AlgU mediates hyperosmotic tolerance in Pseudomonas protegens SN15-2 by regulating membrane stability, ROS scavenging, and osmolyte synthesis

ABSTRACT

Pseudomonas protegens can serve as an agricultural biocontrol agent. P. protegens often encounters hyperosmotic stress during industrial production and field application. The ability of P. protegens to withstand hyperosmotic stress is important for its application as a biocontrol agent. AlgU is a global regulator responsible for stress response and biocontrol ability. However, the specific regulatory role of AlgU in the hyperosmotic adaptation of P. protegens is poorly understood. In this study, we found that the AlgU mutation disrupted the hyperosmotic tolerance of P. protegens. Many genes and metabolites related to cell envelope formation were significantly downregulated in ΔalgU compared with that in the wild-type (WT) strain under hyperosmotic conditions, and we found that the algU mutation caused membrane integrity to be compromised and increased membrane permeability. Further experiments revealed that the cell envelope integrity protein TolA, which is regulated by AlgU, contributes to cell membrane stability and osmotic tolerance in P. protegens. In addition, several genes related to oxidative stress response were significantly downregulated in ΔalgU, and higher levels of intracellular reactive oxygen species were found in ΔalgU. Furthermore, we found that the synthesis of N-acetyl glutaminyl glutamine amide is directly regulated by AlgU and contributes to the hyperosmotic adaptation of P. protegens. This study revealed the mechanisms of AlgU’s participation in osmotic tolerance in P. protegens, and it provides potential molecular targets for research on the hyperosmotic adaptation of P. protegens.

IMPORTANCE

In this study, we found that the extracytoplasmic function sigma factor AlgU is essential for the survival of P. protegens under hyperosmotic conditions. We provided evidence supporting the roles of AlgU in influencing cell membrane stability, intracellular reactive oxygen species (ROS) accumulation, and dipeptide N-acetylglutaminylglutamine amide (NAGGN) synthesis in P. protegens under hyperosmotic conditions. Our findings revealed the mechanisms of AlgU’s participation in hyperosmotic stress tolerance in P. protegens, and they provide potential molecular targets for research on the hyperosmotic adaptation of P. protegens, which is of value in improving the biocontrol ability of P. protegens.

INTRODUCTION

Pseudomonas protegens is an important plant growth-promoting microorganism, and it also has the ability to inhibit plant pathogens (1). P. protegens SN15-2 isolated from the tomato rhizosphere has enormous potential in biological control, as it can effectively inhibit the growth of the pathogen Ralstonia solanacearum (2). Pseudomonas commonly encounters adverse abiotic stresses, such as hyperosmotic, drought, and heat stresses, during industrial production and field application (3). In the fermentation process, hyperosmotic stress seriously affects the growth and metabolism of microorganisms, thus reducing biological manufacturing efficiency. In addition, high salt concentration in the field can drastically decrease the population of microorganisms (4). Thus, osmotic stress adaptation by P. protegens SN15-2 is important for its application as biocontrol agent.

Hyperosmotic stress tolerance involves various genes and mechanisms. Microbial cells regulate gene transcription and protein activity through signal transduction pathways to cope with high osmotic stress. In a previous study, it was found that RpoC influences the osmotic tolerance of Escherichia coli by increasing the production of proline and cell membrane integrity (5). The catabolite control protein CcpA participates in the osmotic stress response of Lactiplantibacillus plantarum by regulating glycine betaine and iron uptake (6). SigB controls the ectoine transporter in Virgibacillus pantothenticus to increase hyperosmotic tolerance (7). The extracytoplasmic function sigma factor AlgU is one of the important global regulators in Pseudomonas (8). AlgU is involved in tolerance to adverse stresses in Pseudomonas aeruginosa and Pseudomonas syringae (9, 10). We previously found that an AlgU deletion mutant decreased the tolerance to NaCl stimulation, which suggests that AlgU controls osmotic stress adaptation in P. protegens SN15-2 (11). However, the specific mechanism of AlgU’s involvement in hyperosmotic stress adaptation in P. protegens needs further exploration.

In this study, we further elucidated the regulatory roles of AlgU in response to osmotic stress in P. protegens. Importantly, we found that AlgU contributes to maintaining the membrane stability of P. protegens under hyperosmotic conditions. A mutant of the AlgU-regulated cell envelope protein TolA exhibited impaired growth under osmotic stress compared with the wild-type (WT) strain. In addition, we found that AlgU is important in reducing the intracellular reactive oxygen species accumulation in P. protegens under hyperosmotic conditions. Interestingly, the genes for N-acetylglutaminylglutamine amide (NAGGN) synthesis were significantly downregulated in ΔalgU compared with that in the wild-type strain under osmotic stress. The elimination of the NAGGN synthesis pathway impaired the growth of P. protegens under hyperosmotic conditions.

RESULTS

algU mutation influences the growth of P. protegens under hyperosmotic conditions

Strains of wild-type P. protegens SN15-2, the algU deletion mutant ΔalgU, and the ΔalgU complement strain ΔalgU-C were cultured in NB medium with or without 400 mM NaCl. The growth curves of ΔalgU and ΔalgU-C were compared with the growth curve of the WT. In NB medium, the growth of ΔalgU was similar to that of the wild-type strain (Fig. 1A). However, in NB medium with 400 mM NaCl, the growth of ΔalgU was slower than that of the WT, whereas the growth of ΔalgU-C was similar to that of the WT (Fig. 1C). In addition, the viability of the strains was measured via counting colony-forming units (CFU) after 24 hours of cultivation. The results show (Fig. 1B) no significant difference between the ΔalgU and WT strains in terms of final biomass after 24 hours of incubation in NB medium. In contrast, the biomass of ΔalgU was reduced by 42.9% compared with that of the WT strain under hyperosmotic conditions (Fig. 1D). These results strongly suggest that AlgU is necessary for the adaptation of P. protegens SN15-2 to hyperosmotic stress.

