Potassium Sulphate Induces Resistance of Rice against the Root-Knot Nematode Meloidogyne graminicola

Background Potassium (K), an important nutrient element, can improve the stress resistance/tolerance of crops. The application of K in resisting plant parasitic nematodes shows that the K treatment can effectively reduce the occurrence of nematode diseases and increase crop yield. However, data on K 2 SO 4 induced rice resistance to Meloidogyne graminicola are still lacking. To evaluate rice resistance against M. graminicola induced by K2SO4 and to further clarify its mechanism is essential for the rational use of K fertilizer to ensure the safety of rice production.

pathway, resulting in slow development and reduced number of RKNs [4]. Soares et al. (2020) con rmed this result by foliar spraying of clay on rice [8]. The treatment of potassium nitrate (KNO3) in roots could induce the activities of resistance-related enzymes and the production of disease-resistant substances in tomato (Solanum lycopersicum L.), and signi cantly reduce the disease index of RKN M. incognita [9]. Xiang et al. (2018) demonstrated that potassium fertilizer application in soil could stimulate the secretion of phenic acid (cinnamic, ferulic and salicylic acids) in soybean (Glycine max (L.) Merr.) root system, and enhance the expression of disease-resistant genes such as phenylalanine ammoniase (PAL) and polyphenol oxidase (PPO), which effectively controlled soybean cyst nematode Heterodera glycines [10]. Application of K 2 SO 4 in the eld can reduce the damage of brown rot disease to potato [11], and induce the expression of defense-related genes in snap bean leaf tissue [12]. However, the data on the induction of host resistance to nematodes by K 2 SO 4 is lacking.
In this work, the effect of K 2 SO 4 on the resistance of rice to M. graminicola was evaluated by pot experiment in greenhouse, and the histopathological changes of rice treated with K 2 SO 4 were observed by microscopy. After the effect was con rmed, its role in priming the basic defense reaction was determined. Then, the expression levels of disease resistance-related genes were quanti ed to preliminarily explore the molecular mechanism. Finally, the potassium (K + ) channel gene OsAKT1 and transporter gene OsHAK5 de cient lines were used to verify whether the OsAKT1 and OsHAK5 proteins are involved in the host resistance against M. graminicola. Results K 2 SO 4 did not impact the mortality and infectivity of M. graminicola at low concentrations The direct toxicity of K 2 SO 4 to M. graminicola was determined by calculating the mortality of nematodes at different concentrations of K 2 SO 4 solution 72 h after treatment. Although the mortality of nematodes increased along with the increase of K 2 SO 4 concentrations ranging from 0.125 mM to 8 mM, there was no signi cant difference compared to the control water treatment (5.0 ± 0.7%), especially at 0.125 mM K 2 SO 4 concentration (5.7 ± 1.4%) (Fig. 1a). These results indicated that the K 2 SO 4 was not toxic to M. graminicola at low concentrations within 72 h.
