NB-LRRs Not Responding Consecutively to Fusarium oxysporum Proliferation Caused Replant Disease Formation of Rehmannia glutinosa

Consecutive monoculture practice facilitates enrichment of rhizosphere pathogenic microorganisms and eventually leads to the emergence of replant disease. However, little is known about the interaction relationship among pathogens enriched in rhizosphere soils, Nucleotide binding-leucine-rich repeats (NB-LRR) receptors that specifically recognize pathogens in effector-triggered immunity (ETI) and physiological indicators under replant disease stress in Rehmannia glutinosa. In this study, a controlled experiment was performed using different kinds of soils from sites never planted R. glutinosa (NP), replanted R. glutinosa (TP) and mixed by different ration of TP soils (1/3TP and 2/3TP), respectively. As a result, different levels of TP significantly promoted the proliferation of Fusarium oxysporum f.sp. R. glutinosa (FO). Simultaneously, a comparison between FO numbers and NB-LRR expressions indicated that NB-LRRs were not consecutively responsive to the FO proliferation at transcriptional levels. Further analysis found that NB-LRRs responded to FO invasion with a typical phenomenon of “promotion in low concentration and suppression in high concentration”, and 6 NB-LRRs were identified as candidates for responding R. glutinosa replant disease. Furthermore, four critical hormones of salicylic acid (SA), jasmonic acid (JA), ethylene (ET) and abscisic acid (ABA) had higher levels in 1/3TP, 2/3TP and TP than those in NP. Additionally, increasing extents of SA contents have significantly negative trends with FO changes, which implied that SA might be inhibited by FO in replanted R. glutinosa. Concomitantly, the physiological indexes reacted alters of cellular process regulated by NB-LRR were affected by complex replant disease stresses and exhibited strong fluctuations, leading to the death of R. glutinosa. These findings provide important insights and clues into further revealing the mechanism of R. glutinosa replant disease.


Introduction
Replant diseases, also known as consecutive monoculture problems or sick soil syndrome, are widespread in the production of different crops, especially in medicinal crops, fruit tree and vegetables, such as Rehmannia glutinosa, Panax notoginseng, apples, peach, strawberry, soybean etc. [1][2][3][4][5][6], which usually lead to disease aggravation, poor growth status, yield reduction and quality deterioration [7]. Rehmannia glutinosa Libosch, a perennial herbaceous plant, is one of the 50 traditional Chinese medicines with high value [8]. However, consecutive monoculture of R. glutinosa results in its abnormal growth and a significant decline in the yield and quality of tuberous roots. Moreover, it cannot be replanted on replant disease and are closely related to immune resistance [48]. However, a comprehensive survey on the relationship between plant hormones and replant disease in R. glutinosa is still unknown.
Given the interferences from complex field environment and vegetative propagation, such as various pathogens, variable light and temperature conditions, maternal resistance, size of vegetative mass, germination rate and site of shoot, pot experiment in controlled conditions with regenerated plantlets is thus an ideal research method. Here, the soils from the sites in which R. glutinosa had never been planted and had been consecutively planted R. glutinosa were proportionally mixed to control the stress levels of replant disease. The regenerated plantlets of R. glutinosa with feeble immunity were acclimatized and transplanted in a phytotron. The dynamic changes of pathogenic microbes in the rhizosphere soils were investigated. At same time, NB-LRRs expression, plant hormone contents and physiological index levels were examined in the roots of R. glutinosa. These research results will provide important insights for further revealing the mechanism of formation of replant disease.

Changes in the Numbers of Pseudomonas spp. and Fusarium oxysporum in Rhizosphere Soils of Replant Disease R. glutinosa
To understand the changes in the beneficial PS and pathogenic FO in rhizospheres of replanted R. glutinosa, the numbers of them were showed in Figure 1. The numbers of PS were 3.87 × 10 8~5 .99 × 10 8 (cell·g −1 soil), the numbers of FO were 0.71 × 10 8~2 .68 × 10 8 (cell·g −1 soil), and the ratio of FO to PS were 0.16~0.55. The maximum difference ranged between 0 DAP and 9 DAP, where PS was in TP soil and FO was in NP soil ( Figure 1A,B). Moreover, only the numbers of FO decreased significantly over time in control NP ( Figure 1E). The inhibiting effect of NP soil on FO proliferation was thus stronger than the promoting effect of replanted soil on FO proliferation. During 0~9 DAP, the FO numbers and ratio of FO to PS in TP soil were significantly lower than that in NP soil at 0 DAP, and the reversing results with significant difference were at 9 DAP ( Figure 1B,C). The PS numbers were significantly lower only in NP soil than that in TP soil at 0 DAP, and the reversed trends with no significant difference were at 9 DAP ( Figure 1A). Furthermore, there were the significant positive correlations between the addition level of replant soils and the changes of the FO numbers and the FO/PS values, and between the FO numbers and the FO/PS values (Tables 1 and 2). The results evidenced that R. glutinosa replant disease (1/3TP, 2/3TP and TP) could significantly promote the FO proliferation in rhizospheres soils. Interestingly, at 0 DAP, the PS numbers in 1/3TP, 2/3TP and TP soil were 1.04, 1.11 and 1.21 times than that in NP soil respectively, and the FO numbers in NP soil were 1.39, 1.44 and 1.83 times than that in 1/3TP, 2/3TP and TP soil respectively. The changes were obviously different from that at 9 DAP. The results suggested that the soil had a strong ability that could restore balances among microbial communities, and this means that the appropriate ration between FO and PS in rhizosphere soil presented in R. glutinosa grows healthly, during fallow stage. NP: Soil that was never planted with R. glutinosa for at least 10 years. TP: Soil that was consecutively planted with R. glutinosa in the same soils for three years.

