Syncytia formed by adult female Heterodera schachtii in Arabidopsis thaliana roots have a distinct cell wall molecular architecture


Author for correspondence:
P. E. Urwin
Tel: +44 113 343 2909


  • Plant-parasitic cyst nematodes form a feeding site, termed a syncytium, through which the nematode obtains nutrients from the host plant to support nematode development. The structural features of cell walls of syncytial cells have yet to be elucidated.
  • Monoclonal antibodies to defined glycans and a cellulose-binding module were used to determine the cell wall architectures of syncytial and surrounding cells in the roots of Arabidopsis thaliana infected with the cyst nematode Heterodera schachtii.
  • Fluorescence imaging revealed that the cell walls of syncytia contain cellulose and the hemicelluloses xyloglucan and heteromannan. Heavily methyl-esterified pectic homogalacturonan and arabinan are abundant in syncytial cell walls; galactan could not be detected. This is suggestive of highly flexible syncytial cell walls.
  • This work provides important information on the structural architecture of the cell walls of this novel cell type and reveals factors that enable the feeding site to perform its functional requirements to support nematode development.


Plant-parasitic nematodes can be broadly classified as either ecto- or endoparasites and, within each grouping, the feeding nematode may be sedentary or migratory within the host organ. Sedentary endoparasitic nematodes are the most economically important species and include nematodes of the genus Meloidogyne (root-knot nematodes) and the genera Heterodera and Globodera (cyst nematodes). Sedentary endoparasitic nematodes form complex interactions with the host and the development of a plant-derived feeding site in plant roots is pivotal to this interaction. Root-knot nematodes form feeding sites composed of giant cells derived from a variable number (3–10) of parenchymal cells (Bird, 1961). A giant cell complex is structurally distinct from the syncytial feeding site formed by a cyst nematode, which is initiated from a single pericycle or procambium cell. Cyst nematodes are dependent on syncytia as their sole sources of nutrition, and failure to develop a fully functional syncytial feeding site can prevent the development of adult female nematodes (Sobczak et al., 1997). Infective second-stage juveniles (J2) of cyst nematodes migrate towards the developing vascular cylinder in the elongation zone of plant roots. On approaching the vascular cylinder, the nematode’s behaviour becomes exploratory, probing root cells by insertion of their hollow stylet to select a suitable initial syncytial cell (ISC; Wyss, 1992). If the nematode does not detect an adverse response from the root cell, for example collapse of the protoplast (Wyss, 1992) or the deposition of callose (Golinowski et al., 1997), a syncytium is initiated. Observations of Heterodera schachtii infecting Arabidopsis thaliana have suggested that successful syncytial development is most likely to occur when the ISC is a procambial cell (Golinowski et al., 1997). The nematode secretes effector proteins through its stylet into the ISC to trigger the formation of a syncytium (Vanholme et al., 2004).

Within 24 h of selecting an ISC, a proliferation of cell cytoplasm occurs with a concomitant reduction in vacuole volume (Magnusson & Golinowski, 1991). Similar changes are induced in neighbouring cells, which fuse with the ISC through openings in the cell wall, formed by the widening of pre-existing plasmodesmata (Grundler et al., 1998). Once the syncytium has become established, cell wall openings are no longer formed via plasmodesmata, but through the local dissolution of the syncytial cell wall and the wall of neighbouring cells. Following cell wall degradation, the fusion of plasma membranes leads to protoplasts of neighbouring cells being incorporated into the syncytium (Sobczak & Golinowski, 2011). In the early stages of feeding site development, the syncytium grows by continually incorporating neighbouring procambial cells, preferentially incorporating cells adjacent to xylem or phloem vessels (Golinowski et al., 1996). Metaxylem precursor cells that have not differentiated into vessels will also be incorporated into the developing syncytium (Golinowski et al., 1996). Xylem vessels that have differentiated cannot be incorporated and the syncytium develops around these vessels (Golinowski et al., 1996). A syncytium is typically composed of c. 200 cells, and extensive hypertrophy of the syncytial elements results in expansion of the feeding site. The syncytium is most highly hypertrophied at the region closest to the head of the nematode. During syncytial development, procambial and pericycle cells proliferate around the developing feeding site; cells not incorporated form a secondary peridermal tissue layer surrounding the syncytium (Golinowski et al., 1996).

