Loss of cytosolic glucose-6-phosphate dehydrogenase increases the susceptibility of Arabidopsis thaliana to root-knot nematode infection

Abstract

Background and Aims

Root knot nematodes (RKNs, Meloidogyne spp.) are microscopic roundworms with a wide host range causing great economic losses worldwide. Understanding how metabolic pathways function within the plant upon RKN infection will provide insight into the molecular aspects of plant–RKN interactions. Glucose-6-phosphate dehydrogenase (G6PDH), the key regulatory enzyme of the oxidative pentose phosphate pathway (OPPP), is involved in plant responses to abiotic stresses and pathogenesis. In this study, the roles of Arabidopsis cytosolic G6PDH in plant–RKN interactions were investigated.

Methods

Enzyme assays and western blotting were used to characterize changes in total G6PDH activity and protein abundance in wild-type Arabidopsis in response to RKN infection. The susceptibility of wild-type plants and the double mutant g6pd5/6 to RKNs was analysed and the expression of genes associated with the basal defence response was tested after RKN infection using quantitative reverse transcription PCR.

Key Results

RKN infection caused a marked increase in total G6PDH activity and protein abundance in wild-type Arabidopsis roots. However, the transcript levels of G6PDH genes except G6PD6 were not significantly induced following RKN infection, suggesting that the increase in G6PDH activity may occur at the post-transcriptional level. The double mutant g6pd5/6 with loss-of-function of the two cytosolic isoforms G6PD5 and G6PD6 displayed enhanced susceptibility to RKNs. Moreover, reactive oxygen species (ROS) production and gene expression involved in the defence response including jasmonic acid and salicylic acid pathways were suppressed in the g6pd5/6 mutant at the early stage of RKN infection when compared to the wild-type plants.

Conclusions

The results demonstrated that the G6PDH-mediated OPPP plays an important role in the plant–RKN interaction. In addition, a new aspect of G6PDH activity involving NADPH production by the OPPP in plant basal defence against RKNs is defined, which may be involved in ROS signalling.

INTRODUCTION

Root-knot nematodes (RKNs, Meloidogyne spp.) are sedentary plant-parasitic nematodes with a broad host range of over 2000 plants, and RKN damage causes great economic losses each year (Wesemael et al., 2011). RKN infection induces the formation of specific feeding cell structures in the infected roots, disrupting the normal growth of the root system by inhibiting absorption of nutrients and moisture (Wyss et al., 1992). Once hatched from eggs, infective second-stage juveniles (J2s) of RKNs are attracted to and locate host roots guided by root exudates or other environmental signals (Bird, 2004). J2s penetrate the epidermis of root cells in the elongation zone with a hollow needle-like stylet, and then migrate intercellularly through the cortex until they reach the vascular cylinder. There, they secrete effector molecules from oesophageal gland cells and induce the formation of a multinucleate, hypertrophied feeding cells, known as giant cells (GCs) (Mitchum et al., 2013). RKNs rely entirely on these GCs as a major sink tissue to meet their dietary requirement for nutrients, water and energy (Caillaud et al., 2008). During the compatible parasitic interaction, the nematode secretes effector molecules that are believed to play key roles in the parasitism of plants, from suppression of host plant defences to manipulating host plant metabolic processing (Hewezi and Baum, 2013; Marella et al., 2013; Mitchum et al., 2013; Cabello et al., 2014).

To defend themselves against attack by pathogens including nematodes, plants have developed a battery of defence mechanisms such as induction of pathogen-associated molecular pattern (PAMP)-triggered immunity (PTI) and effector-triggered immunity (ETI), generation of reactive oxygen species (ROS), activation of a hypersensitive reaction (HR), synthesis of pathogenesis-related (PR) proteins, callose deposition and cell-wall fortification (Torres et al., 2002; Zipfel, 2014). Similar to the response to the fungal PAMP chitin or the bacterial PAMP flagellin, plant-parasitic nematodes can elicit a nematode-associated molecular pattern (NAMP) response (Vieira et al., 2011). Ascarosides, a group of nematode-specific pheromones, can trigger plant defence responses including mitogen-activated protein kinase (MAPKs)-dependent salicylic acid (SA)- and jasmonic acid (JA)-mediated signalling pathways (Manosalva et al., 2015). In addition, several studies indicate that plants recognize NAMPs via surface-localized pattern recognition receptors (PRRs), and the receptor kinase BAK1 (Teixeira et al., 2016), SlSERK3 (Peng and Kaloshian, 2014), and a leucine-rich repeat receptor-like kinase NILR1 (Mendy et al., 2017), leading to the activation of PTI. In addition, the phytohormones SA, JA and ethylene play crucial roles in plant basal defence responses against RKNs (Molinari et al., 2014; Li et al., 2015b; Hu et al., 2017a).

