TY - JOUR
AU - Goffner, Deborah
AB - Abstract Xylogenic cultures of zinnia (Zinnia elegans) provide a unique opportunity to study signaling pathways of tracheary element (TE) differentiation. In vitro TEs differentiate into either protoxylem (PX)-like TEs characterized by annular/helical secondary wall thickening or metaxylem (MX)-like TEs with reticulate/scalariform/pitted thickening. The factors that determine these different cell fates are largely unknown. We show here that supplementing zinnia cultures with exogenous galactoglucomannan oligosaccharides (GGMOs) derived from spruce (Picea abies) xylem had two major effects: an increase in cell population density and a decrease in the ratio of PX to MX TEs. In an attempt to link these two effects, the consequence of the plane of cell division on PX-MX differentiation was assessed. Although GGMOs did not affect the plane of cell division per se, they significantly increased the proportion of longitudinally divided cells differentiating into MX. To test the biological significance of these findings, we have determined the presence of mannan-containing oligosaccharides in zinnia cultures in vitro. Immunoblot assays indicated that β-1,4-mannosyl epitopes accumulate specifically in TE-inductive media. These epitopes were homogeneously distributed within the thickened secondary walls of TEs when the primary cell wall was weakly labeled. Using polysaccharide analysis carbohydrate gel electrophoresis, glucomannans were specifically detected in cell walls of differentiating zinnia cultures. Finally, zinnia macroarrays probed with cDNAs from cells cultured in the presence or absence of GGMOs indicated that significantly more genes were down-regulated rather than up-regulated by GGMOs. This study constitutes a major step in the elucidation of signaling mechanisms of PX- and MX-specific genetic programs in zinnia. Xylogenesis is one of the most remarkable examples of cell specialization in higher plants. The xylem is the principal water-conducting tissue, transporting water from the root system to the aerial portions of the plant. To ensure this critical function, long files of cells divide and elongate, secondary cell wall material is deposited, the end walls between cells are hydrolyzed, and cell content is destroyed. In angiosperms, the resulting hollow structure, or xylem vessel, is composed of single units called tracheary elements (TEs). TEs are characterized by the different types of secondary wall thickening that are laid down: annular, helical, reticulate, and pitted. Protoxylem (PX) TEs with annular or helical secondary cell wall thickening differentiate while an organ is still expanding, whereas metaxylem (MX) TEs with reticulate or pitted secondary cell wall thickening differentiate after organ expansion has ceased (Esau, 1977). The relative proportions of PX and MX TEs that make up a given vascular bundle vary among plant organs and within a given organ throughout plant growth (Fahn, 1990; Pesquet et al., 2003). Until now, the underlying molecular mechanisms, signaling events, and positional information dictating the formation of either PX or MX were largely unknown. Significant headway has recently been made in a study that shows that two NAC family transcription factors, vascular-related NAC-domain VND6 and VND7, act as critical switches for MX and PX vessel formation in Arabidopsis (Arabidopsis thaliana; Kubo et al., 2005). Overexpression of VND6 and VND7 in Arabidopsis led to MX and PX formation, respectively, not only in the xylem but also in unexpected cell types, such as guard cells. Although this elegantly illustrates that genetic programs specify PX and MX, the upstream factors are still largely unknown. That said, very recently, Mähönen et al. (2006) pointed out the critical role of cytokinins in regulating PX-MX specification. The in vitro system in which isolated mesophyll cells of zinnia (Zinnia elegans) transdifferentiate into TEs in the presence of exogenous auxin and cytokinin has yielded an enormous amount of biochemical, cytological, and molecular information concerning TE formation (Fukuda and Komamine, 1980a). Indeed, zinnia cultures mimic many aspects of xylogenesis in planta, including the formation of PX- and MX-like TEs (Falconer and Seagull, 1985; Pesquet et al., 2003). Therefore, the zinnia system provides a unique opportunity to search for actors involved in upstream events that determine PX or MX formation. Among the likely candidate signaling molecules are sugars. The study of the dynamics of cell surface molecules in zinnia cultures showed that a xylan, an arabinan, and a relatively unbranched rhamnogalacturonan were rapidly secreted by cells into the inductive medium (Stacey et al., 1995). By using monoclonal antibodies as probes for specific cell wall components, an arabinogalactan protein, a fucosylated xyloglucan, and a methyl-esterified pectin were identified in TE cell walls both in vitro and in planta. More recently, a glycosylphosphatidylinositol-anchored glycoprotein with a fasciclin domain, ZeFLA11, originally identified in in vitro-differentiating TEs, has been localized exclusively in differentiating MX cells (Dahiya et al., 2005b). Beyond these descriptive studies of polysaccharides and polysaccharide-containing molecules during TE differentiation, physiological roles in TE signaling have been assigned in only a few isolated cases. First, a proteoglycan-designated xylogen has been identified as a coordinator molecule, which, upon polar secretion from differentiating vascular cells, draws neighboring undifferentiated cells into the vascular differentiation pathway to direct continuous vascular development. The polypeptide backbone of the xylogen is a hybrid-type molecule with properties of both arabinogalactan proteins and nonspecific lipid transfer proteins (Motose et al., 2004). Second, a putative oligosaccharide secreted during TE differentiation in zinnia cultures was shown to regulate the PX-MX transition (Roberts et al., 1997). Oligosaccharides play an important role in various aspects of normal plant growth and development and in response to biotic stress (Albersheim and Darvill, 1985). During TE formation, it is known that active degradation of primary cell wall components, including pectin, occurs (Ohdaira et al., 2002), demonstrating that wall-derived oligosaccharidic breakdown products are generated and may subsequently act in directing cell fate determination. A specific class of oligosaccharide-signaling molecules, galactoglucomannan oligosaccharides (GGMOs), has been shown to exhibit a wide range of biological activities when supplemented to plant tissue or cell cultures. GGMOs derived from galactoglucomannans, which are structural constituents of both primary and secondary cell walls of higher plants (Dey, 1980; Akiyama et al., 1983), are abundant in the thickened secondary walls of gymnosperms (Capek et al., 2000; Lundqvist et al., 2002). GGMOs are composed of a backbone of randomly dispersed β-1,4-Man and Glc residues and some minor Gal substitutions and have been isolated from wood (Dey, 1978, 1980), kiwifruit (Actinidia deliciosa; Schröder et al., 2001), and tobacco (Nicotiana tabacum) cell cultures (Sims et al., 1997). Auxtová et al. (1995) and Auxtová-Šamajová et al. (1996) showed that GGMOs isolated from poplar (Populus spp.) wood resulted in inhibition of auxin-induced cell elongation growth in pea (Pisum sativum) stems. Recently, this inhibitory effect was shown to be associated with complex changes in various glycosidase, glycanase, and peroxidase activities in different compartments of the cell (Bilisics et al., 2004; Lišková et al., 2006). From a structural point of view, the Gal side chains attached to the β-1,4-glucopyranosyl-mannopyranosyl backbone are essential for GGMO-mediated inhibition of elongation growth because cleavage of α-linked Gal residues with α-galactosidase significantly reduced GGMO-induced inhibition (Kollárová et al., 2006). GGMOs also inhibited auxin-induced adventitious root formation (Kollárová et al., 2005). Other diverse physiological effects of a more stimulatory nature have also been reported for GGMOs. They include enhanced (1) adventitious root induction in the absence of auxin (Kollárová et al., 2005); (2) survival of spruce (Picea abies) zygotic embryos (Lišková et al., 1995); (3) viability of spruce protoplasts isolated from callus cultures (Lišková et al., 1995); and (4) nonspecific resistance to local viral infection on the leaf surface of cucumber (Cucumis sativus) challenged with Tobacco necrosis virus (Slováková et al., 2000). It is clear that the mode of action of these compounds must be complex given the fact that either a stimulatory or inhibitory effect may be observed, depending on the concentration applied, the time of application (Auxtová et al., 1995; Auxtová-Šamajová et al., 1996; Kollárová et al., 2005), their possible interaction with auxin or GA (Auxtová et al., 1995; Auxtová-Šamajová et al., 1996; Kollárová et al., 2006), and the plant species studied (Auxtová et al., 1995). In this article, we describe the effect of exogenous GGMOs on xylogenic zinnia cultures. Two major effects were observed: an increase in the cell population density and a decrease in the relative proportion of PX/MX TEs. Moreover, biochemical evidence is provided indicating that glucomannans accumulate in walls of zinnia cells during in vitro differentiation. Finally, to determine the effect of GGMOs on global gene expression, macroarrays constructed from differentiating TEs were probed with cDNAs derived from zinnia cells in the presence and absence of GGMOs. These results highlight a novel physiological role for GGMOs as signaling molecules in channeling TEs into a MX genetic program and provide a means to identify potential actors of the MX and PX pathways. RESULTS GGMOs Increase the Cell Population Density of Xylogenic Zinnia Cultures As a first step in characterizing the effect of GGMOs on zinnia cell cultures, the population densities of living and dead cells were calculated every 2 d over a 10-d culture period. The initial population density of living cells was approximately 1.5 × 105 cells mL−1 of culture medium, which is optimal for promoting TE differentiation under our culture conditions. At culture initiation, 30% to 35% of the cells were dead due to the damage caused by the mechanical isolation procedure. In both control and GGMO-treated cultures, the population density of living cells gradually increased over time (Fig. 1 Figure 1. Open in new tabDownload slide Population density of living cells in xylogenic zinnia cultures in the absence (•) or presence (□) of 50 μg mL−1 GGMOs. Cell