To obtain the nutrients required to support nematode development and the production of eggs following mating, female H. schachtii extract the contents of syncytia through their stylet and associated feeding tube. Female cyst nematodes have a high demand for nutrients and have been calculated to take up an amount of solutes that is equivalent to four times the syncytium volume per day (Sijmons et al., 1991). The H. schachtii life cycle is c. 6 wk on A. thaliana, and syncytia remain functional for as long as the nematode continues to feed (Sobczak & Golinowski, 2011).

Transcriptome analysis has revealed extensive changes in the expression of genes involved in cell wall modification during syncytial development (Puthoff et al., 2003; Ithal et al., 2007; Szakasits et al., 2009). However, little is known about syncytial cell wall architectures (Sobczak & Golinowski, 2011) and most knowledge has originated from the study of transmission electron micrograph images (Golinowski et al., 1996; Sobczak et al., 1997; Sobczak & Golinowski, 2011). Primary cell walls are fibrous composites comprising cellulose microfibrils together with a range of noncellulosic heteropolysaccharides (Sarkar et al., 2009). Xyloglucan is an abundant noncellulosic polysaccharide in dicotyledons, such as A. thaliana, and has been proposed to cross-link the cellulose microfibrils and to contribute to cell wall rigidity (Hayashi & Kaida, 2011). Pectins are generally abundant in primary cell walls and are a complex group of polysaccharides, which include homogalacturonan (HG), rhamnogalacturonan-I (RG-I) and rhamnogalacturonan-II (RG-II), which form a multifunctional cell wall matrix (Caffall & Mohnen, 2009). HG consists of a linear chain of (1→4)-α-linked galacturonic acid residues, which can be methyl esterified during polysaccharide synthesis (Wolf et al., 2009). RG-II possesses an HG backbone with complex, but structurally conserved, side chains, whereas RG-I is a highly heterogeneous set of polymers and consists of a backbone of alternating galacturonic acid and rhamnose residues with arabinan, galactan and arabinogalactan side chains (Caffall & Mohnen, 2009). In addition to the polysaccharide components, cell walls also contain structural glycoproteins, such as extensins (Kieliszewski & Lamport, 1994), and cell surfaces, including plasma membranes, may contain arabinogalactan-protein (AGP) proteoglycans that are implicated in cell wall functions (Ellis et al., 2010). The structural composition of cell walls differs between different plant species (reviewed by Cosgrove, 2005); therefore, the structural architecture of uninfected A. thaliana root sections was examined in this study to provide a basis for comparison with the composition of syncytial cell walls.

The isolation of syncytial cell walls is difficult, and therefore in situ analyses are required to understand syncytial-specific cell wall molecular architectures. This study, using in situ fluorescence imaging and a range of monoclonal antibodies to cell wall glycans, in addition to a cellulose-binding module, demonstrates that the cell walls of syncytia induced by H. schachtii in roots of A. thaliana are methyl-HG and arabinan rich. This structural specialization is likely to generate a flexibility of the syncytial cell walls to reflect functional requirements.