NB-LRR Lists Response to Replant Disease Stresses in R. glutinosa
To reveal the responding mode of R. glutinosa NB-LRRs during replant disease formation, the expression levels of NB-LRRs were determined by qRT-PCR. Of 35 expressed NB-LRRs in acclimatization stage, seven were upregulated and 11 were downregulated ( Figure 2). In contrast to NP, 22 NB-LRRs were upregulated and six NB-LRRs were downregulated in TP at 3 DAP, which accounted for 80% of the 35 NB-LRRs. The stage from 0 to 3 DAP was thus the key stage to NB-LRRs Figure 1. The influences of different replant disease stress levels on the numbers and ratio of FO to PS in R. glutinosa rhizosphere soils over time. NP, 1/3TP, 2/3TP and TP are the gradient treatments of replant disease stresses. Four samples in each group are compared with NP (A-C) and 0 DAP (D-F) respectively, and different lower-case letters indicate significant differences (p < 0.05; LSD). Data represented as the mean ± SD (n = 3). DAP: Days after planting. PS: Pseudomonas spp. FO: Fusarium oxysporum. NP: Soil that was never planted with R. glutinosa for at least 10 years. TP: Soil that was consecutively planted with R. glutinosa in the same soils for three years.

NB-LRR Lists Response to Replant Disease Stresses in R. glutinosa
To reveal the responding mode of R. glutinosa NB-LRRs during replant disease formation, the expression levels of NB-LRRs were determined by qRT-PCR. Of 35 expressed NB-LRRs in acclimatization stage, seven were upregulated and 11 were downregulated ( Figure 2). In contrast to NP, 22 NB-LRRs were upregulated and six NB-LRRs were downregulated in TP at 3 DAP, which accounted for 80% of the 35 NB-LRRs. The stage from 0 to 3 DAP was thus the key stage to NB-LRRs that responded to replant disease stress. In addition, only 1 NB-LRR was upregulated and 29 NB-LRRs were downregulated at 6 DAP, indicating that replant disease interference to the expression of NB-LRRs. In 3 DAP, the Pearson correlation analyses displayed that 35 NB-LRRs had no significant correlation with the PS numbers. The significant positive correlation was only between 12 NB-LRRs and the FO numbers in NP, 1/3TP, 2/3TP (Table 3). However, these NB-LRRs that have significantly positive correlation with the FO numbers, have ineffectively prevented the death of R. glutinosa plants at 6 DAP. The results displayed that the 12 upregulated NB-LRRs might not respond to the pathogenic FO in replanted R. glutinosa. Therefore, the six downregulated NB-LRRs (RgNB5, RgNB14, RgNB26, RgNB29, RgNB34 and RgNB35) in TP at 3 DAP, were screened as candidates for responding to R. glutinosa replant disease. In addition, RgNB14 and RgNB26 of the 6 NB-LRRs were upregulated expression, and the remaining 4 NB-LRRs were downregulated or presented a decreasing trend at expression level in acclimatization stage. The two groups of NB-LRRs represented whether they responded to abiotic stress or not. Noticeably, the expression levels of RgNB5 and RgNB29 were continuously downregulated with addition levels of replant soils at 3 DAP. as candidates for responding to R. glutinosa replant disease. In addition, RgNB14 and RgNB26 of the 6 NB-LRRs were upregulated expression, and the remaining 4 NB-LRRs were downregulated or presented a decreasing trend at expression level in acclimatization stage. The two groups of NB-LRRs represented whether they responded to abiotic stress or not. Noticeably, the expression levels of RgNB5 and RgNB29 were continuously downregulated with addition levels of replant soils at 3 DAP. shown. Two or four samples in each group are compared with 3 DAA and NP respectively, and different lower-case letters indicate significant differences (p < 0.05; LSD). Data represented as the mean ± SD (n = 3). DAA: Days after acclimatization. DAP: Days after planting. NP: Soil that was never planted with R. glutinosa for at least 10 years. TP: Soil that was consecutively planted with R. glutinosa in the same soils for three years.

