1.0 Introduction

1.1 Root mucilage definition

Roots excrete small and high molecular weight compounds or exudate through the cell walls of their root caps initially thought to lubricate their root soil interface, thus easing soil penetration (Watt et al. 1993; Bardi et al. 2009). Within this exudate is a battery of low molecular compounds including: amino acids, extracellular DNA, proteins, organic acids, sterols, fatty acids and phytochemicals (Carvalhais et al. 2011). These compounds contained in exudate have been implicated in numerous properties and roles including antibacterial activities, neutrophil-like extracellular traps, signalling and avoidance, nutrient acquisition, heavy metal hyper-accumulation including zinc and aluminium and regulation of fungal composition and diversity (Baetz et al. 2014; Broeckling et al. 2007). It has been suggested that 29% to 40% of a plant’s photosynthetic product is excreted due to this exudate; thus far no explanation has been provided for this (Narasimhan et al. 2003; Lynch and Whipps 1990; McNear 2013; Bais et al. 2006). Of particular interest is the high weight molecule constituent or the polysaccharides of this exudate known as mucilage. It appears that mucilage plays an important role in roots, possibly initiating fungi symbiosis; however, little research has occurred to determine the exact purpose mucilage plays in roots, the polysaccharide content of mucilage and how it may alter to different nutrient deficiencies. A key study revealed that numerous species are embedded within a mucilaginous layer excreted by the root cap and epidermal layers behind the caps (Bacic et al. 1986; Figure 1). It is estimated that 20% to 25% of the total carbon fixed by Zea mays is secreted as high molecular weight molecules in mucilage (Chadboud 1984), though this figure is unknown for other species.

Figure 1 Known areas of mucilage secretion. (A) Illustrates probable locations of mucilage production which range from the zone of maturation to the apical meristem, gradually increasing coverage of the root’s epidermis with mucilage. Mucilage can cover root caps by up to 10 mm. (B) A cross section of the root where mucilage is apoplastically transported from the cortex to the epidermis, adapted from (Oades 1978; Siyavula 2012).

1.2 Basic structure of cell walls

A typical plant cell wall is constructed from long-chains of interconnecting monosaccharide or sugars forming tough cellulose microfibrils. This produces the tensile strength and rigidity that the cell wall requires (Albersheim et al. 2010; Alberts et al. 2008). These cellulose microfibrils are linked to a pectin-rich matrix of cross-linking polysaccharides through cellulose tethers (Smith 1977). This matrix serves to link or separate these microfibrils provide mechanical support, cell adhesion and spacing (Cassab 1998). Primary cell walls, present in all plants, provide tensile strength preventing the cell from bursting (Ma et al. 2009). Secondary cell walls mostly only present in dicots and in the order Commelinoid of monocots contain high levels of lignin providing compressive strength which prevents the cell from collapsing (Albersheim et al. 2010).

1.3 Dicotyledon and monocotyledon cell wall structure

The differences in cell wall composition of dicotyledons (dicots) and monocotyledons (monocots) have long been known. Dicotyledons comprise of both primary and secondary cell walls which contain elevated levels of xyloglucan and pectin compared to monocots (McNeil et al. 1984). Within Commelinoid monocots xyloglucan and pectin presence is low with the main constituent being glucuronoarabinoxylan, the second most abundant hemicellulose in primary cell walls (Vogel 2008). It has been indicated that both dicots and Commelinoid monocots secrete comparable levels of polysaccharides particularly xyloglucan (Morel et al. 1986). This is puzzling considering their differences in cell wall polysaccharide composition. Polysaccharides are produced by the rough endoplasmic reticulum and are then packaged into vesicles by Golgi bodies. These vesicles then follow through the plant’s secretory pathway. Vesicles then merge with the plasma membrane and are excreted by a plasma membrane transporter into the cell wall. The polysaccharides are then apoplastically secreted into the surrounding soil interface, through the cell wall (Northcote and Pickett-Heaps 1966; Smith and Smith 1990).

1.4 Root polysaccharide specific antibodies

mAbs are highly specific to their targeted antigens making them desirable tags for molecules. Polysaccharide specific mAbs have recently been developed and have been widely adopted for cell wall research (for reviews see Carpita 1996; Pattahil et al. 2010). However, mAbs have not been adopted to understand mucilage’s constituent parts and dynamics as the majority of mucilage research was undertaken prior to mAb development. Gas chromatography, mass spectrometry and methylation experiments had determined the major polysaccharides of mucilage that were xyloglucan, mannan, pectin and AGP. Unlike their constituent molecules, monosaccharides, polysaccharides are diverse having numerous properties only differing in their hydroxyl frequency and position (Kamerling 2007; Robyt 1998), making them of interest to research. There is a wide array of mAbs available to confidently identify these polysaccharides. The most effective mAbs available for these polysaccharides are LM14 which specifically target glycans of AGP, LM19 targets partially-methylated or de-esterified pectins, LM21 targets hetreomannans and LM25 targets galactosylated xyloglucans.

2.0 Background

2.1 Root mucilage structural analysis

A key study determined that Zea mays mucilage was composed of arabinogalactan proteins, xyloglucan, arabinogalactan, glycan and pectin (Bacic et al. 1986). Using gas chromatography, mucilage was formed 94% (w/v) of polysaccharides, 6% proteins and trace amounts of phenolic acids (0.17%; Bacic et al. 1986; Read and Gregory 1997). Researchers undertook acid hydrolysis revealing that fucose, arabinose, xylose, galactose and glucose were major components of mucilage. Trace amounts of mannose and starch were also detected in mucilage, though starch levels were highly variable within the investigation (Bacic et al. 1986). Starch is also found to be released from lysed root cap cells after α-amylase treatment. Low levels of uronic acid were detected indicating low proportions of pectic polysaccharides (Bacic et al. 1986). This result was surprising as pectin is abundant within primary cell walls of both dicots and monocots. It was suggested that monocots and dicots were both expelling similar cellulosic polymers enclosed in hydrophilic uronic acids containing pectic-like substances (Wright 1975). High levels of non-sterilise mucilage originated from cell wall polymers in the preparation rather that from the mucilage itself (Bacic et al. 1986; McCulley and Boyer 1997).

Low levels of pectin cause the majority of the gelatinous properties of the mucilage which does not preclude gel formation capacity in other species such as Melastoma malabathricum (Watanabe et al. 2008). AGPs can form gel and non-cellulose polysaccharides can interact with AGPs forming highly viscous gelatinous solutions (Oades 1978). Low levels of glycosyl diversity in mucilage not only provides lubrication for the root cap but it has been suggest that mucilage may help to establish fungi interactions (Cannesan et al. 2012). Low levels of uronic acids in mucilage are characteristic of monocots indicating low presence of pectins (Bacic et al. 1986). Other studies have observed that low uronic acids are also present within Arabidopsis thaliana contrasting the understanding that dicot roots contain high levels of uronic acids indicating pectins (Chaboud and Rougier 1984).

β-glycosyl Yariv reagent has a strong affinity to AGPs which are present within mucilage. By supplementing β-glycosyl Yariv reagent to Z. mays cryostat sections experimenters could determine where mucilage was being excreted. This binding was restricted to the interface between a thick outer periclinal epidermal cell wall and mucilage (Bacic et al. 1986). Mucilage therefore was determined to be excreted from root caps extending to the root’s zone of maturation. Other studies have suggested mucilage is only secreted at root caps (Walker et al. 2003). Α-L-arabinofuranosyl-directed antibody was also added to the sections and was bound to inner and outer regions of the outer periclinal epidermal cells within the zone of elongation. Nonetheless, weak binding was identified within the anticlinal inner periclinal walls of the epidermis, cortical cell walls and root cap (Walker et al. 2003).

2.2 Comparison of Dicots and monocots root mucilage

A comparative study revealed that Vigna unguiculata mucilage was composed of 86% polysaccharides, whereas, Triticum aestivum (Commelinoid) was composed of 91.5% polysaccharides primarily: AGP, xyloglucan, mannose and pectin (Moody et al. 1988). Through methylation analyses amino acid composition between both dicots and monocots were found to be similar (Morel 1986; Popper and Fry 2003). For T. aestivum, arabinose, xylose, glucose and galactose formed major components of its mucilage, whereas, in V. unguiculata arabinose, galactose and xylose were the major components. V. unguiculata contained significantly higher amounts of uronic acids dissimilar to T. aestivum, indicating low levels of pectin (Carpita 1996). On further investigation, monosaccharide constituents of V. unguiculata were fucose, xylose, mannose and rhanose. T. aestivum and Oryza sativa contained comparable levels of xylose, arabinose, galactose with lower levels of fucose, mannose and trace amounts of rhamnose. Small concentrations of starch were detected but experimenters postulated that this was due to lysed cells from mucilage collection (Moody et al. 1988; Sealey et al. 1995).

2.3 Location of mucilage secretion

Mucilage is secreted primarily from the root caps with small contributions from epidermal cell walls of the zone of elongation (Figure 5; Moody et al. 1988). As expected, xylose was secreted in lower concentrations in dicots when compared to Commelinoid monocots. This may reflect the low levels of hetero-xyloglucan. Low levels of hetero-glycogen presented maybe a feature of dicot’s cell walls and secretions (Wright 1975; Read and Gregory 1997). Within Triticum, heteroxylans were excreted in higher amounts in dicots when based on relative proportions of 4-linked xylopyranosyl with a low concentration of fucose containing polymers (Clarke et al. 1979; Chaboud 1983). Nevertheless, Triticum secretions contained small amounts of pectic arabinan; deduced from the presence of 2,3,5, 5 and 3-linked arabinofuranosyl residues. V. unguiculata mucilage differs to Triticum aestivum in composition by containing higher amounts of arabinan, rhamnogalacturonan, pectic polysaccharides with the addition of various types of glycans (Moody et al. 1988). Glucuronic acid is major acidic monosaccharide of Z. mays and accounts for 24% of Triticum mucilage (Gollner et al. 2010). This observation may be associated with levels of heteroxylans such as glucuronoarabinoxylan and AGPs. Triticum mucilage contained 76% acidic monosaccharides for example galacturonic acid signifying associations with AGPs or low concentrations of pectins. This implicates that root mucilage contains polysaccharides analogous to cell wall preparations (Moody et al. 1988). The cell walls of dicots typically contain high levels of pectins where monocots, including graminaceous species contain low levels of pectins. These are composed of high concentrations of glucuronoarabinoxylans (Sterling et al. 2006). The authors verified that AGPs have a high water binding capacity reflecting their role as a gelling and anti-desiccant agent. It is probable that other polymers in mucilage interact non-covalently forming continuous networks which immobilise gel and water (McCully 1999; Knee et al. 2001).

Figure 2 Areas where mucilage excretion occurs. (A-D) Mucilage can form droplets surrounding the root cap and extend along the apical meristem. Solute concentration was 0.5 mg mL^-1. (A) These images were taken from Z. mays after 3-4 days of growth. (B) Image taken from L. angustifolius roots. (C) Illustrates mucilage’s hydration effect on Z. mays roots which indicates the presence of surfactants. (D) Mucilage forming stable membrane bridges on the root caps of Z. mays indicating high cohesion properties of mucilage (Read and Gregory 1997). (E) Hair root taken from Lysinema ciliatum illustrating high levels of mucilage excretion from epidermal cells. Arrows indicate where the root was held within the soil (McCully 1999). (F-H) Shows a time series of mucilage forming a droplet surrounding the root cap of Z. mays and surrounding the epidermal layers (m; mucilage, rc; root cap; McCully and Sealey 1996), white lines represent 50 µm.