Materials and Methods

Plant and nematode culture

Seeds of Arabidopsis thaliana (L.) Heynh. ecotype Columbia-0 were sterilized in 20% (v/v) domestic bleach on a rotational mixer at room temperature for 20 min, followed by five washes in sterile water. Sterile seeds were grown on half-strength Murashige and Skoog medium (Duchefa, Haarlem, the Netherlands) solidified with 1% plant agar (Duchefa). Plants were grown in Sanyo Environmental Test Chambers at 20°C under 16 h : 8 h light : dark cycles. The average light intensity was 140 μmol m−2 s−1 with a humidity of c. 30%. Four seeds were sown per 10-cm plate and plates were stored upright to facilitate the downward growth of roots. Heterodera schachtii Schmidt cysts were extracted from soil, sterilized and hatched as described in Urwin et al. (1997). Following hatching, juvenile H. schachtii nematodes were sterilized in 0.1% chlorhexidine digluconate and 0.5 mg ml−1 hexadecyltrimethylammonium bromide (CTAB) for 32 min on a rotational mixer at room temperature. The nematodes were then washed three times in filter-sterilized tap water with 0.01% (v/v) Tween-20 and re-suspended to a concentration of 1 nematode μl−1. At 2–3 wk post-germination A. thaliana plants were infected with J2 H. schachtii, with 100 nematodes per plant.

Procedures for in situ cell wall analysis

At 14 d post-infection (dpi) with nematodes, root sections infected with female adult nematodes were dissected and fixed overnight in 2.5% glutaraldehyde at 4°C. Following three washes in 1 × phosphate-buffered saline (PBS), root samples were dehydrated using an ethanol series (10% 30 min; 20% 30 min; 30% 30 min; 50% 30 min; 70% 1 h; 90% 1 h; 100% 1 h). Samples were infiltrated with LR White acrylic resin (10% 30 min; 20% 30 min; 30% 30 min; 50% 30 min; 70% 1 h; 90% 1 h; 100% 1 h; 100% overnight; 100% 8 h; 100% overnight; London Resin Company, London, UK) with all incubations performed at 4°C. Individual samples were incubated in 100% LR White in gelatine capsules (Agar Scientific, Stansted, UK) for 5 d at 37°C. Transverse sections (0.5 μm) of root samples were collected using an ultramicrotome and mounted onto Vectabond™-coated (Vector Laboratories, Peterborough, UK) eight-well glass slides (MP Biomedicals, Cambridge, UK). Transverse sections were collected in a sequential manner; for each section incubated with an antibody, a sequential section was incubated with the LM11 antibody to localize xylem vessels before staining with Calcofluor-White. Sections were blocked with 5% (w/v) milk protein in PBS (MP/PBS) for 30 min, followed by incubation in five-fold diluted primary antibodies in MP/PBS for 2 h. Primary antibodies were used in the form of hybridoma cell culture supernatants and are detailed in Table 1. After three washes in PBS, sections were incubated with anti-rat immunoglobulin G linked to fluorescein isothiocyanate (FITC; Sigma-Aldrich, Gillingham, Dorset, UK) diluted 100-fold in MP/PBS for 1.5 h in the dark. Following antibody treatment, all sections were washed three times in PBS before 5 min of incubation in 1 mg ml−1 Calcofluor-White (Fluorescent Brightener 28; Sigma-Aldrich) in the dark. Following several washes in PBS, sections were mounted in glycerol-based Citifluor AF1 antifade (Agar) and examined on an Olympus BX61 microscope equipped with a Hamamatsu ORCA285 camera and Volocity software (PerkinElmer, Waltham, MA, USA). Sections treated with the carbohydrate-binding module, CBM3a protein (Blake et al., 2006), were blocked in 5% (w/v) MP/PBS for 30 min, followed by three washes in PBS, and then incubated with CBM3a at 10 μg ml−1 for 1.5 h, followed by three washes in PBS. This was followed by incubation with a 100-fold dilution of mouse anti-His antibody (Sigma-Aldrich) for 1.5 h. Washes in PBS were carried out before incubation in a 50-fold dilution in MP/PBS of anti-mouse antibody conjugated to FITC (Sigma-Aldrich) for 1.5 h in the dark. Following washes in PBS, sections were mounted in Citifluor and visualized as above. Control experiments were performed for each antibody treatment in which the primary antibody was omitted.

