• In plants, UDP-glucuronic acid is synthesized by the oxidation of UDP-glucose by UDP-glucose dehydrogenase or the oxygenation of free myo-inositol by myo-inositol oxygenase (MIOX). In Arabidopsis, myo-inositol oxygenase is encoded by four genes. Transcriptome analysis of syncytia induced by the cyst nematode Heterodera schachtii in Arabidopsis roots revealed that MIOX genes are among the most strongly upregulated genes.
• We have used β-glucuronidase (GUS) analysis, in situ reverse transcription polymerase chain reaction (RT-PCR), and real-time RT-PCR to study the expression of all four MIOX genes in syncytia induced by H. schachtii in Arabidopsis roots. All these methods showed that MIOX genes are strongly induced in syncytia. GeneChip data were analysed for the expression of genes related to the MIOX pathway (mapman).
• Two complementary double mutants were used to study the importance of MIOX genes. Results of the infection assay with double mutants in two combinations (Δmiox1+2, Δmiox4+5) showed a significant reduction (P < 0.05) in the number of females per plant when compared with the wild-type. Furthermore, syncytia in double mutants were significantly smaller than in wild-type plants.
• Our data demonstrate an important role of the MIOX genes for syncytium development and for the development of female nematodes.
Nematodes are a group of animals that include free-living bacterial feeders such as the intensively studied worm Caenorhabditis elegans as well as many pathogens of animals and plants. Plant-parasitic nematodes attack mainly the roots of a variety of plants, often causing severe damage to crop plants either directly or as virus vectors. Some of the most economically important species are the cyst and root-knot nematodes within the family Heteroderidae. The worldwide crop losses caused by nematode damage have been estimated at over $100 billion yr−1 (Sasser & Freckman, 1987). Cyst nematodes (genera Heterodera and Globodera) (Hussey & Grundler, 1998) enter the plant roots as second stage juveniles (J2) and establish a specialized feeding structure (Jones & Northcote, 1972), which is initiated from a single root cell and then expands by incorporating up to several-hundred neighbouring cells by local cell wall dissolution. The nematode feeds only from this syncytium which is thus a severe nutrient sink for the plant. Adult male cyst nematodes leave the root to mate with females. After mating, the female cyst nematode continues to feed but dies once egg development is completed, leaving several-hundred eggs contained within its enlarged body. This subsequently hardens to form a cyst, which protects the eggs until infective J2 hatch in favourable conditions. Root-knot nematodes (genus Meloidogyne) induce a different feeding structure which is composed of several giant cells (Jones & Payne, 1978).
The development of the syncytium from the initial syncytial cell inside the central cylinder is probably initiated through secretions from the nematode and a coordinated expression of plant genes. Such plant genes include, for example, expansins and cellulases that are important for the degradation of cell walls leading to incorporation of new cells into the growing syncytium (Goellner et al., 2001; Wieczorek et al., 2006, 2008). Syncytial cell walls also undergo modification – a process that requires synthesis of new cell wall polysaccharides. Protuberances are produced at the interface between syncytia and xylem vessels and these are thought to be important for the transport of water and solutes (Jones & Northcote, 1972). The outer cell walls of the syncytium are strengthened, which also requires the synthesis of new cell wall polysaccharides. The cells that are incorporated into the syncytium undergo drastic changes in structure and activity. This includes fragmentation of the central vacuole into many small ones, accumulation of mitochondria and ribosomes in a dense granular cytoplasm and proliferation of the endoplasmic reticulum (Golinowski et al., 1996; Sobczak et al., 1997). To cope with this high metabolic activity, the nuclei and nucleoli are enlarged and contain endoreduplicated DNA (Niebel et al., 1996).
These drastic changes in cell morphology imply an underlying global change in gene expression and a variety of methods have been used to identify genes that are specifically induced in nematode feeding sites (reviewed by Gheysen & Fenoll, 2002). During recent years microarrays have become the methodology of choice to allow a global view of the changes in gene expression in nematode feeding sites. The transcriptome of dissected galls induced by the root-knot nematode Meloidogyne javanica on tomato roots was studied using microarrays containing 12 500 cDNAs (Bar-Or et al., 2005). A similar approach has been used to analyse whole root tips of soybean infected with the cyst nematode Heterodera glycines (Alkharouf et al., 2006). Recently, Laser Capture Microdissection has been used to study gene expression in syncytia induced by H. glycines in soybean roots (Ithal et al., 2007a; Klink et al., 2007).
The sugar beet cyst nematode Heterodera schachtii completes its life cycle on Arabidopsis thaliana roots in vitro within 6 wk (Sijmons et al., 1991) and this interaction has been established as a model system. The translucent Arabidopsis roots growing on artificial media facilitate study of the development of this and other nematode species inside the root (Wyss & Grundler, 1992).
We have recently analysed the transcriptome of syncytia induced by H. schachtii in roots of Arabidopsis at 5 and 15 d postinfection (dpi) (Szakasits et al., 2009) and it was observed that two myo-inositol oxygenase (MIOX) genes were among those most strongly upregulated. MIOX converts myo-inositol to glucuronic acid which can be further phosphorylated to produce glucuronic acid-1-phosphate which is then converted to UDP-glucuronic acid (UDP-GlcA). The gene/s coding for the glucuronokinase which phosphorylates glucuronic acid in Arabidopsis is/are not known yet. However, the UDP-glucuronic acid pyrophosphorylase has recently been identified as UDP-sugar pyrophosphorylase (USP, At5g52560). This is probably a single gene and knockouts do not produce viable pollen (Schnurr et al., 2006, R. Tenhaken, unpublished).