This would determine the capacity for specific interaction locations between polysaccharides on the root surface and micro-organisms seeking symbiosis with plants (Moody et al. 1988). The major conclusions of this research and others were that the xyloglucans were secreted in similar levels in dicots and Commelinoids, though xyloglucans are only major components within monocots and dicots. Xyloglucans are present in trace amounts in primary cell walls of Commelinoids (Roy et al. 2002; Geshi et al. 2013). Cell wall analyses of coleoptiles contain differing levels of xyloglucans in monocots and dicots (Iijima et al. 2004; Chaboud and Rougier 1984).

2.4 Alterations in mucilage polysaccharide derivatives

Gel filtration chromatography revealed that roots of L. sativum released: galactose, arabinose, glucose which was found to be comparable to Z. mays. Fucose formed from 6-deoxyhexose was 4% in L. sativum and was 12% higher in Z. mays (Ray et al. 1988). The protein content was 25% which is significantly higher compared to Z. mays (3%) and other reported cases which contrast with results of other research. Conversely, levels of protein in Z. mays have been recorded as high as 30% (Hinch and Clarke 1980; Kato and Nevins 1984). As a comparison, rhamnose was 5.8% higher in T. aestivum whereas, in Lepidium arabinose was 1.25%, xylose was 1.2%, mannose was 3.7% and galactose was 7.3% higher in all within unfractionated mucilage samples. A constituent of arabinogalactan, hydroxyproline was in low concentrations within L. sativum. Amino acid analysis determined that glutamine, asparagine, serine and lysine were abundant in L. sativum mucilage (Ray et al. 1988; Fry et al. 1993; Fry 1998). Eluting the unfractionated polysaccharide solutions for ion exchange chromatography produced an unbound fraction of polysaccharides which represented 28% of total carbohydrate with a reduction in uronic acid. Further eluting the samples using a salt gradient uncovered a peak of polysaccharide at 15% with enriched uronic acid. The use of a shallower gradient suggested that the polysaccharide content was heterogeneous (Ray et al. 1988; Osborn et al. 1999). The total recovery of carbohydrates was 60% and uronic acid was 40%.

Polysaccharide heterogeneity was echoed by the variation within the monosaccharide content. High molecular weight material was enriched with galactose and lower weight materials contained decreasing levels of galactose in correlation with: xylose, arabinose and mannose (Ray et al. 1988; Narasimhan et al. 2003). Collecting mucilage from L. sativum was not undertaken due to difficulties in working with the plant’s root size. Moreover, mucilage from dicots does not form droplets unlike Z. mays mucilage (Figure 5 E and 6; Miller and Moore 1990). SDS-PAGE analysis of mucilage revealed several bands ranging between 29 to 200,000 Da. It remains unclear whether these proteins are structurally important to mucilage. It is possible that these proteins could be extraneous proteins from cut or damaged root caps (Ray et al. 1988). Fractionation and ion exchange chromatography was unable to separate these proteins from polysaccharides within secretions. Recently, protease has successfully removed these proteins without interfering with the polysaccharide’s molecular weight (Ray et al. 1988). L. sativum contained trace levels of hydroxyproline, suggesting that AGPs were not major components of its mucilage. This finding specifies that AGPs may not cause aggregation of the polysaccharide components in Lepidium mucilage (Nguema-Ona et al. 2013; Gerwig et al. 1978).

Figure 3 Sections of Z. mays roots imaged using epifluorescence microscopy. (10) Details of root cap which demonstrates portions of epidermal cells. Mucilage is present in the periplasmic space (arrows) with the excreted mucilage (em) leaving a pore between two epidermal cells (x 2,600). (11) Highlights secretory cells producing cytoplasmic mucilage which can be seen as dark regions between the cells (x 320). (12) Periplasmic pockets of mucilage as seen in the whitened regions (x 320). (13) The periplasmic mucilage is uniformly fluorescing (x 1320). (14) The same image in (12) but at a higher magnification highlighting areas of mucilage section out of the cells, grazed cell wall; w (x 1320). (15-16) Pockets of mucilage within the cells of the root caps having a fibril appearance (x 100; McCully and Sealey 1996).

In the fractions of L. sativum, low levels of uronic acid was uncovered suggesting the presence of pectin and hemicellulose polymers. Furthermore, the high levels of uronic acid in conjunction with rhamnose recorder suggest the presence of pectic acids (Ray et al. 1988). Rhamnose can interrupt uronic acid backbones with monosaccharides creating side chains (Choct 1997; Chadboud 1984). The experimenters postulated that mucilage containing elevated levels of fucose was being excreted through epidermal cell walls behind the root cap. In opposition to the finding that dicot and monocot mucilage comprises of high levels of 1,4,β-D-glycan, the study demonstrated using α-amylase treatment that there were no levels of 1,4,β-D-glycan in L. sativum mucilage (Figure 6). Glucose was a major component of the mucilage indicating the presence of glycan (Wright and Northcote 1974). Lectin binding studies suggest that mucilage inhibits agglutination of erythrocytes (Rougier et al. 1979; Gooch 2011). L-fucose binding proteins from Lotus tetragonolobus and Ulex europaeus and D-glucose and D-mannose could bind to lectin from Arachis hypogaea (Alkafafy et al. 2012). Some studies suggest that the same lectins and mucilage have a weak affinity, indicating that in vivo galactose residues could be inhibited by the conformation of the heterogeneous polysaccharide’s present in mucilage (Dick-Perez et al. 2011). L. sativum mucilage was composed of uronic acid (48%) indicating that mucilage was highly charged and capable of aggregating and forming cross-links through cations. Mucilage can form layers during histological and electron microscopy investigations verifying mucilage’s autonomous aggregation (Miki et al. 1980; Voinicius et al. 2013).

2.5 Border cells

One additional feature increasing production and dispersal of mucilage to several millimetres at root caps are border cells. These aggregated cells have broken off from the root cap and continue to proliferate (Miyaska and Hawes 2001). Border cells have been implicated in antibacterial defence of root caps (Baetz and Martinoia 2014). Remarkably, border cells even produce extracellular DNA into mucilage, trapping bacteria and preventing root cap infection. These cells can excrete a mixture of antimicrobial proteins, antioxidants, AGPs and pectins into the mucilage matrix (Knee et al. 2001). In response to low levels of nitrates, phosphates and secondary metabolites, border cells increase mucilage production at the root caps (Driouich et al. 2013).

2.6 How mucilage can stimulate nutrition uptake

In situ antimony electrodes have demonstrated that mucilage can alter surrounding soil pH when faced with nutrient deficiency through direct chelation as well as indirectly through microbial activities (Marschner et al. 1986). If plants cannot increase nitrate uptake through redox alterations they increase polysaccharide secretions by up to 30% to possibly attract symbiotic partners (Mary et al. 1993). Mucilage can alter pH by differential potentials in cations and anions. Mucilage can also liberate nutrients through dissolution of insoluble minerals in soil where they are released by altering redox potential as a result of calcium. Once nutrients have been released they are reabsorbed by the roots (Marschner et al. 1986). Subjected to low nitrate levels, plants excrete protons in the surrounding soil apoplastically into mucilage to increase overall positive charge so that anions (negatively charge); phosphates or nitrates can be absorbed by roots (Skene et al. 1996; Grinsted et al. 1982), thus soil acidity increases. If the plant requires cations from phosphates, calcium or magnesium (Brown 1978), hydroxyls are excreted to create an overall negative charge to attract cations.

The presence of heavy metals including zinc and aluminium can increase mucilage production. Mucilage can absorb and capture metal ions preventing roots from absorbing harmful ions (Watanabe et al. 2008). The effects of pH and redox potentials were observed from cultured Lupinus alba and T. aestivum in agar and using bromocresol purple as an pH indicator (Marschner et al. 1986). When nitrate and ammonia levels increase in mucilage, pH decreased from 7.5 to 4.5 enabling cations to be absorbed by roots. Pseudotsuga menziesii and Picea abies develop a distinct pH gradient along their root mucilage in accordance with nitrate or ammonia availability. In low pH (4.5) the soil surrounding the root caps and epidermis behind the root caps can increase up to pH 5.4 to chelate nitrogen (Jentschke and Godbold 2000; Hinsinger et al. 2009).

Iron deficiency in dicots and monocots can lead to enhanced proton secretion within the root cap (Romheld and Marschner 1981), though this remains to be determined for other macronutrients. In Phleum pratense, low molecular compounds are released undertaking phosphate mobilisation from sparingly soluble iron or aluminium phosphates through citric or malic acid in mucilage of clustered roots. Using Trifolium pratense and Lolium perenne researchers uncovered that high phosphorous intake by roots was neither caused by high root growth or high root to shoot ratio but correlated with increases in mucilage with elevated levels of protons (Gahoonia and Nielsen 1998).

2.7 Research Aims

Dicots and Commelinoids may produce similar mucilage from their cell walls even though their cell wall components are completely different, diverging over 140 million years ago (Chaw et al. 2004). Xyloglucan is a minor component of Commelinoid cell walls though it is secreted in high amounts in monocot mucilage (Lynch and Staehelin 1995). However, the reasons underlining this similarity remain unknown. Collecting and analysing mucilage from A. thaliana and T. aestivum ‘Cadenza’ through epitope detection chromatography allows a precise measure of the species of polysaccharide present. Additionally, very little is known of the effects on mucilage when macronutrients are individually removed. To elucidate this gap of understanding, plants were axenically grown on medias lacking a source of nitrogen, potassium and phosphorous. A range of mAbs were utilised to individually tag the major polysaccharides: AGPs, mannan, pectins and xyloglucan to act as biomarkers in mucilage. Elucidating these questions may develop a new, more effective means for plants to attract and regulate their fungi symbiosis, thus developing a more effective means of crop fertilising.

· Determine polysaccharide composition of Arabidopsis thaliana and Triticum aestivum mucilage through mAb epitope detection grown in full nutrient media.

· Determine polysaccharide content alterations in mucilage from A. thaliana and T. aestivum seedlings grown without nitrate, phosphate and potassium.

3.0 Materials and Methods

3.1 Plant and media preparation

To examine the effects of excluding nitrogen, potassium and phosphorous on mucilage, GB11 media at pH 5.6 was modified and stored at 4 ºC (Rippka et al. 1979). GB11 was modified by excluding NaNO₃ for the nitrogen deficient group and K₂HPO₄ 3H₂O was removed for both the potassium and phosphorus groups. To compensate for alteration in the molarity of removing K₂HPO₄ 3H₂O, Na₂HPO₄ (Sigma-Aldrich, Munich, Germany) at 6.21g in 250 mL of dH₂O (stock solution) for the potassium deficient group and CH₃CO₂K (Sigma-Aldrich, Munich, Germany) at 1.7g in 250 mL of dH₂O (stock solution) for the phosphorous deficient group. As a control, plants were grown in unmodified BG11 media. For each group 5 plates were assigned with 5 seeds placed on to the agar’s surface. Triticum aestivum ‘Cadenza’ (Knox Laboratory, University of Leeds) was grown in liquid GB11 media excluding plant agar and sucrose. Wild type Arabidopsis thaliana (Knox Laboratory, University of Leeds) was cultured in BG11 media with plant agar (0.9%) and sucrose (10 g/L). For A. thaliana, 50 mL of BG11 media at 60 ºC was added to each 100 mL square petri dishes as described (Xu et al. 2013). For T. aestivum, 100 mL of liquid BG11 was added to 250 mL conical flasks. For each group 2 250 mL conical flasks were assigned with 20 seeds.