Table 1.   List of primary antibodies used for the immunolabelling of plant cell walls
Primary antibodyTargeted component of the cell wallReference
  1. AGP, arabinogalactan protein; HG, homogalacturonan.

LM11Xylan McCartney et al. (2005)
LM15Xyloglucan Marcus et al. (2008)
LM21Mannan Marcus et al. (2010)
LM19De-esterified pectin HG Verhertbruggen et al. (2009a)
LM20Methyl-esterified HG Verhertbruggen et al. (2009a)
LM6Arabinan Willats et al. (1998)
LM16Processed arabinan Verhertbruggen et al. (2009b)
LM5Galactan Jones et al. (1997)
LM2AGPs Smallwood et al. (1996)
JIM13AGPs Knox et al. (1991)

Some cell wall epitopes may be masked by pectic HG (Marcus et al., 2008). To unmask epitopes, pectic HG was enzymatically degraded before immunolabelling with LM15 and LM21. Sections were initially incubated in 0.1 M sodium carbonate (pH 11.4) for 1.5 h to remove methyl esters; following three washes with deionized water, sections were incubated in recombinant microbial pectate lyase (Cellvibrio japonicas; Megazyme, Bray, County Wicklow, Ireland) at 10 μg ml−1 in 50 mM N-cyclohexyl-3-aminopropane sulfonic acid (CAPS), 2 mM CaCl2 buffer (pH 10) for 2 h at room temperature. Following three washes in deionized water, antibody detection procedures were performed, as already described.


The in situ analysis of internal cell walls within plant organs requires the preparation of thin sections and the application of molecular probes for defined cell wall components – most notably polysaccharides. Transverse sections of A. thaliana roots encompassing feeding sites of the cyst nematode H. schachtii were prepared for immunofluorescence analysis; sections were taken through the most extensively developed region of each syncytium. Figure 1(a) shows a syncytium associated with an adult female nematode at 14 dpi within a region of an A. thaliana root. Fluorescent staining of cross-sections through the syncytium revealed that all cell walls of the stele were readily stained with Calcofluor-White and the cell walls of xylem vessels and the syncytium were strongly stained (Fig. 1b). The hypertrophied syncytial cells are clearly visible and the outer syncytial cell wall is thickened. The most strikingly obvious feature of nematode-infected root sections is the extensive enlargement of the vascular cylinder. The crystalline cellulose-directed CBM3a binds widely to stele cell walls (data not shown). The modifications to the root cell architecture associated with syncytial development occur in the vascular cylinder; therefore, this study focuses specifically on the vascular region of the root.

Figure 1.

Structure of a cyst nematode feeding site in the root of Arabidopsis thaliana. (a) Female adult Heterodera schachtii and syncytium at 14 d post-infection (dpi). Dotted line indicates the location of a typical transverse section; arrowheads indicate the boundaries of the feeding site; asterisk indicates syncytium; arrow indicates adult female cyst nematode. (b) Calcofluor-White staining of a 0.5-μm transverse section through a nematode-infected root section. Asterisks indicate syncytium elements; arrow indicates location of xylem vessels; Pc, procambium; Pd, periderm; Xy, xylem. Bars: (a) 20 μm; (b) 50 μm.

