Rice “Instant Inoculation” With Endophytic Fungus Causes Excessive Immune Response To Fusarium Proliferatum And Plant cell Death


 Aims: This article explored the effects and possible mechanism of different inoculation patterns of endophytic fungi Phomopsis liquidambaris B3 on the plant immune response and defense results to pathogen Fusarium. proliferatum.Methods: Rice seedlings pre-inoculated, instant-inoculated and non-inoculated with B3 were infected with F. proliferatum and grown in controllable conditions. The disease severity, F. proliferatum infection and plant growth were evaluated, and the activity of defensive enzymes, defense-related substances, reactive oxygen content, and defense hormone SA content were determined. Chemical treatment experiments and plant cell staining to verify plant cell death.Results: Pre-inoculation of endophytic fungus Ph. liquidambaris B3 may triggered the SA-dependent defense pathways of plants, increased defense-related enzyme activities and up-regulated the expression of defense genes, and thus decreased F. proliferatum colonization. However, instant-inoculation Ph. liquidambaris B3 inhibited its own root colonization, caused excessive burst of ROS and leaves cell death. Conclusion: For a specific plant pathogen, the instant-inoculation Ph. liquidambaris B3 when F. proliferatum infected leads to excessive burst of reactive oxygen species (ROS) and plant cell death. We recommend pre-inoculation of biological control agent and other prevention and control measures before pathogen infection to achieve the best control effect rather than instant control method applicated in the agricultural crop management.


