XTH20 and XTH19 regulated by ANAC071 under auxin flow are involved in cell proliferation in incised Arabidopsis inflorescence stems

Summary

One week after partial incision of Arabidopsis inflorescence stems, the repair process in damaged tissue includes pith cell proliferation. Auxin is a key factor driving this process, and ANAC071, a transcription factor gene, is upregulated in the distal region of the incised stem. Here we show that XTH20 and the closely related XTH19, members of xyloglucan endotransglucosylase/hydrolases family catalyzing molecular grafting and/or hydrolysis of cell wall xyloglucans, were also upregulated in the distal part of the incised stem, similar to ANAC071. XTH19 was expressed in the proximal incision region after 3 days or after auxin application to the decapitated stem. Horizontal positioning of the plant with the incised side up resulted in decreased Pro_DR5:GUS, ANAC071, XTH20, and XTH19 expression and reduced pith cell proliferation. In incised stems of Pro_35S:ANAC071‐SRDX plants, expression of XTH20 and XTH19 was substantially and moderately decreased, respectively. XTH20 and XTH19 expression and pith cell proliferation were suppressed in anac071 plants and were increased in Pro_35S:ANAC071 plants. Pith cell proliferation was also inhibited in the xth20xth19 double mutant. Furthermore, ANAC071 bound to the XTH20 and XTH19 promoters to induce their expression. This study revealed XTH20 and XTH19 induction by auxin via ANAC071 in the distal part of an incised stem and their involvement in cell proliferation in the tissue reunion process.

Introduction

Injuries caused by wind, herbivores, and chewing insects are one of the critical environmental factors affecting plant survival. Therefore, plants have developed primary physical barriers such as cuticle, bark, and thorns. Defense responses are activated by wound response hormones such as ethylene and jasmonic acid to prevent further damage when injuries occur (León et al., 2001). Unlike other organs, stems provide essential structural support and deliver nutrients, water, and physiological and chemical information between organs through vascular tissues (Satoh, 2006; Kehr and Buhtz, 2008). Therefore, damaged stems need to be repaired as soon as possible to recover their physiological functions. This ability has long been used for agricultural grafting. The mechanism of the tissue reunion process has been explored to improve grafting efficiency and overcome tissue incompatibility. Turnbull et al. (2002) developed micrografting in young Arabidopsis seedlings to study long‐distance signaling pathways including flowering, systemic acquired resistance, and abiotic stress responses. In mature Arabidopsis, homografts and heterografts were used as a model system to study the compatibility and ontogeny of graft union formation (Flaishman et al., 2008).

Although regeneration of vascular tissue in damaged stems has been widely observed (Flaishman et al., 2003), adherence between incised or separated tissues is also required for the tissue repair process. Auxin is required to heal the hypocotyl graft between the scion and root stock in Arabidopsis (Yin et al., 2012). Analysis of partially incised hypocotyls of cucumber and tomato also indicates the importance of endogenous gibberellin and Mn, Zn, and B mineral elements for repairing incised hypocotyls (Asahina et al., 2002, 2006). Microarray analysis of Arabidopsis inflorescence stems, which was used to better understand the molecular basis of tissue repair, showed the upregulation of genes involved in cell division, cell wall modification, and phytohormone synthesis and signaling (Asahina et al., 2011). According to analyses of the pin1‐1 mutant, which is lacking a functional auxin efflux carrier for polar transport of auxin, and decapitated plants with or without indole acetic acid (IAA) application, auxin: (i) is a key factor controlling cell division in pith tissue during the repair of incised Arabidopsis stems, and (ii) induces expression of ANAC071, a NAM, ATAF1/2, CUC2 (NAC)‐type transcription factor, in the distal region of the incision (Asahina et al., 2011). Recently, the involvement of ARF6 and ARF8 has been clarified in auxin signaling for this process (Pitaksaringkarn et al., 2014). The NAC domain was originally characterized from consensus sequences from petunia NAM and from Arabidopsis ATAF1, ATAF2, and CUC2 (Olsen et al., 2005). NAC family genes are plant‐specific transcription factors and are expressed in various developmental stages and tissues (Kubo et al., 2005; Mitsuda et al., 2005). The control of downstream target genes by NAC has been demonstrated in dehydration stress responses (Tran et al., 2004), secondary wall biosynthesis (Zhong et al., 2007), and defense responses in Arabidopsis (Bu et al., 2008).

Using Pro_35S:ANAC071‐SRDX transgenic plants, we have shown that ANAC071 controls cell proliferation in pith tissue after an incision is made (Asahina et al., 2011). However, the control of downstream targets by ANAC071 remains unknown. The target genes should be included in the upregulated genes reported in our previous microarray analysis. Indeed, XTH20 (xyloglucan endotransglucosylase/hydrolases20) was upregulated in the microarray analysis of the tissue reunion process (Asahina et al., 2011). The XTH family is implicated as having a principal role in the construction and restructuring of xyloglucan crosslinking within the cellulose/xyloglucan framework by catalyzing molecular grafting and/or hydrolysis of xyloglucans. XTHs are also involved in various physiological processes, mainly by participating in cell enlargement (Nishitani and Tominaga, 1992; Okazawa et al., 1993; Fry, 2004). XTH20 is induced by auxin in roots (Vissenberg et al., 2005; Osato et al., 2006). Here we found that XTH20 was also induced in decapitated plants with IAA application as well as when ANAC071 was induced. These results suggested that XTH20 may be a downstream target of ANAC071 as a consequence of auxin action in incised stems. We also show the involvement of XTH in cell proliferation in plant tissue.