Xyloglucan and mannan epitopes are abundant in syncytial cell walls

The LM11 xylan probe binds specifically to secondary cell walls, which are found in xylem vessels (Fig. 2b). No binding was observed in the cell wall of the syncytium, but this antibody provides a useful marker to distinguish between xylem vessels and syncytial elements. To aid orientation within the stele, Calcofluor-White and LM11 fluorescence images were merged, as shown in Fig. 2, and this approach is used in subsequent figures. A failure to detect the xylan-specific epitope in cell walls of the syncytium indicates that the thickening of the outer syncytial cell wall is not a result of the synthesis of a secondary cell wall. The treatment of syncytial sections with pectate lyase before antibody treatment does not affect binding of the LM11 antibody, indicating that the failure to detect the LM11 epitope in syncytial cell walls is not a result of masking by pectic HG (data not shown). Both xyloglucan and heteromannan epitopes were present in syncytial cell walls (Fig. 3). The LM15 xyloglucan epitope was only detected in the cell walls of infected and uninfected sections following enzymatic removal of pectic HG, indicating that the xyloglucan is effectively masked from protein access in these cell walls. The LM15 epitope was abundantly detected in syncytial cell walls, xylem vessels and cell walls of adjacent cells, but was absent from the walls of cells distal to the syncytium (Fig. 3f). The LM21 heteromannan epitope was also effectively masked by pectic HG, and pectate lyase pretreatment revealed the epitope in the wall of all stele cells (with the exception of xylem vessels), including the syncytial cell walls (Fig. 3j,k). The LM21 heteromannan epitope appeared to be more abundant than the LM15 xyloglucan epitope in the periderm cell walls surrounding the syncytium. In root sections from uninfected plants, both epitopes were detected after pectate lyase treatment, but appeared to be less abundant (Fig. 3a,b).

Figure 2.

Immunolabelling of Arabidopsis thaliana xylem vessels in syncytium sections. The LM11 antibodies bind specifically to xylem vessels (b). To distinguish xylem vessels from syncytium cells, Calcofluor-White images (a) and LM11 antibody staining images (b) were merged (c). Bars, 50 μm.

Figure 3.

Immunolabelling of hemicelluloses in nematode-infected and uninfected Arabidopsis thaliana root sections. LM15 and LM21 antibodies were used to localize xyloglucan and mannan, respectively. Uninfected roots were incubated with LM15 (a) and LM21 (b) following treatment with pectate lyase (PL). Infected sections were incubated with LM15 (e, f) and LM21 (i, j) before and following PL treatment. Images with increased magnification of the syncytium (area of magnification indicated by dotted lines in f) treated with LM15 (g) and LM21 (k) are also shown. Images of Calcofluor-White and LM11 staining of uninfected (d) and infected (h, l) root sections sequential to the sections treated with LM15 and LM21 are also shown. In control experiments, primary antibodies were omitted from the immunolabelling procedure (c). Arrows indicate internal syncytium walls; arrowheads indicate the location of the external syncytium wall. Bars: (g, k) 20 μm; all other images, 50 μm.

Syncytial cell wall pectic HG is extensively methyl esterified

HG is the major component of pectin and is generally synthesized in a methyl-esterified form that can be de-esterified in muro by the action of pectin methylesterases. Monoclonal antibodies LM19 and LM20 are directed to unesterified and methyl-esterified HG, respectively (Verhertbruggen et al., 2009a; Marcus et al., 2010), and these were used in conjunction with section pretreatment with sodium carbonate which removes methyl esters from HG. LM20 bound extremely strongly to the stele and syncytial cell walls of untreated sections (Fig. 4h) and LM19 only bound effectively after the removal of methyl esters (Fig. 4g). These results indicate that syncytial cell walls and the walls of cells in the stele contain pectic HG with a high level of methyl esterification. Some unesterified HG was detected in the stele of uninfected roots (Fig. 4b), but it was notable that no LM19 epitope was detected in the syncytial cell walls before section pretreatment (Fig. 4f). Overall, these observations indicate an elevation of HG methyl ester levels in steles hosting syncytia.

Figure 4.

Immunolabelling of pectin homogalacturonan (HG) in nematode-infected and uninfected Arabidopsis thaliana root sections. The LM19 antibody localized unesterified HG in uninfected (b) and nematode-infected (f) sections. Sections were treated with 0.1 M sodium carbonate (pH 11.4) to remove the methyl groups in pectin HG and re-treated with LM19 (c, g). Methyl-esterified HG was localized with the LM20 antibody (d, h). Calcofluor-White staining of the root sections treated with the LM19 antibody is shown (a, e). Bars, 50 μm.

