Microbial Community Dynamics and Rice Adaptation in Saline–Alkali Soils: Insights into Plant-Microbe Interactions

Abstract

The saline–alkali soil environment profoundly influences the diversity and composition of soil microbial communities, reshaping their ecological network structures. As a vital staple crop, rice (Oryza sativa L.) plays a crucial role in global food security, highlighting the urgent need to improve its cultivation efficiency in saline–alkali soils. However, the mechanisms by which rice roots recruit beneficial microorganisms from native soils under prolonged saline–alkali stress remain largely unclear, and limited research has been conducted on the effectiveness of microbial inoculants in enhancing rice salt tolerance. This study investigated microbial communities in a saline field subjected to over a decade of continuous rice cultivation. Plant growth-promoting microorganisms were isolated and screened from the rhizosphere. The findings revealed long-term salt stress significantly altered microbial diversity and community composition, although the overall microbial network structure remained resilient. A total of 21 plant growth-promoting strains were identified, indicating that rice roots under sustained salt stress selectively recruit beneficial microbes that contribute to plant growth and stress adaptation. Further experimental validation demonstrated that synthetic microbial communities outperformed individual strains in promoting rice seedling growth under high-salinity conditions, likely due to synergistic microbe and microbe–plant interactions. In conclusion, while saline–alkali conditions disrupt native microbial communities, rice exhibits adaptive capacity by selectively enriching growth-promoting microorganisms. The application of synthetic microbial consortia presents a promising strategy to enhance rice resilience and productivity in saline–alkali environments.

1. Introduction

The world is facing escalating challenges driven by rapid population growth, climate change, and the decreasing availability of arable land. According to projections by the United Nations, the global population is expected to reach 9.3 billion by 2050 and 10.1 billion by 2100 [1,2], rendering food security an increasingly critical global concern. Key constraints on the Earth’s carrying capacity include total land area, the extent of arable and forested lands, food production capabilities, and timber supply [3]. More than 412 million hectares of agricultural land worldwide are affected by soil salinization and alkalization [4,5]. Compounding these issues, global water scarcity, intensified soil salinity and alkalinity, and other abiotic stresses contribute significantly to crop productivity declines, with potential yield reductions exceeding 50% under severe stress conditions [6]. To address these constraints, safeguarding existing arable land while expanding the utilization of marginal lands offers a viable strategy to increase cultivable land area and relieve the mounting pressure on global food systems [7]. However, current research indicates that food production from existing arable land, under prevailing consumption and production trends, will be insufficient to meet future global demand. This has been demonstrated through food flow modeling in self-sufficient regions [8,9]. Consequently, promoting adaptive cultivation strategies and developing crop varieties with enhanced stress resilience is imperative [10]. In particular, advancing research on the remediation of saline–alkali soils and the breeding of salt-tolerant crops is both urgent and of considerable practical significance for ensuring long-term global food security.

Oryza sativa is one of the most important staple crops globally, and enhancing its productivity in saline–alkali soils is essential for addressing the decline in arable land and meeting the rising demand for food production [11,12]. In recent decades, considerable progress has been made in improving rice salt tolerance through genetic engineering, including incorporating salt stress-responsive genes. Additionally, molecular marker-assisted breeding has become a pivotal tool for identifying salt-tolerance-related loci and facilitating targeted genetic improvement [13,14,15]. Despite these advancements, the widespread adoption of genetically improved rice varieties remains limited, primarily due to concerns regarding food safety, nutritional quality, environmental risks, and the economic costs associated with genetically modified organisms (GMOs) [16]. Consequently, increasing attention is being directed toward understanding salt tolerance mechanisms from an ecological perspective. Emerging research suggests that rice performance under saline–alkali conditions is influenced by its genetic traits and complex interactions with environmental factors. These ecological dynamics play a critical role in shaping rice adaptability and resilience to salt stress, offering new opportunities for sustainable improvement in salt-affected agroecosystems.

