- Top of page
- Materials and Methods
- Supporting Information
Plants are continuously attacked by a variety of pathogens and pests. One group that preferentially attacks the roots is the plant parasitic nematodes. They have developed life styles that make them important pests for many cultivated plants. Migratory nematodes browse the root surface or cortex to parasitize single root cells, whereas sessile nematodes retrieve nutrients only from feeding sites that they induce in the roots of their host plants. Nematode infections can severely reduce the yield of crop plants, and their economic impact has been estimated at $157 billion yr−1 (Abad et al., 2008). Root-knot nematodes, especially the genus Meloidogyne, induce several giant cells from which they feed alternately. Their name is derived from the galls that form around their feeding sites through enhanced division of root cells. Cyst-forming nematodes are the second most economically important group of sessile plant parasitic nematodes, which induce a feeding site that is a syncytium. The female nematodes produce several hundred eggs inside their body, which thereby enlarges to a lemon shape and hardens to form a cyst when the female dies. The eggs can survive in the cyst for many years, making these nematodes difficult to eradicate. The infective juveniles (J2) hatch from the eggs under favourable conditions and infect the roots of host plants.
The development of the syncytium starts from a single root cell inside the vascular cylinder, which is selected by the J2 and pierced with its stylet. It is commonly agreed that the nematode injects proteins produced in its oesophageal glands into the root cell to induce the development of the syncytium, although the nature of these secretions is still largely unknown (Hewezi & Baum, 2013). From this initial syncytial cell the syncytium develops through local cell wall dissolution of neighbouring cells (Wyss & Grundler, 1992). In its final size, a syncytium associated with female nematodes consists of several hundred root cells whose nuclei enlarge through endoreduplication. Furthermore, the ultrastructure of the syncytial elements shows drastic changes as the large central vacuole is replaced by several small vacuoles and increasing cytoplasm containing large numbers of ribosomes and mitochondria (Sobczak et al., 1999). The ultrastructure of syncytia implicates a high metabolic activity, which is necessary to fulfil the needs of the developing nematode, which continuously withdraws nutrients from its feeding site. During the course of these processes, the osmotic pressure of syncytia increases to exceed that of adjacent cells by several fold.
The incorporation of root cells into syncytia requires the dissolution of cell walls, and it has been shown that a number of plant proteins are involved in this process, including expansins, cellulases and pectinases. The expression of these proteins must be tightly regulated, and some of the genes encoding expansins and cell wall-degrading enzymes are specifically induced in syncytia or in the surrounding tissue (Wieczorek et al., 2006, 2008; Szakasits et al., 2009). In addition to processes that degrade cell walls, the development of syncytia also requires the synthesis of new cell wall materials. The outer cell wall of syncytia is thickened to withstand the high osmotic pressure inside syncytia (Golinowski et al., 1996). In addition, cell walls of syncytia that are associated with female nematodes have been shown to develop pronounced cell wall ingrowths at the interface with xylem vessels (Siddique et al., 2012). Similar cell wall ingrowths have been found in the transfer cells of plants (Jones & Northcot, 1972). It has been proposed that their function is to increase the surface, thus allowing a higher exchange of solutions.
In Arabidopsis, the major precursor of cell wall polysaccharides is UDP-glucuronic acid, which can be produced through two different pathways. Under normal growth conditions, the enzyme UDP-glucuronic acid dehydrogenase (UGD) supplies the majority of UDP-glucuronic acid from UDP-glucose. A second pathway involves myo-inositol oxygenase (MIOX), which converts myo-inositol to d-glucuronic acid, which is thereafter converted into d-glucuronic acid-1-phosphate and, finally, into UDP-glucuronic acid, catalysed by glucuronokinase and UDP-sugar pyrophosphorylase (USP), respectively (Supporting Information Fig. S1).