Results

Differences in sugar composition in incised and non‐incised Arabidopsis inflorescence stems

In our microarray analysis of Arabidopsis inflorescence stems, upregulation of some enzymes for the metabolism of cell wall polysaccharides was observed (Asahina et al., 2011). The families of upregulated genes included the enzymes that degrade cell wall polysaccharides such as polygalacturonase, pectate lyase, pectin methylesterase, glycosyl hydrolase, and β‐glucosidase. In contrast, FUT3, a homolog of FUT1 (xyloglucan fucosyltransferase) (Vanzin et al., 2002) was the only upregulated glycosyltransferase. To comprehensively survey cell wall polysaccharides for this study, we sequentially treated Arabidopsis cell walls with α‐amylase, endo‐polygalacturonase, and Na₂CO₃ to extract the pectin fractions and then treated them with 1 m KOH and 4 m KOH to extract hemicellulose fractions; the insoluble residue was considered to be a cellulose fraction. Monosaccharide composition analysis was performed on these fractions. Differences in monosaccharide composition were present only in the pectin fractions, and the arabinose content increased in incised stems (Table S1).

Glycosyl‐linkage analysis was also performed on the α‐amylase fraction, because this fraction contains the majority of pectic polysaccharides and because half of the cell wall pectin is released by this treatment in Arabidopsis (Zablackis et al., 1995). Glycosyl‐linkage analysis showed that terminal and 5‐linked arabinofuranosyl residues increased in incised stems (Table S2). These results suggest that branched arabinan side chains of pectin increase in incised stems, because the backbones of pectic arabinans specifically consist of α‐1,5‐l‐arabinofuranose.

Although these results showed an increase in pectin in incised stems, our previous microarray data (Asahina et al., 2011) did not show the upregulation of the genes involved in pectin biosynthesis including the GAUT1 (Atmodjo et al., 2011) and ARAD1 (Harholt et al., 2006) gene families. In addition to the upregulation of genes for the enzymes responsible for degrading cell wall polysaccharides, we did note the upregulation of XTH20 and Expansin10 (Cho and Cosgrove, 2000), which are involved in the modification of the cross‐link between xyloglucan and cellulose microfibril. These results indicated that the modification and reconstruction of the cell wall network might be promoted during tissue reunion.

Spatial and temporal expression of ANAC071, XTH20, and XTH19 in incised Arabidopsis inflorescence stems

The expression levels of ANAC071, XTH20, and XTH19, which is phylogenetically closest to XTH20, were determined in various organs of Arabidopsis by quantitative RT‐PCR. These genes were highly expressed in incised inflorescence stems. XTH20 and XTH19 were also highly expressed in roots, but the concomitant expression of ANAC071 was observed only in incised stems (Figure 1a–c). The tissue‐specific expression of ANAC071, XTH20, and XTH19 was examined in more detail using promoter‐reporter markers. The 1.5‐kb upstream region of each gene was fused with the β‐glucuronidase (GUS) reporter gene and introduced into Arabidopsis. Pro_ANAC071:GUS was highly expressed in the distal region 1 and 3 days after the incision was made (Figure 1d,e), confirming our previous quantitative RT‐PCR results showing that ANAC071 is specifically expressed in the distal region of incised inflorescence stems (Asahina et al., 2011). The expression of XTH20 and XTH19 was higher in the distal than proximal regions of the incision at 1 day (Figure 1p,q), but expression of XTH19 was similar in both regions at 3 and 5 days (Figure 1q). In Pro_XTH20:GUS and Pro_XTH19:GUS transgenic plants, GUS activity was higher in the distal than the proximal region for Pro_XTH20:GUS, but expression was similar in both regions for Pro_XTH19:GUS (Figure 1h–o). XTH19 was also expressed in the siliques and flowers of intact plants (Figure 1c). These results indicated that the expression of ANAC071 and the expression of XTH20 are highly correlated.

Figure 1

Tissue‐specific expression of ANAC071, XTH20, and XTH19.

(a–c) The expression of ANAC071 (a), XTH20 (b), and XTH19 (c) was examined by quantitative RT‐PCR in roots, rosette leaves, cauline leaves, incised and non‐incised inflorescence stems, siliques, and flowers in wild‐type plants. Actin7 was used as an internal control.

(d–o) Promoter activity of ANAC071, XTH20, and XTH19. GUS activities were detected in incised inflorescence stems of Pro_ANAC071:GUS, Pro_XTH20:GUS, and Pro_XTH19:GUS transgenic plants. The photographs were taken 1 day (d, h, and l), 3 days (e, i, m), 5 days (f, j, n), and 7 days (g, k, o) after the incision was made (indicated by arrowhead). Bar = 0.5 mm.

