The Plant Journal (2019) 100, 661–676 doi: 10.1111/tpj.14479 Craterostigma plantagineum cell wall composition is remodelled during desiccation and the glycine-rich protein CpGRP1 interacts with pectins through clustered arginines Niklas U. Jung1, Valentino Giarola1, Peilei Chen1, John Paul Knox2 and Dorothea Bartels1,* 1 Institute of Molecular Physiology and Biotechnology of Plants (IMBIO), Faculty of Natural Sciences, University of Bonn, Kirschallee 1, Bonn D-53115, Germany, and 2 Centre for Plant Sciences, Faculty of Biological Sciences, University of Leeds, Leeds LS2 9JT, UK Received 21 March 2019; revised 27 June 2019; accepted 23 July 2019; published online 27 July 2019. *For correspondence (e-mail dbartels@uni-bonn.de). SUMMARY Craterostigma plantagineum belongs to the desiccation-tolerant angiosperm plants. Upon dehydration, leaves fold and the cells shrink which is reversed during rehydration. To understand this process changes in cell wall pectin composition, and the role of the apoplastic glycine-rich protein 1 (CpGRP1) were analysed. Cellular microstructural changes in hydrated, desiccated and rehydrated leaf sections were analysed using scanning electron microscopy. Pectin composition in different cell wall fractions was analysed with monoclonal antibodies against homogalacturonan, rhamnogalacturonan I, rhamnogalacturonan II and hemicellulose epitopes. Our data demonstrate changes in pectin composition during dehydration/rehydration which is suggested to affect cell wall properties. Homogalacturonan was less methylesterified upon desiccation and changes were also demonstrated in the detection of rhamnogalacturonan I, rhamnogalacturonan II and hemicelluloses. CpGRP1 seems to have a central role in cell adaptations to water deficit, as it interacts with pectin through a cluster of arginine residues and de-methylesterified pectin presents more binding sites for the protein
pectin interaction than to pectin from hydrated leaves. CpGRP1 can also bind phosphatidic acid (PA) and cardiolipin. The binding of CpGRP1 to pectin appears to be dependent on the pectin methylesterification status and it has a higher affinity to pectin than its binding partner CpWAK1. It is hypothesised that changes in pectin composition are sensed by the CpGRP1
CpWAK1 complex therefore leading to the activation of dehydration-related responses and leaf folding. PA might participate in the modulation of CpGRP1 activity. Keywords: desiccation tolerance, resurrection plant, Craterostigma plantagineum, cell wall, glycine-rich protein, pectin, leaf folding, lipid binding, phosphatidic acid. Linked article: This paper is the subject of a Research Highlight article. To view this Research Highlight article visit https://doi.org/10.1111/tpj.14582. INTRODUCTION Most land plants are not able to survive prolonged periods of water shortage and even a mild water deficit can lead to irreversible damage and plant death (Zhang and Bartels, 2018). Water plays a crucial role in the maintenance of cell turgor, transport of solutes and nutrients and it mediates hydrophobic and hydrophilic interactions essential for macromolecular structures. Plants, as sessile organisms, face dehydration periods and have different strategies to reduce water loss and adapt to low water availability (Verslues and Juenger, 2011). Desiccation tolerance is rare in vegetative tissues of vascular plants, but it is common in seeds and pollen. Only approximately 300 species of vascular plants have been reported to survive extensive dehydration during their vegetative growth phase. These plants are generally called resurrection plants (Porembski, 2000). Most angiosperm resurrection plants are native to central and southern Africa, Australia, south America and India (Gaff, 1971; Gaff and Bole, 1986). The resurrection plant Craterostigma plantagineum (Cp) has been extensively studied (Giarola et al., 2017). Lindernia brevidens (Lb) and Lindernia subracemosa (Ls) are closely related to C. plantagineum (Rahmanzadeh et al., 2005) and their genome sequences have recently been deciphered (VanBuren et al., 2018). The interest for © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. 661 662 Niklas U. Jung et al. comparative studies of these plants are different degrees in desiccation tolerance. L. brevidens is desiccation-tolerant like C. plantagineum (Phillips et al., 2008) but L. subracemosa is desiccation-sensitive (Seine et al., 1995). Cell walls are the outermost structures of plant cells and essential to maintain cell and organ integrity and functionality. Mechanical stress is built up when the cell wall folds up during dehydration. When the vacuole shrinks and the cell contents are drawn inwards, tension builds up between the plasma membrane and the rigid cell wall (Levitt, 1987). The protection of the plasma membrane during dehydration is essential for the plant to survive. Iljin (1957) proposed that any plant could tolerate desiccation if the mechanical stress can be limited. In desiccation-sensitive species the cell wall loses its integrity upon desiccation. When resurrection plants encounter desiccation the leaves start to fold and the cell volume is reduced (Farrant, 2000; Farrant et al., 2003; Willigen et al., 2003). Previous studies showed dehydration-induced changes in cell wall architecture, cell wall composition and variations of hemicellulose polysaccharides and pectin-associated arabinans in resurrection plants (Vicre
et al., 1999, 2004; Moore et al., 2006, 2008). It has been proposed that high levels of pectic-arabinans, arabinogalactan-proteins and arabinoxylans ensure the required cell wall plasticity upon dehydration for the resurrection plant Myrothamnus flabellifolia (Moore et al., 2013). Changes in xyloglucan and modifications of pectin structures have been reported for Craterostigma wilmsii (Vicre
et al., 1999, 2004). Glycine-rich proteins (GRPs) have a high-glycine-content with glycine residues arranged in (Gly)n-X repeats. In addition, a cysteine-rich region, an oleosin domain, RNA-recognition motifs, a cold-shock domain or zinc-finger motifs are found in GRPs. Although several GRPs have been characterised, the function of the glycine-rich domains in these proteins is poorly understood (Czolpinska and Rurek, 2018). GRPs are classified according to the arrangement of the glycine-rich repeats and the presence of additional domains (Sachetto-Martins et al., 2000; Fusaro et al., 2001; Bocca et al., 2005; Mangeon et al., 2010). Class I GRPs share a high-glycine-content region with (GGX)n repeats. A C-terminal cysteine-rich region is present in the class II GRPs. The class III GRPs have a lower glycine content and may have an additional oleosin domain. Class IV GRPs are known as RNA-binding GRPs with either an RNA-recognition motif or a cold-shock domain and in some cases GRPs have additional zinc-finger motifs. Class V GRPs are similar to class I GRPs but show mixed patterns of glycine repeats. The expression patterns and the subcellular localisation of the different proteins within the GRP-superfamily are highly diverse thus suggesting that these proteins have different functions (Mangeon et al., 2010). GRPs with an apoplastic signal peptide have been proposed to be an important component of cell wall structures (Condit and Meagher, 1986, 1987; Keller et al., 1988). For example, the French bean PvGRP1.8 protein is part of the cell wall and plays a role in the protoxylem repair system (Ringli et al., 2001). According to microarray results GRPs could also be implicated in maintaining protoxylem structures (Yokoyama and Nishitani, 2006). Glycine-rich proteins have been proposed to connect the secondary cell wall thickenings between protoxylem elements (Ryser et al., 2004). The AtGRP9 protein from Arabidopsis thaliana interacts with a cinnamyl alcohol dehydrogenase (AtCAD5) and may be involved in lignin biosynthesis (Chen et al., 2007). Besides class I GRPs also GRPs from other classes are plant cell wall components. The class V glycine-rich protein 1 (BhGRP1) from Boea hygrometrica was proposed to be important for cell wall integrity during dehydration, whereas the class II GRP NtCIG1 protein from tobacco was proposed to enhance callose deposition in cell walls (Ueki and Citovsky, 2002; Wang et al., 2009). The C. plantagineum glycine-rich protein 1 (CpGRP1) belongs to class II GRPs. CpGRP1 is highly abundant in the apoplast of desiccated leaves and interacts with the C. plantagineum cell wall-associated protein kinase 1 (CpWAK1) (Giarola et al., 2016). WAKs contain an extracellular pectin binding domain and an intracellular serine/ threonine protein kinase domain. WAKs have been suggested to link the cytoplasm to the extracellular matrix and to activate signalling pathways in response to pectin changes (He et al., 1999; Anderson et al., 2001; Kohorn and Kohorn, 2012). The CpGRP1