Up to 1,000 A. thaliana seeds were added to an Eppendorf tube with 900 µL of dH₂O and 100 µL of household bleach. This mixture was left for 20 minutes on a rocker. Subsequently, 6 washes occurred within an Airstream laminar air flow cabinet (Esco, Rotherham, UK) using dH₂O, seeds were placed on sterile filter paper. Using sterile tweezers, 5 seeds were placed onto the agar on each plate. Plates were then placed in 4 ºC storage for 2 days. After storage, A. thaliana were grown in a plant growth room at 20 ºC with a day length of 12 hours for 14 days. A. thaliana were 5 cm from the light source, evenly spaced and turned daily to ensure even levels of light reaching each plate. To sterilise T. aestivum, 20 seeds were placed into a 15 mL falcon tube with 1 mL of household bleach and 9 mL of dH₂O, 6 washes using dH₂O followed. Seeds were placed into 100 mL of liquid BG11 media and incubated within an Orbi-Safe environmental shaker (Sanyo, Munich, Germany) at 22 ºC and at 110 rpm with a day length of 12 hours for 8 days.

3.2 Analysis of root growth

To investigate the effects on root growth, A. thaliana and T. aestivum root length, root architecture and fresh weight were recorded. Root length was measured using a digital calliper (Fisher Scientific, Roskilde, Denmark) by placing each end of the calliper at the top of the root near the leaves and at the root cap. An Explorer Pro (Ohaus, Parsippany, New Jersey, US) balance was used to measure the fresh weight of the seedlings. pH was measured using a Seven Compact meter to 3 decimal places (Mettler Toledo, Leicester, UK). The balances were zeroed once every group was weighed using a cleaned tray. An 8 mega pixel camera with an exposure setting of 1.0 on a black background was used to capture images of root architectures of both species.

3.3 Polysaccharide extraction and mAbs

The agar was cut into squares using 4 cm x 2 cm card templates. Subsequently, the cut agar was pummelled and left overnight on a Platform STR 8 rocker at 30 Rev/min (Stuart Scientific, Villepinte, France) with 12 mL of sterile water to extract the polysaccharides from A. thaliana media. After 1 hour the water from each tube was removed and stored at -20ºC. Subsequently, more water was added and 24 hours. After 24 hours extraction, the liquid was stored at 4 ºC prior to analysis. Once cut and pummelled, liquid was removed, filtered using 20 µm mesh laboratory grade sieve (Retsch, Haan, Germany) to prevent agar contamination abd placed into a fresh 15 mL falcon tube. A lyophiliser LyPro 6000 (Heto, Allerod, Denmark) was used to freeze dry and concentrate the signal from the 1 hour, overnight A. thaliana samples and T. aestivum samples. The freeze drying process was left overnight. Once powder was formed, 1.8 mL dH₂O was added with 200 µL of x10 PBS. The liquid was then stored at -20 ºC until ELISA and EDC analyses. The liquid from the polysaccharides mixture from T. aestivum media was filtered using the same laboratory sieve and placed into storage at -20ºC. Aliquots (1 mL) were used per microtiter plate. Initial screening using 20 mAbs listed in Table 2 was used to determine the most prominent candidates to use for the RE printing, ELISA and EDC analyses. All antibodies used were sourced from the Knox Cell Wall Laboratory (University of Leeds, UK).

3.4 Root exudate printing

A piece of nitrocellulose membrane was laid on top of 14 day old agar with seedlings removed and left overnight at 4ºC. The nitrocellulose was blocked with 5% milk powder in 1 mL x10 phosphate buffer solution, dH₂O (4 mL) and sodium azide (10 µL) left for 1 hour. Each primary antibody was added at 1 in 10 dilution in 5% milk powder and PBS for 1 hour at room temperature while shaking using a Platform STR 8 rocker at 30 Rev/min (Stuart Scientific, Villepinte, France). Membrane was washed with tap water 3 times and left for 5 minutes per wash. Horseradish peroxidase was added at 1 in 1,000 dilution to 5% of milk power and PBS for 1 hour at room temperature while shaking. Membrane washing was repeated. Colour reagent was formed from 25 mL of dH₂O, 5 mL of chloronaphthol solution (5 mg/mL) and 30 µL of hydrogen peroxide. The nitrocellulose was then placed in-between blotting paper, left overnight and visualised using GeneSys v.1.4.1.0 (SynGene, Cambridge, UK) with an exposure time of 750 ms. An Epson SX400 (Epson, London, UK) flatbed scanner was also used to image prints that the GeneSys system could not reflect that was presented on the nitrocellulose sheets. A contrast ratio of 168:71 was used for each sheet.

3.5 Enzyme-linked immuno-sorbent assay

Plates were initially coated with polysaccharide at a neat concentration (1.8 mL of mucilage sample and 200 µL x10 PBS) at 125 µL per well at the top row only with the other wells being filled with 100 µL of x1 PBS. From the top, neat wells were titrated moving down each row using a 1 in 5 dilution, each step taking 25 µL resulting in dilutions of 1:5, 1:25, 1:125, 1:625, 1:3,125 and 1:15,625. Once the H row had been reached 25 µL of solution was removed leaving the bottom row to act as a control without an antigen. This was then left overnight covered with tin foil. Microtitre plates (Thermo Fisher Scientific, Roskilde, Denmark) were washed 3 times using a 200M HT microplate washer (Titertek Berthold, Pforzheim, Germany) and dried by knocking the plate against paper towels. The plates were blocked with 5% milk powder in PBS and NaN₃ (0.025%). Aliquots (300 µL) of this elution was added to each well and incubated for 2 hours at room temperature. Plates were washed 9 times and shaken dry. Each primary antibody was added to 4 columns of the plates. One in 10 dilution of primary antibodies (1 mL per antibody) were added in 5% milk powder and PBS, 100 µL of this solution was added per well. The plates were incubated for a further hour at room temperature and covered using tin foil. Plates were washed a further 9 times and shaken dry. Anti-rat IgG-HRP secondary antibody was added to the wells at 1 in 1,000 dilution in 5% milk powder and PBS (100 µL per well) and incubated at room temperature for 1 hour and covered using tin foil. Plates were washed 9 times and shaken dry. The following substrate was added at 150 µL per well; 18 mL dH₂O, 2 mL sodium acetate buffer (1 M) at pH 6.0, 200 µL tetramethyl benzidine (10 mg/mL in DMSO) and 20 µL 6% (v/v) hydrogen peroxide. Subsequently, 30 µL per well of 2.5 M sulphuric acid was added to each well and then read 450 nm on MultiSkan FC spectrometer (Thermo Scientific, Roskilde, Denmark).

3.6 Epitope detection chromatography

Aliquots (1.8 mL) of the A. thaliana and T. aestivum media extracts were diluted in 2.5 mL of dH₂O were injected into a weak anion exchange chromatography column (1 mL Hi-trap ANX FF, GE Healthcare, UK) using a BioLogic LP system (Biorad, Hertfordshire, UK). Samples were eluted to 1 mL per minute with sodium acetate buffer (20 mM) at pH 4.5 from 0 to 17 minutes. A step alteration to 50 mM of sodium acetate (pH 4.5) for a further 17 minutes with a linear gradient from 0% to 100% with NaCI (0.6 M) to 48 minutes which was proceeded by 8 minutes at 50 mM acetate with 600 mM NaCI (elution sample 1) (Cornuault et al. 2014). Forty eight 1 mL fractions were retained. Injection loop (5 mL) and columns were washed with 10 mL of NaOH (0.1 N) between sample injections. Columns were then equilibrated and injection loops washed with 10 mL of acetate buffer (20 mM) before the next injection. A 2 step salt gradient was used to fractionate xyloglucan sub-populations (Cornuault et al. 2014). Xyloglucan samples were eluted with sodium acetate (20 mM) buffer at pH 4.5 from 0 to 25 minutes with a change to 50 mM sodium acetate buffer (pH 4.5) at 25 minutes with a linear gradient from 0% to 50% NaCI (600 mM) to 75 minutes which was proceeded by 50% to 100% of NaCI (600 mM) to 88 minutes, further proceeded by NaCI (600 mM) to 96 minutes (elution sample 2). The fractions were adjusted to pH 7.0 by supplementing 50 µL Na₂CO₃ (1 M) and 100 µL aliquots were incubated in microtitre plate wells overnight at 4 ºC for subsequent ELISA protocol (Cornuault et al. 2014).

3.7 Statistical analysis

The root lengths and fresh weights of A. thaliana and T. aestivum were analysed using Minitab 16 statistical software (Minitab Ltd, Coventry, UK). Descriptive data were analysed initially followed by the normal distribution using the Anderson-Darling test and equal variance using the Levene’s test. This led to the One-way Independent ANOVA or its non-parametric equivalent the Krusal-wallis test which examined the p-values relative to their control group. To determine where significance differences laypost-hoc Tukey or Mann-Whitney U tests were used. For the ELISA data all neat concentrations were used for statistical analysis. This was preceded by Mann-Whitey U tests to analyse the differences within the groups. For ELISA and EDC data, Excel 2013 (Microsoft, Washington, US) was used to gain means and standard errors which were then transferred to the GraphPad to generate the figures (GraphPad Ink, California, US).

4.0 Results

4.1 Xyloglucan, AGP, pectin and mannan produced the strongest signals in both plants

To investigate which mAbs had the highest affinity to root polysaccharide secretion, 20 mAbs targeting plant polysaccharides were chosen in a wide screen of both plants (Figure 4). LM5 targeted (1→4)-β-D-galactan, LM7 targeted non-blockwise partially methylated homogalacturonan, JIM7 and LM20 targeted partially methylated homogalacturonan, LM8 targeted xylogalacturonan, LM11 targeted (1→4)-β-D-xylan and arabinoxylans, LM12 targeted feruloylated xylan and pectin, LM13 targeted linearised (1→5)-α-L-arabinan, JIM15 and LM14 targeted AGP glycan, LM15 targeted XXXG motif of xyloglucan, LM16 targeted processed arabinan, JIM18 targeted glycophospholipid, LM18 and LM19 targeted partially methylated and de-esterified homogalacturonan, LM21 and LM22 targeted hetreomannan, LM24 and LM25 targeted galactosylated xyloglucan. Of these antibodies screened LM14, LM19, LM21 and LM25 had the highest and most consistent signals in both species (Figure 4). LM14’s signal appeared similar in both plants (0.454 au and 0.4378 au) as well as LM25 (1.481 au and 1.853 au). LM19’s signal was higher within A. thaliana (1.125 au neat) when compared to T. aestivum (0.240 au neat). This pattern was repeated for LM21 having a 76.5% reduction in signal in T. aestivum. LM11 had the third highest signal (1.356 au neat) within T. aestivum, however, it did not appear in A. thaliana and it is for this reason why LM11 was not selected (Figure 4).

Figure 4 Screening of 20 antibodies targeting a range of epitopes demonstrating the highest signals which were derived from LM14, LM19, LM21 and LM25 in A. thaliana’s mucilage (A). This was repeated for the T. aestivum control sample. One in five dilution steps were used to titrate the exudate extraction of both plants. (B) LM25 seems to have the highest signal in both plants closely followed by LM21, LM19 and LM14, though LM25 and LM21 signals overlap in Triticum. Signals from Arabidopsis appear to be slightly higher than in Triticum.

4.2 Nutrient deficiencies decrease Arabidopsis root length but slightly increase Triticum root length

To determine the wider consequences of excluding each nutrient, root lengths were recorded to examine the variations caused by each deficiency. When A. thaliana was grown in full nutrition, root lengths were 29.8% longer compared to –N, 58.6% longer compared to -K and 41.9% longer compared to -P. A. thaliana subjected to –K exhibited stunted growth unlike under other conditions (Figure 5, left). The Anderson-Darling test revealed that –N (P= 0.274) and -K (P= 0.323) had normality. The Krusal-Wallis (H=42.2, P= <0.05) test determined that there was a significant difference between the groups. A Mann-Whitney U test uncovered that roots elongated more rapidly when subjected to full nutrition (W=5604.0, P= <0.05). Roots grew significantly shorter in –N and –K compared to full nutrition (W=4816.0, P= 0.015). However, there were no significant differences between the nutrient deficiencies (W=3974.5, P= 0.0794).