(p, q) The expression of XTH20 (p) and XTH19 (q) in non‐incised and distal and proximal regions of incised inflorescence stems at 1, 3, and 5 days after the incision was made. Mean ± standard deviation (SD) (n = 3).

Induction of ANAC071, XTH20, and XTH19 by auxin in incised Arabidopsis inflorescence stems

Similar to previous results (Asahina et al., 2011), the expression of Pro_ANAC071:GUS in response to auxin was higher in the distal part of the incision. The expression of XTH20 and XTH19 was also induced by IAA application (Figure 2a,b). In Pro_XTH20:GUS transgenic plants, GUS activity was high in the distal region of the incision after IAA application, which was similar to the pattern with Pro_ANAC071:GUS (Figure 2d,f). In Pro_XTH19:GUS transgenic plants, GUS activity was detected in both the distal and proximal regions of the incision after IAA application, but the level of Pro_XTH19:GUS was higher in the distal region (Figure 2h). These results indicated that ANAC071, XTH20, and XTH19 are auxin‐inducible genes in incised Arabidopsis stems and that the expression of XTH20 is more closely correlated with that of auxin‐induced ANAC071 than is XTH19.

Figure 2

Effect of auxin on the induction of ANAC071, XTH20, and XTH19 in the tissue reunion process.

(a, b) Expression levels of XTH20 (a) and XTH19 (b) in incised (black bar) and non‐incised (gray bar) stems. The expression levels of XTH20 and XTH19 were examined by quantitative RT‐PCR in non‐decapitated plants and in decapitated plants 1 day after the incision was made after decapitation followed by the application of distilled water (D.W.) or 10⁻³ m indole acetic acid (IAA) at the top of the cut stems. Mean ± standard deviation (SD) (n = 3).

(c–h) Promoter activity of ANAC071, XTH20, and XTH19. GUS activity was detected 1 day after an incision was made in inflorescence stems of Pro_ANAC071:GUS, Pro_XTH20:GUS, and Pro_XTH19:GUS transgenic plants after decapitation followed by application of distilled water (D.W.) (c, e, g) or 10⁻³ m IAA (d, f, h). Bar = 0.5 mm.

Effect of gravity on tissue repair in Arabidopsis incised inflorescence stems

Auxin is involved in gravity responses in stems and roots, and thus the effect of the gravity response on the tissue repair process was examined. After incision of the stems, the plants were incubated in a horizontal position with the incision site facing either downward or upward, in the same or opposite direction as the gravity vector, respectively. Pith cell number 7 days after the incision was made did not increase at the cut surface for those plants grown with an upward‐facing incision (Figure 3a–d). In addition, GUS activity in Pro_ANAC071:GUS, Pro_XTH20:GUS, Pro_XTH19:GUS, and Pro_DR5:GUS (to monitor the auxin response) transgenic plants decreased when the incision sites were upward facing as compared with downward‐facing incisions and control (i.e. vertical) stems (Figure 3e–p). All of these results indicated a good correlation between the auxin response and ANAC071, XTH20, and XTH19 expression.

Figure 3

Effects of gravity on the tissue reunion process.

(a–c) Tissue reunion phenotypes observed 7 days after an incision was made in plants grown vertically (a) or horizontally with upward‐facing (b) and downward‐facing (c) incisions. Arrowheads indicate the position of the incision. g, gravity; pi, pith; co, cortex; vb, vascular bundle. Bar = 100 μm.

(d) Cell density in pith tissues calculated from sectioned samples as grown in (a–c). Mean ± standard deviation (SD) (n = 7). *P < 0.05 (t‐test). The calculation is described in Figure S5.

(e–p) Promoter activity of DR5, ANAC071, XTH20, and XTH19. GUS activity was detected in incised inflorescence stems of Pro_DR5:GUS, Pro_ANAC071:GUS, Pro_XTH20:GUS, and Pro_XTH19:GUS transgenic plants 3 days after an incision was made. Immediately after the incision was made, plants were oriented vertically as a control (e, h, k, n) or horizontally with upward‐facing (f, i, l, o) or downward‐facing (g, j, m, p) incisions. Bar = 0.5 mm.

Control of the expression of XTH20 and XTH19 by ANAC071 in incised Arabidopsis stems

As described above, the expression of ANAC071, XTH20, and XTH19 was correlated with the incision response (Figure 1) and with the auxin response (Figures 2 and 3), and ANAC071 is a key auxin response factor that controls tissue reunion (Asahina et al., 2011). Hence, the expression levels of XTH20 and XTH19 were examined in incised stems of Pro_35S:ANAC071‐SRDX transgenic plants to determine whether XTH20 and XTH19 are controlled by ANAC071. The expression level of XTH20 in incised stems substantially decreased in Pro_35S:ANAC071‐SRDX transgenic plants (Figure 4a). Moreover, the expression level of XTH19 in incised stems of Pro_35S:ANAC071‐SRDX transgenic plants decreased moderately (Figure 4b). Furthermore, decreased expression of XTH20 and XTH19 in the anac071 T‐DNA insertion mutant and increased expression in Pro_35S:ANAC071 transgenic plants were observed (Figure 4c,d). These results indicated that XTH20 and XTH19 may be downstream genes under the control of ANAC071 in incised Arabidopsis stems.