Plant defence reactions are associated with upregulation of primary metabolic pathways (Rojas et al., 2014). Microarray expression profiling of the roots of host plants infected by nematodes reveals changes in expression of genes in pathways involved in primary metabolism, such as photosynthesis, carbohydrate metabolism including the pentose phosphate pathway and the tricarboxylic acid cycle, mitochondrial electron transport, ATP biosynthesis, nitrogen assimilation, amino acid metabolism, and nucleotide and lipid metabolism (Li et al., 2015a; Xing et al., 2017; Shukla et al., 2018). Genetic evidence suggests a role for lipid metabolism in plant defence responses against parasitic nematodes. For instance, maize (Gao et al., 2008) or Arabidopsis (Ozalvo et al., 2014) mutants with disrupted synthesis of oxylipins are much more susceptible than wild-type plants to RKN infection. However, only a few studies have focused on the contribution of plant primary metabolism to basal defence during plant–nematode interactions.

The oxidative pentose phosphate pathway (OPPP) is an essential metabolic pathway that generates nicotinamide adenine dinucleotide phosphate (NADPH), which is required for carbon fixation, fatty acid and amino acid synthesis, and nitrogen assimilation, as well as for maintaining the redox balance of plant cells (von Schaewen et al., 1995; Wakao and Benning, 2005). Glucose-6-phosphate dehydrogenase (G6PDH), the key NADPH-generating enzyme of the OPPP, is responsible for the conversion of glucose-6-phosphate to 6-phosphogluconolactone. In plants, biochemical and genome-wide analyses suggest that G6PDH molecular variants are divided into cytosolic and plastidic isoforms (Kruger and von Schaewen, 2003; Wakao and Benning, 2005). Extensive studies reveal key roles of G6PDH in responses to biotic and abiotic stresses, such as drought stress (Landi et al., 2016; Wang et al., 2016), salt stress (Dal Santo et al., 2012), cold stress (Lin et al., 2013), ROS production and HR (Asai et al., 2011; Yang et al., 2014) and pathogen infection (Scharte et al., 2009; Yang et al., 2014; Stampfl et al., 2016). NADPH oxidases at the plasma membrane are mainly responsible for ROS production in plants (Torres et al., 2002). Previous studies provide experimental evidence for a role of NADPH oxidase-derived ROS in establishing a biotrophic relationship between Arabidopsis and plant-parasitic nematodes (Siddique et al., 2014; Teixeira et al., 2016). Such oxidative bursts are usually accompanied by transient oxidation of the cytosol (decreased NADPH levels) that triggers redox signalling and activation of the OPPP. However, it is unclear whether the enzyme G6PDH of the OPPP participates in the plant–RKN interaction during the early stage of parasitism.

In this study, we elucidated the role of G6PDH in response to RKN infection in the model plant Arabidopsis. We determined the gene expression and enzyme activity of G6PDH in nematode-infected roots and provided evidence that activation of cytosolic G6PDH is required to cope with RKN stress. Furthermore, analysis of mutants showed that the functional loss of cytosolic G6PDH resulted in enhanced susceptibility to RKN infection, which may be caused partly by suppressing the basal defence response.

MATERIALS AND METHODS

Plant materials and culture

The wild-type Arabidopsis ecotype Columbia (Col-0) used in this study was provided by Dr Y. R. Bi (School of Life Sciences, Lanzhou University). The two single T-DNA insertion mutants g6pd5 (SALK-045083) and g6pd6 (SALK-016157) (Wakao and Benning, 2005) were obtained from the Arabidopsis Biological Resource Center, and the homozygous knockout lines were identified by a PCR-based reverse genetic screen. The double mutant g6pd5/6 (Wakao et al., 2008) was generated using pollen from g6pd5 plants to fertilize the g6pd6 plants and was confirmed by RT-PCR. Seeds were surface-sterilized in 15 % bleach for 15 min, extensively rinsed with sterilized water, and then placed on plates of half-strength Murashige and Skoog (1/2 MS) agar medium (pH 5.7) containing 1 % (w/v) sucrose and 0.8 % (w/v) agar (Biosharp). After 2–4 d at 4 °C, plates were transferred to a growth chamber where they were kept at an angle of approx. 85° in racks to promote unidirectional root growth. The seedlings were maintained at 22 °C, 16/8 h photoperiod, and photosynthetic photon flux density of 100–120 µm m^‒2 s^‒1. After 12 d of growth, Arabidopsis seedlings were used for the nematode inoculation assays.