Plant Hormones Response to Replant Disease Stresses in R. glutinosa
Four plant hormones including ABA, ET, JA and SA were measured, and their contents in 1/3 TP, 2/3TP and TP treatments fluctuated obviously from 0 DAA to 9 DAP (Figure 3). In the acclimatization stage (from 0 DAA to 3 DAA), the contents of plant hormones decreased significantly except for SA. The sensitive stages for rapid alteration of hormones was from 0 to 3 DAP based on the comparison between 3 DAP and 3 DAA (0 DAP). At 3 DAP, the four hormones were significantly activated under replant disease stresses compared with the control NP by 0.56 to 1.53 times for ABA, 0.59 to 2.02 times for JA, 0.50 to 1.50 times for ET and 0.37 to 1.19 times for SA (p < 0.05), which showed that the four plant hormones responded strongly to replant disease stress levels. From 0 to 3 DAP, the four hormone contents in 1/3TP, 2/3TP and TP have significantly increased at 3 DAP compared with that at 0 DAP, excepting SA treated with TP soils. Of which, ABA contents increased by 65.26% to 173.64%, JA by 60.19% to 203.50%, ET by 65.26% to 173.84% and SA by 3.91% to 66.39% under replant disease stresses, but only increased by 8.50%, 0.33%, 9.73% and −24.72% in NP, respectively (p < 0.05). Notably, there were significant negative correlations between FO and SA (p < 0.05) and significant positive correlations among the four hormones (p < 0.01) ( Table 4). The results showed that the FO in replant disease rhizosphere soil inhibited the SA biosynthesis in root. Taking together these changes showed that the complex replant disease stress, including biotic and abiotic stress, promoted the four hormones release, but FO inhibited SA biosynthesis, resulting in R. glutinosa death at 6 DAP.  and TP are gradient treatments. Two or four samples in each group are compared with 3 DAA (0 DAP) and NP respectively, and different lower-case letters indicate significant differences (p < 0.05; LSD). 3 DAP with NP, 1/3TP, 2/3TP and TP gradient treatments is compared with 3 DAA (0 DAP), and the asterisk indicate significant differences (p < 0.05; LSD). Data represented as the mean ± SD (n = 3). DAA: Days after acclimatization. DAP: Days after planting. NP: Soil that was never planted with R. glutinosa for at least 10 years. TP: Soil that was consecutively planted with R. glutinosa in the same soils for three years.

Physiological Response to Replant Disease Stresses in R. glutinosa
The physiological indexes including root activity; superoxide dismutase (SOD), peroxidase (POD) and catalase (CAT) activities; and malondialdehyde (MDA) and hydrogen peroxide (H2O2) Figure 3. The four plant hormone contents in R. glutinosa roots during acclimatization stage and planting stage with different replant disease stress levels. 0 DAA, 3 DAA (0 DAP), NP, 1/3TP, 2/3TP and TP are gradient treatments. Two or four samples in each group are compared with 3 DAA (0 DAP) and NP respectively, and different lower-case letters indicate significant differences (p < 0.05; LSD). 3 DAP with NP, 1/3TP, 2/3TP and TP gradient treatments is compared with 3 DAA (0 DAP), and the asterisk indicate significant differences (p < 0.05; LSD). Data represented as the mean ± SD (n = 3). DAA: Days after acclimatization. DAP: Days after planting. NP: Soil that was never planted with R. glutinosa for at least 10 years. TP: Soil that was consecutively planted with R. glutinosa in the same soils for three years.

Physiological Response to Replant Disease Stresses in R. glutinosa
The physiological indexes including root activity; superoxide dismutase (SOD), peroxidase (POD) and catalase (CAT) activities; and malondialdehyde (MDA) and hydrogen peroxide (H 2 O 2 ) contents were investigated in R. glutinosa with different replant disease stress levels ( Figure 4). In the acclimatization stage, there were no significant changes in the POD activities. The H 2 O 2 contents were significantly increased, and the others that used to eliminate toxicity of H 2 O 2 were decreased significantly. In contrast to 3 DAA (0 DAP), the root activities, SOD activities, POD activities and MDA contents in NP tended to increase consecutively from 3 DAP to 9 DAP, which reported the normal response of plantlets planted in healthy soil. In addition, root activity, SOD activity, CAT activity and H 2 O 2 contents in NP, 1/3TP, 2/3TP and TP at 3 DAP were significantly different from those at 3 DAA (0 DAP). Of which H 2 O 2 contents significantly decreased by 0.45 to 0.75 times, and root activity, SOD activity and CAT activity significantly increased by 0.36 to 2.75 times, 3.12 to 7.33 times, 2.76 to 4.26 times at 3 DAP, respectively. However, there were no gradient changes for the four physiological indexes at 3 DAP. Moreover, only the H 2 O 2 contents showed a gradient change at 6 DAP. Furthermore, there were significant positive correlations between the FO numbers and the CAT activity and the MDA content only at 6 DAP (Table 5). Overall, these results displayed that these physiological indexes acting downstream of life activities were affected by complex replant disease stresses and did not show significant correlation with the FO numbers until 6 DAP. four physiological indexes at 3 DAP. Moreover, only the H2O2 contents showed a gradient change at 6 DAP. Furthermore, there were significant positive correlations between the FO numbers and the CAT activity and the MDA content only at 6 DAP (Table 5). Overall, these results displayed that these physiological indexes acting downstream of life activities were affected by complex replant disease stresses and did not show significant correlation with the FO numbers until 6 DAP. and TP are gradient treatments. Two or four samples in each group are compared with 3 DAA (0 DAP) and NP respectively, and different lower-case letters indicate significant differences (p < 0.05; LSD). 3 DAP with NP, 1/3TP, 2/3TP and TP gradient treatments is compared with 3 DAA (0 DAP), and the asterisk indicate significant differences (p < 0.05; LSD). Data represented as the mean ± SD (n = 3). DAA: Days after acclimatization. DAP: Days after planting. NP: Soil that was never planted with R. glutinosa for at least 10 years. TP: Soil that was consecutively planted with R. glutinosa in the same soils for three years.  NP respectively, and different lower-case letters indicate significant differences (p < 0.05; LSD). 3 DAP with NP, 1/3TP, 2/3TP and TP gradient treatments is compared with 3 DAA (0 DAP), and the asterisk indicate significant differences (p < 0.05; LSD). Data represented as the mean ± SD (n = 3). DAA: Days after acclimatization. DAP: Days after planting. NP: Soil that was never planted with R. glutinosa for at least 10 years. TP: Soil that was consecutively planted with R. glutinosa in the same soils for three years.

