View all hidden authors and organizations

Edited by Jeffery L. Dangl of the University of North Carolina at Chapel Hill, approved on October 5, 2021 (reviewed on May 26, 2021)

Although plant growth promoting bacteria (PGPB) can improve plant performance, only a few mechanisms have been identified so far. We show the sulfur metabolism in PGPB Enterobacter. SA187 and Arabidopsis plants play a key role in plant salt stress tolerance. Salt stress induces a sulfur starvation response in plants, which is attenuated by SA187. Arabidopsis sulphur metabolism mutants are allergic to salt stress, but can be rescued by SA187. Most plant sulfur metabolism occurs in the chloroplast and is related to the stress-induced accumulation of reactive oxygen species inhibited by SA187. This work shows that the salt stress tolerance of plants requires coordinated regulation of the sulfur metabolism pathways of beneficial microorganisms and host plants.

Enterobacter SA187 is a root endophytic bacteria that can maintain plant growth and yield under abiotic stress conditions. In this work, we compared the metabolic circuits of Arabidopsis and SA187 in free-living and endogenous interaction states. The interaction between SA187 and Arabidopsis induces great changes in bacterial gene expression chemotaxis, flagella biosynthesis, quorum sensing, and biofilm formation. In addition to changing bacterial carbon and energy metabolism, various nutrient and metabolite transporters and the entire sulfur pathway are up-regulated. Under salt stress, Arabidopsis is similar to plants under sulfate starvation, but it is different when it is colonized by SA187 that is reprogrammed with the Arabidopsis sulfur regulator. Therefore, SA187 partially or completely rescued the salt hypersensitivity of multiple mutants of Arabidopsis sulfur metabolism, just like adding sulfate, L-cysteine ​​or L-methionine. SA187 partially rescued many components of sulfur metabolism in chloroplasts. Finally, the salt-induced accumulation of reactive oxygen species and the hypersensitivity of LSU mutants were inhibited by SA187. LSU encodes a central regulator that links sulfur metabolism with chloroplast superoxide dismutase activity. The salt stress tolerance of Arabidopsis requires coordinated regulation of sulfur metabolism pathways in beneficial microorganisms and host plants, which may be a common mechanism used by different beneficial microorganisms to mitigate the harmful effects of different abiotic stresses on plants.

Plant growth promoting bacteria (PGPB) have the ability to establish a mutually beneficial relationship with plants, thereby promoting plant growth. PGPB can use different strategies to reduce the vulnerability of plants to biotic and abiotic environmental stress (1). PGPB can live in the soil (rhizosphere) around plant roots, attach to the surface of roots, stems, or leaves, or act as an endophyte in plant tissues. PGPB can promote plant growth through different mechanisms, such as absorbing nutrients (phosphate, nitrogen, iron, etc.) from the soil, regulating plant hormone levels (auxin, ethylene, abscisic acid, etc.), or enhancing plant growth. Resistance to pathogens is developed by activating a defense mechanism called induced systemic resistance (ISR) or producing antimicrobial agents (2, 3). In the past few decades, PGPB, as a biotechnology tool, has received attention as a substitute for traditional fertilizers and pesticides in agriculture (4).

Monitoring changes in metabolic pathways is an effective method to identify beneficial plant-microbe interaction factors and regulatory processes (5). After PGPB interacts with the host plant, both interacting organisms trigger global changes in gene expression and metabolic pathways. RNA sequencing (RNA-Seq) has been proven to provide highly accurate and sensitive transcriptional analysis for microorganisms and host plants, and helps greatly improve the biological insights of the interacting parties (6⇓ –8). Nevertheless, the application of RNA-Seq in the context of plant-microbe interactions is mainly focused on the response of the host plant (9⇓ ⇓ –12). Many studies have investigated the transcriptome changes of different PGPB species cultivated under conditions designed to simulate endogenous lifestyles (13⇓ ⇓ ⇓ ⇓ ⇓ ⇓ –20). However, studies on the transcriptional changes of PGPB after host interaction under plant conditions are still limited (21).

In the past few years, the potential of the endogenous microbiota of desert plants as biological fertilizers has been studied (22, 23). Strains isolated from desert plants have proven that they are effective in promoting the growth of important agronomic crops such as rapeseed, cucumber and alfalfa (24, 25). One of these strains is Enterobacter. SA187 is an endogenous PGPB, isolated from the root nodules of the endogenous desert plant Indigofera argentea in the Jizan region of Saudi Arabia (26). SA187 has been described as promoting multi-stress tolerance under desert cultivation conditions of the crop plant alfalfa and laboratory conditions of the model plant Arabidopsis thaliana (12). Our recent report on Arabidopsis shows that the promotion of salt stress tolerance by SA187 is mediated by activating the plant ethylene signaling pathway. In addition, the results indicate that SA187 induces salt stress tolerance through the production of 2-keto-4-methylthiobutyric acid (KMBA) (12), which is known to be converted to ethylene in plants (27). In order to deepen our understanding of the molecular mechanism of the beneficial association between SA187 and Arabidopsis, here we compare the transcriptomes of SA187 and Arabidopsis before and after the establishment of a functional interaction under normal and salt stress conditions. We show that the endogenous colonization of Arabidopsis by SA187 is accompanied by genetic reprogramming of host and bacterial primary and secondary metabolic pathways, revealing that bacterial and plant sulfur metabolic pathways are key players in beneficial plant-microbe interactions.

In order to understand the process of endogenous interaction induction with Arabidopsis roots in SA187, RNA-Seq analysis was performed as shown in Figure 1A. The interaction of SA187 with Arabidopsis roots was assessed by electron microscopy (SI Appendix, Figure S1A), and the level of bacterial proliferation was similar under non-salt and salt conditions, and was assessed by qRT-PCR (SI Appendix, Figure S1B). Approximately 16 to 20 million double-ended reads were obtained from a single bacterium, and 60 to 82 million reads were obtained from double SA187-Arabidopsis RNA-Seq (SI appendix, Table S1). Due to the small proportion of bacteria/plants in RNA obtained from double samples, most sequencing reads are mapped to the Arabidopsis TAIR10 (Arabidopsis Information Resource) genome, and only <1% of bacterial reads (SI appendix, Table S1) . Although the number of bacterial reads obtained was small, in-depth sequencing analysis confirmed the quality of all RNA-Seq libraries (SI appendix, Figure S1C). Pearson's linear regression coefficients, FPKM (per thousand base fragments per million mapped reads) profile, and the distribution of mapped reads show a good correlation between biological repetitions (for free-living and endogenous samples, R2> 0.98 and ~ 0.90, respectively) with the exception of copying SB1 (R2 = 0.73), and was therefore excluded from further analysis (SI appendix, Figure S1 D and E).