To determine whether K 2 SO 4 directly impact the infection and development of M. graminicola, the newly hatched second stage juveniles (J2s) were soaked in 0.5 mM K 2 SO 4 solution for 48 h before inoculation. At 14 days post inoculation (14 dpi), no difference was observed in the total numbers of juveniles and adult females between the K 2 SO 4 -soaked and the water-soaked groups, which indicated that there was no direct effect of K 2 SO 4 on the infection of M. graminicola. At the same time, most nematodes developed into adult females, and there was no signi cant difference in the numbers of third stage juvenile (J3), fourth stage juvenile (J4) or adult females between the two groups (Fig. 1b, c). These results indicated that K 2 SO 4 did not suppress the development of M. graminicola. K 2 SO 4 treatment induced resistance of rice to M. graminicola To evaluate the resistance of rice against M. graminicola induced by K 2 SO 4 ,different concentrations of K 2 SO 4 solution treated the rice roots 1 d before inoculation, and preliminary plant infection rate was assayed at 14 dpi. The results showed that compared to the water treatment, K 2 SO 4 treatment could effectively reduce the numbers of both galls and nematodes (juveniles and adult females) at the concentration of 0.125 mM, and there was no signi cant difference among the K 2 SO 4 treatments. After applying with K 2 SO 4 solution at a concentration of 0.5 mM, the numbers of both root galls and nematodes were decreased signi cantly (57.2 ± 4.4%, 59.2 ± 6.6%) at 14 dpi (Fig. 2a, b). In addition, compared to the untreated control plants, the development of nematodes treated with K 2 SO 4 was evidently delayed. At 14 dpi, the proportion of adult female in plants treated with K 2 SO 4 (70.9 ± 5.6%) was signi cantly lower than that of plants in water control (90.7 ± 5.1%), and the proportion of juveniles (J3 + J4) (27.0 ± 6.3%) in plants treated with K 2 SO 4 was higher than that of control plants (6.0 ± 3.2%) (Fig. 2c, e). These results indicated that the low concentrations of K 2 SO 4 could effectively reduce the infection of nematodes and delay the development of nematodes. Therefore, the 0.5 mM K 2 SO 4 solution was used in all the subsequent experiments. We also calculated the male/female ratio of nematodes, indicating that the ratio in rice roots treated with K 2 SO 4 was lower than that in the untreated roots (Fig. 2d). Meanwhile, there was no signi cant difference in the fresh weight and height of rice shoots or roots between K 2 SO 4 treatment and water treatment (Fig. 2f, g), suggesting that application of K 2 SO 4 had no toxic effect on rice growth. All these results indicate that the application of K 2 SO 4 can induce rice resistance against M. graminicola. K 2 SO 4 -induced resistance of rice to M. graminicola was achieved by neither impacting the attractiveness of rice roots to nematodes nor inhibiting the developmental phenotype of giant cells In nature, attraction of the host roots to nematodes is of great importance in establishing infection [13]. Here, the nematodes in the range of 5 mm around the root tips treated with K 2 SO 4 solution were counted.
At 10 hours post inoculation (hpi), the numbers of nematodes attracted to the K 2 SO 4 treated tips (26.2 ± 3.1) were not different from the water treated tips (25.7 ± 2.8) (Fig. 3a, b). The results indicated that K 2 SO 4 treatment did not affect the attraction to M. graminicola.
The invaded J2 can induce the plant to form giant cell at the feeding site by secreting effectors from esophageal gland through stylet [14], and to acquire the nutritional needs of subsequent growth and development. Then the giant cells develop to form the root galls. In this experiment, the root galls of 7 dpi were collected and used for histological observation. The average cross-sectional area of giant cells in K 2 SO 4 treated group was not signi cantly different from the untreated group (Fig. 3c, e). We also counted the numbers of giant cells in feeding sites; there was no signi cant difference between the two treatments ( Fig. 3d). All these results indicated that the application of K 2 SO 4 did not in uence the development of giant cells. K 2 SO 4 induced resistance to M. graminicola in Rice by priming basal defense Callose deposition is one of the markers of plant defense response, which can play a role in resistance to pathogen infection [15]. We observed the presence of callose in the gall tissue induced by RKNs after staining with aniline blue. The results showed that compared with the water treatment group, more callose was deposited in the K 2 SO 4 treatment group, and the callose spots were larger and more dense ( Fig. 4a, b), with 67.9% of increase in the average area of callose spots (Fig. 4a). These results suggest that increased callose deposition is one of the pathways of K 2 SO 4 -induced resistance in rice plants.
Hydrogen peroxide (H 2 O 2 ) is a signaling molecule that mediates plant defense response to biological or abiotic adversity [16]. Our analyses indicated that the H 2 O 2 response of rice roots could also be induced by K 2 SO 4 alone, because the H 2 O 2 level at 24 hpi was signi cantly higher than that of the control group.