Fig 1.

AlgU regulates the tolerance of P. protegens SN15-2 to hyperosmotic stress. Growth curves of WT, ΔalgU, and ΔalgU-C in NB medium (A) and NB medium with 400 mM NaCl (C) at 30°C. OD_600nm was measured with a UV-visible spectroscope every 4 hours. Biomass of WT, ΔalgU, and ΔalgU-C after 24 hours of incubation in NB medium (B) and NB medium with 400 mM NaCl (D) at 30°C. Experiments were performed in biological and technical triplicates. Error bars indicate the standard deviations (n = 3). Statistical significance was calculated using t-tests. ***, P < 0.001.

Transcriptome analysis of the ΔalgU mutant strain under hyperosmotic stress

To further investigate the mechanism of AlgU’s participation in the adaptation of P. protegens to hyperosmotic stress, RNA sequencing (RNA-seq) was conducted to compare gene expression in the ΔalgU and WT strains cultured in NB medium containing 400 mM NaCl. RNA-seq identified a total of 788 differentially expressed genes (DEGs) in the ΔalgU strain compared with in the WT strain under hyperosmotic conditions (Fig. 2A). Among the genes identified, 549 genes were downregulated, and 239 genes were upregulated. The DEGs were classified using Gene Ontology (GO) functional annotation, and the results are presented in Fig. 2B. The upregulated genes were mainly involved in integral components of the membrane, plasma membrane, ATP binding, hydrolase activity, metal ion binding, and transmembrane transport, whereas the downregulated genes were mainly involved in integral components of the membrane and plasma membrane, metal ion binding, oxidoreductase activity, ATP binding, hydrolase activity, and DNA binding. To further investigate the global transcriptional profile of the AlgU-induced osmotic stress response, a Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of the DEGs was also performed. The top 20 KEGG enrichment pathways of the upregulated and downregulated genes in the mutant strain are shown in Fig. 2C and D, respectively. Few of the pathways were significantly enriched since we used a strict adjust-p threshold. Under hyperosmotic conditions, the upregulated genes in ΔalgU were significantly enriched in ABC transporters, the citrate cycle (tricarboxylic acid cycle), quorum sensing, carbon fixation pathways in prokaryotes, and pyruvate metabolism. Meanwhile, downregulated genes were significantly enriched in starch and sucrose metabolism, benzoate degradation, and glyoxylate and dicarboxylate metabolism. Collectively, the GO functional annotation analysis and KEGG pathway enrichment analysis of these DEGs suggest that AlgU regulates multiple cell functions in P. protegens SN15-2 under hyperosmotic culture conditions. Details of the DEGs are presented in Table S2. Additionally, 16 DEGs were randomly selected for qRT-PCR to verify the RNA-seq data. The results of the expression levels determined via qRT-PCR were consistent with the RNA-seq data (Fig. S1).

Fig 2.

Comparative transcriptome profile of ΔalgU and wild-type P. protegens under hyperosmotic conditions. (A) Volcano map of DEGs between ΔalgU and WT under hyperosmotic conditions. Red dots represent upregulated genes, and blue dots represent downregulated genes. The DEGs were defined based on the criteria: fold change >2 and adjusted P value < 0.05. (B) GO functional annotation analysis of DEGs. The ordinate represents the GO term, and the abscissa is the number of differential genes in the term. (C) KEGG pathway enrichment analysis of upregulated genes. (D) KEGG pathway enrichment analysis of downregulated genes. KEGG enrichment analysis was performed with “KOBAS” using Fisher’s exact test, and Benjamini-Hochberg was used for multiple testing corrections. Adjusted P value < 0.05 represents the KEGG pathway significantly enriched. The rich factor represents the ratio of DEG numbers annotated in this pathway term to all gene numbers annotated in this pathway term. The size and color of the dot represent the number of DEGs and the adjusted P value, respectively. WH, WT under hyperosmotic conditions; UH, ΔalgU under hyperosmotic conditions.

Metabolomic analysis of the algU mutant strain under hyperosmotic stress

Non-targeted metabolomic experiments using liquid chromatography mass spectrometry (LC-MS) were conducted to determine the role of AlgU in the adaptation of P. protegens SN15-2 to hyperosmotic stress. A principal component analysis (PCA) showed that the ΔalgU and WT groups were separated from each other (Fig. 3A), indicating an obvious effect of AlgU on the levels of intracellular metabolites in P. protegens under hyperosmotic culture conditions. In this study, a total of 248 significant differentially accumulated metabolites (DAMs), comprising 177 upregulated metabolites and 71 downregulated metabolites, were identified via orthogonal partial least squares discriminate analysis (OPLS-DA) with the criteria of variable importance in the projection (VIP) >1 and P < 0.05 (Fig. 3B). Metabolites were identified and assigned to putative molecular classes using the Human Metabolome Database (HMDB). The upregulated metabolites were mainly involved in organic acids and derivatives; organoheterocyclic compounds; nucleosides, nucleotides, and analogs; lipids and lipid-like molecules; and organic oxygen compounds (Fig. 3C), whereas the downregulated metabolites were mainly involved in organic acids and derivatives, lipids and lipid-like molecules, organoheterocyclic compounds, and organic oxygen compounds (Fig. 3D). Details of the DAMs are presented in Table S3. The results suggest that AlgU is important for responding to hyperosmotic stress in P. protegens SN15-2, as it affects the accumulation of many types of metabolites. The differential accumulation of a large number of lipids and lipid-like molecules in ΔalgU suggests that AlgU might regulate the cell membrane’s lipid composition under hyperosmotic stress conditions.

Fig 3.

