals3m after 0 hr (E) and 24 hr induction (F). Asterisks,

nlsYFP signal.

(G–I) Callose biosynthesis and degradation at the neck

region of the PD leads to controlled mobility of the

signaling molecules, such as proteins and miRNA species

(G), whereas increased callose deposition narrows the PD

aperture (H). SHR, miR165, and unknown intercellular

signal are trafficking via PD in the root tip (I). DT, desmo-

tubule; PM, plasma membrane.

Scale bar, 50 mm. See also Figure S6.

Developmental Cell

Callose Biosynthesis Regulates Cell Signaling

YFP signal. This sensor system was introduced into J0571;

p6xUAS::icals3m line. In the resulting line (subsequently called

pSPR1::nlsYFP_165mu_tgt/UAS::MIR165Amu/J0571; p6xUAS::

icals3m), miR165Amu is constitutively transcribed in the ground

tissue and icals3m can be induced in the same domain. Under

noninduced condition, nlsYFP was not detected in the root meri-

stem indicating that ground tissue-derived miR165Amu is effec-

tively suppressing its expression (Figure 7E). This is consistent

with the previously reported non-cell-autonomous action of

miR165/6 and miR165Amu (Miyashima et al., 2011). In contrast,

after 24 hr of induction of the icals3m in the ground tissue, nlsYFP

signal was observed in the stele cells (Figure 7F). This suggests

thatmovement ofmiR165Amu from the ground tissue is inhibited

by accumulation of callose in the ground tissue. Taken together,

our data indicate that induced callose production by icals3m

limits SHR movement from the stele-to-endodermis. In addition,

it also effectively inhibits the mobility of MIR165A products,

presumably the mature form of miR165, from the ground tissue

to the stele.

DISCUSSION

Regulated cell-to-cell signaling is essential in any multicellular

organism to allow coordination of developmental programs.

Previously, plant development has been shown to involve three

types of cell-to-cell communication: the interactions of extracel-

lular ligandswith PM-localized receptors (Clark et al., 1997; Stahl

et al., 2009), the polar transport of the phytohormone auxin

through membrane localized transporters (Galweiler et al.,

1998; Robert and Friml, 2009), and the regulated transport of

transcription factors between cells (Koizumi et al., 2011; Xu

et al., 2011). Based on molecular, genetic and developmental

analysis, we highlight here the central role of regulated callose

biosynthesis for symplastic trafficking through the PD in medi-

ating cell-to-cell signaling in plants.

Developmenta

Controlled Activity of the CALS/GT48 Gene FamilyRegulates PD ApertureThe deposition or presence of callose in the neck regions of the

PD has been implicated in regulating the PD size exclusion limit.

The cellular level of callose is regulated by a balance between the

activity of callose synthases, which produce callose and

b-1,3-glucanases, which degrade callose. By decreasing the

endogenous levels of b-1,3-glucanases decreased cell-to-cell

movement of viral proteins and GFP has been observed in

tobacco (Iglesias and Meins, 2000) and Arabidopsis (Levy

et al., 2007). Likewise Benitez-Alfonso et al. (2009) showed

that increased ROS leads to a decreased PD aperture and

reduced diffusion of GFP, due to increased callose production.

However, the role of callose synthases and the underlying

molecular components regulating PD are not well characterized.

Genes that encode putative callose synthases have been identi-

fied in higher plants on the basis of their similarity with the yeast

FKS genes (see, for instance, Cui et al., 2001). Nevertheless, very

few results are available on the structure and function of the

corresponding proteins. The most convincing biochemical data

linking callose biosynthesis to CALS proteins has been obtained

in barley (Li et al., 2003) and tobacco pollen tubes (Brownfield

et al., 2007). The callose synthase enzymes are generally consid-

ered to be membrane-bound complexes, primarily based on the

fact that attempts to purify them usually result in the copurifica-

tion of several polypeptides (Colombani et al., 2004). However,

evidence for their actual organization as complexes and data

on the corresponding subunit composition, stoichiometry and

regulation are lacking.

Arabidopsis has 12 putative callose synthase genes, which are

members of the glycosyltransferase family 48 (GT48) and display

divergent but overlapping expression patterns throughout plant

development (Hong et al., 2001). Besides the reports on callose

biosynthesis in cellular locations other than PD (Thiele et al.,

2008; Toller et al., 2008; Chen et al., 2009), Guseman et al.

l Cell 21, 1144–1155, December 13, 2011 ª2011 Elsevier Inc. 1151

Developmental Cell

Callose Biosynthesis Regulates Cell Signaling

(2010) recently reported the involvement of a member of the

GT48 family, CALS10/GSL8, in the biosynthesis of PD associ-

ated callose. Furthermore, Xie et al. (2011) and Barratt et al.