Nematode culture, attraction and inoculation

Meloidogyne incognita and M. hapla were reared on tomato ‘Zhongshu-4’ in a galsshouse at 22−28 °C. Nematode eggs and pre-parasitic second-stage juveniles (J2s) were collected as previously described (Hu et al., 2017a). Nematode attraction assays were conducted following the method of Hu et al. (2017b). Briefly, 1 mL 23 % Pluronic F-127 (Wang et al., 2009) containing 150 J2s and one root was added into each well of a 12-well tissue culture plate. The plates were transferred to room temperature to allow the gel to solidify. The number of J2s touching the root surface up to 5 mm from the root tip was counted at 2 h using an Olympus SZX-16 dissecting microscope (Olympus Corporation). The experiment was repeated three times with 24 replicates each time. For inoculation, hatched J2s were sterilized in 0.01 % mercuric chloride and 0.002 % sodium azide for 10 min and immediately washed three times in sterilized water; 200 surface-sterilized J2s were then applied near the root tips on the surface of MS agar medium in the Petri plates. At 48 h after inoculation (hai), the roots were collected and were stained with acid fuchsin (Byrd et al., 1983), and the number of nematodes per plant was counted using a stereomicroscope. The number of galls was counted at 21 d after inoculation (dai). In each experiment, 40 plants were tested for each genotype. The experiment was repeated twice.

Quantitative reverse transcription-PCR

RNA was extracted from the roots of mutants and wild type at 2 dai using Trizol (Invitrogen) according to the manufacturer’s protocols. RNA samples were digested using RNase-free DNase I (Invitrogen) to eliminate any contaminating genomic DNA. The purified RNA was transcribed into cDNA using PrimeScript RT reagent Kit (Thermo Fisher). Quantitative reverse transcription PCRs (qRT-PCRs) were performed in the LightCycler 480 System with FastStart Universal SYBR Green Master (ROX) (Roche) according to the procedure described by the manufacturer. The gene-specific primers used are listed in Supplementary Data Table S1. All PCR cycles began with 10 min at 95 °C, followed by 40 two-step cycles comprising 10 s at 95 °C and 1 min at 60 °C. The relative expression levels of specific genes were calculated by the 2^-∆∆Ct method using Arabidopsis 18S gene expression as a reference control. All experiments were performed with three independent biological replicates and three technical repetitions.

Determination of total G6PDH activity

Briefly, 1 g of roots of the inoculated and/or mock-inoculated Arabidopsis seedlings was frozen in liquid nitrogen and ground in a mortar with 1 mL of extraction buffer containing 50 mm Hepes-Tris (pH 7.8), 3 mm MgCl₂, 1 mm EDTA, 1 mm phenylmethylsulfonyl fluoride (PMSF) and 1 mm dithiothreitol (DTT). The homogenates were centrifuged at 13 000 g for 20 min at 4 °C and the supernatant was used for enzyme assays. G6PDH total activity was determined according to the method described by Hauschild and von Schaewen (2003). Absorbance was measured at 340 nm with a spectrophotometer. G6PDH activities were expressed in nmol NADPH/ (min mg⁻¹ protein). A minimum of six technical replicates was performed per experiment.

ROS detection

Diaminobenzidine (DAB) staining was used to detect H₂O₂, a type of ROS. Nematode-infected Arabidopsis seedling roots were incubated in DAB solution (1 mg mL⁻¹ in water, pH 5.8) at room temperature for 5 h in the dark. After cleaning, the stained roots were mounted on glass microscope slides and photographed under an Olympus compound microscope using Cellsens Standard image software (Olympus Corporation). At least 20 replicate plants were analysed and the experiments were repeated twice.

Protein extraction and immunoblot analysis

Total proteins from Arabidopsis roots were extracted by grinding tissues with quartz sand in homogenization buffer (50 mm Tris, pH 7.2, 60 mm β-glycerophosphate, 15 mm nitrophenyl phosphate, 15 mm EGTA, 15 mm MgCl₂, 2 mm DTT, 0.1 mm vanadate, 50 mm NaF, 20 µg mL⁻¹ leupeptin, 1 mm PMSF and 0.1 % Triton X-100). After centrifugation at 12 000 g for 30 min, protein concentration was assessed using the Bradford method (Bradford, 1976). Forty mnicrograms of protein was solubilized and separated by 11.5 % SDS-PAGE and blotted onto Immobilon-P membranes (Millipore). To detect the total G6PD protein level, a 1: 4000 dilution of polyclonal G6PDH antibody (ab993) was added and incubated with the membrane overnight. Alkaline phosphatase-conjugated goat anti-rabbit IgG (Sigma-Aldrich) was used as the secondary antibody. The reaction was detected using a chemiluminescence assay.