CpWAK1 complex may play a role in sensing dehydration-induced cell wall changes and thus activate dehydration-induced signalling pathways (Giarola et al., 2016). A similar complex is known from A. thaliana where the cysteine-rich region of AtGRP3 interacts with the cell wall-associated kinase 1 (AtWAK1). This complex has been proposed to be involved in pathogen defence mechanisms (Park et al., 2001). Phosphatidic acid (PA) belongs to the group of phospholipids and consists of a glycerol backbone, a saturated fatty acid, an unsaturated fatty acid and a phosphate group and has different roles in plants (Hou et al., 2016). Phospholipase C and phospholipase D are involved in the synthesis of PA from phosphatidylinositol-(4,5)-bisphosphate (PIP2), phosphatidylcholine and phosphatidyl-ethanolamine. These reactions are crucial for the maintenance of the PA pool, which serves as a source for the biosynthesis of other phospholipids (Munnik, 2001; Ufer et al., 2017). PA can also function as signalling molecule in response to environmental cues. PA was shown to bind to other proteins and thus modify the activity of proteins (Hou et al., 2016). Changes in the C. plantagineum cell wall during a desiccation/rehydration cycle were investigated with a set of monoclonal antibodies that recognise different epitopes of homogalacturonan (HG), rhamnogalacturonan I (RG-I), rhamnogalacturonan II (RG-II) and hemicelluloses. Our © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2019), 100, 661–676 Modifications of pectin in response to desiccation 663 findings suggest that C. plantagineum cell walls undergo reversible remodelling of pectic polymers upon desiccation. A cluster of arginine residues within the CpGRP1 protein enables its binding to pectin and PA. HG methylesterification is reduced upon dehydration thus promoting CpGRP1 binding to pectin. Changes in the pectin organisation provide plasticity to the cell wall which is required for the extensive folding and expansion of cell walls during desiccation/rehydration. RESULTS Morphological characterisation of leaf structures Microscopic views of the C. plantagineum leaf surfaces and transverse sections during a desiccation/rehydration cycle are shown in Figure 1(a,b). Untreated and rehydrated leaf tissues are almost identical suggesting that leaves can fully recover from desiccation-induced cellular changes after 48 h of rehydration. In contrast to the hydrated and rehydrated samples, the epidermis of the desiccated leaf is extensively folded. In this compact structure leaf glands are trapped in the epidermal folds. Leaf folding mainly occurs during late dehydration when the relative water content is below 60%. Cell walls in desiccated tissues were slightly thicker than cell walls in untreated and 48 h-rehydrated tissues as shown by cell wall staining (Figure 1c,d). These findings suggest that changes in pectin composition between the different samples might be involved in cell wall adaptations to water-stress conditions. Pectin and hemicellulose profiles determined in a desiccation/rehydration cycle Panels of monoclonal antibodies allow the monitoring of changes in cell wall polysaccharides (Table 1). In enzymelinked immunosorbent assays (ELISAs), JIM5, JIM7, LM20 and LM19 were used to analyse differences in the methylesterification pattern of C. plantagineum HG fractions upon desiccation and rehydration (Table 2). The abundance of the pectin and hemicellulose epitopes is correlated with colour intensity. JIM5, JIM7 and LM20 displayed a stronger binding to the untreated and rehydrated (1 and 2) samples than to the desiccated sample. LM19, which detects fully de-methylesterified HG, displayed a stronger signal to desiccated samples than to untreated samples. These results indicate changes in the HG methylesterification status during the desiccation/rehydration cycle and suggest homogalacturonan synthesis in the recovery process. Changes in the abundance of RG-II were analysed using the 42-6 antibody which binds RG-II monomers, crosslinked RGII and an unknown pectic component (Table 1). The antibody bound more strongly to the untreated and rehydrated samples than to the desiccated samples (Table 2). LM25, LM15, LM11, LM6 and LM5 monoclonal antibodies were used to analyse changes in the hemicelluloses and the rhamnogalacturonan I (Tables 1 and 2). LM25 and LM15 indicate strong signals for xyloglucan in desiccated leaves which is reversed during rehydration. LM11, which detects xylan, bound stronger to the desiccated samples, whereas LM6, which detects (1?5)-a-arabinan, displayed a slightly weaker binding to pectins of desiccated leaves than to pectins of untreated leaves. No significant changes were observed in the binding of the LM5 antibody which detects (1?4)-b-galactan. Results obtained with RG-I, RG-II and hemicellulose antibodies suggest changes in the C. plantagineum cell wall architecture during the desiccation/rehydration cycle. Interaction between CpGRP1 and pectin The glycine-rich cell wall protein CpGRP1 interacts with the CpWAK1 kinase protein in the apoplast and it has been hypothesised that CpGRP1 binds to pectin (Giarola et al., 2016). The glycine-rich domain of CpGRP1 contains a positively charged arginine cluster through which it may interact with cell wall polysaccharides, such as pectins (Figure 2a,b). Three different approaches have been used to analyse a possible interaction of CpGRP1 and pectins: Blue-native page gel-shift assays, ELISA and dot-blot assays using either commercial pectin (from citrus peel; Sigma-Aldrich, USA) or the 1,2-cyclohexanediaminetetraacetic acid (CDTA)-pectin fraction isolated from C. plantagineum or closely related species. For binding experiments the full-length recombinant CpGRP1 protein or fragments of the CpGRP1 protein corresponding to Nor C-terminal domains or to the mutated N-terminal domain (two arginines mutated to glycine, a352g_c358g) were used (Figure 2). When the CpGRP1 full-length protein was incubated with commercial pectin and calcium, the electrophoretic mobility was retarded compared to the CpGRP1 protein in a native page (Figure 3). A similar mobility shift was observed for the N-terminal fragment, but not for the C-terminal fragment. The mutated N-terminal fragment did also show a mobility shift in the presence of pectin and calcium but in contrast to the non-mutated fragment, two protein bands were detected. No mobility shift for any of the proteins was observed in the absence of calcium. The interaction between CpGRP1 and pectins was further investigated using pectins extracted from C. plantagineum leaves (Figure 4a). Microtiter plate wells were coated with CDTA-soluble pectins or KOH-soluble cell wall fractions and then incubated with the CpGRP1 fulllength protein, the N-terminal or the mutated N-terminal polypeptide. No complex was observed between CpGRP1 and the KOH cell wall fractions. The strongest interaction was observed with the N-terminal polypeptide, followed by the mutated N-terminal polypeptide and the full-length protein. Each of the tested proteins showed the highest value for the CDTA-pectin fraction isolated from desiccated C. plantagineum leaves. No differences were observed in © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2019), 100, 661–676 664 Niklas U. Jung et al. Figure 1. Morphological characterisation of Craterostigma plantagineum leaf structures using scanning electron microscopy. Surface images (a), transverse section images (b) and images of cell walls stained with silver nitrate (c) are shown. Micrographs were taken in a desiccation/rehydration cycle. From left to right: Untreated (RWC = 100%), Partially dehydrated (RWC = 50%), Desiccated (RWC = 2%), 24 h after rehydration (Rehydrated 1) and 48 h after rehydration (Rehydrated 2). Representative micrographs are shown. Three different samples were used for the evaluation. (d) Thickness of C. plantagineum cell walls in a desiccation/rehydration cycle. Images from (c) were used for the evaluation. Cell wall thickness increases slightly during dehydration and the starting point (untreated samples) is restored after 48 h of rehydration. U = Untreated, PD = Partially dehydrated, D = Desiccated, R 1 = Rehydrated 1, R 2 = Rehydrated 2. The asterisk indicates the levels of significance in comparison with the untreated sample (one-way ANOVA, Holm
Sidak method): *P < 0.05. the presence or absence of calcium. In addition, pectin fractions from the two closely related species differing in desiccation tolerance, L. brevidens and L. subracemosa, were prepared and the CpGRP1 binding to those fractions was analysed (Figure 4b). The CpGRP1 full-length protein and the N-terminal polypeptide showed the strongest interaction with C. plantagineum pectin, followed by pectin from L. brevidens and L. subracemosa. To confirm that these differences are due to the pectin composition and not due to the amount of isolated HG the galacturonic acid (GA) content was determined for all different fractions. No significant differences in the GA content were detected which supports comparable HG contents (Table 3). In a second step the JIM5, JIM7, LM20 and LM19 antibodies were used to analyse the L. brevidens and L. subracemosa HG fractions. The methylesterification profile of HG for the desiccation-tolerant plant L. brevidens was similar to the one of C. plantagineum (Table 4). For L. subracemosa the signal intensity detected with the antibodies in the rehydrated samples was lower than the signal in the untreated sample suggesting that this plant is not able to restore cell wall architecture during rehydration which is consistent with the phenotype (Table 4). To confirm the ELISA analyses a series of dot-blot experiments was carried out. The interaction of the CpGRP1 protein and pectin was either detected using a 6x His-tag © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2019), 100, 661–676 Modifications of pectin in response to desiccation 665 Table 1 Monoclonal antibodies used in this study for pectin and hemicellulose characterization HGa RG-IIb RG-Ib Hemicelluloses Antibody Specificity Reference JIM5 JIM7 LM20 LM19 42-6 LM6 LM5 LM25 LM15 LM11 Partially or de-methylesterified HG Partially methylesterified HG Methylesterified HG Fully de-methylesterified HG B-RG-II, RG-II monomers, unknown pectic fragment (1?5)-a-arabinan (1?4)-b-galactan Xyloglucan (XXLG, XLLG) Xyloglucan (XXXG) (1?4)-b-xylan Knox et al. (1990) Knox et al. (1990) Verhertbruggen et al. (2009) Verhertbruggen et al. (2009) Zhou et al. (2018) Willats et al. (1998) Jones et al. (1997) Pedersen et al. (2012) Marcus et al. (2008) McCartney et al. (2005) a Homogalacturonan. Rhamnogalacturonan. b antibody or the JIM5 antibody (Figure 5a). The dot-blot analyses confirmed the interaction between the CpGRP1 full-length protein and the two N-terminal protein fragments with pectins. BSA (Carl Roth, Karlsruhe, Art.