Experimental setup: SA187 was incubated for 4 hours at 28°C in 1/2MS without (B) or with 100 mM NaCl (SB). Arabidopsis seedlings germinate for 5 days with or without SA187, then transfer to fresh 1/2MS plates for 12 days, without (P, PB) or 100 mM NaCl (SP, SPB), and grow vertically for 12 days (22 °C, 16/8 hour light/dark cycle) (A). SA187 DEG Volcano diagram under free life and endogenous conditions with and without salt: In order to analyze the influence of salinity, SA187 in 1/2MS 100 mM NaCl (salt stress) is compared with that in 1/2MS (simulation) The growth of SA187 is compared with free-living and plant-related conditions (B). MA graph: Representation of the fold change of gene expression (M axis) and the average gene expression value (A axis) of each comparison. The blue dotted line indicates that the differential expression fold change = 2 [log2(fold change) = 1]. These genes show that M> 1 is up-regulated, and M <-1 is down-regulated. M = log2(log10(FPKM2)/(log10(FPKM1)), A = 1/2log2(log10(FPKM1)*log10(FPKM2)). The zero value of FPKM is ignored. The scatter plot is by R (Rstudio version 1.0) Generated. 136) (C). Venn diagram shows common or specific up-regulated and down-regulated genes under simulated and salt stress conditions, comparing the endogenous and free-living status of SA187 (q value <0.05) (D). KO function analysis: Under non-salt (PB vs. B) and salt stress conditions (SPB vs. SB) (E), metabolic processes are regulated in response to the adaptation of plant-related lifestyles. Metabolic processes are up-regulated and down-regulated in response to plant associations under control and salt stress conditions (left) or specific to control conditions (center) or salt stress (right) (F). y-axis = KEGG function category. The x-axis = the number of genes or KO identifiers in KEGG comes from the knowledge base of each regulatory gene subset in our analysis. Up-regulated metabolic processes are represented by red, and down-regulated metabolic processes are represented by blue.

By applying a differential expression cutoff value of q=0.05 and a fold change >2, RNA-Seq data showed that the endogenous interaction between SA187 and Arabidopsis induced significant changes in the bacterial transcriptome, regardless of the presence or absence of salt. Under salt-free conditions, compared with free-living bacteria, endophytic SA187 has 1,050 genes up-regulated and 587 genes down-regulated (PB and B) (Table 1 and Figure 1B). Under salt stress, the comparison of endogenous and free life showed 696 up-regulated genes and 633 down-regulated genes (SPB and SB) (Table 1 and Figure 1 B and C). The addition of 100 mM NaCl to the growth medium only slightly affected the gene expression of free-living and endogenous SA187 (SB vs. B, SPB vs. PB) (Table 1 and SI appendix, Figure S2). According to these data, the core set of 519 up-regulated and 404 down-regulated genes in SA187 participated in the endogenous lifestyle under salt and non-salt conditions (Figure 1D), and similar biological processes were changed in all endogenous samples ( Figure 1 E and F). A total of 50 key genes related to the endogenous lifestyle of SA187 were selected for qRT-PCR verification. The vast majority of qRT-PCR expression patterns of SA187 gene confirmed the reliability of RNA-Seq results (SI appendix, Figure S3 A ​​and B and S4A).

Standardized DEG in Enterobacter genus (FPKM, q value <0.05). SA187

In order to gain insight into the biological functions of the identified differentially expressed genes (DEG), we used the Kyoto Encyclopedia of Genes and Genomes to perform KEGG orthologous (KO) and Brite mapper analysis (SI appendix, Figure S5).

During the endogenous interaction between SA187 and Arabidopsis thaliana, the biological processes involved in host plant colonization, such as chemotaxis, flagella assembly, quorum sensing (QS) and biofilm formation, are all up-regulated (SI Appendix, Figure S5 And S6 and data set) 4). Interestingly, the genome of SA187 contains five paralogous genes encoding flagellin (FliC) (26). Although three of the five genes are poorly expressed under all conditions, they have the highest homology with Pseudomonas aeruginosa flg22 motif (28), SA187PBcda_000004844 (fliC.4) and SA187PBcda_000004845 (fliC.4) ( (77.3% identical) of the two flagellin genes are induced six to eight times when plants interact (SI Appendix, Figures S3A and S4A and Data Set 4). In addition, many methyl-accepting chemotactic proteins and the two-component system (TCS) CheA/CheB are involved in the signal cascade that controls flagella assembly, and the chemotactic proteins CheW and CheZ are also upregulated when plants interact (SI appendix, figure). S3A and data sets 4 and 5).

In the endogenous state, the TCS of SA187 participating in QS, QseC/QseB and FusK/FusR was increased (data sets 3 and 5). The AI-2 transporter (LsrBCDA), along with the lsr operon transcription repressor LsrR and terminator LsrF, are also up-regulated (data sets 1 and 3). In addition, many genes involved in the formation of biofilms were induced, including the subunits of the curly fiber biosynthetic complex: CsgB and CsgC, and the poly-N-acetylglucosamine (PGA) synthase subunits PgaA and PgaC (Data Set 6) .

In the endogenous state of SA187, the phosphotransferase system (PTS) involved in the import of glucose, sucrose, 2-O-α-mannosyl-D-glyceric acid, cellobiose and β-glucoside is strongly upregulated (SI Appendix ,figure 1). S6 and data set 2). Many ABC transporters have also been upregulated, including sulfate CysPUWA, maltose/maltodextrin MalEFGK, D-methionine MetNIQ, and signaling molecule autoinducer 2 (AI-2) LsrABCD (SI Appendix, Figure S6 and Data Set 1 ) Transporter. Interestingly, it was found that subunits involved in the transport of osmoprotective agents, such as glycine betaine, proline, spermidine and putrescine, as well as ABC.SP.A, ABC.SP.P1 and ABC.SP.S, are regulated (Data sets 1 and 3).

During the beneficial interaction, a large number of bacterial genes for carbohydrate metabolism are upregulated. Although the glycolysis/gluconeogenesis and pentose phosphate pathways are down-regulated, the tricarboxylic acid (TCA) cycle and glyoxylate pathway are up-regulated (SI Appendix, Figure S5 and Data Sets 10 and 11). Among the down-regulated genes, we identified phosphofructokinase (PfkB) (data set 11), which is one of the key regulatory points that make glycolysis irreversible, and pyruvate dehydrogenase (PoxB), which is involved in glycolysis. Another rate-limiting step (data set 11). On the other hand, citrate synthase was up-regulated (data set 11), indicating that SA187 preferentially uses the TCA cycle to generate energy in the form of GTP and NADH in plants.

Oxidative phosphorylation is a metabolic process that produces energy required for almost all important processes. It is upregulated during interactions, including O-cytochrome (CyoCDE) and several subunits of F0/F1-ATP synthase, which convert across membranes. The proton gradient enters ATP (data set 12). Similarly, several subunits of NADH-dehydrogenase (complex I) and succinate dehydrogenase (SDH) (complex II) are also part of the TCA cycle and are also strongly upregulated during beneficial interactions (SI appendix, Figure S6 and data set) 11 and 12).