In the case of nematode challenge, compared with the rice without K 2 SO 4 treatment, the level of H 2 O 2 in treated rice was increased by 78.2% and 118.7% at 8 and 24 hpi, respectively, but it decreased at 72 hpi ( Fig. 4c). In addition, the expression of OsRbohB, an H 2 O 2 synthetic gene known to be involved in plant immune response [17], was increased in the K 2 SO 4 -treated rice at 8 and 24 hpi. Especially, at 8 hpi, the expression level of OsRbohB in the inoculated plant treated with K 2 SO 4 was signi cantly higher than that in the non-treated (P ≤ 0.05) (Fig. 4d), which was consistent with the results of H 2 O 2 content (Fig. 4c).
These results suggest that the application of K 2 SO 4 can induce defense against M. graminicola by activating the rapid production of reactive oxygen species (ROS). K 2 SO 4 treatment promoted the expression of disease resistance-related genes in inducing rice resistance against M. graminicola While the plant is attacked by the pathogens, it will activate the autoimmune response system to defend. Immune response is a complex process involving the transcription and expression of many disease resistance-related genes [18]. We analyzed the expression of several common disease resistance-related genes in different treatments and at different time points.
Firstly, we analyzed two salicylic acid (SA) pathway-related genes: SA transcription factor OsWRKY45 and biosynthesis gene OsICS1. The expression of both OsWRKY45 and OsICS1 was signi cantly increased by K 2 SO 4 treatment alone at 72 hpi, however, after inoculation, K 2 SO 4 treatment only induced OsWRKY45 at 24 hpi. OsICS1 was not prominent at any time points (Fig. 5a). Therefore, SA pathway is not the main pathway in K 2 SO 4 -induced rice resistance against M. graminicola. Secondly, we analyzed two jasmonic acid (JA) pathway-related genes and two ethylene (ET) pathway-related genes. After treatment with K 2 SO 4 , JA transcription factor gene OsJAMYB was signi cantly increased in both inoculated and non-inoculated groups at 8 hpi (Fig. 5b). The expression pattern of JA biosynthesis gene OsAOS2 was similar to that of OsJAMYB, but it was induced more than OsJAMYB, at 8 hpi (Fig. 5b). Although the ET signaling gene OsEIN2 was not expressed much, it was expressed signi cantly higher in K 2 SO 4 -treated inoculated rice than in untreated. The expression of OsEIN2 was slowly increased at all the Page 7/21 three time-points, while the expression of ET biosynthesis gene OsACS1 was increased sharply in the K 2 SO 4 -treated and MG-inoculated group at 72 hpi, which was 6.98 times higher than that of untreated but MG-inoculated group (Fig. 5c). These results suggest that K 2 SO 4 plays an active role in inducing the expressions of JA and ET pathways-related genes in rice resistance to RKNs. Thirdly, we analyzed the expression of pathogenesis-related (PR) genes. The expression of both OsPR1a and OsPR1b in the K 2 SO 4 -treated rice was signi cantly increased compared to the untreated group at 8 hpi. The expression of OsPR1b continued to be increased in the inoculated rice treated with K 2 SO 4 after 8 hpi, and was still signi cantly higher than that in the untreated at 24 hpi (Fig. 5d). These results suggest that K 2 SO 4 positively regulates the expression of these two genes in rice resistance to RKNs.
Finally, we analyzed the expression of two brassinolide (BR) pathway-related genes. The expression of BR receptor gene OsBRI1 was signi cantly increased in both inoculated and uninoculated K 2 SO 4 -treated groups at 24 hpi and 72 hpi, while the expression level of BR biosynthesis gene OsDwarf in the inoculated group treated with K 2 SO 4 was signi cantly higher than the untreated at 24 hpi (Fig. 5e). This indicates that BR pathway also plays an active role in the resistance process, and K 2 SO 4 positively regulates the expression of BR pathway-related genes.
The resistance of rice against nematodes induced by K 2 SO 4 required the participation of potassium absorption and transportation systems To determine the role of potassium (K + ) absorption and transportation systems in disease resistance process, we rst compared the expression of K + channel gene OsAKT1 and K + transporter gene OsHAK5 between K 2 SO 4 application and non-application, and between nematode inoculation and non-inoculation.