-Nr. 8076.2) and the LEA-like 11-24His recombinant protein (Petersen et al., 2012) were used as negative controls and did not show any interaction. Figure 5(b) presents the species-specific interactions between the CpGRP1 full-length protein, the N-terminal fragment and the mutated N-terminal fragment with the CDTA fractions of C. plantagineum, L. brevidens and L. subracemosa. Mutation of arginines in the N-terminal fragment led again to weaker interactions with pectins. Quantification of signal intensities of the dot blots fully supports the ELISA results (Figure 5c). The binding of CpGRP1 and CpGRP1 protein fragments to pectins from the desiccated samples was stronger than the binding to pectins from the untreated or rehydrated samples in all experiments. The highest signal intensity was obtained for pectin isolated from C. plantagineum. To investigate and compare the CpGRP1
pectin interaction, the pectin binding of two other apoplastic proteins, CpWAK1 and the C. plantagineum germin-like protein 1 (CpGLP1), was investigated. Figure 6 demonstrates that the CpGRP1 binds much stronger to pectin than CpWAK1 or CpGLP1. Pre-treatment of pectin with CAPS-buffered solutions (pH 7–11) or with 0.1 M sodium carbonate reduced the extent of HG methylesterification. CpGRP1 full-length protein bound more strongly to de-methylesterified than to methylesterified pectin from C. plantagineum (Figure 7a). Table 2 Analysis of changes in the cell wall composition of Craterostigma plantagineum leaves in a desiccation/rehydration cycle 0.00 0.50 1.00 1.50 2.00 2.50 JIM5, JIM7 and LM20 detect varying levels of methylesterification in HG. LM19 detects fully de-methylesterified HG. The 42-6 antibody detects RG-II, crosslinked RG-II and an unknown pectic epitope. LM6 and LM5 detect pectic arabinan and pectic galactan, epitopes present in the RG-I, respectively. LM25 and LM15 detect xyloglucan and LM11 detects xylan. For specificity of different antibodies see Table 1. The colour scale in relation to absorbance values is shown bottom left. Results for HG were generated analysing the CDTA fractions in 1:5 dilutions and results for RG-I and RG-II were generated analysing the KOH fractions in 1:5 dilutions. Values shown are means of three biological replicates  SD. The letters indicate the levels of significance (one-way ANOVA, Holm
Sidak method): a,b,c,d P < 0.05; ABCD P < 0.01; aA = significantly different from ‘Untreated’ sample, bB = significantly different from ‘Desiccated’ sample, cC = significantly different from ‘Rehydrated 1’ (24 h) sample, dD = significantly different from ‘Rehydrated 2’ (48 h) sample. a Homogalacturonan. b Rhamnogalacturonan. © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2019), 100, 661–676 666 Niklas U. Jung et al. We hypothesize that the de-methylesterification of HG provides more binding sites for the CpGRP1 protein. Highly methylesterified commercial apple pectin (Apple Pectin Powder; Solgar, USA) was used to confirm the effectiveness of the de-methylesterification procedure as indicated by the specificities of the LM19 and LM20 probes. Figure 7(b) shows that untreated apple pectin is recognised by the LM20 antibody but not by LM19, whereas the sodium carbonate-treated and hence de-methylesterified apple pectin is recognised by the LM19 antibody and not by LM20. As lipids are a major component of cell membranes and are modified during desiccation/rehydration in C. plantagineum (Gasulla et al., 2013) it was tested whether CpGRP1 may also bind to lipids. CpGRP1 binding to different lipids was analysed in protein-lipid overlay assays (Figure 8a). The LEA-like 11–24 protein was used as a positive control and it was used with or without an additional Histag to account for possible His-tag background binding (Petersen et al., 2012). The CpGRP1 full-length recombinant protein showed strong binding to PA and cardiolipin (CL), but not to any other lipid tested (Figure 8b). The Nterminal fragment of CpGRP1 and the full-length CpGRP1 showed similar binding to PA, whereas the binding of the C-terminal fragment of CpGRP1 to PA was very weak. These results indicate that the binding of CpGRP1 to PA is mainly mediated by the N-terminal region of the protein. Mutations of two arginines in this region led to weaker binding, suggesting the involvement of positively charged arginines in the CpGRP1
PA interaction. DISCUSSION Pectin fractions are remodelled upon dehydration Folding of cell walls and changing of texture and chemical composition is an adaptation to desiccation in resurrection plants (Bartels and Hussain, 2011). Intensive cell wall folding of desiccated cells is also in place in C. plantagineum as shown by scanning electron microscopy (SEM) (Figure 1). Similar observations have been made for L. brevidens and L. subracemosa, which are closely related to C. plantagineum and C. wilmsii (Farrant, 2000; Phillips et al., 2008). Although the leaves of the desiccation-sensitive plant L. subracemosa are folded upon dehydration, they became brown and necrotic and did not recover after rewatering, which points to alterations in the composition of the cell wall of C. plantagineum. To address this question pectins and hemicelluloses associated with cell wall flexibility were analysed. C. plantagineum showed a lower degree of methylesterification in HG upon dehydration, which is reversed during rehydration (Table 2). L. brevidens did show a similar methylesterification status as C. plantagineum, which is in agreement with the fact that this plant is also desiccation tolerant. In contrast, no significant changes in the HG fraction were detected for the desiccation-sensitive plant L. subracemosa upon rehydration (Table 4). The degree of methylesterification may be of importance for desiccation tolerance, at least among the Linderniaceae. HG is thought to be synthesised in the Golgi apparatus in a highly methylesterified form and is de-methylesterified in the cell Figure 2. (a) Craterostigma plantagineum CpGRP1 amino acid sequence and protein domains. The different protein domains of CpGRP1 are indicated by coloured boxes above the sequence. Red: predicted signal peptide; blue: Semi-repetitive glycine-rich region; the two different amino acid tandem repeat motifs are shown with solid or dashed arrows respectively. Yellow: Cysteine-rich region; the six cysteine residues are marked by black triangles. Different protein fragments are marked by dotted lines. Additionally, the full-length CpGRP1 recombinant protein and the other polypeptides carried six histidine residues. (b) Electrostatic surface model of the recombinant CpGRP1 protein. Clustered arginines in the glycine-rich domain of the protein build up a negative cluster. Arginines at positions 118 and 120 were mutated to glycine (red boxes). The image in (a) was modified from Giarola et al. (2016). © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2019), 100, 661–676 Modifications of pectin in response to desiccation 667 Figure 3. Analysis of CpGRP1
pectin interaction by Blue-native page gel-shift assay. The full-length recombinant CpGRP1 protein, a CpGRP1 N-terminal protein fragment and the mutated N-terminal protein fragment were used for this analysis. The proteins were detected with CpGRP1 or His-tag antibodies and SDS-gels were stained using Coomassie brilliant blue. Figure 4. Quantification of the protein
pectin interaction using the CpGRP1 full-length protein, the N-term. fragment and the mutated N-term. fragment and pectin isolated from (a) Craterostigma plantagineum, (b) Lindernia brevidens and Lindernia subracemosa during dehydration and rehydration. All signals shown were detected using 1:5 dilutions of CDTA fractions. No signals were detected in KOH fractions. The LEA-like 11–24 protein was incubated with commercial pectin and 2 mM Ca2+ and used as a negative control (Neg.). The CpGRP1 full-length protein was incubated with commercial pectin and Ca2+ and used as a positive control (Pos.). U = Untreated, D = Desiccated, R 1 = Rehydrated 1 (24 h), R 2 = Rehydrated 2 (48 h). Values shown are means of three biological replicates  SD. The star indicates the levels of significance in comparison with the untreated sample (one-way ANOVA, Holm
Sidak method): *P < 0.05; **P < 0.01. © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2019), 100, 661–676 668 Niklas U. Jung et al. Table 3 Quantification of galacturonic acid content in the different CDTA-pectin fractions in lg ll
1 Craterostigma plantagineum Lindernia brevidens Lindernia subracemosa Untreated Desiccated Rehydrated 1 Rehydrated 2 2.01  0.31 1.81  0.24 2.05  0.25 2.12  0.22 1.55  0.26 1.92  0.21 2.00  0.24 1.58  0.27 1.65  0.27* 1.62  0.41 1.60  0.21 2.02  0.20 Values shown are means of three biological replicates  standard deviation (SD). The star indicates the levels of significance in comparison to the untreated sample (one-way ANOVA, Holm