Comparative metabolomic profile of ΔalgU and wild-type P. protegens under hyperosmotic conditions. (A) PCA plots of metabolites in positive (POS) and negative (NEG) ion modes between ΔalgU and WT. (B) Volcano map of DAMs between ΔalgU and WT. Red dots represent upaccumulated metabolites, and blue dots represent downaccumulated metabolites. The DAMs were defined using the criteria: VIP values (from the OPLS-DA model) >1 and P value (from t-test) <0.05. The size of the points indicates the VIP value. (C) HMDB classification annotation analysis of upregulated DAMs. (D) HMDB classification annotation analysis of downregulated DAMs. The names of the HMDB superclass levels and the percentage of metabolites in order of the number of metabolites were shown.

Mutation in algU compromises the membrane stability of P. protegens in a hyperosmotic environment

Plasma membrane composition can influence the capacity of cells to survive hyperosmotic stress. The extracytoplasmic sigma factor AlgU and its homolog RpoE have been reported to be involved in the envelope stress response via membrane remodeling (12, 13). In this study, the membrane integrity and membrane permeability of ΔalgU were examined and compared with those of the WT strain (Fig. 4C and D). We did not detect any differences in propidium iodide (PI) or 1-N-phenylnaphthylamine (NPN) uptake between the ΔalgU strain and the WT strain under normal culture conditions. However, under hyperosmotic culture conditions, ΔalgU significantly increased PI uptake, suggesting that AlgU plays an important role in the maintenance of the membrane integrity of P. protegens under hyperosmotic conditions. In addition, the NPN uptake factor of ΔalgU was higher than that of the WT under hyperosmotic conditions, implying that the mutation in algU led to an increase in membrane permeability. Also, the transcriptome data identified many differentially expressed genes related to membrane homeostasis (Fig. 4A); this may be the mechanism by which AlgU affects the membrane integrity and permeability under hyperosmotic conditions. The RNA-seq data revealed that several genes related to lipid biosynthesis and metabolism were downregulated by at least 2-fold in ΔalgU compared with that in the WT, including fatty acid synthesis-related genes (fabG and CFA), a glycerolipid metabolism-related gene (glpK), and glycerophospholipid metabolism-related genes (pcs and clsB). Peptidoglycan is a major constituent of the bacterial cell wall, and it endows strength and rigidity to the cell wall (14). Several genes related to peptidoglycan biosynthesis were significantly downregulated in ΔalgU compared with that in the WT, such as pgpG, lysM, and ycbB. mdoH and opgG, related to osmoregulated periplasmic glucans synthesis, were also significantly downregulated in ΔalgU. Additional genes related to LPS biosynthesis, outer membrane proteins, and lipoproteins were observed to be downregulated in ΔalgU. A set of downregulated genes involved in contributing to cell envelope integrity (tolA, creD, and tolC) were also observed in the comparative transcriptomic data. In the metabolomic data, several phosphatidylserines (PSs), phosphatidylethanolamines (PEs), and glycerophosphoinositols, which play central roles in cell membrane structure, were downregulated in ΔalgU compared with that in the WT (Fig. 4B). Correspondingly, the precursors of glycerophospholipids (glycerol-3-phosphate and glycerol-2-phosphate) were upregulated in ΔalgU compared with that in the WT. Overall, AlgU could regulate the cell membrane components and membrane stability of P. protegens under hyperosmotic stress conditions.

Fig 4.

AlgU influences the membrane stability of P. protegens in hyperosmotic conditions. (A) Differentially expressed genes relevant to cell membrane composition in ΔalgU compared with WT under hyperosmotic conditions. The Log₂FC (ΔalgU vs WT) represents log2-transformed fold change value of gene expression between strains ΔalgU and wild-type P. protegens under hyperosmotic conditions. (B) Differentially accumulated metabolites relevant to cell membrane composition in ΔalgU compared with WT under hyperosmotic conditions. VIP represents variable importance in the projection obtained by the OPLS-DA model. The dendrogram of metabolites clustering is displayed on the left side. Each row represents a metabolite, and the color represents the relative expression amount of the metabolite in the group of samples. The VIP bar graph of metabolites is displayed on the right side. The color of the bar indicates the significant difference in metabolites between the two groups. The darker the color, the greater the log₁₀(P-value). (C) Uptake of propidium iodide (PI) as an indicator of cell membrane integrity in WT, ΔalgU, and ΔalgU-C after 24 hours of cultivation in NB and NB with 400 mM NaCl. (D) Uptake of 1-N-phenylnaphthylamine (NPN) as an indicator of cell membrane permeability in WT, ΔalgU, and ΔalgU-C after 24 hours of cultivation in NB and NB with 400 mM NaCl. Experiments were performed in biological and technical triplicates. Error bars indicate the standard deviations (n = 3). Statistical significance was calculated using t-tests. **, P < 0.01; ***, P < 0.001.

The cell envelope integrity protein TolA contributes to osmotic tolerance in P. protegens

Our RNA-seq data indicated that tolA was significantly downregulated in ΔalgU under hyperosmotic stress. Previous studies have shown that TolA is involved in maintaining the integrity of the outer membrane (15). In this study, we constructed a tolA deletion mutant to further investigate the role of TolA in the hyperosmotic adaptation of P. protegens SN15-2. Growth and viability analyses were conducted for ΔtolA in a hyperosmotic environment. The result shows that the growth of ΔtolA was slower than that of the WT (Fig. 5A). The living bacterial count of ΔtolA was decreased by 29.6% compared with that of the WT after 24 h of cultivation (Fig. 5B), which means that tolA is important for the hyperosmotic adaptation of P. protegens. The membrane integrity and permeability of ΔtolA were also assayed under hyperosmotic culture conditions by PI and NPN staining. The uptake of PI by ΔtolA was significantly higher than that by the WT (Fig. 5C), which means that the tolA mutation decreased the membrane integrity. The uptake of NPN by ΔtolA was significantly higher than that by the WT (Fig. 5D), which means that the tolA mutation increased the membrane permeability. Collectively, these results indicate that tolA, which is regulated by AlgU, participates in maintaining outer membrane stability and tolerance to hyperosmotic stress in P. protegens.

Fig 5.