The availability of sugar is considered to be the main determinant of the adjustment of microbial metabolism to an endogenous state (16). After interacting with Arabidopsis, the strong up-regulation of SA187 ABC transporter and PTS indicates that the plant host actively absorbs a variety of nutrients. These results are consistent with reports of Bacillus amyloliquefaciens FZB42, SQR9 or Enterobacter. 638 (13, 15, 16), indicating that glucose, maltose or inositol in corn root exudates induced the expression of the corresponding bacterial ABC and PTS transport systems. To evaluate whether the endogenous SA187 gene expression profile is mainly driven by the sugar availability of the host plant, we analyzed the expression of 40 endogenous SA187 marker genes during growth in a medium supplemented with 1% sucrose. Among the marker genes, 43% showed an expression profile similar to the endogenous state (SI Appendix, Figure S3 CE and S7), indicating that nutrient availability is the main driver of SA187 metabolic regulation. Nevertheless, 57% of bacterial genes did not respond to the addition of sucrose to the growth medium (SI appendix, Figure S7), indicating that other factors are also responsible for modifying the SA187 metabolic network during the Arabidopsis interaction.

Our previous work identified the methionine salvage pathway intermediate KMBA as an important metabolite provided by SA187, which can induce salt tolerance in Arabidopsis (12). In addition, studies have shown that the active metabolism of SA187 is required because heat-inactivated bacteria also lose its beneficial effects (12). However, it is currently unclear where the sulfur that forms KMBA and whether only the methionine salvage pathway is upregulated. Our comparative transcriptome analysis clearly shows that after interaction with Arabidopsis, the SA187 aslA gene that mobilizes soil sulfur is up-regulated in parallel with the ABC sulfate transporter CysPUWA (Figure 2 and SI appendix, Figure S6). After cells ingest SA187, adenylate sulfate transferase (CysND) converts sulfate into APS (adenosine 5'-phosphate sulfate), and then adenylate kinase (CysC) produces PAPS (3'-adenosine phosphate) -5'-phosphoric sulfate), which can then be converted to sulfite (Figure 2 and SI appendix, Figure S6). The up-regulation of the SsuABC transporter expands the possibility of alkane sulfate in the environment where SA187 is used. Consistently, FMNH2-dependent sulfonate monooxygenase (SsuD) and FMN reductase, which provides FMNH2 (SsuE) for SsuD, which convert alkane sulfonates to sulfites, are also upregulated (Figure 2 and SI appendix) , Figure S6 and data sets 1, 9 and 10). Cysteine ​​kinase (CysK) catalyzes the synthesis of L-cysteine ​​after the enzymatic reduction of sulfite to sulfide. The enzymes involved in the conversion of L-cysteine ​​to L-methionine, CBS, TCH and MetE, together with the MetE transcriptional regulator MetR, are also highly upregulated when SA187 interacts with Arabidopsis (Figure 2 and SI appendix, Fig. S6 and data sets 9 and 10). In addition, genes encoding the methionine salvage pathway TyrB, MtnN, MtnK, MtnC, MtnD and SpeD enzymes showed significant up-regulation during plant interactions (Figure 2 and SI appendix, Figure S6 and data sets 9 and 10). These results It shows that the entire sulfur metabolism of SA187, not just the methionine salvage pathway, is strongly induced when interacting with Arabidopsis.

Compared with free-living bacteria, during the colonization of Arabidopsis endophytes, a heat map (A) was constructed using the up-regulated (red) and down-regulated (blue) genes encoding sulfur metabolism transporters and enzymes. Arabidopsis thaliana up-regulates the SA187 sulfur metabolism pathway. APS = adenosine 5'-phosphate sulfate, PAPS = 3'-adenosine phosphate-5'-phosphate sulfate, KMBA = 2-keto-4-methylthiobutyric acid (B).

Next, we performed inductively coupled plasma emission spectroscopy (ICP-OES) analysis on Arabidopsis branches and roots cultured for 21 days without salt and 100 mM NaCl in the absence or presence of SA187. Interestingly, under salt stress, we found that the sulfur content in the roots of Arabidopsis thaliana planted with SA187 was significantly higher, but not in the buds (Figure 3A and SI appendix, Figure S8A). Salt stress strongly affects plant growth and photosynthesis, which may be due to the amount of iron-sulfur cluster protein in the photosynthetic apparatus of chloroplasts, and its interference leads to abnormal electron transfer and reactive oxygen species (ROS) generation (29). Sulfur metabolism is also related to the successful establishment of plant-microbe interactions between beneficial and pathogenic microorganisms (30, 31).

The effect of SA187 inoculation in Arabidopsis on 1/2MS 100 mM NaCl. Sulfur levels in buds and roots of Arabidopsis thaliana grown under salt stress in the presence or absence of SA187. The value represents the average of three biological experiments, each experiment in three technical replicates (n = 9). Error bars represent SE. The asterisk indicates a statistically significant difference based on the Mann-Whitney U test (***P <0.001) compared with the simulated plant (A). The total fresh weight benefit index of 21-day-old plants treated with SA187, 100 nM L-methionine, 100 nM L-cysteine ​​or 1 mM MgSO4 on 100 mM NaCl. The value represents the average of 3 biological experiments, each experiment repeated 10 times (n = 30). Error bars represent SE. The asterisk indicates a statistically significant difference based on the Mann-Whitney U test compared with the simulated plant (B) (*P <0.05; **P <0.01). Uncolonized (grey) and SA187 colonized (green) 21-day-old wild-type, sultr1, apk1/2, apr2, fry1, sir-1 or amIR-LSUc plants on 1/2MS total fresh weight beneficial index 100 mM sodium chloride . The value represents the average of 3 biological experiments, each experiment repeated 12 times (n = 36). Error bars represent SE. Asterisks indicate statistically significant differences based on Mann-Whitney U test (**P <0.01; ***P <0.001) compared with uninoculated wild-type plants (C). In 0 (control) and 100 mM NaCl (salt stress), the accumulation of superoxide radicals (visible by nitro blue tetrazolium staining) in wild-type and amiR-LSUc mutant plants (D).

It has been reported that an increase in the sulfate concentration in the medium helps plants resist salt stress conditions (32). Therefore, we compared the endogenous SA187 plants under salt stress conditions (control, 1/2 MS 100 mM NaCl, 0 mM MgSO4) with the addition of MgSO4 to the medium (treatment, 1/2 MS 100 mM NaCl, 1 mM sulfuric acid). magnesium). Sulfate addition showed significant beneficial effects on plants similar to SA187 colonization (Figure 3B and SI appendix, Figure S8B). Since our dual transcriptome analysis showed that not only the sulfate uptake was up-regulated in SA187 (12), but also the synthesis of cysteine ​​and methionine was up-regulated. Therefore, we determined the internal cystine in Arabidopsis thaliana colonized by SA187 (12). (Oxidation product of 2 cysteine ​​molecules) and methionine levels. Compared with buds, the cystine and methionine levels in roots planted by SA187 are significantly higher, especially under salt stress conditions (SI appendix, Figure S9). We also tested whether 100 nM L-cysteine ​​or L-methionine has a beneficial effect on plant growth under salt conditions. L-methionine, L-cysteine, and MgSO4 showed similar beneficial effects, the same range as SA187 when planting plants (Figure 3B). The beneficial effects of these compounds have also been observed in plants grown under non-salt conditions (SI Appendix, Figure S8B).