Sidak method): *P < 0.05. Table 4 Analysis of changes in cell wall methylesterification of Lindernia brevidens (Lb) and Lindernia subracemosa (Ls) leaves in a desiccation/rehydration cycle 0.00 0.50 1.00 1.50 2.00 2.50 JIM5, JIM7 and LM20 detect varying levels of methylesterification in HG. LM19 detects fully de-methylesterified HG. For specificity of different antibodies see Table 1. The colour scale in relation to absorbance values is shown bottom left. Results were generated analysing the CDTA fractions in 1:5 dilutions. Values shown are means of three biological replicates  standard deviation (SD). The letters indicate the levels of significance (one-way ANOVA, Holm
Sidak method): a,b,c,d P < 0.05; ABCD P < 0.01; aA = significantly different from ‘Untreated’ sample, bB = significantly different from ‘Desiccated’ sample, cC = significantly different from ‘Rehydrated 1’ (24 h) sample, dD = significantly different from ‘Rehydrated 2’ (48 h) sample. a Homogalacturonan. wall (Zhang and Staehelin, 1992; Staehelin and Moore, 1995; Sterling, 2001). Thus, the de-methylesterification of HG in the cell wall is a one-way process and there is no evidence of methylesterification of pectin in cell walls. The increasing binding of antibodies detecting methylesterified HG in rehydrated samples suggests that pectin is synthesised de novo in C. plantagineum and L. brevidens during the recovery process (Tables 2 and 4). Pectin de-methylesterification is catalysed by a large enzyme family of pectin methylesterases (PMEs) (Fleischer et al., 1999; Le Gall et al., 2015). Little information is known about the activities of PMEs in C. plantagineum. PMEs are thought to be crucial in influencing cell wall properties and to be important for the modulation of apoplastic Ca2+ concentrations (Micheli, 2001; Wu and Jinn, 2010; Wu et al., 2010, 2018). Ca2+ mediates the interaction between de-methylesterified pectin chains leading to the formation of “egg-box” pectin structures (Grant et al., 1973; Jarvis, 1984; Moore et al., 1986; Lloyd, 1991). An increase in apoplastic Ca2+ was observed in C. wilmsii upon dehydration suggesting enhanced cell wall strength in Craterostigma upon dehydration (Vicre
et al., 1999). RG-II is crosslinked by borate (Kobayashi et al., 1996). The interaction between RG-II, borate and HG is important for the physical and biochemical properties of the cell wall (O’Neill et al., 2001). The antibody 42-6 detects RG-II monomers, borate-crosslinked RG-II and an unknown pectic component (Zhou et al., 2018). We found significant changes in RG-II during desiccation/rehydration (Table 2). The decrease in the 42-6 signals in response to dehydration suggests pectin remodelling which could affect cell wall rigidity. Xyloglucan connects cellulose fibrils and contributes to cell wall extensibility (Moore et al., 1986; Fry, 1989). Higher abundance of xyloglucan and de-methylesterified pectins are both known to strengthen the cell wall. This fits well with the increase in xyloglucan and de-methylesterified pectins in desiccated C. plantagineum leaves (Table 2) and C. wilmsii (Vicre
et al., 1999). Increased detection of xylan, mostly present in secondary cell walls, was observed upon dehydration in C. plantagineum. Higher xylan contents may contribute to cell wall strength, as xylan together with xyloglucan connects cellulose fibrils. A. thaliana plants with reduced xylan contents have weakened cell walls © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2019), 100, 661–676 Modifications of pectin in response to desiccation 669 Figure 5. Evaluation of CpGRP1 interaction with Craterostigma plantagineum, Lindernia brevidens and Lindernia subracemosa pectin fractions during dehydration and rehydration by dot-blot analyses. (a) Binding of CpGRP1 full-length, N-terminal fragment and mutated N-terminal fragment to commercial pectin. BSA and the LEA-like 11-24 protein were used as negative controls. In the left-hand column the protein was spotted on the membrane first and then incubated with pectin, the right-hand column represents the opposite experiment, the pectin was spotted first and then the protein was added. (b) Interaction of CpGRP1 fulllength, N-terminal fragment. and mutated N-terminal fragment with CDTA fractions. From each fraction 1 lg of pectin was spotted. Quantification of signals from (b) is shown in (c). U = Untreated, PD = Partially dehydrated, D = Desiccated, R 2 = Rehydrated 2 (48 h), Cp = Craterostigma plantagineum, Lb = Lindernia brevidens, Ls = Lindernia subracemosa. Values shown are means of three technical replicates  SD. (Brown et al., 2007; Wu et al., 2009). (1–>4)-b-galactan, the most flexible component of the cell wall, decreases the ability of pectin molecules to crosslink and also (1–>5)-aarabinan side chains are motile (Ha et al., 1996; Jones et al., 1997). These components do not change or even decrease upon dehydration in C. plantagineum (Table 2). Our results demonstrate major changes in cell wall components. The flexible cell wall components are not changing or are even decreasing upon drying. Conversely, components associated with stiffer and stronger cell walls increase. These findings are in agreement with the results published by Vicre
et al. (1999) but are different to the results by Jones and McQueen-Mason (2004). They found an increase in cell wall flexibility and a higher activity of expansins. We propose that the C. plantagineum cell wall folds in a distinct way and this process is tightly influenced by different factors. Some connections are strengthened, while more flexibility is added to others. Different mechanisms need to be in place to protect the cells from being damaged by the mechanical tension during the dehydration process. CpGRP1 binds to de-methylesterified pectin through clustered arginines Glycine-rich proteins have been connected to cell wall properties of plants (Condit and Meagher, 1986, 1987; © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2019), 100, 661–676 670 Niklas U. Jung et al. Figure 6. Quantification of protein
pectin interaction using the CpGRP1 full-length recombinant protein, the CpWAK1 extracellular recombinant protein fragment and the CpGLP1 full-length recombinant protein together with CDTA-pectin extracts from C. plantagineum leaves. Values shown are means of three biological replicates  SD. The asterisk indicates the levels of significance in comparison to the control sample (one-way ANOVA, Holm
Sidak method): **P < 0.01. Keller et al., 1988). In C. plantagineum the class II glycinerich protein CpGRP1 accumulates in the apoplast of dehydrated leaf tissues (Giarola et al., 2015). The glycine-rich protein 1 from the resurrection plant Boea hygrometrica (BhGRP1) was proposed to be correlated with cell wall flexibility (Wang et al., 2009). The A. thaliana glycine-rich protein 3 (AtGRP-3), structurally similar to CpGRP1, interacts with a cell wall-associated protein kinase 1 (AtWAK1). This interaction is involved in plant-pathogen defence mechanisms and requires the C-terminal cysteine-rich domain (Park et al., 2001). Giarola et al. (2016) showed that CpGRP1 binds to a WAK protein of C. plantagineum, CpWAK1, and proposed that the CpGRP1-CpWAK1 complex could be involved in dehydration-induced mechanisms. WAKs can bind to cell wall pectins (Kohorn and Kohorn, 2012) but a GRP
pectin interaction has not been characterised so far. Here we showed that the CpGRP1 protein binds to pectins which requires the glycine-rich domain of CpGRP1 (Figure 3). The substitution of two arginines within the glycine-rich domain by glycines was sufficient to reduce the protein
pectin interaction (Figures 3, 4 and 5). The involvement of an arginine cluster in pectin binding was also shown for Phaseolus vulgaris polygalacturonase inhibitor proteins (Spadoni et al., 2006). The glycine-rich domain of the CpGRP1 protein is important for pectin binding. The repetitive character of this Figure 7. Quantification of CpGRP1
pectin interaction after pectin demethylesterification. (a) ELISA to quantify GRP1
pectin interaction after pre-treatment of coated homogalacturonan with CAPS-buffers (pH 7
11) or with sodium carbonate. (b) Binding of LM19 and LM20 to different concentrations of apple pectin with and without sodium carbonate pre-treatment. Values shown are means of three biological replicates  SD. The asterisk indicates the levels of significance in comparison with the untreated sample (one-way ANOVA, Holm
Sidak method): **P < 0.01. domain leads to multiple arginines with a distinct spacing. This spacing could be important for the spatial organisation of a positive amino acid cluster, which plays a crucial role in pectin binding. Similarly, the binding to PA seems to be mediated through the N-terminal domain of the protein and the arginine cluster. Multiple binding capacities of CpGRP1 suggests that this protein may coordinate different ligands involved in cell wall/membrane modifications during the folding process. The role of Ca2+ in the protein
pectin binding is not clear. The gel-shift assay suggests Ca2+ to be important for the interaction but the ELISAs showed that Ca2+ was not essential for the interaction between CpGRP1 and pectin (Figures 3 and 4). One explanation could be that gel shift assays are less sensitive than ELISAs and therefore in gel © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2019), 100, 661–676 Modifications of pectin in response to desiccation 671 Figure 8. Protein