Figure 4

Control of the expression of XTH20 and XTH19 by ANAC071 during the tissue reunion process.

(a, b) Expression of XTH20 (a) and XTH19 (b) was examined by quantitative RT‐PCR in inflorescence stems of wild‐type (incised = blue; non‐incised = red) and Pro_35S:ANAC071‐SRDX transgenic (incised = green; non‐incised = purple) plants 0–7 days after the incision was made. Mean ± standard deviation (SD) (n = 3).

(c, d) Quantitative RT‐PCR analysis of XTH20 (c) and XTH19 (d) in incised (black bar) and non‐incised (gray bar) inflorescence stems of wild‐type, anc071, and Pro_35S:ANAC071 transgenic plants 3 days after the incision was made. Mean ± SD (n = 3).

Binding of ANAC071 to XTH20 and XTH19 promoters

The Arabidopsis NAC protein family targets the recognition sequence CATGT and the core binding site CACG to bind with promoters of downstream genes (Bu et al., 2008). The promoters of XTH20 and XTH19 at −923 bp and −102 bp upstream from their respective coding sequences (Figure 5a,b) had sequences similar to the recognition site (CATG) and core binding site (CACC) (Figure 5c,d). Hence, we examined whether ANAC071 binds to the promoters of XTH20 and XTH19 using an electrophoretic mobility shift assay (EMSA). Purified His‐ANAC071 fusion protein was incubated with labeled DNA probes, and an EMSA was performed. A high‐affinity DNA‐protein complex was detected ((Figure 5e,f); lane 2 arrowhead), and DNA binding was reduced by the addition of an excess of unlabeled oligonucleotide probes corresponding to the XTH20 or XTH19 promoter, which acted as specific competitors ((Figure 5e,f); lanes 3 and 4). Moreover, Agrobacterium‐mediated transient expression of ANAC071 driven by Pro_35S in the leaves of Pro_XTH20:GUS and Pro_XTH19:GUS transformants showed significantly higher GUS activity than the empty vector control (Figure S1). These results suggest that ANAC071 controls XTH20 and XTH19 expression by binding to their respective promoters.

Figure 5

Binding of ANAC071 to XTH20 and XTH19 promoters.

(a, b) Illustration of selected oligonucleotides containing the cis‐elements of the NAC recognition site (CATGT and CATG) and a presumed core binding site (CACC) in the promoters of XTH20 (a) (−923 bp) and XTH19 (b) (−102 bp). Black boxes indicate exons.

(c, d) Oligonucleotides used in the EMSA. The probes for XTH20 (c) and XTH19 (d) (pXTH20 and pXTH19, respectively) contained the cis‐element of the NAC recognition site (CATGT and CATG) and the presumed core binding site (CACC) in the XTH20 and XTH19 promoter regions.

(e, f) The EMSA showed binding of His‐ANAC071 to the CATGT, CACC, and CATG motifs in XTH20 (e) and XTH19 (f) promoters. Protein–DNA complexes were detected when His‐ANAC071 fusion protein was incubated with labeled pXTH20 or pXTH19 probes (lane 2). Unlabeled pXTH20 or pXTH19 (lanes 3 and 4) were used for 100‐fold (lanes 3) and 200‐fold (lanes 4) excess levels of competitors. Arrowheads indicate the position of the protein–DNA complex.

ANAC071, XTH19, and XTH20 control cell proliferation in pith tissue of incised Arabidopsis stems

Because XTH20 and XTH19 were controlled by ANAC071, the tissue reunion phenotype was examined in the anac071 mutant and in Pro_35S:ANAC071 transgenic plants, as well as in the xth19 and xth20 mutants 1, 3, 5, and 7 days after the incision was made (Figures 6, S2 and S3). The anac071 mutant had fewer pith cells than wild‐type (Figure 6a,b,i), as was previously described for Pro_35S:ANAC071‐SRDX transgenic plants (Asahina et al., 2011). Introduction of Pro_35S:ANAC071 into the anac071 background restored cell division in pith tissue (Figure 6c,i). Moreover, two lines of Pro_35S:ANAC071 transgenic plants showed higher cell numbers in pith tissue than in wild‐type control plants (Figures 6a,d,e,i). These results suggested that ANAC071 controls the tissue reunion process by controlling cell proliferation in the pith tissue.

Figure 6

Phenotypes of anac071, xth20, and xth19 single mutants and the xth20xth19 double mutant and the effect of ANAC071 overexpression on cell division in pith tissue.

(a–h) Tissue reunion phenotype observed 7 days after the incision was made. Pro_35S:ANAC071 was introduced into anac071 mutant (Pro_35S:ANAC071 (anac071)) or two wild‐type lines (Pro_35S:ANAC071 #2‐3 and #4‐1). Arrowheads indicate the position of the incision. pi, pith; co, cortex; vb, vascular bundle; WT, wild‐type. Bar = 100 μm.