In order to gain a deeper understanding of the changes in Arabidopsis plants induced by endophytes SA187, we generated root and shoot transcriptomes of 21-day-old plants that were grown in 100 mM NaCl in the absence or presence of SA187. Or treated with 100 nM KMBA, compared with plants treated with 100 nM ACC as a positive control (Table 2 and Figure 4 AC). In total, we obtained approximately 40 to 64 million double-ended readings. Pearson's linear regression coefficient, FPKM curve and the distribution of mapped readings show a good correlation between biological replicates: 0.98 ≥ R2> 0.87. By applying the differential expression cutoff value of q=0.05 and the fold change>2, RNA-Seq data showed that SA187 colonization induced the down-regulation of 813 genes and the up-regulation of 89 genes in the buds, while 240 down-regulation and 187 up-regulation in SA187 Regulatory genes were identified in the planted roots (Table 2). The SA187-derived active compound KMBA induced the down-regulation of 910 genes and the up-regulation of 178 genes in shoots, while 343 down-regulated and 85 up-regulated genes were identified in KMBA-treated roots (Table 2). The ethylene precursor ACC induced the down-regulation of 1,684 genes and the up-regulation of 283 genes in shoots, while 968 down-regulated and 322 up-regulated genes were identified in ACC-treated roots (Table 2 and Figure 4C). It is worth noting that DEG is specific to shoot and/or root tissues, and the number of common DEGs between plant tissues is very small (Figure 4E). Despite the overall beneficial growth under salt stress, this observation supports the hypothesis of the tissue-specific role of the beneficial microorganism SA187.

Normalized DEG in Arabidopsis Col-0 (FPKM, q value <0.05)

The effect of SA187 inoculation in Arabidopsis sprouts and roots on 1/2MS 100 mM NaCl, experimental setup (A). Phenotypic evaluation of the beneficial effects of three treatments on the growth of 21-day-old Arabidopsis seedlings under the condition of 1/2MS 100 mM NaCl: SA187 was inoculated and 100 nM KMBA or 100 nM ACC was added to the growth medium. Col-0 = untreated Arabidopsis as a control (B). Volcano map of Arabidopsis root and shoot DEG, on 1/2MS 100 mM NaCl, for each of the three treatments: SA187 (SPB and SP), 100 nM KMBA (SP KMBA and SP) or 100 nM ACC (SP ACC and SP) (C). Using a heat map constructed using sulfur and ethylene signaling Arabidopsis genes, each of the three treatments was statistically significantly up-regulated (red) and down-regulated (blue): SA187, 100 nM KMBA, 100 nM ACC ( D). The Venn diagram shows the common or specific up-regulated and down-regulated genes of Arabidopsis sprouts and roots, comparing the endogenous and free-living status of SA187 (q value <0.05) (E).

Gene ontology (GO) term enrichment analysis showed that the up-regulated gene set in buds is rich in proteins that are functional in jasmonic acid metabolism, while root genes show enrichment in root morphogenesis (SI appendix, figure) S8). Interestingly, the down-regulated genes in shoots and roots showed significant enrichment in defense responses and salicylic acid processes (SI appendix, Figure S8). More refined genes through genetic analysis revealed significant regulation of sulfur metabolism genes in roots and shoots (Figure 4D). In shoots and roots, sulfate transporter (SULTR), ATP sulfurylase (ATPS), APS kinase (APK), APS reductase (APR), sulfate reductase SIR1 and several sulfur-regulated transcription factors (such as SLIM1 , EIL2, EIN3) and EIL1 are differentially regulated compared with non-colonized plants (Figure 4D and SI appendix, S4B). Interestingly, many of these genes are part of sulfur regulators, which not only regulate sulfur metabolism, but are also key factors in glucosinolate biosynthesis and ethylene signaling pathways (33).

We found that SA187 increased the sulfur level in Arabidopsis salt-stressed roots and enhanced the gene expression of many members of the sulfur regulator, indicating that SA187 may directly affect the sulfur absorption or metabolism of Arabidopsis (Figure 4D). To test the role of the sulfate assimilation pathway in the beneficial interaction of SA187 with Arabidopsis, we analyzed many mutants of Arabidopsis sulfur metabolism. As shown in Figure 3C, sultr1;2 (34), apk1 apk2 (35), apr2 (36), sir1 (37) and Fried1 (38) mutants are allergic to salt stress, but can be significantly rescued by SA187 colonization. It should be noted that under normal growth conditions, the growth of all mutants was affected to a certain extent (SI Appendix, Figure S8D), but SA187 only exerted a substantial rescue on the strong growth inhibition phenotype under salt stress Effect (Figure 3C). These results indicate that SA187 is closely related to sulfur metabolism to give Arabidopsis salt stress tolerance.

Most of the sulfur assimilation enzymes are located in the chloroplast, which indicates that a complete chloroplast physiology is required for normal operation. In this regard, we noticed that the Arabidopsis thaliana colonized by SA187 continued to grow under salt stress, and did not show the down-regulation of photosynthesis-related genes, as shown by the genes encoding electron transport carriers in Photosystem I and II, such as low quantum Yield photosystem ii 1 (LQY1), xyloglucan endoglucosylase/hydrolase 6 (XTH6), acetyl-coenzyme A: (z)-3-hexen-1-ol acetyltransferase (CHAT), Ferulic acid cutin (DCF) deficiency and halogen acid dehalogenase-like hydrolase (HAD) superfamily proteins (SI appendix, Figure S4B).

Salt stress induces the production of ROS in chloroplasts, and one of the most significant effects on salt-growing plants is the down-regulation of the entire photosynthetic mechanism (29). To test the possibility of SA187 protecting Arabidopsis from excessive ROS levels, we monitored O2– levels in SA187 colonized plants grown under 0 or 100 mM NaCl. As seen by nitro blue tetrazolium staining, SA187 did not change the level of superoxide radicals under non-stress conditions (Figure 3D). However, the previous results of our group showed that after SA187 colonization, the increase in ROS levels in non-colonized Arabidopsis plants induced by salt stress was significantly reduced (Figure 3D) (12). Since sulfur metabolism is closely related to the protection of ROS through the synthesis of glutathione, one possible explanation for these results may be that plants planted with SA187 have higher redox capacity due to the increased GSH/GSSG ratio. Therefore, we determined the GSH and GSSG levels in the shoots and roots of SA187 colonized and uncolonized plants in the absence and presence of salt stress (SI Appendix, Figure S10). As expected, GSH/GSSG levels in plants were significantly reduced under salt stress (SI appendix, Figure S10 B and E). Compared with uncolonized plants, the shoots and roots of SA187 planted plants showed higher GSH/GSSG ratios under non-stress conditions, but only the roots of SA187 planted plants showed a strongly enhanced GSH/GSSG ratio under salt stress ( SI appendix, Figure S10 B and E). In summary, these results support the view that SA187-induced sulfur metabolism in Arabidopsis enhances glutathione levels to reduce ROS damage.