UDP-GlcA is an important precursor for several nucleotide sugars which are used for the synthesis of cell wall polysaccharides. In addition to the MIOX pathway, UDP-GlcA can also be produced from UDP-glucose by UDP-glucose dehydrogenase (UGD) that is encoded by four genes in Arabidopsis (Klinghammer & Tenhaken, 2007). The regulation and importance of these different pathways for the production of UDP-GlcA is largely unknown (Seifert, 2004).
A strong indication for the role of the MIOX pathway in providing nucleotide sugars for cell-wall polymers comes from the analysis of Arabidopsis mutant lines for MIOX1 and MIOX2. Both MIOX knockout lines showed a drastic reduction of 3H-inositol incorporation into cell walls when grown in liquid MS medium containing 3H-inositol. This effect was not seen with the MIOX5 knock-out line, which is not surprising because the MIOX5 gene is expressed in seedlings only at a very low level (Kanter et al., 2005). Similarly, overexpression of MIOX4 resulted in increased incorporation of MIOX-derived sugars into cell walls (Endres & Tenhaken, 2009). In addition to producing UDP-GlcA, MIOX might also be involved in the production of ascorbate (Lorence et al., 2004). Overexpression of the flower-specific MIOX4 gene in Arabidopsis plants resulted in elevated ascorbate levels, suggesting a role for glucuronic acid and the myo-inositol oxygenase pathway in vitamin C biosynthesis in plants. However, Endres & Tenhaken (2009) were unable to repeat these experiments.
The finding that two MIOX genes were strongly upregulated in syncytia prompted us to further investigate the role of MIOX genes in the interaction with cyst nematodes. Our results showed that all Arabidopsis MIOX genes were expressed in syncytia and that they play an important role in syncytium development and in the development of female cyst nematodes.
Materials and Methods
Arabidopsis thaliana (L.) Heynh. seeds were surface sterilized for 20 min in 6% (w : v) sodium hypochlorite and subsequently washed three times with sterile H2O. Seeds were placed into sterile Petri dishes (9 cm) on a modified Knop medium with 2% sucrose (Sijmons et al., 1991). Seeds were grown in a growth chamber at 25°C in a 16 h light : 8 h dark cycle.
RNA isolation for quantitative real time reverse transcription polymerase chain reaction (qRT-PCR) analysis
Root segments containing syncytia were excised at 15 dpi and immediately frozen in liquid nitrogen. Control root segments were collected and frozen as described for Affymetrix analysis (described in the next section). Two biological replicates were done, each consisting of approx. 60 syncytia or a corresponding number of root segments. Total RNA was isolated using an RNeasy Plant Mini Kit (Qiagen) according to the manufacturer's instructions, including DNaseI (Qiagen) digestion. Quality and quantity of the RNA was assessed using an Agilent 2100 bioanalyser (Agilent Technologies, Palo Alto, CA, USA). Reverse transcription was performed with a SuperScript III reverse transcriptase (Invitrogen) and random primers (oligo(dN)6) according to the manufacturer's instructions.
Affymetrix GeneChip analysis
Syncytial samples were collected at 5 dpi and 15 dpi by microaspiration. Root segments cut from the elongation zone of 12-d-old uninfected plants were used as controls. Care was taken to avoid any root tips or lateral root primordia. RNA was isolated as described above. Biotin-labelled probes were prepared according to the Affymetrix protocol with some modifications (for details see Szakasits et al., 2009).
Quantitative RT-PCR of gene expression in syncytia
Quantitative real time RT-PCR of MIOX gene expression in syncytia was performed with an ABI PRISM 7300 Sequence Detector (Applied BioSystems) as follows. Each qRT-PCR sample contained 12.5 µl Platinum SYBR Green qPCR SuperMix with UDG and ROX (Invitrogen), 2 mm MgCl2, 0.5 µl forward and reverse primer (10 µm), 2 µl cDNA and water to make a 25 µl total reaction volume. The primers used were as described by Kanter et al. (2005): MIOX1 (5′-CACACCAACTCTTTTGGTCGC-3′; 5′-GTACGATTTAGCTTCTCGTATTCTTC-3′; E = 0.88; R2 = 0.990), MIOX2 (5′-TGATATGAATTTCTTGGGCCATT-3′; 5′-ATCTTGTTAAGTTT TCCATACTCTTTCC-3′; E = 0.83; R2 = 0.995), MIOX4 (5′-GAGATGAATGCATTTGGCCG-3′; 5′-TTTATCTAATTTTCCATATTCAGCCC-3′; E = 0.92; R2 = 0.993), and MIOX5 (5′-GAGATGAACGCATTTGG-TCGT-3′ 5′-CTTGTCCAATTTTCCATACTCACTT-3′; E = 0.87; R2 = 0.990). All samples were diluted 1 : 3 and were analysed in three technical replicates. Control reactions with no cDNA template ruled out false positives and dissociation runs were performed to assess the possible formation of primer dimers. The UBP22 gene was used as an internal reference as described previously (Hofmann & Grundler, 2007). Results were obtained using the Sequence Detection Software SDS v2.0 (Applied BioSystems). Relative expression was calculated by the (1+E)−ΔΔCt method.