Replant Disease Promotes Proliferation of Fusarium oxysporum in R. glutinosa Rhizospheres Soil
Previous studies have revealed that R. glutinosa replant disease induced rhizosphere microbes' adverse chemotaxis [49,50], resulting in the increase of pathogenic FO abundance and the decrease of beneficial PS abundance, thus the identified PS and FO were usually used as a characteristic mark in the R. glutinosa replant disease study [10,28]. In this study, the PS numbers decreased and the FO numbers increased in 1/3TP, 2/3TP and TP soils from 0 to 9 DAP. Moreover, the significant positive correlations were presented between the addition levels of replant soils and the changes of the FO numbers and the FO/PS values. These results revealed that replant disease promoted FO proliferation in rhizospheres soils. Interestingly, the focus on maximum difference between the PS and FO numbers displayed that the inhibiting effect of NP soil on FO proliferation was stronger than the promoting effect of replanted soil on FO proliferation. Furthermore, the number of PS in NP soil was significantly lower than that in TP soil at 0 DAP, while the number of FO in NP soil was significantly greater than that in 1/3TP, 2/3TP and TP soils at 0 DAP. The changes were obviously different from that at 9 DAP.
These results are different from those in the field [10], which might be related to the difference between the air-dried soil used in this experiment and the field soil after fallow cultivation in the literature [51]. A valuable clue was thus presented that soil may have a strong ability restored microbe balance in rhizosphere soil during the fallow stage.

NB-LRRs Failed to Respond Timely and Effectively to Pathogenic Fusarium oxysporum in Replanted R. glutinosa
Pathogens almost always occupy extracellular niches [37]. Early studies had unraveled that Fusarium oxysporum could inhibit the release of ATP after invading the host plant [52], which is required for NB-LRR activation [37], but low doses toxins secreted by FO could induce the biosynthesis of phytoalexins in the host plants [53]. In this study, the upregulated expressions of NB-LRRs were mainly appeared at 3 DAP in TP and at 6 DAP in NP, which were consistent with the changes of the FO relative values ( Figure 5). According to the original data associated with the relative values, we found that the upregulated expression of NB-LRRs was the strongest when the number of FO was at the lowest level (about 0.95 × 10 8 cell·g −1 soil), and gradually decreased with the increasing of FO number in six days after transplanting. These results exhibited that NB-LRR expressions have a typical regulars with "promotion in low concentration and suppression in high concentration" when countered FO invasion, which was consistent with Scott et al. (2019) [54].
numbers and the FO/PS values. These results revealed that replant disease promoted FO proliferation in rhizospheres soils. Interestingly, the focus on maximum difference between the PS and FO numbers displayed that the inhibiting effect of NP soil on FO proliferation was stronger than the promoting effect of replanted soil on FO proliferation. Furthermore, the number of PS in NP soil was significantly lower than that in TP soil at 0 DAP, while the number of FO in NP soil was significantly greater than that in 1/3TP, 2/3TP and TP soils at 0 DAP. The changes were obviously different from that at 9 DAP. These results are different from those in the field [10], which might be related to the difference between the air-dried soil used in this experiment and the field soil after fallow cultivation in the literature [51]. A valuable clue was thus presented that soil may have a strong ability restored microbe balance in rhizosphere soil during the fallow stage.