(i) Cell density in pith tissue was calculated from sectioned samples. Mean ± standard deviation (SD) (n = 7). *P < 0.05 (t‐test compared with WT).

To monitor the effects of ANAC071 on auxin flow and/or responses in stems, the expression of IAA5, the incision‐inducible auxin response gene (Asahina et al., 2011), was examined in incised and non‐incised stems. IAA5 expression decreased in the non‐incised stems of anac071 and increased in those of Pro_35S:ANAC071 transgenic plants. In contrast, IAA5 expression in incised stems was not substantially different among wild‐type, anac071, and Pro_35S:ANAC071 transgenic plants (Figure S4). These results suggest that changes in auxin transport and/or responses may not be the main factor in the control of cell division by ANAC071.

In terms of general plant morphology, rosette leaf number, length, and width were reduced in the anac071 mutant and slight increases in those traits were observed in Pro_35S:ANAC071 transgenic plants (Figure S5), as was an increase in the size of mesophyll and epidermal cells in rosette leaves of Pro_35S:ANAC071 transgenic plants (Figure S6). These results suggest that cell growth is likely to be involved in the promotion of leaf growth by ANAC071.

To study the function of XTH20 in the tissue reunion process, we examined the xth20 T‐DNA insertion mutant. In the xth20 single mutant 7 days after the incision was made, the number of pith cells was not significantly different from wild‐type (Figure 6a,f,i). We hypothesized that XTH20 may act redundantly with XTH19. The single mutant of XTH19 had similar pith cell numbers as wild‐type control plants (Figure 6a,g,i). Furthermore, we generated the xth20xth19 mutant, which exhibited fewer pith cells 7 days after the incision was made than wild‐type (Figure 6h,i). In those incised stems 3 days after the incision was made, the expression of CYCB1;1 was not significantly different between wild‐type and xth20xth19 (Figure S7). These results support our hypothesis that XTH20 and XTH19 are functionally redundant and are involved in the cell proliferation of pith tissue during the tissue reunion process.

The primary root length of xth20xth19 was significantly decreased as compared with that of xth20, xth19, and wild‐type plants (t‐test, P < 0.05) (Figure S8a,b). In addition, the lengths of meristematic zones and epidermal cells of xth20xth19 roots were shorter than those of wild‐type (Figure S9a–f). These results are consistent with the finding that XTH20 and XTH19 were highly expressed in intact roots (Figure 1b,c). Because the levels of CYCB1;1 expression were not significantly different between wild‐type and xth20xth19 (Figure S10), these root‐expressed XTHs are likely to be involved in root growth mainly via the promotion of cell growth.

Discussion

Auxin is involved in the induction of ANAC071, XTH20, and XTH19

Auxin is produced in the apical meristem and is transported to all plant organs (Robert and Friml, 2009). Auxin plays an important role in cell proliferation and vascular differentiation (Berleth et al., 2000; Flaishman et al., 2003; Teale et al., 2006). Histochemical analysis of Pro_DR5:GUS transformants, which can be used to monitor auxin accumulation, showed high GUS activity in the distal region of the incision 3 days after the incision was made (Figure 3e). Tissue reunion is strongly inhibited by decapitation of or TIBA (an inhibitor of auxin polar transport) application to wild‐type plants and on pin1‐1 mutants, whereas the application of 10⁻³ m IAA to decapitated plants leads to the recovery of the reunion process (Asahina et al., 2011). These results indicate that auxin is a key factor in the regulation of tissue repair in the Arabidopsis incised inflorescence stem. Xie et al. (2000) reported that NAC1 is induced by auxin and promotes lateral root formation. Previous studies have suggested that some XTH genes, including XTH20 and XTH19, are regulated by auxin in the roots (Vissenberg et al., 2005; Osato et al., 2006). Application of IAA to decapitated stems increased both XTH20 and XTH19 expression (Figure 2a,b), and the direction of gravity controlled their expression (Figure 3l,o) as well as the expression of ANAC071 (Figure 3i). However, XTH19 was also expressed in the proximal region of the incision (Figure 1m,q) and 2(h)) and was constitutively expressed in the silique and flower (Figure 1c). Thus ANAC071, XTH20, and XTH19 were induced by auxin in the tissue reunion process, but XTH19 may also be controlled by other factors.

ANAC071 acts as a regulator in the tissue reunion process

The plant‐specific NAC transcription factor family, with 105 members predicted in the Arabidopsis genome (Ooka et al., 2003), is important for various physiological processes such as lateral root formation, auxin signaling, responses to abiotic stress (Olsen et al., 2005), xylem formation (Kubo et al., 2005), and thickening of the secondary wall (Mitsuda et al., 2005). We have previously reported the use of Pro_35S:ANAC071‐SDRX to show that ANAC071 controls the tissue reunion process in incised inflorescence stems (Asahina et al., 2011). Here we show that the expression levels of ANAC071 in the anac071 mutant and in Pro_35S:ANAC071 transgenic plants and the expression of Pro_35S:ANAC071‐SRDX were correlated with the expression of XTH20 and XTH19 (Figure 4a–d). The anac071 mutant showed strong inhibition of the reunion process (Figure 6b,i). Complementation of anac071 with ANAC071 also showed recovery of cell proliferation in pith tissue (Figure 6c,i). Moreover, wild‐type plants that over‐expressed ANAC071 had more cell proliferation than did wild‐type control plants (Figure 6a,d,e,i), but auxin flow and/or responses monitored by IAA5 expression were not largely enhanced in Pro_35S:ANAC071 transgenic plants (Figure S4).These results suggest that ANAC071 is an important transcription factor that controls its downstream genes and pith cell proliferation in the tissue reunion process.