β-glucuronidase (GUS) reporter analysis
Promoter regions immediately upstream of the initiation codon of MIOX4 (2185 bp) and MIOX5 (1188 bp) were amplified by PCR using 50 ng Arabidopsis Col-0 genomic DNA as template. Primer pairs used for MIOX4 (5′-ATAAAGCTTTATTTAACCAAAAATGGCATC-3′; 5′- ATACCATGGCTTTTCGAAGAAAGGTTTTTA-3v) and MIOX5 (5′-ATATAAGCTTGGAGGAAGATGAGACTGA-3′; 5′-ATATCCATGGCATCTTCCAAAAAAAAACAAAGT-3′) included Hind III and NcoI restriction sites (underlined) for subsequent cloning into the vector pCAMBIA1303. Promoter::GUS constructs were introduced into Agrobacterium tumefaciens GV3101 for transformation of Arabidopsis Col-0 plants by the floral dip method (Clough & Bent, 1998). Transformed plants were selected on half-strength Murashige and Skoog (MS) medium (Murashige & Skoog, 1962) containing 50 µg ml−1 hygromycin, then transferred to soil for seed collection.
For analysis of GUS expression, Arabidopsis seeds were surface-sterilized for 5 min in 95% ethanol followed by 5 min in 10% (v : v) commercial bleach and subsequently washed three times in sterile H2O. Seedlings were grown vertically on half-strength MS medium supplemented with 1% (w : v) sucrose and 0.6% (w : v) plant agar (Duchefa, Haarlem, the Netherlands) in 10-cm square Petri dishes. Roots of 12-d-old plants were inoculated with c. 35 sterile H. schachtii J2 applied to 1 cm2 pieces of GF/A paper (Whatman, Maidstone, Kent, UK) at each of three infection points per plant.
The GUS expression was analysed at 5, 7, 10, 15 and 20 dpi. Four plants from each of four transgenic lines per construct were analysed at each time-point. Root systems were separated from aerial tissue and submerged in 100 mm NaPO4 buffer (pH 7.0) containing 10 mm EDTA, 0.01% Triton X-100, 0.5 mm K3(Fe(CN)6), 0.5 mm K4(Fe(CN)6) and 1 mg ml−1 5-bromo-4-chloro-3-indolyl glucuronide (X-gluc; Melford Laboratories Ltd, Ipswich, UK). Tissue was vacuum-infiltrated for 5 min then incubated in the dark for 16 h at 37°C. Stained tissue was mounted in glycerol and viewed using bright field optics on a Leica DMRB microscope. Images were captured with an Olympus C-5050 digital camera.
In situ RT-PCR
In situ RT-PCR was carried out according to Koltai & Bird (2000) and Urbanczyk-Wochniak et al. (2002). Syncytia at 10 dpi were dissected from roots and immediately put into cold fixation solution (63% ethanol, v : v; 2% formalin, v : v). After 24 h, syncytia were embedded in 4% low-melting agarose and 25 µm thick sections were prepared using a vibratome (VT 100; Leica, http://www.leica.com/). A RT-PCR was then carried out using digoxigenin-labelled dUTP. After a staining reaction with nitro blue tetrazolium (NBT)/5-bromo-4-chloro-3-indolyl phosphate (BCIP) substrate, cross-sections were photographed under an inverted microscope (Axiovert 200M; Zeiss, http://www.zeiss.com/) with an integrated camera (AxioCam MRc5; Zeiss). For full details see Wieczorek et al. (2006).
Single knockout mutants of all four alleles of MIOX were obtained from the GABI-KAT stock centre (450D10 for Δmiox1) and the SALK institute (040608 for Δmiox2, 018395 for Δmiox4, 112535 for Δmiox5). To obtain the Δ1+2 and Δ4+5 double mutants, the respective homozygous lines were crosspollinated and the resulting heterozygous generation was analysed via PCR for the presence of each intact and disrupted wild-type allele. This generation was allowed to self-pollinate to produce a generation in which the desired genotype of a homozygous double knockout will appear with a frequency of 1/16. A PCR analysis (primer pairs used for screening of single and double mutants are shown in Table 1) was used to identify these individuals. To verify true-breeding, six to eight individuals of their offspring were confirmed as homozygous double knockouts.
Table 1. Primer pairs used for screening of single and double mutants
Wild type MIOX1
Wild type MIOX4
Wild type MIOX5
Quantitative RT-PCR of MIOX gene expression in mutants
Seedlings were grown on wet filter paper or on MS medium. Seven-day-old seedlings were harvested for extraction of RNA. Real time PCR was performed on a Stratagene MX3000 real-time cycler using the SybrGreen method. One reaction (30 µl) consisted of 1× PCR-buffer, a 1:200 000 dilution of SybrGreen stock (Roche), 200 nmol each primer, betaine at a final concentration of 0.6 m, and 1 U Taq polymerase (recombinant wild type). Primers were 5′-CATGTACCTTGTTGCGAAGGAG-3′ (forward); 5′-ACCATTTTAGCTTGGACGGA-3′ for MIOX1; 5′-GCTGTCGTTGGCGATACATTTC-3′; 5′-AGGGTCGTGCCATTCTTCTTAG-3′ for MIOX2; 5′-GGCTGTTGTTGGTGACACATTC-3′; 5′-CGTGTAGCCACTTCAGATTCTC-3′ for MIOX4; 5′-GGCTGTTGTTGGTGACACATTTC-3′; 5′-TAAGCTCCAGCCTTGTGCAATG-3′; for MIOX5, and 5′-GTAACAAGATGGATGCCACCACC-3′; 5′-CCTCTTGGGCTCGTTGATCTG-3′ for EF1α (used for internal normalization). Relative expression was calculated by the ΔΔCt method.
