A salt-tolerant growth-promoting phyllosphere microbial combination from mangrove plants and its mechanism for promoting salt tolerance in rice

Abstract

Background

Mangrove plants growing in the high salt environment of coastal intertidal zones colonize a variety of microorganisms in the phyllosphere, which have potential salt-tolerant and growth-promoting effects. However, the characteristics of microbial communities in the phyllosphere of mangrove species with and without salt glands and the differences between them remain unknown, and the exploration and the agricultural utilization of functional microbial resources from the leaves of mangrove plants are insufficient.

Results

In this study, we examined six typical mangrove species to unravel the differences in the diversity and structure of phyllosphere microbial communities between mangrove species with or without salt glands. Our results showed that a combination of salt-tolerant growth-promoting strains of Pantoea stewartii A and Bacillus marisflavi Y25 (A + Y25) was constructed from the phyllosphere of mangrove plants, which demonstrated an ability to modulate osmotic substances in rice and regulate the expression of salt-resistance-associated genes. Further metagenomic analysis revealed that exogenous inoculation with A + Y25 increased the rice rhizosphere’s specific microbial taxon Chloroflexi, thereby elevating microbial community quorum sensing and ultimately enhancing ionic balance and overall microbial community function to aid salt resistance in rice.

Conclusions

This study advances our understanding of the mutualistic and symbiotic relationships between mangrove species and their phyllosphere microbial communities. It offers a paradigm for exploring agricultural beneficial microbial resources from mangrove leaves and providing the potential for applying the salt-tolerant bacterial consortium to enhance crop adaptability in saline–alkaline land.

Video Abstract

Graphical abstract

The sources of the photos used in this figure are provided below:

Avicennia marina: https://guru.sanook.com/6011/

Acanthus ilicifolius: https://tracuuduoclieu.vn/acanthus-ilicifolius-l.html

Aegiceras corniculatum: https://apps.lucidcentral.org/plants_se_nsw/text/entities/aegiceras_corniculatum.html

Sonneratia apetala: http://www.fpcn.net/a/shuishengzhiwu/20131109/Sonneratia_apetala.html

Excoecaria agallocha: https://efloraofindia.com/2011/03/01/excoecaria-agallocha/

Kandelia candel: https://www.szhb.org/6539.html

Background photo: https://www.discovery.com/nature/the-world-is-waking-up-to-the-importance-of-mangroves

Background

Mangrove ecosystems are located in the coastal habitat of tropical and subtropical intertidal zones. Located at the interface between land and sea, mangroves play an important role in filtering pollution from river basins and act as a protective shield against wind and waves [1]. Due to periodic seawater flooding, mangrove plants have developed a set of salt-tolerant mechanisms different from terrestrial plants or freshwater plants and can adapt to special ecological conditions, such as a high tidal range, high temperature, high salinity, and strong wind.

The special ecosystem of mangrove forests is rich in salt-tolerant and growth-promoting microbial resources [2]. However, the current research on mangrove microorganisms mainly focuses on biogeochemical cycles [3, 4], potential biocontrol agents [5, 6] and the biodegradation of toxic compounds such as microplastics [7, 8], heavy metals [9], and petroleum hydrocarbons [10, 11], while studies on salt-tolerant growth promotion of crops are scarce. The phyllosphere is an important ecological niche of plants, where various microorganisms are colonized. The literature has shown that the area of the global leaf layer is approximately 1 billion km² and that the number of microorganisms colonized on each square centimeter of leaf surface can reach 10⁶–10⁷ [12]. Due to the direct exposure of phyllosphere microorganisms to plenty of variable environmental factors such as temperature, humidity, precipitation, and radiation, phyllosphere microorganisms have developed strong metabolic abilities to cope with changing environments [13]. The phyllosphere microorganisms of mangrove plants have been living in a high-salt environment for a long time and have formed a series of specific morphological structures to adapt to the environment. Based on the presence or absence of salt glands, mangroves can be broadly classified as salt secretors with salt glands and nonsecretors (ultrafiltrators) without salt glands [14]. The leaf epidermal cells and stem epidermal cells of salt-secretors can differentiate into salt glands, which can secrete the excess salt present in the leaves and reduce the damage of salt stress. On the other hand, ultrafiltrators that lack salt glands can maintain low levels of ions through the developed Kelley zone in the inner cortex of the root system by filtration rather than by salt secretion [14]. Due to their salt-secreting structures, salt glands enable the construction of a salt-rich microenvironment and form a small salt lake structure on the leaf surface. The microorganisms on the leaf surface of salt secretors live in a microenvironment where salt stress and nutrients coexist, which is quite different from that of ultrafiltrators without salt glands.

Similar to rhizosphere microbial communities, phyllosphere microorganisms are critical for maintaining plant health [15]. Previous studies have shown that phyllosphere microorganisms are capable of improving plant productivity and maintaining plant adaptability via influencing host function and life history [16]. Thapa et al. noted that lots of members of the phyllosphere microbial community stimulate plant growth by producing various plant hormones or through root secretions [12]. It has been reported that phyllosphere microorganisms can positively affect plant adaptation through varieties of mechanisms, such as promoting plant growth through hormones, regulating plant immune systems, and enhancing tolerance to biological and abiotic stresses [17]. Research has found that Sphingomonas from Arabidopsis leaves can alter the expression of 400 genes, including those involved in signaling and defense responses that promote immunity against Pseudomonas syringae [18]. Other studies have shown that the phyllomicrobiome from tomato “leaf washing” can reduce the infection of P. syringae on tomatoes [19]. Kumar et al. found that microorganisms isolated from the phyllosphere of rice can alleviate drought stress in rice [20]. Arun et al. also reported that the osmotic stress experienced by drought-sensitive rice could be alleviated following the inoculation of microorganisms isolated from the phyllosphere of drought-tolerant rice into the seed or phyllosphere of drought-sensitive rice [21]. These evidences suggest that phyllosphere microorganisms play an important role in promoting plant growth and assisting plants to cope with stress. From this, it can be seen that plant phyllosphere is an important resource bank for mining beneficial microorganisms, and the study of phyllosphere microorganisms will have a profound impact on improving plant growth and alleviating the biotic and abiotic stresses to which plants are subjected. However, most current research has focused on rhizosphere microorganisms with few studies dedicated to the understand of phyllosphere microbiota [22], and the differences in the microbial communities shaped by the presence or absence of salt glands on the leaf microbial community of mangroves are poorly understood.

In addition, it has been found that consortia are more effective than single-strain inoculation for promoting plant development in saline–alkali environment [23]. Therefore, this study aimed to screen combinations of salt-tolerant growth-promoting strains based on the secondary metabolome and salt-tolerant effective reaction of culturable microorganisms and provide microbial resources for the development of salt-tolerant growth-promoting bioformulation products.

Soil salinity is one of the major abiotic constraints on global agroecosystems. At present, approximately 7% of the global land surface area (approximately 1 billion hectares) and more than 20% of arable land (approximately 45 million hectares) are affected by salt stress. Because of climate change and the deterioration of the quality of irrigation water, the saline-alkali soil area in the world will further expand [24]. Soil salinization has severely restricted the cultivation and production of many crops, among which rice (Oryza sativa L.) is considered the most salt-sensitive cereal crop. As one of the world's most important staple crops, the physiological and biochemical indices of rice are affected by salt stress from germination to senescence. Therefore, it is very important to improve the salt tolerance of rice, which is related to food production and food security in many countries, especially Asian countries [25, 26].