NB-LRRs Failed to Respond Timely and Effectively to Pathogenic Fusarium oxysporum in Replanted R. glutinosa
Pathogens almost always occupy extracellular niches [37]. Early studies had unraveled that Fusarium oxysporum could inhibit the release of ATP after invading the host plant [52], which is required for NB-LRR activation [37], but low doses toxins secreted by FO could induce the biosynthesis of phytoalexins in the host plants [53]. In this study, the upregulated expressions of NB-LRRs were mainly appeared at 3 DAP in TP and at 6 DAP in NP, which were consistent with the changes of the FO relative values ( Figure 5). According to the original data associated with the relative values, we found that the upregulated expression of NB-LRRs was the strongest when the number of FO was at the lowest level (about 0.95 × 10 8 cell·g −1 soil), and gradually decreased with the increasing of FO number in six days after transplanting. These results exhibited that NB-LRR expressions have a typical regulars with "promotion in low concentration and suppression in high concentration" when countered FO invasion, which was consistent with Scott et al. (2019) [54].  In antagonistic associations with microbes, plants have evolved to form two strategies of PTI and ETI for fighting microbial pathogens [37]. A well-known view is that immune receptor play a vital role in the recognition to pathogens [55,56], and constitutive downstream proteins are tightly controlled by both positive and negative regulators [39,57]. An integrated understanding for plant immune response is thereby in both the immune receptor and downstream signal transduction. Based on studies till date, there is mounting evidence that two possible interpretations were supported for the relationships of immune response and plant death. (i) Plant immunity has not been triggered [37,38]. (ii) Excessive immunity response often leads to inhibition of normal plant growth and even death [57][58][59]. In this study, the majority of NB-LRRs identified in replanted R. glutinosa roots were upregulated or downregulated in TP at 3 DAP in comparison to 0 DAP. At same time, there were significantly positive correlations found between 12 NB-LRRs and the FO numbers in NP, 1/3TP, 2/3TP and TP during 0~3 DAP, but the FO numbers in 1/3TP, 2/3TP and TP were no different with those in NP at 3 DAP even significantly lower at 0 DAP. In addition, accompanied by the death of replanted R. glutinosa at 6 DAP, only 1 NB-LRR was upregulated (downregulated at 3 DAP) and 29 NB-LRRs were downregulated. Taking together these data showed that the stage of 0~3 DAP was the key stage for NB-LRRs to respond to replant disease stress, while the immune response was obviously inactivated to FO rather than excessive responses of NB-LRRs. Therefore, it was one of important reasons that NB-LRRs were not consecutively responsive to the FO proliferation at the transcriptional level in replanted R. glutinosa roots. These new findings provide insights into the response mechanism of R. glutinosa to replant disease.
One of the big gaps in our understanding of plant immunity is in the downstream signaling pathways after receptor protein activation [37]. Only two identified downstream signaling proteins of EDS1 and non-race-specific disease resistance 1 (NDR1) are required for signaling of all TIR-NB-LRRs and some CC-NB-LRRs, respectively [37]. For NB-LRR, the protein structure consists of a carboxy-terminal LRR domain for effector recognition, NB-ARC (nucleotide-binding adaptor shared by APAF-1, R proteins, and CED-4) and amino-terminal Toll/interleukin-1 receptor (TIR) or Coiled-coil (CC) domains for signal transduction to downstream proteins [41]. In this study, the six downregulated NB-LRRs (RgNB5, RgNB14, RgNB26, RgNB29, RgNB34 and RgNB35) in TP at 3 DAP, were screened as candidates for responding to R. glutinosa replant disease. According to their functional conservation in these NB-LRRs, there were seven resistance in linkage group 1A (R1A), 2 resistance in linkage group 1B (R1B), 1 resistance to Pseudomonas syringae (RPS2), 1 target of AvrB operation 1 (TAO1) and 2 resistance to powdery mildew 8 (RPW8). Based on studies till date, EDS1 were required for downstream signaling of these identified NB-LRRs except for RPS2 (RgNB32) [60][61][62][63]. The results provide valuable clues for studying the signaling pathways that operate downstream of NB-LRR protein activation in replanted R. glutinosa.

Lower Level of SA Biosynthesis Stimulated by Fusarium oxysporum Might Be Cloesly Related to the Formation of R. glutinosa Replant Disease
Generally, the stress factors associated with replant disease were complex and multiple, including pathogens, nematode and abiotic stress [7]. To resist these stress factors, plant hormones are widely involved in resistance levels as important signaling molecules [64]. ABA is mainly associated with abiotic environmental stresses [44]. SA is typically involved in the defense against biotrophs [65]. JA and ET are generally thought to act together, and to play core roles in the defense against necrotrophs [66]. The relationship between the SA and JA-ET pathway is more antagonistic other than cooperative [44]. In this study, the four hormones were significantly activated under replant disease stresses compared with the control NP at 3 DAP, and presented each other significant positive correlation (p < 0.001). The cooperation for SA and JA-ET was consistent with some literatures, such as   [67] and Adie et al. (2007) [68], which was different from the antagonistic relationship between SA and JA-ET. A reasonable explanation for the cooperation of ABA with SA was that ABA increased susceptibility to pathogens in some plant-pathogen interactions [45]. However, these inferences need to be further verified. Previous studies displayed that some plant hormones including ET and ABA involved in the formation mechanism of R. glutinosa replant disease and closely related to immune resistance [48,69]. Some studies have demonstrated that SA was involved in the defense against Fusarium oxysporum and associated with acteoside accumulation, one of the important pharmacodynamic component [46,47]. More importantly, there were significant negative correlations only between ∆FO and ∆SA with the gradient changes in the stress level of replant disease ( Table 4). The results revealed that the SA synthesis, which may be inhibited by Fusarium oxysporum, was involved in the formation of replant disease in R. glutinosa.
Based on the above results, a possible depiction of the immune response and its potential crosstalk with microbes and plant hormones in replanted R. glutinosa was drawn, and it is shown in Figure 6. These findings provide insights into the formation of replant disease. gradient changes in the stress level of replant disease ( Table 4). The results revealed that the SA synthesis, which may be inhibited by Fusarium oxysporum, was involved in the formation of replant disease in R. glutinosa.
Based on the above results, a possible depiction of the immune response and its potential crosstalk with microbes and plant hormones in replanted R. glutinosa was drawn, and it is shown in Figure 6. These findings provide insights into the formation of replant disease. Figure 6. Schematic of effector-triggered immunity response mediated by consecutive monoculture stress in R. glutinosa.