(2011) reported the role of CALS7 in the biosynthesis of callose

at the phloem PD and sieve plates. However, their role in regu-

lating PD aperture remains elusive.

We report here the identification and molecular characteriza-

tion of CALS3, another member of the GT48 gene family. We

identified three gain-of-function, missense mutations in CALS3

that result in transient callose overaccumulation at the early

stages of root development and therefore display various devel-

opmental defects due to impaired cell-to-cell communication. It

seems possible that the activity of the CALS3 protein is normally

under a negative regulation that these gain-of-function muta-

tions are able to overcome. This is supported by results indi-

cating that overexpression of the PD localized b-1,3-glucanase,

At5g42100 partially suppresses the cals3-1d phenotype.

Symplastic Movement of SHR Protein and miR165To gain a more detailed understanding on the role of these

mutations, the effect of cals3-1d solely or in combination with

cals3-2d (to give ‘‘cals3m’’) was analyzed by establishing

a system that allows spatiotemporal control of callose accumu-

lation at PD. Inducible expression of the CALS3 mutant protein

caused enhanced callose accumulation in the zone surrounding

PD in as little as 8 hr as indicated by callose immunolocalization

studies. This callose accumulation often appeared to span the

whole cell wall diameter rather than occupying only the two

neck regions. The enhanced callose accumulation was accom-

panied with a reduced PD aperture and impaired trafficking of

various macromolecules. The cals3-d mutations lead to tran-

siently enhanced callose formation during lateral root develop-

ment. It is possible that a similar transient callose accumulation

may occur during embryonic root development in cals3-d and

result an altered cell-to-cell communication. As we know that

PD apertures and density change over developmental time,

with young cells and tissues showing greater degrees of intercel-

lular transport and connectivity (Kim et al., 2002), further analysis

of the spatial and temporal regulation of callose accumulation

by genes involved in the biosynthesis and degradation of callose

will reveal how PD are regulated (Levy et al., 2007; Guseman

et al., 2010; Fernandez-Calvino et al., 2011).

The stele and the endodermis is a nexus of cell communication

involving the transport of the SHR transcription factor from the

stele into the endodermis and the reciprocal transport of

miR165/6 from the endodermis to the stele. What is particularly

striking about this signaling pathway is that although SHR is

expressed in the stele, it must move into the endodermis to

turn on miR165/6, which, in turn, must move to accomplish its

primary function: suppression of the PHB transcript. To better

understand the nature of this signaling pathway, we transiently

and exclusively increased accumulation of callose in specific

tissues in the root to assay their affects on the bidirectional

signaling of SHR and miR165/6. We found that movement of

both the SHR protein and the product of modified MIR165A,

most likely mature miR165a, were impaired upon induction of

callose in either the endodermis or the stele. This decrease in

movement was accompanied by phenotypic defects associated

with loss of the function of the mobile factor in the target tissues.

1152 Developmental Cell 21, 1144–1155, December 13, 2011 ª2011

Our results indicate that both protein and miRNA can move

through PD to control development.

Rate-Limiting Role of Callose Biosynthesisin Symplastic TraffickingIn the course of this work, we have illustrated several non-cell-

autonomous effects due to constriction of the PD aperture and

macromolecular traffic through the PD. The cals3-d mutations

were identified in the same genetic screens as the phb-7dmuta-

tion and scr-6mutations. The cals3-d, phb-7d and scr-6mutants

all display ectopic metaxylem formation at the protoxylem posi-

tion resulting from impaired bidirectional signaling between the

endodermis and stele. Ectopic metaxylem formation can also

be observed after a 48 hr induction of callose biosynthesis in

the ground tissue, demonstrating that regulation of the PD aper-

ture has non-cell-autonomous consequences.

We observed a tissue interactionwhen icals3mwas expressed

phloem specifically under the APL promoter. This resulted in

a rapid arrest of cell division activity in the root meristem accom-

panied with an expansion of the cell differentiation zone of the

meristem at the expense of the cell division zone. The nature

of themolecular signal(s) underlying this interaction is not known.

Loss of meristem maintenance and pSUC2::GFP unloading

accompanied by callose accumulation in the root meristem

has been earlier reported in GAT1 loss-of-function mutant

(Benitez-Alfonso et al., 2009). GAT1 encodes a plastid localized

thioredoxin m3, and it has been proposed to act as a redox-

regulator of callose synthesis. It is possible that same pathway

required for meristem maintenance is defected both in pAPL::

icals3m line and gat1.