Application of halo-tolerant plant growth-promoting bacteria (HT-PGPB) is a cost-effective method to alleviate salt stress. HT-PGPR can regulate plant physiological and biochemical reactions through synthetic osmosis of exopolysaccharides (EPS), accumulation of osmotic adjustment substances, increasing the level of plant hormones, increasing the production of antioxidant enzymes, and inducing systemic resistance (ISR) in plants, thereby reducing plant salt stress [27, 28]. In addition, PGPB inoculation altered the interaction among the local soil microbial community. In the presence of PGPB, the bacterial community had more complex and compact associations, and the enhanced microbial co-occurrence associations may contribute to the PGP effects of PGPB during phytoremediation [29]. Nonetheless, the mechanism by which HT-PGPB regulates the structure and function of microbial communities and participates in host salt tolerance growth promotion remains unknown.

In this study, we posited that highly efficient HT-PGPB could be derived from mangrove leaves, both with and without salt glands, characterized by high salt and low nutrient contents. These bacteria are hypothesized to assist rice in mitigating salt stress. Our aim was to develop a microbial consortium with functional complementarity and to clarify the biological and rhizosphere microecological mechanisms that enhance salt tolerance in rice. Consequently, we developed a synergistic combination of salt-tolerant growth-promoting bacteria (P. stewartii A and B. marisflavi Y25) from mangrove plant leaves. We found that these synergistic communities (SynComs) can enhance rice’s ability to tolerate salt by modulating the plant’s osmotic substances and influencing the expression of salt-resistant genes. Additionally, these SynComs can utilize their metabolite lysine to promote the salt tolerance of the key rhizosphere microorganism C. islandicus, thereby jointly fostering a novel mechanism for salt tolerance in rice.

Methods

Study site and sampling

This study was conducted in the Zhanjiang Mangrove National Nature Reserve, located in southern China (20°14′–21°35′N, 109°40′–110°35′E), characterized by a tropical climate with an average annual temperature of 23.4 °C and a mean annual precipitation of 1600 mm [30]. Leaf samples were collected from six mangrove species within the Tongminggang watershed (20°98′N, 110°16′E) on March 27, 2021, including three species with salt glands: A. corniculatum, A. marina, and A. ilicifolius; and three mangrove species without salt glands: K. candel, S. apetala, and E. agallocha. Sampling followed the protocol reported by Yao et al. [30]. Briefly, 30 healthy leaves were randomly selected from each plant and its neighboring conspecifics to constitute one replicate. Three replicates were collected for each species, with a minimum distance of 50 m between replicates. Samples were promptly placed in sterile bags and transported in ice boxes. Subsequently, some samples underwent culturable microbial screening, while the remainder were stored at − 80 °C for DNA extraction.

Scanning electron microscopy analysis

We randomly collected fresh leaf samples from six species of mangroves in the field. Immediately, mature leaf tissues of approximately 5 mm × 5 mm were excised using sterile scissors (marked front and back, approximately 10 slices of each type) and fixed in 2.5% glutaraldehyde (w/v) [31]. Subsequently, the samples were dehydrated and dried in the laboratory, after which the microstructure and microbial colonization of the leaves were observed using a scanning electron microscope (Hitachi, S-3400 N, Japan).

Collection of microbial samples and DNA extraction

Microbial collection from leaf surfaces was performed as described by Yao et al. [30]. Briefly, 3 g of each sample was randomly weighed and placed in a 50-mL sterile tube containing 30 mL of sterile PBS, and the samples were alternately sonicated and vortexed multiple times. Then, the leaves were transferred to a new 50-mL sterile tube, and the above steps were repeated. The two supernatants were mixed and centrifuged to obtain a pellet for DNA extraction. Total genomic DNA was extracted from leaf surface samples using the E. Z.N.A.® Soil DNA Kit (Omega Biotek, Norcross, GA, USA).

Amplicon sequencing and data analysis

Primers 338F and 806R were used to amplify the V3–V4 region of the bacterial 16S rRNA gene [32], and the primers ITS1F and ITS2R were used to amplify the ITS1 region of the fungal rRNA gene [30]. The samples were sequenced with the MiSeq PE300 platform (Illumina, San Diego, USA) according to the standard protocols by Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China).

The raw 16S rRNA and fungal ITS sequences were quality-filtered by FASTP version 0.19. 6 [33] and merged by FLASH version 1.2.11 [34]. The operational taxonomic units (OTUs) with 97% sequence similarity cutoff values were clustered using UPARSE version 7.1 [35]. All annotations to chloroplast and mitochondrial sequences were removed. The taxonomy of each OTU representative sequence was then analyzed by the RDP Classifier against the Silva V138/16S rRNA gene database [36] and the Unite8.0/ITS database [37], with a confidence threshold of 0.7. We used the subsampling method in MOTHUR (v1.30.1) to normalize the sequence number of each sample for alpha and beta diversity analysis [38]. PICRUSt2 software was used to predict the relative abundance of related enzymes based on the taxonomy of 16S rRNA gene sequences and fungal ITS gene sequences [39].

Culturable microorganism isolation and taxonomic analysis

Microorganisms were isolated according to published methods with some modifications [40, 41]. Briefly, the leaf surfaces of six mangrove species were first washed individually with sterile PBS buffer. The collected wash fluid was serially diluted, and the 10^–4 dilution was spread onto R2A agar, beef extract peptone agar, Gause’s synthetic agar, and potato dextrose agar (PDA) supplemented with ampicillin. Petri dishes were incubated at 30 °C and colony appearance was continuously monitored (1–2 days for bacteria and 1–7 days for fungi). Individual colonies were repeatedly marked on R2A or PDA agar plates to ensure purity.

Bacterial DNA and fungal DNA were extracted using a bacterial genomic DNA extraction kit (TIANGEN, DP302) and the E.Z.N.A.® Fungal DNA Kit (Omega, D3390-01), respectively. The primer pair 27 F and 1492 R was used for bacterial 16S rRNA gene amplification [42], and the primer pair ITS1/ITS4 was used for fungal ITS gene amplification [43]. PCR products were sequenced by Tsingke Biotech Co., Ltd. Sequencing results were assembled, and repeated strains were removed with DNAMAN, resulting in 24 bacterial strains and 18 fungal strains. Finally, taxonomic identities were determined by aligning the sequences with NCBI BLAST (https://www.ncbi.nlm.nih.gov/Blast.cgi). We mapped the full-length 16S rRNA gene sequences of bacterial isolates and the ITS gene sequences of fungal isolates to OTU representative sequences obtained by high-throughput sequencing and calculated the relative abundance of matched OTUs [40], and the strain with a matching degree of more than 97% was retained. All phylogenetic trees were constructed using MEGA version 7.0 based on neighbor-joining.

Qualitative evaluation of salt tolerance and PGP abilities of isolated microorganisms

The beef extract peptone was prepared with NaCl concentrations (w/v) of 5%, 10%, and 15%. The test strain suspension was inoculated at a cell density of OD₆₀₀ = 0.1, cultured on a shaking table for 72 h (30 °C, 180 r·min⁻¹), and the OD₆₀₀ of the medium was measured to assess the salt tolerance capability of the strains. Potato dextrose agar medium with NaCl concentrations (w/v) of 5%, 10%, and 15% was prepared, 5 mm test fungal cakes were inoculated, and the samples were cultured for 7 days to qualitatively evaluate the salt tolerance of the fungal strains. Phosphate solubilization of the strain was measured according to the methods of Bashan et al. [44], the strains were grown in phosphorus-free SRSM liquid medium with calcium phosphate, and the soluble phosphorus converted by the strains was measured by UV–visible spectroscopy. The ability of the strain to dissolve organic phosphorus was determined according to the method by Wang et al. [45]. The potassium concentration was determined according to the method described by Aliyat et al. [46]. Nitrogen fixation capacity was determined using Ashby medium according to the method by Zhou et al. [47].