The expression of OsAKT1 in the inoculated rice treated with K 2 SO 4 was higher than that in both the untreated and in the non-inoculated within 72 h, especially at 8 h and 72 h (Fig. 6a). This suggests that OsAKT1 expression is affected by K 2 SO 4 and nematodes, which may be involved in the resistance process of rice against nematodes induced by K 2 SO 4 . Meanwhile, the expression of OsHAK5 was upregulated in all the treatments within 24 h after inoculation. Although there was no signi cant difference among the treatments, the expression of OsHAK5 was slightly higher in the inoculated rice than in noninoculated, and the expression of OsHAK5 was slightly higher in the inoculated rice treated with K 2 SO 4 than in the non-treated (Fig. 6a). This indicates that nematode infection and K 2 SO 4 treatment have slight effects on OsHAK5.
Subsequently, we inoculated nematodes on the rice lines de cient in these two genes. Interestingly, in the presence of K 2 SO4 at a concentration of 0.5 mM, both of the two defective plants (OsAKT1-RNAi and OsHAK5-Cr) became more susceptible than the wild-type plants. Speci cally, compared with the two defective plants, the numbers of both galls and nematodes (juveniles and adult females) in the wild-type plants roots were reduced, and the proportion of juveniles (J3 + J4) were increased, while the proportion of adult females was reduced. (Fig. 6b, c). These results indicate that these two genes are associated with the inhibition of the infection and development of nematodes in rice. On the hand, in the case of K + starvation, the susceptibility of the two defect types was still higher than that of wild type (Fig. 6b, c). These results suggest that these two genes also play an independent role in nematode resistance. However, without K 2 SO 4 treatment, the relative susceptibility of OsAKT1 and OsHAK5 de cient lines was weaker than that with the K 2 SO 4 treatment (Fig. 6b, c). These results suggest that although both OsAKT1 and OsHAK5 have independent roles in the process of disease resistance, both were involved in K 2 SO 4induced resistance of rice against M. graminicola.

Discussion
Potassium can enhance plant resistance to biological stresses [19]. The present study showed that K 2 SO 4 could effectively induce the resistance of rice, reduce the numbers of both galls and nematodes and delay the development of M. graminicola. Similar to the effects of silicon and biochar [4,5], K 2 SO 4 treatment had no direct effect on the survival, infection and development of M. graminicola. Similar to the results observed in M. incognita and H. glycines [9,10], application of high concentration of K + was not helpful to the control of M. graminicola. The application of K 2 SO 4 at low concentrations at 24 h before inoculation can effectively reduce the susceptibility of rice to M. graminicola, but has no negative impact on the growth of rice. The reproduction mode of M. graminicola is combining cross-fertilization and facultative meiotic parthenogenesis [20]. Sexual reproduction is conducive to the release of higher genetic diversity [21]. The male/female ratio of nematodes in rice roots treated with K 2 SO 4 was lower than that in the untreated control, indicating that K 2 SO 4 treatment inhibited the sex differentiation of RKNs. The depletion of homoglutathione content increased the proportion of males [22], and the similar result can be obtained without K 2 SO 4 treatment, which is consistent with the view that sex determination of parthenogenetic RKN species is strongly in uenced by environmental factors [23].
Plant parasitic nematodes can be attracted by hosts and promote their infections [13]. Similar to the results observed in β-Aminobutyric acid (BABA) and thiamine treated rice plants, K 2 SO 4 has no signi cant effect on attraction of nematodes [24,25]. However, treatment with metabolites of Aspergillus welwitschiae can signi cantly affect the attractiveness of rice roots to nematodes [26,27]. When the nematode moved and invaded the host plant, the infective J2s migrated to the vascular column cells, and induced the root cells to differentiate into specialized giant cells (GCs) for feeding sites [14]. Previous researches observed that the morphology or size of giant cells may be changed by the treatments of BABA or αβ-Dehydrocurvularin [24,27]. However, K 2 SO 4 treatment had no signi cant effect on the phenotype of giant cells, which is similar to the infection of M. javanica in potassium-treated tomato roots [28].