Long-term exposure to saline–alkali soil environments significantly alters the composition and structure of soil microbial communities, highlighting these ecosystems as reservoirs of diverse and unique microbial resources [17,18]. A growing body of research has demonstrated that salt-tolerant plant growth-promoting microorganisms (PGPM) inhabiting saline–alkali soils play a pivotal role in enhancing plant tolerance to salt stress, thereby contributing to improved growth and productivity [19]. For instance, bacterial strains KS8 and KS28, isolated from saline–alkali soils, produce ACC deaminase and have been shown to significantly improve plant height, root and stem biomass, nutrient uptake (P and K), and the K⁺/Na⁺ ratio when inoculated into the rhizosphere of sunflower [20]. Similarly, Bacillus megaterium, isolated from salinized soils along the Mekong River, produces indole-3-acetic acid (IAA) and enhances root development and biomass accumulation in rice when introduced into the rhizosphere [21,22]. Moreover, endophytic bacteria such as Pseudomonas pseudoalcaligenes ST1 and Bacillus pumilus ST2, isolated from rice grown in saline–alkali soils, exhibit antioxidant properties. Their inoculation enhances rice salt tolerance by scavenging reactive oxygen species (ROS) and increasing the accumulation of osmoprotectants in root tissues [23,24]. Despite these advances, relatively few studies have focused on characterizing microbial communities in saline–alkali soils subjected to long-term rice cultivation, and the targeted isolation and screening of potential PGPM from such environments remain limited. Therefore, we hypothesize that long-term salt stress enhances the modularity and functional redundancy of soil microbial networks, driving the formation of more stable interaction architectures that improve microbial community resilience to persistent environmental pressures. Furthermore, plants growing under prolonged saline conditions selectively recruit salt-tolerant, plant-growth-promoting microorganisms to mitigate salt stress. Based on this hypothesis, this study aims to (I) characterize changes in composition, diversity, and network structure of microbial communities in paddy soils under long-term artificial salinity stress; (II) evaluate their associations with key environmental factors; (III) isolate and identify potential plant-growth-promoting microorganisms (PGPM) enhancing rice salt tolerance; and (IV) assess the effects of individual PGPM strains versus synthetic microbial communities on rice salt tolerance through pot experiments.

2. Materials and Methods

2.1. Experimental Site and Material

The saline–alkali experimental field is situated at the Agricultural Science Base of Hainan University, located in Haikou City, Hainan Province, China (20.05° N, 110.32° E) (Figure 1). This area, positioned in the northern region of Hainan Island, experiences a tropical monsoon climate. During the experimental period, the mean annual temperature was 25.1 ± 0.4 °C, with annual precipitation ranging from 1786 ± 62 mm. (https://data.cma.cn) The experimental site was established in 2012 and comprises three distinct treatment conditions: a non-saline control, a moderately saline field with 0.3% salinity, and a highly saline field with 0.5% salinity. The field is primarily utilized for screening rice varieties and evaluating their salt tolerance and quality traits. Rice is cultivated twice annually, and the soil type across the experimental plots is classified as Gleysols. The experimental plots measured 5.10 m × 1.70 m (length × width) in dimension.

2.2. Soil Sample Collection

Soil sampling was conducted in July and December 2023 at three treatment sites: the control field (CK), the 0.3% salinity field (03), and the 0.5% salinity field (05). Samples were collected at two time points, before rice planting (Q) and after rice harvest (H), with all sampling performed in the evening. Before collection, surface vegetation, dead branches, and leaf litter were cleared. The topsoil was removed using a sterile shovel, and soil cores were extracted from a 5–6 cm depth using a soil auger. Sampling locations were randomly selected along an S-shaped transect to ensure spatial representativeness. Collected samples were placed in sterile metal containers, and visible root debris was immediately removed. The soil from each sampling point was thoroughly homogenized, and a representative subsample was obtained using the quartering method. Subsamples were then transferred into sterile, sealed bags and promptly transported to the laboratory. A portion of the soil was air-dried for physicochemical property analysis, while the remaining portion was stored at −80 °C for subsequent DNA extraction.