Cell envelope integrity protein TolA influences the tolerance of P. protegens to hyperosmotic stress. (A) Growth curves of WT, ΔtolA, and ΔtolA-C in NB medium with 400 mM NaCl at 30°C. OD_600nm was measured with a UV-visible spectroscope every 4 hours. (B) Biomass of WT, ΔtolA, and ΔtolA-C after 24 h of incubation in NB medium with 400 mM NaCl at 30°C. (C) Uptake of propidium iodide (PI) as an indicator of cell membrane integrity in WT, ΔtolA, and ΔtolA-C after 24 hours of cultivation in NB with 400 mM NaCl. (D) Uptake of 1-N-phenylnaphthylamine (NPN) as an indicator of cell membrane permeability in WT, ΔtolA, and ΔtolA-C after 24 hours of cultivation in NB with 400 mM NaCl. Experiments were performed in biological and technical triplicates. Error bars indicate the standard deviations (n = 3). Statistical significance was calculated using t-tests. ***, P < 0.001; ****, P < 0.0001.

Mutation in algU increases the reactive oxygen species in P. protegens under hyperosmotic conditions

Reactive oxygen species that accumulate during osmotic stress damage cells (16). The RNA-seq data revealed that several genes related to ROS scavenging and oxidative damage repair were significantly downregulated in ΔalgU compared with that in the wild-type strain under hyperosmotic conditions (Fig. 6A), including OsmC family peroxiredoxin-coding genes, the catalase-coding genes katE and katN, the peroxidase-coding gene efeB, ferritin-coding genes, and glutathione S-transferase-coding genes. katE was downregulated by 71.4-fold in ΔalgU compared with that in the wild-type strain. It has been reported that bacterioferritin can protect DNA under hyperthermal and oxidative stresses (17). The bacterioferritin Bfr-coding gene bfr was downregulated by 31.0-fold in ΔalgU compared with that in the WT in this study. Accordingly, we speculated that ROS might accumulate in ΔalgU in large amounts. The intracellular ROS levels of ΔalgU and WT were thus assayed via 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) staining (18). The staining results are shown in Fig. 6B. Under normal culture conditions, no difference in the ROS levels was observed between ΔalgU and the WT. However, under a hyperosmotic culture environment, the ROS levels of ΔalgU were significantly higher than those of the WT. These results suggest that the deletion of algU may lead to oxidative damage in P. protegens under hyperosmotic conditions.

Fig 6.

Deletion of algU increases the reactive oxygen species (ROS) in P. protegens under hyperosmotic conditions. (A) Differentially expressed genes relevant to ROS scavenging and oxidative damage prevention in ΔalgU compared with WT under hyperosmotic conditions. The Log₂FC (ΔalgU vs WT) represents log2-transformed fold change value of gene expression between strains ΔalgU and wild-type P. protegens under hyperosmotic conditions. (B) DCFH-DA assay detecting intracellular ROS levels of WT, ΔalgU, and ΔalgU-C after 24 hours of cultivation in NB and NB with 400 mM NaCl. Experiments were performed in biological and technical triplicates. Error bars indicate the standard deviations (n = 3). Statistical significance was calculated using t-tests. ****, P < 0.0001.

Synthesis of the dipeptide N-acetylglutaminylglutamine amide contributes to hyperosmotic tolerance in P. protegens

NAGGN, which is synthesized by the highly conserved asnO-ngg cluster, acts as an osmoprotectant (19). In a previous study, it was found that in Rhizobium meliloti, the acetyltransferase Ngg was responsible for the synthesis of the intermediate N-acetylglutaminylglutamine (NAGG) from glutamine, and the glutamine amidotransferase AsnO contributed to the conversion of NAGG to NAGGN (20). In this study, we identified genes in P. protegens that are homologous to the asnO-ngg cluster in Rhizobium meliloti (Fig. S2A). The RNA-seq data showed that the expression of ngg and asnO was significantly downregulated in ΔalgU compared with that in the WT strain under hyperosmotic conditions (Fig. S2C). This suggests that AlgU confers osmotolerance, in part, by regulating the synthesis of NAGGN in P. protegens. Cotranscription of ngg and asnO in P. protegens was confirmed via RT-PCR (Fig. S2B). Consistent with the transcriptome results, the RT-qPCR results indicate that ngg and asnO were significantly downregulated in ΔalgU compared with that in the WT under hyperosmotic conditions (Fig. S2D and E). The asnO-ngg operon was thus deleted to explore the role of NAGGN in hyperosmotic adaptation in P. protegens. The growth of ΔasnO ngg and the WT strain in NB medium with or without 400 mM NaCl was examined. Considering that NAGGN is thought to be synthesized from glutamine (Gln) (20), 10 mM Gln was added to the medium to investigate the role of Gln in NAGGN accumulation and hyperosmotic adaptation in P. protegens. Under normal culture conditions, the growth of ΔasnO ngg was similar to that of the WT strain (Fig. 7A and B), and the addition of Gln had no effect on the growth of all strains in the absence of hyperosmotic stress (Fig. 8A and B). Under hyperosmotic culture conditions, the growth of ΔasnO ngg was more affected than that of the WT strain (Fig. 7C and D). The growth rate of ΔasnO ngg was significantly slower than that of the WT strain. The number of viable bacteria of ΔasnO ngg was also significantly lower than that of the WT strain after 24 hours of culture. The addition of Gln improved the growth of the WT strain and the complementation strain ΔasnO ngg-C under hyperosmotic conditions, whereas it had no effect on the growth of ΔasnO ngg (Fig. 8C and D).

Fig 7.

The asnO-ngg gene cluster influenced the tolerance of P. protegens SN15-2 to hyperosmotic stress. Growth curves of WT, ΔasnO ngg, and ΔasnO ngg-C in NB medium (A) and NB medium with 400 mM NaCl (C) at 30°C, and OD_600nm was measured with a UV-visible spectroscope every 4 hours. Biomass of WT, ΔasnO ngg, and ΔasnO ngg-C after 24 h of incubation in NB medium (B) and NB medium with 400 mM NaCl (D) at 30°C. Experiments were performed in biological and technical triplicates. Error bars indicate the standard deviations (n = 3). Statistical significance was calculated using t-tests. **, P < 0.01.