Analysis of cell wall sugars
Leaf material (100–150 mg) was frozen in liquid nitrogen, homogenized and suspended in 70% ethanol. Subsequent extractions with methanol–chloroform and acetone resulted in a cell wall pellet, which was dried and suspended in 800 µl 0.25 m sodium acetate (pH 4.0). The sample was incubated at 80°C for 20 min, chilled on ice, adjusted to a pH of 5.0 with 1 m NaOH, and incubated with α-amylase and pullulanase at 37°C overnight with 0.01% NaN3. The following day, the sample was boiled for 10 min in a water bath, centrifuged at 18 000 g for 5 min, and the supernatant discarded. To remove all residual free sugars, the pellet was washed four times with 1 ml water. After two more washing steps with acetone, the pellet was dried. The weight of the sample was determined and the sample was hydrolysed by autoclaving for 2 h in 250 µl 2 m TFA (Trifluoroacetic acid) and 10 µl 0.5% inositol. Once again the sample was dried and redissolved in 1 ml distilled water per 3 mg of dry sample; 200 µl were then diluted with 300 µl water and analysed via HPLC on a Dionex ICS 3000 system with a PA 20 column. Peak identity was verified with authentic standards.
Heterodera schachtii (Schmidt) cysts were harvested from in vitro stock cultures on mustard (Sinapis alba cv. Albatros) roots growing on Knop medium supplemented with 2% sucrose (Sijmons et al., 1991). Hatching of J2 was stimulated by adding 3 mm ZnCl2. The J2 were resuspended in 0.5% (w : v) Gelrite (Duchefa) and 12-d-old Arabidopsis roots were inoculated under sterile conditions with c. 80–90 J2 per plant. Ten plants per plate were used and experiments were repeated at least three times with c. 40 plants per replicate per line.
The number of males and females per plant were counted at 14 dpi. The data were analysed using single factor ANOVA (P < 0.05). As the F-statistic was greater than F-critical, a Fisher LSD test was applied.
Syncytium size measurement
The size of syncytia (longitudinal sections) was measured at 10 dpi and 14 dpi with H. schachtii. For each line, 10 female syncytia were randomly selected and photographed by an Axiovert 200M (Zeiss) using a Zeiss Axiocam digital camera. The syncytia were outlined using the Axiovision Kontour tool and the area of longitudinal sections was calculated by the software. These individual measurements were used to calculate the average size of syncytia. Data were further statistically analysed using single factor ANOVA (P < 0.05) and LSD.
GeneChip analysis of genes related to the MIOX pathway
We recently performed a transcriptome analysis of syncytia induced by H. schachtii in Arabidopsis roots (Szakasits et al., 2009). This analysis revealed that the genes MIOX4 and MIOX5 were among the most strongly upregulated genes (Tables 2 and S1). MIOX2 was also significantly upregulated in syncytia but had a much higher basal expression level in roots than MIOX4 and MIOX5. MIOX1, the fourth MIOX gene (MIOX3 is a pseudogene) cannot be measured with the Affymetrix ATH1 GeneChip. Because the GeneChip data indicated an important role for the MIOX genes, we studied the expression of all four MIOX genes in detail by using GUS analysis, qRT-PCR and in situ RT-PCR.
Table 2. GeneChip expression profiles of Arabidopsis genes involved in UDP-glucuronic acid (UDP-GlcA) formation during the development of syncytia induced by Heterodera schachtii
Syncytium (5 + 15 dpi)
Control vs syncytium
Data for microaspirated syncytia at 5 d postinfection (dpi) and 15 dpi were combined and compared with control roots. Elongation zone without root tip was used as control. All expression values have been normalized and are on a log2 scale (third and fourth column) and the differences (fold changes) between the pairwise samples displayed (fifth column) are accordingly normalized log2 ratios (see the Materials and Methods section for details).
q-values indicate significance after correction for multiple testing controlling the false discovery rate.
, Significant upregulation or downregulation (false discovery rate < 5%).
Data for MIOX1, UXS2 and UXS4 are not available because the probe sets are ambiguous.
A GUS expression analysis was performed for MIOX4 and MIOX5 (Fig. 1). The general GUS expression pattern was assessed for seven Arabidopsis lines transformed with the MIOX4::GUS construct and 12 lines transformed with the MIOX5::GUS construct. For all lines, expression in uninfected plants was localized to floral tissues with the strongest expression in pollen grains and the stigma. Lower expression was observed in petals, sepals and filament (Fig. 1c). There was generally no expression in other plant tissues, although GUS activity was occasionally observed in < 1% of lateral root bases of MIOX5::GUS lines (Fig. 1b).