2.3. Experimental Design

Soil samples were collected in July and December 2023 from experimental fields with three different salinity levels (CK: salinity < 0.1%; 03: 0.3% salinity; 05: 0.5% salinity) at two time points—before rice planting (Q) and after rice harvest (H). One portion was air-dried for physicochemical analysis, while another was stored at −80 °C for analysis of soil microbial diversity. For the pot experiment using a multi-factorial design (salinity + bacteria), 0.5% salinity soil was air-dried, sterilized, and potted. The salt-tolerant rice variety M58 was used and inoculated with 14 functional bacterial strains divided into four groups (Group A: ACC deaminase-producing bacteria; Group L: IAA-producing bacteria; Group N: ammonia-producing bacteria; Group P: phosphate-solubilizing bacteria) along with their mixed consortia (A_MIX, L_MIX, N_MIX, P_MIX, T_MIX), totaling 19 treatments. Bacterial suspensions (OD600 ≈ 0.6) were inoculated one day after rice transplantation, with the control (CK_SALT) receiving only sterile medium. The 25-day experiment maintained soil salinity at 0.5% while monitoring plant growth and physiological parameters, including plant height, biomass, chlorophyll content, and antioxidant enzyme activity, with all experiments conducted in triplicate.

2.4. Measurement of Physiochemical Properties of Soil

Soil chemical properties were analyzed using standardized protocols and precision instrumentation, as described by [25,26] (Table S1). The assessed parameters included soil organic matter (SOM), total nitrogen (TN), alkali-hydrolyzable nitrogen (AHN), ammonium nitrogen (NH₄+-N), nitrate nitrogen (NO₃⁻-N), total potassium (TK), available potassium (AK), total phosphorus (TP), and available phosphorus (AP). The soil pH and electrical conductivity (EC) were determined according to the method of Ye et al. [27]. The soil chloride (Cl⁻) content was determined by silver nitrate titration. Soluble sodium (Na⁺) was measured using a flame photometer. The concentrations of calcium (Ca²⁺) and magnesium (Mg²⁺) ions were analyzed by EDTA complexometric titration. The soil sodium adsorption ratio (SAR) was calculated based on the concentrations of Na⁺, Ca²⁺, and Mg²⁺, where ion concentrations are expressed in mmol/L [28].

Sodium Adsorption Ratio (SAR)

The sodium adsorption ratio (SAR) was calculated from the concentrations of Na⁺, Ca²⁺, and Mg²⁺ in the soil extract using the standard formula:

SAR = Na⁺/√ [(Ca²⁺ + Mg²⁺)/2]

2.5. Soil DNA Extraction, PCR Amplification, and High-Throughput Sequencing

DNA was extracted from approximately 0.50 g of soil per sample (n = 18) using the E.Z.N.A.^® Soil DNA Kit (Omega, Norcross, GA, USA), following the manufacturer’s protocol. The concentration and quality of the extracted DNA were evaluated via 1% agarose gel electrophoresis. The bacterial 16S rRNA gene V1–V3 hypervariable region was amplified using universal primers 27F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 533R (5′-TTACCGCGGCTGCTGGCAC-3′), while the fungal internal transcribed spacer (ITS) region was amplified using primers ITS1F (5′-CTTGGTCATTTAGAGGAAGTAA-3′) and ITS2R (5′-GCTGCGTTCTTCATCGATGC-3′). Polymerase chain reaction (PCR) conditions included an initial denaturation at 95 °C for 3 min, followed by 30 cycles of denaturation at 95 °C for 30 s, annealing at 55 °C for 30 s, and extension at 72 °C for 45 s, with a final extension at 72 °C for 10 min. PCR amplification was conducted using an ABI GeneAmp^® 9700 thermal cycler (Applied Biosystems, Foster, CA, USA). The amplified products were visualized via 2% agarose gel electrophoresis and purified using the AxyPrep™ DNA Gel Extraction Kit (AXYGEN, Union, CA, USA). Purified amplicons were quantified using the QuantiFluor™-ST fluorescence quantification system (Promega, Madison, WI, USA). High-throughput sequencing of the purified amplicons was carried out on the Illumina NextSeq 2000 platform (Illumina, San Diego, CA, USA) by Shanghai Meiji Biomedical Technology Co., Ltd. (Shanghai, China), following the manufacturer’s standard protocols.