When excluding each nutrient T. aestivum root length increased. When grown in full nutrition length was on average 52.5 mm which was 1.5% smaller compared to –N and –K but was even shorter compared to –P (11.1%; Figure 5, right). The roots elongated without each nutrient but their roots became slimmer and brittle. Anderson-Darling demonstrated that there was no normality in the conditions (P= <0.05). Kruskal-Wallis (H=7.67, P= 0.053) uncovered that T. aestivum subjected to –P grew significantly longer than in other conditions. Mann-Whitney U revealed that when subjected to -P roots grew significantly greater (P= 0.026). There were no statistical differences between –N and –K (W=1771.0, P= 0.977) but –P roots grew longer compared to the other groups (P= 0.02).

Figure 5 Summary of the root lengths of both A. thaliana taken after 2 weeks and T. aestivum taken after 8 days. The yield of A. thaliana was on average 88.3% and for T. aestivum average yield was 56.5%. From the charts there was a difference between –K and the other conditions in A. thaliana (left) and –P and other conditions in A. thaliana (n=56, 52, 51, 53). When T. aestivum was subjected to a lack of phosphorous root length significantly increased (right). When A. thaliana was subjected to a lack of potassium root length significantly decreased (n= 45, 42, 48, 45).

4.3 Nutrient deficiencies dramatically reduce Arabidopsis fresh weight but increase the weight of Triticum

To investigate root strength, fresh weights were measured to determine if excluding each macronutrient would impact upon the root’s capacity to absorb water. As expected A. thaliana grown in full BG11 media was 45.1% heavier compared to –N, 45.5% heavier than –K and 85.5% heavier than –P (Figure 6, left). Anderson-Darling uncovered no normality in the conditions (P= <0.05), indicating high variation within the weights. Kruskal-Wallis (H=105.66, P= <0.05) revealed a significant difference between the conditions. Mann-Whitney U (W=559.5, P= <0.05) revealed that there were significant differences between plants grown in full nutrition and the nutrient deficient groups. There were significant differences between each nutrient deficiency with –P demonstrating the lowest weights in the conditions (W=6028.5, P= <0.05).

Subjected to full nutrition, T. aestivum remained 6.4% heavier compared to –N, however, weight was 30.3% less compared to –K and 47.3% less compared to –P.–P was the heaviest condition which was 56.7% heavier than –N and 13% than –K (Figure 6, right). This observation appears to be in reverse to A. thaliana. Anderson-Darling uncovered that there was no normality in the conditions (P= <0.05). Krusal-Wallis (H=8.50, P= 0.037) uncovered a significant difference between the groups. Mann-Whitney U determined that significant differences lay between the control and -P, -N and -P and -K and –P (W=1722.5, P= >0.05). There were no significant differences between plants grown in full nutrition, -N and -K (W=2088.5, P= >0.05).

Figure 6 Summary of the fresh weights of A. thaliana and T. aestivum. The yield of A. thaliana was on average 88.3% and for T. aestivum was 56.5%. The control had the heaviest roots within the Arabidopsis culture. -P had the lowest weight roots. Of the nutrient deficient groups -K had the heaviest weight (Left; n=56, 52, 51, 53). To contrast, the control for T. aestivum had the lowest weight whereas -P had the heaviest weight. -N had the lightest weight compared to the control for T. aestivum (n= 45, 42, 48, 45).

4.4 Xyloglucan significantly increases with -N in Arabidopsis whereas xyloglucan significantly increases in all nutrient deficiencies in Triticum

The epitopes of the root polysaccharides were monitored from the agar after 1 hour and 24 hours of diffusion in water for Arabidopsis. The signals retrieved from the 1 hour gel fractions were mostly in low concentrations (Figure 7). This suggested that the polysaccharides required more time to diffuse. When comparing the diffusion of both 1 and 24 hours gel extraction samples, there was a significant increase in concentration. LM14’s signal increased from 0.154 au to 0.453 au, LM19 increased from 0.094 au to 1.318 au, LM21 increased from 0.131 au to 1.654 au and for LM25’s signal, it increased from 0.878 au to 1.852 au.

Figure 7 Epitopes detected in each group after 1 hour of gel extraction. Data taken from 3 biological replicates. Top left illustrates epitope detection within the presence of full nutrient BG11 media (n=56). The highest point represents LM25 which binds to xyloglucan and was taken from a neat concentration (125 µL). This point rapidly decreases after the first five-fold dilution. A. thaliana grown without nitrogen; all epitopes are detected at low concentrations which blend into the background (n=42; top right). A. thaliana grown without potassium; epitopes are detected in low concentrations which blend into the background (n=51; bottom left). A. thaliana grown without phosphorous and epitopes that are detected blending into the background (n=45; bottom right).

After 24 hours of diffusion the mAb signals increased and enabled composition analysis of the root polysaccharides (Figure 8). ELISAs allowed a quantitative composition analysis of each condition’s polysaccharide combination and the alterations in relation to each nutrient deficiency. LM14’s signal was 95.8% lower compared to plants grown in full nutrition, decreasing below 0.110 au. Anderson-Darling uncovered that there was normality within the data (P= >0.05). The Levene’s test demonstrated that the data had no equal variance. Kruskal-Wallis (H=11.33, P= 0.010) uncovered that a significant difference was present. Mann-Whitney U revealed that -P had a significantly reduced absorbance compared to the control (W=24.0, P= 0.019) and the other nutrient deficiency regimes (W= 25.0, P= 0.012).

Figure 9 Alterations within epitopes detected in the liquid media of T. aestivum. Data taken from 3 biological replicates. LM25 had the highest peak which is highlighted within T. aestivum. Variation of LM25 increased within plants lacking each essential nutrient. Within the present of full nutrient BG11 media (top left; n=45) all signals blended into the background at the 1 in 3,125 dilution, however, this was not apparent for the other groups. When plants were without nitrogen the LM25 signal increased and had a sharper decline until the 1 in 625 dilution (top right; n=42). When plants were subjected to potassium deficiency LM25 also increased (bottom left; n=48) but remained apparent until the 1 in 15,625 dilution. Variation was greatly increased at the 1 in 3,125 dilution. When subjected to phosphorous deficiency LM25 signal increased (bottom right; n=45). There was also a pattern in the LM14, LM19 and LM21 signals for each of the groups.

Triticum’s LM14 signal was 0.437 au (neat), when compared to the other nutrient regimes there was a 30.8% decrease compared to -N and an 11.5% decrease compared to -K. -P was 11.9% higher compared to the control. LM14 had the lowest signal within -N and the highest in -P (Figure 9). On further analysis, the control (P= 0.543), -K (P= 0.204) and -P (P= 0.325) had normality, -N (P= 0.048) did not. The data did not have equal variance (P= <0.05). Kruskal-Wallis (H= 15.30, P= 0.002) determined that there was a significant alteration between the groups. Mann-Whitney U uncovered that -N was significantly increased compared to the control (W= 54.0, P= 0.020). There was also a significant reduction in -P compared to the control and -N (W= 21.0, P= <0.05). In the presence of full nutrition, LM19’s signal was 0.241 au which was 31.2% higher compared to -N, 2.38% higher compared to -P but 0.83% lower than -K. It appears consistent that LM19 had the lowest and unchanging signal in the groups (Figure 9). The control, -K and -P LM19 followed a similar pattern of signal. Anderson-Darling uncovered that the groups had normal distribution (P= >0.165). The data had equal variance (P= 0.435). The One-way Independent ANOVA (F= 6.48, P= <0.05) test uncovered a significant difference. Tukey’s post-hoc test revealed that -N was reduced compared to the other groups (P= <0.05).

There was a 37.5% decrease in LM21 absorbance when -N was compared to full nutrition. There was also a 12% increase in absorbance in -K and a 26.2% increase in -P compared to the control. LM21’s signal blended with the signals from LM19 and LM14 unlike in Arabidopsis where the signal was the second highest peak (Figure 9, bottom right). Anderson-Darling uncovered that the control (P= 0.446), -P (P= 0.162) and -N (P 0.06) had normality unlike -K (P= 0.01). The data did not have equal variance (P= <0.05). Kruskal-Wallis (H= 10.25, P= 0.01) revealed that there were significant differences between the groups. Mann-Whitney U demonstrated that there were significant differences between -N (W=53.0, P= 0.03) and the other groups and -P (W=21.0, P= <0.05) and the other groups and -N and -P (W=21.0, P= <0.05). There were no significant alterations between the other conditions.

-N LM25’s signal was 29.2% higher compared to plants grown in full nutrition. Furthermore, -K (35.1%) and -P (39.7%) signals were even higher. LM25 signals within the nutrient deficient groups had a sharper decline as the dilutions progressed (Figure 12, top left). -P and -K’s LM25 signals appeared in a similar fashion to the control. -P signal’s decline was steeper and there were elevated signals between 1 in 625 to 1 in 3,125 dilutions (Figure 9, bottom right). There was slightly more variation within the nutrient deficiency groups. The control (P= 0.515), -K (P= 0.199) and -P (P= 0.296) had normality unlike -N (P= 0.01). The data also had equal variance (P= 0.306). The One-way Independent ANOVA (F= 22.20, P= <0.05) highlighted significant alterations between the groups. Tukey’s post-hoc test revealed that all the nutrient regimes had elevated levels of LM25 (P= <0.05). No significant differences lay between the nutrient regimes.

4.5 Xyloglucan increases with nitrogen deficiency along with pectin diffusion

RE prints imaged where the mAbs signals were located along the roots and were diffusing from them. From the RE prints LM14’s signal was concentrated toward the major roots of the plants. Additionally, LM14 was speckled along the major roots, particularly towards the top half of the plant (Figure 10, A). When subjected to -N the LM14’s signal appeared reduced with speckles of signal along the major roots (Figure 11, B). When A. thaliana was subjected to -K and -P trace amounts of LM14 was detected with small dots of signal located towards the bottom of the nitrocellulose, illustrating a reduction (Figure 10, C-D).

The LM19 control group highlights diffusion of the signal as well as the concentrated regions and speckles on the major roots of A. thaliana (Figure 10, E). In -N, diffusion of LM19 appeared to have been prevented. The concentrated regions within the major roots were more spread out through the roots (Figure 10, D). Moreover, LM19’s signal within -N appeared more towards the top of the plants unlike the control, where it appeared more towards to the root ends (Figure 10, F). In -K, LM19’s signal appeared highly concentrated along the major roots with no apparent diffusion (Figure 10, G). When subjected to -P, LM19 diffusion had increased throughout the entire root network with no diffusion at the root caps. Speckles of concentrated LM19 also appeared, following the major roots (Figure 10, H). A. thaliana grown in full BG11 media appeared larger and greener compared to the nutrient deficiencies.

The LM21’s signal within the control appears strong along the major roots with high concentrations located where the leaves and seed testa were previously held in place. LM21 also has a heightened signal toward the majority of root hair ends and 2 root caps (Figure 11, A). LM19 speckles are apparent toward the top of the root network. Within the -N condition, LM21’s signal appears to have been reduced with slight concentrated regions along the major roots (Figure 11, B). The LM21 speckles are no longer apparent on the nitrocellulose. In -K, LM21’s signal is dramatically reduced with only slight indications of LM21 toward the top of the roots with slight speckling toward the root ends (Figure 11, C). In -P plants there was also a reduction. In one plant LM21 appears to have a similar signal as with -N (Figure 11, D). LM21 was also concentrated where the seed testa would have been located.