Fig 8.

Role of glutamine (Gln) in hyperosmotic adaptation of P. protegens. 0 mM or 10 mM Gln was added to the NB medium and NB medium with 400 mM NaCl. After 24 h of culture in NB medium, OD_600nm (A) and living bacteria count (B) were recorded for WT, ΔasnO ngg, ΔasnO ngg-C, ΔalgU, and ΔalgU-C. After 24 h of culture in NB medium with 400 mM NaCl, OD_600nm (C) and living bacteria count (D) were recorded for WT, ΔasnO ngg, ΔasnO ngg-C, ΔalgU, and ΔalgU-C. Experiments were performed in biological and technical triplicates. Error bars indicate the standard deviations (n = 3). Statistical significance was calculated using t-tests. ns, no significance; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

We then determined the NAGGN concentration of the WT strain and derivative mutant strains cultured under hyperosmotic conditions with or without 10 mM Gln (Fig. 9). The deletion of the asnO-ngg operon indeed prevented the synthesis of NAGGN. NAGGN was undetectable in ΔasnO ngg with or without Gln in the hyperosmotic medium, whereas NAGGN accumulated in the WT strain and the complementation strain ΔasnO ngg-C. The addition of Gln significantly increased intracellular NAGGN in the WT strain. These results confirm the notion that NAGGN is synthesized from Gln via the asnO-ngg operon and that it contributes to hyperosmotic adaptation in P. protegens.

Fig 9.

The intracellular NAGGN concentrations of WT, ΔasnO ngg, ΔasnO ngg-C, ΔalgU, and ΔalgU-C cultured under hyperosmotic conditions with 0 mM or 10 mM glutamine (Gln). Intracellular NAGGN was extracted from the cells and the concentration was determined using HPLC and calculated in relation to the dry cell weight (DCW). Experiments were performed in biological and technical triplicates. Error bars indicate the standard deviations (n = 3). Statistical significance was calculated using t-tests. *, P < 0.05; ***, P < 0.001; ****, P < 0.0001.

In addition, we found that extra Gln could not improve the growth of ΔalgU under hyperosmotic conditions (Fig. 8C and D). Moreover, the deletion of algU significantly decreased the synthesis of NAGGN (Fig. 9). These results support the hypothesis that AlgU contributes to the osmotic tolerance of P. protegens partly by regulating the synthesis of NAGGN.

AlgU directly regulates the expression of the asnO-ngg operon

qRT-PCR was used to detect the expression of the asnO gene in the WT strain and its derivative mutants in response to hyperosmotic stress. In the wild-type P. protegens SN15-2, the expression of asnO in NB medium with 400 mM NaCl was significantly higher than that in NB medium, which means that the expression of asnO in the WT strain was upregulated by hyperosmotic stress (Fig. 10A). In ΔalgU, there was no significant difference in the expression of asnO when grown in NB medium or NB medium with 400 mM NaCl (Fig. 10B). However, in the complementation strain ΔalgU-C, expression of asnO was significantly upregulated when grown under hyperosmotic conditions (Fig. 10C). These results reveal that AlgU is necessary for the hyperosmotic induction of asnO-ngg operon expression in P. protegens. Furthermore, an electrophoretic mobility shift assay (EMSA) was performed to determine whether AlgU directly interacts with the asnO-ngg operon. The result shows that the purified His-AlgU bound to the promoter fragment of the asnO-ngg operon in a dose-dependent manner (Fig. 10D). In addition, AlgU failed to bind to the negative control DNA (Fig. S3). Together, these results indicated that AlgU could combine with the promoter region and directly regulate the expression of the asnO-ngg operon.

Fig 10.

AlgU directly regulates the expression of the asnO-ngg operon. Gene expression of asnO was analyzed by qRT-PCR and normalized to the expression of reference gene 16s rRNA. The expression of asnO in NB medium with 400 mM NaCl was significantly higher than that in NB medium in the wild-type P. protegens SN15-2 (A) and ΔalgU-C (C). (B) There was no significant difference between the expression of asnO in NB medium and in NB medium with 400 mM NaCl for ΔalgU. (D) EMSA assay confirmed that AlgU directly binds to the promoter region of asnO-ngg operon. Bands corresponding to AlgU-DNA complexes (Bound DNA) and free DNA (Free DNA) are indicated. The final concentration of AlgU is indicated above each lane. Experiments were performed in biological and technical triplicates. Error bars indicate the standard deviations (n = 3). Statistical significance was calculated using t-tests. ****, P < 0.0001.

DISCUSSION

The ability of P. protegens to resist hyperosmotic stress is important for its application as a biocontrol agent. AlgU is a global regulator responsible for the stress response and biocontrol capability in P. protegens (11). In Pseudomonas fluorescens, AlgU controls the tolerance toward osmotic stress, but the precise mechanism by which it exerts this influence remains unclear (21). In this study, we confirmed that AlgU is a key contributor to the hyperosmotic adaptation of P. protegens. Furthermore, via the combination of an omics analysis and phenotypic experiments, we found that AlgU plays a role in hyperosmotic stress tolerance by influencing cell membrane homeostasis, intracellular ROS accumulation, and dipeptide NAGGN synthesis.