Four representative promoter::GUS lines for each gene were infected with nematodes and stained for GUS activity at different time-points after infection, i.e. 5 dpi, 7 dpi, 10 dpi, 15 dpi and 20 dpi. No MIOX4::GUS expression was seen in the vicinity of nematode infections at 5 dpi (Fig. 1a). However, at 7 dpi, expression was switched on in most of the feeding sites and nearly all of the syncytia showed staining at 10 dpi, which became very intense at 15 dpi (Fig. 1a). The GUS staining was localized to the syncytium. No expression was seen in the roots of MIOX5::GUS plants until 10 dpi, when every syncytium showed some degree of GUS expression (Fig. 1b). Similarly, there was strong GUS expression within c. 50% of syncytia at 15 dpi (Fig. 1b). In some cases the GUS staining extended beyond the syncytium into the surrounding root tissue. Representative pictures are shown for all time points. Additional pictures are available in Figs S3 and S4.
The regulation of all four MIOX genes in syncytia was further studied by using qRT-PCR. For this analysis, syncytia were excised at 15 dpi. For comparison with the GeneChip results, the same control roots were used as in that study (see the Materials and Methods section). For MIOX4 and MIOX5 no transcripts could be detected in control roots and it was therefore impossible to formally calculate a fold-change value for these genes (Table 3). MIOX2 was shown to be upregulated in syncytia compared with controls, although the signal intensity was weak compared with GeneChip data. While no chip data were available for MIOX1, qRT-PCR revealed a strong upregulation in syncytia.
Table 3. Change in expression of MIOX genes as measured by quantitative real-time reverse transcription polymerase chain reaction (qRT-PCR) relative to the expression of AtUBP22
GeneChip data (syncytium vs root)
qRT-PCR 15 dpi
M value (log2)
Ct value (log2± SE)
Transcripts were measured from root segments containing syncytia at 15 d postinfection (dpi) and compared with those from control root segments as used for GeneChips. Asterisks indicate significant difference in expression level between infected and noninfected roots. For both MIOX4 and MIOX5 no RNA was detected in the control, making it impossible to calculate a fold change value (indicated as ∞).
SE, standard error.
5.9* ± 1.1
2.50* ± 0.3
Localization of MIOX gene expression by in situ RT-PCR
Localization of MIOX1, MIOX2, MIOX4 and MIOX5 mRNA was investigated by in situ RT-PCR using fresh sections from syncytia at 10 dpi. Transcripts of all four genes were clearly detected in syncytia (Fig. 2a,d,g,j). In control roots, MIOX1, MIOX2 and to a lesser degree MIOX4 transcripts were detected while there was no detectable expression of MIOX5 (Fig. 2c,f,i,l). No products were observed in controls using syncytium root sections without Taq polymerase (Fig. 2b,e,h,k). Thus, all four MIOX genes were expressed in syncytia.
Transcriptional analysis of genes involved in UDP-GlcA synthesis
The role for myo-inositol oxygenase in plants is still unclear. One proposed function is in the synthesis of UDP-GlcA, which is converted to sugar nucleotides as precursors for cell wall polysaccharides. However, UDP-GlcA can also be produced from UDP-glucose by UDP-glucose dehydrogenase (UGD), which is also encoded by four genes in Arabidopsis. We have therefore calculated normalized expression values for 71 genes involved in UDP-GlcA and myo-inositol metabolism (Tables 2 and S1) using GeneChip data for syncytia provided by Szakasits et al. (2009) and visualized these data using the mapman program (Thimm et al., 2004; http://gabi.rzpd.de/projects/MapMan/). For this analysis we combined the 5 dpi and 15 dpi syncytium data and compared them with the controls (Fig. 3a,b).
It is again clearly evident that the MIOX genes are strongly upregulated in syncytia while the single gene for USP is expressed in control roots and syncytia at approximately the same level. This is also the case for all four UGD genes (see also Table 2), although one (UGD1) is rather downregulated. Similarly, UDP-xylose synthase (UXS) is largely not regulated at the transcript level, except for UXS5, which is upregulated. The upregulation of the MIOX genes in syncytia could have resulted from a higher level of the substrate myo-inositol. We therefore extended our analysis to the expression of genes involved in myo-inositol synthesis from glucose-6-phosphate. Genes leading to the synthesis of myo-inositol were preferentially upregulated: One of two genes for inositol phosphate synthase and one of three genes for inositol monophosphatase were upregulated, while one of three genes for inositol polyphosphatase was downregulated (Table 2, Fig. 3b). This would indicate a preferential production of myo-inositol as a substrate for myo-inositol oxygenase.
Analysis of MIOX mutants
Kanter et al. (2005) have recently described single T-DNA mutants for all four MIOX genes. These mutants were crossed to produce a set of corresponding double mutants (see the Materials and Methods section). The PCR characterization of the double mutants is shown in Fig. 4.
The double mutants were indistinguishable from wild-type plants. Flowers and pollen developed normally, the siliques were filled, and seeds were fertile, and germinated readily.
We have used qPCR to test the expression of the MIOX genes in seedlings of the double mutants (Table 4). MIOX1 and MIOX2 were not detectable or at the limit of detection in the miox1+2 mutants. The same is true for MIOX4 and MIOX5 in the miox4+5 mutants. In the miox1+2 mutants an increase in MIOX4 expression and a decrease in MIOX5 expression was observed.
Table 4. Expression of MIOX genes in Arabidopsis seedlings of double mutants (miox1+2; miox4+5) as measured by quantitative real-time reverse transcription polymerase chain reaction (qRT-PCR) relative to the expression of EF1α
Δmiox1+2 (ΔΔCt(log2) ± SE, n = 3)
Δmiox4+5 (ΔΔCt(log2) ± SE, n = 3)
Transcripts were measured from 7-d-old seedlings in double mutants compared with those from the wild type. It was not reasonable to calculate a fold change (indicated as ∞) for MIOX1 and MIOX2 in miox1+2 and MIOX4 and MIOX5 in miox4+5 as the respective transcripts were either absent or at the detection limit.