Statistical analysis

Data were subjected to one-way ANOVA using SPSS software (SAS Institute). Results are reported as significant or non-significant in Tukey’s t test (Tukey HSD) (P < 0.05).

RESULTS

Total G6PDH activity is enhanced in nematode-infected Arabidopsis wild-type roots

To address whether G6PDH is involved in plant responses to nematode infection, total G6PDH enzyme activity was analysed in uninfected versus infected Arabidopsis wild-type roots at 1, 3, 7 or 14 dai with 200 M. incognita active J2s. G6PDH activity did not show significant change at 1 dai, but was increased 2.5-fold at 3 dai and remained high until 14 dai (Fig. 1A). Changes in protein abundance were also investigated by western blotting analysis using polyclonal antibodies that recognized the cytosolic G6PDH isoforms of Arabidopsis. In uninfected plants, total G6PDH levels were low; however, nematode infection resulted in a rapid and persistent enhancement of G6PDH protein levels over an experimental period of 4 d (Fig. 1B), suggesting that G6PDH is involved in a primary stress response of plants to RKNs.

Fig. 1.

Changes in G6PDH activity and transcript levels in wild-type Arabidopsis (Col-0) responses to root-knot nematode (RKN) Meloidogyne incognita infection. (A) Total G6PDH enzyme activity in root extracts of wild-type plants after RKN infection. Bars indicate s.e. from six technical replicates. Graph shows one representative data set. *P < 0.05 compared to control plants (Col-0) without RKN infection. (B) Protein levels of G6PDH in Arabidopsis roots. Total G6PDH was immunoprecipitated from 50 μg protein extract and subsequently detected by western blotting analysis using a polyclonal G6PDH antibody. The experiment was repeated twice with similar results.

In plants, at least two isoform classes exist, cytosolic and plastidic G6PDH (Von Schaewen et al., 1995). Five members of a gene family comprising six G6PDH isoforms in Arabidopsis have been confirmed to encode active G6PDH (Wakao and Benning, 2005). To ascertain which gene confers the observed increase in G6PDH activity, transcript levels of each G6PDH gene were measured by qRT-PCR at different time points after RKN infection. Interestingly, we found that mRNA expression of all six G6PDH isoforms increases upon RKN infection, with highest induction for G6PD6, one of two cytosolic isoforms, at 2 dai (Fig. 2). It has been shown that Arabidopsis GLYCOGEN SYNTHASE KINASE3 (GSK3)/Shaggy-like kinase ASKα evokes the elevation of cytosolic G6PD6 activity through phosphorylation in response to abiotic and biotic stresses (Dal Santo et al., 2012; Stampfl et al., 2016). This indicates that the increase in total G6PDH activity induced by RKN infection may be also regulated at the post-transcriptional level.

Fig. 2.

Relative expression levels of six putative G6PDH encoding genes analysed by quantitative RT-PCR at 1, 2 and 3 d after root-knot nematode (RKN) Meloidogyne incognita infection (dai). The relative level of mRNA was normalized to reference gene 18S values defined as 1. Error bars represent s.e. *P < 0.05 compared to control plants (Col-0) without RKN infection.

Loss of G6PD5 and G6PD6 function alters the susceptibility of Arabidopsis to RKN

With the aim of understanding the potential role of G6PDH and determining which enzyme activity is up-regulated in nematode-infected roots of Arabidopsis, a genetic approach was adopted using Arabidopsis knockout single mutants (g6pd5, g6pd6) and double mutant (g6pd5/6) for G6PD5 and G6PD6. Homozygous plants for T-DNA insertions of AtG6PD5 and AtG6PD6, respectively, result in the reduction of transcripts and protein levels of the corresponding gene (Wakao and Benning, 2005; Wakao et al., 2008). As our results above indicated that G6PDH activity was induced in infected roots of wild-type plants, we evaluated the changes of total G6PDH activity in mutant roots at 14 dai. After inoculation, total G6PDH activity in the single mutants g6pd5 and g6pd6 was increased by more than two-fold after RKN infection, similarly to that in the infected wild-type roots. However, an approx. 1.5-fold increase in G6PDH activity was observed in RKN-infected double mutant roots (Fig. 3). This may be explained by the redundancy of G6PDH isoforms in increasing activity of the remaining isoform when one is lost. Alternatively, other cytosolic and/or plastidic isoforms could contribute to the increase in total G6PDH activity in the single mutants after RKN infection. Therefore, we used the double mutant for the susceptibility assay in subsequent experiments.

Fig. 3.

Changes of G6PDH activity in Col-0 and mutant plants in response to root-knot nematode (RKN) infection. Total G6PDH enzyme activity in root extracts of wild type, single mutants g6pd5 and g6pd6, and double mutant g6pd5/6 at 14 d after inoculation (dai). Bars indicate s.e. from six technical replicates. *P < 0.05 compared to control plants without RKN infection.