Plant Growth and Treatments
To obtain aseptic plantlets, tuberous roots of R. glutinosa "Wen 85-5" were surface sterilized with 0.1% mercuric chloride solution for 17-20 min, washed five times with the sterile water, and then cultured in sterile bottles with two layers of damp gauze at the bottom. The shoots, approximately 1 cm long, were cut and cultured on hormone-free MS agar medium containing 30 g·L −1 sucrose and 10 g·L −1 agar [47]. The explants were cultured under controlled conditions (25 °C, 4000 lux, 14 h light/10 h dark photoperiod) in a growth chamber for 30 days.

Plant Growth and Treatments
To obtain aseptic plantlets, tuberous roots of R. glutinosa "Wen 85-5" were surface sterilized with 0.1% mercuric chloride solution for 17-20 min, washed five times with the sterile water, and then cultured in sterile bottles with two layers of damp gauze at the bottom. The shoots, approximately 1 cm long, were cut and cultured on hormone-free MS agar medium containing 30 g·L −1 sucrose and 10 g·L −1 agar [47]. The explants were cultured under controlled conditions (25 • C, 4000 lux, 14 h light/10 h dark photoperiod) in a growth chamber for 30 days.
To enhance the quality of plantlets for transplantation, the aseptic plantlets of R. glutinosa with seven to eight leaves were adapted in a phytotron (28 • C, 10,000 lux, 14 h light/10 h dark photoperiod) for 12 h, followed by unscrewing the bottle caps with a small opening to adapt for 6 h, and then removing the cap to adapt for 30 h ( Figure 7A). After carefully washing away the adherent medium on the roots, the plantlets were adapted in sterile water for one day and transplanted into plastic pots.
Pot experiments were performed under controlled conditions (28 • C, 10,000 lux, 14 h light/10 h dark photoperiod) at the Institute of GAP for Chinese Medicinal Materials, Fujian Agriculture and Forestry University. R. glutinosa plantlets after acclimatization were transplanted on 23 August 2018 and grown in plastic pots of 18 cm diameter and 15 cm height (1.38 kg soil per pot). Three plants were planted as three replicate sub-samples in each pot. Four treatments of replant disease levels were constructed by mixing two kinds of soils in different proportions. The soils were collected from the site where R. glutinosa had not been planted for at least 10 years (NP) and where R. glutinosa had been consecutively planted for three years (TP) in Wen County, Jiaozuo City, Henan Province, in the "geo-authentic" zone of R. glutinosa cultivation (34 • 56 N, 112 • 58 E). The air-dried soil samples were taken to the laboratory for this experiment. Four treatments thus included NP, 1/3TP (mixed by 2 NP soils and 1 TP soil), 2/3TP (mixed by 1 NP soil and 2 TP soils), and TP, NP of which was used as the control ( Figure 7B). and grown in plastic pots of 18 cm diameter and 15 cm height (1.38 kg soil per pot). Three plants were planted as three replicate sub-samples in each pot. Four treatments of replant disease levels were constructed by mixing two kinds of soils in different proportions. The soils were collected from the site where R. glutinosa had not been planted for at least 10 years (NP) and where R. glutinosa had been consecutively planted for three years (TP) in Wen County, Jiaozuo City, Henan Province, in the "geo-authentic" zone of R. glutinosa cultivation (34°56′N, 112°58′E). The air-dried soil samples were taken to the laboratory for this experiment. Four treatments thus included NP, 1/3TP (mixed by 2 NP soils and 1 TP soil), 2/3TP (mixed by 1 NP soil and 2 TP soils), and TP, NP of which was used as the control ( Figure 7B).

The Collection of Fresh Root and Rhizosphere Soil Samples
On day zero, three after acclimatization (DAA), the fresh roots were collected after carefully washing with sterile water and drying with absorbent paper. The samples of 0 DAP and 3 DAA are the same. At 3, 6 and 9 DAP, the fresh roots and their rhizosphere soil were carefully collected as described in   [10]. Briefly, the roots and the soil around the roots were carefully dug up using a sterilized fork spade and slightly shaken to remove loosely attached soil. The rhizosphere soil that was tightly attached to roots (1-3 mm zone around the root) was brushed off and collected. All of the collected samples were immediately frozen in liquid nitrogen and then stored at −80 °C for Step 4: Washing the culture medium carefully with sterile pure water; Step 5: Plantlets were adjusted to sterile pure water for 24 h; Step 6: Transplanting three plants per pot; (B) the phenotype changes of R. glutinosa under different replant disease stresses within 9 DAP. In the pot stage, 1/3TP, 2/3TP and TP are compared with NP, and 3 DAP, 6 DAP and 9 DAP are compared with 0 DAP (3 DAA). DAP: Days after planting. NP: Soil that was never planted with R. glutinosa for at least 10 years. TP: Soil that was consecutively planted with R. glutinosa in the same soils for three years.

The Collection of Fresh Root and Rhizosphere Soil Samples
On day zero, three after acclimatization (DAA), the fresh roots were collected after carefully washing with sterile water and drying with absorbent paper. The samples of 0 DAP and 3 DAA are the same. At 3, 6 and 9 DAP, the fresh roots and their rhizosphere soil were carefully collected as described in   [10]. Briefly, the roots and the soil around the roots were carefully dug up using a sterilized fork spade and slightly shaken to remove loosely attached soil. The rhizosphere soil that was tightly attached to roots (1-3 mm zone around the root) was brushed off and collected. All of the collected samples were immediately frozen in liquid nitrogen and then stored at −80 • C for further experiments for soil DNA extraction, absolute quantification of PS and FO, qRT-PCR of NB-LRR and measurement of plant hormones and physiological index.