SE, standard error; n, number of biological replicates.
−0.38 ± 0.46
0.01 ± 0.01
4.65 ± 0.40
−2.39 ± 0.23
In agreement with previous data (Kanter et al., 2005), a substantial loss of MIOX activity was seen in incorporation experiments for the miox1+2 mutant (data not shown). The same experiment cannot serve to assay the decreased activity of miox4+5 because its level in seedlings is negligible. Root anatomy in the double mutants was studied by embedding roots in LR White and staining sections for cellulose with Calcofluor white. No difference between wild-type and the double mutants was observed in uninfected roots and in roots infected with H. schachtii (Fig. S5).
It has also been shown previously that the single mutants had no differences in cell wall sugars compared with wild type (Kanter et al., 2005). We have similarly tested the double mutants (Fig. 5) but both double mutants again showed no significant differences in cell wall composition compared with the wild type.
Although we did not find any difference between the double mutants and the wild-type plants beyond the expression level of the MIOX genes, the strong expression of all MIOX genes in syncytia pointed to an important role for myo-inositol oxygenase in syncytium development. In that case, a strong downregulation of MIOX genes should severely impair syncytium development and lead to enhanced resistance against H. schachtii. To test this hypothesis, the functional role of the MIOX genes in syncytia was evaluated in a nematode infection assay using T-DNA insertion lines. Plants were grown on Knop medium and infected with H. schachtii J2 larvae as described in the Materials and Methods section. Two weeks after inoculation, when females and males could be clearly distinguished, the number of males and females was counted. From that, the infection rate per plant and the ratio of males to females was calculated. No significant difference was observed (Fig. 6) between wild-type plants and any of the single mutants (miox1, miox2, miox4, and miox5). However, because all four MIOX genes were strongly upregulated in syncytia, we reasoned that single mutants might not show a significant effect, because the three remaining genes might compensate for the loss of one gene. Indeed, when we tested the double mutants in the nematode infection assay described above, both double mutants (miox1+2 and miox4+5) showed a significant reduction (P < 0.05) in the number of females per plant compared with the wild type (Fig. 6) while the number of males per plant was not significantly different from the wild type in any of the mutants.
In addition to the number of males and females we measured the size of syncytia at 10 dpi and 14 dpi in the double mutants. Syncytia in the roots of both double mutants were significantly smaller than syncytia in wild-type roots at both time-points (Fig. 7).
Expression of MIOX genes in response to nematode infection
Arabidopsis contains a total of four active genes in the MIOX gene family, MIOX1, MIOX2, MIOX4 and MIOX5 (MIOX3 is a pseudogene). According to GeneChip data available at Genevestigator (http://www.genevestigator.ethz.ch; Zimmermann et al., 2004), MIOX4 and MIOX5 are only strongly expressed in pollen under normal growth conditions while MIOX2 is, in addition, expressed in roots and seedlings. The single USP gene is expressed in pollen but also in all other tissues. The UGD genes are also expressed in almost all tissues but with varying strength (Table S2).
In this paper, we studied the expression of all four MIOX genes in syncytia induced by the beet cyst nematode H. schachtii in roots of Arabidopsis using different techniques. Three of the four MIOX genes are unambiguously represented on the Affymetrix ATH1 GeneChip (MIOX1 is also included but the probes for this gene are not specific) and according to our GeneChip data, MIOX4 and MIOX5 were strongly upregulated in syncytia with barely detectable expression in uninfected roots. The MIOX2 gene is expressed in control roots but its expression was also eightfold upregulated in syncytia. The GeneChip results were supported by GUS analysis, qPCR and in situ RT-PCR. The discrepancy between GeneChip data and qPCR data for MIOX2 can be attributed to the fact that different syncytium material was used in the two cases. For the transcriptome analysis we used aspirated syncytial cell contents while qPCR analysis was carried out with syncytia that were excised from infected roots and thus contained additional root tissue that would cause a dilution effect in the expression of syncytium specific genes. However, obtaining syncytium material by microaspiration is very time-consuming and was therefore not used for the qRT-PCR analysis. MIOX4 and MIOX5 expression was not detected in control roots, which is in agreement with expression data for different Arabidopsis tissues available at Genevestigator (Zimmermann et al., 2004) and the expression data provided by Kanter et al. (2005).
Expression of MIOX4 and MIOX5 in syncytia was also confirmed by promoter::GUS analysis. The expression in roots was generally confined to the syncytium with only a few instances of GUS expression in lateral roots of some MIOX5::GUS lines. In addition, in situ RT-PCR confirmed a strong expression of all four MIOX genes in syncytia while control roots showed weak expression in the central cylinder for MIOX1 and MIOX2, a very weak expression for MIOX4, and no expression for MIOX5. In conclusion, our expression analysis has shown a strong expression of all four MIOX genes in syncytia by a variety of different methods.