In most cases, host plant resistance is characterized by local necrosis of plant cells at the site of infection hypersensitive reaction (HR), which is typically characterized by the outbreak of ROS [29]. In the induced resistance to root-knot nematode M. javanica, the production of H 2 O 2 may be helpful for tomato and cucumber plants to defense against RKNs infection [30,31]. Ascorbate oxidation activated systemic defense against M. graminicola, which was also related to the accumulation of H 2 O 2 [32]. Overexpression of respiratory burst oxidase homolog B (RbohB), a gene related to ROS generation, will enhance the resistance of Arabidopsis to nematodes. [33]. On the contrary, the interference or silencing of RBOH1 gene, which encoding NADPH oxidase for the production of apoplastic H 2 O 2 , increased the susceptibility of tomato to M. incognita [34]. In the process of rice resistance to nematodes, K 2 SO 4 treatment signi cantly increased the expression of OsRbohB after inoculation, and stimulated the accumulation of H 2 O 2 within 24 h, indicating that K 2 SO 4 primed the basic defense of rice against M. graminicola by stimulating hydrogen peroxide accumulation.
Callose deposition plays a vital role in the defense of plants against various biological stresses [15]. Ellinger et al. [35] observed that spraying pathogen elicitor g22 can induce callose deposition in plant leaves, and increase the resistance of Arabidopsis to powdery mildew in the early stage. Bian et al. [36] found that validamycin A treatment signi cantly enhanced the deposition of callose in Arabidopsis leaves to defense against different fungi. Similarly, K 2 SO 4 treatment can also enhance the deposition of callose in rice root galls, so it is possible that this way is also used to prevent nematode infection. Validamycin A induced broad-spectrum resistance was involved in SA and JA/ET signaling pathways.
However, further research is needed to elucidate the pathway involved in K 2 SO 4 -induced resistance.
The complex defense signaling pathways involved in plant response to biological stress are regulated by different hormones [37]. The deletion of genes related to defense signaling pathways (e.g., JA, ET and BR) will result in the reduction of plant defense against pathogens (e.g., M. graminicola, Botrytis cinerea and M. incognita) [34,36,37]. Previous studies have shown that after the JA pathway is antagonized by gibberellin and monocrotaline, rice is more susceptible to RKNs [38,39]. The resistance of rice to M. graminicola induced by BABA or ascorbate oxidation depends on JA and ET pathways [24,32]. The enhancement of rice resistance to M. graminicola induced by biochar and silicon is related to the increased transcription levels of ET pathway-related genes [4,5]. Our results showed that the expressions of JA pathway-related genes were up-regulation after the K 2 SO 4 treatment in the early stage; meanwhile the expressions of ET pathway-related genes were very low and were not signi cantly expressed until the JA biosynthesis gene OsAOS2 was highly expressed. This was consistent with the view that the ET pathway requires JA biosynthesis [37]. As an important part of the plant defense system, PR protein can cause HR when it is accumulated in the infected site [40]. OsPR1a and OsPR1b can be induced by exogenous plant hormones JA, SA and ET, respectively, play an important role in rice disease resistance [37,41]. Our research shows the expression of these two genes were signi cantly up-regulated in inoculated or non-inoculated rice after treated with K 2 SO 4 in a short period of time, while H 2 O 2 was also accumulated, both of which were involved in HR [29,40]. Brassinosteroids (BRs) play an active role in plant innate immunity and enhance the level of H 2 O 2 in the apoplast [42,43]. Nahar et al. [44] observed that high concentration of exogenous epibrassinolide signi cantly reduced the susceptibility of rice to M. graminicola. In our study, at 24 hpi when H 2 O 2 accumulated signi cantly, K 2 SO 4 treatment signi cantly increased the expression of these two BR pathway-related genes in rice. These results were consistent with the view that BRs enhance H 2 O 2 accumulation [43]. K 2 SO 4 positively regulates the expression of resistance-related genes during rice defense against M. graminicola infection, and its process may be related to the accumulation of H 2 O 2 and HR. However, further research is needed to elucidate the role of different pathways in K 2 SO 4 -induced hormone metabolisms.