CALS7 was recently identified as a phloem-specific member

of the GT48 gene family, and loss-of-function mutations in

CALS7 resulted in reduced callose deposition at the PD of sieve

plates and radial walls of the SEs (as recently reported by Xie

et al. [2011] and Barratt et al. [2011]). We report here that during

root development CALS7 is highly and specifically enriched in

phloem SEs that have a large number of symplastic connections

to the neighboring cells. CALS3 is expressed in phloem and in

meristematic tissues. Enhanced CALS3 expression during

phloem development can suppress a loss-of-function mutation

in CALS7. This indicates that the two proteins are at least to

a certain extent functionally redundant as callose synthases

that function during SE development, and the callose biosyn-

thesis rate during SE development is under specific transcrip-

tional regulation. Given the diversity of the various structural

forms of PD connecting plant tissues and the expanding list of

the molecular components involved in PD function, it is possible

that tissue-specific control of the PD composition is an important

regulatory mechanism during plant development.

EXPERIMENTAL PROCEDURES

Plant Materials

Arabidopsis thaliana lines were in Columbia (Col-0) or in C24 background.

Seeds were sterilized and germinated on long day conditions on plates con-

taining one-half MS; 1% sucrose; 0,05% MES; 1% phytagel or 1% agar.

The media was supplemented with either 20 mM NPA (N-(1-naphthyl)phthala-

mic acid) (TCI) or 10 mM 17-b-estradiol (Sigma) (except in cals7-1 suppression

experiment where 1 mM 17-b-estradiol was used to reduce accumulation of

background callose due to induction) as indicated in the text. Inductions

Elsevier Inc.

Developmental Cell

Callose Biosynthesis Regulates Cell Signaling

were performed by transferring transgenic as well as wild-type seedlings (to

mimic induction) to a normal plate or a plate containing NPA or 17-b-estradiol.

For anatomical, histological, and reporter gene analyses, primary roots of 4- to

6-day-old vertically grown seedlings were used.

Genetic Screening and Positional Cloning

The cals3-1d-3d mutants were isolated in two genetic screens based on EMS

mutagenesis of themobile phloemmarker pSUC2::GFP (Imlau et al., 1999) and

the protoxylem marker pAHP6::GFP (Mahonen et al., 2006) described by

Carlsbecker et al. (2010). The cals3-1d (C24) was crossed to Col-0 to generate

F2 mapping population. Detailed information on mapping procedures is

provided in Supplemental Experimental Procedures.

Transgenic Work and Inducible System

Various tissue-specific markers were introduced into mutant backgrounds by

crossing and analyses were done on segregating F2 generations. Most trans-

genic constructs were generated using the gateway or multisite gateway

system (Invitrogen) and all were introduced into plants using the floral dip

method (Clough and Bent, 1998). PCR-based site directed mutagenesis was

used to introduce cals3-1d and -2d mutations to the genomic fragment con-

taining the wild-type CALS3 gene. Modified version of pER8, an estrogen-

receptor based chemical-inducible system was used to create icals3m,

icals3-1d and iCALS3 (Zuo et al., 2000). Effector genes were subcloned into

pDONR(221) (Invitrogen) and subsequently recombined into the inducible

system. Detailed information on DNA manipulations and transgenic work is

provided in Supplemental Experimental Procedures.

Histological Analyses

Transverse sectioning, fuchsin staining, and confocal imaging were performed

as described by Mahonen et al. (2000), except analysis was done on a Leica

SP5 confocal using a solid state blue laser for GFP and YFP (480 nm/270

mW). Fluorescent intensity was measured with the Image J (http://rsbweb.

nih.gov/ij). GUS staining was carried out as described by Miyawaki et al.

(2004). Xylem cell type analysis was carried out with 5-day-old seedlings

embedded to chloral hydrate solution. For plasmolysis, the seedlings express-

ing p35S::GFP-CALS3 were incubated in 0.8 M mannitol for 1.5 hr and then

stained with propidium iodide in 0.8 M mannitol prior to microscopy.

Aniline Blue Staining

Aniline blue staining was performed using aniline blue fluorochrome

(Biosupplies Australia Pty. Ltd.) as described by Ingram et al. (2011). A stock

solution (0.1 mg/ml in H2O) was diluted 1:3 ratio with 67 mM K3PO4 pH 9.5.