Myo-inositol is produced from glucose-6-phosphate through the rate-limiting conversion to myo-inositol-3-phosphate catalysed by myo-inositol-1-phosphate synthase (MIPS), followed by dephosphorylation to myo-inositol by myo-inositol monophosphatases (IMP; Loewus & Loewus, 1982). In addition to being converted to UDP-glucuronic acid, myo-inositol serves as a precursor for phytic acid, phosphatidylinositol phosphate, myo-inositol phosphates and sphingolipids, which have been implicated in a variety of cellular processes (Irvine & Schell, 2001; Tan et al., 2007). Furthermore, it has been shown that myo-inositol is important for embryo development as a precursor for phosphatidylinositol and phosphatidylinositides, which are essential for auxin-regulated embryogenesis (Luo et al., 2011). Arabidopsis contains small gene families with three genes each for MIPS and IMP. MIPS1 is expressed in most Arabidopsis tissues and developmental stages, whereas MIPS2 and MIPS3 have been found to be especially expressed in vascular or related tissues (Donahue et al., 2010).
Myo-inositol is also a precursor for galactinol through the coupling with UDP-galactose, which is catalysed by galactinol synthase (GS). Galactinol can be further reacted with sucrose to produce raffinose; this reaction recycles myo-inositol. Although there are 10 GS genes in the Arabidopsis genome, there is only one gene coding for raffinose synthase (RS). Galactinol and raffinose have been proposed as osmoprotectants in plants. In line with this, several GS genes are induced by abiotic stresses, and the overexpression of GS increases the drought resistance of transgenic plants (Taji et al., 2002). However, the Arabidopsis RS mutant, which is unable to produce raffinose, but accumulates higher levels of galactinol, did not show any difference in cold acclimation or freezing tolerance, indicating that raffinose is not involved, but that galactinol might play a role (Zuther et al., 2004).
Arabidopsis possesses a small gene family of four genes encoding UGD (UGD1, UGD2, UGD3 and UGD4; Klinghammer & Tenhaken, 2007). UGD1 is weakly expressed in roots, whereas the other three genes (UGD2, UGD3 and UGD4) are strongly expressed in roots. We have recently studied the expression of these genes in syncytia using promoter::GUS lines (Siddique et al., 2012). All four genes were expressed in syncytia, UGD2 and UGD3 as early as 1 d post-inoculation (dpi), whereas the expression of UGD1 and UGD4 was detected starting at 2 dpi. A mutant analysis revealed that the single mutants Δugd2 and Δugd3 support the development of fewer and smaller females and smaller syncytia when compared with wild-type plants. The double mutant ΔΔugd23 showed an even stronger effect than the single mutants. The ultrastructure of syncytia in the ΔΔugd23 double mutant revealed an electron-translucent cytoplasm with degenerated cellular organelles and an absence of cell wall ingrowths in syncytia associated with female nematodes. Thus, UGD2 and UGD3 are needed for cell wall ingrowth formation in syncytia (Siddique et al., 2012).
Four genes in Arabidopsis encode MIOX. MIOX1 and MIOX2 are expressed preferentially in seedlings, whereas MIOX4 and MIOX5 are highly expressed in pollen (Kanter et al., 2005). A quadruple (miox1/2/4/5) mutant that incorporates T-DNA insertions in all four MIOX genes has been described (Endres & Tenhaken, 2011). This mutant showed a severe reduction in transcripts for all four MIOX genes. However, except for MIOX2, transcripts for the other three MIOX genes could still be detected at a level of 2–14% of their abundance in the wild-type. The miox1/2/4/5 mutant did not show any visible phenotype and produced viable pollen. However, it was found that the incorporation of myo-inositol-derived sugars into cell walls was strongly (> 90%) inhibited. All four MIOX genes are expressed at high levels in syncytia (Siddique et al., 2009). Double mutants of the four MIOX genes showed a significantly reduced development of H. schachtii, indicating the importance of the MIOX pathway for the development of syncytia. We therefore suspected that this might be caused by impairment in cell wall biosynthesis; however, we could not detect differences in the cell wall composition of miox double mutants or at the ultrastructural level. Here, we have extended this work using the quadruple mutant. We provide evidence that the importance of MIOX in syncytium development is not the production of cell wall precursors, but rather the removal of excess myo-inositol from syncytia.