viability was determined by Evans blue staining. Data represent the means of three replicates ±sd. Figure 1. Open in new tabDownload slide Population density of living cells in xylogenic zinnia cultures in the absence (•) or presence (□) of 50 μg mL−1 GGMOs. Cell viability was determined by Evans blue staining. Data represent the means of three replicates ±sd. ), but to a significantly greater extent in GGMO-treated cultures. The effect of GGMOs (50 μg mL−1) on cell population density was observed within 2 d. The main difference was observed on day 6, when GGMO-treated cells reached the value of 3.8 × 105 live cells mL−1 (a 2.3-fold increase with respect to initial density of cells) compared with 2.5 × 105 live cells mL−1 (a 1.4-fold increase over initial density of cells) for control cells. A similar GGMO-mediated increase in cell population density was observed at 20 μg mL−1 and, to a lesser extent, at 100 μg mL−1. Exogenous GGMOs had no effect on the population density of dead cells compared to the control (data not shown), suggesting that GGMOs act by stimulating cell division and not by protecting against cell death. GGMOs Do Not Affect the Two Bursts of TE Differentiation in Xylogenic Cultures It has been previously described that TE differentiation in vitro occurs in two bursts in long-term cultures (Falconer and Seagull, 1985). To determine whether GGMOs affect the timing of the two bursts of TE differentiation and/or their amplitude (the effect on the percentage of differentiation per se), percentage of TE differentiation was determined over a 10-d period in the presence and absence of various concentrations of GGMOs. Because the results were the same at 20 and 50 μg mL−1, the representative concentration of 50 μg mL−1 is shown in Figure 2 Figure 2. Open in new tabDownload slide TE differentiation in xylogenic zinnia cultures in the absence (•) or presence (□) of 50 μg mL−1 GGMOs. Data represent the means of three replicates ±sd. Figure 2. Open in new tabDownload slide TE differentiation in xylogenic zinnia cultures in the absence (•) or presence (□) of 50 μg mL−1 GGMOs. Data represent the means of three replicates ±sd. . The first burst of TE differentiation in control cultures started at day 3 and was maximal at day 4. The second burst reached its highest values at days 8 to 9. As non-TE cells actively divide, especially between days 5 and 7 (Fig. 1), the overall proportion of TEs decreases in the culture at day 6 (Fig. 2). In the presence of exogenous GGMOs, neither the timing nor the amount of TE differentiation of the two peaks was modified. GGMOs Modify the Relative Proportion of PX- and MX-Like TEs As a prerequisite for characterizing the effect of GGMOs on secondary cell wall patterning, we first monitored the dynamics on PX and MX formation in control cultures over a 10-d culture period under our experimental conditions. Different patterns were observed as previously described: annular, annular-helical, helical, helical-reticulate, reticulate, scalariform, and pitted (Bierhorst, 1960; Falconer and Seagull, 1985; Roberts and Haigler, 1994; Pesquet et al., 2005). For the sake of clarity, we have classified the different types of TEs as follows: PX-like TEs have annular, annular-helical, and helical thickening; reticulate TEs have helical-reticulate and reticulate thickening; and MX-like TEs have scalariform and pitted thickening. On day 3, the reticulate were most abundant (46%), followed by PX (36%) and MX (18%) TEs (Fig. 3 Figure 3. Open in new tabDownload slide Relative proportions of TE secondary wall thickenings in xylogenic zinnia cultures (% of TE type/total TEs). White bars, Annular and helical PX TEs; gray bars, reticulate; black bars, scalariform and pitted MX TEs. Data represent the means of three replicates ±sd. Figure 3. Open in new tabDownload slide Relative proportions of TE secondary wall thickenings in xylogenic zinnia cultures (% of TE type/total TEs). White bars, Annular and helical PX TEs; gray bars, reticulate; black bars, scalariform and pitted MX TEs. Data represent the means of three replicates ±sd. ). Over the culture period, the relative proportions of PX and reticulate TEs decreased, whereas the proportion of MX TEs increased. Although the actual values varied among cultures, the tendency was always the same; the MX TEs were always less abundant during the first burst and much more abundant in the second. Interestingly, during the second burst (day 8), 95% of the living TEs, as judged by the presence of an intact nucleus (Fig. 4B Figure 4. Open in new tabDownload slide A and B, Micrographs of MX TEs cultured on semisolid medium for 4 (A) or 10 (B) d. First-burst MX TEs (A) are much smaller than second-burst (B) MX TEs. Nuclei are readily visible in B, indicating that they have recently differentiated and PCD is not yet completed. N, Nucleus. C, Micrographs of the zinnia mesophyll cells differentiating on semisolid medium over an 8-d period. The same two cells were observed over time (a PX and a MX TE). Magnification bars = 10 μm. Figure 4. Open in new tabDownload slide A and B, Micrographs of MX TEs cultured on semisolid medium for 4 (A) or 10 (B) d. First-burst MX TEs (A) are much smaller than second-burst (B) MX TEs. Nuclei are readily visible in B, indicating that they have recently differentiated and PCD is not yet completed. N, Nucleus. C, Micrographs of the zinnia mesophyll cells differentiating on semisolid medium over an 8-d period. The same two cells were observed over time (a PX and a MX TE). Magnification bars = 10 μm. ) and Evans blue staining, were MX. These second-burst MX TEs are much longer and wider than first-burst TEs (Fig. 4, A and B). To determine morphological changes of the same mesophyll cells during TE transdifferentiation, a time course was carried out over an 8-d time period in semisolid media and the secondary wall thickening of 48 individual differentiating TEs was monitored (Fig. 4C). These results showed that wall patterns are conserved throughout secondary cell wall deposition, indicating that cell fate determination to PX or MX is an early event. This also suggests that each TE cell type in vitro, despite the lack of positional information encountered in planta, is derived from a specific ontogenic program. We conclude that there is no continuum between PX and MX TE formation in vitro, and, therefore, this cannot explain the PX/MX decrease over time (Fig. 3). Moreover, Figure 4C revealed that MX differentiation may occur without an increase in cell volume, whereas mesophyll cells that undergo cell expansion may differentiate into PX. Thus, cell expansion and MX differentiation are not always intimately linked as previously described (Roberts et al., 1997; Pesquet et al., 2005). To ascertain the effect of GGMOs on secondary wall patterning, the relative proportions of PX, reticulate, and MX TEs were determined in the presence or absence of 50 μg mL−1 GGMOs over a 7-d culture period (Fig. 5 Figure 5. Open in new tabDownload slide Effect of GGMOs on percentage of TE secondary wall patterning in xylogenic zinnia cultures in the absence (A) or presence (B) of 50 μg mL−1 GGMOs. White bars, Annular and helical PX TEs; gray bars, reticulate TE; black bars, scalariform and pitted MX TEs. Data represent the means of three replicates ±sd. Figure 5. Open in new tabDownload slide Effect of GGMOs on percentage of TE secondary wall patterning in xylogenic zinnia cultures in the absence (A) or presence (B) of 50 μg mL−1 GGMOs. White bars, Annular and helical PX TEs; gray bars, reticulate TE; black bars, scalariform and pitted MX TEs. Data represent the means of three replicates ±sd. ). Starting at day 4 up to day 6, the proportion of MX TEs is greatly increased by exogenously added GGMOs. At day 7, when the second burst of TE has begun in control cultures, the differences between GGMO-treated cultures and controls are less apparent, but exist nevertheless. As the number of total TEs is not modified by GGMOs, this suggests that GGMOs are acting as a switch from PX to MX formation. Interestingly, there is no concomitant increase in cell expansion in the presence of GGMOs, and, therefore, as stated above, there appears not to be a strict cell expansion requirement for MX production (data not shown). GGMOs Orient Longitudinally Divided Cells toward MX Differentiation Exogenous GGMOs have two major effects on xylogenic zinnia cultures: an increase in cell population density, presumably by stimulating cell division (Fig. 1), and an increase in the relative proportion of MX TEs (Fig. 5). To determine whether there is a causal relationship between them, we examined the relationship between the subset of TEs derived from cell division and their propensity to become either a PX or MX in the presence/absence of GGMOs (Fig. 6 Figure 6. Open in new tabDownload slide Effect of GGMOs on PX and MX formation of zinnia cells that have divided in the transverse or longitudinal plane of division. Cells were cultured for 6 d. Percentages are indicated in bold italics for GGMO-treated cultures. Note that GGMOs increased the relative proportion of MX TEs derived from longitudinally, but not transversely, divided cells. Magnification bars = 10 μm. Figure 6. Open in new tabDownload slide Effect of GGMOs on PX and MX formation of zinnia cells that have divided in the transverse or longitudinal plane of division. Cells were cultured for 6 d. Percentages are indicated in bold italics for GGMO-treated cultures. Note that GGMOs increased the relative proportion of MX TEs derived from longitudinally, but not transversely, divided cells. Magnification bars = 10 μm. ). Although it is well known that mesophyll cells may indeed transdifferentiate without prior cell division (Fukuda and Komamine, 1980b; Fig. 4C), a large proportion do divide first. In control cultures, the large majority of TEs that underwent prior cell division resulted from transverse (71%) as compared to longitudinal (29%) division. Exogenous GGMOs did not influence the plane of cell division. In control cultures, when cells divided along a transverse axis, 60% formed MX and 40% PX TEs. GGMOs had no effect on the transversely divided population. When TEs resulted from longitudinal division, 73% became PX, whereas only 27% became MX TEs in control cultures. Interestingly, GGMOs increased the proportion of MX TEs derived from longitudinally divided cells (Fig. 6). Together, these results indicate that (1) the plane of division has a profound effect on whether a cell will differentiate into either a PX or MX; and (2) GGMOs do not directly channel divided cells into the MX-enriched transverse plane of division, but act downstream in orienting the cell fate of