lipid overlay assays using the recombinant CpGRP1 full-length protein, the N. term. polypeptide, the mutated N-term. polypeptide and the Cterm. polypeptide. Anti-His antibody was used to detect the different protein fragments. The LEA-like 11-24 protein (with and without His-tag) served as a positive control and was detected with an antibody specific to LEA-like 11-24. (a) Investigation of CpGRP1 binding properties to different lipids: phospatidylcholine (PC), phosphatidic acid (PA), diacylglycerol 18:3 (di18:3-DAG), diacylglycerol 16:0 (di16:0-DAG), monogalactosyldiacylglyceride (MGDG), digalactosyldiacylglyceride (DGDG), phosphatidylglycerin (PG), phosphatidylethanolamine (PE) and cardiolipin (CL). (b) Evaluation of differences in PA binding strength of protein fragments using filters containing 10, 5, 1, 0.5 and 0.1 ng of PA. shift assays Ca2+ might be required to crosslink pectins and therefore increases the strength of the interaction. In A. thaliana a Ca2+-induced modification of the pectin structure supports the interaction with AtWAK1 (Decreux and Messiaen, 2005). Ca2+ seems not to be necessary for CpGRP1
pectin binding in vitro, but a role of Ca2+ in planta cannot be excluded. To compare the CpGRP1
pectin interaction with other apoplastic proteins, CpWAK1 and CpGLP1 were used. CpWAK1 and CpGLP1 were chosen, because they are exposed to pectin in a similar way as CpGRP1 due to their localisation. Both proteins are able to bind pectin (Figure 6) but the interaction of CpGRP1 with pectin is much stronger. To exclude the pectin binding of CpGRP1, CpWAK1 and CpGLP1 is just due to a high electrostatic charge, two other proteins with high lysine and arginine frequencies were used as additional controls (Figure 5). The recombinant LEA-like 11-24 protein (Petersen et al., 2012) has a lysine frequency of 9.1% which is more than double compared with CpGRP1, which has a lysine frequency of 3% and an arginine frequency of 7.4%. BSA (UniProtKB accession number P02769) has a lysine frequency of 9.9% and an arginine frequency of 4.3%. Despite the high lysine and arginine frequencies LEA-like 11–24 and BSA did not show any interaction to pectin, proving a high electrostatic charge alone is not sufficient for a protein
pectin interaction and a particular spatial arrangement of the amino acid residues is required. The strong pectin binding of CpGRP1 seems to be specific, as other apoplastic proteins or proteins with a high electrostatic charge did not show any or a weak protein
pectin binding. After identifying the protein domains important for pectin binding, we investigated the pectin fractions with regards to CpGRP1 binding. Both the CpGRP1 full-length protein and the N-term. fragment showed a slightly stronger binding to pectin samples isolated from desiccated C. plantagineum leaves than to pectins isolated from untreated leaves (Figure 4a). This could be due to the methylesterification level of HG which decreased upon dehydration. Previous reports demonstrated the importance of de-methylesterified HG stretches for protein binding (Spadoni et al., 2006; Chevalier et al., 2019). Pectin with a lower degree of methylesterification provides more binding sites for proteins. The small N-term. fragment showed always a higher signal in ELISA pectin binding assays than the full-length CpGRP1 protein (Figure 4a,b). Likely, more protein molecules can bind to the de-methylesterified pectin and this explains the stronger signal. To demonstrate a link between the degree of pectin methylesterification and protein binding, the C. plantagineum pectin was treated with solutions of different pH values to stepwise de-methylesterify the HG. Our results demonstrate that de-methylesterification of HG provides more binding sites for CpGRP1 (Figure 7). Reversible methylesterification of HG in the C. plantagineum cell wall during dehydration might be © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2019), 100, 661–676 672 Niklas U. Jung et al. Figure 9. Model depicting the interaction between the Craterostigma plantagineum glycine-rich protein 1 (CpGRP1), the C. plantagineum cell wall associated protein kinase 1 (CpWAK1) and cell wall pectins and their role in regulating growth and dehydration-stress responses (modified from Giarola et al., 2016). CpGRP1 is a part of the pectin matrix. Under non-stress conditions cell wall pectins (jagged blue lines) with a high degree of methylesterification (red dots) are bound to CpWAK1 and are involved in cell expansion. Upon dehydration, CpGRP1 is more abundant. Clustered arginines in the N-terminal domain of CpGRP1 are involved in binding PA and de-methylesterified stretches of pectin, which are more abundant upon dehydration. The binding of CpGRP1 to CpWAK1 is most likely mediated by cysteines in the C-terminal domain of CpGRP1. CpGRP1 might be either bound to PA or pectin with both competing for CpGRP1 binding. Ca2+ levels increase upon dehydration which may lead to the formation of egg-box structures resulting in an increase in rigidity of the cell walls. responsible for the reduction of mechanical stress and might provide more binding sites for proteins. However, the protein-binding experiments using pectin from L. brevidens and L. subracemosa suggest that other factors in the structure of pectins also contribute to protein
pectin interactions. No significant changes were detected in the degree of methylesterification between C. plantagineum, L. brevidens and L. subracemosa, but the binding of the CpGRP1 protein to the pectin fractions from the three species was different (Figures 4 and 5). There seems to be a correlation between desiccation tolerance and the CpGRP1-binding capacity to pectin, as the signal is strongest for the C. plantagineum pectin, weaker for the L. brevidens pectin and even weaker for the pectin isolated from the non-tolerant L. subracemosa. The reasons for these differences are unknown. Pectins and hemicelluloses of the C. plantagineum cell wall are modified, especially the HG and xyloglucan. These changes strengthen the cell wall. A higher proportion of de-methylesterified HG provides more binding sites for the CpGRP1 protein. CpGRP1 interacts through arginines with HG or PA and via cysteines in the C-terminal part of CpGRP1 with the extracellular domain of CpWAK1. These results are now integrated in a model how CpGRP1, CpWAK1, pectins and PA may interact and what are their possible roles in regulating cell shrinkage and expansion (Figure 9). We propose, that CpGRP1 is an essential factor in cell wall adaptations to desiccation. EXPERIMENTAL PROCEDURES Cultivation of plants C. plantagineum Hochst., L. brevidens Skan, and L. subracemosa De Wild plants were grown as described in Bartels et al. (1990) and Phillips et al. (2008). Plants were subjected to desiccation and rehydration treatments and the hydration status of leaves was determined by calculating the relative water content (RWC) with the formula: RWC [%] = (Initial weight
dry weight)/(full turgor weight
dry weight) 9 100 (Bernacchia et al., 1996). For dehydration, fully-grown plants were dehydrated in pots until partial dehydration (PD, 50%) or desiccation (D, 2%) samples. Completely desiccated plants were rehydrated for either 24 or 48 h. Leaves from dehydrated and rehydrated plants were ground in liquid nitrogen with a mortar and a pestle and stored at
80°C. Molecular biology techniques and DNA sequencing analysis Molecular biology techniques were performed according to Green and Sambrook (2012). DNA sequencing was carried out by GATC Biotech (https://www.gatc-biotech.com/en/index.html) and primer synthesis by Eurofins MWG Operon (http://www.eurofinsgenomics. eu). All primers used are listed in Table S1. Scanning electron microscopy C. plantagineum leaves were frozen in liquid nitrogen and sputtered with palladium for 2 min using a sputter-coater (SCD 040; Balzer, http://www.oerlikon.com/balzers). Leaves were fixed on the sample holder and analysed under the electron microscope at 9100 and 9400 magnification (Cambridge Stereoscan S 200; Cambridge Instrument Company, UK). © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2019), 100, 661–676 Modifications of pectin in response to desiccation 673 Cell walls were stained as follows: leaf material was cut with a razor blade and immediately immersed in cold FAA (10% (v/v) formalin, 10% (v/v) acetic acid, 30% (v/v) water and 50% (v/v) ethanol) solution for at least 24 h at 4°C. After fixation samples were incubated for 30 min in 85% (v/v) ethanol, 50% (v/v) ethanol/50% (v/v) acetone and finally in 100% (v/v) acetone solutions. Samples were embedded using the Agar Low Viscosity Kit (Plano GmbH, Wetzlar, https://www.plano-em.de). Cell wall structures were stained with a 30% (w/v) silver nitrate solution for 10 min. After staining samples were incubated in 10% (v/v) HCl for 5 min before analysing the surface under the SEM (Blockface imaging). Critical point drying (CPD 020; Balzers, http://www.oerlikon. com/balzers) was performed according to Svitkina et al. (1984) before analysing with the SEM. Protein analyses The sequence encoding the N-terminal fragment (aa 22–120) and the C-terminal fragment (aa 121–156) of CpGRP1 (GenBank accession number ALQ43973.1) was amplified with primers from the pET28 CpGRP1His plasmid to add a NcoI site