The Extraction of Soil DNA and Its Method Comparison
Approximately 5 g of soil of each sample was weighed for the extraction of soil DNA. Three extraction methods were compared according to the electrophoretic strips. Method I referred to the conventional cetyltrimethylammonium bromide (CTAB) method [70]. Methods II and III used different extraction methods based on the optimization for removing humic acid. PCR and gel electrophoresis were used to evaluate the different DNA extraction methods for PS and FO. For PCR, specific primers of PS (PS for: 5 -GGTCTGAGAGGATGATCAGT-3 , PS rev: 5 -TTAGCTCCACCTCGCGGC-3 ) and FO (ITS1-F: 5 -CTTGGTCATTTAGAGGAAGTAA-3 , AFP308R: 5 -CGAATTAACGCGAGTCCCAA-3 ) were synthesized with reference to   [10] (SunYa Biotechnology Co., Ltd. Fuzhou, China). Conventional PCR was performed using a Thermo cycler instrument (ThermoFisher Scientific A24812, Waltham, MA, USA) to detect transcript abundance. Each 20-µL reaction contained 0.4 µM each primer, 0.5 U 2× Es taq MasterMix enzyme, cDNA and nuclease-free water. The amplification procedure was 95 • C for 2 min, 95 • C for 5 s, and 58 • C to anneal for 30 s, for 30 cycles. Horizontal gel electrophoresis (1% gel) was used to evaluate the effect of the extraction with a DL2000 DNA Marker (Takara, Japan) in 130 V, 200 mA (Liuyi DYY-12, Beijing, China). The results showed that the extraction effect of method III was relatively best (Supplementary Figure S1A).

Absolute Quantification of Pseudomonas spp. and Fusarium oxysporum
To accurately evaluate the shift of rhizosphere microorganisms to avoid the expression changes of reference genes in different growth phases and soil samplings [71], the characteristic rhizosphere microorganisms of beneficial PS and pathogenic FO were detected based on absolute quantification PCR (AQ-PCR).

Construction of the Recombinant Plasmid
Amplifications of PS and FO specific primers were performed by conventional PCR (as mentioned above) and touchdown PCR (the same 20-µL reaction; 95 • C for 2 min, 95 • C for 5 s, 50-60 • C to anneal for 30 s, 72 • C for 30 s, for eight cycles; 95 • C for 2 min, 95 • C for 5 s, 58 • C to anneal for 30 s, for 30 cycles), respectively (Supplementary Figure S1B). The bright electrophoretic strips were cut and extracted using gel pure DNA kits following the manufacturer's instructions (Magen D2111-02, China). The gel extraction solutions were concentrated to over 50 ng·µL −1 of cDNA using a concentrator for approximately 25 min at 1400 rpm (Eppendorf Concentrator Plus AG5305, V-AQ mode). DNA fragments were inserted into a vector and ligated overnight at 16 • C using the pMD19-T vector cloning kit following the manufacturer's instructions (Takara 6013, Japan). Then, 5 µL of vector DNA solution and 50 µL of E. coli DH 5α were blended and incubated in ice for 30 min. After heat shock at 42 • C for 60 s, the vectors were kept on ice for 3 min. The solution with 600 µL of liquid LB culture medium was closed using Parafilm and the culture was shaken for 60 min (37 • C, 200 rpm). Then, 45 µL of 5-Bromo-4-chloro-3-indolyl β-d-galactoside (X-Gal), 10 µL of isopropyl-l-d-thiogalactopyranoside (IPTG) and 200 µL of the culture solution were smeared evenly on solid LB culture medium containing 0.4% Ampicillin (Amp) and then incubated at 37 • C for 14 h. We selected 1-2 white single colonies and added 300 µL of liquid LB culture medium containing 0.4% Amp for co-culture (37 • C, 200 rpm) for approximately 14 h until the solutions were turbid. PCR and 1% gel electrophoresis as aforementioned were used to identify the size of DNA fragment comparing with a DL2000 DNA Marker (Takara, Japan). The solutions containing the appropriate size of DNA fragments were chosen to extract plasmids using Hipure Plasmid Micro Kit following the manufacturer's instructions (Magen, China). 10 µL solutions of plasmids were used for sequencing in Biosune, China (987 bp for PS, 433 bp for FO; Supplementary  Table S1).

Establishment of Standard Curve
The plasmid solutions of s containing the right size of DNA fragments were amplified again using plasmids primer RV-M/M13-47 by qRT-PCR (BIO-RAD CFX96, USA) (20-µL reaction contained 0.4 µM each primer, 0.5 U SYBR Premix EX Taq II (2×), cDNA and nuclease-free water following the Takara RR820A instructions; 95 • C for 2 min, 95 • C for 5 s, 55 • C to anneal for 30 s, 72 • C for 30 s, for 30 cycles). The DNA concentration of target genes were detected by a NanoDrop2000 spectrophotometer (Thermo Scientifi, USA) and then diluted to 0, 1, 2, 3, 4 ng/µL. The standard curves were drawn based on the DNA concentration of target genes and Ct values (Supplementary Figure S2). The calculation of plasmid copy number was based on Shirima et al. (2017) [72].