longitudinally divided cells into MX TEs. β-1,4-Mannan Compounds Accumulate in Zinnia Cell Culture Medium and Are Localized in TE and Non-TE Cell Walls To determine whether GGMOs or structurally similar compounds naturally exist in zinnia cell cultures, we characterized mannan-containing polysaccharides secreted into the culture medium by immunoblot assays with a polyclonal antibody that specifically recognizes β-1,4-mannosyl residues present within glucomannan, mannan, and galactomannan (Handford et al., 2003). In TE-inductive media, mannan epitopes accumulate between days 3 and 6, whereas in control cultures without hormones, only trace amounts were detected (Fig. 7 Figure 7. Open in new tabDownload slide Immunoblot assays of mannan-containing epitopes in zinnia cultures. I, TE-inductive cultures; C, control cultures without hormones that do not differentiate. Commercial mannan (M) and oligomannan [(Man)6] were also spotted onto membranes as controls. Figure 7. Open in new tabDownload slide Immunoblot assays of mannan-containing epitopes in zinnia cultures. I, TE-inductive cultures; C, control cultures without hormones that do not differentiate. Commercial mannan (M) and oligomannan [(Man)6] were also spotted onto membranes as controls. ). This accumulation is concomitant with secondary wall formation in TE-inductive conditions. Competition experiments using glucomannan were performed to confirm signal specificity. The signal was substantially reduced, but not completely lost (data not shown). This is probably due to the fact that glucomannan was not completely soluble and therefore effective under the experimental conditions used for immunoblot assays (5 mg mL−1 glucomannan in phosphate-buffered saline [PBS]). At the electron microscopy level, we further investigated the location of mannan epitopes in cells cultured in both TE-inductive and noninductive media (Fig. 8 Figure 8. Open in new tabDownload slide Immunogold labeling of mannan-containing epitopes in zinnia cultures visualized by transmission electron microscopy. Cells were cultured in noninductive (A) or TE-inductive (B–E) medium for 4 d. C and D, A transverse/oblique (C) and a tangential (D) view of two successive hoops of the secondary wall of a single TE. The primary cell wall is not in the focal plane in D. E, TE from competition experiments treated with mannan and glucomannan. Note the very weak background signal indicated by dashed arrows. CWI, Primary cell wall; CWII, secondary cell wall; ML, middle lamella. Magnification bars = 0.1 μm (A and B); 0.3 μm (C–E). Figure 8. Open in new tabDownload slide Immunogold labeling of mannan-containing epitopes in zinnia cultures visualized by transmission electron microscopy. Cells were cultured in noninductive (A) or TE-inductive (B–E) medium for 4 d. C and D, A transverse/oblique (C) and a tangential (D) view of two successive hoops of the secondary wall of a single TE. The primary cell wall is not in the focal plane in D. E, TE from competition experiments treated with mannan and glucomannan. Note the very weak background signal indicated by dashed arrows. CWI, Primary cell wall; CWII, secondary cell wall; ML, middle lamella. Magnification bars = 0.1 μm (A and B); 0.3 μm (C–E). ). In noninductive medium, the primary cell wall showed weak labeling (Fig. 8A). In TE-inductive cultures, mannan polysaccharides were localized in both TE and non-TE cells (Fig. 8, B–D). In non-TEs, labeling was mainly located in the primary walls and middle lamella region of two adjacent, joined cells (Fig. 8B). In the other wall regions of these joined cells, very little label could be detected, suggesting that the mannan polysaccharides presumably secreted into these wall regions ended up in the culture medium (data not shown). In TEs, mannan-containing epitopes were densely and homogeneously distributed throughout the thickened secondary cell walls (Fig. 8, C and D). The TE primary walls were weakly labeled (Fig. 8C). When sections were labeled in the presence of mannan and glucomannan competitors, the labeling was almost completely abolished from TE walls, confirming the specificity of antibody labeling (Fig. 8E). Walls of Zinnia Cells Contain Glucomannans When Cultured in TE-Inductive Medium Polysaccharide analysis using carbohydrate gel electrophoresis (PACE) in combination with specific hydrolases is a fast and quantitative method to analyze cell wall polysaccharide composition and structure (Goubet et al., 2002). We have used PACE in combination with Man5A, a recombinant mannanase that specifically hydrolyzes mannan, glucomannan, and galactomannan, to investigate the mannan polysaccharide structure of cell walls of zinnia cells cultured in TE-inductive versus noninductive medium (Fig. 9 Figure 9. Open in new tabDownload slide PACE fingerprint of mannan polysaccharides in zinnia cell walls. Glucomannan can be detected within 5 d of culture in inductive conditions (I), but not in the control culture without hormones (C). The cell walls were treated with specific mannanase Man5A similarly as commercial polysaccharides (mannan [M] and glucomannan [GM]). Asterisks show the bands that comigrate with GM hydrolysis. Figure 9. Open in new tabDownload slide PACE fingerprint of mannan polysaccharides in zinnia cell walls. Glucomannan can be detected within 5 d of culture in inductive conditions (I), but not in the control culture without hormones (C). The cell walls were treated with specific mannanase Man5A similarly as commercial polysaccharides (mannan [M] and glucomannan [GM]). Asterisks show the bands that comigrate with GM hydrolysis. ). Zinnia cell wall-derived oligosaccharides were compared with commercial mannan and glucomannan standards that were hydrolyzed by Man5A under the same conditions (Fig. 9, lanes M and GM, respectively). Up to day 4, the products of Man5A-mediated hydrolysis could not be detected either in control or inductive conditions. From day 5 on, digestible glucomannan in cell walls was readily detectible uniquely in TE-inductive cultures, as indicated by bands that comigrate with the glucomannan standard. Judging by band intensities, the amount of glucomannan increased over time. The relative band intensities observed for zinnia glucomannan were different compared with the konjak glucomannan standard, indicating that the ratio of Glc to Man is likely to be species dependent. It should be noted that some faint bands could be detected in both control and TE-inductive cultures. However, these bands were considered nonspecific because they were also detected in walls that were not hydrolyzed by Man5A (data not shown) and their intensity did not vary over time (Fig. 9). Global Gene Expression Is Modified in Zinnia Cultures by an Exogenous Supply of GGMOs As a first step in unraveling the effect of GGMOs in zinnia cultures, we have compared global gene expression using zinnia macroarrays (Pesquet et al., 2005) in the presence or absence of 50 μg mL−1 GGMOs in TE-inductive medium. Theoretically, the 800-gene macroarray contains PX and MX genes because the library was made from cultures containing both types of TEs (Pesquet et al., 2003). Two comparisons were made: the first with a pool of RNA from cells cultured in the presence versus absence of GGMOs for 12 h, and the second from cells cultured for 60 h, again, in the presence or absence of GGMOs. Several differentially expressed genes were confirmed by reverse transcription-PCR experiments (data not shown). Very few genes exhibited altered expression levels within 12 h when considering a 2-fold difference in expression as the significance threshold (Table I Table I. Genes with rapidly modified transcript levels in xylogenic zinnia cultures in the presence of 50 μg mL−1 GGMOs (within 12 h) Gene . Accession No. . Closest Arabidopsis Homologs . Fold Change . Macroarray Signal . . . . . . Control . GGMOs . Up-regulated by GGMOs     TED4-2 DV017432 At1g48750 2.0 9,054 18,122     TED4-3 DV017226 At3g18281 2.1 9,315 19,212     60S ribosomal protein L41 DV017564 At3g56020 2.6 15,997 41,197 At3g08520 At3g11120 Down-regulated by GGMOs     Heat shock protein hsp70 DQ336881 At4g24280 2.2 64,380 28,630     Pathogenesis-related protein AB091075 At1g24020 2.2 54,761 25,338     PI1 AB091073 At2g38870 2.3 195,293 86,470     PI2 AB091074 At2g38870 2.2 55,975 25,472 Gene . Accession No. . Closest Arabidopsis Homologs . Fold Change . Macroarray Signal . . . . . . Control . GGMOs . Up-regulated by GGMOs     TED4-2 DV017432 At1g48750 2.0 9,054 18,122     TED4-3 DV017226 At3g18281 2.1 9,315 19,212     60S ribosomal protein L41 DV017564 At3g56020 2.6 15,997 41,197 At3g08520 At3g11120 Down-regulated by GGMOs     Heat shock protein hsp70 DQ336881 At4g24280 2.2 64,380 28,630     Pathogenesis-related protein AB091075 At1g24020 2.2 54,761 25,338     PI1 AB091073 At2g38870 2.3 195,293 86,470     PI2 AB091074 At2g38870 2.2 55,975 25,472 Open in new tab Table I. Genes with rapidly modified transcript levels in xylogenic zinnia cultures in the presence of 50 μg mL−1 GGMOs (within 12 h) Gene . Accession No. . Closest Arabidopsis Homologs . Fold Change . Macroarray Signal . . . . . . Control . GGMOs . Up-regulated by GGMOs     TED4-2 DV017432 At1g48750 2.0 9,054 18,122     TED4-3 DV017226 At3g18281 2.1 9,315 19,212     60S ribosomal protein L41 DV017564 At3g56020 2.6 15,997 41,197 At3g08520 At3g11120 Down-regulated by GGMOs     Heat shock protein hsp70 DQ336881 At4g24280 2.2 64,380 28,630     Pathogenesis-related protein AB091075 At1g24020 2.2 54,761 25,338     PI1 AB091073 At2g38870 2.3 195,293 86,470     PI2 AB091074 At2g38870 2.2 55,975 25,472 Gene . Accession No. . Closest Arabidopsis Homologs . Fold Change . Macroarray Signal . . . . . . Control . GGMOs . Up-regulated by GGMOs     TED4-2 DV017432 At1g48750 2.0 9,054 18,122     TED4-3 DV017226 At3g18281 2.1 9,315 19,212     60S ribosomal protein L41 DV017564 At3g56020 2.6 15,997 41,197 At3g08520 At3g11120 Down-regulated by GGMOs     Heat shock protein hsp70 DQ336881 At4g24280 2.2 64,380 28,630     Pathogenesis-related protein AB091075 At1g24020 2.2 54,761 25,338     PI1 AB091073 At2g38870 2.3 195,293 86,470     PI2 AB091074 At2g38870 2.2 55,975 25,472 Open in new tab ). Among them were two members of the TE differentiation 4 (TED4) family, TED4-2 and TED4-3, which were up-regulated within the first 12 h of culture. TED4-2 is identical to the TED4 gene originally described by Demura and Fukuda (1993, 1994), whereas TED4-3 only shares 57% identity with TED4 at the nucleotide level. TED4 is a stage 2 marker gene (Demura et al., 2002). Stages 1, 2, and 3 of the transdifferentiation process were originally defined by Fukuda (1997). Stage 1 corresponds to the dedifferentiation