and a XhoI site at the 50 and 30 ends, respectively (CpGRP1_NTERM_R/T7 promoter and CpGRP1_CTERM_F/T7 terminator, Table S1). The sequence encoding the extracellular domain of CpWAK1 (aa 31– 315) was amplified with primers from a CpWAK1 cDNA clone (GenBank accession number KT893872.1; Giarola et al., 2016) to add a XhoI site at the 30 end (pJET1.2 fwd and CpWAK1_XhoI_R, Table S1). A NcoI site is already present in the CpWAK1 sequence. The sequence encoding the C. plantagineum germinlike protein 1 (CpGLP1, Dulitz, 2016) without the signal peptide (aa 27–226) was excised from a pAD vector using EcoRI and SalI restriction enzymes. N- and C-terminal fragments of CpGRP1 as well as the CpWAK1 extracellular fragment were cloned between the NcoI and the XhoI sites of the pET28a expression vector (NOVAGEN, http://www.novagen.com) to create the corresponding protein-His-tag translation fusion constructs (pET28 CpGRP1_N-terminalHis, pET28 CpGRP1_C-terminalHis, and pET28 CpWAK1_extracellularHis, respectively). The CpGLP1 fragment was cloned between EcoRI and SalI sites of the pET28a vector to generate the pET28 CpGLP1_His fusion construct. Overexpression constructs were transformed into BL21 (DE3) Escherichia coli cells (Amersham Pharmacia Biotech, https://www.ge lifesciences.com) and protein overexpression was induced by adding 1 mM isopropyl-1-thio-b-D-galaytopyranoside (IPTG). The recombinant proteins were purified from bacteria 5 h after IPTG € hrig, induction using affinity chromatography (Kirch and Ro 2010). Purified protein fragments were concentrated using Amicon Ultracel-10K centrifugal concentrators (MILLIPORE, http:// www.millipore.com) and desalted with PD10-columns (http:// www.gelifesciences.com) before freeze-drying. Freeze-dried proteins were used for Blue-native page gel-shift assays, ELISA pectin binding and dot-blot assays. The CpGRP1 sequence coding for two arginine residues (arginine [118] and [120]) in the CpGRP1 N-terminal fragment was mutated with primers (CpGRP1_NTERM_a352g_c358g_F and CpGRP1_NTERM_a352g_c358g_R. Table S1) using the QuikChange II site-directed mutagenesis kit according to the manufacturer0 s instructions (Agilent, https://www.agilent.com). Separation of proteins was performed as described in He (2011) based on the method first described by Laemmli (1970). The separation of the CpGRP1 C-terminal fragment was performed as described by Scha€gger and von Jagow (1987) using a peptide page. Proteins of SDS-PAGE gels were visualised with Coomassie brilliant blue R250 (0.1% (w/v) Coomassie R250, 50% (v/v) methanol, 10% (v/v) glacial acetic acid, 40% (v/v) water). Immunoblot analyses were performed according to Towbin et al. (1979). CpGRP1 or 6x-HisTag polyclonal antibodies (http://www.thermofisher.com) were used at a dilution of 1:5000. Detection of proteins was performed using the ECL Western Blotting detection kit (GE Healthcare, http:// www.gehealthcare.com). Signals were visualised using the Azure Biosystems c300 chemiluminescent detection system (http:// www.biozym.com). Extraction of cell wall components Cell wall components were obtained by following extractions as described by Cornuault et al. (2014). The 1,2-cyclohexanediaminetetraacetic acid (CDTA) fraction was obtained by vortexing 1 mg freeze-dried material in 1 ml 50 mM CDTA pH 7.5 for 1 h. The KOH fraction was obtained by vortexing the pellet from the CDTA extraction in 1 ml 4 M KOH, 1% (w/v) NaBH4 solution for 1 h. Acetic acid (80%) was used to neutralise the pH of the KOH fraction after extraction. All samples were stored at
20°C until use. Enzyme-linked immunosorbent assay Isolated CDTA and KOH fractions were incubated in microtiter plates (NUNC-Immuno MicroWell 96 well solid plates, flat bottom, http://www.sigmaaldrich.com) overnight at 4°C. Plate wells were washed vigorously six times with PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2 mM KH2PO4) and then blocked using 200 ll per well of 4% (w/v) non-fat milk powder in PBS for 2 h at room temperature. The plates were washed nine times with PBS and padded dry. Primary antibodies were added and incubated overnight at 4°C. The plates were washed 12 times with PBS and shaken dry. Then 150 ll secondary peroxidasecoupled anti-rabbit or anti-rat IgG antibodies (http://www.sigmaa ldrich.com) were added at a 5000-fold dilution in 4% milk powder/PBS for 1 h at room temperature. After washing 12 times with PBS, microtitre plates were developed using 150 ll of substrate solution (0.1 M sodium acetate buffer, pH 6, 0.1% (w/v) tetramethyl benzidine, 0.006% (v/v) H2O2) in each well. The enzyme reaction was stopped by adding 40 ll of 2.5 M H2SO4 to each well, and the absorbance at 450 nm was determined for each well. Analysis of the cell wall composition. The experiment was performed as described in Cornuault et al. (2014) with minor changes. Nine different rat monoclonal antibodies (mAB) were used in this study: JIM5, JIM7, LM20, LM19, LM25, LM15, LM11, LM5 and LM6 and one rabbit mAb: 42-6 (Table 1). CDTA and KOH 1:5 dilutions were used for ELISAs as they showed the most appropriate signal intensity (Figure S1). Ten-fold dilution of hybridoma cell culture supernatants in 4% (w/v) non-fat milk powder/ PBS and a 1:10 000 dilution for 42-6 were used as primary antibodies (150 ll each well). CpGRP1, CpWAK1, CpGLP1 pectin binding assay. After incubating CDTA and KOH fractions overnight, the recombinant proteins were dissolved in 4% (w/v) non-fat milk powder at a concentration of 1 lg ll
1 and incubated in the plates overnight at 4°C. The protein was detected using a 6x-His-Tag polyclonal antibody (http:// www.thermofisher.com) at a 1:5000 or 1:10 000 dilution for CpGRP1, 1:10 000 for CpWAK1 and 1:10 000 for CpGLP1. Pre-treatment of pectin was performed by incubating the plates with 50 mM CAPS-buffered solution at alkaline pH (adjusted between 7 and 11 with HCl and KOH) or with 0.1 M sodium carbonate. © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2019), 100, 661–676 674 Niklas U. Jung et al. Determination of GA The GA content was determined according to Blumenkrantz and Asboe-Hansen (1973) and Verma et al. (2014). All CDTA and KOH fractions were analysed for their GA content by m-hydroxydiphenyl method, using GA as a standard (Figure S2). Blue-native page gel-shift assays For gel-shift assays 0.5 lg of CpGRP1 were incubated with 0.5 lg of citrus pectin (http://www.sigmaaldrich.com) and 2 mM CaCl2 for 4 h at room temperature. The samples were mixed with 59 sample buffer (15.5 ml 1 M Tris
HCl, pH 6.8; 2.5 ml of 1% bromophenol blue solution; 7 ml of water; 25 ml of glycerol) and separated in 15% polyacrylamide gels without SDS and without stacking gel. A Tris/ boric acid buffer (89 mM Tris, 89 mM boric acid; pH 9.25) was used for gel preparation and gel electrophoresis. Gels were either stained with Coomassie brilliant blue or incubated in 50 mM Tris/1% (w/v) SDS buffer prior to immunoblotting. Proteins were detected using a 6x-His-Tag polyclonal antibody (http://www.thermofisher.com) at a 1:5000 dilution. Buchholz and T. Dieckmann for providing plant material, H-J. Ensikat for help with the SEM, S. Raj and P. Do€ rmann for help with lipid-binding experiments and M. Kobayashi (Graduate School of Agriculture, Kyoto University, Kyoto) for providing the 42-6 antibody. S. Manduzo and A. Sergeeva helped with providing the expression constructs for the recombinant CpGLP1 and CpWAK1, respectively. The work was supported by the Deutsche Forschungsgemeinschaft (BA712-18-1). N. Jung is a member of the GRK2064. CONFLICT OF INTEREST The authors confirm that there is no conflict of interest to declare. AUTHOR CONTRIBUTIONS NJ and PC planned and designed the research and conducted the experiments. NJ wrote the manuscript. DB, PK and VG designed the research, supervised the work and corrected the manuscript. Dot-blot pectin binding assays SUPPORTING INFORMATION Dot-blot assays were either based on proteins or pectins immobilised on a nitrocellulose membrane. Dots of 1.5 ll of either polygalacturonic acid (PGA) or of the CDTA-pectin fraction or of the CpGRP1 recombinant protein were spotted on a nitrocellulose membrane and allowed to dry. Membranes were blocked with 4% (w/v) non-fat milk powder in TBST for 2 h and were then washed with TBST three times for 5 min. The membranes were incubated with CpGRP1 recombinant protein (1 lg ml
1; membranes with PGA and CDTA spots) or pectin (2 mg ml