2.6. Isolation, Screening, and Functional Identification of Potential PGPM

Root-soil samples were collected from experimental fields with 0.3% and 0.5% salinities during rice growth to isolate potential plant growth-promoting microorganisms (PGPMs). Four culture media (R1, R2, R3, and R4), known for their effectiveness in isolating salt-tolerant microorganisms, were selected for microbial isolation and cultivation [29] (Table S2). The isolation procedure involved adding approximately 0.50 g of rice root soil into a 200 mL Erlenmeyer flask containing 50 mL of sterile water, followed by thorough mixing on a shaker (15 min at 180 rpm). The resulting soil suspension was serially diluted to 10⁻⁵, and the diluted suspension was plated onto the four solid media, then incubated at 29 °C in the dark. Once colonies exhibiting distinct color and morphology appeared, differential colonies were selected and purified through at least two rounds of streaking to ensure single-strain isolation. The isolated strains were then evaluated for plant growth-promoting (PGP) functions, including phosphorus (P) solubilization, ACC deaminase production, IAA production, and ammonia production.

2.6.1. Phosphorus (P) Solubilization

For phosphate-solubilizing activity evaluation, bacterial strains were streaked onto pre-solidified Pikovskaya’s (PVK) agar medium and incubated at 28 °C for 3 days. The formation of clear halo zones around colonies indicated positive phosphate solubilization [30].

2.6.2. ACC Deaminase Activity

Bacterial strains were inoculated into Dworkin-Foster (DF) minimal salt medium containing 2.5 mmol L⁻¹ 1-aminocyclopropane-1-carboxylic acid (ACC) as the sole nitrogen source. Cultures were incubated at 28 °C with shaking at 180 rpm for 72 h. The ability to grow in ACC-supplemented medium, as evidenced by increased OD600 values compared with controls, indicated ACC deaminase activity [31].

2.6.3. IAA Production Capacity

Bacterial strains were inoculated in LB liquid medium supplemented with 1 g L⁻¹ L-tryptophan (50 mL in 250 mL Erlenmeyer flasks) and cultured at 28 ± 2 °C with shaking at 150 rpm for 24 h. For IAA detection, 1 mL of bacterial culture was mixed with an equal volume of Salkowski reagent (1:1, v/v). The mixture was incubated in darkness at 28 °C for 30 min, with the development of a pink color indicating positive IAA production [32].

2.6.4. Ammonia Production

Bacterial strains were cultured in peptone broth at 28 ± 2 °C with 180 rpm agitation for 48 h. Following incubation, 0.5 mL of Nessler’s reagent was added to each culture tube. Indicated by color development from yellow to deep brown, confirmed ammonia production capacity [33], using appropriate functional assay media for PGPMs (Table S3).

Subsequently, the single strains exhibiting PGP functions were identified. The bacterial 16S rRNA V1–V9 hypervariable regions were amplified using universal PCR primers 27F (5′-AGAGTTTGATCCTGGCTCA-3′) and 1492R (5′-GGTTACCTTGTTACGACTT-3′). The PCR conditions were as follows: initial denaturation at 98 °C for 3 min, followed by 40 cycles consisting of denaturation at 98 °C for 10 s, annealing at 57 °C for 10 s, and extension at 72 °C for 30 s, with a final extension at 72 °C for 5 min. PCR products were analyzed via 1% agarose gel electrophoresis, and the bands were visualized using a gel imaging system (Vilber Lourmat QUANTUM-ST5, Eberhardzell, Germany) to assess concentration and quality. PCR products meeting sequencing criteria were stored at low temperatures and sent to Liuhe Huada Company (Guangzhou, China) for sequencing. Sequence data for each strain were obtained and aligned using the MAFFT algorithm for similarity comparison. Sequences with >97% similarity were classified within the same species. Sequences showing similarity <97% were individually compared in the NCBI 16S rRNA database (https://blast.ncbi.nlm.nih.gov/Blast.cgi, accessed on 18 August 2024) to identify the genus of each strain. A phylogenetic tree was then constructed using MEGA11 software (version 12.0.14).