of mesophyll cells, stage 2 to the transition from procambial initials to TE precursors, and stage 3 to TE differentiation sensu stricto. Besides the TED4 genes, the 60S ribosomal protein L41 was also up-regulated within 12 h, suggesting an increase in protein synthesis. Interestingly, of the four rapidly down-regulated genes, two of them encode protease inhibitors (PIs). PI1 and PI2 correspond to ZePI1 (97% nucleotide identity) and ZePI2 (89% nucleotide identity), respectively (Demura et al., 2002). These PIs are marker genes for stage 1 corresponding to the process where mesophyll cells lose their photosynthetic capacity and become competent for TE differentiation. In conclusion, during the first 12 h, GGMOs up-regulated two stage 2 genes (TED4-2, TED4-3) and down-regulated two stage 1 genes (PI1, PI2), suggesting an acceleration of the TE differentiation process. At 60 h, which corresponds to the onset of TE secondary wall deposition, the large majority of genes that showed modified expression were repressed rather than induced as a result of exogenously supplemented GGMOs to the culture medium (Table II Table II. Genes with modified transcript levels in xylogenic zinnia cultures in the presence of 50 μg mL−1 GGMOs for 60 h Gene . Accession No. . Closest Arabidopsis Homologs . Fold Change . Macroarray Signal . . . . . . Control . GGMOs . Up-regulated by GGMOs     α-l-Arabinofuranosidase DV017381 At3g10740 2.1 10,563 21,700     Calmodulin-1 DV017264 At2g27030 2.1 9,053 18,927 At3g43810 At2g41110 At3g56800     Unknown gene 10 DV017178 – 2.2 13,312 29,204 Down-regulated by GGMOs     ADP ribosylation factor 1-like DQ336887 At5g14670 3.3 68,204 20,623     Auxin-regulated protein (zIAA8) AY090553 At2g22670 2.7 51,003 19,131     Caffeoyl-CoA O-methyltransferase U13151 At4g34050 3.1 111,257 36,195     Caffeoyl-CoA O-methyltransferase-1 DV017416 At4g34050 2.1 10 880 5 174     Calreticulin-1 DV017386 At1g56340 3.2 47,455 14,885     Calreticulin-2 DV017418 At1g56340 3.3 41,386 12,390     Cinnamyl alcohol dehydrogenase DQ336885 At4g39330 2.1 69,689 32,949     Cysteine proteinase-1 (CP-1) DV017552 At4g35350 2.0 11,777 5,774     Cysteine proteinase-3 (CP-3) DV017587 At1g20850 2.6 26,067 10,092     Cysteine proteinase-6 (CP-6) DV017427 At4g35350 2.8 73,320 26,418     Cysteine proteinase-7 (CP-7) DV017543 At1g20850 3.3 64,805 19,918     Elongation factor 1-α DQ336889 At5g60390 2.3 43,593 18,811     F1-ATPase α-subunit DQ336888 AtMg01190 5.8 181,576 31,440     Fiber protein Fb15 AU308895 At4g30010 2.0 28,804 14,199     Glyceraldehyde-3-P dehydrogenase DQ336886 At3g04120 2.6 74,483 28,294     Pentameric polyubiquitin DQ336883 At5g20620 2.7 67,365 25,051     Phe ammonia lyase DQ336884 At2g37040 2.3 47,477 20,949     PSII component CP43 DQ336882 AtCg00280 2.3 66,379 28,733     PSII component CP47 DQ336891 AtCg00680 3.1 106,682 37,790     S-adenosine-l-Met synthetase DQ336890 At2g36880 2.6 59,041 22,425     TED4-3 DV017226 At3g18281 2.3 110,909 48,676 Gene . Accession No. . Closest Arabidopsis Homologs . Fold Change . Macroarray Signal . . . . . . Control . GGMOs . Up-regulated by GGMOs     α-l-Arabinofuranosidase DV017381 At3g10740 2.1 10,563 21,700     Calmodulin-1 DV017264 At2g27030 2.1 9,053 18,927 At3g43810 At2g41110 At3g56800     Unknown gene 10 DV017178 – 2.2 13,312 29,204 Down-regulated by GGMOs     ADP ribosylation factor 1-like DQ336887 At5g14670 3.3 68,204 20,623     Auxin-regulated protein (zIAA8) AY090553 At2g22670 2.7 51,003 19,131     Caffeoyl-CoA O-methyltransferase U13151 At4g34050 3.1 111,257 36,195     Caffeoyl-CoA O-methyltransferase-1 DV017416 At4g34050 2.1 10 880 5 174     Calreticulin-1 DV017386 At1g56340 3.2 47,455 14,885     Calreticulin-2 DV017418 At1g56340 3.3 41,386 12,390     Cinnamyl alcohol dehydrogenase DQ336885 At4g39330 2.1 69,689 32,949     Cysteine proteinase-1 (CP-1) DV017552 At4g35350 2.0 11,777 5,774     Cysteine proteinase-3 (CP-3) DV017587 At1g20850 2.6 26,067 10,092     Cysteine proteinase-6 (CP-6) DV017427 At4g35350 2.8 73,320 26,418     Cysteine proteinase-7 (CP-7) DV017543 At1g20850 3.3 64,805 19,918     Elongation factor 1-α DQ336889 At5g60390 2.3 43,593 18,811     F1-ATPase α-subunit DQ336888 AtMg01190 5.8 181,576 31,440     Fiber protein Fb15 AU308895 At4g30010 2.0 28,804 14,199     Glyceraldehyde-3-P dehydrogenase DQ336886 At3g04120 2.6 74,483 28,294     Pentameric polyubiquitin DQ336883 At5g20620 2.7 67,365 25,051     Phe ammonia lyase DQ336884 At2g37040 2.3 47,477 20,949     PSII component CP43 DQ336882 AtCg00280 2.3 66,379 28,733     PSII component CP47 DQ336891 AtCg00680 3.1 106,682 37,790     S-adenosine-l-Met synthetase DQ336890 At2g36880 2.6 59,041 22,425     TED4-3 DV017226 At3g18281 2.3 110,909 48,676 Open in new tab Table II. Genes with modified transcript levels in xylogenic zinnia cultures in the presence of 50 μg mL−1 GGMOs for 60 h Gene . Accession No. . Closest Arabidopsis Homologs . Fold Change . Macroarray Signal . . . . . . Control . GGMOs . Up-regulated by GGMOs     α-l-Arabinofuranosidase DV017381 At3g10740 2.1 10,563 21,700     Calmodulin-1 DV017264 At2g27030 2.1 9,053 18,927 At3g43810 At2g41110 At3g56800     Unknown gene 10 DV017178 – 2.2 13,312 29,204 Down-regulated by GGMOs     ADP ribosylation factor 1-like DQ336887 At5g14670 3.3 68,204 20,623     Auxin-regulated protein (zIAA8) AY090553 At2g22670 2.7 51,003 19,131     Caffeoyl-CoA O-methyltransferase U13151 At4g34050 3.1 111,257 36,195     Caffeoyl-CoA O-methyltransferase-1 DV017416 At4g34050 2.1 10 880 5 174     Calreticulin-1 DV017386 At1g56340 3.2 47,455 14,885     Calreticulin-2 DV017418 At1g56340 3.3 41,386 12,390     Cinnamyl alcohol dehydrogenase DQ336885 At4g39330 2.1 69,689 32,949     Cysteine proteinase-1 (CP-1) DV017552 At4g35350 2.0 11,777 5,774     Cysteine proteinase-3 (CP-3) DV017587 At1g20850 2.6 26,067 10,092     Cysteine proteinase-6 (CP-6) DV017427 At4g35350 2.8 73,320 26,418     Cysteine proteinase-7 (CP-7) DV017543 At1g20850 3.3 64,805 19,918     Elongation factor 1-α DQ336889 At5g60390 2.3 43,593 18,811     F1-ATPase α-subunit DQ336888 AtMg01190 5.8 181,576 31,440     Fiber protein Fb15 AU308895 At4g30010 2.0 28,804 14,199     Glyceraldehyde-3-P dehydrogenase DQ336886 At3g04120 2.6 74,483 28,294     Pentameric polyubiquitin DQ336883 At5g20620 2.7 67,365 25,051     Phe ammonia lyase DQ336884 At2g37040 2.3 47,477 20,949     PSII component CP43 DQ336882 AtCg00280 2.3 66,379 28,733     PSII component CP47 DQ336891 AtCg00680 3.1 106,682 37,790     S-adenosine-l-Met synthetase DQ336890 At2g36880 2.6 59,041 22,425     TED4-3 DV017226 At3g18281 2.3 110,909 48,676 Gene . Accession No. . Closest Arabidopsis Homologs . Fold Change . Macroarray Signal . . . . . . Control . GGMOs . Up-regulated by GGMOs     α-l-Arabinofuranosidase DV017381 At3g10740 2.1 10,563 21,700     Calmodulin-1 DV017264 At2g27030 2.1 9,053 18,927 At3g43810 At2g41110 At3g56800     Unknown gene 10 DV017178 – 2.2 13,312 29,204 Down-regulated by GGMOs     ADP ribosylation factor 1-like DQ336887 At5g14670 3.3 68,204 20,623     Auxin-regulated protein (zIAA8) AY090553 At2g22670 2.7 51,003 19,131     Caffeoyl-CoA O-methyltransferase U13151 At4g34050 3.1 111,257 36,195     Caffeoyl-CoA O-methyltransferase-1 DV017416 At4g34050 2.1 10 880 5 174     Calreticulin-1 DV017386 At1g56340 3.2 47,455 14,885     Calreticulin-2 DV017418 At1g56340 3.3 41,386 12,390     Cinnamyl alcohol dehydrogenase DQ336885 At4g39330 2.1 69,689 32,949     Cysteine proteinase-1 (CP-1) DV017552 At4g35350 2.0 11,777 5,774     Cysteine proteinase-3 (CP-3) DV017587 At1g20850 2.6 26,067 10,092     Cysteine proteinase-6 (CP-6) DV017427 At4g35350 2.8 73,320 26,418     Cysteine proteinase-7 (CP-7) DV017543 At1g20850 3.3 64,805 19,918     Elongation factor 1-α DQ336889 At5g60390 2.3 43,593 18,811     F1-ATPase α-subunit DQ336888 AtMg01190 5.8 181,576 31,440     Fiber protein Fb15 AU308895 At4g30010 2.0 28,804 14,199     Glyceraldehyde-3-P dehydrogenase DQ336886 At3g04120 2.6 74,483 28,294     Pentameric polyubiquitin DQ336883 At5g20620 2.7 67,365 25,051     Phe ammonia lyase DQ336884 At2g37040 2.3 47,477 20,949     PSII component CP43 DQ336882 AtCg00280 2.3 66,379 28,733     PSII component CP47 DQ336891 AtCg00680 3.1 106,682 37,790     S-adenosine-l-Met synthetase DQ336890 At2g36880 2.6 59,041 22,425     TED4-3 DV017226 At3g18281 2.3 110,909 48,676 Open in new tab ). Of the three up-regulated genes, one was a putative α-l-arabinofuranosidase that may be involved in cell wall remodeling resulting from the cleavage of terminal Ara residues of Ara-containing wall polysaccharides and another was UG-10, a gene of unknown function. After 60 h, several genes exhibited reduced expression levels in the presence of GGMOs. Among them were several genes involved in protein stability/degradation, including four different Cys proteinases (CP-1, 3, 6, 7) and a pentameric polyubiquitin. CP-1 is identical to ZeCP4 described by Demura et al. (2002), whereas CP-3, 6, and 7 are distinct Cys proteinases. An auxin-regulated protein was also down-regulated by GGMOs. This gene is identical to zIAA8, a member of the Aux/IAA family, which was localized in primary vasculature in planta and expressed within 3 h in inductive zinnia cultures (Groover et al., 2003). Several lignin-associated genes also have lower levels of expression in the presence of GGMOs, including caffeoyl CoA O-methyltransferase, Phe ammonia lyase, and cinnamyl alcohol dehydrogenase. S-adenosine-l-Met synthetase was also down-regulated in response to GGMOs, suggesting that it may be implicated in methyl transfer of lignin precursors. It is tempting to speculate that all these gene family members may be more closely associated with lignification in PX TEs in vitro. In general, the fact that the majority of genes were down-regulated by GGMOs is in keeping with the fact that the library from which these genes were obtained was constructed from first-burst, PX-enriched TE cultures. Finally, we have observed the down-regulation of two photosynthesis-related genes: PSII components CP43 and CP47. The inactivation of the photosynthetic machinery is not considered to be a rapid response during zinnia TE differentiation (Demura et al., 2002). The down-regulation of such genes could suggest that stage 1, or dedifferentiation, has been shortened by GGMOs. As mentioned above, GGMOs affect both density in viable cells and secondary wall patterning. In an attempt to pinpoint the effect of GGMOs on wall patterning, we compared differences in global gene expression in two experimental conditions that favor MX formation; that is, in the presence/absence of GGMOs after 60 h (see above) and second-burst (5 d cultures) versus first-burst TE cultures (60 h) without exogenous