The cell envelope is the first line of defense against environmental stress in Gram-negative bacteria (22). Peptidoglycan maintains the rigidity of the cell envelope and protects cells from hyperosmotic stress (23). The outer membrane protein is an essential component of bacteria’s cell envelope. Lipopolysaccharide, as an important component of the outer membrane of Gram-negative bacteria, is responsible for hyperosmotic adaptation (24). Fatty acids and phospholipids make up the composition of the cell membrane (25). In our current study, transcriptome and metabolomic data revealed that a large number of genes and metabolites related to cell envelope formation were downregulated in ΔalgU compared with that in the WT strain under hyperosmotic conditions. It was consistently found that membrane integrity and permeability were significantly altered in ΔalgU under hyperosmotic conditions. As the barrier between the intracellular and harsh extracellular environment, the proper formation and maintenance of the envelope are crucial for the survival of bacteria (26). The sigma factor RpoE, a homolog of AlgU, is known to be important for envelope maintenance in Escherichia coli (27). Here, our findings illustrate that AlgU regulates hyperosmotic tolerance by affecting cell envelope homeostasis in P. protegens.

Among the differentially expressed genes related to cell envelope stabilization, tolA, which belongs to the Tol-Pal system, attracted our attention. It has been reported that the Tol-Pal system is necessary for maintaining the stability of the outer membrane in Gram-negative bacteria (28). In a previous study, the deletion of tol-pal altered the cell morphology of Acinetobacter baumannii and weakened its resistance to detergent sodium dodecyl sulfate (SDS) (29). Hitherto, there are few reports on whether tolA is involved in hyperosmotic adaptation. In this study, we found that tolA knockout hampered the growth and membrane stability of P. protegens in a hyperosmotic environment. These results provide a new perspective on the mechanism of AlgU’s participation in hyperosmotic adaptation in P. protegens.

Excessive ROS can cause functional changes and damage to nucleic acids, lipids, and proteins (30). Osmotic stress can stimulate the accumulation of ROS and cause oxidative damage to cells (31). Catalase and peroxidase play important roles in ROS detoxification in cells (32). OsmC family proteins are considered antioxidant enzymes, as they can reduce peroxides (33). Glutathione S-transferases act as detoxification enzymes and can protect cells from oxidative injury caused by ROS (34). Our previous study showed that AlgU is important for oxidative stress defense in P. protegens (11). Here, in a hyperosmotic environment, the deletion of algU reduced the expression of many genes that contribute to oxidative damage prevention and increased the accumulation of intracellular ROS. We speculate that the adversely decayed growth of ΔalgU under hyperosmotic conditions may be partly caused by the accumulation of intracellular ROS.

Compatible solutes are low-molecular-weight compounds that can increase the osmotic pressure in cells without interfering with their normal metabolic function (35). The accumulation of compatible solutes through synthesis or import is an effective strategy for bacteria to resist hyperosmotic stimulation (36). The dipeptide NAGGN was first discovered in Sinorhizobium meliloti as a compatible solute with the ability to help strains adapt to hyperosmotic environments (37). Subsequently, the accumulation of NAGGN was observed in Pseudomonas. NAGGN was also observed to be accumulated during the hyperosmotic adaptation of P. aeruginosa (19). In Pseudomonas syringae, the growth of an NAGGN-deficient mutation was severely impaired compared with that of the wild-type strain in a hyperosmotic medium containing 3% NaCl (38). In this study, the RNA-seq data revealed that the expression of the asnO-ngg operon involved in NAGGN synthesis was significantly downregulated in ΔalgU compared with that in the WT. Further analysis demonstrated that NAGGN is important for hyperosmotic adaptation in P. protegens. Glutamine is known to be converted to NAGGN by the acetyltransferase Ngg and the glutamine amidotransferase AsnO in S. meliloti (20). Similarly, we found that the exogenous addition of glutamine is a feasible method to improve the hyperosmotic resistance of P. protegens by increasing the biosynthesis of NAGGN. As a precursor, glutamine can be converted to NAGGN in P. protegens by proteins encoded by ngg and asnO.

In P. syringae, transcriptome analysis indicates that the genes relate to NAGGN synthesis are upregulated by AlgU (39). Here, we found that AlgU is required for the synthesis of NAGGN in P. protegens. In addition, EMSA confirmed the direct binding of AlgU to the promoter of the asnO-ngg operon, which had not been shown before. AlgU can directly regulate the expression of asnO-ngg and the synthesis of NAGGN in P. protegens. Interestingly, in P. aeruginosa, the NAGGN operon is directly regulated by the MarR-type regulator PA3458 (40). Our finding suggests that the regulatory mechanism of NAGGN synthesis in P. protegens is different from that in P. aeruginosa. The regulatory relationship between AlgU and NAGGN synthesis found in this study is an important supplement to the study of the mechanism by which AlgU affects hyperosmotic tolerance.

To summarize, this study reveals the mechanisms of AlgU’s participation in osmotic tolerance in P. protegens. The deletion of algU destroys the cell membrane of P. protegens in hyperosmotic environments. The cell envelope integrity protein TolA regulated by AlgU contributes to osmotic tolerance in P. protegens. Several significantly downregulated genes related to oxidative damage response and higher intracellular ROS levels are also part of the reason why ΔalgU is more sensitive to hyperosmotic stress. In addition, the protective effect of the compatible solute NAGGN in hyperosmotic adaptation in P. protegens is clarified. The extracytoplasmic function sigma factor AlgU can directly regulate the synthesis of NAGGN, thereby affecting the tolerance of P. protegens to high osmotic environments. This research also provides potential molecular targets for research on hyperosmotic adaptation in P. protegens. In the future, we will explore other mechanisms that are involved in the high osmotic tolerance of P. protegens.

MATERIALS AND METHODS

Bacterial strains and culture conditions

The strains and plasmids used in this study are detailed in Table 1. E. coli was grown in an Luria-Bertani (LB) medium (10 g tryptone, 5 g yeast extract, 10 g NaCl, and 1,000 mL deionized water, pH 7.0) at 37°C. P. protegens was grown in Nutrient Broth (NB) medium (5 g tryptone, 3 g beef powder, 10 g sucrose, and 1,000 mL deionized water, pH 7.0) at 30°C. When necessary, NaCl was added to NB medium to achieve a final concentration of 400 mM to create hyperosmotic culture conditions. Antibiotics were used where appropriate at the following final concentrations: kanamycin at 50 µg/mL and tetracycline at 100 µg/mL for P. protegens and 20 µg/mL for E. coil.