Introduction
In the sustainable intensi cation of global agricultural production, crop health and food security are constantly receiving attention. In the long history of cultivation, humans have adopted farming strategies such as crop rotation to avoid soil-borne diseases, and targeted used microorganisms to against herbivores (Finkel et al. 2017). At present, the use of bene cial microorganisms to improve crop health and alleviate and protect plants from stress has become a research hotspot (Zandi et al. 2021;Sahoo et al. 2013). Currently, the Organic Materials Review Institute (OMRI) lists 174 'microbial inoculants' products and 274 'microbial products201', as crop fertilizers or crop management tools could be applied in agriculture (Finkel et al. 2017). Surprisingly, most bene cial microorganisms are much less effective in eld applications than in the laboratory due to environmental suitability, the persistence of bene cial microorganisms in the soil and other several limiting factor (Schreiter et al. 2014;Weidner et al. 2015).
This inevitably raises the question of how to achieve the precise application and management of bene cial microbes to ensure and increase the control effect in consideration of different actual characteristics (for example, application time and place of bene cial microbes, local climate and soil conditions). Thus, understanding and exploring the physiological changes and immune responses of host plants colonized by bene cial microorganisms and pathogens will be critical for realizing the pertinence and precise application and developing the management options that maximize bene cial microorganisms for crop diseases control and save agricultural production costs.
In order to survive, plants have evolved a complex and sophisticated immune system to recognize and adapt to the attack of pathogenic microorganisms (Jones and Dangl 2006;Cui et al. 2015). When pathogens invade plants, plant cell walls are rst attacked (Lee et al. 2019). Rice blast fungus Magnaporthe grisea can successfully invade host cells by producing appressorium to penetrate the plant cell wall (Park et al. 2002). Lignin, calluses, colloids and another effective ingredient for thicking and augmenting structural barriers of pathogen invasion (Hematy et al. 2009;Luna et al. 2010). In addition, the vast commensal and mutualistic microbial communities surrounding and, in the plants, also provide related necessary services Han et al. 2020). A recent study reported that the symbiotic microbe Chitinophagaceae and Flavobacteriaceae, which lives in the roots of beet seedlings, were found to be enriched in the endosphere, and the change of endophytic root microbiome constitutes a second microbiological layer of plant defense to protect plants once pathogen bypasses the rst layer and Studies have shown that bene cial symbiotic microorganisms can survive and reproduce in plants without causing obvious pathological effects of host plants, but assist plants in obtaining nutrients, promote plant growth, and induce plant resistance (Pieterse et al. 2014;Lugtenberg et al. 2009). The most typical representatives of these microorganisms are AM, root nodule symbioses and plant-growthpromoting bacteria (PGPR) (Peer et al. 1991;Barrio et al. 2020;Mun et al. 2020). Generally, induced resistance (IR) enhanced defensive capacity of the entire plant against a broad spectrum of pathogens (Walters et al. 2013). It is generally believed that the colonization of rhizosphere non-pathogenic or bene cial microorganisms in plant roots triggered ISR, which is mediated by JA-ET-sensitive pathway in plants, and enhances resistance to pathogen in non-induced site (Spoel and Dong 2012). SAR is triggered upon local activation of a PTI or ETI response by necrotizing pathogens or insects and predominantly mediated by SA and the accumulation of pathogenesis-related (PR) proteins (Corina et al. 2009). In fact, most immune responses in plants are highly similar or identical due to bene cial and pathogenic microbes share physiological features and an evolutionary proximity ). Transcriptome analysis of Arabidopsis. thaliana leaves live with commensal microbes showed that these symbiotic microorganisms do activate the rst layer of plant immune response. Nearly 400 genes were up-regulated in treated plants and partly overlapped with Pseudomonas syringae induced, while A. thaliana does not show disease symptoms (Vogel et al., 2016). Like pathogens, bene cial microorganisms also need to overcome or evade the immune response of plants in order to establish a long-term and intimate mutually bene cial interaction with the host. It has been reported that the colonization of PGPR in plant roots may require local suppression of PTI to prevent the production of antibacterial compounds by MAMP triggered (Wang et al. 2012;Zamioudis et al. 2012;Pieterse et al. 2014). However, how the differential time colonization of bene cial microorganisms regulates plant immune response and the underlying mechanisms upon pathogens challenge is still unclear. In particular, it is unknown whether the difference colonization time determines the winner in the race between plants and pathogens.
In this study, we explored the differences between bene cial endophytic microbes and pathogen inoculation sequence in plant immune response in order to seek the crop precise management for maximum protection of plants. A bene cial commensal fungus, Ph. Liquidambaris, namely B3, isolated from the inner bark of the stem of Bischo a polycarpa, as a root endophyte in rice with no visible infection symptoms (Chen et al. 2011). Previous studies have shown that Ph. Liquidambaris can control Rice spikelet rot disease (RSRD), and improve crop yields (Zhu et al. 2020). However, the difference of Ph. Liquidambaris inoculation time on the growth and ability of plants to resist pathogens is still unclear. Thus, this study addresses the following questions: (a) Whether different inoculation times of Ph. Liquidambaris affect plant disease resistance (b) Whether different inoculation times of Ph. Liquidambaris and pathogen infection cause the difference of plant immune system response or immune pathway.

Fungal stains
Ph. liquidambaris strain B3 and the plant pathogen F. proliferatum Ff-1 caused rice spikelet rot disease, were obtained from the Jiangsu Key Laboratory for Microbes and Functional Genomics, China. Ph. liquidambaris strain B3 was labeled with a green uorescent protein (GFP) by the vector plasmid pCT74 and colonized in root of rice as root endophyte. Ph. liquidambaris strain B3 was stored and activated as previously described . Fungal hyphae were collected and washed twice with sterile distilled water and suspend again in sterile distilled water as fungal inoculum. The plant pathogen F. proliferatum Ff-1 was stored and activated on PDA (200g − 1 potato extract, 20 g − 1 glucose, pH 7.0) for 5 d at 28°C. Add sterile distilled water and use sterile spreading rod to scrape spore suspension (10 6 spores mL − 1 ) as plant pathogen inoculants.