As expected, LM25 appears more concentrated toward the root caps and the epidermal layers behind the caps. LM25 was also observed along the major roots. However, LM25 is expressed above the cap roots but in lower concentrations (Figure 11, E). Interestingly, when plants were subjected to -N LM25’s signal appears to be concentrated at the root caps, testa and at the top of the roots. LM25’s signal does not appear along the roots in a less concentrated form (Figure 11, F). LM25’s signal is not just present at the root caps and root hairs but more spread out along the root network. In –K, LM25’s signal appears faded along the major roots with no concentrated signals at the root caps but instead just behind them (Figure 11, G). In -P, LM25’s signal is reduced throughout the roots but remains concentrated at the root caps in a similar fashion to the control (Figure 11, H). When subjected to -N, plants appear to have longer roots with more root hairs. However, -N plants have smaller leaves with a yellow hue. -K and -P plants have smaller paper-like root systems with yellow hues on their leaves. -P plants did appear to have the characteristic red hue which is not apparent in Figure 11.

Figure 11 RE prints of A. thaliana using LM21 and LM25 (scale bar 20 mm). To the right of the RE prints are photographs of the plants prior to RE printing. (A) LM21 appears to be spread out through plants grown on full nutrient BG11 media. There was also a highly concentrated signal towards the right-hand corner which may be due to the seed testa. (B) -N plants have a reduced LM21 signal excluding the top left-hand side where the LM21 signal is elevated along the root. (C) The LM21 signal is reduced throughout the root system and is weakly expressed toward the top of the root. (D) When plants are subjected to a lack of phosphorus the LM21 signal is reduced excluding the right-hand side. LM21 is mostly revealed towards the seed testa and is slightly expressed throughout the centre of the root. (E) With plants grown in the presence of full nutrition, the LM25 signal is highly expressed in the root cap and just behind the root caps. (F) A. thaliana grown in -N increases the expression of LM25. The LM25 signal is highly expressed throughout the entire root system not just located toward the root cap. This signal is highly concentrated at both the upper and lower parts of the roots. (G) A. thaliana cultured deficient in potassium showing more evenly distributed LM25 signal. (H) LM25 signal appears toward the upper roots as well as the root caps.

4.6 Xyloglucan is heightened in nutrient deficiencies becoming more neutral in T. aestivum and more acidic in A. thaliana

EDC was undertaken to examine the epitopes present within the nutrient deficiencies. LM14’s signal was in low concentrations where it was expected to be present (Figure 12). When acidity was high LM14’s signal increased slightly. Subjected to LM19 EDC uncovered that -P had high concentrations of neutral pectic polysaccharides caused by 3 data points. Subsequently, -K had 3 smaller peaks more grouped toward the acidic residues which is localised where pectin was expected. –N and the control blended into background noise (Figure 12).

LM21 epitope detection highlights that high levels of neutral mannans are present in the control which were flushed out of the column within 10 fractions. The control also had 2 smaller peaks towards the centre of the fractions (Figure 13). The other conditions blended in with the background; however, all conditions produced a small bump between fractions 25 and 40. There was also a small peak produced by -N. The most interesting result, LM25 released a lot of neutral xyloglucans particularly in the control which then returned to nominal levels until fraction 25 where LM25’s signal dramatically increased until fraction 40 (Figure 13). This was the location where xyloglucan was expected to be localised. Furthermore, the nutrient deficiencies formed a bump between fractions 15 and 35. This may suggest that their xyloglucan is slightly less acidic and abundant compared to full nutrient growth.

Since T. aestivum is larger higher, signals derived from the mAbs were detected. LM14 varied between the conditions with -P and -K producing comparable signals, although -K shifted becoming more acidic. All conditions had stronger signals than the control, though there was a small control peak at fraction 35 suggesting more acidic AGPs (Figure 13). All conditions increased together from fractions 20 to 40 slightly decreasing, which may indicate the presence of AGPs as it was in the correct location. LM19’s signal frequently appeared slightly shifted to the left to where pectins should be located. This indicated a slight decrease of pectic acidity (Figure 13). The highest peaks were -P indicated by 5 data points and was preceded by the control. -N had a small peak at the start suggesting the release of neutral pectic polysaccharides. Low levels of mannans were released from the column until fraction 15. This did not occur in A. thaliana which released high levels of mannans (Figures 12 and 13). -P was the highest signal producing 2 peaks in the slightly acidic regions to the more acidic residues. All groups produced a bump between fractions 20 and 45.

All the conditions produced 3 distinct peaks when subjected to LM25. The first peak specified the presence of neutral xyloglucans which could also be seen in A. thaliana. The second narrow peak indicated smaller and fewer xyloglucans which was a lot higher in A. thaliana. In A. thaliana there was a bump indicating these smaller xyloglucans. The third peak which was not so present in A. thaliana suggested higher levels of acidic residues (Figures 12 and 13). These residues maybe linked to pectic polysaccharides. This third peak was within the expected location of xyloglucan though the second peak is not.

Though 3 different peaks were present, the conditions followed the same order with -P at the top preceded by -N, -K and the control having consistently the lowest signal (Figure 13). Notably, -K and -P swapped rankings with -P peaking more neutral than -K. -P’s peak was wider compared to the other peaks.

5.0 Discussion

This study utilised a novel approach to determine what major polysaccharides were present in Arabidopsis and Triticum root mucilage and how they altered with the exclusion of N, K and P. Previous research had determined that xyloglucan, pectin, AGP and mannan were major mucilage components. Although the quantitative levels of the polysaccharides and how they altered to differing deficiencies remains to be determined. The results obtained from this research suggest that xyloglucan either remained unaffected by the nutrient deficiencies or increased in concentration in both species. When subjected to -N, -K or -P the other polysaccharides were suppressed. When plants were grown in full media xyloglucan had the highest signal closely followed by mannan, pectin and AGP. Pectin rapidly became more diffuse in plants grown without phosphorus. In Arabidopsis, xyloglucan remained unaltered in acidity but large quantities of neutral derivatives were apparent. In Triticum, xyloglucan became more acidic and released high levels of neutral residues compared to full nutrient plants.

5.1 An update of the polysaccharide constituents of root mucilage

There have been many areas of confusion within mucilage research. Early studies suggested that mucilage contained many other components including proteins and extracellular DNA. Two recent studies in particular offer a clear definition divorcing mucilage and the overall term root exudate (Lynch and Whipps 1990; McNear 2013). Some early and more recent studies use these terms interchangeably, causing this confusion. There has been no direct measuring of polysaccharide levels within mucilage utilising antibody affinity, instead their counterpart monosaccharides have been measured. Little dicotyledonous species have been used in mucilage research and there has been a lack of research using A. thaliana as a model. In both A. thaliana and T. aestivum, xyloglucan had the highest signal, closely followed by mannan, pectin and then considerably lower AGP. This is despite the RE prints clearly highlight AGP along the upper roots of A. thaliana (Figure 10, A). In A. thaliana, xyloglucan, mannan and pectin signals became entangled at the 1 in 125 dilution, whereas AGP signal remained constant until the 1 in 625 dilution merged with the other signals at the 1 in 3,125 dilution (Figure 8, top left). Previous research had determined that arabinose, xylose, galactose and glucose were major components in V. unguiculata, L. sativum and Z. mays mucilage, suggesting the presence of the major polysaccharides used within this study (Moody et al. 1988). The monosaccharides form most of the component parts of the polysaccharides outlined in this investigation compared to other polysaccharides identified in low concentrations determined by wide screening (Figure 4). In many other studies the same monosaccharides were observed, all supporting the observations in this investigation. Some research suggested that mannan and pectic polysaccharides were only present as minor contributors (Bacic et al. 1986). This result contrasted with the results of this study’s RE prints which illustrated pectin along the roots and being diffused (Figure 10, E). Uronic acids which has an affinity to pectin had been detected in higher amounts compared to their homogalacturonan counterparts (Carpita 1996), thus revealing an area of conflict.

β-glycosyl Yariv reagents have been used to confirm the presence of AGP, reportedly in high amounts (Bacic et al. 1986), although this study suggests that AGPs were in low amounts in mucilage. One study uncovered that dicots contained 5.5% fewer polysaccharides within their mucilage both secreting equally high levels of xyloglucan and trace amounts of mannan (Moody et al. 1988). Uncertainty remains as to what percentage each major polysaccharide makes up mucilage. Heteroxylans had been documented to be secreted in similar levels in both Commelinoid monocots and dicots (Morel et al. 1986).

Within the wide screening of this study LM11 which binds to arabinoxylan as well as (1→4)-β-D-xylan had a high affinity to T. aestivum mucilage (Figure 4). When subjected to LM24 no binding was present yet there was a high signal when subjected to LM25 and a lower but noticeable LM15 signal in T. aestivum mucilage. This suggests that xyloglucan within both plant’s mucilage had to contain the XXXG motifs. As well as LM19 binding there was LM7 binding which was slight, indicating that pectin was probably partially methylated within the mucilage of T. aestivum (Figure 4). Most epitopes had diffused after being mixed with water for 24 hours unlike in the 1 hour samples in which most signals were below 0.2 au. Xyloglucan had rapidly diffused within the pure samples from the control media. Extending the diffusion time by another 24 hours may have increased these epitope signals.

5.2 Xyloglucan secretion in monocots and dicots

This study which had directly measured xyloglucan levels, confirmed that xyloglucan was secreted in comparable amounts in both Commelinoid monocots and dicots, only differing by 1.8% more in A. thaliana. Notably, a study demonstrated a 1.2% increase in xylose when comparing dicotyledonous and monocotyledonous species (Ray et al. 1988). This suggested that xyloglucan may have an independent function which is further supported by the differences in their cell wall compositions. As Commelinoid monocots contained significantly fewer amounts of xyloglucan compared to dicots, they should not be able to secrete equal levels of xyloglucan at their root caps and epidermal layers behind their caps. It is reasonable to stipulate that xyloglucan has more importance to fungi which is supported by previous research which used Z. mays and not T. aestivum which is a Commelinoid monocot, containing little xyloglucan within their cell walls. Perhaps these fungi are initially attracted by root mucilage having less importance to the plants that secrete mucilage. It would be interesting to examine the effects of mutants which do not secrete xyloglucan from both Commelinoid monocots and dicots and their abilities to attract and maintain fungi interactions. A chemical and structural analysis of xyloglucan would be interesting to determine what feature may exist to make xyloglucan important to fungi.

5.3 Neutrality and acidity of Arabidopsis and Triticum control mucilage

A study uncovered that 75% of monosaccharides were acidic, indicating low levels of pectins and AGPs in T. aestivum (Moody et al. 1988). The research also revealed that Triticum mucilage contained 24% of glucuronic acid. No direct measuring of acidity within A. thaliana’s root polysaccharides had taken place. This study directly measured polysaccharide acidity through EDC. Mannan was uncovered to have eluted in large amounts in the first 10 fractions, indicating the presence of high amounts of neutral mannan residues followed by 3 smaller acidic peaks (Figure 12, I-P). This was also repeated for xyloglucan, however, xyloglucan had high acidity between fractions 22 and 40, whereas in T. aestivum the expected range peaked at fraction 40 (Figure 13, I-P). Since a monosaccharide’s acidity depends on carboxyl groups, perhaps this indicates an alteration in the structure of the secreted polysaccharides. Furthermore, other studies have suggested that increases in uronic acid also cause increases in monosaccharide acidity, containing a terminal hydroxyl group (Smith 1979). If true, targeting uronic acids would act as an indirect indicator of the presence of pectins, yet this was not apparent in early monosaccharide studies (Bacic et al. 1986, Moody et al. 1998).