GGMOs. One may predict that genes up-regulated in both comparisons may be associated with MX production, whereas genes that are down-regulated in both comparisons would more likely be associated with the PX genetic program. Only less abundant genes were found in common in both comparisons (Table III Table III. Commonly down-regulated genes from two independent comparisons in which both favor MX formation Comparison 1, Zinnia cells cultured for 60 h in the presence versus absence of GGMOs (see Table II). Comparison 2, Second-burst (5-d-old) versus first-burst (60-h) zinnia cultures, both in the absence of GGMOs. Gene Down-Regulated by GGMOs . Accession No. . Auxin-related protein, zIAA8 AY090553 Caffeoyl-CoA O-methyltransferase U13151 Caffeoyl-CoA O-methyltransferase DV017416 Calreticulin-1 DV017386 Calreticulin-2 DV017418 Cinnamyl-alcohol dehydrogenase DQ336885 Cysteine proteinase-1 (CP-1) DV017552 Cysteine proteinase-3 (CP-3) DV017587 Cysteine proteinase-6 (CP-6) DV017427 Cysteine proteinase-7 (CP-7) DV017543 F1-ATPase α-subunit DQ336888 PSII CP 47 protein DQ336891 TED4-3/lipid transfer protein DV017226 Gene Down-Regulated by GGMOs . Accession No. . Auxin-related protein, zIAA8 AY090553 Caffeoyl-CoA O-methyltransferase U13151 Caffeoyl-CoA O-methyltransferase DV017416 Calreticulin-1 DV017386 Calreticulin-2 DV017418 Cinnamyl-alcohol dehydrogenase DQ336885 Cysteine proteinase-1 (CP-1) DV017552 Cysteine proteinase-3 (CP-3) DV017587 Cysteine proteinase-6 (CP-6) DV017427 Cysteine proteinase-7 (CP-7) DV017543 F1-ATPase α-subunit DQ336888 PSII CP 47 protein DQ336891 TED4-3/lipid transfer protein DV017226 Open in new tab Table III. Commonly down-regulated genes from two independent comparisons in which both favor MX formation Comparison 1, Zinnia cells cultured for 60 h in the presence versus absence of GGMOs (see Table II). Comparison 2, Second-burst (5-d-old) versus first-burst (60-h) zinnia cultures, both in the absence of GGMOs. Gene Down-Regulated by GGMOs . Accession No. . Auxin-related protein, zIAA8 AY090553 Caffeoyl-CoA O-methyltransferase U13151 Caffeoyl-CoA O-methyltransferase DV017416 Calreticulin-1 DV017386 Calreticulin-2 DV017418 Cinnamyl-alcohol dehydrogenase DQ336885 Cysteine proteinase-1 (CP-1) DV017552 Cysteine proteinase-3 (CP-3) DV017587 Cysteine proteinase-6 (CP-6) DV017427 Cysteine proteinase-7 (CP-7) DV017543 F1-ATPase α-subunit DQ336888 PSII CP 47 protein DQ336891 TED4-3/lipid transfer protein DV017226 Gene Down-Regulated by GGMOs . Accession No. . Auxin-related protein, zIAA8 AY090553 Caffeoyl-CoA O-methyltransferase U13151 Caffeoyl-CoA O-methyltransferase DV017416 Calreticulin-1 DV017386 Calreticulin-2 DV017418 Cinnamyl-alcohol dehydrogenase DQ336885 Cysteine proteinase-1 (CP-1) DV017552 Cysteine proteinase-3 (CP-3) DV017587 Cysteine proteinase-6 (CP-6) DV017427 Cysteine proteinase-7 (CP-7) DV017543 F1-ATPase α-subunit DQ336888 PSII CP 47 protein DQ336891 TED4-3/lipid transfer protein DV017226 Open in new tab ). This set of genes includes the four Cys proteinases, TED4-3, and zIAA8. Overlapping of these genomic comparisons gives further weight to these down-regulated genes as potential actors in PX formation. DISCUSSION GGMO-Like Oligosaccharides Are Naturally Produced in Xylogenic Zinnia Cell Cultures The walls of plant cells are supramolecular composite structures comprising polysaccharides, proteins, and phenolic compounds (Carpita and Gibeaut, 1993). It is well known that the structure and composition of the cell wall undergo dynamic changes during plant development, and these changes are essential to ensure cellular function. In this article, we have focused on a subset of mannan-derived oligosaccharides, GGMOs, and provided evidence for novel biological effects of these compounds in plants. In general, mannans play a structural role in cross-linking cellulose fibrils (Whitney et al., 1998), but the specific roles of other related polysaccharides are unclear. In Arabidopsis, mannans have been localized not only in thickened secondary cell walls of xylem elements, xylem parenchyma, and interfascicular fibers, but also in the thickened walls of the epidermal cell of leaves and stems and, to a lesser extent, in most other cell types (Handford et al., 2003). AtCSLA7 has recently been shown to encode a mannan synthase (Liepman et al. 2005). Interestingly, the corresponding mutant is embryo lethal, thereby suggesting that mannan may be important in signaling during embryogenesis (Goubet et al., 2003). In zinnia cultures, β-1,4-mannans accumulated to significant amounts in the culture medium and the walls of cells cultured in TE-inductive conditions. These results suggest that mannans may be naturally occurring signaling molecules involved in TE differentiation. Moreover, an expansion-inducing factor, originally described as a secreted oligosaccharide, has been partially characterized from conditioned culture medium from 6-d-old zinnia cultures (Roberts et al., 1997). When expansion-inducing factor was added to fresh cultures, early production of MX was observed, providing further evidence that oligosaccharide signaling occurs in xylogenic zinnia cultures. GGMOs Affect Cell Population Density and Relative Proportion of PX/MX TEs Although the antiauxin effect of GGMOs on elongation growth has been well documented (Auxtová et al., 1995; Auxtová-Šamajová et al., 1996; Kollárová et al., 2006), only fragmentary data exist concerning other biological effects. The most closely associated effect reported is an increase in protoplast viability derived from spruce callus, presumably by preventing necrosis (Lišková et al., 1995). In zinnia cultures, the mechanisms are likely to be different because, although exogenous GGMOs resulted in an increased live cell population density, there was no difference in the dead cell population density and, therefore, on necrotic cell death. Although Fosket (1968) reported in earlier studies that cell division was a prerequisite for wound-induced xylogenesis in cultured stem segments of coleus, it is well known that in vitro TE differentiation may take place without prior cell division (Fukuda and Komamine, 1980b). The original cytological studies carried out in the zinnia system indicated that more than 60% of the TEs were formed directly in the G1 phase of the cell cycle without intervening DNA replication or mitosis (Fukuda and Komamine, 1980b). Recently, Mourelatou et al. (2004) reported that almost all cells go through several rounds of division before characteristic features of TE formation are observed. They also confirmed that division is not a prerequisite for TE differentiation, but delay in the early S phase enhanced differentiation. In our study, even if cell division occurs throughout the culture, we provide evidence that there is no requirement of cell division for either PX or MX TE formation (Fig. 4). Concerning the effect of GGMOs on the increased proportion of MX, it is unclear whether it is indirect via cell division or whether the two effects are completely unrelated. In TE cultures that are monitored in long-term experiments (at least 8 d), a second burst in TE formation is observed (Falconer and Seagull, 1985; Fig. 2). This second burst is MX rich and is preceded by an increase in the population density of living cells (Fig. 1). These MX TEs may result from newly divided cells that are competent to differentiate, perhaps due to their position in the cell cycle, which in turn would enable them to perceive residual inductive hormones and/or threshold levels of MX-inductive TE signals excreted by the high-density cell culture. Moreover, when GGMOs were added to cultures at day 0, enhanced increase in cell population density (already observed by day 2) preceded the decrease in PX/MX TE production. Together, these arguments suggest that MX production is favored when a critical density of viable cells is reached. A link between cell division and MX formation has been established in the Arabidopsis mutant wol (wooden leg) that is mutated in a CRE-family His kinase, cytokinin receptor gene (Mähönen et al., 2000; Inoue et al., 2001), and the double mutant wol/fass (Mähönen et al., 2000). The vascular cylinder of the wol primary root consists solely of PX cell files with no MX and phloem. To investigate whether the defects in vascular differentiation are related to aberrant cell division, fass was introduced into the wol background. Indeed, the fass mutant undergoes an excessive number of cell divisions in all tissues (Torrese-Ruiz and Jurgens, 1994). In the wol/fass double mutant, the presence of MX was restored. The authors thus concluded that WOL is not necessary for MX differentiation per se, but has an indirect influence on xylem differentiation by controlling the number of cells in the vascular cylinder (Mähönen et al., 2000). Recently, by performing a suppressor screen on wol, Mähönen et al. (2006) identified a new factor in PX/MX signaling. AHP6 encodes a pseudophosphotransfer protein and functions by facilitating PX specification via negative regulation of the cytokinin-signaling pathway. This implies that, in the absence of cytokinin, PX is the default identity. The plane of cell division has a profound effect on PX and MX formation. Interestingly, cells that divided along a transverse axis of division differentiated into PX or MX at roughly equal frequencies and were not affected by GGMOs. Cells that divided along the longitudinal axis resulted in relatively higher amounts of PX. GGMOs appear to channel these cells into a more MX-rich TE differentiation pathway, but the mechanisms involved in this apparent sensitivity to GGMOs is unclear. Exogenous GGMOs Enabled the Uncoupling of Cell Expansion and Wall Patterning in Zinnia TEs Little is known about how cell wall expansion is integrated with secondary wall formation. Earlier studies provided evidence that stem elongation and patterns of secondary cell wall deposition in vessels in planta were linked. For example, if stem elongation was inhibited, the production of annular and spiral vessels ceased and pitted vessels started to develop (Goodwin, 1942; Smith and Kersten, 1942; Brower and Hepler, 1976). Furthermore, in Cyperus alternifolius, the development of vessels with wall thickening that is intermediary between PX and MX occurs at the end of intercalary growth of the internode, indicating that the extent of organ elongation determines the TE wall patterns of the organ in question (Fisher, 1970). These results indicate that vessel size seems to be correlated with the developmental stage rather than the type of TE thickening. In zinnia plants, as is the case for all angiosperms, primary vasculature is characterized first by the development