Plant material and growth conditions
The rice seed used was a common cultivar "Nanjing 5055" grown in Jiangsu Province, China. All rice seeds were surface disinfected with 70% (v/v) ethyl alcohol for 5 min, followed by 10% (v/v) sodium hypochloride (NaClO) for 3 min and washed with sterile distilled water (SDW) to remove excess NaClO and germinated in the dark at 28°C. The 10-day-old rice seedings were transferred to oat tray (8/15 cm, diameter/height) lled with sterilized soil substrate composed by the rice soil and vermiculite (2:1) and mixed thoroughly. For the inoculated group (E+), each rice seedling root was inoculated with 2 mL of the fungal inoculum by injector, and the control treatment (E-) was treated with 2 mL SDW.
In order to study the in uence of Ph. liquidambaris B3 different inoculation time on plant immune response, experiments were carried out, containing ve treatment: (a) infected with F. proliferatum Ff-1 after inoculation with Ph. liquidambaris B3 three days (BF, B3 pre-inoculation), (b) infected with F. proliferatum Ff-1 and inoculated with Ph. liquidambaris B3 at the same time (FB, B3 instant-inoculation), (c) infected with F. proliferatum Ff-1 alone (F), (d) inoculated with P. liquidambaris B3 alone (B), (e) noninoculated control(CK). The oat trays were randomly placed at a control incubator (28°C for 16-h light and 8-h dark) with nutrient solution every 4 d. Fresh plant was collected at 0, 2, 4, 6, 10 d for experiment after inoculated. In addition, pathogenicity experiments were carried out to prove that Ff-1 can colonize and cause disease in rice seedlings (Fig. S1).

Endophytic and pathogenic colonization
Fresh leaves and roots were collected and rinsed thoroughly with SDW to extract the rice DNA using an AxyPrep Multisource Genomic DNA Miniprep Kit (Axygen Biosciences, CA, USA). Ph. liquidambaris colonization of rice root was quanti ed with a speci c B3 ITS primer (Bf1/ Br1) set as previously described (Zhu et al. 2020). The quantitative detection of F. proliferatum in rice leaves was determined using speci c primer (Ff1/ Fr1) as previously described (Zhu et al. 2020).

Determination of chlorophyll content
Fresh leaves were collected, and 0.1 g samples were immediately ground in liquid N. Chlorophyll was extracted with 20 mL 80% acetone at 4°C in the dark for 24 h. The solution was centrifuges at 12,500 gat 4°C for 10 min. The supernatants were separated and analyzed for chlorophyll content, and the absorption value was measured at 645 nm and 663 nm as per Hunt et al. (2005). Each sample was ampli ed in triplicate in each experiment.

Defense-enzyme assay
Fresh leaves and roots were collected, and 0.1 g samples were immediately ground in liquid N, and extracted with 1 mL 0.1M phosphate buffer solution (pH 7.4, 8.0 g L − 1 NaCl, 0.2 g L − 1 KCl,1.44 g L − 1 Na 2 HPO 4 , and 0.24g L − 1 KH 2 PO 4 ), and centrifuged at 12,500 rpm at 4°C for 10 min. Enzyme activities of superoxide dismutase (SOD) and polyphenol oxidase (PPO) was measured using commercial kits (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) following the manufacturer's instructions. One unit of SOD activity was de ned as the quantity when the inhibition rate is 50%. One unit of PPO activity was de ned as an absorbance change of 0.01 per minute. Enzyme activities of chitinase and β-1,3-glucanase were detected with the 3,5-dinitrosalicylic acid (DNS) method according to Sridevi et al (2008). Colloidal chitin and laminarin (Sigma-Aldrich, Missouri, USA) were used as the substrates in chitinase and β-1,3-glucanase determined, respectively. One unit of enzyme activity (U) was de ned as the amount of reducing sugar released per hour from chitinase and β-1,3-glucanase. Each sample was ampli ed in triplicate in each experiment.