Seedlings were incubated in the staining solution for 2 hr at room temperature

and imaged in Citifluor/PBS solution (Citifluor Ltd.) on a Leica SP5 or SP2

confocal using a UV laser.

TEM Analysis, Callose Immunogold, and Immunofluorescence

Localization

Root tips were high-pressure frozen, freeze-substituted in acetone containing

0.5% glutaraldehyde and 0.3% uranyl acetate and embedded in Lowicryl

HM20 or K11M at �50�C. Ultrathin resin sections were labeled with mouse

anti-callose antibodies and 12 nm gold coupled to goat anti-mouse IgG or

Cy3 coupled to goat anti-mouse IgG. Detailed information is provided in

Supplemental Experimental Procedures.

In Situ Hybridization

For riboprobe and LNA in situ hybridizations, tissue samples were prepared

from 4- to 5-day-old roots; sectioning, preparation of the probes, and in situ

hybridization were performed as described by Mahonen et al. (2000), except

that the tissues were embedded in Histosec wax (Merck) and the 7 mm thick

sections were mounted on SuperFrost Plus or Ultra Plus slides (Fisher Scien-

tific). Templates for RNA probes were amplified using T7-labeled primers were

transcribed using a digoxigenin (DIG) RNA Labeling Kit (Roche). The PHB

sense and antisense probe templates were amplified as described by

Carlsbecker et al. (2010). LNA probes with the complementary sequences to

miR165 were synthesized and 50 digoxigenin labeled by Exiqon (Vedbaek).

As a control probe, we used a miR165 LNA probe with four nucleotide

changes (termed miR165MM), 50-ggCgAatgTagcGtggtccga-30, synthesized

Developmenta

and 50 digoxigenin labeled by Exiqon. Hybridization, mounting, and imaging

were performed as described by Carlsbecker et al. (2010). Primer sequences

are provided in Supplemental Experimental Procedures.

RNA Analysis

For quantification of mRNA, root tips or shoots from 5- to 7-day-old seedlings

were harvested and total RNAs were extracted with miRNeasy Mini Kit (Quia-

gen) and were DNase treated on column (QIAGEN). cDNAs were synthesized

from 500 ng of total RNAs primed with random hexamer primers using First

strand synthesis kit (Roche). For semiquantitative PCR, 5–20 ng cDNA and

a different number of PCR cycles were used depending on transcript abun-

dance from three independent biological replicates. Quantitative PCRs

(qPCRs) were performed using 10 ng of cDNA with 250 nM of gene-specific

primers and real-time PCR mix (Roche) in a LightCycler 480 Real-Time PCR

System (Roche) at 55�Cannealing temperature for 40 cycles. For each deregu-

lated gene, the difference in Cp value between 17-b-estradiol treated and

mock-treated RNAs (ΔCp) was computed and then normalized with internal

housekeeping gene Actin to calculate ΔΔCp. The final fold-change in expres-

sion level upon induction calculated from ΔΔCp was determined from twelve

technical replicates derived from three independent biological samples.

Primer sequences are provided in Supplemental Experimental Procedures.

Statistical Analyses

All the data are shown asmean ±SEM, except Table S1, where data are shown

as mean ± SD. Comparison of differences among groups in Figure 5D was

carried out by Pearson’s Chi-square test (**p < 0.01).

ACCESSION NUMBERS

The GenBank accession number for the CalS3 mRNA sequence is HM594867.

SUPPLEMENTAL INFORMATION

Supplemental Information includes six figures, one table, and Supplemental

Experimental Procedures and can be found with this article online at doi:10.

1016/j.devcel.2011.10.006.

ACKNOWLEDGMENTS

We thank K. Kainulainen, M. Herpola, R. Ursache, I. Sevilem, T. Svitkina, and

A. Stout for technical assistance, M. Bonke, A. Bishopp, M. Prigge, P. Benfey,

ABRC, and NASC for materials, S. El-Showk for comments on the manuscript,

V. Vaten for technical assistance with figure preparation, and C. van der

Schoot for discussions. This work was supported by the grants by the

Academy of Finland, Tekes and ESF to Y.H., S.L., and A.V., Viikki Graduate

School in Molecular Biosciences to A.V., European Molecular Biology Organi-

sation (EMBO, ALTF 450-2007) to J.D., FORMAS and Carl Trygger’s Founda-

tion for Scientific Research to A.C., Human Frontier Science Program Long-

Term Fellowship and Academy of Finland to A.P.M., World Class University

Program (R33-10002, NRF) and the Next-Generation BioGreen 21 Program

(SSAC #2011, RDA) of Korea to J.K., and NSF grant 0920327 (awarded to

K.L.G.) to S.W. V.B. was supported by the Swedish Centre for Biomimetic

Fiber Engineering (Biomime). A.V. designed and performed experiments to

screen the cals3-d mutations and identified and characterized them and the

CALS7 knockout lines at the molecular level, including construction and anal-

ysis of the icals3m lines in various genetic backgrounds. J.D. participated in

the identification and characterization of cals3-dmutations, optimized the cal-

lose staining method, participated in the immunological analyses of callose

deposition, and analyzed the subcellular localization of 35S::GFP-CALS3.