RG-I epitopes are differentially detected in relation to syncytial cell walls

In general, the second major pectic polysaccharide in primary cell walls in terms of abundance is RG-I, which is a complex multidomain polymer that is highly heterogeneous. In uninfected root sections, the antibody LM6, directed to arabinan side chains of RG-I, bound strongly to cells within the vascular cylinder, with the exception of xylem vessels (Fig. 5b). A similar pattern of binding was observed in syncytial sections, with binding occurring in syncytial cell walls (Fig. 5h). The epitope of the LM16 antibody is a product of arabinofuranosidase activity. It is likely that the antibody recognizes a galactosyl residue or galactan stub on the backbone of RG-I, which is unmasked by the removal of an arabinan side chain (Verhertbruggen et al., 2009b). Unlike LM6, LM16 did not bind to the outer syncytial cell wall, but only occurred in restricted regions of cell walls within the syncytia and was particularly abundant in regions adjacent to cell wall degradation (Fig. 5i). LM16 also bound to phloem cells in infected sections, and the use of this probe to identify phloem cells confirms the previously reported proliferation of phloem sieve elements in response to cyst nematode infection (Juergensen et al., 2003). The LM5 galactan epitope was specifically absent from syncytial cell walls, although binding was observed in other cells of the vascular cylinder (Fig. 5g), indicating a specific and novel element of syncytial cell wall architecture that is distinct from adjacent nonsyncytial cell walls and suggests a syncytial cell wall-specific configuration of RG-I.

Figure 5.

Immunolocalization of pectin rhamnogalacturonan-I (RG-I) in nematode-infected and uninfected Arabidopsis thaliana root sections. The LM5 antibody was used to localize galactan in transverse sections (a, g, j). Arabinan was localized with the LM6 antibody (b, h, k) and processed arabinan was recognized by the LM16 antibody (c, i, l). LM16 also binds to phloem vessels in uninfected root sections and nematode-infected sections, in which the number of phloem vessels has increased (c, i); images with increased magnification of the syncytium are shown in the bottom row. Calcofluor-White and LM11 antibody images (in which the xylem vessels can be distinguished from the syncytial elements) of sections sequential to those treated with the LM5, LM6 and LM16 antibodies are also shown (d, e, f). Arrows indicate internal syncytium walls; arrowheads indicate the location of the external syncytium wall. Bars: (a–i) 50 μm; (j–l) 20 μm.

The JIM13 AGP glycan epitope is specifically detected at syncytial cell surfaces

AGPs are highly glycosylated proteoglycans that are frequently associated with plant cell walls and membranes (Ellis et al., 2010). In uninfected root sections, the LM2 and JIM13 AGP glycan epitopes were both detected at the surface of pericycle cells and weak binding was observed in phloem vessels (Fig. 6b,c). In contrast, these two AGP glycan epitopes occurred differentially at the surface of syncytial cells, with the JIM13 epitope being present at all syncytial cell surfaces (probably at the plasma membrane) and the LM2 glycan epitope not detected (Fig. 6). The JIM13 AGP glycan epitope is therefore the only glycan epitope that has so far been detected at syncytial cell surfaces, but is absent from adjacent nonsyncytial cells.

Figure 6.

Localization of arabinogalactan proteins (AGPs) in nematode-infected and uninfected Arabidopsis thaliana root sections. AGPs were localized using the LM2 antibody, which binds to pericycle cells and phloem vessels (b, e). The JIM13 antibody localized to pericycle cells in uninfected and nematode-infected sections in addition to binding to the syncytium (c, g). Images of Calcofluor-White- and LM11-treated sections sequential to the uninfected (a) and infected root sections (d, f) treated with the LM2 and JIM13 antibodies are shown. Bars, 50 μm.


