The Plant Journal (2007) 49, 669–682
doi: 10.1111/j.1365-313X.2006.02986.x
A plasmodesmata-associated b-1,3-glucanase in Arabidopsis Amit Levy, Michael Erlanger, Michal Rosenthal and Bernard L. Epel* Department of Plant Sciences, George S. Wise Faculty of Life Sciences, Tel Aviv University, Tel Aviv 69978, Israel Received 28 June 2006; revised 7 September 2006; accepted 10 October 2006. *For correspondence (fax þ972 3 6409380; e-mail
[email protected]).
Summary Plasmodesmal conductivity is regulated in part by callose turnover, which is hypothesized to be determined by b-1,3-glucan synthase versus glucanase activities. A proteomic analysis of an Arabidopsis thaliana plasmodesmata (Pd)-rich fraction identified a b-1,3-glucanase as present in this fraction. The protein encoded by the putative plasmodesmal associated protein (ppap) gene, termed AtBG_ppap, had previously been found to be a post-translationally modified glycosylphosphatidylinositol (GPI) lipid-anchored protein. When fused to green fluorescent protein (GFP) and expressed in tobacco (Nicotiana tabacum) or Nicotiana benthamiana epidermal cells, this protein displays fluorescence patterns in the endoplasmic reticulum (ER) membrane system, along the cell periphery and in a punctate pattern that co-localizes with aniline blue-stained callose present around the Pd. Plasma membrane localization was verified by co-localization of AtBG_ppap:GFP together with a plasma membrane marker N-[3-triethylammoniumpropyl]-4-[p-diethylaminophenylhexatrienyl] pyridinium dibromide (FM4-64) in plasmolysed cells. In Arabidopsis T-DNA insertion mutants that do not transcribe AtBG_ppap, functional studies showed that GFP cell-to-cell movement between epidermal cells is reduced, and the conductivity coefficient of Pd is lower. Measurements of callose levels around Pd after wounding revealed that callose accumulation in the mutant plants was higher. Taken together, we suggest that AtBG_ppap is a Pd-associated membrane protein involved in plasmodesmal callose degradation, and functions in the gating of Pd. Keywords: Nicotiana tabacum, Arabidopsis thaliana, plasmodesmata, cell–cell communication, proteomics, callose.
Introduction Plasmodesmata (Pd) are co-axial membranous channels that cross walls between adjacent plants cells, interconnecting the cytoplasm of these cells, thus allowing direct cell-to-cell transport of soluble cytoplasmic macromolecules (proteins and RNA molecules), as well as small molecules (Heinlein and Epel, 2004; Oparka, 2004). A group of cells that are interconnected through open Pd but symplasmically isolated from other cells forms a communicating ‘symplast domain’ that acts as an isolated developmental and physiological unit (Rinne and van der Schoot, 1998). The control of transport through Pd serves as an important element in regulating the direct cell-to-cell transport between cells and in the organization and functioning of symplasmic domains. The fact that many plant viruses exploit Pd as conduits for spread of infection between cells makes this transport system a crucial point for defense against virus spread. ª 2006 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd
Although the ultrastructure of Pd is well described, their molecular composition is still mostly unknown. The list of proteins identified as structural or functional components of the Pd is only partial, and includes a reversible glycosylated polypeptide (Sagi et al., 2005), a Ca2þ-dependent protein kinase (Yahalom et al., 1998) and the cytoskeleton proteins actin, myosin and centrin (reviewed by Heinlein and Epel, 2004). The cell wall that surrounds the Pd has a specialized structure devoid of cellulose and hemicellulose but containing non-esterified pectins and callose (Roy et al., 1997; Turner et al., 1994). This specialized wall sheath, which may be an essential part of the Pd, also requires further characterization. Pd are dynamic transport channels that can be modulated. During the sink-to-source transition in leaf development, Pd structure changes from a single tunnel (simple Pd) to a branched one, and the transfer rate of proteins through the 669
670 Amit Levy et al. channels decreases (Liarzi and Epel, 2005; Oparka et al., 1999; Roberts et al., 2001). Beyond these general developmental changes, the functional state of a single plasmodesma is dynamic, and can change in a transient manner from ‘closed’ to ‘open’ to ‘dilated’ (Oparka and Roberts, 2001; Zambryski and Crawford, 2000). Two different mechanisms are assumed to produce these focused changes in the tunnels. The first model suggests that the conductivity changes because of alterations in the plasmodesmal cytoskeleton proteins actin, myosin and centrin. This model gains strength from the finding that actin filament disruption increases Pd permeability, allowing dextrans up to 20 kDa (Stokes radius ¼ 3.3 nm) to move from an injected tobacco mesophyll cell to surrounding cells within 3–5 min, while no movement was seen when the dextrans were injected alone or with treatments stabilizing the actin (Ding et al., 1996). The second model suggests that changes in the wall sheath surrounding the Pd cause changes to its structure that alter its conductivity. These changes are hypothesized to be mediated by callose synthesis and hydrolysis (Olesen and Robards, 1990; Radford and White, 2001; Radford et al., 1998; Ruan et al., 2004; Sivaguru et al., 2000). Callose, a poly-sugar molecule in the form of b-1,3-glucan, is reversibly and transiently deposited in cell walls as a result of stresses and during many developmental processes (Kauss, 1996). These include stress responses such as wounding (Radford et al., 1998) or aluminum toxicity (Sivaguru et al., 2000) and developmental processes such as cotton fiber elongation (Ruan et al., 2004) or bud dormancy induction caused by a short photoperiod (Rinne and van der Schoot, 1998). Callose deposition occurs in the wall surrounding the Pd at both ends of the channel, compressing the plasma membrane inward, thus creating a narrowed neck region (Radford et al., 1998), which reduces the free space available for the passage of molecules through the Pd. During dormancy, callose is also deposited inside the Pd channel, creating an inner plug (Rinne et al., 2001). The enzyme that degrades callose, b-1,3-glucanase, is also an important factor in the regulation of conductivity through Pd. This was demonstrated in studies employing a tobacco mutant with decreased levels of class I b-1,3-glucanase that was generated by antisense transformation. In this mutant line, higher levels of callose were found, and the susceptibility to virus infection was decreased. Examining the movement of dextrans and peptides between the cells revealed that the size exclusion limit, which is the size of the largest dye capable of moving through the Pd, is lower in the mutant plants (Iglesias and Meins, 2000). When the b-1,3glucanase coding sequence was cloned into the TMV replicon, the virus spread faster through the cells, and cloning of the gene in an antisense formation led to the opposite results (Bucher et al., 2001). In order to characterize the molecular composition of Pd and to understand the function of each Pd component in the
regulation of intercellular communication, a Pd isolation protocol was developed in our laboratory, and putative Pdassociated proteins from Arabidopsis thaliana were identified; one being a b-1,3-glucanase. We show here that this protein, when expressed as a GFP fusion, targets to the endoplasmic reticulum (ER) membrane and Pd. Moreover, we present functional evidence showing that, in mutants that lack this b-1,3-glucanase, the movement of GFP between cells is reduced, and the amount of callose in the Pd is elevated. Results Proteomic studies reveal a putative Pd-associated b-1,3glucanase Proteins in an A. thaliana Pd-enriched fraction were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS–PAGE) (Figure 1). Coomassie blue-stained bands were cut, in-gel reduced, alkylated and proteolysed overnight with trypsin, and the tryptic peptides were resolved by reverse-phase chromatography on a C18 column (see Experimental procedures). Mass spectrometry was performed with ion-trap mass spectrometers operating in the positive mode, using repetition of a full MS scan followed by collision-induced dissociation (CID) of the most dominant ion selected from the first MS. The MS data were compared to simulated proteolysis and CID of the proteins in the non-redundant National Center for Biotechnology Informatic (NR-NCBI) database using the SEQUEST software (LCQ; Thermo, www.thermo.com). One of the proteins identified that was present in band 7 (Figure 1) was a 45 kDa, 425 amino acid long b-1,3-glucanase (At5g42100) that was
Figure 1. Detection of AtBG_ppap. SDS–PAGE separation of the Arabidopsis Pd-rich fraction. The arrow indicates the 45 kDa band identified as AtBG_ppap.