Effects of selected HT-PGPB strains on rice seed germination

The experimental protocol followed the method described by Yang et al. with some adjustments [31]. Initially, the two strains of bacteria were activated and centrifuged to obtain the bacterial cells. Bacteria were resuspended to the appropriate concentration (strain A: OD₆₀₀ = 0.05, strain Y25: OD₆₀₀ = 0.20) using either 100 mmol·L⁻¹ NaCl solution or sterile water, and the control group was treated with the same amount of sterile 100 mmol·L⁻¹ NaCl solution or sterile water. Healthy, full rice seeds (Oryza Sativa L. spp. japonica) were selected and subjected to routine surface sterilization. Each treatment consisted of three Petri dishes, each containing 20 seeds, and 10 mL of bacterial suspension or sterile water was added. Seeds were treated with suspensions of strain A, strain Y25, and a combination of both (containing 5 mL of each strain), and the germination and seedling growth of rice seeds were observed after 12 days of inoculation. Shoot length and root length were measured to assess salinity tolerance and growth-promoting effects under salt stress.

Metabolite analysis of selected HT-PGPB

Untargeted metabolomics analysis was performed using LC‒MS. The isolated HT-PGPB were grown in beef extract peptone liquid medium for 24 h at 30 °C. The microorganisms were collected after centrifugation and stored at − 80 °C as samples for analysis. Using the UHPLC-Q Exactive HF-X system of Thermo Fisher Scientific as the instrument platform, the conditions for LC‒MS analysis were performed according to the method described by Peng MW et al. [48]. Data were preprocessed by Progenesis QI software (Waters Corporation, Milford, USA). Metabolites were identified with the HMDB (http://www.hmdb.ca/), Metlin (https://metlin.scripps.edu/), and Majorbio databases. In addition, differential metabolites between the two groups were mapped into their biochemical pathways through metabolic enrichment and pathway analysis based on a database search (KEGG, http://www.genome.jp/kegg/).

Effects of selected HT-PGPB strains on rice seedlings

Microbial cultivation was carried out according to the method described above. The bacteria were suspended in sterile 150 mmol·L⁻¹ NaCl solution or sterile water to the optimum concentration of strain, and the control group was treated with the same amount of sterile 150 mmol·L⁻¹ NaCl solution or sterile water. Healthy rice seedlings (Oryza Sativa L. spp. japonica) with similar shapes and sizes were selected as the objects of the pot experiment, and each treatment was repeated three times, with three clusters of rice seedlings planted in each pot. We provide a detailed description in the Supplementary Information (Appendix S1). After the pot experiment, the rhizosphere soil was obtained according to the method of Edwards et al. [49]. The plants and soil were removed from the pots, the bulk soil was manually shaken, and the rhizosphere soil and plants were transferred to a sterile centrifuge tube containing sterile PBS [50]. The tube was shaken and centrifuged, the plant roots were removed, and finally, the rhizosphere soil was obtained.

Determination of physiological and biochemical characteristics and gene expression of rice under different inoculation

The total soluble sugar (TSS) in rice leaves was determined using the anthrone–sulfuric acid method based on Huang A and Singh RP’s methods [51, 52]. The method for measuring malondialdehyde (MDA) was based on that of Toscano et al. [53]. The method for measuring proline (Pro) was based on the ninhydrin colorimetric method of Muhammad et al. [54]. Metal concentrations were analyzed by inductively coupled plasma‒optical emission spectrometry (ICP‒OES) with the simultaneous measurement of sodium (Na) and potassium (K) according to Keshishian et al. with adjustments [55]. According to the method described previously [56], qRT-PCR was performed to study the expression of three genes related to rice salt tolerance. The primer sequences are shown in Table S5. We used OsActin1 as the normalized reference gene and measured the transcript levels of OsSOS1, OsPIN1, and OsCIPK15. Three replicates were collected for each treatment.

DNA extraction and metagenome sequencing analysis of rice rhizosphere soils

Total genomic DNA was extracted from rice rhizosphere soil samples using the E.Z.N.A.® Soil DNA Kit (Omega Biotek, Norcross, GA, USA) according to the manufacturer’s instructions. All DNA samples were fragmented to an average size of approximately 400 bp using Covaris M220 (Gene Company Limited, China) for paired-end library construction. A paired-end library was constructed using NEXTFLEX® Rapid DNA-Seq (Bioo Scientific, Austin, TX, USA). Paired-end sequencing was performed on an Illumina NovaSeq instrument (Illumina, USA) at Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China) using NovaSeq Reagent Kits according to the manufacturer’s instructions.

Metagenome assembly and analysis

The data were analyzed on the Majorbio Cloud Platform (www.majorbio.com). Approximately 84.7 GB of raw reads were generated after Illumina sequencing for all 12 samples. In addition, 41.0–51.1 million clean reads per sample were obtained. Metagenomic data were assembled using MEGAHIT (version 1.1.2) [57], which makes use of succinct de Bruijn graphs, and a total of 3,901,366 contigs were generated. Contigs with a length ≥ 300 bp were selected for further gene prediction and annotation.

Open reading frames (ORFs) were obtained using MetaGene [58], and 4,728,401 ORFs were obtained. All predicted genes were aligned pairwise using CD-HIT [59], and those for which more than 90% of their length could be aligned to another gene with more than 95% identity (no gaps allowed) were removed as redundancies except for the longest gene, resulting in a nonredundant gene catalog comprising 2,482,533 genes with an average length of 417.62 bp. High-quality reads were aligned to the nonredundant gene catalogs to calculate gene abundance with 95% identity using SOAPaligner [60] (version 2.21).

For taxonomy and functional annotation, Diamond [61] (version 0.8.35) was used for the alignment of the representative sequences from the nonredundant gene catalog against the NR and KEGG (Kyoto Encyclopedia of Genes and Genomes) databases with an e-value cutoff of 1e⁻⁵. The correlation between species abundance and functional abundance was calculated using R software (v3.3.1).

Effects of Chloroflexus islandicus alone or in combination with (A + Y25) on rice seed seedlings under salt stress

C. islandicus (Fig. S8e) was provided by Professor Wenjun Li, School of Life Sciences, Sun Yat-sen University. The strain was isolated from Taggejia hot spring in Xizang Province [62]. The experimental protocol is the same as above. We provide a detailed description of microbial inoculation in the supplementary information (Appendix S2). The germination and seedling growth of rice seeds were observed after 12 days of inoculation. Shoot length, root length, fresh weight, and dry weight were measured to evaluate salt tolerance and growth promotion effects under salt stress. To explore the potential effects of (A + Y25) on C. islandicus, we determined the effects of exogenous addition to C. islandicus, see Appendix S3 for details.