Cyst nematodes alter the expression of host plant genes to induce the development of a complex feeding cell, which results in extensive alterations to the morphology of root vascular cells. Revealing aspects of the syncytial cell wall molecular architecture is an important step in understanding the structural mechanics and functions of this novel cell type.

The in situ analyses indicated that syncytial cell walls contain cellulose, xyloglucan, heteromannan, methyl-esterified pectic HG and pectic arabinan (Table 2). A failure to detect xylan indicates that the cell walls of the syncytia do not possess a secondary cell wall.

Table 2.   Overview of the immunolocalization of primary antibodies to components of the cell wall in Arabidopsis thaliana root sections infected with Heterodera schachtii
Primary antibodyComponent of the CWRegions of the root in which antibody binding was observed
CW within the syncytiaOuter syncytial CWPericycle cellsProcambiumPhloem elementsXylem vessels
  1. The cell wall component against which each antibody is directed is shown, together with each cell type in which the antibody was detected. +, detection of antibody; −, antibody not detected; AGP, arabinogalactan protein; CW, cell walls; HG, homogalacturonan.

LM19De-esterified pectin HG+++++
LM20Methyl-esterified pectin HG++++++
LM16Processed arabinan+/−+

Cellulose provides the fibrous components of the syncytial cell walls, and this is likely to be cross-linked with xyloglucan and/or heteromannan to provide the load-bearing structural component. Xyloglucan and heteromannan were both effectively masked by pectic HG in the cell walls. This masking could influence the ability of proteins, such as expansins or other modifying enzymes, to gain access to the hemicellulose component of the syncytial cell wall. The expression of expansins in the syncytia of H. schachtii has been well documented (Wieczorek et al., 2006); however, analysis of expansin promoter-β-glucuronidase (GUS) lines revealed that, by 15 dpi, only weak GUS staining was visible in the syncytia.

The pectic HG in the vascular cylinder of infected sections is largely methyl esterified, which is unusual and has implications for cell wall function. Stretches of unesterified HG can form cross-links with calcium ions, which could reduce the porosity of the cell wall matrix and also the extension capabilities of primary cell walls (Willats et al., 2001; Derbyshire et al., 2007). The high methyl ester status of HG in steles hosting syncytia could therefore contribute to the capacity for extension and flexibility of cell walls. It should be noted that the high methyl ester HG does not lead to high porosity in the syncytial cell wall. This was demonstrated by the inability of the LM15 and LM21 antibodies to gain access to the hemicellulose components of the cell wall before the degradation of pectin. The heavily methyl-esterified status of HG in the vascular cylinder was surprising, given that a previous study reported the up-regulation of a pectin methyl esterase (PME) during H. schachtii infection (Hewezi et al., 2008). Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) analysis showed that A. thaliana pectin methylesterase protein 3 (AtPME3) was up-regulated in H. schachtii-infected roots at 13 dpi (Hewezi et al., 2008).

RG-I polymers are implicated in cell wall mechanical properties: an abundance of arabinan is associated with cell wall flexibility, whereas galactan is associated with increased cell wall stiffness (McCartney et al., 2000; Jones et al., 2003). The abundance of heavily methyl-esterified HG in the vascular cylinder of infected sections suggests that the syncytial cell wall is flexible, and the relative abundance of the RG-I polysaccharides, arabinan and galactan, provides further evidence for a flexible cell wall. In situ analyses indicate that the cell walls of syncytia are arabinan rich, and the occurrence of the LM16 epitope is suggestive of arabinan metabolism in these cell walls. Contrastingly, pectic galactan could not be detected in syncytium cell walls. Flexibility of syncytial cell walls may be essential in enabling syncytium functionality. The syncytium is a major nutrient sink for solutes derived from the phloem (Bockenhoff et al., 1996); in addition, the syncytium is highly metabolically active. The accumulation of solutes increases the osmotic value and, consequently, the turgor pressure in the feeding site. Pressure probe experiments calculated a turgor pressure of c. 0.4 MPa in root parenchyma cells; in comparison, the pressure of syncytia formed by cyst nematodes in A. thaliana roots was 0.9 MPa (Bockenhoff et al., 1996). In addition to withstanding a high turgor pressure, the structure of the syncytia needs to be maintained during nematode feeding when the contents of the syncytia are ingested through the nematode’s stylet. A flexible syncytial cell wall may be required to enable the syncytial cell to expand and contract in response to these considerable fluctuations in turgor pressure.