ª 2006 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 49, 669–682
A plasmodesmal b-1,3-glucanase 671
Figure 2. Schematic model of At5g42100. The gene contains two exons (black boxes) and one intron (white box). Splicing of the intron results in a 1278 bp coding sequence that encodes the 425 amino acid protein AtBG_ppap. For the GFP fusion, the GFP sequence was inserted after nucleotide 1152. Retention of the intron, which contains a stop codon 63 bp from the start, results in a 1272 bp coding sequence that encodes the 423 amino acid alternatively spliced protein AtBG_ppapas.
named AtBG_ppap (A. thaliana beta-1,3-glucanase_putative Pd-associated protein). Five peptides were identified leading to unique identification as At5g42100. This protein contains the glycosyl hydrolase 17 domain (pfam00332) between amino acids 27 and 346 (CD search http://www.ncbi.nlm. nih.gov/Structure/cdd/wrpsb.cgi; Marchler-Bauer and Bryant, 2004). The SOSUI computer program (http://sosui. proteome.bio.tuat.ac.jp/sosuiframe0E.html; Hirokawa et al., 1998) predicts that the protein contains two transmembrane regions between amino acids 9 and 31 and between amino acids 407 and 422. According to the SignalP program (http:// www.cbs.dtu.dk/services/SignalP; Bendtsen et al., 2004), the protein contains an N-terminal signal peptide, with a cleavage site between amino acids 26 and 27. Three different computer programs predict that AtBG_ppap has the essential characteristics to become glycosylphosphatidylinositol (GPI)-modified: big PI plant predictor (http://mendel.imp.univie.ac.at/sat/gpi/plant_server. html; Eisenhaber et al., 2003); DGPI (http://129.194.185.165/ dgpi/index_en.html; D. Buloz and J. Kronegg, University of Geneva, unpublished data) and GPI-SOM (http://gpi.unibe.ch; Fankhauser and Maser, 2005). This feature was recently confirmed experimentally by cleavage of the GPI anchor with phospholipases C and D (Elortza et al., 2003, 2006; ). The AtBG_ppap gene (Accession number: NM_123575) contains two exons and one intron (Figure 2). Exon 1 is 1209 bp long (403 amino acids) and ends immediately after the predicted GPI x site (the site of the C-terminal peptide cleavage), which is amino acid 401 (nucleotide 1203). The intron is 431 bp long, and the second exon, which contains the C-terminal signal, is 69 nucleotides long (23 amino acids). The resulting spliced coding sequence is 1278 bp, which encodes a protein containing 425 amino acids. A second version of the protein results from retention of the intron, which contains a stop codon 63 nucleotides after its beginning, resulting in a 1272 bp coding sequence (Acces-
Figure 3. Transcription levels of AtBG_ppap in various Arabidopsis tissues. Real-time RT-PCR measurements of AtBG_ppap, showing its relative transcription levels in various Arabidopsis tissues. Transcription is highest in the flowers and siliques, and lowest in the roots. SD is represented by vertical lines within the bars.
sion number: NM_203139) and a 423 amino acid protein. This alternatively spliced protein, which was entitled AtBG_ppapas, was not found to be GPI-anchored by the big-PI Plant Predictor, while both the DGPI and GPI-SOM programs predicted that it has the essential characteristics to become GPI-modified (Figure 2). Quantitative measurements of AtBG_ppap transcript levels in the Arabidopsis tissues using real-time RT-PCR revealed that the gene is not expressed equally in the different tissues (Figure 3). The transcription level of AtBG_ppap was the highest in flowers and siliques. In stem, rosette leaves and cauline leaves, transcript accumulation was considerably lower than in flowers and siliques. In roots, levels were negligible compared with the flowers and siliques (a 34-cycle RT-PCR confirmed that the gene is transcribed in all tissues, including roots; data not shown). Quantitative measurement of the alternatively spliced version of AtBG_ppap (AtBG_ppapas) showed that transcript accumulation was a few orders of magnitude lower than the AtBG_ppap form (data not shown). We therefore focused our research on the normally spliced form of AtBG_ppap, and cautiously suggest that the presence of AtBG_ppapas results from a low efficiency of normal splicing rather than a regulated process of any biological significance, a wellknown phenomenon in plants (Lorkovic et al., 2000). However, more experiments are required to confirm this assumption. AtBG_ppap:GFP localizes to Pd In order to verify plasmodesmal localization of AtBG_ppap, a fusion between the protein and GFP was created. As
ª 2006 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 49, 669–682
672 Amit Levy et al.
(a)
(b)
(a)
(d)
(b) (e) (c)
(d)
(c) (f)
Figure 4. Cellular localization of AtBG_ppap:GFP. (a) Transiently expressed AtBG_ppap:GFP in the epidermis of Nicotiana benthamiana 48 h after microprojectile bombardment. Fluorescence is seen along the periphery of the wall and in a punctate pattern (arrows). (b) Stably expressed AtBG_ppap:GFP in transgenic tobacco epidermis. Punctate fluorescence spans the walls and is seen as paired foci. The areas inside the white boxes are enlarged in the insets. (c,d) Spongy mesophyll cell of AtBG_ppap:GFP-expressing transgenic tobacco. Fluorescence is detected only in wall areas where there is cell–cell contact, and is absent from wall areas without cell–cell contact (d). To emphasize wall partitions, spongy mesophyll cells are also shown with the fluorescence channel turned off (c). Bar in (a) ¼ 10 lm, those in (b)–(d) ¼ 20 lm.
AtBG_ppap is a GPI-anchored protein and both the N- and C-terminal ends of its preprotein are removed, it was necessary to create a fusion protein with GFP within regions that remain after N- and C-processing. The coding sequence for GFP was thus inserted within the AtBG_ppap coding sequence immediately after amino acid 384, which is localized 17 amino acids before the x site (Figure 2). This region into which the GFP was inserted is described as a ‘disordered/unstructured’ and ‘low complexity’ region by the SMART program (http://smart.embl-heidelberg.de; Letunic et al., 2004), and is suspected to serve as a linker between the active glucanase domain and the GPI anchor. AtBG_ppap:GFP was expressed transiently in Nicotiana benthamiana source leaves by microprojectile bombardment (Figure 4a) and stably in Nicotiana tabacum by Agrobacterium transformation (Figure 4b). In both cases, the fusion protein appeared along the entire periphery of epidermal leaf cells, and in a punctate pattern (Figure 4a,b) or as pairs of fluorescent foci on opposite sides of the cell wall (enlarged boxes in Figure 4b), a pattern characteristic of Pd. Spongy mesophyll cells have regions with no cell–cell contact, where walls face an intercellular space and are
Figure 5. AtBG_ppap:GFP in transgenic tobacco co-localizes with aniline blue-stained callose present around plasmodesmata. Aniline-blue stained callose is shown in blue (a), AtBG_ppap:GFP is shown in red (b), and both are shown overlaid (c) in a section of cell wall between epidermal cells. (d)–(f) are enlargements of the areas inside the boxes in (a)– (c), respectively. Bars ¼ 20 lm.
devoid of Pd, and regions with cell–cell contact, where walls contain Pd. AtBG_ppap:GFP fluorescence was almost undetected within cell wall regions where there is no cell–cell contact (Figure 4c,d). This differential labeling of walls further indicates that AtBG_ppap:GFP is associated with Pd or with the wall sheath surrounding Pd. To verify that the AtBG_ppap:GFP fluorescence foci inside cell walls indeed represent Pd, we stained callose using aniline blue. Callose has been widely used as a plasmodesmal marker (Baluska et al., 1999; Bayer et al., 2004; Gorshkova et al., 2003; Sagi et al., 2005 and more). When transgenic tobacco source leaves expressing AtBG_ ppap:GFP were stained by aniline blue, AtBG_ppap:GFP colocalized with the aniline blue-stained callose present around Pd (Figure 5). Control experiments verified that no GFP fluorescence is seen under aniline blue conditions and no aniline blue fluorescence is seen under GFP conditions. AtBG_ppap:GFP is a membrane protein As AtBG_ppap is a GPI-anchored protein, it is expected to target both the ER and the plasma membrane (Udenfriend and Kodukula, 1995). Indeed, when AtBG_ppap:GFP is transiently expressed by Agrobacterium leaf injection, it can be seen to accumulate in ER membranes (Figure 6a). Likewise, as predicted, AtBG_ppap:GFP also localizes to the plasma membrane. AtBG_ppap:GFP transgenic leaves were stained
ª 2006 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 49, 669–682
A plasmodesmal b-1,3-glucanase 673
(a)
(b)
(c)
(d)
(e)
Figure 6. AtBG_ppap:GFP localization in the ER and plasma membranes. (a) A projection of several optical sections of transiently expressed AtBG_ppap:GFP in N. benthamiana plasmolysed epidermal cell 48 h after Agrobacterium infiltration, showing the localization of AtBG_ppap:GFP in the ER membrane system. (b)–(e) AtBG_ppap:GFP fluorescence (shown in green) co-localizes with FM4-64-stained plasma membrane (shown in red) in plasmolysed transgenic tobacco epidermal cells. (b) Overlay of all channels showing that AtBG_ppap:GFP and FM4-64 co-localize (see arrows), indicating that AtBG_ppap:GFP is a plasma membrane protein. (c)–(e) show separate channels of (b). (c) Nomarsky differential interference contrast (DIC) showing the cell wall (CW). (d) DIC and FM4-64 fluorescence overlay. Upon plasmolysis, plasma membrane (PM) withdraws from the cell wall. (e) DIC and GFP fluorescence overlay. Upon plasmolysis, AtBG_ppap:GFP withdraws from the cell wall as well, in the same pattern as the plasma membrane. Bars ¼ 10 lm.