S.W. analyzed the plasmodesmal apertures and SHR movement in various

genetic backgrounds. Y.-D.S. analyzed the callose distribution using immuno-

fluorescence and immunogold labeling techniques. S.M. analyzed the subcel-

lular localization of 35S::GFP-CALS3-1d, performed the analysis of meristem

size following pAPL::icals3m and performed the experiments on the sensor

approach in monitoring miR165A mobility. S.R.Y. participated in molecular

characterization of the CALS3 and CALS7 loci and icals3m lines and

enzymatic analyses of callose synthase activity. C.J.R. performed the in situ

l Cell 21, 1144–1155, December 13, 2011 ª2011 Elsevier Inc. 1153

Developmental Cell

Callose Biosynthesis Regulates Cell Signaling

hybridizations. A.C. isolated the cals3-3d allele. V.B. performed the analysis of

callose synthase activity. R.L. and S.L. participated in identification of the

CALS7 gene and R.L. performed the cloning of pCALS7::YFP. A.P.M. and

B.S. contributed the vector system for icals3m. J.K. participated in optimiza-

tion of the callose staining protocol. E.J. and N.S. participated in the analysis

ofCALS7. K.N. contributed the sensor system formiR165Amobility. A.C. iden-

tified cals3-2d and co-supervised A.V. and C.J.R. K.G. designed the experi-

ments on SHR movement and plasmodesmal aperture. Y.H. participated in

experimental design. A.V., K.G., and Y.H. wrote the manuscript. A.C. and

K.G. contributed equally, S.W., Y-D.S. and S.M. contributed equally. All

authors discussed the results and commented on the manuscript.

Received: August 28, 2010

Revised: June 1, 2011

Accepted: October 11, 2011

Published online: December 12, 2011

REFERENCES

An, Q., Huckelhoven, R., Kogel, K.-H., and van Bel, A.J.E. (2006).

Multivesicular bodies participate in a cell wall-associated defence response

in barley leaves attacked by the pathogenic powdery mildew fungus. Cell.

Microbiol. 8, 1009–1019.

Aoki, K., Kragler, F., Xoconostle-Cazares, B., and Lucas, W.J. (2002).

A subclass of plant heat shock cognate 70 chaperones carries a motif that

facilitates trafficking through plasmodesmata. Proc. Natl. Acad. Sci. USA 99,

16342–16347.

Barratt, D.H., Kolling, K., Graf, A., Pike, M., Calder, G., Findlay, K., Zeeman,

S.C., and Smith, A.M. (2011). Callose synthase GSL7 is necessary for normal

phloem transport and inflorescence growth in Arabidopsis. Plant Physiol. 155,

328–341.

Benitez-Alfonso, Y., Faulkner, C., Ritzenthaler, C., and Maule, A.J. (2010).

Plasmodesmata: gateways to local and systemic virus infection. Mol. Plant

Microbe Interact. 23, 1403–1412.

Benitez-Alfonso, Y., Cilia, M., San Roman, A., Thomas, C., Maule, A., Hearn,

S., and Jackson, D. (2009). Control of Arabidopsis meristem development

by thioredoxin-dependent regulation of intercellular transport. Proc. Natl.

Acad. Sci. USA 106, 3615–3620.

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.

Botha, C.E.J., Hartley, B.J., and Cross, R.H.M. (1993). The ultrastructure and

computer-enhanced digital image analysis of plasmodesmata at the Kranz

mesophyll-bundle sheath interface of Themeda triandra var. imberbis (Retz.)

A. Camus in conventionally-fixed leaf blades. Ann. Bot. (Lond.) 72, 255–261.

Brady, S.M., Orlando, D.A., Lee, J.Y., Wang, J.Y., Koch, J., Dinneny, J.R.,

Mace, D., Ohler, U., and Benfey, P.N. (2007). A high-resolution root spatiotem-

poral map reveals dominant expression patterns. Science 318, 801–806.