of PX vessels of narrow diameter and later by MX vessels of wide diameter (Pesquet et al., 2003; Dahiya et al., 2005a). This ontogenic development is mimicked in long-term xylogenic cultures in which the first burst is characterized mainly by small, PX TEs and the second by large, MX TEs. Until now, cell expansion and the type of TE thickening have never been uncoupled in zinnia cultures, neither by the addition of exogenous factors nor by a change in the physicochemical properties of the medium. For example, elimination of large pH fluctuations by buffering the culture medium altered both cell expansion and TE differentiation (Roberts and Haigler, 1994). This work provides two lines of evidence indicating that there is not a strict correlation between cell expansion and patterning of secondary cell walls and, in particular, MX formation. First, when differentiating TEs on semisolid medium are monitored for up to 8 d, we found that both PX and MX TEs differentiate with or without prior cell expansion. Second, when GGMOs are added to zinnia cultures, there are more first-burst MX TEs without concomitant cell expansion. GGMOs Influence PX and MX Developmental Programs In planta, the formation of PX and MX is under tight spatial and temporal regulation. This has been clearly documented in Arabidopsis (Busse and Evert, 1999a, 1999b) and, more recently, in zinnia (Pesquet et al., 2003; Dahiya et al., 2005a). Although there may be positional cues from neighboring cells that are essential in dictating cell fate, it may be assumed that the formation of PX or MX must be dictated by a specific genetic program inherent to each cell type. By following the cell fate of individual, isolated zinnia cells grown on semisolid medium in vitro, we observed that whether a cell will become a PX or an MX TE is determined at very early stages with no changes in secondary wall patterns over time, despite the absence of positional cues in the zinnia system. However, we show that, given the right signals, PX and MX determination can be altered. This flexibility is nicely illustrated by the fact that exogenous GGMOs can redirect the differentiation of zinnia mesophyll cells from a PX- to a MX-rich population. Genetic approaches are extremely useful in dissecting PX and MX genetic programs. One interesting example is the wilty (wi) maize (Zea mays) mutant (Postlethwait and Nelson, 1957). This mutant was originally characterized by its inability to resist when subjected to growth conditions in which the transpiration rate was high. An anatomical study of the vascular bundles of internodes of wi indicated that the two large MX vessels typically found in mature wild-type maize were either immature or lacking altogether. In contrast, early stage PX formation was normal in the wi background. More recently, nonallelic wilty2, wilty3, and wilty-2445 mutants have been identified and are also characterized by aberrant MX formation (Rock and Ng, 1999). The identification of these genes would be of extreme interest in characterizing novel underlying mechanisms of MX formation. GGMOs and Global Gene Expression in Zinnia Cultures Another approach to understanding the genetic control of TE differentiation has been to analyze global gene expression using macroarray and microarray technologies (Demura et al., 2002; Pesquet et al., 2005). To identify potential actors in GGMO-signaling pathways in relation to TE formation, we compared transcriptomic profiles of zinnia cultures in the presence and absence of GGMOs. It has been previously demonstrated that GGMOs interfere with auxin-mediated processes such as elongation growth (Auxtová et al. 1995; Auxtová-Šamajová et al., 1996; Kollárová et al., 2006). The fact that an auxin-induced gene, zIAA8, was down-regulated by GGMOs provides an argument that GGMOs may also interact with auxin-signaling pathways during in vitro TE formation. However, it remains unclear how GGMO-mediated interference with auxin signaling may result in increased cell population density and/or enhanced MX formation. Finally, another feature of GGMO inhibition of auxin-mediated elongation growth was a concomitant increase in wall-bound α-l-arabinofuranosidase activity in pea segments (Bilisics et al., 2004). In our study, one of the only genes that were up-regulated by GGMOs was a putative α-l-arabinofuranosidase. This enzyme could potentially cause significant changes in remodeling of the cell wall network and influence the pattern of secondary wall deposition. The final phase of TE differentiation is programmed cell death (PCD), which is characterized by the participation of several hydrolases (nucleases, proteases) that accumulate in the vacuole and degrade the cellular content when the tonoplast is disrupted (Fukuda, 1997; Kuriyama, 1999). It has been shown that the TED4 protein, which belongs to the nsLTP family, accumulates prior to the release of proteases from dying TEs and may play a role in protecting living cells from harmful proteases (Endo et al., 2001). In our experiments, TED4-3 is up-regulated after 60 h in the presence of GGMOs. Although we found that GGMOs increased cell population density, we did not, on the other hand, observe a protective effect against cell death in zinnia cultures by GGMOs. Our transcriptomic studies indicated that very few genes were more highly expressed in the presence of GGMOs. The observation that the large majority were down-regulated may be explained by the fact that the macroarrays that were used in this study were established with cDNAs from a first-burst PX-rich library (Pesquet et al., 2005). Thus, if GGMOs reduce the relative proportion of PX TEs, it is reasonable to think that levels of PX-TE gene expression would decrease accordingly. When considering Cys proteinases, several were down-regulated by GGMOs. In zinnia cultures, TE differentiation has been correlated with increased Cys proteinase gene expression (Minami and Fukuda, 1995; Ye and Varner, 1996; Demura et al., 2002). Recently, one of them, ZeCP4, has been localized in TEs in vitro, suggesting that it is part of the PCD genetic program (Pesquet et al., 2004). Runeberg-Roos and Saarma (1998) showed that a barley (Hordeum vulgare) vacuolar aspartic proteinase (phytepsin), a plant homolog to cathepsin D implicated in PCD in HeLa cells, is more closely associated with MX formation and that the Cys proteinase, aleurain, is preferentially localized in PX cells in barley roots. These results nicely show a division of labor of proteases involved in either PX or MX PCD. In the presence of GGMOs, four Cys proteinases, CP-1, CP-3, CP-6, and CP-7, were down-regulated. We postulate that these proteinases may be more closely associated with PX rather than MX PCD in zinnia cultures. We provide evidence herein of a novel role for GGMOs in vascular development. By coupling physiological and transcriptomic approaches, we have also identified candidate genes that may be specifically involved in PX and MX TE differentiation programs. It would be of interest in the future to determine to what extent other oligosaccharides (i.e. xyloglucans and oligogalacturonides) may also be involved in signaling pathways of vascular differentiation. MATERIALS AND METHODS Plant Material Seeds of zinnia (Zinnia elegans cv Envy; Hem Zaden BV) were grown in potting soil in controlled-environment chambers under a 16-h photoperiod, 170 μE m−2 s−1 irradiance, at 27°C to 29°C. Xylogenic Cell Cultures Mesophyll cells were isolated from the first true leaves of 14-d-old seedlings of zinnia and cultured in a liquid medium containing 0.1 mg L−1 α-naphthylacetic acid (NAA) and 0.2 mg L−1 benzylaminopurine (BAP; Sigma-Aldrich) as described by Fukuda and Komamine (1980a). For cultures in semisolid conditions (Fig. 4C only), the medium was supplemented with 0.75% agarose (type VII: low gelling temperature; Sigma-Aldrich). Cell Staining and Counting Vital staining and population density of living cells were performed with 0.2% (w/v) Evans blue. To check division, cells were stained with 0.01% (w/v) Calcofluor White (fluorescent brightener 28; Sigma-Aldrich) and 0.001% (w/v) DAPI (Sigma-Aldrich), and dividing cells were identified by cell walls separating daughter cells. To determine cell viability and cell population density, 200 cells from each of three repetitions (i.e. 600 cells/treatment) were counted using light microscopy. The number of viable and dead TEs was counted after 60 h with 0.001% fluorescein diacetate (FDA; Sigma-Aldrich). Ten independent experiments have been done (only one is shown in this article). Cells were observed using an inverted microscope (DMIRBE; Leica) with bright-field optics or epifluorescence illumination (FDA, excitation filter BP 450–490 nm, suppression filter LP 515 nm; Calcofluor White, excitation filter BP 270–380 nm, emission filter BP 410–580 nm). Image acquisition was performed using a CCD camera (Color Coolview; Photonic Science). GGMOs GGMOs were prepared by partial acid hydrolysis of pure galactoglucomannan from spruce (Picea abies L. Karst) secondary cell walls and purified by paper and gel permeation chromatography, and precipitation with saturated Ba(OH)2 (Capek et al., 2000). As pure galactoglucomannan did not contain Xyl, Ara, and uronic residues, the resulting GGMO fraction was devoid of oligoxyloglucan, oligogalacturonic acid, and oligoarabinan. The pure GGMO fraction used in this study had a degree of polymerization 4 to 8. This mixture has already been shown to be highly active in various biological assays (Bilisics et al., 2004). It contained Man (70.4%), Glc (25.1%), and Gal (4.5%). GGMOs (20, 50, or 100 μg mL−1) and hormones (NAA and BAP) were added to the culture medium immediately after cell isolation. Aqueous stock solutions of GGMOs were filter sterilized prior to use. Gene Expression Analysis RNA Extraction One milliliter of Extract-All (Eurobio) was added to the cell pellet obtained after centrifugation of 10 mL of cell suspension (2 min at 150 g). Cells frozen in liquid N2 were conserved at −80°C. RNA was isolated according to the Extract-All protocol, then incubated with 5 units of RQ1 RNase-free DNase I (Promega) for 1 h at 37°C and again treated with Extract-All to remove DNase. RNA was quantified spectrophotometrically at 260 nm using a Biophotometer (Eppendorf) and verified by gel electrophoresis. Macroarray Probe Synthesis and Macroarray Hybridization First-strand cDNA probes were synthesized from 7 μg of total RNA from cell cultures for three different comparisons: (1) a TE-inductive medium for 60 h in the presence/absence of 50 μg mL−1 GGMO; (2) 60 h versus 5 d in a TE-inductive medium without GGMOs; and (3) cells cultured for 24 h without