with FM4-64 (N-[3-triethylammoniumpropyl]-4-[p-diethylaminophenylhexatrienyl] pyridinium dibromide), a plasma membrane marker (Bloch et al., 2005), and cells were plasmolysed. Following plasmolysis, AtBG_ppap:GFP could be seen to recede from the cell wall along with the FM4-64 fluorescently stained plasma membrane (Figure 6b–e). GFP cell-to-cell spread is slower in AtBG_ppap mutant plants It is hypothesized that non-selective plasmodesmal conductivity is a function of the steady-state level of callose accumulation in the wall sleeve surrounding Pd, and that this steady-state level is the net result of glucanase versus synthase activities (Heinlein and Epel, 2004). Assuming no change in callose synthase activity, a lower b-1,3-glucanase activity in the Pd is expected to increase callose levels and reduce cell-to-cell transport. To test this hypothesis, we measured the diffusion of GFP between leaf epidermal cells in wild-type and AtBG_ppap knockout mutants. As GFP movement between cells is diffusive (Crawford and Zambryski, 2000; Liarzi and Epel, 2005; Oparka et al., 1999), it is expected that, in AtBG_ppap knockout mutant plants, there will be less callose hydrolysis, resulting in a narrowing of the cytoplasmic sleeve and hence a slower diffusion of GFP. Two independent Arabidopsis lines carrying a T-DNA insertion that disrupts the AtBG_ppap locus (SALK_
019116.47.55.x, Alonso et al., 2003; SAIL_115_G04, Sessions et al., 2002) were employed to test this hypothesis. PCR analysis and sequencing of the two T-DNA lines established that the insertion in SALK_019116.47.55.x disrupted the gene at 344 bp downstream of the ATG start codon, and that the insertion in SAIL_115_G04 is located 625 bp from the ATG start (Figure 7a; data not shown). Plants of the two lines, homozygous for the insertion, do not transcribe AtBG_ppap RNA, as determined by RT-PCR (Figure 7b). Epidermal cells of vegetative wild-type and mutant leaves were transfected with pGFP by microprojectile bombardment of gold particles, a method that wounds the leaf and induces callose formation (Hunold et al., 1994). Forty-eight hours postbombardment, the number of the cells in a cluster created by GFP movement was analyzed by scanning the epidermal cells using a confocal laser scanning microscope (CLSM; see Experimental procedures). The cells counted are those showing green fluorescence in the nucleus and the cytoplasm (Figure 8). A box-plot analysis, comparing the sizes of the GFP clusters in leaves of wild-type and knockout mutants, shows that, in wild-type plants, the diffusion of GFP is more extensive (Figure 8). A one-way ANOVA analysis revealed no statistical difference between SALK and SAIL mutants (P ¼ 0.949), but showed a statistically significant difference between the mutants and wild-type (P < 0.05). In a second analysis, the exponential decay parameter b and the coefficient of conductivity of Pd, C(Pd), were
ª 2006 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 49, 669–682
674 Amit Levy et al. Table 1 Changes in Pd conductivity in AtBG_ppap-deficient mutants
(a)
(b)
Plant type
Mean ba
SEb
GFP
C(Pd)c
Wild-type SALK mutant SAIL mutant
)1.41 (31)d )1.66 (31) )1.88 (15)
0.072 0.068 0.194
0.7 0.59 0.52
a Average decay parameter b, also termed impedance, which describes the slope of the gradient formed by differences in plasmodesmal conductivity. b SE, SEM b value. c The coefficient of conductivity of plasmodesmata for the cell-to-cell spread of GFP, calculated as )1/mean b value. d The number in parentheses is the number of cells analyzed.
GFP
Figure 7. Identification of AtBG_ppap T-DNA insertion lines. (a) A schematic model showing the T-DNA insertions in SALK and SAIL mutants (SALK_019116.47.55.x and SAIL_115_G04 respectively). Numbers indicate the coding sequence location of the insertions. (b) RT-PCR of the homozygous SALK mutant, homozygous SAIL mutant, wildtype (WT) and heterozygous SAIL mutant (Het) Arabidopsis leaves. The upper panel shows the amplification of AtBG_ppap coding sequence. No amplification is found in the homozygous mutants. The lower panel shows the amplification of control gene Ubiquitin10.
calculated for GFP spread in vegetative leaves of the wildtype and AtBG_ppap knockout mutants, according to the protocol described by Liarzi and Epel (2005) (Table 1). The
Figure 8. Movement of bombarded GFP between Arabidopsis epidermal cells in wild-type and AtBG_ppap-deficient mutants. (a) A box-plot diagram of the GFP cluster sizes (determined by the number of cells in a cluster), two days after bombardment, in SALK and SAIL mutants, and in wild-type plants. The plot represents the statistical distribution of the results. The thick lines across the boxes represent the median cluster sizes, and the ends of boxes indicate quartile results. Circles represent outliers. The median cluster sizes are 6.5 for the SALK mutant (n ¼ 36), 8 for the SAIL mutant (n ¼ 62) and 15 for wild-type (n ¼ 97). (b)–(d) Average GFP clusters of SALK mutants (b), SAIL mutants (c) and wild-type plants (d), 48 h after GFP bombardment. GFP is produced in the intensely fluorescent cell and moves to the neighboring cells that show weaker fluorescence. Bar ¼ 50 lm.
(a)
(b)
C (Pd) value in wild-type plants was found to be higher than the GFPC (Pd) value in the SALK and SAIL mutants (0.70, 0.59 and 0.52, respectively). A one-way ANOVA of b showed no difference between the SALK and SAIL mutants, but showed a statistically significant difference between the mutants and the wild-type (P < 0.05). The results suggest that the conductivity of Pd is downregulated in AtBG_ppapdeficient mutants. To test whether the decreased cell-to-cell movement of GFP results from increased callose accumulation in Pd, we examined the amount of Pd callose in the epidermis of Arabidopsis vegetative source leaves 2 days after wounding, employing an aniline blue fluorescence assay. Quantification of 1487 fluorescent sites in AtBG_ppap-deficient
(c)
(d)
ª 2006 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 49, 669–682
A plasmodesmal b-1,3-glucanase 675
(a)
(b)
Figure 9. Callose accumulation in AtBG_ppap-deficient mutants and wildtype Arabidopsis plants in response to wounding. Epidermal cells of (a) an AtBG_ppap knockout mutant and (b) a wild-type plant, stained with aniline blue two days after wounding and observed under a fluorescent microscope. The fluorescence intensity of callose deposits around Pd was measured by drawing a small circle around Pd foci [see circle in (a)] and determining the minimum fluorescence intensity inside the circle (background) and the maximum intensity of the Pd foci (maximum) using the IMAGEJ software. Pd intensity was determined by subtracting the background from the Pd maximum intensity [for the circle in (a), the maximum Pd intensity is 143, the minimum intensity is 103, therefore the Pd intensity is 40]. Two days after wounding, the mean Pd callose intensity was 40.4 0.42 (mean intensity SEM in arbitrary units) in AtBG_ppap knockouts and 27.9 0.404 in wild-type plants (n ¼ 1487 and 779, respectively). Bar ¼ 30 lm.