Determination of AQ-PCR
Ct values of recombinant plasmid containing 5 gradient concentration (as control) and soil DNA extracts were detected using specific primers of PS and FO by qRT-PCR (20-µL reaction was as mentioned above). The copy numbers of PS and FO were calculated based on the standard curve. All reactions were replicated three times.

qRT-PCR Analysis of NB-LRRs
For 35 previously identified NB-LRRs (Supplementary Table S2), RNA extraction of roots, reverse transcription and qRT-PCR analysis (BIO-RAD CFX96, USA) were conducted as described by   [41]. All reactions were replicated three times. The data were normalized on the basis of the 18S rRNA threshold cycle (Ct) value. The samples with the NP treatments were used as the controls at the same sampling time, and their normalized Ct values were set to 1. The relative gene expression of the other treatments was calculated using the 2 −∆∆CT method [73].

Measurement of ABA, SA, ET and JA
The contents of abscisic acid (ABA), salicylic acid (SA), ethylene (ET) and jasmonate (JA) were determined using a one-step double-antibody sandwich enzyme-linked immunosorbent assay (ELISA). Briefly, 1.0 g of fresh root was ground in 5 mL of phosphate-buffered saline (PBS) (0.01 M, pH 7.4) with an ice-cooled mortar and centrifuged at 2500 rpm for 20 min at 4 • C to obtain a supernatant for the ELISA analysis following the protocol described in Zhao et al. (2006) [74]. The mouse monoclonal antigen and antibodies against free ABA, SA, ET and JA were provided by MLBIO Co. Ltd., Shanghai, China. The hormone content was measured at 450 nm using a microplate reader (BIO-Tek ELX800, USA). Calculations of the ELISA data were performed as described in Wang et al. (2012) [75]. The recovery percentages obtained by using internal standards during extraction and analysis were all >90%.

Measurement of Root Activity and the Physiological Index
To determine root activity, the methodology described by Zhang et al. (2013) [76] was followed with modifications. Approximately 0.5 g of fresh root was mixed with 10 mL of a half-and-half blend of 0.4% triphenyl tetrazolium chloride (TTC) and 1/15 M PBS (pH 7.4) and incubated at 37 • C for 1 h, and then the reaction was stopped by 2 mL of 1 M H 2 SO 4 . The roots were homogenized in ethyl acetate with a capacity of 10 mL. The absorbance of the final solution was measured at 415 nm (Pgeneral T6-1650E, China). A standard curve was used to determine the concentration of root activity in the extract.
The determination of SOD, POD and CAT activities and MDA content were as described by   [48] and Deenamo et al. (2018) [77], respectively. The pretreatment was the same, and then 0.5 g of roots was homogenized in 5 mL of precooled PBS (0.05 M, pH 7.8) with a small amount of quartz sand. Extracts were centrifuged for 15 min at 13 000 rpm. The supernatant was used for the measurement of the four indexes. The colorimetric wavelengths were 560 nm, 470 nm and 240 nm for SOD, POD and CAT, and 600 nm, 532 nm, 450 nm for MDA (Pgeneral T6-1650E, China).
The H 2 O 2 content was determined by the KI method [78]. In short, 0.2 g of roots was homogenized in 0.8 mL of precooled 0.1% TCA (trichloroacetic acid) with liquid nitrogen. Extracts were centrifuged for 20 min at 19,000 rpm. Then, 0.5 mL of supernatant was added to 2 mL of KI (1 M) and 0.5 mL of PBS (100 M) for reaction at darkness for 1 h. The absorbance of the final solution was measured at 390 nm (Pgeneral T6-1650E, China). A standard curve was used to determine the concentration of H 2 O 2 in the extract.

Statistical Analysis
Multiple comparisons (LSD) and Pearson correlation were analysed with SAS statistical software (V9.1 SAS Institute Inc., Cary, NC, USA). Each value with three replicates represented as the mean ± SD. p < 0.05 was considered as significant between any two groups. A heatmap was generated via the hierarchical clustering method using the MeV 4.9.0 tool [79].

Conclusions
Our results indicated that R. glutinosa replant disease promoted Fusarium oxysporum proliferation in rhizospheres soil, but NB-LRRs were not consecutively responsive to the FO proliferation at the transcriptional level in R. glutinosa roots. The analysis on the relationships between Fusarium oxysporum numbers and NB-LRRs expression showed that the NB-LRRs responded to the Fusarium oxysporum invasion with a typical phenomenon of "promotion in low concentration and suppression in high concentration", and 6 NB-LRRs were identified as candidates for responding to R. glutinosa replant disease. Salicylic acid (SA), jasmonic acid (JA), ethylene (ET) and abscisic acid (ABA) involved in the formation mechanism of R. glutinosa replant disease. Importantly, salicylic acid, as an important signaling molecule, which may be inhibited by Fusarium oxysporum, was involved in the formation of replant disease in R. glutinosa. Concomitantly, the physiological indexes acting downstream of life activities were affected by complex replant disease stresses and exhibited strong fluctuations, resulting in the death of R. glutinosa. These findings provide important insights and clues into further revealing the mechanism of replant disease.

Conflicts of Interest:
The authors declare no conflict of interest.