Statistical analysis

Student's t test (two-tailed) was used for comparisons between two groups. Analysis of variance (ANOVA) followed by Tukey's HSD test was used to compare means between multiple groups. Rarefaction curves and the Chao1 and Shannon indices were calculated with MOTHUR v1.30.1 [38]. Principal coordinate analysis (PCoA) based on Bray‒Curtis was used to visualize dissimilarity in microbial communities and analysis of similarity (ANOSIM) was used to analyze the differences in microbial community structure between different groups [63]. We performed differential abundance (DA) analysis of microbial taxa using Student's t-test and ANCOM-BC2, and ANCOM-BC2 analysis was performed according to previous studies [64].

Results

Analysis of phyllosphere microbial community structure of six mangrove species with and without salt glands

Microscopic colonization of phyllosphere microorganisms from six mangrove species

We used scanning electron microscopy (SEM) to investigate the microstructures of the surfaces of leaves from six mangrove species. Avicennia marina, Acanthus ilicifolius, and Aegiceras corniculatum have salt glands on their leaves, whereas Sonneratia apetala, Excoecaria agallocha, and Kandelia candel do not. Our results suggest that the presence of salt glands influences the distribution of leaf surface microorganisms (Fig. S1 and S2). Even though the adaxial leaf surface of mangrove plants faces more adverse environmental conditions, such as seawater splashing, it was found that the salt glands on the adaxial leaf surface provided specific colonization sites for leaf surface microorganisms (Fig. S1a-c). Short rod-shaped bacteria, spherical bacteria, and filamentous fungi were observed to colonize and accumulate around the salt glands, whereas the distribution of microorganisms on the leaf surfaces without salt glands was rare (Fig. S1d-f). Furthermore, the abaxial leaf surface exhibited stomatal structures, which also served as colonization sites for leaf surface microorganisms (Fig. S2). Overall, a larger number and more diverse microbial community was observed to colonize around the salt gland structures. However, little is known about the specific microorganisms that preferentially colonize around the salt glands and whether they possess unique functionalities. Therefore, diversity analysis and screen identification of phyllospheric microorganisms from the six mangrove species with and without salt glands were conducted.

Diversity and structural differences of phyllosphere microbial communities in mangrove species with and without salt glands

To optimize and cluster the sequences from all samples of six mangrove species, a total of 610,534 bacterial reads and 1,220,268 fungal reads were retained, and the rarefaction curve of both the Sobs index and Shannon diversity index reached a plateau (Fig. S3), indicating that the sequencing depth was sufficient for subsequent analysis.

The phyllosphere microbial diversity of mangrove plants differed between mangrove species with and without salt glands (Fig. 1a–d). For bacterial communities, the Shannon and Chao1 indices of mangrove species without salt glands were significantly higher than those of mangrove species with salt glands (*P ≤ 0.05, Fig. 1a and b). For fungal communities, compared to mangrove species with salt glands, mangrove species without salt glands have a significantly higher Shannon index (*P ≤ 0.05, Fig. 1c), but there is no significant difference in the Chao1 index between the two groups (Fig. 1d). This may be because the salt gland structures form a high-salinity environment caused by the secretion of salt, which has a filtering effect on leaf surface microorganisms. Principal coordinate analysis (PCoA) and ANOSIM based on Bray‒Curtis distance were used to analyze the similarity and heterogeneity of the bacterial and fungal communities between mangrove species with and without salt glands (Fig. 1e and f). This result indicated that the bacterial (ANOSIM, r = 0.8009, P = 0.001) and fungal (ANOSIM, r = 0.5089, P = 0.001) communities significantly differed between mangrove species with and without the salt glands.

Fig. 1

Effects of the with and without salt glands on the leaf surface of mangrove species on the structure, composition, and function of thephyllosphere microbial community. Alpha diversity metrics: Shannon and Chao1 indices for a–b bacterial and c–d fungal communities. Principal coordinates analysis (PCoA) of e bacterial and f fungal communities based on Bray–Curtis distances. ANOSIM was used to identify significant differences between treatment communities. The ellipses represent 95% confidence intervals. Differentially abundant g bacterial and h fungal genera (only the top 15 genera in terms of relative abundance are displayed) between mangrove species with and without salt glands. Predicted functions related to stress resistance and growth promotion in the i bacterial and j fungal community. Two-tailed Student’s t test: *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. The error bars represent the mean ± SDs

Differences in the composition and functionality of the phyllosphere microbial communities between mangrove species with and without salt glands

By analyzing the composition of the phyllosphere microbial community in mangrove species with and without salt glands, we found that the composition of the microbial community differed between mangrove species with and without salt glands (Fig. S4). Specifically, Fulvimarina, unclassified Rhodobacteraceae, Salinisphaera, and unclassified Sphingomonadaceae, Pesudokineococcus, Zunongwangia, Kushneria, and Halomonas were significantly enriched in mangrove species with salt glands (*P ≤ 0.05, Fig. 1g and Fig S5a). In contrast, Pantoea, Pseudomonas, and Bacillus were significantly enriched in mangrove species without salt glands. (*P ≤ 0.05, Fig. 1g and Fig S5a). In the fungal community, Cladosporium was the most abundant taxon for both types of mangrove species (Fig. S4b). Unclassified_o_Tremellales and unclassified Capnodiales were significantly enriched in mangrove species with salt glands (Fig. 1h and Fig. S5b), whereas Phaeophleospora, Toxicocladosporium, Nigrospora, and Aureobasidium were significantly enriched in mangrove species without salt glands (Fig. 1h and Fig. S5b). These results suggest that there are distinct compositional differences in the phyllosphere microbial communities of the two groups of mangrove species.

To explore functional differences, we predicted potential enzyme functional disparities in nutrient cycling and stress resilience between these microbial communities (Fig. 1i and j). In the bacterial community, the relative abundances of catalase, superoxide reductase, nitrogenase, nitrite reductase (NO-forming), and acid phosphatase were significantly enriched in mangrove species without salt glands, while nitrite reductase (NADH), 3-phytase, and cellulase were notably enriched in mangrove species with salt glands (P ≤ 0.05, Fig. 1i). In the fungal community, peroxidase, beta-glucosidase, and saccharolysin were more abundant in mangrove species without salt glands, while cellulase activity was significantly higher in species with salt glands (P ≤ 0.01, Fig. 1j). These predictive functional analyses suggest that microbial communities in mangrove species without salt glands may have a higher potential for stress resilience.

Synthesis of salt-tolerant growth-promoting microbial flora

To delve deeper into the function of mangrove phyllosphere microbial communities, we isolated culturable microorganisms from these habitats and matched them to community profiles derived from high-throughput sequencing. Using various media, we isolated 24 bacterial and 18 fungal strains from six mangrove species (Fig. 2a, b and Table S1). We calculated the relative abundance of each strain in the microbial community by mapping the sequence of these 42 strains to the OTU representative sequence of the microbial community. The results showed that 20 bacterial strains and 10 fungal strains matched possible OTUs, while 4 bacterial strains and 8 fungal strains did not match similar OTUs (Table S2). We further tested the salt tolerance and PGP properties of these isolated strains (Table S3), and the results showed that 90% of the isolated strains could grow in 5% (w/v) NaCl, and more than 40% of isolates showed phosphorus solubilizing ability. Notably, two prominent isolates, P. steartii A (Fig. 2c and d) from the Pantoea genus and B. marisflavi Y25 from the Bacillus genus (Fig. 2e and f), showed remarkable salt tolerance (10% (w/v) NaCl and 15% (w/v) NaCl, respectively) and PGP properties (Table S3). Pantoea and Bacillus genus were significantly enriched in mangroves without salt glands (*P ≤ 0.05, Fig. 1g) and were commonly distributed in these three mangroves species (Fig. S6), suggesting their preference for without salt glands environments. OTU matching results showed that the similarity between P. steartia A and OTU1046 Pantoea sp. was 99.30%. However, we did not find an OTU with a similarity greater than 97% to B. marisflavi Y25, we found that its closest OTU256 Bacillus sp. (similarity 96.23%) was only distributed in three mangrove plants without salt glands, and a previous study showed that close strains have similar environmental adaptability and growth characteristics [41].