hormones. Hormones were added after 24 h with or without GGMOs and cells were harvested 30 min, 4 h, and 12 h after hormone addition. For hybridizations, RNA was pooled from two to three cultures per treatment. Total RNA was first annealed with 500 ng of ABgene (Epsom) anchored oligo NVdT15 and then reverse transcribed for 1 h at 42°C using 10 units of ABgene-iT blend reverse transcriptase in the presence of 40 μCi [α-33P]dCTP. RNA was then hydrolyzed with 0.04% (w/v) SDS, 20 mm EDTA (pH 8.0), and 360 mm NaOH for 30 min at 65°C and neutralized with 260 mm Tris-HCl (pH 8.0) and 160 mm HCl. To remove nonincorporated radiolabeled dCTP, cDNAs were precipitated with 140 mm sodium acetate (pH 5.3), 0.07% (w/v) mussel glycogen (Sigma-Aldrich), and 60% isopropanol for 1 h at −20°C, washed with 70% ethanol, and dried. cDNA labeling efficiency was quantified by scintillation counting. The specific activity of all probes was adjusted with water to about 160 cpm μL−1. The zinnia macroarrays used in this study have been described in detail by Pesquet et al. (2005). Membranes were prehybridized in 3× SSC, 0.5% (w/v) SDS, 10% (w/v) polyethylene glycol 6000, 0.2% (w/v) low-fat milk for 2 h at 65°C, and hybridized overnight with fresh prehybridization solution at the same temperature. Membranes were then washed with 3× SSC/0.5% (w/v) SDS for 15 min at 65°C and placed into a phosphor imager cassette (Molecular Dynamics; Amersham-Pharmacia) for 72 h. Image scanning was performed at 50 μm pixel−1 with a Storm820 scanner and the resulting hybridization signals were quantified using ImageQuant 5.0 (Molecular Dynamics; Amersham-Pharmacia). The average signal intensity values from the duplicate set were calculated for each gene showing a differential expression in the different culture conditions. Among these genes, those that were localized at different positions on the array due to library redundancy were averaged together. Reproducibility of hybridizations has previously been estimated by comparing raw signal intensity values from the duplicate spots, and genes that showed greater than a 2-fold difference in average signal values for each comparison were defined as differentially expressed genes. Preparation of Biological Material for Transmission Electron Microscopy Cells were centrifuged for 2 min at 150g after 4 d of culture. Pellets were rinsed with fresh medium and embedded in 1% (w/v) agarose (type VII: low gelling temperature; Sigma-Aldrich). Cells were fixed with 2.5% (v/v) glutaraldehyde in 0.05 m cacodylate buffer (pH 7.0) for 2 h at room temperature. They were then washed extensively with water and dehydrated in an ascending aqueous ethanol series (20%, 40%, 60%, 80% ethanol two times for 15 min each, and absolute ethanol two times for 30 min). The material was infiltrated with LR White resin and polymerized overnight at 70°C. Ultrathin sections (80–90 nm) were collected on gold grids. Immunogold Labeling of Mannan Polysaccharide Epitopes for Electron Microscopy The β-1,4-mannan polyclonal antibodies used for electron microscopy and immunoblot assays have been previously characterized by Handford et al. (2003). They specifically recognize glucomannan, mannan, or galactomannan epitopes. For immunolabeling experiments, sections were incubated in a blocking solution, PBST/bovine serum albumin (BSA), composed of 1% (w/v) BSA in PBST (0.14 m NaCl, 2.7 mm KCl, 7.8 mm Na2HPO4, 1.5 mm KH2PO4, 2% [v/v] Tween 20, pH 7.2), for 2 h at room temperature. For competition experiments, a 10-min preincubation was performed with the primary antibodies and 20 mg L−1 β-1,4-mannan and 200 mg L−1 konjak glucomannan (Megazyme) in PBST/BSA. The grids were then incubated overnight with the β-1,4-mannan polyclonal antibodies (produced from rabbit) diluted 1:5 (v/v) with PBST/BSA, with or without competitors. The grids were then washed with PBST/BSA before being transferred on a droplet of secondary antibody (10-nm gold-conjugated anti-rabbit) diluted 1:50 (v/v) in PBST/BSA for 2 h. The grids were washed and stained with an aqueous solution of uranyl acetate (5% [w/v]) at room temperature. Observations were made using a transmission electron microscope (Hitachi H600) at 75 kV and images acquired on Kodak electron microscope film. Immunoblot Assays For immunoblot assays, aliquots of medium were harvested from day 0 to day 6 from cell cultures with and without hormones and centrifuged (5 min at 13,200 rpm). The supernatants were boiled for 15 min and then spotted (6 μL) onto nitrocellulose membranes (Hybond ECL; Amersham). Mannan and oligomannan [(Man)6; 6 nmol; Megazyme] were spotted as controls. The membranes were air dried at room temperature for at least 30 min. After blocking for 1 h in PBS (137 mm NaCl, 2.7 mm KCl, 18 mm KH2PO4/Na2HPO4, pH 7.4) containing 5% (w/v) fat-free milk powder (5% m PBS), membranes were incubated for one night in primary antibodies diluted (1:300) in 5% m PBS. For competition experiments, primary antibodies were first incubated with 5 mg mL−1 glucomannan in PBS for 4 h. To decrease the background, a preincubation of the antibody with nitrocellulose membrane was performed. After washing, membranes were incubated for 1 h in secondary antibody (anti-rabbit horseradish peroxidase conjugate; Sigma) diluted 1:1,000 in 5% m PBS. Membranes were thoroughly washed prior to development in SuperSignal West Pico chemiluminescent substrate (Pierce). Preparation of Cell Wall for PACE Experiment A suspension aliquot was taken on day 0 to day 6 from zinnia cell cultures with and without hormones. Cells were incubated for 30 min in ethanol:water (9:1 [v/v]) at 65°C to inactivate enzymes. Cells were ground in a Mixer Mill MM200 (Glen Creston). The homogenate was centrifuged at 5,000 rpm for 15 min. The pellet was briefly washed with ethanol, methanol:chloroform (2:3 [v/v]; 1 h), ethanol:water (6:4 [v/v]; 2 × 5 min), and ethanol (5 min). The remaining pellet, containing cell walls, was dried overnight at 80°C (Goubet et al., 2002). Before hydrolysis with α-1,4-mannanase (Man5A; Cellvibrio japonicus), the cell walls were treated with 100 μL ammonium solution for 30 min and then dried. Polysaccharide Fingerprinting Polysaccharide fingerprints were determined by PACE. Mannan polysaccharides or cell walls were treated with Man5A in a total volume of 100 μL for 4 h. The reaction was buffered in 0.1 m ammonium acetate, pH 6. Controls without substrates or enzymes were performed under the same conditions to identify any unspecific compounds in the enzymes, polysaccharides/cell walls, or labeling reagents. The reactions were stopped by boiling for 20 min. Man5A was analyzed for the purity or/and specificity using the PACE method as described by Goubet et al. (2002). 8-Aminonaphthalene-1,3,6-trisulfonic acid (ANTS) was purchased from Molecular Probes. Acrylamide solution with a ratio of acrylamide to N,N′-methylenebisacrylamide (29:1) was obtained from Severn Biotech Ltd. Derivatization using ANTS, electrophoresis, and imaging was carried out according to Goubet et al. (2002). Sequence data from this article can be found in the GenBank/EMBL data libraries under accession numbers DQ336881 to DQ336891. ACKNOWLEDGMENTS We wish to thank Guylaine Keller for advice on amplifying cDNA clones, the “Génopole de Toulouse” for spotting facilities, and especially Nathalie Ladouce for help in spotting macroarrays. We acknowledge Caroline Deswarte and Sabine Guillaumie for assistance with hybridizations, and we also thank Prof. Harry Gilbert (University of Newcastle, UK) for kindly providing Man5A, and Magalie Pichon for critically reading this manuscript. LITERATURE CITED Akiyama Y, Eda S, Hori M, Kato K ( 1983 ) A galactoglucomannan from extracellular polysaccharides of suspension-cultured cells of Nicotiana tabacum. Phytochemistry 22 : 1177 – 1180 Crossref Search ADS Albersheim P, Darvill AG ( 1985 ) Oligosaccharins. Sci Am 253 : 58 – 64 PubMed Auxtová O, Lišková D, Kákoniová D, Kubačková M, Karácsonyi Š, Bilisics L ( 1995 ) Effect of galactoglucomannan-derived oligosaccharides on elongation growth of pea and spruce stem segments stimulated by auxin. Planta 196 : 420 – 424 Crossref Search ADS Auxtová-Šamajová O, Lišková D, Kákoniová D, Kubačková M, Karácsonyi Š, Bilisics L ( 1996 ) Inhibition of auxin stimulated short-term elongation growth of pea stem segments by galactoglucomannan-derived oligosaccharides. J Plant Physiol 147 : 611 – 613 Crossref Search ADS Bierhorst DW ( 1960 ) Observations of tracheary elements. Phytomorphol 10 : 249 – 305 Bilisics L, Vojtaššák J, Capek P, Kollárová K, Lišková D ( 2004 ) Changes in glycosidase activities during galactoglucomannan oligosaccharide inhibition of auxin induced growth. Phytochemistry 65 : 1903 – 1909 Crossref Search ADS PubMed Brower DL, Hepler PK ( 1976 ) Microtubules and secondary wall deposition in xylem: the effects of isopropyl N-phenylcarbamate. Protoplasma 87 : 91 – 111 Crossref Search ADS PubMed Busse JS, Evert RF ( 1999 a) Vascular development in Arabidopsis. Int J Plant Sci 160 : 1 – 13 Crossref Search ADS Busse JS, Evert RF ( 1999 b) Vascular transition in Arabidopsis. Int J Plant Sci 160 : 241 – 251 Crossref Search ADS Capek P, Kubačková M, Alföldi J, Bilisics L, Lišková D, Kákoniová D ( 2000 ) Galactoglucomannan from the secondary cell wall of Picea abies L. Karst. Carbohydr Res 329 : 635 – 645 Crossref Search ADS PubMed Carpita NC, Gibeaut DM ( 1993 ) Structural models of primary cell walls in flowering plants: consistency of molecular structure with the physical properties of the walls during growth. Plant J 3 : 1 – 30 Crossref Search ADS PubMed Dahiya P, Dimitra Milioni D, Wells B, Stacey N, Roberts K, McCann MC ( 2005 a) A RING domain gene is expressed in different cell types of leaf trace, stem, and juvenile bundles in the stem vascular system of zinnia. Plant Physiol 138 : 1383 – 1395 Crossref Search ADS PubMed Dahiya P, Findlay K, Roberts K, McCann MC ( 2005 b) A fasciclin-domain containing gene, ZeFLA11, is expressed exclusively in xylem elements that have reticulate wall thickenings in the stem vascular system of Zinnia elegans cv Envy. Planta 3 : 1 – 11 Demura T, Fukuda H ( 1993 ) Molecular cloning and characterization of cDNAs associated with tracheary element differentiation in cultured zinnia cells. Plant Physiol 103 : 815 – 821 Crossref Search ADS PubMed Demura T, Fukuda H ( 1994 ) Novel cell vascular cell-specific genes whose expression is regulated temporally and spatially during vascular system development. Plant Cell 6 : 967 – 981 PubMed Demura T, Tashiro G, Horiguchi G, Kishimoto N, Kubo M, Matsuoka N, Minami A, Nagata-Hiwatashi M, Nakamura K, Okamura Y, et al ( 2002 ) Visualization by comprehensive microarray