Defense-related substance assay
The total phenol content in rice plants was determined using the Folin-phenol method (Zhao et al. 2016). Fresh plant samples were fully grinded and extracted in 90% ethanol solution and lled 25 mL to volume.
2 mL ethanolic extract and 2 mL foline-phenol solution were added in 10 mL centrifuge tube, then add 2 mL 10% Na 2 CO 3 within 3 min and mix thoroughly in the dark for 1 h for detected at 700 nm. The control was determined with distilled water as extract solution. Each sample was ampli ed in triplicate in each experiment. Histochemical marker staining assay Trypan blue dye and Evans blue staining was used to observe plant cell death according to Morcillo et al.
(2019) described, with some modi cation. In brief, fresh plant leaf tissues were collected at 5 d after F. proliferatum infection and placed into 5 mL centrifuge tubes containing trypan blue or evans blue and boiled for 3 min, then stained at room temperature for 2 h for leaves. The leaves were treated with 95% ethanol until they were completely colorless.

The detection of O 2
•− by NBT staining and DAB staining for H 2 O 2 was performed as described by Wohlgemuth et al. (2002) and Bindschedler et al. (2006). In brief, fresh rice leaves were collected at 5 days after F. proliferatum infection and placed in 0.5 mg mL − 1 NBT or 1 mg mL − 1 DAB staining at 37 ° C for 2 h in darkness. Then the samples boil in a water bath for 10 min and treat with 95% ethanol until the leaves were completely colorless.
Quanti cation of plant hormone SA At 0, 2, 4, 10 d after treated, the leaves and roots of plant were samples for SA content determined according to Yuan et al. (2016). Brie y, 0.1 g samples were immediately ground in liquid N, and added 0.5 mL methyl alcohol vortex shocked for 1 min followed by ultrasonicated for 5 min and extracted at 4°C for 24 h. The methyl alcohol extracted was centrifugated at 14,000 rpm for 10 min at 4°C and the precipitation was resuspension by methyl alcohol. The methyl alcohol supernatant was collected after twice centrifugation and then 10 µL of 0.2 M NaOH was added followed by quickly rot-evaporate to remove methyl alcohol. Resuspend the precipitate with 250 µL 5% trichloroacetic acid and added 0.8 mL of ethyl acetate: cyclohexane (1:1, v/v). Collected the supernatants after extract twice and added 60 µL 0.2 M NaAc (pH 5.5) followed by quickly rot-evaporate, and dissolve the residue by 600 µL of methyl alcohol for detection.

RNA extraction and quantitative RT-PCR analysis
Real-time quantitative PCR (RT-qPCR) was performed to determine the expression levels of the genes of SA in plant. Plants were harvested at 0, 2, 4, 10 d after treated. Total RNA was extracted from plant samples using TRIzol reagent (Invitrogen, CA, USA), according to the manufacturer's recommendations.
The primers used in the study are listed in Table S1.
qRT-PCR was performed using the StepOne Real- proliferatum Ff-1 (F). At 7 d after treated, the leaves were collected for further detection.

Statistical analysis
All experiments performed in the study were at least three times. All statistical analyses were performed using SPSS 18.0 (SPSS, Inc., Chicago, USA) and the nal data were expressed as the mean with standard error (SE). When three or more groups were compared, a one-way ANOVA was performed followed by a Tukey's multiple-comparison test. The data were considered signi cantly different at P ≤ 0.05. Graphs and images were assembled using Adobe Photoshop CS6 (CA, USA).