Fig. 2

Identification and characterization of salt-tolerant and growth-promoting microbial taxa. a Phylogenetic tree of identified isolated bacteria. b Phylogenetic tree of identified isolated fungi. Green dots are outgroups. c Colony morphology of P. stewartii A on R2A agar. d 16S rRNA phylogenetic tree of P. stewartii A. e Colony morphology of B. marisflavi Y25 on R2A agar. f 16S rRNA phylogenetic tree of B. marisflavi Y25

We further investigated the growth-promoting effects of bacterial strains A and Y25, both individually and in combination, on rice seedlings under water and salt stress conditions. Under water conditions, both the individual strains A and Y25, as well as their combination A + Y25, significantly enhanced the shoot length, root length, and overall plant height of the rice seedlings. Interestingly, no significant difference was observed between the growth promotion imparted by individual strains and the combined inoculum (P > 0.05; Fig. 3a and c). Upon exposure to 150 mmol·L⁻¹ NaCl, the strains significantly bolstered the shoot, root, and overall lengths (*P ≤ 0.05; Fig. 3b and d). Notably, A + Y25 showed the most effective root growth promotion under salinity stress, exceeding the effects of individual strains (Fig. 3d). Overall, the combination A + Y25 demonstrated growth-promoting effects under both water and salt stress conditions, with a noticeable improvement in mitigating the negative impact of salt stress on rice seedlings, especially in terms of root growth.

Fig. 3

Growth-promoting effects of A and Y25 (alone or in combination) on rice seedlings under water and salt stress. a Phenotype of the effect of the strain on the growth of rice seedlings under water conditions. b Phenotype of the effects of the strain on the growth of rice seedlings under salt stress. c Effects of the strain on the growth of rice seedlings under water conditions. d Effects of the strain on the growth of rice seedlings under salt stress. Different lowercase letters indicate significant differences, ANOVA, Tukey’s HSD test, *P ≤ 0.05. The error bars represent the mean ± SDs

Further probing into the metabolic profiles of strains A and Y25 highlighted their repertoire of metabolites associated with salt tolerance and nutrient provision (Fig. S7). Particularly in strain Y25, we identified elevated levels of L-glutamate, aspartic acid, and betaine metabolites frequently associated with the salinity stress response in plants, with betaine playing a pivotal role [65,66,67]. Conversely, strain A could produce more L-lysine compared with strain Y25 (Fig. S7). Given the distinctive features of strains A and Y25, as well as the superior root growth-promoting ability of A + Y25 under salt stress, we aimed to further explore the effect and biological mechanisms of this bacterial combination in enhancing the salt tolerance and growth of rice seedlings under salt stress through soil cultivation experiments.

Effects and biological mechanism of the combination A + Y25 promoting rice salt tolerance and growth under salt stress

Building on our preliminary findings, we identified the optimal bacterial consortium, A + Y25. By establishing a rice cocultivation system, we delved deeper into the biological mechanisms by which A + Y25 alleviates salt stress and promotes growth under soil cultivation conditions (Fig. 4). Inoculation with A + Y25 led to a significant increase in the height of rice seedlings (**P ≤ 0.01, Fig. 4b), with growth rates of 24.58% and 25.42% under water and 150 mmol·L⁻¹ NaCl treatments, respectively. Notably, under salt stress, inoculation with A + Y25 effectively restored the seedlings’ stature to that observed under nonstress conditions (Fig. 4a and b). This evidence underscores the ability of this bacterial consortium to counteract the inhibitory effects of salt stress and restore a near-normal growth phenotype in rice.

Fig. 4

Biological mechanism underlying the growth promotion and enhanced salt tolerance in rice under salt stress by A + Y25 in a soil cultivation experiment. a Growth phenotype of rice under different treatments. b Height promotion results for rice plants. c Soluble sugar content. d Malondialdehyde content. e Proline content. f K⁺ content. g Na⁺ content; different letters indicate significant differences between groups, ANOVA, Tukey’s HSD test, P ≤ 0.05. h Expression of salt-responsive genes in rice leaves. Two-tailed Student's t test, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001. The error bars represent the mean ± SDs

Furthermore, we found that the A + Y25 treatment significantly improved a series of physiological indicators of rice under salt stress, while there was no significant effect under the water treatment. Under salt stress, inoculation with A + Y25 significantly increased the content of soluble sugar (*P ≤ 0.05, Fig. 4c) and proline (*P ≤ 0.05, Fig. 4e) and significantly decreased the content of malondialdehyde in the rice acid content (*P ≤ 0.05, Fig. 4d). The combination A + Y25 also significantly improved the ion balance of rice under salt stress. The contents of Na⁺ and K⁺ are important indicators for evaluating the response of plants to salt stress. Under effective regulation, compared with the control, the content of K⁺ significantly increased (*P ≤ 0.05, Fig. 4f), and the content of Na⁺ significantly decreased (*P ≤ 0.05, Fig. 4g).

Gene expression analyses further elucidated the role of A + Y25. Under 150 mmol·L⁻¹ NaCl, rice leaves inoculated with A + Y25 manifested elevated expression levels of critical genes, including the Na⁺ transporter OsSOS1, the auxin efflux carrier OsPIN1, and the calcineurin B-like protein-interacting protein kinase OsCIPK15, when juxtaposed against control treatments. Collectively, these genes orchestrate pivotal roles in the plant salt stress response, mediating Na⁺/K⁺ ion transport, auxin secretion, and overall plant stress regulation. Thus, A + Y25 augments salt tolerance in rice, possibly by upregulating these key stress-responsive genes. Our findings indicate that A + Y25 not only modulates osmolytes but also orchestrates a resilience response by regulating key stress genes and fine-tuning ion balance, thereby bolstering rice growth and salinity tolerance under stressful conditions.

Regulation of A + Y25 on the structure and function of the rice rhizosphere microbial community under salt stress

We further observed the effect of inoculation with A + Y25 on the structure and function of the rice rhizosphere microbial community from the perspective of soil microecology (Fig. 5). The results of metagenomic data showed that inoculation with A + Y25 significantly reduced the Shannon index of the microbial community under salt stress (*P ≤ 0.05, Fig. 5a and b). Moreover, principal coordinate analysis (PCoA) and ANOSIM based on Bray–Curtis distance were used to analyze the similarity and heterogeneity of microbial communities under water conditions and salt stress, respectively (Fig. 5c and d). The results showed that the application of A + Y25 and the control were separated under salt stress, but there was no significant separation after inoculation with A + Y25 under water conditions; therefore, inoculation with A + Y25 significantly reduced the diversity of microorganisms under salt stress, but the overall community structure was not significantly affected. In addition, we found that the abundance of unclassified Chloroflexi, the most dominant species in the microbial community, increased significantly after inoculation with A + Y25 under salt stress (*P ≤ 0.05, Fig. 5e). The analysis of species and functional contributions showed that unclassified Chloroflexi species played an important role in regulating functions such as metabolic pathways, biosynthesis of secondary metabolites, and microbial metabolism in diverse environments at KEGG pathway level 3 (Fig. 5f). These results suggested that A + Y25 had a significant effect on the diversity of the rice rhizosphere microbial community under salt stress, especially because it enhanced the abundance of unclassified Chloroflexi and regulated the overall function of the microbial community through this microbial population, improving the growth environment of rice.