analysis of gene expression programs during transdifferentiation of mesophyll cells into xylem cells. Proc Natl Acad Sci USA 99 : 15794 – 15799 Crossref Search ADS PubMed Dey PM ( 1978 ) Biochemistry of plant galactomannans. Adv Carbohydr Chem Biochem 35 : 341 – 376 Crossref Search ADS Dey PM ( 1980 ) Biochemistry of α-D-galactosidic linkages in the plant kingdom. Adv Carbohydr Chem Biochem 37 : 283 – 372 Crossref Search ADS Endo S, Demura T, Fukuda H ( 2001 ) Inhibition of proteasome activity by the TED4 protein in extracellular space: a novel mechanism for protection of living cells from injury caused by dying cells. Plant Cell Physiol 42 : 9 – 19 Crossref Search ADS PubMed Esau K ( 1977 ) Anatomy of Seed Plants, Ed 2. Wiley, New York Fahn A ( 1990 ) Plant Anatomy, Ed 4. Pergamon Press, Oxford Falconer MM, Seagull RW ( 1985 ) Xylogenesis in tissue culture: taxol effects on microtubule reorientation and lateral association in differentiating cells. Protoplasma 128 : 157 – 166 Crossref Search ADS Fisher JB ( 1970 ) Xylem derived from the intercalary meristem of Cyperus alternifolius. Bull Torrey Bot Club 97 : 58 – 66 Crossref Search ADS Fosket DE ( 1968 ) Cell division and the differentiation of wound-vessel members in cultured stem segments of Coleus. Proc Natl Acad Sci USA 59 : 1089 – 1096 Crossref Search ADS PubMed Fukuda H ( 1997 ) Tracheary elements differentiation. Plant Cell 9 : 1147 – 1156 PubMed Fukuda H, Komamine A ( 1980 a) Establishment of an experimental system for the study of tracheary element differentiation in single cells isolated from the mesophyll of Zinnia elegans. Plant Physiol 65 : 57 – 60 Crossref Search ADS PubMed Fukuda H, Komamine A ( 1980 b) Direct evidence for cytodifferentiation to tracheary elements without intervening mitosis in a culture of single cells isolated from the mesophyll of Zinnia elegans. Plant Physiol 65 : 61 – 64 Crossref Search ADS PubMed Goodwin RH ( 1942 ) On the development of xylary elements in the first internode of Avena in dark and light. Am J Bot 29 : 818 – 828 Crossref Search ADS Goubet F, Jackson P, Deery M, Dupree P ( 2002 ) Polysaccharide analysis using carbohydrate gel electrophoresis (PACE): a method to study plant cell wall polysaccharides and polysaccharide hydrolases. Anal Biochem 300 : 53 – 68 Crossref Search ADS PubMed Goubet F, Misrahi A, Park SK, Zhang Z, Twell D, Dupree P ( 2003 ) AtCSLA7, a cellulose-synthase-like putative glycosyltransferase, is important for pollen tube growth and embryogenesis in Arabidopsis. Plant Physiol 131 : 547 – 557 Crossref Search ADS PubMed Groover A, Pattishall A, Jones A ( 2003 ) IAA8 expression during vascular cell differentiation. Plant Mol Biol 51 : 427 – 435 Crossref Search ADS PubMed Handford MG, Baldwin TC, Goubet F, Prime TA, Miles J, Yu X, Dupree P ( 2003 ) Localisation and characterisation of cell wall mannan polysaccharides in Arabidopsis thaliana. Planta 218 : 27 – 36 Crossref Search ADS PubMed Inoue T, Higuchi M, Hashimoto Y, Seki M, Kobayashi M, Kato T, Tabata S, Shinozaki K, Kakimoto T ( 2001 ) Identification of CRE1 as a cytokinin receptor from Arabidopsis. Nature 409 : 1060 – 1063 Crossref Search ADS PubMed Kollárová K, Henselová M, Lišková D ( 2005 ) Effect of auxins and plant oligosaccharides on the root formation and elongation growth of mung bean hypocotyls. Plant Growth Regul 46 : 1 – 9 Crossref Search ADS Kollárová K, Lišková D, Capek P ( 2006 ) Further biological characteristics of galactoglucomannan oligosaccharides. Biol Plant 50 : 232 – 238 Crossref Search ADS Kubo M, Udagawa M, Nishikubo N, Horiguchi G, Yamaguchi M, Ito J, Mimura T, Fukuda H, Demura T ( 2005 ) Transcription switches for protoxylem and metaxylem vessel formation. Genes Dev 19 : 1855 – 1860 Crossref Search ADS PubMed Kuriyama H ( 1999 ) Loss of tonoplast integrity programmed in tracheary element differentiation. Plant Physiol 121 : 763 – 774 Crossref Search ADS PubMed Liepman AH, Wilkerson CG, Keegstra K ( 2005 ) Expression of cellulose synthase-like (csl) genes in insect cells reveals that CslA family members encode mannan synthases. Proc Natl Acad Sci USA 102 : 2221 – 2226 Crossref Search ADS PubMed Lišková D, Auxtová O, Kákoniová D, Kubačková M, Karácsonyi Š, Bilisics L ( 1995 ) Biological activity of galactoglucomannan-derived oligosaccharides. Planta 196 : 425 – 429 Crossref Search ADS Lišková D, Capek P, Kollárová K, Slováková L', Kákošová A ( 2006 ) The potential of carbohydrates in plant regulation. In JA Teixeira da Silva, ed, Floriculture, Ornamental and Plant Biotechnology: Advances and Topical Issues, Ed 1. Global Science Books, London, pp 373–378 Lundqvist J, Teleman A, Junel L, Zacchi G, Dahlman O, Tjerneld F, Stalbrand H ( 2002 ) Isolation and characterization of galactoglucomannan from spruce (Picea abies). Carbohydr Polym 48 : 29 – 39 Crossref Search ADS Mähönen AP, Bishopp A, Higuchi M, Nieminen KM, Kinoshita K, Tormakangas K, Ikeda Y, Oka A, Kakimoto T, Helariutta Y ( 2006 ) Cytokinin signaling and its inhibitor AHP6 regulate cell fate during vascular development. Science 311 : 94 – 98 Crossref Search ADS PubMed Mähönen AP, Bonke M, Kauppinen L, Riikonen M, Benfey PN, Helariutta Y ( 2000 ) A novel two-component hybrid molecule regulates vascular morphogenesis of the Arabidopsis root. Genes Dev 14 : 2938 – 2943 Crossref Search ADS PubMed Minami A, Fukuda H ( 1995 ) Transient and specific expression of a cysteine endopeptidase during autolysis in differentiating tracheary elements from Zinnia mesophyll cells. Plant Cell Physiol 36 : 1599 – 1606 PubMed Motose H, Sugiyama M, Fukuda H ( 2004 ) A proteoglycan mediates inductive interaction during plant vascular development. Nature 429 : 873 – 878 Crossref Search ADS PubMed Mourelatou M, Doonan JH, Mccann MC ( 2004 ) Transition of G1 to early S phase may be required for zinnia mesophyll cells to trans-differentiate to tracheary elements. Planta 220 : 172 – 176 Crossref Search ADS PubMed Ohdaira Y, Kakegawa K, Amino S ( 2002 ) Activity of cell-wall degradation associated with differentiation of isolated mesophyll cells of Zinnia elegans into tracheary elements. Planta 215 : 177 – 184 Crossref Search ADS PubMed Pesquet E, Barbier O, Ranocha P, Jauneau A, Goffner D ( 2004 ) Multiple gene detection by in situ RT-PCR in isolated plant cells and tissues. Plant J 39 : 947 – 959 Crossref Search ADS PubMed Pesquet E, Jauneau A, Digonnet C, Boudet AM, Pichon M, Goffner D ( 2003 ) Zinnia elegans: the missing link from in vitro tracheary elements to xylem. Physiol Plant 119 : 463 – 468 Crossref Search ADS Pesquet E, Ranocha P, Legay S, Digonnet C, Barbier O, Pichon M, Goffner D ( 2005 ) Novel markers of xylogenesis in zinnia are differentially regulated by auxin and cytokinin. Plant Physiol 139 : 1821 – 1839 Crossref Search ADS PubMed Postlethwait SN, Nelson OE ( 1957 ) A chronically wilted mutant of maize. Am J Bot 44 : 628 – 633 Crossref Search ADS Roberts AW, Donovan SG, Haigler CH ( 1997 ) A secreted factor induces cell expansion and formation of metaxylem-like tracheary elements in xylogenic suspension cultures of zinnia. Plant Physiol 115 : 683 – 692 Crossref Search ADS PubMed Roberts AW, Haigler CH ( 1994 ) Cell expansion and tracheary element differentiation are regulated by extracellular pH in mesophyll cultures of Zinnia elegans L. Plant Physiol 105 : 699 – 706 Crossref Search ADS PubMed Rock CD, Ng PPF ( 1999 ) Dominant wilty mutants of Zea mays (Poaceae) are not impaired in abscisic acid perception or metabolism. Am J Bot 86 : 1796 – 1800 Crossref Search ADS PubMed Runeberg-Roos P, Saarma M ( 1998 ) Phytepsin, a barley vacuolar aspartic proteinase, is highly expressed during autolysis of developing tracheary elements and sieve cells. Plant J 15 : 139 – 145 Crossref Search ADS PubMed Schröder R, Nicolas P, Vincent SJF, Fischer M, Reymond S, Redgwell RJ ( 2001 ) Purification and characterisation of a galactoglucomannan from ripe kiwifruit (Actinia deliciosa). Carbohydr Res 331 : 291 – 306 Crossref Search ADS PubMed Sims IM, Craik DJ, Bacic A ( 1997 ) Structural characterisation of galactoglucomannan secreted by suspension-cultured cells of Nicotiana plumbaginifolia. Carbohydr Res 303 : 79 – 92 Crossref Search ADS PubMed Slováková L', Lišková D, Capek P, Kubačková M, Kákoniová D, Karácsonyi Š ( 2000 ) Defence responses against TNV infection induced by galactoglucomannan-derived oligosaccharides in cucumber cells. Eur J Plant Pathol 106 : 543 – 553 Crossref Search ADS Smith GH, Kersten H ( 1942 ) The relation between xylem thickenings in primary roots of Vicia faba seedlings and elongation as shown by soft x-ray irradiation. Bull Torrey Bot Club 69 : 221 – 234 Crossref Search ADS Stacey NJ, Roberts K, Carpita NC, Wells B, McCann MC ( 1995 ) Dynamic changes in cell surface molecules are very early events in the differentiation of mesophyll cells from Zinnia elegans into tracheary elements. Plant J 8 : 891 – 906 Crossref Search ADS Torrese-Ruiz RA, Jurgens FG ( 1994 ) Mutations in the FASS gene uncouple pattern formation and morphogenesis in Arabidopsis development. Development 120 : 2967 – 2978 Crossref Search ADS PubMed Whitney SEC, Brigham JE, Darke AH, Reid JSG, Gidley MJ ( 1998 ) Structural aspects of the interaction of manna-based polysaccharides with bacterial cellulose. Carbohydr Res 307 : 299 – 309 Crossref Search ADS Ye ZH, Varner JE ( 1996 ) Induction of cysteine and serine proteases during xylogenesis in Zinnia elegans. Plant Mol Biol 30 : 1233 – 1246 Crossref Search ADS PubMed Author notes 1 This work was supported in part by the European Community (Marie Curie fellowship no. QLK3–CT–2001–60067to A.B.-K.) and the Slovakia Agency for the Support of Science and Technology (grant no. 51–013304), and by grants from the Biotechnology and Biological Sciences Research Council for work performed at Cambridge University. Preparation of galactoglucomannans was supported by the Slovak Grant Agency for Science (grant no. 2/4145/04). * Corresponding author; e-mail goffner@scsv.ups-tlse.fr; fax 33–5–62–19–35–02. The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instruction for Authors (www.plantphysiol.org) is: Deborah Goffner (goffner@scsv.ups-tlse.fr). www.plantphysiol.org/cgi/doi/10.1104/pp.106.085712 © 2006 American Society of Plant Biologists This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Galactoglucomannans Increase Cell Population Density and Alter the Protoxylem/Metaxylem Tracheary Element Ratio in Xylogenic Cultures of Zinnia
JF - Plant Physiology
DO - 10.1104/pp.106.085712
DA - 2006-10-10
UR - https://www.deepdyve.com/lp/oxford-university-press/galactoglucomannans-increase-cell-population-density-and-alter-the-sTPv0vGcsD
SP - 696
EP - 709
VL - 142
IS - 2
DP - DeepDyve
ER -