Previous studies have drawn similarities between pollen tapetal cells and syncytia (Hussey & Grundler, 1998; Karimi et al., 2002). Tapetal cells supply nutrition to developing pollen grains (Majewska-Sawka et al., 2004); analogous to syncytia, tapetal cells act as transfer cells, are multinucleate, contain a dense and granular cytoplasm and possess cell wall ingrowths (Jones & Northcote, 1972; Mascarenhas, 1975). However, the carbohydrate composition of tapetal cell walls does not show similarities with syncytium cell walls. Antibodies directed against methyl-esterified HG bound to the cell walls of Beta vulgaris tapetal cells, but the LM6 and LM5 epitopes were not detected (Majewska-Sawka et al., 2004). Syncytia are also structurally distinct from other transfer cells. Primary cell walls of epidermal transfer cells of Vicia faba were found to be relatively abundant in methyl-esterified HG, but both arabinan and galactan were detected at similar levels (Vaughn et al., 2007). In comparison, syncytia are rich in methyl-esterified HG, abundant in arabinan, whereas galactan was not detected. It should be noted that the composition of cell walls is highly diverse between different plant species (Wolf et al., 2012); therefore, these differences in cell wall architecture of syncytia, tapetal cells and transfer cells may be indicative of differences between the plant species studied, as opposed to the specific cell type.

At 14 dpi, the syncytium reaches its maximum size (Urwin et al., 1997); therefore, this study is a characterization of the cell wall architectural requirements for syncytium function, as opposed to syncytium formation. It may be expected that the cell wall modifications required for syncytial development would be completed by 14 dpi; however, the detection of the LM16 epitope in regions of the cell wall adjacent to areas in which the cell wall has been degraded indicates that cell wall degradation is ongoing. Detection of the LM16 antibody is suggestive of arabinan metabolism, as this epitope arises following the enzymatic removal of arabinan side chains from the RG-I backbone (Verhertbruggen et al., 2009b). This may indicate that the degradation of cell walls within the syncytium is still occurring, or the epitope may have arisen during cell wall processing at an earlier stage of syncytial development.

The identification of the JIM13 glycan epitope as specific to syncytial cell surfaces, and not adjacent surrounding cells (although present in pericycle cells), has the potential for a specific marker for syncytial cells. AGPs are complex proteoglycans and their functions remain uncertain (Seifert & Roberts, 2007; Ellis et al., 2010). In this case, the JIM13 epitope may reflect a certain cell type specification system, may be involved in specific processes of cell wall alteration or may represent an aspect of cell wall structural maintenance. One of the difficulties in determining a function for AGPs associated with the JIM13 antibody is the specificity of the antibody epitope. JIM13 binds to the carbohydrate component on AGPs and this epitope may be displayed on different members of the AGP family.

In conclusion, this study provides insights into aspects of the molecular architecture of the cell walls of cyst nematode feeding sites formed in the roots of A. thaliana. The high methyl ester status of HG and the presence of a specific arabinan-rich RG-I (with no detectable galactan elements) are indicative of flexible cell walls that are likely to contribute to the capacity of a syncytium to sustain high turgor pressure and to withstand fluctuations in pressure during nematode feeding.


We would like to thank Sue Marcus for technical assistance. The authors acknowledge the award of a UK Biotechnology and Biological Sciences Research Council (BBSRC) postgraduate studentship to L.J.D.