Fig. 5

Effects of inoculation with A + Y25 on the structure and potential function of the rice rhizosphere microbial community. Alpha diversity indicators: a Shannon indices and b Chao1 indices. Principal coordinate analysis (PCoA) based on Bray‒Curtis distances under c water and d salt stress. ANOSIM was used to identify significant differences between treatment communities. Ellipses represent 95% confidence intervals. e Effects of A + Y25 on the abundance of rice rhizosphere microbial communities under water treatment and salt stress (top 10 relative abundances at the genus level). f Species and function contribution analysis. After inoculation with A + Y25, association analysis between taxa and the relative abundance of KEGG pathway level 3 (top 10 relative abundances) functions was carried out under salt stress. Two-tailed Student's t test was used to analyze the significant differences, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001. The error bars represent the means ± SDs

To study the regulatory effect of A + Y25 on the function of the rice rhizosphere microbial community, we used the KEGG database to annotate the functions of rice rhizosphere microorganisms in detail, in which a total of 5070 KO functional categories were annotated. On this basis, we further conducted a functional difference analysis between the samples inoculated with A + Y25 and those not inoculated with A + Y25 to analyze their effects and differences under water and salt stress conditions (Fig. 6). The results showed that under water conditions, 67 KO functional categories were upregulated, whereas 155 KO functional categories were downregulated (Fig. 6a). Under salt stress, 497 KO functional categories were upregulated and 367 KO functional categories were downregulated (Fig. 6b). This finding indicated that inoculation with the A + Y25 strain combination under salt stress had a significant effect on the function of the soil microbial community. Further special attention was given to KO functional categories directly related to rice growth, stress resistance, and plant growth promotion, including ion transport (22 categories), nitrogen metabolism (51 categories), and phosphorus cycling (46 categories), and the results of significant changes are shown (Fig. 6c).

Fig. 6

Regulation of rice rhizosphere microbial community function after inoculation with A + Y25 under salt stress. a Changes in KO functional categories under water conditions. b Changes in KO functional categories under salt stress; up represents upregulated expression, down represents downregulated expression, and Nosig represents the amount without significant change. c KO functional categories related to ion transport, nitrogen metabolism, and phosphorus cycle with significant changes. d The relative abundance of the top 10 KEGG pathways and their significant differences. e Genes that are significantly altered in quorum sensing pathways. Two-tailed Student's t test was used to analyze the significant differences, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001

Regarding ion transport, the relative abundances of K02115 (ATPF1G, atpG) and K02133 (ATPeF1B, ATP5B, ATP2) increased after inoculation with A + Y25 under water conditions, whereas under salt stress, the relative abundances of K02112 (ATPF1B, atpD) and K00348 (nqrC) decreased, while that of K02133 (ATPeF1B, ATP5B, ATP2) increased. These findings indicated that inoculation with the A + Y25 strain was able to regulate the expression of ion transport-related genes in microbial communities under different environmental conditions. In terms of phosphorus cycling, A + Y25 inoculation under water conditions augmented the abundances of K07768 (senX3) and K05815 (ugpE). Under salt stress, however, K06167 (phnP), K04750 (phnB), K05781 (phnK), and K06163 (phnJ) decreased, while K22906 (phy) and K02037 (pstC) surged, underscoring the participation of A + Y25 in the phosphorus cycle via differential P-cycle gene regulation. In the nitrogen metabolism pathway, the relative abundances of K03385 (nrfA) and K00261 (GLUD1_2, gdhA) decreased significantly after inoculation with A + Y25 under water conditions, while under salt stress, K01948 (CPS1), K15876 (nrfH), K00374 (narI, narV), K00266 (gltD), and K15371 (GDH2) significantly increased. These findings suggest that the A + Y25 strain may help rice better cope with salt stress by adjusting the nitrogen metabolism of the microbial community.

We further performed enrichment analysis on KEGG pathways (Fig. 6d), and the results showed that the KEGG-enriched pathways did not change significantly after inoculation with A + Y25, while the relative abundance of the quorum sensing pathway increased significantly (*P ≤ 0.05, Fig. 6d). A total of 10 genes were involved in significant changes in quorum sensing, mainly K07667 (kdpE), K07173 (luxS), K07782 (sdiA) and K20260 (phzA_B) upregulation and K18139 (mtrE), K01658 (trpG), K01658 (trpE), K10915 (cqsA), and K10557 (lsrD) downregulation (*P ≤ 0.05, Fig. 6e). This may indicate that in a high-salt environment, the microbial community can better adapt to adversity by improving its internal communication and coordination mechanisms and simultaneously enhance its colonization of the rice rhizosphere under salt stress by forming a biofilm structure to assist rice in salt resistance. The above results indicated that inoculation with A + Y25 could help rice jointly resist salt stress by altering the structure and function of the rhizosphere microbial community.

Effects of C. islandicus on salt tolerance of rice seedlings under salt stress

Under salt stress, C. islandicus inoculation could significantly increase the fresh weight, dry weight, and root length of the rice seedlings and could promote their salt tolerance and growth to a certain extent (Fig. S8a-c). This is the first reported evidence that C. islandicus promoted salt tolerance and growth in plants under salt stress. Moreover, the combined treatment with C. islandicus and A + Y25 can also significantly promote the root length, shoot length, fresh weight, and dry weight of rice seedlings. Interestingly, A + Y25 synergistically increased the promoting effect of C. islandicus on rice shoot growth, but it had the opposite effect on root length. Studies have shown that high doses of nitrogen supply inhibit root growth and the ratio of root length to bud length decreases [68]. Another study found that with the increase of nitrogen application, corn root length decreased [69]. Therefore, we speculated that the combined treatment of C. islandicus with A + Y25 may release more nitrogen-containing nutrients that can be utilized by rice and promote the growth of rice shoots; however, at the same time, this combination has an inhibitory effect on the roots of rice, which needs further study. It is also suggested that attention should be given to the application ratio of C. islandicus to A + Y25, as well as the interaction between microorganisms in the subsequent bactericide application. In general, C. islandicus alone and in the presence of A + Y25 can promote salt tolerance and growth of rice seedlings under salt stress. This further verified that A + Y25 collaborated to promote salt tolerance and growth of rice by regulating functional microorganisms such as C. islandicus in rhizosphere. We further analyzed the effects of four key metabolites of strains A and Y25 (lysine, glutamate, aspartate, and betaine) on the growth of C. islandicus under salt stress. Interestingly, only lysine significantly promoted salt tolerance and growth of C. islandicus (Fig. S8d), indicating that lysine, the metabolite of A and Y25, plays an important role in regulating salt tolerance and increasing the abundance of C. islandicus in rhizosphere.

Discussion

The structure of salt glands is an important adaptive structure for mangroves to cope with high-salt and high-humidity environments. Salt glands can effectively regulate the salt balance in plants and reduce the impact of salt stress on plants by actively excreting salt [70]. However, although the physiological differences between mangrove species with and without salt glands have been extensively studied [71,72,73], the identity and functional differences of leaf surface microorganisms of these two types of mangrove species have not been fully studied.