1; membranes with CpGRP1 spots) in TBST overnight at 4°C followed by three washing steps. The membranes were incubated with 6x-His-tag antibody (1:5000 dilution) or JIM5 antibody (1:10 dilution) overnight at 4°C and then washed again three times. Immunodetection was performed as described above and quantification of dot intensity was done using ImageJ (http://www.imagej.net). Additional Supporting Information may be found in the online version of this article. Protein-lipid overlay assay Protein-lipid overlay assays were performed according to Deak et al. 1999 and Ufer et al. 2017 to analyse lipid-binding properties of CpGRP1. Electrostatic surface modelling The Phyre2 (http://www.sbg.bio.ic.ac.uk/phyre2/html/page.cgi?id=in dex) (Kelley et al., 2015) web service was used to create a PDB-file from the CpGRP1 protein sequence, which was then used for electrostatic surface modelling using the website http://www.charmm-gui. org/?doc=input/pbeqsolver (Im et al., 1998; Jo et al., 2008a,b). ACCESSION NUMBERS The GenBank accession numbers used here are KT893871. 1 (CpGRP1), KT893872.1 (CpWAK1) and JQ067608.1 (LEAlike 11–24). The UniProtKB accession numbers used here are A0A0S2ZYI4 (CpGRP1), A0A0S2ZYM7 (CpWAK1), P02769 (BSA) and O23764 (CpLEA-like 11-24). ACKNOWLEDGEMENTS Part of this work was inspired by discussions with J. Renaut and colleagues within the SMARTWALL project. We thank C. Figure S1. Standard-curve to calculate the galacturonic acid (GA) content in the different pectin fractions. Figure S2. LM25 binding to different dilutions of the KOH fraction extracted from C. plantagineum to determine the most appropriate dilution for ELISA studies. Table S1. List of primers used in this study. REFERENCES Anderson, C.M., Wagner, T.A., Perret, M., He, Z.H., He, D. and Kohorn, B.D. (2001) WAKs. Cell wall-associated kinases linking the cytoplasm to the extracellular matrix. Plant Mol. Biol. 47, 197–206. Bartels, D. and Hussain, S.S. (2011) Resurrection plants: physiology and molecular biology. In Plant Desiccation Tolerance (L€ uttge, U., Beck, E. and Bartels, D., eds). Berlin: Springer, pp. 339–357. Bartels, D., Schneider, K., Terstappen, G., Piatkowski, D. and Salamini, F. (1990) Molecular cloning of abscisic acid-modulated genes which are induced during desiccation of the resurrection plant Craterostigma plantagineum. Planta, 181, 27–34. Bernacchia, G., Salamini, F. and Bartels, D. (1996) Molecular characterization of the rehydration process in the resurrection plant Craterostigma plantagineum. Plant Physiol. 111, 1043–1050. Blumenkrantz, N. and Asboe-Hansen, G. (1973) New method for quantitative determination of uronic acids. Anal. Biochem. 54, 484– 489. Bocca, S.N., Magioli, C., Mangeon, A., Junqueira, R.M., Cardeal, V., Margis, R. and Sachetto-Martins, G. (2005) Survey of glycine-rich proteins (GRPs) in the Eucalyptus expressed sequence tag database (ForEST). Genet. Mol. Biol. 28, 608–624. Brown, D.M., Goubet, F., Wong, V.W., Goodacre, R., Stephens, E., Dupree, P. and Turner, S.R. (2007) Comparison of five xylan synthesis mutants reveals new insight into the mechanisms of xylan synthesis. Plant J. 52, 1154–1168. Chen, A.P., Zhong, N.Q., Qu, Z.L., Wang, F., Liu, N. and Xia, G.X. (2007) Root and vascular tissue-specific expression of glycine-rich protein AtGRP9 and its interaction with AtCAD5, a cinnamyl alcohol dehydrogenase, in Arabidopsis thaliana. J. Plant. Res. 120, 337–343. Chevalier, L.M., Rioux, L.E., Angers, P. and Turgeon, S.L. (2019) Study of the interactions between pectin in a blueberry puree and whey proteins. Functionality and application. Food Hydrocolloids, 87, 61– 70. © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2019), 100, 661–676 Modifications of pectin in response to desiccation 675 Condit, C.M. and Meagher, R.B. (1986) A gene encoding a novel glycine-rich structural protein of petunia. Nature, 323, 178–181. Condit, C.M. and Meagher, R.B. (1987) Expression of a gene encoding a glycine-rich protein in petunia. Mol. Cell. Biol. 7, 4273–4279. Cornuault, V., Manfield, I.W., Ralet, M.C. and Knox, J.P. (2014) Epitope detection chromatography: a method to dissect the structural heterogeneity and inter-connections of plant cell-wall matrix glycans. Plant J. 78, 715–722. Czolpinska, M. and Rurek, M. (2018) Plant glycine-rich proteins in stress response. An emerging, still prospective story. Front. Plant Sci. 9, 302. Deak, M., Casamayor, A., Currie, R.A., Downes, C.P. and Alessi, D.R. (1999) Characterisation of a plant 3-phosphoinositide-dependent protein kinase1 homologue which contains a pleckstrin homology domain. FEBS Lett. 451, 220–226. Decreux, A. and Messiaen, J. (2005) Wall-associated kinase WAK1 interacts with cell wall pectins in a calcium-induced conformation. Plant Cell Physiol. 46, 268–278. Dulitz, J.S. (2016) Isolation and characterization of the Craterostigma plantagineum germin-like protein 1 CpGLP1. Master Thesis. University of Bonn, Germany. Farrant, J.M. (2000) Comparison of mechanisms of desiccation tolerance among three angiosperm resurrection plants. Plant Ecol. 151, 29–39. Farrant, J.M., Bartsch, S., Loffell, D., Vander Willigen, C. and Whittaker, A. (2003) An investigation into the effects of light on the desiccation of three resurrection plants species. Plant Cell Environ. 26, 1275–1286. Fleischer, A., O’Neill, M.A. and Ehwald, R. (1999) The pore size of non-graminaceous plant cell walls Is rapidly decreased by borate ester cross-linking of the pectic polysaccharide rhamnogalacturonan II. Plant Physiol. 121, 829–838. Fry, S.C. (1989) The structure and functions of Xyloglucan. J. Exp. Bot. 40, 1–11. Fusaro, A., Mangeon, A., Junqueira, R.M., Rocha, C.A.B., Coutinho, T.C., Margis, R. and Sachetto-Martins, G. (2001) Classification, expression pattern and comparative analysis of sugarcane expressed sequences tags (ESTs) encoding glycine-rich proteins (GRPs). Genet. Mol. Biol. 24, 263–273. Gaff, D.F. (1971) Desiccation-tolerant flowering plants in southern Africa. Science, 174, 1033–1034. Gaff, D.F. and Bole, P.V. (1986) Resurrection grasses in India. Oecologia, 71, 159–160. € rmann, P. and Gasulla, F., Dorp, K., Dombrink, I., Za€hringer, U., Gisch, N., Do Bartels, D. (2013) The role of lipid metabolism in the acquisition of desiccation tolerance in Craterostigma plantagineum: a comparative approach. Plant J. 75, 726–741. Giarola, V., Krey, S., Frerichs, A. and Bartels, D. (2015) Taxonomically restricted genes of Craterostigma plantagineum are modulated in their expression during dehydration and rehydration. Planta, 241, 193–208. Giarola, V., Krey, S., den Driesch, Bv. and Bartels, D. (2016) The Craterostigma plantagineum glycine-rich protein CpGRP1 interacts with a cell wall-associated protein kinase 1 (CpWAK1) and accumulates in leaf cell walls during dehydration. New Phytol. 210, 535–550. Giarola, V., Hou, Q. and Bartels, D. (2017) Angiosperm plant desiccation tolerance: hints from transcriptomics and genome sequencing. Trends Plant Sci. 22, 705–717. Grant, G.T., Morris, E.R., Rees, D.A., Smith, P.J.C. and Thom, D. (1973) Biological interactions between polysaccharides and divalent cations: the egg-box model. FEBS Lett. 32, 195–198. Green, M.R. and Sambrook, J. (2012) Molecular Cloning. A Laboratory Manual. Cold Spring Harbor: Cold Spring Harbor Laboratory Press. Ha, M.A., Evans, B.W., Jarvis, M.C., Apperley, D.C. and Kenwright, A.M. (1996) CP-MAS NMR of highly mobile hydrated biopolymers. Polysaccharides of Allium cell walls. Carbohyd. Res. 288, 15–23. He, F. (2011) Laemmli-SDS-PAGE. Bio-101, e80. https://doi.org/10.21769/bio protoc.80 He, Z.H., Cheeseman, I., He, D. and Kohorn, B.D. (1999) A cluster of five cell wall-associated receptor kinase genes, Wak1-5, are expressed in specific organs of Arabidopsis. Plant Mol. Biol. 39, 1189–1196. Hou, Q., Ufer, G. and Bartels, D. (2016) Lipid signalling in plant responses to abiotic stress. Plant Cell Environ. 39, 1029–1048. Iljin, W.S. (1957) Drought resistance in plants and physiological processes. Annu. Rev. Plant Physiol. 8, 257–274. Im, W., Beglov, D. and Roux, B. (1998) Continuum solvation model. Computation of electrostatic forces from numerical solutions to the PoissonBoltzmann equation. Comput. Phys. Commun. 111, 59–75. Jarvis, M.C. (1984) Structure and properties of pectin gels in plant cell walls. Plant Cell Environ. 7, 153–164. Jo, S., Kim, T., Iyer, V.G. and Im, W. (2008a) CHARMM-GUI: a web-based graphical user interface for CHARMM. J. Comput. Chem. 29, 1859–1865. Jo, S., Vargyas, M., Vasko-Szedlar, J., Roux, B. and Im, W. (2008b) PBEQSolver for online visualization of electrostatic potential of biomolecules. Nucleic Acids Res. 36, W270–W275. Jones, L. and McQueen-Mason, S. (2004) A role for expansins in dehydration and rehydration of the resurrection plant Craterostigma plantagineum. FEBS Lett. 559, 61–65. Jones, L., Seymour, G.B. and Knox, J.P. (1997) Localization of pectic galactan in tomato cell walls using a monoclonal antibody specific to (1[->]4)[beta]-D-Galactan. Plant Physiol. 113, 1405–1412. Keller, B., Sauer, N. and Lamb, C.J. (1988) Glycine-rich cell wall proteins in bean. Gene structure and association of the protein with the vascular system. EMBO J. 7, 3625–3633. Kelley, L.A., Mezulis, S., Yates, C.M., Wass, M.N. and Sternberg, M.J.E. (2015) The Phyre2 web portal for protein modeling, prediction and analysis. Nat. Protoc. 10, 845 EP. € hrig, H. (2010) Affinity purification and determination of Kirch, H.H. and Ro enzymatic activity of recombinantly expressed aldehyde dehydrogenases. Methods Mol. Biol. 639, 282–291. Knox, J.P., Linstead, P.J., King, J., Cooper, C. and Roberts, K. (1990) Pectin esterification is spatially regulated both within cell walls and between developing tissues of root apices. Planta, 181, 512–521. Kobayashi, M., Matoh, T. and Azuma, Ji (1996) Two chains of rhamnogalacturonan II are cross-linked by borate-diol ester bonds in higher plant cell walls. Plant Physiol. 110, 1017–1020. Kohorn, B.D. and Kohorn, S.L. (2012) The cell wall-associated kinases, WAKs, as pectin receptors. Front. Plant Sci. 3, 88. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680–685. Le Gall, H., Philippe, F., Domon, J.M., Gillet, F., Pelloux, J. and Rayon, C. (2015) Cell wall metabolism in response to abiotic stress. Plants, 4, 112– 166. Levitt, J. (1987) Responses of Plants to Environmental Stresses. London: Academic Press. Lloyd, C.W. (1991) The Cytoskeletal Basis of Plant Growth and Form. London: Academic Press. Mangeon, A., Junqueira, R.M. and Sachetto-Martins, G. (2010) Functional diversity of the plant glycine-rich proteins superfamily. Plant Signal. Behav. 5, 99–104. Marcus, S.E., Verhertbruggen, Y., Herve