mutants and 779 fluorescent sites in wild-type plants revealed that callose accumulation was 45% higher in AtBG_ppap mutants (Figure 9). An independent-samples t test showed the results to be significantly different at P < 0.001. Although the background levels varied between repeats with different leaves, the net intensity in the Pd was the same. These results are in agreement with the hypothesized role of callose in regulating Pd function, and support the hypothesis AtBG_ppap functions as a plasmodesmal protein. Discussion Identification of a Pd-associated Arabidopsis b-1,3-glucanase b-1,3-glucanases, enzymes that catalyze the hydrolysis of b1,3-glucan linkages, hydrolyse mostly callose in plants, and play a role in various important physiological processes. For example, they are found in the pistils, where callose is known to be abundant in the growing pollen tube wall and as plugs behind the growing tube protoplast, which presumably maintains the two nuclei in the proximity of the tube tip (Kauss, 1996). In the style, developmentally regulated b-1,3-glucanases were hypothesized to function in regulating pollen tube growth or in defense against pathogen attack during fertilization (Delp and Palva, 1999; Ori et al., 1990). In anthers, b-1,3-glucanases were found to be expressed just before microspores are released, where they function in degradation of the callose that surrounds the microspore tetrad, and contribute to the release of the
microspores as pollen grains (Hird et al., 1993; Tsuchiya et al., 1995). Some b-1,3-glucanases are involved in stress processes through their function as ‘pathogenesis-related‘ proteins (Leubner-Metzger and Meins, 1999; Stintzi et al., 1993). b-1,3-glucanase enzymes have been localized to the cell wall, the intercellular space, the plasma membrane and the vacuole (Benhamou et al., 1989; Hu and Rijkenberg, 1998; Keefe et al., 1990). During dormancy release in birch (Betula pubescens), b-1,3-glucanases were found peripherally in spherosome-like vacuoles, in close proximity to the Pd (Rinne et al., 2001). In Arabidopsis, five b-1,3-glucanase genes have been studied. BG2 and BG3 are 37 and 30 kDa acidic proteins, respectively, whose genes are upregulated after pathogen infection. BG2 is located extracellularly, and both proteins are purported to function as part of the plant defense system (Dong et al., 1991; Uknes et al., 1992). A6 is a 53 kDa basic protein whose gene is expressed in the tapetum cells in anthers just before microspores are released (Hird et al., 1993). BG4 and BG5 are 38 kDa proteins with an acidic and basic pI, respectively. BG4 is expressed in the style and septum of the ovary, and may play a role in the plant reproductive process (Delp and Palva, 1999). In tobacco, based on their sequences homologies, the b-1,3-glucanase proteins have been classified into three classes. Class I are basic proteins localized in the vacuole of mesophyll and epidermal cells, while classes II and III are acidic, extracellular isoforms (Beffa and Meins, 1996). Here we describe the isolation and characterization of a 45 kDa b-1,3-glucanase present in an enriched Pd fraction from Arabidopsis and termed AtBG_ppap. AtBG_ppap was identified by MS/MS analysis of a tryptic digest of a protein isolated from a Pd-enriched fraction. The mature form of AtBG_ppap with its N- and C-terminal ends removed (see below) is predicted to be a neutral protein (pI 6.18), and thus does not fit into the tobacco basic/acidic classification scheme. According to the Arabidopsis Information Resource (TAIR), the Arabidopsis genome contains 48 b-1,3-glucanase genes. A bioinformatic analysis, using a sensitive prediction program for the compatibility of plant proteins with GPI lipid-anchoring motif requirements (big-PI Plant Predictor) (http://mendel.imp.univie.ac.at/sat/gpi/plants/pred/ plants.athal.class.html#glyco), predicted that 18 Arabidopsis b-1,3-glucanase proteins, including AtBG_ppap (At5g42100), are GPI-anchored. GPI-anchored proteins undergo post-translational modifications, in the course of which both their C-terminal and N-terminal ends are spliced. The tobacco class I isoforms are also synthesized as prepro-enzymes, which similarly undergo post-translational processing in which their C- and N-terminal ends are removed (Beffa and Meins, 1996). However, the big-PI Plant Predictor did not find the class I GLA protein to be GPI-anchored. Alignment (Corpet, 1988) between the
ª 2006 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 49, 669–682
676 Amit Levy et al. predicted GPI-anchored b-1,3-glucanases of Arabidopsis (see Figure in Supplementary Material) shows that AtBG_ppap shares the highest homology with At1g32860 (61% identity and 73% similarity). Both of these proteins contain a C terminus that is rich in glycine and serine, and is described as a ‘disordered/unstructured’ region by the SMART program (http://smart.embl-heidelberg.de; Letunic et al., 2004). The C terminus of both proteins is clearly distinguishable from that of the other proteins in this group, and may indicate a unique property of these proteins.
1,3-glucanase-deficient mutant plants, reduced trafficking of fluorescent probes between cells and induced callose deposition was also found, although callose intensity was not measured around the Pd (Iglesias and Meins, 2000). An important difference, however, exists in the intercellular localization found for these proteins. While AtBG_ppap was found to be localized in the Pd, tobacco class I b-1,3glucanases were found to be vacuolar proteins (Keefe et al., 1990). It is possible that some members of the class I family are indeed primarily vacuolar, but other undetected members may localize in Pd.
A novel plasmodesmal b-1,3-glucanase Several lines of evidence presented here support the hypothesis that AtBG_ppap is a Pd-associated protein. First, the protein was identified in a Pd-enriched fraction from Arabidopsis. Secondly, AtBG_ppap:GFP fluorescence appears in a punctate pattern and as pairs of fluorescent foci on opposite sides of the cell walls (Figure 4), a pattern similar to that of plasmodesmal marker MPTMV:GFP or the Pd-associated protein AtRGP2:GFP (Sagi et al., 2005). Third, AtBG_ ppap:GFP fluorescence co-localizes with aniline-blue stained callose present around Pd (Figure 5). In addition to plasmodesmal localization, our results indicate that AtBG_ppap also has a plasmodesmal function. Monitoring the movement of GFP between cells in AtBG_ ppap knockout mutants and wild-type plants revealed that GFP diffused more slowly in AtBG_ppap mutant plants. The number of cells into which GFP diffused from a GFPtransformed cell is approximately twice as high in wild-type as in the AtBG_ppap-deficient mutants (Figure 8), and the conductivity of Pd [GFPC (Pd)] in these mutants is downregulated (Table 1). These results strongly suggest that the Pd are in a more ‘closed’ state in AtBG_ppap knockout mutants, resulting in reduced cell-to-cell spread of GFP. Using an assay for the direct measurement of callose present around the Pd, we found that, after wounding induced by cutting leaves, the level of callose around Pd was 45% higher in AtBG_ppap knockout mutant plants than in wild-type controls (Figure 9). These results show that AtBG_ppap deficiency enhances callose accumulation in response to wounding. Several studies have shown that accumulation of callose around Pd results in decreased cellto-cell movement of fluorescent dyes (Radford and White, 2001; Sivaguru et al., 2000). Taken together, our results strongly suggest that AtBG_ppap functions in degrading the callose associated with Pd, and that the reduced cell-to-cell spread of GFP and the elevated callose levels in AtBG_ppap mutant plants reflect a decrease in callose degradation. Our results also support the hypothesis that the level of steadystate callose accumulation around the Pd is a function of b1,3-glucan synthase versus glucanase activities. The functional results we obtained are similar to those described for class I b-1,3-glucanases in tobacco. In class I b-
GPI anchoring of AtBG_ppap in the Pd Bioinformatic analysis predicts that AtBG_ppap is a GPI-anchored protein. This prediction was confirmed by two different proteomic analyses aimed at identifying GPI-anchored proteins in Arabidopsis (Elortza et al., 2003, 2006). In these analyses, membrane fractions were treated with phosphatidylinositol phospholipase C (PI-PLC) (Elortza et al., 2003) and phospholipase D (PLD) (Elortza et al., 2006) enzymes in the presence of Triton X-114. These enzymes hydrolyse the phosphatidylinositol, releasing a soluble GPI protein from the membrane. Proteins that were enriched in the aqueous phase upon PI-PLC or PLD treatments were identified by liquid chromatography MS/MS. AtBG_ppap (At5g42100) was identified as GPI-anchored in both analyses. GPI anchoring, found in all eukaryotic organisms, is a way to attach proteins to the outer leaflet of the plasma membrane (Udenfriend and Kodukula, 1995). GPI-anchored proteins have a cleavable N-terminal secretion signal for translocation into the ER. They also have a C-terminal transmembrane region that is thought to function as a signal for transamidase, which cleaves the C-terminal hydrophobic peptide and transfers the protein to a prefabricated GPI anchor. From the ER, the GPI-anchored protein is transported to the cell membrane via the Golgi (Udenfriend and Kodukula, 1995). Our results confirm that AtBG_ppap:GFP indeed localizes to the ER and the plasma membrane. GPI anchoring can serve as a link between the plasma membrane and the cell wall, as the protein is facing the wall while attached to the membrane. Callose deposition is hypothesized to occur in a collar around Pd and to push the membrane inward, reducing the channel diameter (Olesen and Robards, 1990; Radford et al., 1998). The proposed GPIanchored localization of AtBG_ppap in the outer plasma membrane of the Pd facing the wall is therefore very appropriate in terms of function. It will be very interesting to determine whether GPI anchoring in the Pd is a wider phenomenon with functional significance. At present, the mechanism by which AtBG_ppap accumulates in the Pd is unknown. This targeting may be the result of a special targeting mechanism or an internal property of the anchored enzyme, which holds it in the