Nematode infection assays indicated that the double mutant g6pd5/6 had a more than two-fold (110 %) increase in the number of galls compared with the wild-type plants at 21 dai (Fig. 4A and B), suggesting that the double mutant is more susceptible to RKN infection than the wild type. We also evaluated the susceptibility of these plants to another RKN species, M. hapla. Similar differences in nematode susceptibility as in M. incognita were observed in the double mutant compared to the wild type (Fig. S1). Previous studies reported that plant nematodes exhibit a rapid response to root exudates of some Arabidopsis mutants, and are attracted to root tips, which increases efficiency of nematode infection (Fudali et al., 2013; Kammerhofer et al., 2015). To exclude the possibility that the increased susceptibility of mutants was the result of alterations in attractiveness, we tested the attraction of RKNs (M. incognita or M. hapla) to Arabidopsis roots. No differences in root attractiveness were found between the wild type and the double mutant (Fig. 4C). Subsequently, we investigated whether the root penetration of J2s is enhanced in mutants. However, we did not observe any difference in the number of J2s inside the roots between double mutant g6pd5/6 and wild type at 2 dai (Fig. 4D).

Fig. 4.

Susceptibility to RKN infection. (A) Percentage of root galls on Arabidopsis double mutant g6pd5/6 relative to the wild-type Col-0 at 21 d after inoculation (dai). Values are mean deviances from control (= 100 %) ± s.e. (n = 30). Asterisks represent statistically significant differences with wild-type Arabidopsis (Col-0) using Student’s t-test (*P < 0.05). (B) Nematode development (dark red) inside acid fuchsin-stained roots of the double mutant g6pd5/6 and the wild-type Col-0 at 21 dai. (C) Number of nematodes touching the terminal 5 mm of the root was counted at 1 h after placing roots in Pluronic F-127 gel containing 100 J2s. Values are the mean ± s.e. of one representative experiment (n = 24). (D) Total number of nematodes inside the root tips of wild-type and mutant seedlings at 24 dai. Values are the mean ± s.e. of one representative experiment. The experiments were repeated three times with similar results.

ROS production is reduced in g6pd5/6 double mutant at the early stage of RKN infection

Early findings indicated a role for G6PDH in the HR and defence responses against pathogens (Asai et al., 2011; Yang et al., 2014). Nematode infection induces an ROS burst in plants that is closely connected with the HR and induction of immunity (Waetzig et al., 1999; Siddique et al., 2014). Therefore, we tested whether the susceptibility of mutants to RKNs is associated with reduced ROS production. As shown in Fig. 5, a deep brown DAB staining was observed in the region of nematode invasion in the root of a Col-0 plant at 1 dai (Fig. 5A). Significantly less browning near the nematode was found in the g6pd5/6 plant than in the wild type (Fig. 5B). These results show that G6PDH may play a role in ROS-induced HR and defence reaction to RKN infection.

Fig. 5.

ROS production in RKN Meloidogyne incognita-infected roots. (A,B) DAB staining of ROS at 1 dai in the infected roots of the wild type (A) and double mutant g6pd5/6 (B). N, nematode. (C) Quantitative RT-PCR analysis of NADPH oxidase gene expression in Arabidopsis roots after M. incognita infection. Data were normalized by using reference gene 18S expression values (the value in RKN-infected roots of Col-0 set at 1). Error bars represent s.e. *P < 0.05 compared to wild-type plants (Col-0).

ROS are mainly generated by NADPH oxidases localized in the plasma membrane (Torres et al., 2002), and loss of NADPH oxidase activity impairs plant resistance to RKN infection (Teixeira et al., 2016). To explore whether RKN infection reduces the NADPH oxidase in mutant roots which may contribute to RKN susceptibility, the expression of NADPH oxidase coding genes RbohD and RbohF at 2 dai was assessed. We found that the RbohD isoform showed similar expression patterns in the double mutant roots compared to the infected wild type, but expression of RbohF exhibited a significant reduction (Fig. 5C). Expression of the APX1, CAT1 and GR1 genes, which encode antioxidant enzymes for scavenging ROS, was also evaluated following RKN infection. Expression of all three genes was inhibited in g6pd5/6 plants compared to wild-type roots at 2 dai with RKNs (Fig. 5C).