- Top of page
- Materials and Methods
- Supporting Information
The development of nematode-induced syncytia is accompanied by cell wall dissolution, but new cell wall polysaccharides are also required to enable syncytium expansion and full functionality. The major precursor of plant cell wall polysaccharides in plants is UDP-GlcA, which can be produced through two different pathways. Under normal growth conditions, the enzyme UGD supplies the majority of UDP-glucuronic acid from UDP-glucose. This hypothesis is also supported by our recent work in the context of plant–nematode interaction, which demonstrated that UGD genes are necessary for the formation of cell wall appositions in syncytia associated with female nematodes (Siddique et al., 2012). Alternatively, the MIOX biosynthetic pathway produces UDP-GlcA from myo-inositol. The Arabidopsis genome contains four MIOX genes and all four are strongly expressed in syncytia (Siddique et al., 2009), pointing to an important function for the MIOX pathway in syncytia. Indeed, we showed that double mutants with T-DNA insertions in two of the four MIOX genes were less susceptible to infection by H. schachtii (Siddique et al., 2012). However, the quadruple mutant, which has been tested in the present study, did not show a further decrease in susceptibility relative to the double mutants tested previously. As UDP-GlcA is an important precursor of several plant cell wall polysaccharides, we reasoned that the miox mutants might be affected by cell wall modifications. However, we could not detect differences in cell wall composition or ultrastructure of syncytia in the double mutants (Siddique et al., 2009) or quadruple mutant (Fig. S2). This was surprising, but we found that this could be explained by an upregulation of UGD genes in the quadrupule mutants. As UGD is involved in the alternative pathway for UDP-GlcA production as a precursor of cell wall polysaccharides, the upregulation of this pathway could at least partially rescue the MIOX quadruple mutant. It is also possible that residual expression of the three MIOX genes in miox1/2/4/5, together with the upregulation of two UGD genes, meets the minimal level of production of cell wall polysaccharides required for syncytium development. Nevertheless, the reduction in MIOX transcript levels in the double and quadruple mutants reduced significantly the susceptibility to H. schachtii. Therefore, the importance of the MIOX pathway does not seem to be a result of defects in cell wall synthesis. This raised the question of whether MIOX might be involved in other pathways important for syncytium development, other than the production of UDP-GlcA. This could be tested by knocking out the genes coding for glucuronokinase and/or USP, respectively, which act downstream of MIOX. Unfortunately, no mutants for glucuronokinase are available and the mutant for USP is pollen sterile and thus no homozygous mutants can be obtained. The available USP RNAi lines tested did not show any decrease in susceptibility, which might be caused by the fact that the USP enzyme activity level is only downregulated to c. 25% of the wild-type level (Kotake et al., 2007). All these results suggest that the reduced susceptibility of miox mutants is not caused by a decrease in cell wall polysaccharides. Another explanation for the reduced susceptibility of miox mutants might be a reduced level of AsA in syncytia. It has been postulated that the MIOX pathway is involved in the production of AsA (Lorence et al., 2004). However, our results showed that the level of AsA in the syncytia and roots of the miox quadruple mutant was not significantly different from that of wild-type tissues. This showed that the MIOX pathway was not involved in AsA production in roots and syncytia, and thus the change in susceptibility of the MIOX quadruple mutant was not caused by a reduced AsA level.
Thus far, therefore, our results could not explain the reduced susceptibility of miox mutants. Therefore, we took an unbiased approach and carried out metabolite profiling of syncytia developing in the roots of the quadruple mutant. The work reported here was performed with plants grown on Knop medium containing 1% sucrose (Sijmons et al., 1991), which is generally used for research on Arabidopsis–nematode interactions, because it allows the easy observation and extraction of syncytia. Not much is known about gene expression in syncytia from soil-grown Arabidopsis plants.
Compared with the wild-type, the content of myo-inositol was increased in the syncytia induced in mutant roots, as was also confirmed by high-performance liquid chromatography (HPLC). Similarly, the contents of myo-inositol phosphate (direct precursor of myo-inositol) and glucose-6-phosphate (a precursor for myo-inositol phosphate), were much higher in syncytia developing in the quadruple mutant.