Jammes et al. (2005) performed a gene expression analysis of galls induced by the root-knot nematode Meloidogyne incognita in roots of Arabidopsis. Only two MIOX (MIOX1 and MIOX5) genes were included on the microarray that was used in that study and the observed signal intensities were generally much weaker than those reported here. This could be attributed to the fact that they dissected the galls, so that mRNA from giant cells was diluted in the samples by other tissues. Nothing is known about the expression of the other two MIOX genes in response to Meloidogyne species and it is therefore not clear if the MIOX pathway is also important for gall formation and the development of giant cells.
Ithal et al. (2007a) have recently reported the use of Laser Capture Microdissection to collect material for transcriptome analysis of syncytia induced by H.glycines in soybean roots. According to the Unigene database (http://www.ncbi.nlm.nih.gov/sites/entrez?db=unigene), there are five known MIOX clusters in soybean. Two of these have sequence similarity to MIOX4 and one each to MIOX1, MIOX2 and MIOX5 of Arabidopsis (Table S3). Examination of the online supplemental data showed that out of 12 MIOX probesets present on the chip (Table S3) one of them was upregulated in syncytia (Gma.3888.2.S1_at) showing that it might play a role in development and maintenance of the syncytium. Although Ithal et al. (2007a) annotated the ProbesetID Gma.3888.2.S1_at as representing a gene involved in tRNA processing, Unigene and Affymetrix data identify it as a MIOX gene. However, a more detailed study is needed to assess the exact function of this gene. An earlier study by the same authors (Ithal et al., 2007b) which used excised syncytia, showed that Gma.17873.1.S1_s_at, which corresponds to the MIOX2 gene of Arabidopsis, was 20 times upregulated in syncytia compared with controls at 5 dpi.
Characterization of MIOX double mutants
The single mutants used to develop the double mutants were characterized by Kanter et al. (2005). These authors could not find any phenotypic differences between the single mutants and wild-type Arabidopsis plants. The same is true for the double mutants described in this paper under the growth conditions used. We used qPCR to verify lack of MIOX expression in the respective double mutants, This showed that expression of MIOX1 and MIOX2 dropped to the detection limit in the miox1+2 mutant just as did MIOX4 and MIOX5 in the miox4+5 mutant. The increase in MIOX4 expression that was observed in the miox1+2 mutant could indicate the possibility that some of the MIOX isoforms can replace each other. However, this needs to be further investigated in different tissues.
As UDP-glucuronic acid, the final metabolite of the MIOX pathway, is a precursor for cell wall polysaccharides, we tested the cell wall composition in double mutants. As has been found for single mutants (Kanter et al., 2005), we could not find any difference between double mutants and wild-type plants in the cell wall contents for nine different sugars. Furthermore, we looked at the root anatomy of double mutants, either uninfected or infected with H. schachtii. Again, there was no difference between the double mutants and the wild type.
MIOX genes are important for syncytium development
To test the importance of the MIOX genes for syncytium function, we used a set of MIOX T-DNA insertion mutants. As expected, infection of single knock-out mutants compared with wild-type plants showed no significant differences in the number of males and females per plant. This can be explained by the fact that all four MIOX genes are strongly expressed in syncytia allowing the other three genes to compensate for the loss of one gene function. By contrast, double mutants showed a significant reduction in the number of females per plant compared with wild-type plants, showing an important role of MIOX genes for the development of female nematodes. Both double mutants gave similar results, indicating that a high level of MIOX enzymes is necessary for optimal syncytial development and function. A poorly developed syncytium will not provide enough nutrients to support the development of an adult female nematode (Müller et al., 1981). In line with this, we found that the size of female syncytia was significantly smaller in both double mutants compared with the wild type. The MIOX level in single mutants seems to be sufficient for normal function of the syncytium. These results also suggest that all four MIOX genes have most likely the same function, at least in syncytia. Thus, MIOX genes are important for the normal development of syncytia and for the development of female H. schachtii nematodes.
All Arabidopsis MIOX genes are strongly expressed in pollen in addition to syncytia (Kanter et al., 2005). This is confirmed by publicly available microarray data at Genevestigator (Zimmermann et al., 2004) for MIOX2, MIOX4, and MIOX5. By contrast, UGD genes are only expressed at a low level in both tissues, which suggests that UDP-GlcA might be predominantly produced via the MIOX pathway in both tissues and only at a lower level via the UGD pathway. The importance of the MIOX pathway for pollen development is supported by analysis of a knockout mutant in the USP gene that does not produce viable pollen (Schnurr et al., 2006; R. Tenhaken, unpublished).
In Arabidopsis, the major central intermediate for precursors of cell-wall polymers is UDP-GlcA based on polymer analysis of cell walls (Zablackis et al., 1995). Myo-inositol oxygenase (MIOX) is generally supposed to play a role in UDP-GlcA synthesis although other roles (e.g. ascorbic acid synthesis) cannot be ruled out. If we assume that the MIOX pathway in syncytia is important for the production of UDP-glucuronic acid and, ultimately, cell-wall polysaccharides – although we cannot exclude other functions of myo-inositol oxygenase in syncytia – one might speculate that the alternative UDP-glucose dehydrogenase pathway is for some reason blocked, and that UDP-glucuronic acid has to be produced mainly through the MIOX pathway. This is indeed supported by work on UDP-glucose dehydrogenase from soybean nodules, indicating that this enzyme is subject to feedback inhibition by UDP-xylose, one of the nucleotide sugars produced from UDP-glucuronic acid (Stewart & Copeland, 1998). Biochemical feedback inhibition of the UGD enzymes by UDP-xylose (Hinterberg et al., 2002) is superimposed on the transcriptional control of UGD gene expression (Seitz et al., 2000). This mechanism might explain why no explicit changes in the mRNA levels of the UGD genes were detected in syncytia. In addition, one of the UXS genes is upregulated, hinting at an accumulation of UDP-Xylose, a prerequisite for feedback inhibition of UGDs.