Brownfield, L., Ford, K., Doblin, M.S., Newbigin, E., Read, S., and Bacic, A.

(2007). Proteomic and biochemical evidence links the callose synthase in

Nicotiana alata pollen tubes to the product of the NaGSL1 gene. Plant J. 52,

147–156.

Bucher, G.L., Tarina, C., Heinlein, M., Di Serio, F., Meins, F., Jr., and Iglesias,

V.A. (2001). Local expression of enzymatically active class I beta-1, 3-gluca-

nase enhances symptoms of TMV infection in tobacco. Plant J. 28, 361–369.

Carlsbecker, A., Lee, J.Y., Roberts, C.J., Dettmer, J., Lehesranta, S., Zhou, J.,

Lindgren, O., Moreno-Risueno, M.A., Vaten, A., Thitamadee, S., et al. (2010).

Cell signalling by microRNA165/6 directs gene dose-dependent root cell

fate. Nature 465, 316–321.

Chen, X.-Y., Liu, L., Lee, E., Han, X., Rim, Y., Chu, H., Kim, S.-W., Sack, F., and

Kim, J.-Y. (2009). The Arabidopsis callose synthase gene GSL8 is required for

cytokinesis and cell patterning. Plant Physiol. 150, 105–113.

Citovsky, V., Knorr, D., Schuster, G., and Zambryski, P. (1990). The P30move-

ment protein of tobacco mosaic virus is a single-strand nucleic acid binding

protein. Cell 60, 637–647.

1154 Developmental Cell 21, 1144–1155, December 13, 2011 ª2011

Clark, S.E., Williams, R.W., and Meyerowitz, E.M. (1997). The CLAVATA1 gene

encodes a putative receptor kinase that controls shoot and floral meristem size

in Arabidopsis. Cell 89, 575–585.

Clough, S.J., and Bent, A.F. (1998). Floral dip: a simplified method for

Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16,

735–743.

Colombani, A., Djerbi, S., Bessueille, L., Blomqvist, K., Ohlsson, A., Berglund,

T., Teeri, T., and Bulone, V. (2004). In vitro synthesis of (1-3)-beta-D-glucan

(callose) and cellulose by detergent extracts of membranes from cell suspen-

sion cultures of hybrid aspen. Cellulose 11, 313–327.

Cui, X., Shin, H., Song, C., Laosinchai, W., Amano, Y., and Brown, R.M., Jr.

(2001). A putative plant homolog of the yeast beta-1,3-glucan synthase

subunit FKS1 from cotton (Gossypium hirsutum L.) fibers. Planta 213,

223–230.

Dolan, L., Janmaat, K., Willemsen, V., Linstead, P., Poethig, S., Roberts, K.,

and Scheres, B. (1993). Cellular organisation of the Arabidopsis thaliana

root. Development 119, 71–84.

Dong, X., Hong, Z., Chatterjee, J., Kim, S., and Verma, D.P. (2008). Expression

of callose synthase genes and its connection with Npr1 signaling pathway

during pathogen infection. Planta 229, 87–98.

Ehlers, K., and Kollmann, R. (2001). Primary and secondary plasmodesmata:

structure, origin, and functioning. Protoplasma 216, 1–30.

Emery, J.F., Floyd, S.K., Alvarez, J., Eshed, Y., Hawker, N.P., Izhaki, A., Baum,

S.F., and Bowman, J.L. (2003). Radial patterning of Arabidopsis shoots by

class III HD-ZIP and KANADI genes. Curr. Biol. 13, 1768–1774.

Erwee, M.G., and Goodwin, P.B. (1985). Symplast domains in extrastelar

tissues of Egeria densa Planch. Planta 163, 9–19.

Fernandez-Calvino, L., Faulkner, C., Walshaw, J., Saalbach, G., Bayer, E.,

Benitez-Alfonso, Y., and Maule, A. (2011). Arabidopsis plasmodesmal pro-

teome. PLoS ONE 6, e18880.

Guseman, J.M., Lee, J.S., Bogenschutz, N.L., Peterson, K.M., Virata, R.E., Xie,

B., Kanaoka, M.M., Hong, Z., and Torii, K.U. (2010). Dysregulation of cell-to-

cell connectivity and stomatal patterning by loss-of-function mutation in

Arabidopsis chorus (glucan synthase-like 8). Development 137, 1731–1741.

Galweiler, L., Guan, C., Muller, A., Wisman, E., Mendgen, K., Yephremov, A.,

and Palme, K. (1998). Regulation of polar auxin transport by AtPIN1 in

Arabidopsis vascular tissue. Science 282, 2226–2230.