Recently, it has been reported that members of the low-sulfur up-regulated (LSU) gene family (LSU1-LSU4) play a role in protecting photosynthetic mechanisms under salt stress conditions (39). The LSU family has been shown to directly interact and activate the chloroplast superoxide dismutase FSD2, thus playing a role in regulating the levels of plastid superoxide free radicals (39). To test the potential interaction of SA187 with the LSU network in Arabidopsis, we tested the amiR-LSUc line, which expresses an artificial microRNA (amiRNA) ( 39). As expected, the lsuC line exhibited stronger ROS accumulation under salt stress conditions (Figure 3D), confirming its role in ROS protection. Interestingly, after SA187 was planted, the accumulation of ROS in salt-stressed lsuC plants was completely suppressed (Figure 3D). This result is supported by phenotypic analysis, indicating that the reduced growth of amIR-LSUc plants is strongly supplemented after SA187 planting (Figure 3C), indicating that SA187 inhibits the accumulation of ROS in Arabidopsis plastids induced by salt stress through the LSU pathway.

Many PGPBs have been described as interacting with and having beneficial effects on plant growth and stress tolerance, but understanding the underlying molecular mechanisms is hindered by the inherent complexity of heterogeneous natural systems. In addition, different strains and species use different strategies to help plant growth under different conditions. The mechanism of interaction depends not only on the interacting microorganisms, but also on the species, ecotype, developmental and physiological state of the plant, and non-biological or biological conditions. biology. Environment (4). In addition, the transition from controlled growth to anarchic proliferation can lead to a change from the lifestyle of beneficial bacteria to the lifestyle of pathogenic bacteria (40). Enterobacter SA187 provides salt tolerance for alfalfa crops in open field trials, and also provides salt tolerance for Arabidopsis under controlled laboratory conditions, and can be used as a good model system to understand how beneficial microorganisms induce plants Stress tolerance. Compared with the small (5 to 10%) beneficial effects under non-salt conditions, SA187 increases plant growth to approximately 40 to 50% under salt conditions (12). In addition, it is recommended that the bacterial metabolite KMBA, which is converted to ethylene in plants, promotes plant growth under salt stress. Our current analysis has determined that bacterial and plant sulfur metabolism pathways are the most important in the beneficial interaction of SA187 with Arabidopsis. Our work reveals the importance of coordinated regulation of bacterial and plant sulfur metabolism in plant stress tolerance. In addition, our data provide a mechanism to understand how sulfur metabolism and chloroplast function are related to ethylene signaling in the context of salt stress in Arabidopsis.

Analysis of the metabolic process of SA187 during the endogenous interaction with Arabidopsis thaliana revealed two major sets of genes. On the one hand, we identified a set of genes that are mainly involved in the initial contact and colonization of host plants, such as chemotaxis, bacterial motility, QS and signal transduction, and biofilm formation. The other set of genes is related to the transportation and exchange of nutrients and the metabolism of carbon, sulfur and energy.

The first step in plant root endophyte colonization involves chemotaxis and subsequent use of cellulose-like fibers, colanic acid, and adhesion proteins to attach bacterial cells to the roots (17). Chemotaxis and bacterial motility are usually the opposite of biofilm formation (15, 16, 18), which is controlled by QS, which coordinates gene expression in a cell density-dependent manner (41). In order to successfully colonize plant roots, beneficial bacteria produce cellulose or cellulose-like fibers to form biofilms (13, 15, 18). After SA187 stably colonized Arabidopsis thaliana, it was observed that the bacterial gene expression of cellulose and crimped fibers was reduced, and the factors involved in biofilm formation were up-regulated. The simultaneous up-regulation of bacterial genes involved in flagellar biosynthesis and chemotaxis indicates that even after the endogenous state is established, a heterogeneous population of SA187 still exists. Although some bacterial cells may have been sedentary and form biofilms, other active cells may still spread through growing plant tissues.

In the beneficial plant-microbe interactions, both parties benefit from the association that shows a wide exchange of nutrients, metabolites, and information. However, so far, little is known about how to coordinate the expression of a large number of genes in each partner. During the salt stress period of Arabidopsis plants, the inhibition of growth and development is related to the accumulation of ROS and the down-regulation of photosynthesis and carbohydrate metabolism. It is important that SA187 is planted to maintain photosynthesis and growth of plants under salt stress conditions (12).

In addition to being a component of the essential amino acids cysteine ​​and methionine, sulfur metabolism is also related to the production of various secondary metabolites. Cysteine ​​is a component of glutathione, and methionine is an important precursor of S-adenosylmethionine (SAM). SAM is a methyl donor and precursor for many processes, including sulfur Synthesis of glucoside, polyamine and ethylene. Figure 5). For this reason, the absorption and metabolism of sulfur are strictly controlled. In the roots, sulfur is absorbed in the form of sulfate, which is mainly controlled by the expression of the SULTR transporter. We found that SA187 increased the levels of sulfur and cysteine ​​and methionine in salt-stressed roots of Arabidopsis thaliana (Figure 3A and SI appendix, S8 A and B), indicating that SA187 directly affects the absorption and absorption of sulfur by Arabidopsis thaliana. metabolism. Once the plant roots absorb sulfate, ATPS converts sulfate to APS, then firstly converts to sulfite by APR, and converts to sulfide by sulfite reductase (SIR1), and then generates cysteamine in the following reaction Acid and methionine. To test the role of sulfur metabolism and the beneficial interaction between SA187 and Arabidopsis during salt stress, we analyzed sultr1;2 (34), apk1 apk2 (35), apr2 (36), sir1 (37) and Fried1 (38) ) The behavior of mutant plants in terms of salt stress and SA187. Although the growth of all mutants under non-salt conditions was hardly affected (SI Appendix, Figure S8C), they showed strong growth inhibition under salt stress (Figure 3C), supporting sulfur metabolism in salt stress tolerance An important point of view. Perhaps due to the redundancy of the sulfur transporter and APR genes, SA187 can almost completely rescue the salt growth inhibitory phenotype of sultr1;2 and apr2 mutants. However, mutants of the sulfite reductase SIR1 obtained different results, which is the bottleneck of sulfur metabolism, limiting cysteine, methionine and their derived downstream products (such as glutathione or polyamines) Availability (37). SA187 could hardly alleviate the severe growth inhibition phenotype of sir1 mutant, indicating that a complete sulfur assimilation pathway is necessary for the salt stress tolerance induced by SA187 in Arabidopsis.