ANAC071 controls XTH20 and XTH19

The expression of XTH20 and XTH19 in both incised and non‐incised tissue decreased in Pro_35S:ANAC071‐SRDX plants (Figure 4a,b). The expression of XTH20 and XTH19 decreased in anac071 plants, particularly 3 days after the incision was made, but substantially increased in Pro_35S:ANAC071 transgenic plants (Figure 4c,d). These results suggested that ANAC071 controls the expression of XTH20 and XTH19 in the tissue reunion process. Bu et al. (2008) demonstrated the direct binding of ANAC019 to the VSP1 promoter, which contains the recognition site (CATGT) and core binding site (CACG) for the NAC protein family. The promoters of XTH20 and XTH19 also contain a similar recognition site (CATGT and CATG) and core binding site (CACC), which differs from the usual core binding site (CACG) for NAC proteins (Figure 5c,d). EMSA assays (Figure 5e,f) and the promotion of the expressions of XTH20 and XTH19 by the transient expression of ANAC071 (Figure S1) suggest that the ANAC071 protein binds to the promoters of XTH20 and XTH19 to promote their expression and that the CACC sequence is important for recognition by ANAC071. The expression pattern of XTH20 (Figure 1i,p) was different from that of XTH19 (Figure 1m,q), suggesting that some unknown factor(s) participates in the expression of XTH19 in the proximal region of the incised stem.

XTH20 and XTH19 act redundantly in the tissue reunion process

Sugar composition and sugar linkage analysis of the fractions of cell wall polysaccharides showed the increase in arabinan side chains of pectin in incised stems (Tables S1 and S2). ARAD1 is the only glycosyltransferase gene that is known to be involved in the synthesis of pectic arabinan (Harholt et al., 2006), but its upregulation was not observed in the previous microarray analysis (Asahina et al., 2011). In contrast, the genes for pectin modification and degradation enzymes were upregulated, and hence pectin turnover might be involved in the tissue reunion process. Although no obvious changes were observed in the composition of hemicellulosic sugars in incised stems (Table S1), remodeling of the hemicellulose network with cellulose might occur in the tissue reunion. Actually, some of the genes for altering xyloglucan‐cellulose cross‐linking, including XTH20 and Expansin10, were noted in the previous microarray analysis (Asahina et al., 2011).

The function of XTHs is to split and reconnect xyloglucan molecules in plant cell walls, thereby affecting cell enlargement (Nishitani and Tominaga, 1992; Okazawa et al., 1993; Fry, 2004). Hyodo et al. (2003) suggested the involvement of XTH9 in cell elongation in the flowering stem. Rose et al. (2002) revealed the gene structure and phylogenetic tree of the Arabidopsis XTH family, showing that XTH20 and XTH19 are close homologs. Vissenberg et al. (2005) reported the differential expression of XTH17, XTH18, XTH19, and XTH20, which have highly conserved sequences in the promoter region. In our experiment, the xth20 single mutant did not show an inhibition of the reunion process (Figure 6f). Thus, XTH20 may act redundantly with other genes that regulate this process. Therefore, we made a double mutant of XTH20 and XTH19. xth20xth19 showed strong inhibition of the reunion process (Figure 6h) and a lower pith cell number (Figure 5i) without changes in CYCB1;1 expression (Figure S7). In addition, the length of primary roots, as well as the lengths of meristematic zones and epidermal cells, also decreased in xth20xth19 (Figures S8a,b and S9a–f) without changes in CYCB1;1 expression (Figure S10). These results indicated that XTH20 and XTH19 act redundantly to control cell proliferation in the reunion process as well as primary root elongation possibly by a mechanism that does not involve controlling the cell cycle.

During cytokinesis, the nascent cell wall produces a cell plate that fuses to the pre‐existing cell wall of the mother cell. This process involves the restructuring of xyloglucan cross‐links in the cellulose/xyloglucan framework, which could possibly include a role for XTH20 and XTH19. Yokoyama and Nishitani (2001) have also hypothesized a function for XTHs in cell plate formation during cytokinesis by showing the localization of XTH in the cell plates of BY‐2 cells. During cell plate formation, xyloglucan molecules that are produced and transported from the Golgi apparatus may be required as materials for cell plates and nascent cell walls (Moore and Staehelin, 1988). Future research will elucidate how XTHs control cell proliferation and/or cytokinesis by modulating the xyloglucan molecules and the cellulose/xyloglucan framework in the cell plate and/or cell wall.