G6PDH is involved in Arabidopsis basal defence reaction to RKN infection

As susceptibility was elevated but ROS production was decreased in double mutant g6pd5/6 plants, we further investigated the role of G6PDH in the early defence reaction to RKNs. The expression of genes involved in basal defence response was analysed in RKN-infected wild-type and mutant roots at 2 dai by using qRT-PCR. Transcript levels of WRKY11 and WRKY23 in RKN-infected g6pd5/6 roots were significantly lower than in wild type, but expression of WRKY33 showed a slight increase (Fig. 6). The SA-responsive genes PR1 and PR2 also exhibited lower expression in nematode-infected g6pd5/6 roots than in the wild type. In addition, expression of the JA-dependent gene PDF1.2 was also significantly decreased in g6pd5/6 plants at 2 dai compared to Col-0 plants (Fig. 6).

Fig. 6.

Quantitative RT-PCR analysis of defence-related gene expressions in Arabidopsis roots after Meloidogyne incognita infection. Data were normalized by using reference gene 18S expression values (the value in RKN-infected roots of Col-0 set at 1). Error bars represent s.e. *P < 0.05 compared to wild-type plants (Col-0).

DISCUSSION

It has been widely shown that G6PDH can be induced in plants in response to various abiotic or biotic stress conditions, implying that OPPP has a role in the plant’s response to environment stresses. In the present study, an increase in total G6PDH activity and protein levels was observed in Arabidopsis following RKN infection, suggesting that G6PDH, as a key enzyme for NADPH production by the OPPP, also plays an important role in the host plant–nematode interaction.

In higher plants, G6PDH is classified into plastidic and cytosolic types based on their distribution in cell compartments. Phylogenetic analysis further supports the evolutionary characteristics of the G6PDH protein family (Wakao and Benning, 2005; Cardi et al., 2016). Several studies provided experimental evidence indicating that cytosolic and plastidic G6PDH have different regulatory mechanisms and play distinct roles in plant growth and stress tolerance (Wakao and Benning, 2005; Wakao et al., 2008; Meyer et al., 2011; Cardi et al., 2016). In Arabidopsis, several G6PDH isoforms show different gene expression patterns in tissues. For example, plastidic G6PDH isoforms of AtG6PD2 and AtG6PD3 show highest expression patterns in roots, while expression of the cytosolic isoform gene AtG6PD5 is prevalent in leaves and the mRNA from another cytosolic AtG6PD6 is ubiquitous (Wakao and Benning, 2005; Wakao et al., 2008). Because the tissue-specific gene expression patterns of cytosolic G6PDH often do not match with the enzyme activity (Wakao and Benning, 2005; Wakao et al., 2008), it is commonly accepted that the major regulatory mechanisms of Arabidopsis G6PDH isoforms activity occur at the post-transcriptional level. For instance, G6PDH activity was shown to be regulated by the cellular redox status, the product NADPH/NADP⁺ ratio and protein phosphorylation (Wendt et al., 2000; Debnam et al., 2004; Dal Santo et al., 2012; Stampfl et al., 2016). Convincing evidence suggests that the Arabidopsis ASKα targets cytosolic G6PDH isoforms by phosphorylating the conserved residue Thr-467, thereby stimulating its activity and salt stress tolerance (Dal Santo et al., 2012). Interestingly, although ASKα can phosphorylate G6PD5 and G6PD6 in an in vitro assay, only G6PD6 activity is enhanced by phosphorylation via ASKα, implying distinct mechanisms responsible for regulating activity of two cytosolic G6PDH isoforms (Dal Santo et al., 2012). Moreover, recent studies further indicate that an ASKα–G6PD6 module also contributes to the PTI response and enhances resistance to a bacterial pathogen in Arabidopsis (Stampfl et al., 2016). The above studies suggest that phosphorylation of cytosolic G6PDH isoforms by ASKα or associated kinases plays a key role in plant adaption to abiotic and biotic stresses. During plant–nematode interactions, several host protein kinases such as INTERACTING PROTEIN KINASE (IPK), BAK1, SlSERK3, NILR1 and MAPK are involved in induction of innate immunity to parasitic nematodes (Peng and Kaloshian, 2014; Hewezi et al., 2015; Sidonskaya et al., 2016; Teixeira et al., 2016; Mendy et al., 2017). In the present study, we found that the steady protein levels of the cytosolic G6PDH remain constant upon RKN infection, and the expression levels of AtG6PD6 and AtG6PD5 only increased at 2 dai. This suggests that the activity of cytosolic G6PDH isoforms depends greatly on regulation at the post-transcriptional level during initial/early RKN stress. Thus, it is possible that ASKα-mediated phosphorylation of cytosolic G6PDH isoforms contributes to the increase in RKN-induced G6PDH activity.