TABLE 1.

Strains and plasmids used in this study

Strain or plasmid	Descriptiona	Reference or source
Strains
P. protegens SN15-2	Wild type	Laboratory collection
ΔalgU	P. protegens SN15-2 algU gene knockout strain	(11)
ΔalgU-C	ΔalgU complemented with pME6032-algU; Tetr	(11)
ΔtolA	P. protegens SN15-2 tolA gene knockout strain	This study
ΔtolA-C	ΔtolA complemented with pME6032-tolA; Tetr	This study
ΔasnO ngg	P. protegens SN15-2 asnO ngg gene knockout strain	This study
ΔasnO ngg-C	ΔasnO ngg complemented with pME6032-asnO ngg; Tetr	This study
E.coli BL21(DE3)	F– ompT hsdSB (rB–, mB–) gal dcm (DE3)	Laboratory collection
E. coli DH5α	λ-ϕ80dlacZΔM15Δ(lacZYA-argF)U169 recA1 endA1 hsdR17(rK- mK-) supE44 thi-1 gyrA relA1	Laboratory collection
Plasmids
p2P24Km	Suicide plasmid with sacB used for homologous recombination; derivative of pEX18; Kmr	(41)
p2P24-tolA	p2P24-Km carrying a deleted tolA gene; Kmr	This study
p2P24-asnO ngg	p2P24-Km carrying a deleted asnO ngg gene; Kmr	This study
pME6032	Expression vector containing tetracycline resistance marker; Tetr	(42)
pME6032-tolA	pME6032 containing the intact tolA gene; Tetr	This study
pME6032-asnO ngg	pME6032 containing the intact asnO ngg gene; Tetr	This study
pET28a(+)	Used for protein heterologous expression	Laboratory collection
pET28a(+)-algU	pET28a(+) containing the algU expression cassette	This study

^a Km^r and Tet^r indicate kanamycin and tetracycline resistance, respectively.

Marker-less deletion and complementation of P. protegens genes

The suicide plasmid p2P24-Km containing the sucrose-sensitive gene sacB was employed to construct deletion mutants by using a previous method (41). The plasmid p2P24-Km was digested by EcoR I and Hind III. The upstream and downstream homologous arms of the target gene obtained via polymerase chain reaction were cloned into linearized p2P24Km using an EZ-HiFi Seamless Cloning Kit (GenStar). The constructed suicide vector containing the homologous arms of the target gene was then introduced into P. protegens SN15-2 via electroporation. A kanamycin-resistant colony was subsequently plated on 10% sucrose containing an LB solid medium to generate the mutant. The plasmid pME6032, which contains the tac promoter, was used to complement genes into P. protegens according to a previous method (42). The plasmid pME6032 was linearized with EcoR I and Xho I. The full-length sequence of the target gene amplified via PCR was cloned into the linearized pME6032 using an EZ-HiFi Seamless Cloning Kit (GenStar) to construct the complementation vector. The complementation plasmid was electroporated into the deletion mutant. All the mutants were confirmed via PCR and DNA sequencing. Details of the primer pairs used in this study are shown in Table S1.

Growth and viability analyses

The overnight cultures of P. protegens SN15-2 and derivative mutants were adjusted to OD₆₀₀ = 0.8. Then, the suspensions were inoculated into fresh NB medium and NB medium with 400 mM NaCl at a ratio of 1:100. The cultures were sampled at 4-hour intervals, and the OD₆₀₀ values were recorded. To analyze cell viability, the cells cultured for 24 hours were diluted appropriately and plated on NB agar plates to calculate the total colony forming units.

Cell membrane damage analysis

The WT strain and derivative mutant strains were grown in NB medium with or without 400 mM NaCl to the stationary phase. Cells were collected via centrifugation and washed twice with a NaCl solution isosmotic to the medium. For a cell membrane integrity analysis, the samples were incubated with propidium iodide (PI) at a final concentration of 20 mg/mL for 30 min at room temperature (43); then, fluorescence was measured using a Tecan GENios Microplate Reader (536 nm excitation wavelength and 617 nm emission wavelength). For a cell membrane permeability analysis, the samples were incubated with 1-N-phenylnaphthylamine (NPN) at a final concentration of 10 µM for 5 min at room temperature (44); then, fluorescence was measured at a 350 nm excitation wavelength and a 420 nm emission wavelength. Probe uptake factor = [(fluorescence intensity of sample with probe – fluorescence intensity of the sample without probe)/(fluorescence intensity of wash solution with probe – fluorescence intensity of the wash solution without probe)]/OD₆₀₀ of the sample (45).

Reactive oxygen species analysis

Cells were collected via centrifugation after 24 hours of cultivation and washed twice using an isosmotic NaCl solution. The ROS content was assayed using the probe DCFH-DA according to a previous method (46). The samples were incubated with DCFH-DA at a final concentration of 10 µM for 30 min at room temperature. The fluorescence intensity was measured using a Tecan GENios Microplate Reader, with an excitation wavelength of 488 nm and an emission wavelength of 525 nm.

Analysis of intracellular N-acetyl glutaminyl glutamine amide (NAGGN) content

Strains were grown in hyperosmotic NB liquid medium supplemented with 0 or 10 mM glutamine (Gln). After 24 hours of culture, the cells were harvested and washed twice. Extraction and quantitative analyses of the NAGGN were conducted according to previous studies (38, 47). The collected cell pellets were extracted with 70% ethanol under vigorous shaking for 10 hours at a temperature of 30°C. The ethanol extracts were dried under vacuum at a temperature of 40°C and resuspended with ultrapure water. Then, 20 μL of the filtered resuspended extract was subjected to high performance liquid chromatography (HPLC) analysis using an LC-20 A and an Agilent C18 NH₂ column (250  ×  4.6 mm, 5 µm) at a flow rate of 1 mL 10% acetonitrile per minute. The NAGGN was determined at 210 nm using a UV detector.