Experimental procedures

Plant materials and growth conditions

The Arabidopsis T‐DNA insertion mutants anac071 (SALK_012841), xth20 (SALK_066758), and xth19 (GABI Kat 369E06) were obtained from the Arabidopsis Biological Resource Center at Ohio State University (Alonso et al., 2003). The homozygous lines were selected by PCR using specific primers (Table S3), and expression levels of the relevant genes were confirmed as compared with wild‐type plants (Figure S3a–c).

To generate the transgenic plants, the amplified fragments were cloned into pENTR/D‐TOPO vector (Invitrogen, http://www.lifetechnologies.com/th/en/home/brands/invitrogen.html; pENTR^™ Direction TOPO^® Cloning Kits) (Table S3). The inserts from entry clones were then integrated into binary vectors pKGWFS7 for the expression of GUS and pK2GW7 for overexpression using LR Clonase (Invitrogen), and the constructed plasmids were used for Agrobacterium‐mediated transformation of Arabidopsis (Col‐0) (see below). Transformants were selected by kanamycin, and the expression levels of the relevant genes were confirmed as compared with wild‐type plants.

Arabidopsis seed materials were vernalized in sterilized distilled water at 4°C for 3 days and then germinated on half‐strength Murashige and Skoog (MS) medium at 22°C under continuous light conditions (32 μmol m⁻² s⁻¹). After 3 weeks of germination, the plants were transferred to soil and grown under the same conditions until bolting.

To generate incised plants, the first internode of the inflorescence stem was incised to half of its diameter under a stereomicroscope using a microsurgery knife (Surgical Specialties Co., http://www.surgicalspecialties.com.au/) and grown for an additional 7 days under conditions as described (Asahina et al., 2011).

Microscopic analysis

Incised inflorescence stem samples were immersed overnight in fixative solution (2.5% [v/v] glutaraldehyde and 1% [w/v] paraformaldehyde in 0.1 m phosphate buffer, pH 6.8–7.4). After fixation, samples were washed twice in the same buffer, dehydrated with an ethanol series, and embedded in Technovit 7100 resin (Heraeus Kulzer, http://heraeus-kulzer.com/en/int/home_4/worldmap_1.aspx). Tissue sections (2.0 μm thick) were prepared using an ultra‐microtome glass knife (Reichert EM‐ULTRACUT, Leica, http://www.leica.com/). The sections were stained with 0.1% (w/v) toluidine blue O for 10 min. The tissue reunion region was observed with a light microscope (DMRB, Leica). Cell density of the sectioned sample 7 days after the incision was made was calculated with the method shown in Figure S11.

Decapitation and application of phytohormone

For experiments requiring decapitation, after each plant bolted, the distal region of the first internode was removed, and 10⁻³ m IAA or distilled water as a mixture with lanolin (1:1, v/v) was immediately applied. The remainder of the first internode was partially incised 24 h after application as described in Asahina et al. (2011).

Histochemical staining with 5‐bromo‐4‐chloro‐3‐indolyl‐β‐d‐clucuronide (X‐Glu)

Plant tissues were softened with 90% (v/v) acetone on ice for 20 min and then immersed in staining solution (100 mm phosphate buffer [pH 7], 10 mm EDTA, 1 mm K₃Fe(CN)₆, 1 mm K₄Fe(CN)₆·3H₂O, 0.1% [w/v] Triton X‐100, and 0.1 m X‐Glu) under vacuum for 30 min. The plant tissue samples were then incubated at 37°C for 8 h (Jefferson, 1987) and cleared with 70% (v/v) ethanol for approximately 12 h. Expression positions were observed using a stereomicroscope (Leica MCSLIII).

RNA preparation and quantitative RT‐PCR

Total RNA was extracted from Arabidopsis incised and non‐incised flowering stems with a RNAqueous RNA isolation kit with plant RNA Isolation aid (Ambion, http://www.lifetechnologies.com/th/en/home/brands/ambion.html). Complementary DNA was synthesized from total RNA (50 μg) by the QuantiTect Reverse Transcription Kit (Qiagen, http://www.qiagen.com/). Real‐time PCR using TaqMan technology or SYBR Green I reagents (Qiagen) was performed using cDNA as a template on a sequence detector system (model 7000; Applied Biosystems, http://www.appliedbiosystems.com/absite/us/en/home.html) with specific primers (Table S3). To normalize and confirm expression levels across various stages of incision treatment, ACT7 was used as an internal standard with three biological replications per sample.