Pre-inoculated of Ph. Liquidambaris alleviates the disease symptoms caused by F. proliferatum infection and reduce pathogen colonization
First of all, we wondered whether and how the time of B3 inoculation affected the growth and disease development after rice plants infected F. proliferatum. Here, we found that the rice steam length of plants instant-inoculated B3 signi cantly higher than pre-inoculated, while the rice stems were slender and prone to lodging, and the leaves were chlorosis and withered quickly after 10 d infected with F. proliferatum (Fig. 1a). However, pre-inoculation with B3 alleviated abnormal growth caused by F. proliferatum infection and promoted root growth (Fig. 1b, 1d). The fresh weight and dry weight of rice plants were signi cantly higher than those of the FB treated with 5.65% and 12.08%, respectively. However, there was no signi cant difference between rice instant-inoculation with B3 and rice infected with F. proliferatum alone ( (Fig. 1f).
To determine whether the time of B3 inoculation affects pathogens and B3 colonization in host plants, we evaluated the F. proliferatum and B3 colonization concentration in rice leaves and roots (Fig. 1). Compared with F. proliferatum infection alone, pre-inoculation with B3 remarkedly reduced the number of pathogens in rice leaves, while no conspicuous effect in rice leaves instant-inoculation with B3 (Fig. 1g).
On the contrary, we found that the amount of B3 colonization was signi cantly lower in rice roots instantinoculation with B3 than that in pre-inoculation with B3. In addition, F. proliferatum infection also reduced the amount of B3 colonization in rice roots (Fig. 1h).
Then we measured the chlorophyll content in rice leaves explore whether B3 colonization time affects photosynthesis and causes plant growth differences (Fig. 2). We found that pathogen infection signi cantly inhibited plant chlorophyll accumulation, while instant-inoculation with B3 did not alleviate this inhibition. The content of chlorophyll a and carotenoids in rice leaves instant-inoculated with B3 were signi cantly decreased by 19.10% and 37.93%, compared with rice pre-inoculated with B3 ( Fig. 2a,   2d). In addition, B3 inoculation increased the chlorophyll content in rice leaves, which may promote the photosynthesis of plants.
In conclusion, these results indicate that different inoculation time of B3 affects the growth and disease development of rice after F. proliferatum infection.
Different inoculation time of Ph. Liquidambaris causes opposite defense responses in rice to F. proliferatum Then, to explore whether B3 inoculation time affects the immune response of plants to F. proliferatum infection, the accumulation of resistant substances like phenols and the activities of defense-related enzymes were measured (Fig. 3, 4). Here, compared to instant-inoculation with B3, the total phenol content in rice leaves pre-inoculation with B3 increased signi cantly from 0 to 4 d after F. proliferatum infection, and was higher than that instant-inoculation with B3 and F. proliferatum infection alone (Fig. 3a). During the experiment period, the total phenol content in the roots of all the treated rice plants decreased. However, the total phenol content in the rice roots instant-inoculation with B3 immediately decreased, while rice pre-inoculated with B3 increased temporarily at 4 d after F. proliferatum infection. In addition, the inoculation of B3 increased the total phenol content in rice root, compared with F. proliferatum infection alone (Fig. 3b).
The results of defense-related enzyme activities showed that pre-inoculation with B3 signi cantly enhanced the chitinase activity in rice leaves from 0 to 4 d after F. proliferatum infection, while the chitinase activity in rice plants instant-inoculation with B3 continued to decrease during 0-10 d, even lower than that of plants F. proliferatum infection alone (Fig. 4a, 4b). Similarly, the β-1,3-glucanase activity in rice leaves pre-inoculation with B3 was signi cantly enhanced than plants instant-inoculation with B3 from 0 to 6 d, but there was no signi cant difference from the plants F. proliferatum infection. However, no signi cant difference was found in the activities of chitinase and β-1,3-glucanase in rice roots between B3 inoculation time (Fig. 4c, 4d).