There was an interesting contrast when comparing A. thaliana and T. aestivum as the peak in xyloglucan noted in A. thaliana was just a fraction of that recorded in T. aestivum. T. aestivum had 2 major peaks; the first peak highlighted the secretion of neutral residues followed by a larger second peak which was located in the predicted acidity of xyloglucan. This is unusual as xyloglucan should be secreted in similar conditions from both plants. In T. aestivum, mannan gradually became more acidic starting from fraction 20 and reduced at fraction 45. A pure sample of mannan run through an anion-exchange chromatography column for EDC may have provided an expected detection range predicting mannan’s acidity would be localised. Higher signals from pectin and AGP were detected in T. aestivum which may be due to T. aestivum having a larger mass compared to A. thaliana. T. aestivum was grown in liquid culture allowing direct sampling of the media without much processing. One study uncovered that elevated levels of AGPs and glucuronoarabinoxylans marked higher acidity within mucilage (Moody et al. 1988). AGP was detected within range of the expected fractions, demonstrating AGP’s growing acidity with 3 additional peaks indicating spikes in acidity. However, pectin was uncovered to be less acid than expected (Figure 12, A-H). Ideally pectin would increase in acidity from fraction 30, whereas this study revealed that acidity began to increase from fraction 20. It would be interesting to investigate structural differences using NMR of the secreted polysaccharides from both Commelinoid monocots and dicots which may occur.

5.4 Alterations in root polysaccharides in -N, -K and -P deficiencies

One of the more novel facets of this investigation was to exclude 3 macronutrients and monitor the alterations in the root polysaccharide secretions. The control which contained all micro and macronutrients in the form of BG11 media was compared to the deficiencies. Xyloglucan was the most interesting polysaccharide exhibiting major alterations. In A. thaliana, xyloglucan increased from 1.8 au to 2.9 au in -N while sharply declining. In –K, Arabidopsis xyloglucan levels slightly decreased yet remained considerably higher compared to the other polysaccharides; in -P xyloglucan levels slightly increased (Figure 8, top right). In T. aestivum xyloglucan rapidly increased in -P and increased but at a lower rate in -N and -K. The rate of decline in -K and -P was similar with comparable levels of variation (Figure 9, top right).

5.5 Xyloglucan as an important secretion

From the RE prints, this observation was repeated and xyloglucan secretion could clearly be seen along the roots, root caps and older roots (Figure 11, F). Xyloglucan was secreted from more areas of root unlike the control where xyloglucan was secreted mostly from the root caps. As expected A. thaliana grown without nitrogen exhibited longer root growth, yellowing of the leaves and weakened structure unlike the control where leaves were green and roots were thicker. For -K and -P xyloglucan was secreted within the root caps at a slightly elevated level but it is also secreted from the older root growth. Comparable to -N Arabidopsis, -K and -P Arabidopsis had weakened structure, reduced leaf sizes and longer roots. Although growth was weaker, xyloglucan levels increased despite the condition of the plants. When plants are under stress they have weakened growth and consequently conserve resources and energy usage until favourable conditions return (Alberts et al. 2008; Franklin et al. 2014). Fresh weight of T. aestivum increased with nutrient deficiencies. This may be caused by a higher rate of water diffusion due to culturing T. aestivum in liquid. If plants are increasing xyloglucan levels in unfavourable conditions, it is plausible that xyloglucan must have major importance otherwise the plants would not be secreting these high energy molecules. For the other polysaccharides, their levels decreased supporting this argument, however, xyloglucan levels either remained the same or increased compared to full nutrient grown plants. When plants are prevented from retrieving macronutrients particularly phosphates and nitrates, plants attempt to attract fungi symbiosis in the form of a hyphae network or nodules on the roots which can retrieve nutrients for the plants at a cost of supplying the fungi with sugars (Franklin et al. 2014; Hinch and Clarke 1980). This symbiosis is widespread and well-documented, yet the initial mechanisms which entice fungi remain unknown. Since this study has verified that xyloglucan levels remain constant or are rapidly secreted during macronutrient deficiency, then perhaps this polysaccharide is the key to further understanding this symbiosis.

The results from the wide screen highlighted no signals from LM24 but a high signal from LM25 and a weaker signal from LM11. This revealed that the xyloglucan was galactosylated, formed heterogeneously from (1→4)-β-D-xylans with XXXG motifs and possibly including arabinoxylans. Xyloglucans were slightly acidic within the control becoming more neutral within the deficiencies in Arabidopsis (Figure 12, M-P). Interestingly, all deficiencies followed the same pattern forming a bump starting from the fractions 15 to 30. It was expected that xyloglucan would appear between fractions 30 to 40. The neutral residues were highly released within the control but were reduced in the deficiencies. In Triticum, there were 3 distinct peaks, the first revealing a high release of neutral xyloglucans, where -K had the highest signal followed by -P and -N. The third peak is in an analogous position to the major peaks within Arabidopsis although it highlights an unexpected release of slightly acidic residues. The third peak was located within the expected range of xyloglucan which illustrates a high release of acidic xyloglucans (Figure 13, M-P). -P had the highest release preceded by -N and -K with the control having the lowest release. Therefore, by excluding -N, -K or -P increases the release of neutral residues in the expected range of xyloglucan.

5.6 AGP, pectin and mannan variations in -N, -K and -P mucilage

When A. thaliana was subjected to -N, -K and -P AGP, pectin and mannan concentrations decreased which at times, blended into the background noise. This is as expected when plants are under stress. Though AGP, pectin and mannan declined sharply, within -P the polysaccharides had distinctive signals. These signals followed the same pattern of expression whereas AGP had the lowest signal followed by pectin and mannan which had the strongest signal. However, the RE prints illustrate clear variations between the polysaccharides. These polysaccharides remained unaffected by the nutrient deficiencies in T. aestivum, only slightly altering in -P where AGP and mannan were elevated. It would be interesting to determine if Arabidopsis polysaccharides would remain unchanged if cultured in liquid media as with T. aestivum.

From the RE prints AGP is concentrated towards the older and major roots in the control whereas in –K and –P AGP is undetectable (Figure 10, A-D). In -N, AGP becomes more speckled and fragmented along the major roots. It is feasible to conceive that this verifies that Arabidopsis is conserving the polysaccharides and cutting back on unessential resources. The RE prints demonstrate a rapid increase in pectin diffusion in -P media. In -K concentrations of pectin within the roots rapidly increased which may be explained by the lack of diffusion (Figure 10, G-H). Pectin diffusion is prevented resulting in pectin retention in the roots which should be secreting pectin. Although diffusion rapidly increased without P, concentrated pectin signals were still evident in the major roots. However, in -N pectin levels slightly decreased and became more concentrated towards the upper root hairs. Mannan concentrations were considerably reduced in -K but remained minimally present in -N and -P with small dots of concentrated mannans (Figure 11).

Though there were changes in AGP, pectin and mannan were evident in the RE prints they were not so in the ELISAs. One possible explanation is that nitrocellulose was placed in the same position as the roots prior to gel extractions for the ELISAs, possibly absorbing the majority of the polysaccharide present on the nutrient deficiencies. Xyloglucan alterations were documented within all experiments in this study. Since AGPs are still secreted in -N plants, this illustrates a possible role for AGP in -N. It is reasonable to propose that each polysaccharide has a particular role to play during each deficiency which may explain the differing epitope concentrations within the roots and diffusing out of them. It is feasible that each polysaccharide is necessary in differing amounts to attract certain fungi which could be specialised to acquire each macronutrient. Little research has been conducted to understand polysaccharide alterations in the absence of each macronutrient; most research indicates that all mucilage is secreted from the root cap or epidermal layers behind the caps. From this AGP, pectin and mannan is secreted from the entire root system. While xyloglucan measurements from this investigation fit into this paradigm, xyloglucan increased in -N media in both plants.

When subjected to -N in A. thaliana, AGP became more acidic whereas -K and -P AGP levels decreased. There is a high release of neutral pectins in -K but -P the roots release few neutrals with some acidic residues (Figure 12 and 13, E-H and M-P). In the control there was a large amount of neutral mannans released but when subjected to the deficiencies, neutral residues were completely suppressed. There was a bump emerging between fractions 25 and 40 which was unexpected. In all deficient media, neutral release of AGP increased compared to the control and were present in the expected range in T. aestivum. The deficiencies reduce the acidity of the secreted polysaccharides (Figure 12 and 13, A-D). Pectin also became more neutral, releasing 10 fractions earlier than expected. Yet again -P had the greatest impact releasing less acidic residues compared to the control.

The bump observed in mannan from A. thaliana was also present in the same location for T. aestivum (Figure 12 and 13, I-L). -P had the highest signal, demonstrating a high release of both neutral and acidic mannans unlike the control and -K. For -N, neutral sugars were released and after the twentieth fraction declined to background levels. The changes in acidity of the polysaccharides resulted from the alterations observed in mucilage pH when roots modify cation or anion availability (Marschner et al. 1986). Root mucilage increases cations within the surrounding soil, in an attempt to attract negatively charged nitrates and phosphates (Mary et al. 1993). In the presence of heavy metals, mucilage from both dicots and monocots can alter redox potentials and pH, capturing the offending ions thus preventing their absorption into the roots (Watanabe et al. 2008; Romheld and Marschner 1981). This study observed many alterations in the mucilage acidity in the absence of N, K and P. It is possible that both species are attempting to attract their deficient macronutrients by altering their mucilage’s pH, though this study had demonstrated a rise in neutral residues as well as acidic residues. Together with the rise in xyloglucan, could these be indicators of the roots attempting to form fungi symbiosis or alter the local charge within these deficient environments. It would be valuable to directly monitor the alterations in mucilage charge state and pH.

5.7 Study limitations

There were issues of detecting sufficient signals for the ELISAs and EDC for LM14, LM19 and LM21. Perhaps 10 Arabidopsis seeds per plate with dialysis after gel extractions would have been sufficient to improve the signal. A. thaliana was grown on agar and not in liquid similar to Triticum. If A. thaliana had been grown in liquid media maybe the LM14, LM19 and LM21 signals would have been enhanced. A pilot study could have been undertaken to examine the effects of the media on root growth. Other media including Murashige and Skoog are specifically designed for Arabidopsis. However, this medium would have been more challenging to produce and modify, excluding each macronutrient.

Triticum aestivum are Commelinoid monocots which contain low xyloglucan and high xylan within their cell walls. Perhaps monocots with comparable amounts of xyloglucan to dicots including Zea mays could have been utilised for a comparative analysis of monocots and dicots. For some RE prints, background noise was an issue resulting in vivid patterns. The prints which were affected could have been repeated with smaller nitrocellulose sheets. One major limitation of RE printing is reflecting what is present on the nitrocellulose after revealing without presenting bias. Though the same parameters can be applied to each sheet, it may underrepresent or over-represent what is present depending upon the strongest and weakest signals.

5.8 Conclusion and future research

The majority of mucilage research occurred 30 years ago using monosaccharides to imply the presence of polysaccharides. This current study has directly tagged these polysaccharides using mAbs and monitored their modifications during -N, -K and -P conditions. The results of this investigation demonstrated that xyloglucan is an important component in mucilage in both Triticum aestivum and Arabidopsis thaliana. When both species were subjected to the macronutrient deficiencies, xyloglucan rapidly increased in A. thaliana and at a slower rate in T. aestivum. AGP, pectin and mannan rapidly deceased to background levels in A. thaliana and remained unaltered in T. aestivum. Current knowledge suggests that plants secrete unknown amounts of sugars which may attract fungi symbiosis in deficient media. These fungi will then absorb the nutrients in a form that plants cannot utilise. This initial attraction may be encouraged by xyloglucan. This investigation has highlighted that xyloglucan is secreted in similar amounts in both T. aestivum and A. thaliana and increases during -N, -K and -P stress. Xyloglucan was also altered becoming more acidic in Triticum and more neutral in Arabidopsis suggesting structural reorganisation had taken place.