Is the high level of galactinol the reason for the reduced susceptibility of miox mutants?
Galactinol was another metabolite highly abundant in syncytia induced in the miox quadruple mutant. It is produced through the coupling of myo-inositol and UDP-galactose. A further reaction with sucrose produces raffinose and recycles myo-inositol. In the miox quadruple mutant, the high myo-inositol content favours the synthesis of galactinol, but did not result in an increase in the level of raffinose. Thus, it was possible that the high galactinol content in the miox mutants was the reason for the reduced susceptibility. We tested this possibility by increasing the content of galactinol independent of mutations in the MIOX pathway. Lines that overexpressed a GS gene from cucumber reached approximately two-fold higher galactinol levels, but did not show any statistically significant change in susceptibility to H. schachtii. However, a mutant of RS (RS14), with blockage in the final step leading to raffinose, supported smaller syncytia than did wild-type plants. Similarly, the female nematodes developing on these plants were also much smaller. The difference between the RS14 mutant and the GS-overexpressing lines was that the RS14 mutant had an approximately four-fold higher galactinol level (Zuther et al., 2004), thus indicating that this high galactinol level might have been responsible for the observed effects. However, it could not be excluded with certainty that the reduced raffinose level in the RS14 mutant might have been the reason for the observed effects. We therefore treated the seedlings with 1, 5 and 10 mM galactinol to directly increase the galactinol level. This treatment also led to smaller syncytia and smaller female nematodes and to a decrease in the number of nematodes, which showed that the effect in the RS14 mutant was caused by the higher level of galactinol and not the reduced level of raffinose. There was no statistically significant difference between the three galactinol concentrations (with the exception of 1 mM which did not result in smaller syncytia). The explanation for this effect might be that only a limited amount of galactinol is taken up into syncytia. Recently, galactinol has been shown to be involved in resistance signalling in addition to its involvement in abiotic stress responses. A cucumber GS gene was found to be associated with priming induced by Pseudomoas chlororaphis O6 root colonization, leading to an increase in galactinol content. This effect could be copied by exogenous galactinol application (Kim et al., 2008). In Arabidopsis, one GS gene was specifically induced by Botrytis cinerea infection and by priming with P. chlororaphis O6 root colonization. This resistance was mediated through the JA pathway (Kim et al., 2008; Cho et al., 2010). Thus, galactinol might be a signalling compound for induced resistance, although it could not be excluded that the active compound might be a product of galactinol, such as raffinose. This view is supported by our results, showing that the expression of the Thi2.1 gene was induced in syncytia formed in roots of the miox quadruple mutant. According to our transcriptome analysis (Szakasits et al., 2009), Thi2.1 is not regulated in wild-type syncytia and roots. This gene codes for an antimicrobial thionin peptide and has been shown previously to be involved in the resistance against pathogens and is regulated through JA (Epple et al., 1997; Bohlmann et al., 1998). However, the JA/ethylene marker gene PDF1.2 (Thomma et al., 1999) and the SA marker gene PR1 (Uknes et al., 1992) were not upregulated in syncytia of the miox quadruple mutant, indicating that the galactinol effect in syncytia does not depend on the ethylene or SA pathways.
Further work is needed to discover which other genes might be upregulated by galactinol and might be involved in the reduced susceptibility of the miox quadruple mutant to H. schachtii. It is also possible that other defence-related genes are activated during the early stages of infection and syncytium development. Furthermore, using mutants of different resistance pathways should indicate whether galactinol acts only through the JA pathway in syncytia during plant–nematode interaction.
In this work, we have analysed the importance of the MIOX pathway for the development of syncytia induced by H. schachtii. Our results showed that the downregulation of the MIOX pathway in syncytia led to an increased galactinol level. The increased galactinol level might be responsible for the reduced susceptibility of miox quadruple and double mutants, most probably via an upregulation of defence-related genes.