SA187-induced sulfur metabolism and ethylene signaling model in Arabidopsis: After absorbing sulfate, it is converted into APS, sulfite, sulfide and cysteine, which can be converted into glutathione (GSH) or methionine acid. GSH can block salt-induced ROS production in plastids. Methionine can be converted into SAM, ACC, and ethylene, the latter regulating sulfur regulators through transcription factors SLIM1, EIL2, and EIN3. The ethylene-regulated sulfur regulator contains most of the genes for sulfur metabolism, thereby forming a feedback loop between the sulfur state and the genes encoding sulfur metabolizing enzymes. The sulfur state is transmitted by retrograde signals induced by PAP levels to regulate nuclear-encoded plastids and sulfur metabolism genes. SULTR = sulfate transporter, ATPS = ATP sulfurylase, APR = APS reductase, APK = APS kinase, PAR = PAPS reductase, SIR1 = sulfite reductase, FRY1 = 5'-diphosphate, nucleotide Phosphatase, OTL = O-acetylserine thiolase, TyrB = aromatic-amino acid transaminase, BCAT4 = methionine aminotransferase, MAT = methionine adenosyltransferase, ACS = ACC synthase, ACO = ACC oxidase, ETR1 = ethylene receptor Body 1, SLIM1 = Sulfur restriction 1 (central transcriptional regulator of plant sulfur response and metabolism), EIN3 = Ethylene-insensitive 3, EIL2 = Ethylene insensitive 3 like 2, CRL = Chloroquine resistance transporter (CRT) like protein , APS = adenosine 5'-phosphate, PAPS = 3'-adenosine phosphate-5'-phosphate, PAP = 3'-polyadenosine 5'-phosphate, AMP = 5'-adenosine monophosphate, KMBA = 2 -Keto-4-methylthiobutyric acid, SAM = S-adenosylmethionine, ACC = 1-aminocyclopropane-1-carboxylic acid, GSH = glutathione, ROS = reactive oxygen species.

APS can also be converted into PAPS by several APS kinases. PAPS is used as a sulfate donor for the enzymatic reaction of various sulfotransferases (SOT) with various substrates (including glucosinolates and plant hormones). This produces the by-product PAP (Figure 5). Unstressed plants contain very low levels of PAP, mainly due to the activity of FRY1, which dephosphorylates PAP to AMP. However, under stress conditions, PAP levels accumulate and act as a retrograde signal to coordinate nuclear chloroplast gene expression during sulfate starvation (42). The central role of FRY1 in linking sulfur metabolism with PAP retrograde signaling inspired us to test whether SA187 mediates salt stress tolerance through the Arabidopsis PAP retrograde signaling pathway. In fact, as previously reported, fry1 mutant plants are allergic to salt stress (38). However, SA187 can significantly inhibit the salt stress phenotype of fry1 plants, indicating that the sulfur monitoring FRY1 pathway is also involved in the beneficial effects of SA187 on Arabidopsis during salt stress (Figure 5).

In our previous work, we discovered that SA187-derived methionine metabolite KMBA plays an important role in mediating ethylene-induced salt stress tolerance in plants (12). The plant ethylene signal is linked to sulfur metabolism through SAM, which is a precursor of ethylene. There is evidence that genes related to sulfur metabolism are also regulated by the ethylene alternative pathway composed of the ethylene receptor ETR1 and the transcription factor SLIM1 (43). SLIM1 is part of the EIN3 transcription factor family, also known as EIL3. Although it is clear that only the SLIM1 homodimer can bind to the promoter of the sulfur regulator, SLIM1/EIL3 can also form homodimers and heterodimers with other members of the EIN3 family to fine-tune sulfur metabolism (44) . Recently, Dietzen et al. (45) revealed the important functions of EIL1 and EIL3 transcription factors in regulating sulfur metabolism under sulfur-limited conditions. Although the transcriptome analysis of our data was compared with the data of Dietzen et al. (45) It was revealed that the DEG of SA187 planted roots and eil1 eil3 plants only 5% overlap, and it was found that SA187 planted branches overlapped 33%. These results need further study, but support the model that EIL transcription factors are involved in mediating some beneficial responses of Arabidopsis sulfur metabolism to SA187 colonization.

The sulfur regulator contains 27 genes, all of which are involved in sulfur metabolism, and most of them are located in the chloroplast (33). Our transcriptome analysis shows that SA187 regulates several genes of sulfur regulators in Arabidopsis, especially ChaC-like family proteins and ATSDI1, which are significant for sulfate starvation in the metatranscriptomics study of Henríquez-Valencia et al. One of the responding genes. (33) (Figure 4D). To test the hypothesis that KMBA produced by bacteria induces ethylene to regulate sulfur regulators, we performed transcriptome analysis on Arabidopsis plants treated with KMBA or ACC under salt stress. The comparison of the transcript levels of sulfur regulators between KMBA and ACC-treated Arabidopsis and SA187 planted plants (Figure 4A-D and SI appendix, Figure S4A) supports the model of SA187 produced by KMBA that regulates the sulfur metabolism of Arabidopsis thaliana in the following ways: ethylene Signal path (Figure 5).

Sulfur metabolism is not only related to protein synthesis and ethylene signal transduction, but also an important precursor of glutathione biosynthesis, thus providing a key factor for ROS detoxification. It is known that salt stress induces ROS accumulation in Arabidopsis and other plants, and is considered to be one of the main harmful factors for plants under these conditions. Therefore, we tested whether SA187 can protect Arabidopsis from the increase in ROS levels under salt stress. In fact, SA187 can reduce the level of ROS induced by salt stress in Arabidopsis. Interestingly, some of the sulfur regulators are the four LSU genes, which encode chloroplast-targeting proteins that directly interact and activate Fe-superoxide dismutase FSD2. Importantly, sa187 can almost completely suppress the hypersensitive salt phenotype of the amiR-LSUc mutant. These data support the role of LSU as a protein center, linking sulfur metabolism with salt stress-induced chloroplast dysfunction (40), and provide a further link between SA187 and Arabidopsis sulfur metabolism during salt stress tolerance.