Hong, Z., Delauney, A.J., and Verma, D.P. (2001). A cell plate-specific callose

synthase and its interaction with phragmoplastin. Plant Cell 13, 755–768.

Iglesias, V.A., andMeins, F., Jr. (2000). Movement of plant viruses is delayed in

a b-1,3-glucanase-deficient mutant showing a reduced plasmodesmatal size

exclusion limit and enhanced callose deposition. Plant J. 21, 157–166.

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 unloading

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

Ingram, P., Dettmer, J., Helariutta, Y., and Malamy, J.E. (2011). Arabidopsis

Lateral Root Development 3 is essential for early phloem development and

function, and hence for normal root systemdevelopment. Plant J. 68, 455–467.

Kim, I., Hempel, F.D., Sha, K., Pfluger, J., and Zambryski, P.C. (2002).

Identification of a developmental transition in plasmodesmatal function during

embryogenesis in Arabidopsis thaliana. Development 129, 1261–1272.

Koizumi, K., Wu, S., Macrae-Crerar, A., and Gallagher, K.L. (2011). An essen-

tial protein that interacts with endosomes and promotes movement of the

SHORT-ROOT transcription factor. Curr. Biol. 21, 1559–1564.

Levy, A., Erlanger, M., Rosenthal, M., and Epel, B.L. (2007). A plasmodesmata-

associated b-1,3-glucanase in Arabidopsis. Plant J. 49, 669–682.

Li, J., Burton, R.A., Harvey, A.J., Hrmova, M., Wardak, A.Z., Stone, B.A., and

Fincher, G.B. (2003). Biochemical evidence linking a putative callose synthase

gene with (1 —> 3)-beta-D-glucan biosynthesis in barley. Plant Mol. Biol. 53,

213–225.

Lucas, W.J., and Lee, J.-Y. (2004). Plasmodesmata as a supracellular control

network in plants. Nat. Rev. Mol. Cell Biol. 5, 712–726.

Elsevier Inc.

Developmental Cell

Callose Biosynthesis Regulates Cell Signaling

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.

Mahonen, A.P., Bishopp, A., Higuchi, M., Nieminen, K.M., Kinoshita, K.,

Tormakangas, K., Ikeda, Y., Oka, A., Kakimoto, T., and Helariutta, Y. (2006).

Cytokinin signaling and its inhibitor AHP6 regulate cell fate during vascular

development. Science 311, 94–98.

Matsuzaki, Y., Ogawa-Ohnishi, M., Mori, A., and Matsubayashi, Y. (2010).

Secreted peptide signals required for maintenance of root stem cell niche in

Arabidopsis. Science 329, 1065–1067.

McConnell, J.R., Emery, J., Eshed, Y., Bao, N., Bowman, J., and Barton, M.K.

(2001). Role of PHABULOSA and PHAVOLUTA in determining radial patterning

in shoots. Nature 411, 709–713.

Miyawaki, K., Matsumoto-Kitano, M., and Kakimoto, T. (2004). Expression of

cytokinin biosynthetic isopentenyltransferase genes in Arabidopsis: tissue

specificity and regulation by auxin, cytokinin, and nitrate. Plant J. 37, 128–138.

Miyashima, S., Koi, S., Hashimoto, T., and Nakajima, K. (2011). Non-cell-

autonomous microRNA165 acts in a dose-dependent manner to regulate

multiple differentiation status in the Arabidopsis root. Development 138,

2303–2313.

Nakajima, K., Sena, G., Nawy, T., and Benfey, P.N. (2001). Intercellular move-

ment of the putative transcription factor SHR in root patterning. Nature 413,

307–311.

Nakajima, K., Furutani, I., Tachimoto, H., Matsubara, H., and Hashimoto, T.

(2004). SPIRAL1 encodes a plant-specific microtubule-localized protein

required for directional control of rapidly expanding Arabidopsis cells. Plant

Cell 16, 1178–1190.

Prigge, M.J., Otsuga, D., Alonso, J.M., Ecker, J.R., Drews, G.N., and Clark,

S.E. (2005). Class III homeodomain-leucine zipper gene family members

have overlapping, antagonistic, and distinct roles in Arabidopsis development.

Plant Cell 17, 61–76.

Rinne, P.L., van den Boogaard, R., Mensink, M.G., Kopperud, C., Kormelink,

R., Goldbach, R., and van der Schoot, C. (2005). Tobacco plants respond to

the constitutive expression of the tospovirus movement protein NS(M) with

a heat-reversible sealing of plasmodesmata that impairs development. Plant

J. 43, 688–707.

Robert, H.S., and Friml, J. (2009). Auxin and other signals on the move in

plants. Nat. Chem. Biol. 5, 325–332.