ª 2006 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 49, 669–682
A plasmodesmal b-1,3-glucanase 677 channels. Alternatively, it may be that its accumulation in Pd is simply the result of much lower rates of turnover in the protected environment within the Pd. Possible roles of AtBG_ppap in symplasmic regulation Our results suggest that AtBG_ppap is active in degrading callose around Pd and in widening the channels. Closure or sealing of Pd by callose accumulation and their re-opening by callose breakdown mediated by b-1,3-glucanase provide an important regulation mechanism for the symplasmic communication between cells (Rinne and van der Schoot, 2003). This regulation has a crucial role in both development and defense, and may indicate possible roles for AtBG_ppap. For example, in birch, a perennial plant, exposure to short days induces dormancy and Pd closure by specific structures around the Pd channel entrances within the shoot apical meristem (SAM), resulting in symplasmic isolation of the cells and the blockage of signaling networks (Rinne and van der Schoot, 1998; Rinne et al., 2001). These structures contain callose, as determined by immunolocalization and tannic acid staining, which visualize putative glucan synthase complexes (Rinne and van der Schoot, 1998; Rinne et al., 2001). After adequate exposure to cold, dormancy is broken and the SAM regains its potential. In this process, symplasmic connection is restored and callose disappears (Rinne et al., 2001). Rinne et al. have found that b-1,3glucanase proteins were upregulated during this process, and that b-1,3-glucanases were localized in spherosome-like vacuoles or lipid bodies in vicinity of the cell membrane and in contact with Pd. They suggested that these b-1,3-glucanases are delivered to the Pd and take part in degrading the plasmodesmal callose sphincters. It would be of interest to determine whether, upon release from dormancy in perennial plants, an AtBG_ppap homologue is involved in callose degradation, and whether this regulation is at the transcriptional or translational level. Symplasmic blockage also occurs during the hypersensitive response to certain pathogens, with the apparent aim of restricting the spread of the invading pathogen. During this response, callose is deposited at the cell walls around the infection sites (Kauss, 1996). A recent paper studying tobacco plants constitutively expressing the movement protein (MP) of tomato spotted wilt virus (Rinne et al., 2005) shows that, in MP-expressing cells in source leaves, Pd are obstructed by the formation of sphincters that contain callose deposits. This obstruction was accompanied by a reduction in cell-tocell movement. Temperature shift treatments (from 22 to 32C) restored the transport capacity through Pd, and levels of b-1,3-glucanase were found to increase. The authors suggest that the removal of callose from the Pd of MP-expressing cells is mediated by b-1,3-glucanase. It should be noted that an increase in callose accumulation as a result of pathogen-induced stress is not necessarily due
solely to an increase in callose synthase activity alone, but could be the result of a shift in the balance between callose synthesis and callose hydrolysis. Thus, a role is suggested for Pd b-1,3-glucanase during viral pathogenesis in both symplasmic blockage and symplasmic recovery. The hypothesis that the activity of b-1,3-glucanase breaks down the callose around Pd and thus promotes cell-to-cell movement is further supported by the finding that increased levels of class I b-1,3-glucanase promote virus spread, while decreased levels in tobacco mutants result in decreased virus spread (Bucher et al., 2001; Iglesias and Meins, 2000). The regulation of symplasmic communication by Pd is of central importance in plant development, involving temporal regulation of different developmental domains (Oparka et al., 1999; Rinne and van der Schoot, 2003; Zambryski, 2004). With the identification of AtBG_ppap as a Pd-associated protein (henceforth to be annotated as AtBG_pap), it will be possible to analyze its role in these processes. Additionally, as callose accumulation is dependent on the equilibrium between callose synthase and callose hydrolysis, another major future goal will be the identification of a Pd-associated b-1,3-glucan synthase, the suggested ‘second half’ of this plasmodesmal regulatory system.
Experimental procedures Plant material Nicotiana benthamiana and N. tabacum cv. Samson plants were grown in 10 cm pots in a mix of equal volumes of potting mixture and vermiculite (Pecka Hipper Gan, Rehovot, Israel) at 25C under long-day conditions (16 h light/8 h dark cycles). Arabidopsis thaliana ecotype Columbia plants were grown in potting mixture in 6 cm pots at 22C under long-day conditions. For measurement of callose levels, plants were grown under short-day conditions (8 h light/16 h dark).
Cell wall preparation Four- to six-week-old A. thaliana cry2 mutants grown under longday conditions were harvested and stored at )80C. The lateflowering cry2 mutant was used in order to obtain a maximum amount of plant material. Plants were pulverized in liquid nitrogen, and homogenized with 2 ml g)1 tissue of buffer A (0.25 M sucrose, 4 mM EDTA, 10 mM EGTA, 20 mM Tris/HCl, pH 8.5, 0.02% azide) containing a cocktail of protease inhibitors [1.5 lM aprotonin, 0.01 units/ml a2-macroglobulin, 2 mM phenylmethyl-sulfonyl fluoride (PMSF) (Roche, Mannheim, Germany), 42 lM leupeptine, 14.5 lM pepstatine (Sigma-Aldrich, http://www.sigmaaldrich.com/), 2.5 mM 1.10 orthophenantroline (Merck, www.merck.de) and 14 lM E64 (Fluka, www.sigmaaldrich.com/Brands/Fluka_Riedel_Home.html)]. The homogenate was filtered through a 16 lm nylon cloth, and the wall fraction (WF) retained on the nylon filter was re-homogenized with 2 ml g)1 (starting tissue) of buffer B (0.25 M sucrose, 2 mM EDTA, 10 mM EGTA, 20 mM Tris/HCl, pH 7.5, 0.02% azide) and protease inhibitors as above. The homogenate was re-filtered through a 5 lm nylon cloth, and the WF was re-suspended with 2 ml g)1 tissue of buffer B and protease inhibitors as above.
ª 2006 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 49, 669–682
678 Amit Levy et al. The WF kept at 0–4C was sonicated (560 W and a 0.5 inch flat tip) for 10–15 min, with 5 sec on/5 sec off pulses. The WF was then centrifuged at 10 000 g for 2 min at 4C, and the wall pellet was resuspended with buffer B containing PMSF and 1.10 orthophenantroline proteases at the concentrations above. Sonication was repeated, and the WF was pelleted at 1000 g for 2 min, the supernatant discarded and the pellet compacted at 10 000 g for 1 min, and supernatants discarded. The WF was resuspended and resonicated three or four additional times as described for the second sonication, until no green supernatant was observed and the wall pellet was a grayish color. After the fourth sonication, the suspension was filtered through a plastic kitchen sieve to remove large unbroken pieces of wall (mainly vascular tissue). After the last sonication and centrifugation, the pellet was re-suspended in two volumes of buffer B without proteases and filtered through a 5 lm nylon cloth. The WF captured on the nylon filter was then centrifuged at 10 000 g for 1 min to compact the wall, and stored at )80C. All procedures were performed at 4C.
Pd preparation Cellulase preparation. Digestion buffer (10 mM MES, pH 5.5, 4.4% mannitol, 0.02% azide) was warmed to 55C, and 0.6% w/v of Cellulase R10 (Karlan www.karlan.com) was added and stirred until dissolved. The mixture was heated at 55C for 5 min, filtered through 6 lm nylon cloth and cooled to 4C. Protease inhibitors were added as described above for cell wall preparation, and the solution was centrifuged (75 600 g for 40 min at 4C) to clear the solution of any particulate matter. Wall digestion. WF was washed in 1 ml g)1 (fresh tissue) of digestion buffer, and centrifuged at 8000 g for 1 min at 4C. The washed wall was then re-suspended in 1 ml g)1 (fresh tissue) of heat-cured cellulase solution (the enzyme solution was prewarmed to 37C) and vortexed until no clumps were visible. The solution was homogenized using a Teflon glass homogenizer, transferred to an Erlenmeyer flask and gently shaken in a rotary shaker (100 r.p.m.) at 37C for 1.5 h. The solution was then centrifuged at 5860 g for 5 min at 4C. Both the supernatant and the pellet were collected separately. The supernatant (Sup I) was adjusted to pH 7 with 18 ll ml)1 1 M Tris, pH 8. The pellet was resuspended in 1 ml g)1 tissue of MMA buffer (10 mM 3-[N-morpholino] propane-sulfonic acid [MOPS], pH 7.5, 4.4% mannitol, 0.02% azide), homogenized and centrifuged as before (5860 g, 5 min, 4C), and the supernatant collected (Sup II). Sup I and Sup II were combined and filtered through a 6 lm nylon cloth, and the filtered supernatant was centrifuged at 75 600 g for 40 min. The pellet was resuspended in 100 ll g)1 (fresh tissue equivalent) of MOPS washing buffer (10 mM MOPS, pH 7.5, 0.02% azide) using a Teflon glass homogenizer, and centrifuged twice at 7 000 g for 5 min at 4C in Corex tubes. The supernatant containing Pd was then centrifuged at 75 600 g for 40 min at 4C and the supernatant was discarded. The compacted Pd pellet was stored at )80C. PAGE separation The Pd fraction (equivalent of 50 g plant tissue) was dissolved in 50 ll dissolving buffer (9 M urea, 40 mM Tris, 2% SDS, 50 mM DTT and 20% glycerol). The sample was incubated at room temperature for at least 1 h and separated by one-dimensional SDS–PAGE on a 12.5% gel. Gels were stained with PhastGelTM BlueR (Amersham
Pharmacia, www.amershambiosciences.com) solution for 1 h on a rotary shaker (25 r.p.m.). Mild distaining was carried out with 7% acetic acid in 5% methanol. Several changes of distaining solution were required till the background was cleared. Bands were excised from the gel and used for protein sequencing.