Previous studies suggest that the two cytosolic isoforms G6PD5 and G6PD6 exhibit redundancy in single mutants but no change at transcript levels (Wakao and Benning, 2005; Wakao et al., 2008). A compensatory increase in G6PDH activity was also observed in the roots of two single mutants g6pd5 and g6pd6 upon RKN infection, suggesting that other G6PDH isoforms are sufficient to compensate the total G6PDH activity. However, the RKN-induced G6PDH activity in double mutant plants was much lower than that of wild type or single mutants, indicating that the increase in total G6PDH activity after RKN infection might be attributed to the cytosolic G6PDH isoforms. The increase in susceptibility of the double mutant to RKNs further supports the involvement of cytosolic isoforms in the response to RKN infection in plants. In plant cells, the complete OPPP reaction is mainly present in plastids (Schnarrenberger et al., 1995). Ribulose-5-phosphate generated by the cytosolic OPPP is converted to xylulose 5-phosphate (Xul-5-P) by ribulose-5-phosphate 3-epimerase (RPE). Xul-5-P is then transported into the plastid via Xul-5-P/phosphate translocator (XPT) where remaining OPPP reactions are completed (Favery et al., 1998; Eicks et al., 2002). These findings indicated that the cytosolic OPPP reactions are connected with the plastidic OPPP via XPT in the inner envelope membrane. This may explain the results reported by Wakao et al. (2008) that plastidic G6PDH activity decreased in the single and double mutants of cytosolic G6PDH during seed development. Based on these findings together with the implication of up-regulated transcriptional levels of all G6PDH isoforms at 2 dai with RKNs and the prominent total enzyme activities of G6PDH, we cannot rule out the possibility that both plastidic and cytosolic G6PDH activities are activated by RKN infection. Additional research is needed to determine whether the plastidic G6PDH isoforms in Arabidopsis are also involved in the stress response to RKN infection.

ROS-mediated HR efficiently restricts pathogen growth and evokes a plant defence reaction of nearby tissues as well. An ROS burst is often detected during nematode penetration, migration and feeding cell formation (Waetzig et al., 1999; Melillo et al., 2006; Das et al., 2008; Siddique et al., 2014), suggesting that ROS may have regulatory roles both in defence signalling (Melillo et al., 2006; Das et al., 2008) and in aiding establishment of feeding sites during plant–nematode interactions (Siddique et al., 2014). Suppressing G6PDH activity via its competitive inhibitor (glucosamine 6-phosphate) or a cytosolic G6PDH mutation reduced ROS generation and the HR upon elicitor treatment (Pugin et al., 1997; Scharte et al., 2009; Stampfl et al., 2016). Conversely, elevating G6PDH activities through overexpression of cytosolic isoform G6PD6 in Arabidopsis or ectopic overexpression of a plastidic G6PDH from Arabidopsis in the cytosol of tobacco results in the induction of disease resistance to a pathogen (Scharte et al., 2009; Yang et al., 2014; Stampfl et al., 2016). These results indicate that G6PDH participates in the plant defence response via ROS-mediated signal transaction. Similarly, we also found that the double mutant used for disruption of cytosolic G6PDH reduced H₂O₂ production at the early stage of RKN infection and increased the susceptibility compared to the control plant. These findings indicate that G6PDH is involved in the ROS signal transduction to induce a defence response to RKNs. Normally, H₂O₂ production is characteristic of PTI that reflects the extent of plant HR. More recently, it has been shown that an ROS burst is involved in the recognition of nematode-associated patterns through NILR1 and is required for activation of BAK1-dependent PTI pathways in response to RKN infection (Mendy et al., 2017). Hence, the enhanced susceptibility of the double mutant g6pd5/6 to RKNs may be due to impairment in the PTI response.

Additionally, qRT-PCR analysis revealed significantly lower expression levels of RbohF in nematode-infected roots of the double mutant than in wild-type roots, suggesting that RKN-induced ROS reduction in the double mutant may be associated with down-regulation of NADPH oxidase. This is consistent with previous findings that an NADPH oxidase inhibitor (DPI) interfered with a defence-induced ROS burst of the host plant after RKN infection (Melillo et al., 2006), and that the Arabidopsis double mutant rbohD/F displayed enhanced susceptibility to RKN infection (Teixeira et al., 2016). In plants, the OPPP is the major source of NADPH via a key step catalysed by G6PDH. In turn, ROS production through NADPH oxidase is dependent on effective NADPH regeneration in the cytosol, which requires an enhanced OPPP operation and increased G6PDH activity (von Schaewen et al., 1995; Wakao and Benning, 2005). Many reports conclude that G6PDH plays a role in supplying NADPH for ROS generation (Scharte et al., 2009; Asai et al., 2011). Taken together, these findings support that the G6PDH activity of the OPPP is essential for ROS production via NADPH oxidases in Arabidopsis, and could therefore activate an ROS-mediated defence reaction of plant–RKN interactions.