In this study, we identified pronounced compositional disparities between the phyllosphere microbial communities in mangrove species with and without salt glands. Such variations potentially underscore the adaptive responses of phyllosphere microorganisms to the physiological attributes of plants with or without salt glands. This aligns with previous studies demonstrating that the composition and function of bacterial communities residing on leaf surfaces are typically modulated by host physiological traits and environmental parameters [74]. For instance, the abundance of Rhodobacteraceae [75], Salinisphaera [76], and Sphingomonadaceae [77] in mangrove species with salt glands might be associated with their heightened salt tolerance and adaptability to high-salinity environments. The presence of these bacteria might assist mangrove species with salt glands in better managing the salt stress. Conversely, Pantoea and Bacillus were markedly enriched on the leaves of mangrove species without salt glands. These microorganisms might provide critical support for the survival and growth of mangrove species without salt glands in saline conditions; for example, certain Bacillus have been identified to facilitate plant root salt ion exclusion via organic acid secretion, thus bolstering plant salinity resistance [78]. On the other hand, Bacillus [24] and Pantoea [79] might also counteract salt stress by enhancing plant nutrient provision and delivering essential nitrogen and phosphorus for plant growth. According to the results of culturable microorganisms, the colonization of microorganisms has a certain selectivity. Our results indicate that Bacillus and Pantoea have more abundant colonization in mangrove species without salt glands. However, the abundance of Pantoea and Bacillus in the phyllosphere of these plants might also resonate with these plants' distinct ecological niches. Salt glands serve as a hallmark adaptive feature in mangroves for high-salinity environments, sustaining ionic balance within the plant by accumulating and excreting salts [80]. In contrast, mangrove species without salt glands might rely on interactions with phyllosphere microorganisms to confront saline settings, perhaps by modulating plant ion uptake and distribution or by refining nutrient provision.

For fungal communities, although Cladosporium is the most abundant group in both types of mangrove species, the differences in other major fungal taxa in the phyllosphere of mangrove species with and without salt glands may reflect their respective survival strategies and special adaptability to the environment [81]. In addition, we found significant differences in potential functional properties between microbial communities on the leaf surface of mangrove species with and without salt glands. The microbial communities of mangrove species with salt gland have shown a high ability to process nitrates and degrade complex organic matter, which may reflect the need for nutrient cycling and coping with pollution risks in high-salt environments. In contrast, microorganisms on the leaf surface of mangrove species without salt glands have shown a greater ability to resist stress and promote nutrient cycling. For example, small molecules of nitric oxide (NO) have been identified as endogenous gas transmitters involved in alleviating salt or other types of stress [82]. We found that nitrite reductase (no-forming), which is related to the formation of NO, was significantly enriched in mangrove species without salt glands which may indicate that mangrove species without salt glands are more dependent on their phyllosphere microbial communities in adapting to high-salt environments.

To further investigate the functions of mangrove phyllosphere microbial communities, we isolated culturable microorganisms from the leaf surfaces. Given that salinity is a dominant stressor in marine settings, our findings suggest that mangroves might counteract this pressure by fostering a highly salt-tolerant microbial community in their phyllosphere [2]. Additionally, over 40% of the strains demonstrated phosphorus solubilizing capabilities, suggesting that these microorganisms could enhance mangrove growth and development by improving phosphorus availability [83]. Notably, the two strains we identified, P. stewartii A and B. marisflavi Y25, manifested the most pronounced salt resilience and growth-promoting abilities. Of particular interest, Pantoea and Bacillus, which were significantly more abundant in mangrove species without salt glands than in their counterparts with salt glands, almost exclusively colonized the phyllosphere of mangrove species without salt glands, indicating a marked preference.

We further explored the effects of P. stewartia A and B. marisflavi Y25 on the growth of rice seedlings under salt stress. The results showed that A and Y25 alone and their combination (A + Y25) could be significantly improved under water and salt stress conditions. This growth-promoting effect was more pronounced under salt stress conditions. Furthermore, strain A and strain Y25 also played an important role in improving the tolerance of rice to salt stress, especially the combined strain A + Y25, which had the most significant growth promotion effect on the root under salt stress; this effect was greater than that of the single strain treatment. Using LC‒MS to analyze the secondary metabolites of the two strains, we found that the Y25 strain contained higher contents of L-glutamate, aspartic acid, and betaine, which are known metabolites resistant to salt stress. On the other hand, strain A surpassed strain Y25 in terms of the content of L-lysine, which suggested that strain Y25 and strain A were functionally complementary and explained that the combined strain A + Y25 has a stronger potential to alleviate plant salt stress and promote growth.

Concurrently, under soil cultivation conditions, we delved into the biological mechanism through which the A + Y25 consortium augments rice salinity tolerance. The combined effects of P. stewartii A and B. marisflavi Y25 highlighted their pivotal roles in ameliorating rice growth and physiological responses under salt stress. Inoculation with A + Y25 significantly bolstered the growth of rice seedlings under both hydrated and salt-stressed conditions, primarily attributed to their modulatory impact on a spectrum of intrinsic physiological processes in rice, such as osmolyte adjustment, ion balance refinement, and regulation of key stress-responsive gene expression. Under salt stress conditions, the A + Y25 consortium significantly improved the rice's physiological markers, including enhanced levels of soluble sugars and malondialdehyde while diminishing proline content. We recognize that soluble sugars [84] and malondialdehyde [85] are deemed critical reaction substances for plants countering salt stress, whereas proline, an osmotic regulatory substance, typically accumulates under saline conditions [85]. Previous studies have indicated that microbe inoculations, such as those involving plant growth-promoting rhizobacteria (PGPR), can bolster plant salt tolerance by altering physiological parameters, aligning with the beneficial effects observed in this study upon A + Y25 inoculation [86]. Our findings further elucidate that A + Y25 effectively elevated the K⁺ content while diminishing the Na⁺ content in rice roots under saline stress. This aligns with the findings of Wang et al., suggesting that rice safeguards young leaves from ionic injury by sequestering excessive Na⁺ and Cl⁻ in older leaves, thus fine-tuning ionic equilibrium to better withstand saline conditions [87]. Furthermore, postinoculation with the A + Y25 consortium, there was a marked upregulation in the expression levels of salt stress-responsive genes in rice leaves, namely, OsSOS1, OsPIN1, and OsCIPK15. These genes play instrumental roles in plant Na⁺ transport, auxin efflux, and stress regulation [88]. The aforementioned findings shed light on the primary pathways and key genes modulated by A + Y25 in facilitating rice salinity resistance.