The plants were exhibiting clear signs of stress as the weights of A. thaliana decreased when grown without N, K and P becoming weaker and containing less chlorophyll within their leaves. In Triticum, roots length and fresh weight slightly increased without detrimental effects on their leaves. Xyloglucan was also demonstrated to be secreted evenly along the roots rather than in the root caps or associated epidermal layers. This illustrated the location of xyloglucan secretion which had been reorganised compared to the control. Although the other polysaccharides were secreted significantly less than xyloglucan, the RE prints highlighted elevated diffusion occurring within pectin in -P and increased concentrations of pectin in the roots without K. AGP and mannan signals were higher in older root growth within the control which rapidly decreased within the nutrient deficiencies. AGP became more neutral when secreted from T. aestivum, whereas pectin and mannan became more acidic in the macronutrient deficiencies.

There still remain many gaps in knowledge within this field, particularly as to what percentage each polysaccharide constitutes mucilage. Moreover, NMR could elucidate structural alterations that may be evident in macronutrient deficient mucilage. It is possible that polysaccharides are in different forms as a consequence of lacking each essential nutrient. It is reasonable that xyloglucan has an independent importance from plants and a chemical analysis could elucidate the possible reasons behind xyloglucan’s importance. By using Arabidopsis mutants unable to secrete xyloglucan, it may uncover whether fungi symbiosis can still proceed thus reinforcing xyloglucan’s role in fungi attraction. Regular monitoring of xyloglucan throughout fungi symbiosis using Arabidopsis could determine if xyloglucan is required for the maintenance of this relationship. Using fluorescent mAbs in microscopy would localise each polysaccharide that was being secreted, revealing the status of each polysaccharide prior to secretion. It may be informative to investigate the effects of excluding each micronutrient and how they may alter the polysaccharides within mucilage of Commelinoid monocots, monocots and dicots.

6.0 References

Albersheim, P. Darvil, A. Roberts, K. Sederoff, R. and Staehelin, A. (2010). “Plant cell walls.” New York: Garland Science.

Alberts, B. Johnson, A. Lewis, J. Raff, M. Roberts, K. and Walter, P. (2008). “Molecular biology of the cell.” 5^th ed. New York: Garland Science.

Alkafafy, M., S. Ebada, R. Rashed and H. Attia (2012). "Comparative morphometric and glycohistochemical studies on the epididymal duct in the donkey (Equus nd asinus) adromedary camel (Camelus dromedarius)." Acta Histochemica 114: 434-447.

Bacic, A. Moody, S. F. and Clarke, A. E. (1986). “Structural analysis of secreted root slime from maize (Zea mays L.).” Plant Physiol 80: 771-777.

Baetz, U. and E. Martinoia (2014). "Root exudates: the hidden part of plant defense." Trends Plant Sci 19: 90-98.

Bais, H. P. Weir, T. L. Perry, L. G. Gilroy, S. and Vivanco, J. M. (2006). “The role of root exudates in rhizosphere interactions with plants and other organisms.” A. Revi. 57: 233-266.

Bardi, D. V., Weir, T. L., van der Lelie, D., & Vivanco, J. M. (2009). “Rhizosphere chemical dialogues: Plant-microbe interactions.” Cur. Opin. Biotechnology. 20: 642–650.

Broeckling, C. D. Broz, A. K. Bergelson, J. Manter, D. K. and Vivanco, J. M. (2007). “Root Exudates Regulate Soil Fungal Community Composition and Diversity.” Applied and Environ. Microbiol. 74: 738-744.

Brown, J. C. (1978). “Mechanism of iron uptake by plants.” Plant Cell and Environment 1: 249-257.

Caesan, M. A., C. Durand, C. Burel, C. Gangneux, P. Lerouge, T. Ishii, K. Laval, M. L. Follet-Gueye, A. Driouich and M. Vicre-Gibouin (2012). "Effect of arabinogalactan proteins from the root caps of pea and Brassica napus on Aphanomyces euteiches zoospore chemotaxis and germination." Plant Physio 159: 1658-1670.

Carpita, N. C. (1996). “Structure and biogenesis of the cell wall of grasses.” Au. Rev. Plant Physio Plant. Mol. Biol. 47: 445-476.

Carpita, N. C. and Gibeaut, D. M. (1993). “Structural models of primary cell walls in flowering plants: consistency of molecular structure with the physical properties of the walls during growth.” Plant J. 1: 1-30.

Carvalhais, L. C. Deis, P. G. Fedoseyenko, D. Hajirezaei, M. Borriss, R. and Wirén, N. V. (2011). “Root exudation of sugars, amino acids, and organic acids by maize as affected by nitrogen, phosphorus, potassium, and iron deficiency.” J. Plant Nutrition and Soil Sci. 174: 3-11.

Cassab, G. I. (1998). “Plant cell wall proteins.” Plant Physio. Plant Molec. Biol 49: 281-309.

Clausen, M. H., W. G. T. Willats, et al. (2003). "Synthetic methyl hexagalacturonate hapten inhibitors of anti-homogalacturonan monoclonal antibodies LM7, JIM5 and JIM7." Carbohydr Res 338: 1797-1800.

Chaboud, A. and M. Rougier (1984). "Identification and Localization of Sugar Components of Rice (Oryza sativa L.) Root Cap Mucilage." Journal of Plant Physiology 116: 323-330.

Chaw, S. M. Chang, C. C. Chen, H. L. and Li, W. H. (2004). "Dating the monocot-dicot divergence and the origin of core eudicots using whole chloroplast genomes." J Mol Evol 58: 424-441.

Chaboud, A. (1983). “Isolation, purification and chemical composition of maize root cap mucilage.” Plant and Soil 73: 395–402.

Choct, M. (1997). “Feed non-starch polysaccharides: chemical structures and nutritional significance.” Feed Milling International 5: 13-26.

Clarke, A. E., R. L. Anderson and B. A. Stone (1979). "Form and function of arabinogalactans and arabinogalactan-proteins." Phytochemistry 18: 521-540.

Cornuault, V., I. W. Manfield, M.-C. Ralet and J. P. Knox (2014). "Epitope detection chromatography: a method to dissect the structural heterogeneity and inter-coections of plant cell-wall matrix glycans." Plant J 78: 715-722.

Dick-Perez, M. Zhang, Y. Hayes, J. Salazar, A. Zabotina, O. A. and Hong, M. (2011). “Structure and Interactions of Plant Cell-Wall Polysaccharides by Two- and Three-Dimensional Magic-Angle-Spiing Solid-State NMR”. Biochem 50: 989-1000.

Driouich, A., M. L. Follet-Gueye, M. Vicre-Gibouin and M. Hawes (2013). "Root border cells and secretions as critical elements in plant host defense." Curr Opin Plant Biol16: 489-495.

Franklin, O. Nasholm, T. Hogberg, P. and Hogberg, M. H. (2014). “Forests trapped in nitrogen limitation – an ecological market perspective on ectomycorrhizal symbiosis.” New Phytologist. 203: 657-666.

Fry, S. C. (1998). “Oxidative scission of plant cell wall polysaccharide by ascorbate- induced hydroxyl radicals.” Biochem. J 332: 507-515.

Fry, S. C. York, W. S. Albersheim, P. Darvill, A. Hayashi, T. Joseleau, J. Kato, Y. Lorenes, E. P. Maclachlan, G. A. McNeil, M. Mort, A. J. Reid, G, Seitz, H. U. Selvendran, R. S. Voragen, A. G. J. and White, A. R. (1993). “An unambiguous nomenclature for xyloglucan-derived oligosaccharides.” Physiologia Plantarium 89: 1-3.

Gahoonia, T. S. and Nielson, N. E. (1998). “Direct evidence on participation of root hairs in phosphorus (³²P) uptake from soil.” Plant and Soil 198: 147-152.

Geshi, N. J. N. Johansen, A. Dilokpimol, A. Rolland, K. Belcram, S. Verger, T. Kotake, Y. Tsumuraya, S. Kaneko, T. Tryfona, P. Dupree, H. V. Scheller, H. Hofte and G. Mouille (2013). "A galactosyltransferase acting on arabinogalactan protein glycans is essential for embryo development in Arabidopsis." Plant J 76: 128-137.

Gerwing, G. J. Kamerling, J. P. and Vliegenthart, J. F. G. (1978). “Determination of the d and l configuration of neutral monosaccharides by high-resolution capillary g.l.c.” Carbohydrate Research. 62: 349-357.

Göllner, E. M., W. Blaschek and B. Classen (2010). "Structural Investigations on Arabinogalactan-Protein from Wheat, Isolated with Yariv Reagent." Journal of Agricultural and Food Chemistry 58: 3621-3626.

Gooch, J. (2011). Growth Curve. Encyclopedic Dictionary of Polymers J. Gooch, Springer New York: 896-896.

Grinsted, M. J. Hedley, M. J. White, R. E. and Nye, P. H. (1982). “Plant induced changes in the rhizosphere of rape (brassica napus var. emerald). Seedlings.” New Physiology 117: 371-375.

Hinch, J. M. and Clarke, A. E. (1980). “Adhesion of fungal zoospores to root surface is mediated by carbohydrate determinates of root slime physiology.” Plant Pathology16:303-308.

Hinsinger, P., A. G. Bengough, D. Vetterlein and I. M. Young (2009). "Rhizosphere: biophysics, biogeochemistry and ecological relevance." Plant and Soil 321: 117-152.

Iijima, M., T. Higuchi and P. W. Barlow (2004). "Contribution of root cap mucilage and presence of an intact root cap in maize (Zea mays) to the reduction of soil mechanical impedance." A Bot 94: 473-477.

Jentschke, G. and Godbold, D. L. (2000). “Metal toxicity and ectoycorrhizas.” Physiologia Plantarum 109: 107-116.

Jones, L. Seymour, G. B. and Knox, J. P. (1997). "Localisation of pectin galactan in tomato cell walls using a monoclonal antibody speific to (1-4)-β-D-Galactan." Plant Physiol 113: 1405-1412.

Kamerling, J. P. (2007). “Comprehensive Glycoscience, Four-Volume Set: from Chemistry to Systems Biology: From Chemistry to Systems Biology.” 1^st ed. Toronto: Elsevier Science Ltd.

Knee, E. M. Gong, F. Gao, M. Teplitski, M. Jones, A. R. Foxworthly, A. Mort, A. J. and Bauer, W. D. (2001). “Root mucilage from pea and its utilisation by rhizosphere bacteria as a sole carbon source.” Amer. Phytopathological Soc 6: 775-784.

Kato, Y. and Nevins, J. Y. (1984). “Enzymatic dissolution of Zea shoot cell wall polysaccharides.” Plant Physiology 75:740-744.

Knox, J. P. Day, S. and Roberts, K. (1989). "A set of cell surface glycoproteins forms a marker of cell position, but not cell type, in the root apical meristem of Daucus carota L." Development 106: 47-56.

Knox, J. P. Linstead, P. J. Peart, J. Cooper, C. and Roberts, K. (1991). "Developmentally-regulated epitopes of cell surface arabinogalactan-proteins and their relation to root tissue pattern formation." Plant Journal 1: 317-326.

Knox, J. P. Peart, J. and Neill, S. J. (1995). "Identification of novel cell surface epitopes using a leaf epidermal strip assay system." Planta 196: 266-270.

Lynch, M. A. and Staehelin, L . A. (1995). “Immunocytochemical localization of cell wall polysaccharides in the root tip of Avena sativa.” Protoplasma 188: 115-127.

Lynch, J. M. and J. M. Whipps (1990). "Substrate flow in the rhizosphere." Plant and Soil 129: 1-10.

Ma, W. Muthreich, N. Liao, C. Franz-Wachtel, M. Shutz, W. Zhang, F. Hochholdinger, F. and Li, C. (2009). “The mucilage proteome of maize (Zea mays L.) primary roots.” J. Proteome Res 9: 2968-2976.