In-gel proteolysis, chromatography and mass spectrometry Coomassie blue-stained bands were cut and in-gel reduced with 10 mM DTT, incubated at 60C for 30 min, alkylated with 10 mM iodoacetamide at room temperature for 30 min, and proteolysed with trypsin overnight at 37C, using modified trypsin (Promega, http://www.promega.com/) at a 1:100 enzyme-to-substrate ratio. The tryptic peptides were resolved by reverse-phase chromatography on a 1 · 150 mm C18 column (Vydac, www.vydac.com). The peptides were eluted using linear 80 min gradients from 5% to 95% acetonitrile containing 0.1% acetic acid. Mass spectrometry was performed wusing ion-trap mass spectrometers (LCQ; Thermo) operating in the positive mode using repetition of a full MS scan followed by CID of the most dominant ion selected from the first MS. The MS data ware compared to simulated proteolysis and CID of the proteins in the NRNCBI database using SEQUEST software (LCQ; Thermo).
RT-PCR Source rosette leaves and stems were cut from newly flowering Arabidopsis plants, while cauline leaves, roots, flowers and siliques were cut from fully flowering plants. For each organ type, two samples were collected, with two repeats, and stored at 80C. RNA was purified from 50–100 mg frozen tissue using the SV total RNA isolation system kit according to manufacturer’s instructions (Promega). cDNA first-strand synthesis was performed as previously described (Caldelari et al., 2001). AtBG_ppap and Ubiquitin10 coding sequences were amplified by PCR reactions using specific primers: for AtBG_ppap, forward primer 5¢-CCGATAACCATGGCTTCTTCTTCTCTGCAGTC-3¢ and reverse primer 5¢-CTATCATCCTAGGTTACAACCGAAGCTTGATGATGCAAAG-3¢; for Ubiquitin10, forward primer 5¢-CGATTACTCTTGAGGTGGAG-3¢ and reverse 5¢-AGACCAAGTGAAGTGTGGAC-3¢. AtBG_ppap was amplified using the following program: 1 cycle of 95C for 2 min, followed by 33 cycles of 94C for 15 sec, 58C for 30 sec and 72C for 1 min, and a final elongation step of 4 min at 72C. Ubiquitin10 was amplified using the following program: 1 cycle of 95C for 4 min, followed by 35 cycles of 94C for 20 sec, 55C for 30 sec and 72C for 30 sec, and a final elongation step of 5 min at 72C.
Real-time PCR Real-time RT-PCR of AtBG_ppap was performed in a fluorescence temperature cycler (LightCycler; Roche, www.roche.com), using actin 2 þ 8gene primers for normalization. cDNA corresponding to 50 ng of RNA served as a template in a 10 ll reaction containing 5 pM gene-specific primers and 5 ll of QuantiTect SYBR green PCR mix (Qiagen, http://www.qiagen.com/). The samples were loaded into capillary tubes and incubated in the fluorescence thermocycler (LightCycler) for an initial denaturation of 15 min at 95C. The PCR reaction for AtBG_ppap consisted of 45 cycles of 15 sec at 95C, 20 sec at 55C, 23 sec at 72C and 5 sec at 82C for measurement of SYBR Green fluorescence. The reaction for actin 2 þ 8 consisted of 45 cycles of 15 sec at 95C, 20 sec at 55C and 12 sec at 72C. The actin 2 þ 8 SYBR Green fluorescence was measured at the end of each cycle. The amplification of specific transcripts was confirmed
ª 2006 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 49, 669–682
A plasmodesmal b-1,3-glucanase 679 by producing melting curve profiles at the end of each run and subjecting the amplification products to electrophoresis on an agarose gel. The primers used for AtBG_ppap were: forward 5¢CCCCAAACACGTTTCT-3¢ and reverse 5¢-CGTATAGGCGTCCTCA-3¢, and those for actin were: forward 5¢-GGTAACATTGTGCTCAGTGGTGG-3¢ and reverse 5¢-AACGACCTTAATCTTCATGCTGC-3¢.
Construction of plant expression plasmids For the expression of AtBG_ppap in plants, we used the pRTL2/ newGFPm (pGFP-MRC) plasmid (Rodriguez-Concepcion et al., 1999). Because AtBG_ppap is a GPI-anchored protein, and both its N- and C-terminal ends are cleaved post-translationally, its coding sequence was divided into two parts, 1–1152 (Glu-N¢) and 1153– 1278 (Glu-C¢), which were cloned at both sides of the GFP (Figure 2). Glu-C¢ was PCR-amplified using forward primer 5¢-CATTAGTGAGCTCTATCAGCCAGTCACGGGTAACC-3¢ containing a SacI site, and reverse primer 5¢-CTATCATCCTAGGTTACAACCGAAGCTTGATGATGCAAAG-3¢ containing an AvrII site, and cloned between the SacI and XbaI sites in pGFP-MRC, in-frame with GFP at its 3¢ end (XbaI and AvrII generate the same 5¢ protruding end). GluN¢ was PCR-amplified using forward primer 5¢-CCGATAACCATGGCTTCTTCTTCTCTGCAGTC-3¢ and reverse primer 5¢CGAATTACCATGGAGATGCCACCACCGCTGGA-3¢, both containing a NcoI site, and cloned in-frame into the unique NcoI site present at the 5¢ beginning of the GFP coding sequence. For expression in plants, the plasmid was digested using SphI, and the resulting cassette, containing the CaMV 35S promoter, AtBG_ppap fused to GFP, and the nos transcriptional terminator, was purified from the agarose gel and its protruding 3¢ ends cleaved with T4 polymerase (Fermentas, www.fermentas.com) to give blunt ends. This cassette was subcloned into the SmaI site in pCAMBIA 2300 (CAMBIA, www.cambia.org) to create plasmid pCambiaAtBG_ppap:GFP.
Expression in plants Stable expression of pCambiaAtBG_ppap:GFP in N. tabacum cv. Samson plants using a modified leaf disk method, and transient expression of pCambiaAtBG_ppap:GFP in Nicotiana benthamiana plants using Agrobacterium tumefaciens leaf injection were both performed as described by Sagi et al. (2005). Particle bombardment of pCambiaAtBG_ppap:GFP into vegetative N. benthamiana source leaves was performed as described below.
FM4-64 membrane staining and plasmolysis Plasma membrane staining was performed by incubating 7 · 3 mm pieces of AtBG_ppap:GFP transgenic source leaves in 1:500 diluted N-[3-triethylammoniumpropyl]-4-[p-diethylaminophenylhexatrienyl] pyridinium dibromide (FM4-64; Molecular Probes, www.probes. invitrogen.com) in double-distilled water, for 1.5 h. For plasmolysis, the FM4-64-stained leaf discs were transferred to 0.75 M mannitol solution, and incubated for 10 min. Plasmolysis was determined by monitoring the retraction of FM4-64-stained plasma membrane from the cell wall.
Mutant identification Arabidopsis T-DNA insertion lines (SALK_019116.47.55.x and SAIL_115_G04) were obtained from the Arabidopsis Biological Resource Center, Ohio State University, Columbus, OH, USA. The presence and location of each T-DNA insertion were determined by
PCR analysis using primers for the left border of the insertion and the gene sequence. For the SALK mutant, the left border primer LBa1 (Alonso et al., 2003) and the Glu-N¢ forward primer (see above) were used, and for the SAIL mutant, the left border primer LB3 (Sessions et al., 2002) and the Glu-N¢ reverse primer (see above) were used. Then, to identify homozygous plants, the Glu-N¢ sequence was amplified (see above) and plants with no PCR product were selected. Verification of the homozygous plants was carried out by RT- PCR performed with leaf RNA, and PCR amplification of the AtBG_ppap coding sequence. Ubiquitin10 was used as a positive control (see above).
Particle bombardment and GFP mobility assays Bombardment of GFP into leaves for GFP movement assays was performed as described by Liarzi and Epel (2005), except that source rosette leaves from 3-week-old Arabidopsis plants which had not yet bolted were used (about six leaves were placed on the agar plate for each bombardment), and 1 lm gold particles were used instead of tungsten. To determine the number of cells into which GFP diffused, cells of the lower epidermis of the leaf were analyzed 48 h postbombardment using a CLSM (LSM 510; Zeiss, http://www.zeiss.com/). For each transformed cell, GFP intensity in the nucleus was set just below saturation by setting detector gain values, in fixed laser and amplifier gain values (5% and 1 respectively). Cell clusters were analyzed after amplification by amplifier gain and by laser over a ninefold range. These amplifications were shown to be in the linear range of the instrument (Liarzi and Epel, 2005). GFP movement was determined by counting the cells around the transformed cell that clearly showed fluorescence in the nucleus and the cytoplasm. Cells that synthesized GFP but showed no movement were not analyzed in order to avoid including seriously wounded cells. Calculation of the coefficient of conductivity of Pd was performed in Arabidopsis as described by Liarzi and Epel (2005), with modifications: instead of analyzing all first-degree cells (cells with a common cell wall with the transformed cell; also termed ‘cell 1’) and second-degree cells (cells with common cell wall with cell 1 but not with the transformed cell; also termed ‘cell 2’), the analysis was performed on four cells of each degree. The cells selected are those located above, below and on both sides of the transformed cell in the slice. The reason for this alteration was the big variety in the sizes of cells in the Arabidopsis epidermis and its disordered organization, which makes the identification of clear first- and second-degrees cells difficult. One-way ANOVA and box-plot analyses were performed using SPSS 12.0.1 statistical software (SPSS Inc., Chicago, IL, USA).