Potassium plays key role in maintaining the activity of cytoplasmic enzymes [45]. If potassium de ciency occurs, the normal physiological and biochemical reactions of plants will be correspondingly inhibited [46]. However, most K + in soil is dehydrated and cannot be used by plants [47]. Venkatesan et al. [48] observed that the content of K + in rice was decreased after M. graminicolais infection. In order to improve the utilization e ciency of K + , plants ensure the absorption and transport of K + in vitro and in vivo through K + transporters and channels [49]. Shi et al. [50] observed that the effector protein AvrPiz-t from Magnaporthe oryzae, a pathogen of rice blast, targets the K + channel OsAKT1 to destroy plant immunity, and the deletion of OsAKT1 resulted in the decrease of K + content and the resistance to Magnaporthe oryzae. The high a nity K + transporter HAK5 could interact with Raf-like Kinase ILK1 which is helpful to plant defense against bacterial pathogens, and is necessary for innate immunity [51]. Our data showed that, compared with the wild-type plants, the susceptibility of the OsAKT1 and OsHAK5 defective plants was signi cantly increased without K 2 SO 4 treatment, indicating that these two genes play an independent and positive role in the resistance of rice to M. graminicola. However, the control effects of K 2 SO 4 treatment on the two defective lines were lower than that on the wild-type, indicating that the resistance induced by K 2 SO 4 requires the participation of these two genes. The decrease of resistance to nematodes led by the deletion of OsAKT1 may be similar to how M. oryzae destroys rice systemic immunity [50,52]. The positive response to K 2 SO 4 treatment and nematode infection also showed that OsAKT1 had great initiative in regulating the loss of K + and maintaining the dynamic balance of K + , so as to reduce the internal competition of pathogens for nutritional resources, protect the activity of plant enzymes, and improve the synthesis of defense compounds [19,53]. OsHAK5 is active in mobilization of potassium in rice, not only mediates the absorption of K + in the external environment, but also mediates the transport of K + from root to shoot or from source to sink in plants, and also plays an important role in the control of potassium/sodium concentration ratio [54], but its expression is inhibited when the external K + is su cient [55]. Therefore, in our experiment, the expression level of OsHAK5 in K 2 SO 4 alone treatment was slightly lower than other inoculated treatments within 24 h. These results indicate that the contribution of OsAKT1 and OsHAK5 in the host's resistance to RKNs infection is positive and K + plays a major role in this process.

Conclusion
For the outstanding performance of potassium in plant resistance to biological stress, we further expanded the data of potassium in helping understand rice defense against RKNs in this study. Treatment of K 2 SO 4 at low concentrations could effectively improve the defense of rice against M.
graminicola. K 2 SO 4 -induced resistance was achieved by priming the basic defense of rice, inducing the expression of resistance-related genes against plant diseases, with the involvement of K + channel and transporter. These results laid a foundation for further study on the mechanism of rice to defense against root-knot nematodes. However, the mechanisms of resistance-related genes, the role of K + channel and transporter will be further explored to better understand IR in rice. Thereafter, K 2 SO 4 -induced resistance will provide a theoretical guidance for the integrative control of nematodes in rice.