Sabatini, S., Beis, D., Wolkenfelt, H., Murfett, J., Guilfoyle, T., Malamy, J.,

Benfey, P., Leyser, O., Bechtold, N., Weisbeek, P., and Scheres, B. (1999).

An auxin-dependent distal organizer of pattern and polarity in the

Arabidopsis root. Cell 99, 463–472.

Schlereth, A., Moller, B., Liu, W., Kientz, M., Flipse, J., Rademacher, E.H.,

Schmid, M., Jurgens, G., and Weijers, D. (2010). MONOPTEROS controls

embryonic root initiation by regulating a mobile transcription factor. Nature

464, 913–916.

Simpson, C., Thomas, C., Findlay, K., Bayer, E., and Maule, A.J. (2009). An

Arabidopsis GPI-anchor plasmodesmal neck protein with callose binding

activity and potential to regulate cell-to-cell trafficking. Plant Cell 21, 581–594.

Developmenta

Sivaguru, M., Fujiwara, T., Samaj, J., Baluska, F., Yang, Z., Osawa, H., Maeda,

T., Mori, T., Volkmann, D., and Matsumoto, H. (2000). Aluminum-induced

1—>3-b-D-glucan inhibits cell-to-cell trafficking of molecules through plasmo-

desmata. A new mechanism of aluminum toxicity in plants. Plant Physiol. 124,

991–1006.

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.

Stahl, Y., Wink, R.H., Ingram, G.C., and Simon, R. (2009). A signaling module

controlling the stem cell niche in Arabidopsis root meristems. Curr. Biol. 19,

909–914.

Thiele, K., Wanner, G., Kindzierski, V., Jurgens, G., Mayer, U., Pachl, F., and

Assaad, F.F. (2008). The timely deposition of callose is essential for cytokinesis

in Arabidopsis. Plant J. 58, 13–26.

Thomas, C.L., Bayer, E.M., Ritzenthaler, C., Fernandez-Calvino, L., andMaule,

A.J. (2008). Specific targeting of a plasmodesmal protein affecting cell-to-cell

communication. PLoS Biol. 6, e7.

Toller, A., Brownfield, L., Neu, C., Twell, D., and Schulze-Lefert, P. (2008). Dual

function of Arabidopsis glucan synthase-like genes GSL8 and GSL10 in male

gametophyte development and plant growth. Plant J. 54, 911–923.

van den Berg, C., Willemsen, V., Hendriks, G., Weisbeek, P., and Scheres, B.

(1997). Short-range control of cell differentiation in the Arabidopsis root meri-

stem. Nature 390, 287–289.

Waigmann, E., Lucas, W.J., Citovsky, V., and Zambryski, P. (1994). Direct

functional assay for tobacco mosaic virus cell-to-cell movement protein and

identification of a domain involved in increasing plasmodesmal permeability.

Proc. Natl. Acad. Sci. USA 91, 1433–1437.

Wolf, S., Deom, C.M., Beachy, R.N., and Lucas, W.J. (1989). Movement

protein of tobacco mosaic virus modifies plasmodesmatal size exclusion limit.

Science 246, 377–379.

Xie, B.,Wang, X., Zhu,M., Zhang, Z., andHong, Z. (2011).CalS7 encodes a cal-

lose synthase responsible for callose deposition in the phloem. Plant J. 65,

1–14.

Xu, X.M., Wang, J., Xuan, Z., Goldshmidt, A., Borrill, P.G., Hariharan, N., Kim,

J.Y., and Jackson, D. (2011). Chaperonins facilitate KNOTTED1 cell-to-cell

trafficking and stem cell function. Science 333, 1141–1144.

Zambryski, P., and Crawford, K. (2000). Plasmodesmata: gatekeepers for

cell-to-cell transport of developmental signals in plants. Annu. Rev. Cell Dev.

Biol. 16, 393–421.

Zarsky, V., Cvrckova, F., Potocky, M., and Hala, M. (2009). Exocytosis and cell

polarity in plants - exocyst and recycling domains. New Phytol. 183, 255–272.

Zuo, J., Niu, Q.W., and Chua, N.H. (2000). Technical advance: An estrogen

receptor-based transactivator XVE mediates highly inducible gene expression

in transgenic plants. Plant J. 24, 265–273.

l Cell 21, 1144–1155, December 13, 2011 ª2011 Elsevier Inc. 1155