Excessive ROS may disrupt the balance of the cellular redox state and lead to oxidative damage, which is required for stimulating antioxidative defence systems, such as antioxidative enzymes and antioxidants for scavenging ROS. The ROS burst caused by plant nematodes is often accompanied by up-regulation of defence-related enzymes such as superoxide dismutase (SOD), peroxidase (POD), chitinase (CHT) and β-1,3 glucanase (GLU) (Sahebani and Hadavi, 2009; Molinari et al., 2014; Hu et al., 2017a), suggesting an important role of these protective enzymes in ROS scavenging. However, based on the complicated biotrophic relationship between nematodes and their host plant, it is still not clear whether these enzymes are used for protecting the nematode or plant against oxidative stress. Our results showed that the expression of antioxidative genes APX1, CAT1 and GR1 in RKN-infected g6pd5/6 roots was lower than that of control plants, suggesting the lower levels of ROS in g6pd5/6 plants may not be sufficient to completely activate antioxidative defence systems. Intriguingly, the finding that the sugar beet cyst nematode Heterodera schachtii manipulates ROS production through NADPH oxidase to establish syncytia implies that both nematode and host plant in long-term evolution might develop the ability to fine-regulate ROS signal for their advantage (Siddique et al., 2014).

Transcriptome analysis of RKN-infected host roots showed down-regulated expression of genes involved in JA or SA signalling at feeding sites demonstrating that RKN infection can suppress the JA- or SA-dependent systemic defence response (Ji et al., 2013; Portillo et al., 2013). This is further supported by genetic evidence that mutants deficient in SA or JA biosynthesis or signalling are more susceptible to RKN infection (Gao et al., 2008; Nahar et al., 2011; Fujimoto et al., 2015). Our finding that lower expression of the JA-responsive gene PDF1.2 and SA-related genes PR1 and PR2 in the double mutant g6pd5/6 upon RKN infection compared to control roots suggests that G6PDH may be an important player in JA- or SA-dependent systemic defence responses to RKN infection. Previous studies demonstrated that JA biosynthesis starts in plastids but that completion of biosynthesis depends on NADPH provision from OPPP reactions in peroxisomes (Katsir et al., 2008) that might stem from OPPP reactions as well (Hölscher et al., 2016). Because of the physical compartmentation of OPPP reactions in plants (Eicks et al., 2002; Hölscher et al., 2016), it is not clear how different OPPP reactions coordinate in supplying NADPH to plant cells under environmental stresses. Thus, it is reasonable to assume that the loss of cytosolic G6PDH as an NADPH source may result in the reduced JA levels of the g6pd5/6 double mutant, subsequently inhibiting JA signalling and increasing susceptibility to RKNs. It is generally accepted that the SA and JA/ethylene defence pathways are mutually antagonistic in plant–pathogen interactions in leaves (Koornneef and Pieterse, 2008), but the JA/SA pathway counteraction has not been reported in the host–RKN interaction until now. qRT-PCR data in the present work only covers the first 3 d after nematode infection and not the other stages of RKN infection. Plant nematode infection-triggered changes of endogenous hormone production may occur at distinct stages (Kammerhofer et al., 2015). Therefore, the balance between the JA and SA pathways in the response of plants to RKNs is an interesting subject for future research. In this study, we also demonstrated that the expression of WRKY11 and WRKY23 was suppressed in the double mutant upon RKN infection compared to the control. WRKY genes are widely reported to be involved in plant basal defence to pathogen infection (Pandey and Somssich, 2009; Millet et al., 2010). RKN infection results in significantly up-regulated expression of WRKY11. Additionally, both the single mutants wrky11 and wrky17 and the double mutant wrky11/17 exhibited enhanced RKN susceptibility compared with the wild type. These results suggest that WRKY genes may be important regulators of the host defence response to RKN infection (Teixeira et al., 2016). Taken together, these data clearly indicated that G6PDH is involved in basal defence in Arabidopsis upon RKN infection and the reduced expression of WRKY11 and WRKY33 might contribute to the enhanced susceptibility of g6pd5/6.

We have characterized the role of the cytosolic G6PDH isoforms in regulating the plant immune response to RKN. This paper provides direct evidence that RKN infection activates G6PDH, which in turn stimulates the OPPP reactions to supply enhanced demand for NADPH, needed for ROS signalling and the basal defence system of the host plant.

SUPPLEMENTARY DATA

Supplementary data are available online at https://academic.oup.com/aob and consist of the following. Fig. S1: Susceptibility of Arabidopsis roots to Meloidogyne hapla infection. Table S1: Primers for gene expression used in quantitative reverse transcription PCR.