Plant materials and K 2 SO 4
Wild-type (WT) Rice seeds (Oryza sativa cv. Nipponbare) were friendly offered by the United States Department of Agriculture (GSOR-100) and cultivated at Hunan Agricultural University in Changsha, Hunan Province, China. Transgenic OsAKT1-RNAi lines de cient in K + channel OsAKT1 were collected from the State Key Laboratory for Biology of Plant Diseases and Insect Pests as gifts by Wang et al. [56] and identi ed by qRT-PCR (Supplementary Fig. 1). Transgenic OsHAK5-Cr lines de cient in K + transporter OsHAK5 were generated from NPB by the CRISPR/Cas9 [57] and identi ed by PCR ( Supplementary  Fig. 2). In the experiment, rice seeds were soaked in 4-5% sodium hypochlorite (NaOCl) for 5-8 minutes to disinfect, then washed with distilled water for three times, and then germinated at 28 ℃ in the incubator. Budding seeds are seeded in each polyvinyl chloride (PVC) tube containing Super Absorbent Polymer (SAP) substrates (1:400 (w: v) mixture of sand and SAP) [58]. Rice seedlings were maintained in a greenhouse at 28°C with 75% relative humidity and 16 h / 8 h light / dark photoperiod, and irrigated with 10 ml of K 2 SO 4 -free nutrient solution (NH 4 NO 3 914 g, NaH 2 PO 4 ·2H 2 O 403 g, CaCl 2 886 g, Collection, culture and hatching of M. graminicola RKNs collected from Pingjiang County, Yueyang City, Hunan Province, were identi ed with morphological and molecular methods as Golden and Birch eld [60] and Htay et al. [61], respectively, and subcultured on Nipponbare in a glass greenhouse at about 28 ℃. After the nematode was fully developed and matured, the eggs were isolated from the root galls and incubated at 28 ℃ for 72 h. The hatched J2 was ltered with a 25-µm sieve and diluted with distilled water into a suspension of about 100 nematodes per ml for inoculation [62].

Direct effects of K 2 SO 4 on M. graminicola
Place J2 s (about 100 pieces) in a 6-well culture plate (φ 3.5 cm) containing 1 ml of diluted K 2 SO 4 solution or distilled water. After culturing for 72 h, a few drops of 1 M NaOH were added in the solution. Nematodes that change their shape are considered to be alive, while stiff nematodes that do not respond are classi ed as dead [63]. The experiment was carried out three times with four replicates each time.
In order to determine the direct effect of K 2 SO 4 on the infection and development of M. graminicola, recently hatched J2s were soaked in K 2 SO 4 solution or distilled water for 48 h before inoculation. Then each two-weeks-old rice plant planted in PVC tubes containing SAP substrates was inoculated with approximately 100 J2s and maintained in a greenhouse at 28°C with 75% relative humidity. At 14 dpi, the root samples were carefully cleaned, wrapped with microcloth, bleached with 5% NaOCl for 5-10 minutes, then rinsed with distilled water at least three times to remove residual NaOCl, and then dyed in acid fuchsin staining [37]. After the dye on the root surface was washed with distilled water, the roots samples were placed in glycerin to observe the number of nematodes in the root at each development stage under the stereomicroscope. Microsoft Excel 6.0 (Redmond, Washington, USA) was used to calculate the proportion of nematodes of different ages (adult females or J3 / J4). The experiment was carried out twice with 6 replicates each time.
Evaluation of rice resistance against M. graminicola induced by K 2 SO 4 In order to test whether K 2 SO 4 treatment induced rice resistance to M. graminicola, two-week-old rice roots were irrigated with 10 ml K 2 SO 4 dilution one day before inoculation (approximately 100 J2s per plant). After measuring the height and fresh weight of each plant at 14 dpi, the roots were dyed according to the method described previously, and then the number of nematodes per developmental stage and sex and the galls number were counted. Galls of each treatment were randomly selected to take photos with Dual-Sensor Monochrome and Color Camera (Olympus DP80) under the microscope BX53 research microscope (Olympus Optical Company, Tokyo, Japan) at 4 magni cations. The test was carried out three times with six replicates each time.
Attraction of rice roots to M. graminicola and giant cells phenotype after K 2 SO 4 treatment According to the method described by Wang et al. [64], the trapping experiment of nematode was carried out. First, 11.5 g of Pluronic F-127 powder (sigma Aldrich, Brussels, Belgium) was added into 50 ml sterile water, stirred on ice and placed in a refrigerator at 4 ℃ to dissolve it completely. The roots of 2-week-old rice were soaked in K 2 SO 4 dilution or water. After 24 h, the root tip of about 1 cm was cut and placed in the center of the culture hole (φ 3.5 cm) containing 1 ml Pluronic gel and approximately 100 J2s. After 10 hours at room temperature, under microscope BX53 research microscope, the nematodes near the root were photographed with Dual-Sensor Monochrome and Color Camera, and then the numbers of nematodes within 5 mm from the root tip were counted. The experiment was carried out three times with six replicates each time.