Ph. Liquidambaris inoculation time causes different plant hormone response level after F. proliferatum infection
Salicylic acid (SA) is an important plant hormone and plays a key role in the defense response against pathogen infection (Murphy et al. 2020). To investigated the behavior of SA in different time B3inoculated plants under the pressure of F. proliferatum, we measured SA concentrations in leaves and roots of rice (Fig. 5a, 5b). We found that F. proliferatum infection increased the SA concentration in rice leaves, which may activate the SA-dependent plant resistance immune pathway. However, the SA concentration in pre-inoculation with B3 rice leaves increased sharply after F. proliferatum infection, and reached 1.64 times at 2 d of rice instant-inoculation with B3. In addition, the SA concentration in rice leaves of instant-inoculation with B3 was signi cantly lower than F. proliferatum infection alone (Fig. 5a). Different from the leaves, the SA concentration in rice roots of instant-inoculation with B3 was signi cantly increased from 2 to 6 d (Fig. 5b). Then we evaluated the expression of the important gene OsPAL involved in SA synthesis and the OsPR1a gene related to the SA pathway in rice leaves. Compared with pre-inoculation with B3, after F. proliferatum infection, the expression of OsPAL2 and OsPR1a in rice leaves instant-inoculation with B3 was signi cantly down-regulated by 1.63 and 1.54 times, respectively ( Fig. 5c, 5d).
These results indicate that B3 inoculation time affects the immune response of rice after F. proliferatum infection, including defense enzyme activity, accumulation of resistant substances, and plant hormones response. Instant-inoculation with B3 caused lower defense response than pre-inoculation, which imply that B3 inoculation time led to different levels of plant immune response and even different immune pathways.

Ph. Liquidambaris instant-inoculation leads to an excessive outbreak of plant ROS and abnormal cell death
After infected by pathogen, plant reactive oxygen species (ROS) will burst out from the infected site to activate the plant immune system and inhibit the spread of the pathogen, the main form is H 2 O 2 (Das et al. 2014). We observed that both pre-inoculation and instant-inoculation with B3 enhanced the H 2 O 2 outbreak after F. proliferatum infection (Fig. 6). It is worth noting that the H 2 O 2 in the leaves of rice instant-inoculation with B3 increased strongly from 0 to 2 d, which was 1.20 times and 1.39 times that of pre-inoculation B3, and was always signi cantly higher during the experiment period (Fig. 6a, 6b). On the contrary, this difference was not observed in rice root. In addition, the results of the expression of related genes involved in the synthesis of ROS showed that OsRboha and OsRbohb were signi cantly upregulated in rice leaves after F. proliferatum infection, which were 1.82 times and 2.29 times higher than those of pre-inoculation with B3 (Fig. 6c, 6d). Surprisingly, the enzyme activities involved in the ROS scavenging showed opposite results. During the entire experimental period, the SOD enzyme activity in rice leaves instant-inoculation with B3 was signi cantly lower than that in the pre-inoculated rice, the POD enzyme activity also showed similar changes (Fig. 7a, 7c). On the contrary, this phenomenon did not appear in rice roots (Fig. 7b, 7d).
These results suggest that instant-inoculation with B3 caused a large burst of ROS. Because high levels of ROS may cause cell death, then we performed histochemical staining on rice leaves (Fig. 8). The results showed that the cell death of leaves in pre-inoculation with B3 was signi cantly reduced, while deteriorated in instant-inoculation with B3 (Fig. 8a, 8c). In addition, the H 2 O 2 deposition in leaves of instant-inoculation with B3 was signi cantly higher than pre-inoculation with B3 (Fig. 8a). This seems to Then, we use the H 2 O 2 scavenger (DMTU) for further experiment (Fig. 8). The results showed that the disease in rice leaves with DMTU exogenous addition was alleviated, including reduction of H 2 O 2 deposition and cell death (Fig. 8b). Compared with F. proliferatum infection alone, exogenous addition of H 2 O 2 aggravated the disease and cell death in rice leaves, while disappeared in leaves added with H 2 O 2 and DMTU (Fig. 8b).
Based on the above results, we speculate that rice plants pre-inoculation with B3 may activate and enhance the immune response after pathogen infection which may be closely related to the SA hormonedependent defense pathways, while instant-inoculation with B3 may cause programmed cell death (PCD) due to the excessive outbreak of ROS.