, C., Ordaz-Ortiz, J.J., Farkas, V., Pedersen, H.L., Willats, W.G. and Knox, J.P. (2008) Pectic homogalacturonan masks abundant sets of xyloglucan epitopes in plant cell walls. BMC Plant Biol. 8, 60. McCartney, L., Marcus, S.E. and Knox, J.P. (2005) Monoclonal antibodies to plant cell wall xylans and arabinoxylans. J. Histochem. Cytochem. 53, 543–546. Micheli, F. (2001) Pectin methylesterases. Cell wall enzymes with important roles in plant physiology. Trends Plant Sci. 6, 414–419. Moore, P.J., Darvill, A.G., Albersheim, P. and Staehelin, L.A. (1986) Immunogold localization of xyloglucan and rhamnogalacturonan I in the cell walls of suspension-cultured sycamore cells. Plant Physiol. 82, 787– 794. Moore, J.P., Nguema-Ona, E., Chevalier, L., Lindsey, G.G., Brandt, W.F., Lerouge, P., Farrant, J.M. and Driouich, A. (2006) Response of the leaf cell wall to desiccation in the resurrection plant Myrothamnus flabellifolius. Plant Physiol. 141, 651–662. Moore, J.P., Farrant, J.M. and Driouich, A. (2008) A role for pectin-associated arabinans in maintaining the flexibility of the plant cell wall during water deficit stress. Plant Signal. Behav. 3, 102–104. Moore, J.P., Nguema-Ona, E.E., Vicre
-Gibouin, M., Sørensen, I., Willats, W.G., Driouich, A. and Farrant, J.M. (2013) Arabinose-rich polymers as an evolutionary strategy to plasticize resurrection plant cell walls against desiccation. Planta, 237, 739–754. Munnik, T. (2001) Phosphatidic acid: an emerging plant lipid second messenger. Trends Plant Sci. 6, 227–233. © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2019), 100, 661–676 676 Niklas U. Jung et al. O’Neill, M.A., Eberhard, S., Albersheim, P. and Darvill, A.G. (2001) Requirement of borate cross-linking of cell wall rhamnogalacturonan II for Arabidopsis growth. Science, 294, 846–849. Park, A.R., Cho, S.K., Yun, U.J., Jin, M.Y., Lee, S.H., Sachetto-Martins, G. and Park, O.K. (2001) Interaction of the Arabidopsis receptor protein kinase Wak1 with a glycine-rich protein, AtGRP-3. J. Biol. Chem. 276, 26688–26693. Pedersen, H.L., Fangel, J.U., McCleary, B. et al. (2012) Versatile high resolution oligosaccharide microarrays for plant glycobiology and cell wall research. J. Biol. Chem. 287, 39429–39438. Petersen, J., Eriksson, S.K., Harryson, P., Pierog, S., Colby, T., Bartels, D. € hrig, H. (2012) The lysine-rich motif of intrinsically disordered and Ro stress protein CDeT11-24 from Craterostigma plantagineum is responsible for phosphatidic acid binding and protection of enzymes from damaging effects caused by desiccation. J. Exp. Bot. 63, 4919–4929. Phillips, J.R., Fischer, E., Baron, M., van den Dries, N., Facchinelli, F., Kutzer, M., Rahmanzadeh, R., Remus, D. and Bartels, D. (2008) Lindernia brevidens: a novel desiccation-tolerant vascular plant, endemic to ancient tropical rainforests. Plant J. 54, 938–948. Porembski, S. (2000) Granitic and gneissic outcrops (inselbergs) as centers of diversity for desiccation-tolerant vascular plants. Plant Ecol. 151, 19– 28. € ller, K., Fischer, E., Bartels, D. and Borsch, T. (2005) Rahmanzadeh, R., Mu The Linderniaceae and Gratiolaceae are further lineages distinct from the Scrophulariaceae (Lamiales). Plant Biol. 7, 67–78. Ringli, C., Hauf, G. and Keller, B. (2001) Hydrophobic interactions of the structural protein GRP1.8 in the cell wall of protoxylem elements. Plant Physiol. 125, 673–682. Ryser, U., Schorderet, M., Guyot, R. and Keller, B. (2004) A new structural element containing glycine-rich proteins and rhamnogalacturonan I in the protoxylem of seed plants. J. Cell Sci. 117, 1179–1190. Sachetto-Martins, G., Franco, L.O. and de Oliveira, D.E. (2000) Plant glycinerich proteins. A family or just proteins with a common motif? Biochem. Biophys. Acta, 1492, 1–14. Scha€gger, H. and von Jagow, G. (1987) Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal. Biochem. 166, 368–379. Seine, R., Fischer, E. and Barthlott, W. (1995) Notes on the Scrophulariaceae of Zimbabwean inselsbergs, with the description of Lindernia syncerus sp. nov. Feddes Repert. 106, 7–12. Spadoni, S., Zabotina, O., Di Matteo, A., Mikkelsen, J.D., Cervone, F., de Lorenzo, G., Mattei, B. and Bellincampi, D. (2006) Polygalacturonase-inhibiting protein interacts with pectin through a binding site formed by four clustered residues of arginine and lysine. Plant Physiol. 141, 557– 564. Staehelin, L.A. and Moore, I. (1995) The plant Golgi apparatus. Structure, functional organization and trafficking mechanisms. Annu. Rev. Plant Physiol. Plant Mol. Biol. 46, 261–288. Sterling, J.D. (2001) The catalytic site of the pectin biosynthetic enzyme alpha-1,4-galacturonosyltransferase is located in the lumen of the Golgi. Plant Physiol. 127, 360–371. Svitkina, T.M., Shevelev, A.A., Bershadsky, A.D. and Gelfand, V.I. (1984) Cytoskeleton of mouse embryo fibroblasts. Electron microscopy of platinum replicas. Eur. J. Cell Biol. 34, 64–74 PMID:6539695. Towbin, H., Staehelin, T. and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl Acad. Sci. USA, 76, 4350–4354. Ueki, S. and Citovsky, V. (2002) The systemic movement of a tobamovirus is inhibited by a cadmium-ion-induced glycine-rich protein. Nat. Cell Biol. 4, 478–486. € hrig, H. and Bartels, D. (2017) IdentiUfer, G., Gertzmann, A., Gasulla, F., Ro fication and characterization of the phosphatidic acid-binding A. thaliana phosphoprotein PLDrp1 that is regulated by PLDa1 in a stress-dependent manner. Plant J. 92, 276–290. VanBuren, R., Wai, C.M., Pardo, J., Giarola, V., Ambrosini, S., Song, X. and Bartels, D. (2018) Desiccation tolerance evolved through gene duplication and network rewiring in Lindernia. Plant Cell, 30, 2943–2958. Verhertbruggen, Y., Marcus, S.E., Haeger, A., Ordaz-Ortiz, J.J. and Knox, J.P. (2009) An extended set of monoclonal antibodies to pectic homogalacturonan. Carbohyd. Res. 344, 1858–1862. Verma, C., Singh, R.K., Singh, R.B. and Mishra, S. (2014) Biochemical and in-silico studies on pectin methylesterase from G9 variety of Musa acuminata for delayed ripening. Open Biochem. J. 9, 15–23. Verslues, P.E. and Juenger, T.E. (2011) Drought, metabolites, and Arabidopsis natural variation: a promising combination for understanding adaptation to water-limited environments. Curr. Opin. Plant Biol. 14, 240–245. Vicre
, M., Sherwin, H.W., Driouich, A., Jaffer, M.A. and Farrant, J.M. (1999) Cell wall characteristics and structure of hydrated and dry leaves of the resurrection plant Craterostigma wilmsii, a microscopical study. J. Plant Physiol. 155, 719–726. Vicre
, M., Lerouxel, O., Farrant, J.M., Lerouge, P. and Driouich, A. (2004) Composition and desiccation-induced alterations of the cell wall in the resurrection plant Craterostigma wilmsii. Physiol. Plant. 120, 229–239. Wang, L., Shang, H., Liu, Y., Zheng, M., Wu, R., Phillips, J., Bartels, D. and Deng, X. (2009) A role for a cell wall localized glycine-rich protein in dehydration and rehydration of the resurrection plant Boea hygrometrica. Plant Biol. 11, 837–848. Willats, W.G., Marcus, S.E. and Knox, J.P. (1998) Generation of a monoclonal antibody specific to (1–>5)-alpha-L-arabinan. Carbohyd. Res. 308, 149–152. Willigen, C.V., Pammenter, N.W., Jaffer, M.A., Mundree, S.G. and Farrant, J.M. (2003) An ultrastructural study using anhydrous fixation of Eragrostis nindensis, a resurrection grass with both desiccation-tolerant and -sensitive tissues. Funct. Plant Biol. 30, 281–290. Wu, H.C. and Jinn, T.L. (2010) Heat shock-triggered Ca2+ mobilization accompanied by pectin methylesterase activity and cytosolic Ca2+ oscillation are crucial for plant thermotolerance. Plant Signal. Behav. 5, 1252–1256. € rnblad, E., Singh, S.K., MatWu, A.M., Rihouey, C., Seveno, M., Ho sunaga, T., Ishii, T., Lerouge, P. and Marchant, A. (2009) The Arabidopsis IRX10 and IRX10-LIKE glycosyltransferases are critical for glucuronoxylan biosynthesis during secondary cell wall formation. Plant J. 57, 718–731. Wu, H.C., Hsu, S.F., Luo, D.L., Chen, S.J., Huang, W.D., Lur, H.S. and Jinn, T.L. (2010) Recovery of heat shock-triggered released apoplastic Ca2+ accompanied by pectin methylesterase activity is required for thermotolerance in soybean seedlings. J. Exp. Bot. 61, 2843–2852. Wu, H.C., Bulgakov, V.P. and Jinn, T.L. (2018) Pectin methylesterases. Cell wall remodeling proteins are required for plant response to heat stress. Front. Plant Sci. 9, 1612. Yokoyama, R. and Nishitani, K. (2006) Identification and characterization of Arabidopsis thaliana genes involved in xylem secondary cell walls. J. Plant. Res. 119, 189–194. Zhang, Q. and Bartels, D. (2018) Molecular responses to dehydration and desiccation in desiccation-tolerant angiosperm plants. J. Exp. Bot. 69, 3211–3222. Zhang, G.F. and Staehelin, L.A. (1992) Functional compartmentation of the Golgi apparatus of plant cells. Immunocytochemical analysis of highpressure frozen- and freeze-substituted sycamore maple suspension culture cells. Plant Physiol. 99, 1070–1083. Zhou, Y., Kobayashi, M., Awano, T., Matoh, T. and Takabe, K. (2018) A new monoclonal antibody against rhamnogalacturonan II and its application to immunocytochemical detection of rhamnogalacturonan II in Arabidopsis roots. Biosci. Biotechnol. Biochem. 82, 1780–1789. © 2019 The Authors. The Plant Journal published by Society for Experimental Biology and John Wiley & Sons Ltd, The Plant Journal, (2019), 100, 661–676