Giant cell phenotypes were observed according to the method described by Ji et al. [65]. The galls of rice at 7 dpi was xed in 1× PIPES buffer overnight, then dehydrated with ethanol diluents step by step, and then embedded according to the instructions of Technovit 7100 kit. CryoStar NX50 Cryostat (Thermo Fisher Scienti c, MA, USA) was used to cut gall tissue into 10 µm slices and placed on the slides. After stained with 0.5% toluidine blue for 5 min, the tissue was observed by microscope BX53 research microscope. The experiment was repeated twice with 10 galls per replicate.
Microscopic observation of callose deposition and quanti cation of H 2 O 2 Callose deposition was detected according to Millet et al. [66]. Rice was inoculated with approximately 100 J2s and maintained in a greenhouse at 28°C. At 7 dpi, 20 root galls were randomly taken from each treatment and xed overnight in ethanol acetic acid solution, and then diluted with ethanol for dehydration. The galls were stained with 1% aniline blue. The deposition of callose was detected by Olympus IX83 inverted microscope (Olympus Optical Company, Tokyo, Japan) under ultraviolet light.
Image J software was used to quantify. The experiment was repeated three times.
Four treatments were used for 2-week-old rice plants, including K 2 SO 4 treatment alone, water treatment or K 2 SO 4 treatment before inoculation, and water treatment alone as control. Root samples were collected for quantitative analysis of H 2 O 2 at 8, 24 and 72 hpi. Fresh root sample (100 mg) collected from each group of six rices was treated according to the method described by Ji et al. [24], and then the H 2 O 2 accumulation was detected by the trichloroacetic acid (TCA) method described by Velikova et al. [67]. The experiment was repeated three times.

Relative quantitative analysis of expression levels of disease resistance-related genes
In order to analyze the effect of K 2 SO 4 treatment on the expression level of defense related genes in rice, four treatment groups as described in the section of quanti cation of H 2 O 2 were set up. Total RNA was extracted from root samples of 6 plants using RNeasy plant Mini Kit, and cDNA was synthesized using PrimeScript™ RT reagent Kit with gDNA Eraser (Perfect Real Time) (TAKARA, Japan). The primer pairs for qRT PCR analysis of resistance-related genes and internal reference genes are listed in Table 1. All qRT-PCR analyses were performed using a 7500-fast real-time PCR system (Thermo Fisher technologies, Beijing, China) with two independent biological repeats and three technical replicates. The relative expression level of resistance-related gene was characterized by the fold change compared with the control. The relative expression of genes was calculated by 2 −∆∆ Ct method [68]. To determine whether the K + channel OsAKT1 and transporter OsHAK5 respond to K 2 SO 4 treatment or nematode infection, the expression levels of OsAKT1 and OsHAK5 were determined by the same method in the previous section. The primer sequences for qRT-PCR analysis of OsAKT1, OsHAK5 are listed in Table 1. K 2 SO 4 solutions at a concentration of 0.5 mM were treated and approximately 100 J2s were inoculated on rice roots.
In order to determine whether OsAKT1 and OsHAK5 are involved in the resistance of rice to RKNs induced by K 2 SO 4 , K 2 SO 4 solutions were irrigated on 2-week-old OsAKT1-RNAi lines and OsHAK5-Cr transgenic lines, and approximately 100 J2s were inoculated on plant roots. The wild-type Nipponbare was used as control. At 14 dpi, the numbers of galls and nematodes were counted using the method described previously. The whole experiment was carried out three times with 6 plants in each repeat.
Statistical analysis of data SPSS version 25 (IBM, Armonk, NY, USA) was used to calculate the mean and standard errors of the data.
Statistical signi cance (P ≤ 0.05) was obtained by Duncan´s multiple range tests.

Declarations
Ethics approval and consent to participate The collection of wild-type rice seeds (Oryza sativa cv. Nipponbare) was permitted by the United States Department of Agriculture and complied with relevant institutional, national, and international guidelines