Upregulation of the MIOX pathway would require an increased pool of the substrate myo-inositol in syncytia. Indeed, expression analysis of the genes involved in myo-inositol synthesis (Table 2, Fig. 3b) showed that there was a slight transcriptional upregulation of IPS1 (inositol phosphate synthase 1) and IPS2 and downregulation of IPSP (inositol polyphosphatases). Similarly, there was an upregulation of one of the IMP (inositol monophosphatase) genes. Dephosphorylation of Ins(3)P1 by IMP is probably the major route to free myo-inositol in plants. Smart & Flores (1997) generated transgenic Arabidopis plants overexpressing Ins(3)P1 synthase encoded by a TUR1 cDNA from Spirodela polyrrhiza and found these plants to contain elevated Ins(3)P1 synthase activity with a concomitant fourfold increase in endogenous myo-inositol. Recently, Torabinejad et al. (2009) have shown that knocking out an IMP gene results in decreased levels of myo-inositol (and ascorbic acid) in Arabidopsis.
Osuna et al. (2007) have recently shown that MIOX genes are induced during carbon starvation (while IPS1 and IPS2 are repressed) to scavenge alternative carbon sources. These changes are reversed after the addition of sucrose. This regulation of MIOX and IPS genes is in agreement with a decreasing amount of myo-inositol during carbon starvation and its gradual recovery after the addition of sucrose. According to these data, the expression of MIOX genes might be related to the level of sucrose. However, syncytia have been show to contain high amounts of sucrose (Hofmann et al., 2007) which would rather indicate an induction of MIOX genes by sucrose. Thus, the regulation of MIOX genes in syncytia might not be simply correlated with the level of sucrose.
Is the MIOX pathway in syncytia involved in ascorbate synthesis?
In plants, ascorbic acid (AsA) biosynthesis proceeds mainly via the GDP-Man pathway (Wheeler et al., 1998) and most of the available data on the AsA biosynthesis pathway are consistent with this hypothesis. However, there is growing evidence indicating the existence of other pathways operating in plants that contribute to the AsA pool. Conversion of methyl-d-galacturonate and d-glucuronolactone to AsA in detached leaves of several plant species (Loewus, 1963) and Arabidopsis cell cultures (Davey et al., 1999) have been shown by tracer and feeding studies. Lorence et al. (2004) have recently shown that constitutive expression of MIOX4 resulted in an elevated level of AsA, suggesting a role for the MIOX pathway in AsA synthesis. Single knockout lines for MIOX2 and MIOX5 do not have a lower ascorbate level (Kanter et al., 2005) but this could be explained by the redundancy in the MIOX gene family with MIOX1 and MIOX2 and MIOX4 and MIOX5, respectively, having a very similar expression pattern (Kanter et al., 2005). However, Endres & Tenhaken (2009) have recently measured the AsA level of two MIOX4 overexpression lines by high-pressure liquid chromatography (HPLC) and concluded that although MIOX controls the levels of inositol in plants, it does not increase AsA.
Degradation of phytate by purple acid phosphatase (PAP) results in production of free myo-inositol and free Pi which might also lead to AsA synthesis. Recently, it has been shown that overexpression or knockout of a phytase gene (AtPAP15) resulted in an increase or decrease of foliar ascorbic acid levels, respectively (Zhang et al., 2008). However, examination of our transcriptome data showed that there was a downregulation in the expression of AtPAP15 and only a slight upregulation of AtPAP13 (Table 2). To further investigate the role of ascorbic acid in syncytium development, it would be important to compare ascorbate and phytate levels in wild-type plants and MIOX quadruple mutants in relation to nematode infection.
Our studies have shown that all (four) MIOX genes are expressed in syncytia induced by the cyst nematode H. schachtii in Arabidopsis. Knocking out pairs of Arabidopsis MIOX genes in combination resulted in plants that had smaller syncytia and that could only support a reduced number of female nematodes. To further study the function of the MIOX genes it will be important to develop Arabidopsis lines with a significant downregulation of all four MIOX genes in syncytia. Such lines would also facilitate study of the role of the MIOX pathway at the biochemical level (such as inositol and ascorbic acid measurements). This approach could potentially lead to the development of engineered nematode-resistant crops in the future. In this regard, it would also be interesting to knock down the USP gene in syncytia of Arabidopsis using RNAi as it is not possible to raise homozygous lines because the homozygous state is pollen lethal. The two pollen grains in the pollen tetrad that carry the mutated gene atrophy.
We thank Drs Axel Nagel and Björn Usadel for help with the mapman program and Dr Georg Seifert for helpful discussions. We appreciate the excellent technical assistance of Sabine Daxböck-Horvath. This research was supported by grants P16296-B06, P16897-B06, and P20471-B11 of the Austrian Science Fund (FWF). S.S. was supported by Higher education commission (HEC) of Pakistan. D.P.K. gratefully acknowledges support by the Vienna Science and Technology Fund (WWTF), Baxter AG, Austrian Research Centres (ARC) Seibersdorf and the Austrian Centre of Biopharmaceutical Technology (ACBT).