Callose staining and quantification Callose staining for co-localization of AtBG_ppap:GFP and callose was performed by incubating AtBG_ppap:GFP transgenic tobacco source leaf segments for 15 min in a mixture of 0.1% aniline blue (Fluka, www.sigmaaldrich.com/brands/Fluka_Riedel_home.html) in double-distilled water and 1 M glycine, pH 9.5, at a volume ratio of 2:3. For better visualization, a negative image was generated using ADOBEPHOTOSHOP 7.0 ME software, and the resulting GFP and aniline blue images were converted to red and blue, respectively, using the LSM 5 image browser (Zeiss). After wounding, callose quantification was performed by measuring aniline blue fluorescence intensity. Leaves of wild-type and TDNA insertion mutant plants (SALK_019116.47.55.x; see above) were grown for 2 months under short-day conditions to prevent flowering and allow continuing leaf growth. Five leaf discs from two plants
ª 2006 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 49, 669–682
680 Amit Levy et al. were analyzed for each experiment. Leaf sections of 7 · 3 mm were cut, placed on a 1% agar Petri dish for 2 days, and then immersed in 85% ethanol overnight until bleached, and transferred to aniline blue solution (see above) for 5 h. Aniline blue fluorescence was viewed and photographed under UV conditions in a DMRBE fluorescence light microscope (Leica, www.leica.com). Pictures were taken randomly from the center of the leaf; when a vein was present, the area next to the vein was photographed due to its strong background. Fluorescence intensity analysis of aniline blue in the Pd was performed under non-saturation excitation levels (exposure time 0.5 sec) using the image processing software IMAGEJ version 1.31 (http://rsb.info.nih.gov/ij). Callose fluorescence intensity associated with Pd was determined by subtracting the minimum fluorescence value in the area adjacent to the Pd (background) from the maximum fluorescent value of the Pd foci (Figure 9). A statistical independent samples t test was performed using the SPSS 12.0.1 statistical software.
Microscopy Fluorescence microscopy. Fluorescence was viewed with a DMRBE fluorescence light microscope (Leica); aniline blue-stained callose fluorescence was measured with a band-pass 340–380 nm excitation filter, an RKP 400 dichromatic mirror, and a long-pass 425 nm emission filter; GFP fluorescence was viewed with a bandpass 450–490 nm excitation filter, an RKP 510 dichromatic mirror, and a band-pass 515–560 nm emission filter (Leica). Confocal fluorescence microscopy. Fluorescence was viewed with a CLSM (LSM 510; Zeiss). GFP fluorescence was excited with a 488 nm argon laser, and emission was detected with a 505–530 nm band-pass filter combination. FM4-64 fluorescence was excited with a 514 nm argon laser, and emission was detected with a 585 nm long-pass filter combination. To expose spongy mesophyll for imaging, leaves were torn by hand so that the epidermis at the lower face of leaf sections was peeled off. Acknowledgements We thank U. Hannania and A. Avni (Tel Aviv University, Israel) for kindly supplying pBinGFP plasmid, S Yalovsky (Tel Aviv University, Israel) for kindly supplying pRTL2/newGFPm plasmid, and the Smoler Proteomic Center, Faculty of Sciences, Tel Aviv University, for protein sequencing. This research was supported by Resource Grant Award IS-3222-01C from the US–Israel Binational Agricultural Resource and Development Fund, by the Israel Science Foundation (grant 723/00-17.1) and by the Manna Institute for Plant Biosciences.
Supplementary Material The following supplementary material is available for this article online: Figure S1. Alignment of Arabidopsis GPI-anchored B-1,3-glucanases. This material is available as part of the online article from http:// www.blackwell-synergy.com
References Alonso, J.M., Stepanova, A.N., Leisse, T.J. et al. (2003) Genomewide insertional mutagenesis of Arabidopsis thaliana. Science, 301, 653–657.
Baluska, F., Samaj, J., Napier, R. and Volkmann, D. (1999) Maize calreticulin localizes preferentially to plasmodesmata in root apex. Plant J. 19, 481–488. Bayer, E., Thomas, C.L. and Maule, A.J. (2004) Plasmodesmata in Arabidopsis thaliana suspension cells. Protoplasma, 223, 93–102. Beffa, R. and Meins, F. (1996) Pathogenesis-related functions of plant b-1,3-glucanases investigated by antisense transformation – a review. Gene, 179, 97–103. Bendtsen, J.D., Nielsen, H., von Heijne, G. and Brunak, S. (2004) Improved prediction of signal peptides: SignalP 3.0. J. Mol. Biol. 340, 783–795. Benhamou, N., Grenier, J., Asselin, A. and Legrand, M. (1989) Immunogold localization of [beta]-1,3-glucanases in two plants infected by vascular wilt fungi. Plant Cell, 1, 1209–1221. Bloch, D., Lavy, M., Efrat, Y., Efroni, I., Bracha-Drori, K., Abu-Abied, M., Sadot, E. and Yalovsky, S. (2005) Ectopic expression of an activated RAC in Arabidopsis disrupts membrane cycling. Mol. Biol. Cell, 16, 1913–1927. 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 b-1, 3-glucanase enhances symptoms of TMV infection in tobacco. Plant J. 28, 361–369. Caldelari, D., Sternberg, H., Rodriguez-Concepcion, M., Gruissem, W. and Yalovsky, S. (2001) Efficient prenylation by a plant geranylgeranyl-transferase-I requires a functional CaaL box motif and a proximal polybasic domain. Plant Physiol. 126, 1416–1429. Corpet, F. (1988) Multiple sequence alignment with hierarchicalclustering. Nucleic Acids Res. 16, 10881–10890. Crawford, K.M. and Zambryski, P.C. (2000) Subcellular localization determines the availability of non-targeted proteins to plasmodesmatal transport. Curr. Biol. 10, 1032–1040. Delp, G. and Palva, E.T. (1999) A novel flower-specific Arabidopsis gene related to both pathogen-induced and developmentally regulated plant b-1,3-glucanase genes. Plant Mol. Biol. 39, 565– 575. Ding, B., Kwon, M.O. and Warnberg, L. (1996) Evidence that actin filaments are involved in controlling the permeability of plasmodesmata in tobacco mesophyll. Plant J. 10, 157–164. Dong, X.N., Mindrinos, M., Davis, K.R. and Ausubel, F.M. (1991) Induction of Arabidopsis defense genes by virulent and avirulent Pseudomonas syringae strains and by a cloned avirulence gene. Plant Cell, 3, 61–72. Eisenhaber, B., Wildpaner, M., Schultz, C.J., Borner, G.H.H., Dupree, P. and Eisenhaber, F. (2003) Glycosylphosphatidylinositol lipid anchoring of plant proteins. Sensitive prediction from sequenceand genome-wide studies for Arabidopsis and rice. Plant Physiol. 133, 1691–1701. Elortza, F., Nuhse, T.S., Foster, L.J., Stensballe, A., Peck, S.C. and Jensen, O.N. (2003) Proteomic analysis of glycosylphosphatidylinositol-anchored membrane proteins. Mol. Cell. Proteomics, 2, 1261–1270. Elortza, F., Mohammed, S., Bunkenborg, J., Foster, L.J., Nu¨hse, T.S., Brodbeck, U., Peck, S.C. and Jensen, O.N. (2006) Modification-specific proteomics of plasma membrane proteins: identification and characterization of glycosylphosphatidylinositolanchored proteins released upon phospholipase D treatment. J. Proteome Res. 5, 935–943. Fankhauser, N. and Maser, P. (2005) Identification of GPI anchor attachment signals by a Kohonen self-organizing map. Bioinformatics, 21, 1846–1852. Gorshkova, E.N., Erokhina, T.N., Stroganova, T.A., Yelina, N.E., Zamyatnin, A.A., Kalinina, N.O., Schiemann, J., Solovyev, A.G. and Morozov, S.Y. (2003) Immunodetection and fluorescent microscopy of transgenically expressed hordeivirus TGBp3
ª 2006 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 49, 669–682