Discussion
Bene cial plant microorganisms have been recommended as biological control agents (BCAs) in agricultural production due to their safety and environmental-friendliness. However, low adaptability and e ciency limit their wide use, compared with traditional chemical pesticides (Berg et al. 2019). Studies have shown that different application treatments of bene cial microorganisms affect the eld biological control effect. In the previous research, we found that the treatment of compound bene cial microbial agents did not increase the control effect of RSRD, while single strain showed better (Zhu et al. 2020).
However, few studies focus on different application times on disease control of bene cial plant microorganisms, which may be a key point for affecting the disease control effects of BCAs in actual production. For example, Xu et al. (2017) found that the release time of endophytic Streptomyces endus OsiSh-2 led to signi cant differences on rice blast control. Spraying OsiSh-2 before rice blast occurrence was better with a disease index of 30.48%, which is obviously lower than spraying after (59.25%). In fact, the difference in application time may lead to the reversal of the relationship between pathogen and bene cial microorganisms, which may be an important limiting factor affecting the application effect in actual production. In this study, the "occupancy" time of bene cial microorganisms in the host plant leads to opposed defensive outcomes: (1) Pre-inoculation with B3 activates rice induced resistance, alleviated the abnormal growth and reduced the disease symptom, stimulates and enhances the plant defense ability and immune response to pathogen challenge; and (2) "instant" inoculation with B3 (that is, when suffering from pathogen infection) causes violent defense response accompany with excessive ROS outbreak, resulting in plant cell death.
The potential and importance of bene cial microorganisms, such as PGPR, to promote plant growth and improve health has been widely concerned (Babalola et al. 2010;Shoresh et al. 2010). Consistent with previous results, in our experiments, B3 promoted the growth of rice roots and lateral root development, while F. proliferatum caused the leaf curling and stem morbid elongation, resulting in chlorosis and unhealthy plants. We observed that the disease and growth stress of rice leaves instant-inoculation with B3 were not decreased, while reduced and alleviated in rice leaves pre-inoculation with B3. The signi cant difference in the colonization of F. proliferatum in rice leaves instant-inoculation with B3 and preinoculation with B3 may explain this result. Interestingly, the colonization of B3 in rice root instantinoculation with B3 was signi cantly lower than that pre-inoculation. In this case, the ability of bene cial microorganisms and pathogen to colonize the host plant determines the outcome of this competition.
Due to their superiority in viability or aggressiveness, pathogen rst "pre-emptively" on the surface of plants to occupypriority colonization rights and niches, and produce toxins or stimulating metabolites and activate plant defense responses, which may limit the colonization of endophytes ( Studies have shown that the chloroplast structure was disintegration and chlorophyll synthesis blocked after pathogen invasion, leading to symptoms chlorosis and yellowing of leaves (Zabala et al. 2015).
Here we found that the leaves chlorophyll a and carotenoids content in rice pre-inoculation with B3 was remarkably higher than instant-inoculation, and no difference between instant-inoculation with B3 and F. proliferatum infection alone, which indicates that pre-inoculation of B3 decreased the damage of F. proliferatum. When tobacco plants are infected with Phytophthora nicotianae, photosynthesis is downregulated accompanied by the complete activation of defense responses and HR responses (Scharte et al. 2010). On the contrary, plants with strong photosynthesis accumulate more resistant substances, such as phenolics, which can inhibit the growth of pathogen directly and participate in the formation of physical barriers to plant cells as precursors for lignin synthesis (Farkas et al. 2005). This is consistent with our results. The rice leaves of pre-inoculation with B3 accumulate remarkedly more phenols, which may augment structural barriers of pathogen invasion and improve the rice resistance to F. proliferatum. In addition, plants will secrete hydrolase enzymes such as chitinase and β-1,3-glucanase to degrade the fungal cell wall to destroy the structure of pathogen fungal to limit pathogen colonization on the leaf surface (Cheng et al. 2020). In our study, the hydrolase activities of rice in pre-inoculation with B3 were enhanced after F. proliferatum infection, especially the chitinase activity of rice leaves. In summary, these results suggest that pre-inoculation with B3 alleviates the growth stress after F. proliferatum infection, stimulates the increase in the synthesis of plant basic defense substances, and enhances the immune response of rice.