Marcus, S. E., A. W. Blake, T. A. Benians, K. J. Lee, C. Poyser, L. Donaldson, O. Leroux, A. Rogowski, H. L. Petersen, A. Boraston, H. J. Gilbert, W. G. Willats and J. P. Knox (2010). "Restricted access of proteins to maan polysaccharides in intact plant cell walls." Plant J 64: 191-203.

Marcus, S. E. Verhertbruggen, Y. Herve, C. Ordaz-Ortiz, J. J. Farkas, V. Pederson, H. L. Willats, W. G. T. and Knox, J. P. (2008). "Pectic homogalacturonan masks abundant sets of xyloglucan epitopes in plant cell walls." BMC Plant Biol 8: 60-68.

Mary, B. Fresneau, C. Morel, J. L. and Mariotti, A. (1993). “C and N cycling during decomposition of root mucilage, roots and glucose in soil.” Soil Biol. Biochem 25: 1005-1014.

Marschner, H. Romheld, V. Horst, W. J. and Martin, P. (1986). “Root-induced changes in the rhizosphere: importance for the mineral nutrition of plants.” Pflanzenernaecher Bodnk 149: 441-456.

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.

McCully, M. E. (1999). “Roots in soil: unearthing the complexities of roots and their rhizosphere.” Au. Rev. Plant Physiol 50: 695-718.

McCully, M. E. and Boyer, J. S. (1997). “The expansion of maize root-cap mucilage during hydration. 3. Changes in water potential and water content.” Physiologia Plantarum 99: 169–177. McCully, M. E. and L. J. Sealey (1996). "The expansion of maize root-cap mucilageduring hydration. 2. Observations on soil-grown roots by cryo-scaing electron microscopy." Physiologia Plantarum 97: 454-462.

McNear, D. H. (2013). “The rhizosphere – root, soil and everything in between.” Nature Education Knowledge 4(3): 1.

McNeil, M. Darvill, A. G. Fry, S. C. and Albersheim, P. (1984). “Structure and Function of the Primary Cell Walls of Plants.” Aual Rev. Biochem 53: 625-663.

Miyaska, S. C. and Hawes, M. C. (2001). “Possible role of root border cell sin detection and avoidance of aluminium toxicity.” Plant Physiology 125: 1978-1987.

Miki, N. K. Clarke, K. J. and McCully, M. E. (1980). “A histological and histochemical comparison of the mucilages on the root tips of several grasses.” Canadian J. Botany 58: 2581-2593.

Miller, I. and Moore, R. (1990). “Defective Secretion of Mucilage is the Cellular Basis for Agravitropism in Primary Roots of Zea mays cv. Ageotropic.” Aals of Botany 66: 169-178.

Moody, S. F. Clarke, A. E. and Bacic, A. (1988). “Structural analysis of secreted slime from wheat and cowpea roots.” Phytochemristry 27: 2857-2861.

Morel, J. L. and Guckert, A. (1986). “Measurement of Pb²⁺, Cu²⁺ and Cd²⁺ binding with mucilage exudates from maize (Zea mays L.) roots.” Biol. Fertil. Soils 2: 29-34.

Moller, I., S. E. Marcus, A. Haeger, Y. Verhertbruggen, R. Verhoef, H. Schols, P. Ulvskov, J. D. Mikkelsen, J. P. Knox and W. Willats (2008). "High-throughput screening of monoclonal antibodies against plant cell wall glycans by hierarchical clustering of their carbohydrate microarray binding profiles." Glycoconj J 25: 37-48.

Narasimhan, K., C. Basheer, V. B. Bajic and S. Swarup (2003). "Enhancement of plant-microbe interactions using a rhizosphere metabolomics-driven approach and its application in the removal of polychlorinated biphenyls." Plant Physiol 132: 146-153.

Nguema-Ona, E., M. Vicre-Gibouin, M. A. Caesan and A. Driouich (2013). "Arabinogalactan proteins in root-microbe interactions." Trends Plant Sci 18: 440-449.

Osborn, H. M. I., F. Lochey, L. Mosley and D. Read (1999). "Analysis of polysaccharides and monosaccharides in the root mucilage of maize (Zea mays L.) by gas chromatography." Journal of Chromatography A 831: 267-276.

Northcote, D. H. and Pickett-Heaps, J. D. (1966). “A function of the Golgi apparatus in polysaccharide biogenesis and transport in the root-cap cells of wheat.” Biochem. J 98: 159-167.

Oades, J. M. (1978). “Mucilages at the root surface.” J. Soil Science 29: 1-16.

Pedersen, H. L., J. U. Fangel, B. McCleary, C. Ruzanski, M. G. Rydahl, M. C. Ralet, V. Farkas, L. von Schantz, S. E. Marcus, M. C. Andersen, R. Field, M. Ohlin, J. P. Knox, M. H. Clausen and W. G. Willats (2012). "Versatile high resolution oligosaccharide microarrays for plant glycobiology and cell wall research." J Biol Chem 287: 39429-39438.

Pattahil, S. Avci, U. Baldwin, D. Swees, A. G. McGill, J. A. Popper, Z. Albert, A. Davis, R. H. Cheareddy, C. Dong, R. O’Shea, B. Rossi, R. Leoff, C. Freshour, G. Narra, R. O’Neil, M. York, W. S. and Hahn, M. G. 2010. “A Comprehensive Toolkit of Plant Cell Wall Glycan-Directed Monoclonal Antibodies.” Plant Physiology. 153: 514-525.

Perotto, S. Donovan, N. Drobak, B. K. and Brewin, N. J. (1995). "Differential expression of a glycosyl inositol phopholipid antigen on the peribacteroid membrane during pea nodule development." Molecular Plant-Microbe Interactions 8: 560-568

Popper, Z. A. and Fry, S. C. (2003). “Primary cell wall composition of Bryophytes and Charophytes.” Aals of Botany 91: 1-12.

Ray, T. C. Callow, J. A. and Keedy, J. F. (1988). “Composition of Root Mucilage Polysaccharides from Lepidium sativum.” J. Experimental Botany 39: 1249-1261.

Read, D. B. and P. J. Gregory (1997). "Surface tension and viscosity of axenic maize and lupin root mucilages." New Phytologist 137: 623-628.

Robyt, J. F. (1999). “Essentials of Carbohydrate Chemistry.” New York: Springer.

Römheld, V. and H. Marschner (1981). "Iron deficiency stress induced morphological and physiological changes in root tips of sunflower." Physiologia Plantarum 53: 354-360.

Rippka, R. Deruelles, J. Waterbury, J. B. Herdman, M. and Stainer, R. Y. (1979). “Generic assignments, strain histories and properties of pure cultures of cyanobacteria.” J. General Microbiology 111: 1-61.

Roy, S. S. Mittra, B. Sharma, S. Das, T. K. and Babu, C. R. (2002). “Detection of root mucilage using an anti-fucose antibody.” Aals of Botany 89: 293-299.

Rougier, M., C. Kieda and M. Monsigny (1979). "Use of lectin to detect the sugar components of maize root cap slime." Journal of Histochemistry & Cytochemistry 27: 878-881.

Sealey, L. J. McCully, M. E. and Cay, M. J. (1995). “The expansion of maize root-cap mucilage during hydration.” Physiologia Plantarum 93: 38–46.

Skene, K. R. Kierans, M. Sprent, J. I. and Ravens, J. A. (1996). “Structural aspects of cluster root development and their possible significance for nutrient acquisition in Grevillea robusta (Proteaceae).” Aals of Botany 77: 443-451.

Smith, H. (1979). “The Molecular Biology of Plant Cells.” 14^th ed. California: University of California Press.

Smith, S. E. and Smith, F. A. (1990). “Structure and Function of the Interfaces in Biotrophic Symbioses as They Relate to Nutrient Transport.” New Phytologist 114: 1-38.

Stacey, N. J. Roberts, K. and Knox, J. P. (1990). "Patterns of expression of the JIM4 arabinogalactan protein epitope in cell cultures and during somatic embryogenesis in Daucus carota L." Planta 180: 285-292.

Sterling, J.D., Atmodjo, M.A., Inwood, S.E., Kumar Kolli, V.S., Quigley, H.F., Hahn, M.G., and Mohnen, D. (2006). “Functional identification of an Arabidopsis pectin biosynthetic homogalacturonan galacturonosyltransferase.” Proc Natl Acad Sci 103:5236-5241.

Siyavula. (2012). Anatomy of dicotyledonous plants. Rice University. http://cnx.org/content/m43142/latest/

Tan, L., S. Eberhard, S. Pattathil, C. Warder, J. Glushka, C. Yuan, Z. Hao, X. Zhu, U. Avci, J. S. Miller, D. Baldwin, C. Pham, R. Orlando, A. Darvill, M. G. Hahn, M. J. Kieliszewski and D. Mohnen (2013). "An Arabidopsis cell wall proteoglycan consists of pectin and arabinoxylan covalently linked to an arabinogalactan protein." Plant Cell25: 270-287.

Vogel, J. (2008). "Unique aspects of the grass cell wall." Current Opinion in Plant Biology 11: 301-307.

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.” Carbohydrate Research 344: 1858-1862.

Verhertbruggen, Y. Marcus, S. E. Haeger, A. Verhoef, R. Schols, H. A. McCleary, B. V. McKnee, L. Gilbert, H. J. and Knox, J. P. (2009). "Developmental complexity of arabinan polysaccharides and their processing in plant cell walls." Plant J 59: 413-425.

Voiniciuc, C., G. H. Dean, J. S. Griffiths, K. Kirchsteiger, Y. T. Hwang, A. Gillett, G. Dow, T. L. Western, M. Estelle and G. W. Haughn (2013). "Flying saucer1 is a transmembrane RING E3 ubiquitin ligase that regulates the degree of pectin methylesterification in Arabidopsis seed mucilage." Plant Cell 25: 944-959.

Walker, T. S., H. P. Bais, E. Grotewold and J. M. Vivanco (2003). "Root exudation and rhizosphere biology." Plant Physiol 132: 44-51.

Watanabe, T., S. Misawa, S. Hiradate and M. Osaki (2008). "Characterization of root mucilage from Melastoma malabathricum, with emphasis on its roles in aluminum accumulation." New Phytol 178: 581-589.

Watt, M., M. E. McCully and C. E. Jeffree (1993). "Plant and bacterial mucilages of the maize rhizosphere: Comparison of their soil binding properties and histochemistry in a model system." Plant and Soil 151: 151-165.

Willats, W. G. Orfila, C. Limbert, G. Bucholt, C. van Alebeek, G. W. M. Voragen, A. G. J. Marcus, S. E. Christensen, T. M. I. E. Mikklesen, J. D. Murray, B. S. and Knox, J. P. (2001). "Modulation of the degree and pattern of methyl-esterification of pectic homogalacturonan in plant cell walls. Implications for pectin methyl esterase action, matrix properties, and cell adhesion." J Biol Chem 276: 19404-19413.

Wright, K. (1975). “Polysaccharide of root cap slime from five maize varieties.” Wright, K. Northcote, D. H. (1974). “The relationship of root cap slime synthesis within dictoysomes of maize root cap. Protoplasma 88: 225-239.

Wright, K. and Northcote, D. H. (1974). “The relationship of root-cap slimes to proteins.” Biochem J. 139: 525-534.

Xu, W., G. Ding, K. Yokawa, F. Baluska, Q. F. Li, Y. Liu, W. Shi, J. Liang and J. Zhang (2013). "An improved agar-plate method for studying root growth and response ofArabidopsis thaliana." Sci Rep 3: 1273.

Yates, E. A. Valdor, J. Haslam, S. M. Morris, H. R. Dell, A. Mackie, W. and Knox, J. P. (1996). "Characterisation od carbohydrate structural features recongised by anti-arabinogalactan-protein monoclonal antibodies." Glycobiology 6: 131-139.

Back to top