A plasmodesmal b-1,3-glucanase 681 movement protein reveals its association with endoplasmic reticulum elements in close proximity to plasmodesmata. J. Gen. Virol. 84, 985–994. Heinlein, M. and Epel, B.L. (2004) Macromolecular transport and signaling through plasmodesmata. In International Review of Cytology: A Survey of Cell Biology (Kwang, W.J., ed.). Oxford: Elsevier, Vol. 235, pp. 93–164. Hird, D.L., Worrall, D., Hodge, R., Smartt, S., Paul, W. and Scott, R. (1993) The anther-specific protein encoded by the Brassica napus and Arabidopsis thaliana A6 gene displays similarity to b-1,3glucanases. Plant J. 4, 1023–1033. Hirokawa, T., Boon-Chieng, S. and Mitaku, S. (1998) SOSUI: classification and secondary structure prediction system for membrane proteins. Bioinformatics, 14, 378–379. Hu, G.G. and Rijkenberg, F.H.J. (1998) Subcellular localization of b1,3-glucanase in Puccinia recondita f.sp. tritici-infected wheat leaves. Planta, 204, 324–334. Hunold, R., Bronner, R. and Hahne, G. (1994) Early events in microprojectile bombardment – cell viability and particle location. Plant J. 5, 593–604. Iglesias, V.A. and Meins, 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. Kauss, H. (1996) Callose synthesis. In Membranes: Specialized Functions in Plants (Smallwood, M., Knox, J.P. and Bowles, D.J., eds). Oxford: BIOS, pp. 77–92. Keefe, D., Hinz, U. and Meins, F. (1990) The effect of ethylene on the cell-type-specific and intracellular-localization of b-1,3-glucanase and chitinase in tobacco leaves. Planta, 182, 43–51. Letunic, I., Copley, R.R., Schmidt, S., Ciccarelli, F.D., Doerks, T., Schultz, J., Ponting, C.P. and Bork, P. (2004) SMART 4.0: towards genomic data integration. Nucleic Acids Res. 32, D142–D144. Leubner-Metzger, G., Meins, F. Jr. (1999) Function and regulation of plant b-1,3-glucanases (PR-2). In Pathogenesis Related Proteins in Plants (Datta, S.K. and Muthukrishnan, S., eds). Boca Raton, FL: CRC Press, pp. 49–76. Liarzi, O. and Epel, B.L. (2005) Development of a quantitative tool for measuring changes in the coefficient of conductivity of plasmodesmata induced by developmental, biotic, and abiotic signals. Protoplasma, 225, 67–76. Lorkovic, Z.J., Kirk, D.A.W., Lambermon, M.H.L. and Filipowicz, W. (2000) Pre-mRNA splicing in higher plants. Trends Plant Sci. 5, 160–167. Marchler-Bauer, A. and Bryant, S.H. (2004) CD-Search: protein domain annotations on the fly. Nucleic Acids Res. 32, W327–W331. Olesen, P. and Robards, A.W. (1990) The neck region of the plasmodesmata: general architecture and some functional aspects. In Parallels in Cell to Cell Junctions in Plants and Animals (Robards, A.W., Lucas, W.J., Pitts, J.D., Jongsma, H.J. and Spray, D.C., eds). Berlin: Springer, pp. 145–170. Oparka, K.J. (2004) Getting the message across: how do plant cells exchange macromolecular complexes? Trends Plant Sci. 9, 33–41. Oparka, K.J. and Roberts, A.G. (2001) Plasmodesmata. A not so open-and-shut case. Plant Physiol. 125, 123–126. Oparka, K.J., Roberts, A.G., Boevink, P., Santa Cruz, S., Roberts, L., Pradel, K.S., Imlau, A., Kotlizky, G., Sauer, N. and Epel, B. (1999) Simple, but not branched, plasmodesmata allow the nonspecific trafficking of proteins in developing tobacco leaves. Cell, 97, 743– 754. Ori, N., Sessa, G., Lotan, T., Himmelhoch, S. and Fluhr, R. (1990) A major stylar matrix polypeptide (sp41) is a member of the pathogenesis-related proteins superclass. EMBO J. 9, 3429–3436.
Radford, J.E. and White, R.G. (2001) Effects of tissue-preparationinduced callose synthesis on estimates of plasmodesma size exclusion limits. Protoplasma, 216, 47–55. Radford, J.E., Vesk, M. and Overall, R.L. (1998) Callose deposition at plasmodesmata. Protoplasma, 201, 30–37. Rinne, P.L.H. and van der Schoot, C. (1998) Symplasmic fields in the tunica of the shoot apical meristem coordinate morphogenetic events. Development, 125, 1477–1485. Rinne, P.L.H. and van der Schoot, C. (2003) Plasmodesmata at the crossroads between development, dormancy, and defense. Can. J. Bot. 81, 1182–1197. Rinne, P.L.H., Kaikuranta, P.M. and van der Schoot, C. (2001) The shoot apical meristem restores its symplasmic organization during chilling-induced release from dormancy. Plant J. 26, 249–264. Rinne, P.L.H., van den Boogaard, R., Mensink, M.G.J., Kopperud, C., Kormelink, R., Goldbach, R. and van der Schoot, C. (2005) Tobacco plants respond to the constitutive expression of the tospovirus movement protein NSM with a heat-reversible sealing of plasmodesmata that impairs development. Plant J. 43, 688– 707. Roberts, I.M., Boevink, P., Roberts, A.G., Sauer, N., Reichel, C. and Oparka, K.J. (2001) Dynamic changes in the frequency and architecture of plasmodesmata during the sink–source transition in tobacco leaves. Protoplasma, 218, 31–44. Rodriguez-Concepcion, M., Yalovsky, S., Zik, M., Fromm, H. and Gruissem, W. (1999) The prenylation status of a novel plant calmodulin directs plasma membrane or nuclear localization of the protein. EMBO J. 18, 1996–2007. Roy, S., Watada, A.E. and Wergin, W.P. (1997) Characterization of the cell wall microdomain surrounding plasmodesmata in apple fruit. Plant Physiol. 114, 539–547. Ruan, Y.L., Xu, S.M., White, R. and Furbank, R.T. (2004) Genotypic and developmental evidence for the role of plasmodesmatal regulation in cotton fiber elongation mediated by callose turnover. Plant Physiol. 136, 4104–4113. Sagi, G., Katz, A., Guenoune-Gelbart, D. and Epel, B.L. (2005) Class 1 reversibly glycosylated polypeptides are plasmodesmalassociated proteins delivered to plasmodesmata via the Golgi apparatus. Plant Cell, 17, 1788–1800. Sessions, A., Burke, E., Presting, G. et al. (2002) A high-throughput Arabidopsis reverse genetics system. Plant Cell, 14, 2985– 2994. Sivaguru, M., Fujiwara, T., Samaj, J., Baluska, F., Yang, Z.M., Osawa, H., Maeda, T., Mori, T., Volkmann, D. and Matsumoto, H. (2000) Aluminum-induced 1 fi 3-b-D-glucan inhibits cell-to-cell trafficking of molecules through plasmodesmata. A new mechanism of aluminum toxicity in plants. Plant Physiol. 124, 991– 1005. Stintzi, A., Heitz, T., Prasad, V., Wiedemannmerdinoglu, S., Kauffmann, S., Geoffroy, P., Legrand, M. and Fritig, B. (1993) Plant pathogenesis-related proteins and their role in defense against pathogens. Biochimie, 75, 687–706. Tsuchiya, T., Toriyama, K., Yoshikawa, M., Ejiri, S. and Hinata, K. (1995) Tapetum-specific expression of the gene for an endo-b-1,3glucanase causes male-sterility in transgenic tobacco. Plant Cell Physiol. 36, 487–494. Turner, A., Wells, B. and Roberts, K. (1994) Plasmodesmata of maize root tips: structure and composition. J. Cell Sci. 107, 3351– 3361. Udenfriend, S. and Kodukula, K. (1995) How glycosyl-phosphatidylinositol-anchored membrane-proteins are made. Annu. Rev. Biochem. 64, 563–591.
ª 2006 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 49, 669–682
682 Amit Levy et al. Uknes, S., Mauchmani, B., Moyer, M., Potter, S., Williams, S., Dincher, S., Chandler, D., Slusarenko, A., Ward, E. and Ryals, J. (1992) Acquired-resistance in Arabidopsis. Plant Cell, 4, 645–656. Yahalom, A., Lando, R., Katz, A. and Epel, B.L. (1998) A calciumdependent protein kinase is associated with maize mesocotyl plasmodesmata. J. Plant Physiol. 153, 354–362.
Zambryski, P. (2004) Cell-to-cell transport of proteins and fluorescent tracers via plasmodesmata during plant development. J. Cell Biol. 164, 165–168. 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.
Accession numbers: AtBG_ppap: NM_123575; AtBG_ppapas: NM_203139
ª 2006 The Authors Journal compilation ª 2007 Blackwell Publishing Ltd, The Plant Journal, (2007), 49, 669–682