In summary, our data shows that after SA187's endogenous interaction with Arabidopsis thaliana, the host plant provides a nutrient-rich environment that is conducive to bacterial growth and proliferation. In exchange, SA187 protects its host from the negative effects of salt stress conditions by providing the metabolite KMBA. KMBA can be converted into ethylene or into other compounds such as methionine or cysteine. The ethylene signal in Arabidopsis regulates sulfur regulators, including most sulfur metabolism enzymes, thereby regulating the absorption of sulfate and the production of cysteine ​​and methionine. These cysteine ​​and methionine not only act as The precursor of protein synthesis, but also used to synthesize important antioxidant glutathione or polyamines can help detoxify ROS. SA187 thus protects plants from the negative effects of salt stress and maintains photosynthesis and plant growth. The regulation of sulfur metabolism does not appear to be unique to SA187. Recently, Cheng et al. (46) The APS reductase gene cysH has been determined to be a key factor in the beneficial interaction between Pseudomonas fluorescens and Arabidopsis. They also found that the sulfur metabolism of Arabidopsis thaliana plays an important role in the ISR and growth of Pseudomonas fluorescens. Therefore, the coordination of sulfur metabolism between bacteria and host plants seems to be at the core of protecting plants from abiotic and biotic stress conditions.

Arabidopsis Col-0 was used as the wild type. SALK_053427 (apk1-1) and SALK_093072 (apk2-1) (35), apr2 (36), sir1 (37), fry1 (38) and amIR-LSUc mutant lines sultr1;2 (34), apk1 apk2 double mutant (39) is also used. As described by Saad et al., seedlings of Arabidopsis thaliana grow. (47). In short, the Arabidopsis seeds were surface sterilized in 70% EtOH 0.05% sodium lauryl sulfate on a shaker for 10 minutes, washed twice in 96% EtOH, and then air-dried. The sterilized seeds were sown on half-strength Murashige and Skoog (48) (1/2MS) agar plates (0.9% agar) inoculated with SA187, and stratified in the dark at 4°C for 2 days. After stratification, the seeds germinated vertically for 5 days at 22 °C and 16/8 h light/dark cycle, and the photon flux density was 150 μmol m-2 s-1. Then transfer the uniformly germinated seedlings with a root length of 1 cm to a 1/2MS plate and grow vertically without (non-salt, 1/2MS) or 100 mM NaCl (salt stress, 1/2MS 100 mM NaCl) , And then grow for another 12 d. Transfer six seedlings to each plate. To prepare the SA187 inoculum, the overnight bacterial culture was collected in Luria-Bertani (LB) broth (Sigma), centrifuged at 3,000 rpm for 15 minutes, washed twice in 1/2MS liquid, and in 1/2MS Resuspend to the final optical density (OD600) = 0.2 (~107 colony forming units [CFU]). A plate containing 50 mL of cooled 1/2MS or 1/2MS 100 mM NaCl was inoculated with 0.1 mL of bacterial suspension (~107 CFU) and allowed to stand to solidify. At the end of the experiment, the complete 17-day-old seedlings were collected and immediately immersed in liquid nitrogen, and then stored at -80°C until needed.

Salt stress tolerance determination and different ethylene precursor treatments ACC (100 nM) and KMBA (100 nM) were performed as described by de Zélicourt et al. (12) There are some modifications. In short, transfer 5-day-old colonized seedlings to 1/2MS plates with or without 100 mM NaCl (Sigma). After scanning the plate, use ImageJ software to measure the length of primary roots every 2 days. Lateral root density is evaluated as the number of lateral roots detectable under a stereo microscope divided by the length of primary roots. The fresh weights of shoots and roots were measured 17 days after seedlings were transplanted. Measure the dry weight after drying the shoots and roots at 70°C for 2 days.

Used to prepare pure bacterial samples, Enterobacter spp. SA187 was grown overnight in LB broth until it reached the exponential growth phase. Collect a certain volume of this pre-culture, centrifuge, wash the pelleted cells twice with 1/2MS, and resuspend to a final OD600 = 0.2 (~107 CFU). A total of 500 μL of this bacterial suspension was inoculated into 50 mL of 1/2MS or 1/2MS 100 mM NaCl, and the bacterial culture was incubated at 28°C in the dark for 4 hours. To evaluate the role of sugar in bacterial metabolism, 1% sucrose was added to the broth. After incubation, harvest 20 mL of bacterial culture and store the cell pellet (~1.4 × 1010 CFU) at -80 °C until needed.

The sulfur content of 21-day-old shoots and root dry samples was measured by digestion. 2 mL of freshly prepared 1% HNO3 (trace metal grade, Fisher Scientific) was used and added to pre-weighed shoots and root samples. After digestion, transfer 1 mL of sample to a volumetric test tube and dilute with 1% HNO3 to a final volume of 50 mL. Use ICP-OES (Agilent 5110, Agilent Technologies) to determine sulfur concentration under the following conditions: 1.2 KW RF power, 0.7 L/min atomizer flow, 12.0 L/min plasma flow, 1.0 L/min auxiliary flow, argon , 8 mm viewing height and axial viewing mode.

In order to determine the gene expression in the freely viable SA187 ("B", "SB"), the RiboPure Bacteria kit (Ambion) was used to make the following modifications according to the manufacturer's instructions, and the total RNA of SA187 cultured in liquid medium was extracted. In the cell lysis step, without using beads, a PowerLyzer 24 homogenizer (Mobio) is used to break the sample. The settings are as follows: three cycles, 30 seconds "on", 30 seconds "off". In order to capture both bacterial and plant transcripts, Arabidopsis seedlings ("PB", "SPB") and individual shoots and roots ("SP", "SPB, ""SP ACC", "SP ACC", "SP KMBA") by using the Nucleospin RNA Plant Kit (Macherey-Nagel) as described by de Zélicourt et al. (12). Use Qubit 2.0 fluorometer and RNA BR detection kit (Invitrogen) to evaluate RNA concentration, and use 2100 bioanalyzer and RNA 6000 Nano detection (Agilent) to verify the integrity of total RNA.

Use 1 µg RNA from a bacterial sample and 1 µg RNA from a double sample (plant bacteria) as starting materials. Use Ribo-Zero Magnetic Kit (Bacteria) (Epicentre/Illumina) to remove ribosomal RNA (rRNA) from total RNA from pure bacterial samples. For the removal of rRNA from plant bacterial samples, use a 1:1 mixture of Ribo-Zero Magnetic Kit (bacteria) and Ribo-Zero Magnetic Kit (plant leaves). After removing rRNA, use TruSeq Stranded Total RNA LT Kit (Illumina) to prepare RNA-Seq library. The final complementary DNA (cDNA) library was sequenced using HiSeq2500 on the core laboratory biological science platform (King Abdullah University of Science and Technology [KAUST], Saudi Arabia). Use three biological replicates of each sample.