Electrophoretic mobility shift assay

The plasmid pET28a (+) was used in this study to obtain the His-tagged AlgU protein (48). pET28a (+) was digested with EcoR I and Hind III. The primer pairs used to amplify the AlgU-coding sequence are shown in Table S1. The AlgU-coding fragment obtained via amplification was inserted into linearized pET28a (+) using an EZ-HiFi Seamless Cloning Kit (GeneStar). The expression vector was then transferred into E. coli BL21 (DE3). The E. coli BL21(DE3) containing the vector was cultured to an OD600 of 0.6 at 37°C; then, the culture was supplemented with 0.1 mM IPTG and transferred to a temperature of 16°C for overnight growth to induce the expression of AlgU-His. The total protein was collected using the ultrasonic cell-break method. The His-AlgU proteins were purified using a 1 mL Ni-NTA agarose column (Sangon Biotech) according to a previous method (49). The promoter sequence of the asnO-ngg operon (263 bp) was amplified using the primer list in Table S1. The promoter fragments were incubated with His-AlgU proteins in an EMSA binding buffer (5% glycerol, 1 mM dithiothreitol, 20 mM Tris-HCl [pH 7.5], 2 mM MgCl₂, 0.1 mg/mL bovine serum albumin, 50 mM NaCl, and 0.5 mM Ethylenediaminetetraacetic acid) at a temperature of 30°C for 30 min. Electrophoresis was performed on a 6% native acrylamide gel for 2 hours at 200 V and 4°C. Finally, ethidium bromide was used to stain the gel for the visualization of DNA bands.

RNA-seq analysis

The WT and ΔalgU strains were cultured to the early logarithmic phase in NB medium amended with 400 mM NaCl at 30℃. The cells of three biological replicates of each strain were collected for total RNA extraction using TRIzol Reagent (Invitrogen) as previously described (50). The quantity of the RNA was monitored using an Agilent Bioanalyzer 2100 system (Agilent), and the quality was detected using a Nanodrop 2000 spectrophotometer (Thermo Scientific). rRNA was removed, and the enriched mRNA was randomly broken into small pieces. Then, cDNA libraries were constructed using a SuperScript double-stranded cDNA synthesis kit (Invitrogen) and sequenced on an Illumina Hiseq 4000 sequencing platform (Illumina Inc.). The reads generated from each sample were aligned to the P. protegens SN15-2 reference genome (CP043179) in the National Center of Biotechnology Information (NCBI). Gene expression levels were normalized by transcripts per million reads (TPM) using RSEM (RNA-seq by expectation maximization) (http://deweylab.github.io/RSEM) (51). DEGs were selected with the thresholds of fold change >2 and a false discovery rate (Benjamini-Hochberg adjusted P value) <0.05 using DESeq2 (https://bioconductor.org/packages/release/bioc/html/DESeq2.html) (52). The DEGs were further analyzed using the GO (53) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway databases (54). KEGG enrichment analysis was performed by KOBAS software using Fisher’s exact test and Benjamini-Hochberg was used for multiple testing correction. If the adjusted P value < 0.05, we consider the KEGG pathways to be significantly enriched.

Untargeted metabolomic analysis

Strains were grown to the early logarithmic phase in NB medium with the addition of 400 mM NaCl at 30℃. The cells of six biological replicates were collected and washed twice with an isosmotic NaCl solution. The total metabolites of each sample (50 mg dry weight) were extracted as previously reported (55). The metabolome was examined using LC-MS analyses via a UHPLC-Q Exactive HF-X system (Thermo Scientific) with a Waters ACQUITY UPLC HSS T3 column coupled to a Q-Exactive HF-X Mass Spectrometer (Thermo Scientific). The pretreatment of the raw data was performed by Progenesis QI (Waters). Metabolites were identified based on the HMDB (https://hmdb.ca/) and KEGG database (https://www.kegg.jp/kegg/compound/). Metabolic features detected at least 80% in any set of samples were retained. Each metabolic feature was normalized using the sum normalization. Variables with a relative standard deviation (RSD) exceeding 30% in quality control (QC) samples were excluded. The R package ropls (Version1.6.2, http://bioconductor.org/packages/release/bioc/html/ropls.html) was used to perform PCA and OPLS-DA. The DAMs were screened based on the VIP obtained by the OPLS-DA model and the P-value of student’s t-test with VIP score >1 and P < 0.05. (56, 57).

Quantitative real-time PCR

Cell samples in the early logarithmic phase were collected, and the total RNA of the cells was separated using TRIzol Reagent (Invitrogen). The RNA samples were transcribed to cDNA with a StarScript II RT Mix and a gDNA Remover Kit (GenStar). qPCR was performed on a BioRad CFX96 using a RealStar Green Fast mixture (GenStar). The specific primers employed in this study are detailed in Table S1. The 16 s rRNA gene was selected as the housekeeping gene. The expression level of the target gene was calculated using the 2 ^−ΔΔCt method.

Statistical analysis

All experiments were performed in triplicate and repeated at least three times. A statistical analysis was performed using Student’s t-test. Differences were considered significant at P < 0.05. The values shown are the means of at least three independent replicates with ± standard deviations.

DATA AVAILABILITY

The transcriptome data for this study have been deposited in the Sequence Read Archive with accession number PRJNA1088423. The metabolomics data for this study have been deposited in the MetaboLights with the unique identifier MTBLS9769.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/aem.00596-24.

Supplemental material. aem.00596-24-s0001.docx.

Figures S1 to S3; Table S1.

Table S2. aem.00596-24-s0002.xlsx.

Detailed data of the differentially expressed genes in ΔalgU compared with WT under hyperosmotic conditions.

Table S3. aem.00596-24-s0003.xlsx.

Detailed data of the differentially accumulated metabolites in ΔalgU compared with WT under hyperosmotic conditions.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.