EMSA

The coding region of ANAC071 was amplified by PCR using specific primers (EcoRI sense and SalI antisense; Table S3). The PCR product was digested with EcoRI and SalI and introduced into EcoRI‐ and SalI‐digested pCold ProS2 (TaKaRa, http://www.takara-bio.com/). The His‐Tag fusion protein was produced in E. coli (Origami (DE3); Novagen, http://www.merckmillipore.com/INTL/en/life-science-research/Eseb.qB.nekAAAE_lD53.L6J,nav?RedirectedFrom=http://www.novagen.com/) and purified using His‐Accept resin (Nacalai Tesque Inc., http://www.nacalai.co.jp/global/index.html). Oligonucleotides (see primer sequences in Figure 5c,d) were labeled with biotin using Biotin 3′ End DNA Labeling Kit (Thermo Fisher Scientific, http://www.thermofisher.com/en/home.html). To make double‐stranded biotin‐labeled probes, the complementary oligonucleotide pairs were annealed at 95°C for 5 min and slowly cooled to room temperature overnight. Double‐stranded unlabeled oligonucleotides were made using the same method. The labeled probes were incubated for 1 h at 28°C with 2 μg of purified His‐ANAC071 protein using the Light Shift Chemiluminescent EMSA Kit (Thermo Fisher Scientific) and were supplemented with 25 μg ml⁻¹ poly(dI‐dC), 0.1 μg ml⁻¹ BSA, 2 mm EDTA, and 1% (w/v) NP‐40. The protein‐probe mixtures were separated in 6% (w/v) polyacrylamide native gels in 0.5× TBE buffer and transferred to Biodyne membrane (PALL Gelman Laboratory, http://www.pall.com/main/home.page). Migration of the biotin‐labeled probes was visualized with the Chemiluminescent Nucleic Acid Kit (Thermo Fisher Scientific) and a detection module (LAS‐4000 mini biomolecular imager; Fujifilm, http://www.fujifilm.com/).

Agrobacterium‐mediated transient expression

Agrobacterium tumefaciens harboring binary vector pK2GW7 with or without insertion of the full‐length coding region of ANAC071 were transformed with pCAMBIA containing Pro_35S:Luciferase and selected on LB plates containing kanamycin and spectinomycin. Agrobacterium cells were suspended in the infiltration buffer (10 mm MgCl₂, 10 mm MES, and 150 mm acetosyringone) at 0.5 OD₆₀₀ and infiltrated into Pro_XTH20:GUS and Pro_XTH19:GUS transgenic plants by the leaf disc method. After 3 days of infiltration, the leaves were harvested and kept at −80°C until measurement.

Luciferase activity was measured by using a luciferase assay system (Promega, http://worldwide.promega.com/ E1500), and relative light unit (RLU) was measured with a single‐tube luminometer (MICROTEC, GL‐200). β‐Glucuronidase (GUS) activity was measured with p‐nitrophenyl‐β‐d‐glucuronide (PNPG) as a substrate. Briefly, cell lysates were suspended to 20 mm phosphate buffer (pH 7.4) and 10 mm β‐mercaptoethanol. PNPG was added to a final concentration of 1 mm, and the absorbance at 405 nm was continuously monitored.

Cell wall preparation and extraction of cell wall polysaccharides

Incised or non‐incised stem segments (4 mm in length) sampled 7 days after the incision was made were individually frozen in liquid nitrogen. The material was milled using rapid shaking twice for 30 sec in a bead mill (Retsch MM400 Retsch Co., Ltd., http://www.retsch.com/). The material was incubated in 96% ethanol for 30 min at 70°C to inactivate enzymes. Alcohol‐insoluble residues were prepared by washing the material sequentially in 100% ethanol, chloroform/methanol (1:1) twice, 65% ethanol, 80% ethanol, 100% ethanol, and finally in 100% acetone (all v/v), and air‐dried. Each step was followed by centrifugation. Starch was removed by digestion with α‐amylase from porcine pancreas (Type 1‐A, Sigma‐Aldrich, https://www.sigmaaldrich.com/sigma-aldrich/home.html; 50 U 100 mg⁻¹ cell wall materials, in 50 mm potassium phosphate buffer [pH 7.0]) overnight at 37°C with constant shaking as described by Zablackis et al. (1995). After treatment with α‐amylase, the walls were treated with endo‐polygalacturonase (Megazyme, http://www.megazyme.com/) as described by Ishii et al. (2001). The walls were then extracted with Na₂CO₃ as described by Selvendran and O'Neill (1987). Na₂CO₃‐extracted residues were treated sequentially with 1 m KOH and 4 m KOH to solubilize hemicelluloses (Selvendran and O'Neill, 1987).

Analytical methods

Total sugar and uronic acid were determined by the phenol‐sulfuric acid method (Dubois et al., 1956) with glucose as the standard and by the carbazole method (Galambos, 1967) with galacturonic acid as the standard, respectively. Neutral glycosyl‐residue compositions were determined by GLC of their alditol acetate derivatives (York et al., 1985). Glycosyl‐linkage compositions were determined by a modified Hakomori procedure (York et al., 1985). Briefly, methylated polysaccharides were carboxyl reduced with Superdeuteride (1 m lithium triethylborodeuteride in tetrahydrofuran, Aldrich) as described by York et al. (1985). The glycosyl‐linkage composition was then determined by GC‐MS of the partially methylated, partially acetylated alditol acetate derivatives (York et al., 1985).

Acknowledgements

Special thanks to Dr. Nobutaka Mitsuda and Dr. Masaru Ohme‐Takagi of the National Institute of Advanced Industrial Science and Technology for providing ANAC071‐SRDX transgenic Arabidopsis and Mr. Takashi Yamazaki, Mr. Katsuya Azuma, and Mrs. Rieko Nozawa for technical assistance. This work was supported in part by a Grant‐in‐Aid for Scientific Research on Priority Areas (21027004) and on Innovative Areas (24114006) to S.S.