Bene cial microorganisms form papillary structures that penetrate the epidermis of plants to create activation of salicylic acid-dependent signal transduction to resist soil-borne pathogen Fusarium oxysporum. In addition, genomic analysis showed that Ph. Liquidambaris B3 was closely related to the pathogen Magnaporthe grisea shared a common ancestor of pathogenic fungi, and retained invading nails, which may lead to identi cation of endophytic fungus B3 and pathogen through similar or common defense pathways. Plants evolved a series of antioxidant enzymes such as superoxide dismutase (SOD) and peroxidase (POD) to control the balance of ROS in order to avoid excessive accumulation of ROS in cells (Peter et al. 2019). In the current study, the SOD enzyme activity in rice leaves instant inoculated with B3 was signi cantly lower than that pre-inoculated with B3, and the POD enzyme activity as well. These results Many studies identi ed that induced resistance (IR) enhance plant defense capabilities rather than immediately active expensive defenses, which usually come at the expense of plant growth. In this stress, plants need to allocate a large amount of energy to activate strong defense responses resulting in energy overdraft that no extra energy to allocate for plants growth (Pieterse et al. 2014). In our research, we found that although instant inoculated with B3 after F. proliferatum infection activates an intense immune response leads to excessive accumulation of ROS and cell death. Obviously, this is an ineffective plant defense response. In addition, necrotrophy plant pathogens such as most Fusarium sp. must absorb nutrients from dead plant tissues and cells, which may accelerate the process of plant death (Cachinero et al. 2010). These results raise a referable suggestion, that is, for certain plant pathogens, the application of bene cial microorganisms may be ineffective and meaningless when the disease appeared or has occurred, and such "belated" management measures may even aggravate the disease in actual agricultural production, which is a choice that outweighed the gain. is, prevention is more important than treatment.
Our study emphasizes that the application time of bene cial microorganisms may be an important reason for limiting the effect of plant disease prevention and control in agricultural production. Only accurate and pre-relevant treatments before pathogen infection or disease occurrence can effectively prevent the occurrence and spread of diseases. Although our study aimed at a speci c plant pathogen, it is indispensable to ensure the e ciency of plant disease control and crop yield on related research on different plant pathogens. In addition, the population density of bene cial microorganisms in the rhizosphere or surface of plants also limit the expression of induced resistance in eld crops (Weller et al. 2002). However, the effect of inoculated population density of endophytic fungi B3 on plant disease control is not clear, and further studies are needed. Our research provided new strategies and perspectives for biological control in the large agricultural production environment, that is, considering the precise application factors of biological control agent (BCAs) such inoculation time and inoculation density may be crucial to ensure and improve the effect of farmland crop disease control.

Conclusion
The main mechanisms of biological control seem to be summarized into two types: (1) direct interaction of microorganisms and (2) stimulated and activated the plant immune system. Our results showed that the inoculation time of rhizosphere bene cial microorganisms affects the immune response of plants and leads to the opposite result of disease control. Pre-inoculation of endophytic fungus Ph. liquidambaris B3 may triggered the SA-dependent defense pathways of plants, induced plant system resistance, enhanced and promoted triggering faster and stronger immune responses upon the subsequent pathogen challenges to suppressed F. proliferatum infection in rice leaves, including increased defense-related enzyme activities and resistance substances accumulation and up-regulated the expression of defense genes. However, instant inoculation Ph. liquidambaris B3 (that is, when pathogen infection or has happened) up-regulated the expression of reactive oxygen synthesis genes, caused excessive burst of ROS and cell death. These results emphasized the signi cance of the application time of bene cial microorganisms in plant disease control (here it can be considered that the instant inoculation B3 simulated the situation after pathogen infection in the farmland environment).

Declarations
Compliance with ethical standards

Con ict of interest
The authors declare that they have no con ict of interest.     Comparisons between different treatments were considered signi cant at p < 0.05 following Tukey's multiple-comparison test.