The plant rhizosphere represents a convoluted niche where microbial interactions with plant roots can profoundly impact plant growth, development, and stress responses [89]. Thus, to elucidate the salinity-promotion mechanism of A + Y25 in rice, we focused on the regulatory effects of exogenously inoculated functional microbial agents on the rice rhizosphere microbial community. In this investigation, upon the inoculation of rice with A + Y25 under saline conditions, there was a conspicuous reduction in the microbial diversity of the rhizosphere. Such shifts could likely stem from the intrusion of exogenous salt-tolerant microbial populations into the rhizosphere environment, modulating interactions among indigenous microbial communities. Notably, A + Y25 application did not markedly alter the overarching structure of the microbial communities within the rhizosphere, potentially because A + Y25 primarily affected select key species. Moreover, under salt stress, we observed a significant surge in the abundance of unclassified Chloroflexi. This might be attributed to either the direct provision of nutrients by the metabolites of A + Y25 strains or indirectly through the modulation of root exudates by A + Y25. Previous studies have corroborated that microbial populations might enhance each other's abundance via mutualistic feeding mechanisms [90]. Concurrently, this species exhibited significant associations with ten functions within KEGG level 3 pathways, such as metabolic pathways, biosynthesis of secondary metabolites, and microbial metabolism in diverse environments, potentially leading to an upregulation of microbial community sensing (Fig. 6). Similar to previous studies, Xian et al. identified a strain of Chloroflexi with a wide range of life-promoting activities, which played an important role in potential microbial interactions [62]. This study underscores that exogenous A + Y25 inoculation might enhance rice rhizosphere microbial community sensing and functionality by increasing such microbial abundance, collaboratively enhancing rice salinity resistance and growth.

Furthermore, the study revealed that under salt stress, inoculation with A + Y25 led to more pronounced alterations in the soil microbial community functions than under water conditions, suggesting that rice under salt stress seems to be more reliant on the exogenous A + Y25 consortium. This aligns with the rice phenotypic and physio-biochemical responses, where the most notable growth-promoting effect of the A + Y25 consortium was observed under saline treatment (Fig. 4a). The introduction of A + Y25 modulated the expression of ion transport-related genes within the microbial community, which might contribute to maintaining ion homeostasis within rice cells, thus mitigating the stress exerted by the high-salinity environment. Notably, under salt stress conditions, the inclusion of A + Y25 increased the abundance of genes related to nitrogen metabolism within the community, possibly bolstering the adaptive ability of rice under salt stress. This finding parallels with the findings of Khumairah et al., where salt-tolerant plant growth-promoting rhizobacteria isolated from saline–alkali soils were shown to facilitate rice nitrogen fixation and alleviate salt stress through the production of indole-3-acetic acid and nitrogenase [91]. Additionally, under salt stress, the relative abundance of most of the genes associated with phosphorus cycling in the inoculated rhizosphere microbial community significantly decreased, whereas a few other genes exhibited notable increases. Such modulation could be viewed as an “energy conservation” strategy. Phosphorus is a pivotal nutrient intricately linked to numerous fundamental life processes, including energy transduction and molecular synthesis. Under salt stress, microorganisms might be confronted with augmented energetic constraints, thereby necessitating an optimization of their phosphorus acquisition and utilization strategies [92]. The downregulation of certain genes related to phosphorus cycling could potentially facilitate microbial energy conservation, redirecting it toward other vital stress-response processes.

We found that the enrichment degree of KEGG pathways in rice rhizosphere microbial communities did not change significantly after inoculation with A + Y25 under water conditions. However, the relative abundance of the quorum-sensing pathway increased significantly under salt stress, suggesting that A + Y25 may function by enhancing the quorum-sensing mechanism in response to salt stress. Bacterial quorum sensing (QS) is an intercellular communication process mediated by autoinducers that allows bacteria to coordinate their behavior in a cell-density-dependent manner [93]. Previous studies have also shown that anaerobic ammox bacteria increase the production of acylated-L-homoserine lactone (AHL) signaling molecules to adjust their behavior and facilitate their growth when faced with changes in environmental stress [94]. In addition, Zhu et al. further pointed out that a high concentration of AHLs can promote bacteria to produce more extracellular polymers, which helps to improve the resistance of bacteria to salt stress [95]. However, the physical basis and molecular mechanism of enhanced quorum sensing in salt-stressed bacteria induced by inoculation with A + Y25 are still unclear. A + Y25 may act as the leader of a community to gather functional bacterial compatriots. This led us to consider the role of quorum sensing in microbial resistance to environmental stress and in helping plants become resilient.

In the quorum sensing pathway, we found that the K07667 (kdpE), K07173 (luxS), K07782 (sdiA), and K20260 (phzA_B) genes were significantly upregulated under salt stress. Among them, the K07667 (kdpE) gene encodes the KdpE protein, which may be involved in the cell response to salt stress and potassium ion transport [96]. The K07173 (luxS) gene encodes the LuxS protein, which may be involved in the production of the intercellular signaling molecule AI-2, thereby regulating the quorum sensing response and inducing chemotaxis and biofilm formation [97]. LuxS is considered to be the main synthetase of the AI-2 signaling molecule, which is a quorum-sensing signaling molecule widely found in bacteria. It can play a regulatory role between bacterial species. The K07782 (sdiA) gene encodes the SdiA protein, which may play an important regulatory role in quorum sensing as a global regulator [98]. However, the K20260 (phzA_B) gene encodes the PhzA_B protein, which is related to the synthesis of secondary metabolites, and its upregulation plays an important role in antioxidant stress and enhancement of salt tolerance [99]. On the other hand, in the quorum sensing pathway, the expression of the K18139 (mtrE), K01658 (trpG), K01658 (trpE), K10915 (cqsA), and K10557 (lsrD) genes was inhibited. Among them, the K18139 (mtrE) gene encodes the MtrE protein, which may play a role as a negative regulator in the quorum sensing pathway. The LsrD protein encoded by the K10557 (lsrD) gene is responsible for transporting the exogenous quorum sensing signaling molecule AI-2 into the cell and participating in its degradation [100]. In particular, studies have shown that E. coli enhances the expression of the lsr gene to enhance the response to AI-2 under certain environmental stresses, such as nutritional deficiency [101]. These results suggest that A + Y25 may improve its environmental adaptability and optimize survival strategies by altering the expression of quorum-sensing-related genes under salt stress. The previous analysis indicated that only inoculation with A + Y25 under salt stress could significantly increase the abundance of unclassified Chloroflexi, which was accompanied by a significant increase in quorum sensing function. This suggests that inoculation with A + Y25 under salt stress can play a key role by significantly increasing the number of specific groups of unclassified Chloroflexi, thus significantly improving the quorum sensing function of microbial communities, alleviating salt stress and promoting the growth of rice. This provides valuable insights into how salt-tolerant strains enhance plant resistance to salt stress by regulating rhizosphere microbial communities.

In this study, we conducted an inaugural exploration of the diversity and structural differences in phyllosphere microbial communities present on the leaves of mangrove species with and without salt glands within a mangrove ecosystem. For delving deeper, we isolated microorganisms from the leaf surfaces of mangrove plants. Subsequently, we constructed the salt-tolerant plant growth-promoting consortium A + Y25. This consortium not only modulates osmolytes and governs the expression of crucial stress-responsive genes, improving the ion balance, but also impacts the functionality of microbial communities by augmenting specific microbial taxa. This can effectively ameliorate the response of rice to salt stress and bolster its growth. Our findings offer fresh insights into the intricate interplay between mangrove species and their foliar microbiomes, laying a theoretical groundwork for the potential application of these functional strains in agricultural practices. Microorganisms in agriculture play a pivotal role in enhancing crop adaptability on saline–alkaline lands, enriching soil fertility, and mitigating soil-borne diseases. Specifically, halophilic and salt-tolerant microorganisms, referred to as the “dominant microflora of saline-alkaline lands,” possess unparalleled advantages in rehabilitating and harnessing such lands. Considering that saline–alkaline lands constitute a vital reserve for China's arable land resources, employing halophilic and salt-tolerant microbial agents is of paramount importance in ensuring food security.