The strand-specific pair-end sequencing of RNA-Seq samples was performed using Illumina HiSeq2500, with a read length of 101 base pairs (bp). By using FASTQC (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/) to perform quality control on the readings, readings with a FASTQC quality score> 30 are considered for further analysis. By using Trimmomatic (49) to apply the following parameters to remove low-quality sequences and linker sequences, and additionally trim the 5'- and 3'-ends of bacterial-derived reads: the minimum length is 36 bp, and the average Phred quality score (Q) is greater than 30, Before and after base removal, the base quality is lower than 3, and the sliding window is 4:15. Then use TopHat (version 2.0.9) to map the clean reads to the reference genome (50). The readings from the "B" and "SB" samples are mapped to Enterobacter. The SA187 genome (CP019113) was downloaded from the internal INDIGO data warehouse (51), and the reads from the "PB" and "SPB" samples were mapped to the tandem sequence constructed from SA187 and the Arabidopsis genome. The Arabidopsis TAIR10 genome is downloaded from TAIR (https://www.arabidopsis.org). In this study, only the reads mapped to the bacterial genome in the two samples were considered. Use Cufflinks version 2.2.0 (52) for transcription assembly, quantification and differential expression analysis. To overcome the difference in sequence coverage between cultured bacteria and endogenous samples (SI appendix, Table S1), library size standardization was applied before DEG analysis. Analyze the sequencing saturation to evaluate whether the sequencing depth is sufficient to ensure the quality of the RNA-Seq experiment. Saturation curve is generated by using R package RNA-SeQC (53). The reproducibility of the three biological replicates was evaluated by calculating the Pearson correlation of normalized gene expression per million mapped reads per kilobase reads. DEG analysis is performed by averaging the FPKM of all three biological replicates for each sample. CummeRbund (version 2.0.0) is used to visualize DEG (53). To identify DEG, a threshold value of q ≤ 0.05 is used. In addition, if 0.5>fold change>2, the gene is considered to be regulated.

In the presence or absence of SA187, use Nucleospin RNA plant kit (Macherey-Nagel) under 100 mM NaCl with KMBA or ACC treatment, as described by de Zélicourt et al., from 21-day-old plant tissue buds and roots Extract the total RNA from it. (12). Use Illumina deep sequencing to sequence RNA samples. Three biological replicates were sequenced for each condition. The samples were sequenced using HiSeq4000, and the read length was 150 bps. Use FASTQC (54) to check the quality of the sequencing reads. Trimmomatic version 0.36 (49) is used to filter low-quality readings and bps with the following parameters: minimum length 36 bp; average Phred quality score greater than 30; removing leading and trailing bases with base quality lower than 3; 4:15 sliding window. Use TopHat version 2.0.9 (56) to align the trimmed reads with the TAIR10 (55) genome. Then use FeatureCounts version 1.6.5 (57) to calculate the number of reads for each gene using aligned reads, and use Cufflinks version 2.2.1 (58) to calculate FPKM. Use Cuffdiff in the Cufflinks 2.2.1 (58) package to calculate the differential expression between two samples. To identify DEG, specific parameters (q value 0.05 and statistical correction: Benjamini Hochberg) are used. Genes showing 0.5> fold change> 2 and q value ≤ 0.05 to the simulant are considered to be regulated. Use AgriGO2 (59) to annotate regulatory gene functions. Use Venny2 (60) to complete common genes that are up-regulated/down-regulated between different comparisons.

To verify the results from bacterial single and double RNA-Seq experiments, 50 DEGs were selected and their gene expression was analyzed by qRT-PCR using specific primers (SI Appendix, Table S2). Use the Primer-Blast online tool provided by the National Center for Biotechnology Information (NCBI) (https://www.ncbi.nlm.nih.gov/tools/primer-blast/) to design primers. The translation initiation factor IF-2 (infB) was used as a reference gene (12). The cDNA used as a template for qRT-PCR was synthesized using SuperScriptIII (Invitrogen) and oligo-dT according to the manufacturer's recommendations. The qRT-PCR reaction was carried out in the CFX96 Touch real-time PCR detection system (BIO-RAD) as follows: 95 °C for 10 minutes; 40 × (95 °C for 10 seconds, 60 °C for 40 seconds), and then proceed to the dissociation step ( Melting curve) to verify the PCR product. The reaction was performed in three technical replicates of each biological replicate, and the relative gene expression was calculated using the ΔΔCT method.

For the qRT-PCR analysis of plant tissue roots and shoots, SuperScriptIII (Invitrogen) and oligo-dT were used to synthesize cDNA according to the manufacturer's recommendations. The qRT-PCR reaction was carried out in the CFX384 real-time PCR detection system (BIO-RAD) as follows: 50 °C for 2 minutes; 95 °C for 10 minutes; 39× (95 °C for 10 seconds, 60 °C for 40 seconds), and then the solution Isolation step (melting curve) to verify the PCR product. The reaction is performed in three technical replicates of each biological replicate. The reference genes used in this analysis were UBIQUITIN10 (At4g05320) and ACTIN2 (At3g18780) (33), and the relative gene expression was calculated using Bio-Rad CFX manager software.

RNA-Seq data can be obtained under NCBI/Gene Expression Synthesis (GEO) data set (https://www.ncbi.nlm.nih.gov/gds) under the accession number. GSE124591 (free survival SA187), GSE102950 (Arabidopsis related SA187) (12), GSE133175 (Arabidopsis tissue: bud/root related SA187) and GSE145884 (Arabidopsis tissue: bud/root treated with ACC or KMBA). All other research data is included in the article and/or supporting information.

This work is part of the DARWIN21 project (http://www.darwin21.org) exploring desert microbes. We thank Stan Kopriva (University of Cologne) for providing sultr1;2, apk1, apk2 and apr2 seeds, and Herve Vaucheret (France National Institute of Agriculture, Food and Environment (INRAE)-Versailles) for providing Fried1-1 and Fried1-4 seeds, Ruediger Hell (Heidelberg University) provided sir1-1 seeds, Pascal Falter-Braun (Munich Ludwig Maximilian University) provided amiR-LSUa-c seeds, and all members of the HH laboratory, including Abdul Aziz Eida, Nathalia Rodriguez, Sarah Alghamdi and Naganand Rayapuram provided technical assistance and help in their work. This publication is based on work supported by KAUST, HH No. BAS/1/1062-01-01.

↵1C.A.-B., HA and RJ made equal contributions to this work.

Author contributions: Research on AdZ, MMS and HH design; CA-B., HA, RJ, AdZ, AB, OA, KA, AR, KS and MA-T. Conduct research; CA-B., HA, and KGM analyze data; CA-B., HA, MMS, and HH wrote this paper; AdZ, MMS, and HH supervised and discussed the results with the authors.

The author declares no competing interests.

This article is directly contributed by PNAS.

This article contains online support information at https://www.pnas.org/lookup/suppl/doi:10.1073/pnas.2107417118/-/DCSupplemental.

This open access article is distributed under Creative Commons Attribution 4.0 (CC BY).

Thank you for your interest in advertising on PNAS.

Note: We only ask you to provide your email address so that the people you recommend the page to know that you want them to see it, and that it is not spam. We do not capture any email addresses.

Feedback privacy/legal

Copyright © 2021 National Academy of Sciences. Online ISSN 1091-6490. PNAS is a partner of CHORUS, COPE, CrossRef, ORCID and Research4Life.