2.7. Assessment of Growth-Promoting Ability of Potential PGPM in Salt-Stressed Rice

Based on phylogenetic diversity (16S rRNA gene sequences showing <97% similarity) and comprehensive evaluation of plant growth-promoting (PGP) traits, 14 distinct bacterial strains were selected from the original collection of 21 isolates for subsequent pot inoculation experiments to assess their PGP potential. The soil was collected from the experimental field with a salinity of 0.5%, air-dried, crushed, and screened to remove stones and debris. It was then bagged, autoclaved to sterilize, and placed into clean pot containers. The salt-tolerant rice variety M58 was used as the experimental material. This variety was independently bred in the laboratory and has undergone metabolomics and transcriptomics analyses, identifying several potential functional genes associated with salt stress adaptation [34]. Initially, the M58 rice seeds were sterilized and germinated in petri dishes. Once the shoots reached approximately 5 cm in length, seedlings with uniform growth were selected for transplanting. The following day, after transplanting, single bacterial and synthetic bacterial suspensions, each with an OD600 value of approximately 0.6, were inoculated into the respective pot soils. An equal volume of sterile liquid medium was added to the non-inoculated salt-treated control group (CK_SALT) and the non-salt-treated control group (CK) to maintain nutrient consistency. The experiment lasted for 25 days, during which soil salinity was monitored daily using a salinity meter to ensure the soil salinity remained approximately 0.5%. After the experimental period, rice plants were harvested for morphological analysis, with measurements taken for plant height (cm), total dry weight (g), and fresh weight (g). Additionally, samples were collected from the middle section of the latest fully expanded leaves and from the roots, approximately 2–3 cm from the root tip, to assess chlorophyll content, malondialdehyde (MDA) levels, and antioxidant enzyme activity. These parameters were used to evaluate the potential effects of different strains on rice salt tolerance and their plant growth-promoting (PGP) functions.

2.7.1. Measurement of Rice Growth Parameters

• Plant height was determined by measuring the vertical distance from the base of the stem to the highest leaf collar using a digital vernier caliper (accuracy ±0.01 cm). Measurements were performed on naturally extended seedlings, with 15 randomly selected plants per treatment group. This standardized protocol ensures precise and reproducible phenotyping data collection [35].
• After 25 days of growth, 15 uniformly growing plants were randomly selected per treatment. Bulk soil was gently removed from roots by rinsing with tap water while maintaining root system integrity. In the laboratory, residual particles were further cleaned using deionized water. After brief surface moisture removal (30 s blotting with absorbent paper), fresh weight was determined using a precision analytical balance, with values recorded to four decimal places (0.0001 g) [36].
• Whole seedlings were rinsed with deionized water, blotted dry with filter paper for 30 s, and oven-dried at 80 °C for 48 h until constant weight was achieved. After cooling in a desiccator for 30 min, samples were weighed using an analytical balance with 0.0001 g precision (n = 15) [37].

2.7.2. Measurement of Rice Physiological Parameters

Chlorophyll content, malondialdehyde (MDA) concentration, and antioxidant enzyme activities (superoxide dismutase, SOD; ascorbate peroxidase, APX; glutathione reductase, GR; and catalase, CAT) were determined using standardized commercial kits (Solarbio Science & Technology Co., Beijing, China https://www.solarbio.com) following the manufacturer’s protocols (Table S4). All spectrophotometric measurements were performed in triplicate with appropriate blank controls using a K5600 ultra-micro spectrophotometer (Kaiao, Beijing, China).

2.8. Bioinformatics Analysis

With the ongoing expansion of microbial databases, traditional Operational Taxonomic Unit (OTU) methods, which rely on a 97% similarity threshold, no longer accurately capture microbial species diversity, potentially leading to the loss of valuable information. To overcome this limitation and improve the resolution of microbial community analysis, this study employed the Amplicon Sequence Variant (ASV) approach, which provides more precise sequence and abundance data [38]. Raw sequencing data were demultiplexed using QIIME 2.2023.x-based Meiji Cloud platform, followed by quality control using fastp (v0.19.6) and sequence assembly with FLASH (v1.2.7). Denoising and optimization of the sequencing data were performed with DADA2 to generate ASV sequences and abundance tables. Bacterial and fungal ASV sequences were annotated using the SILVA 138/16S_bacteria and UNITE 9.0/ITS_fungi databases, with annotation performed using the classify-sklearn algorithm and a confidence threshold of 0.7. Alpha diversity analysis was conducted using mothur-1.30, while beta diversity analysis was performed using the R package “vegan” (version 4.3.0). The LEfSe multilevel species differential abundance analysis was carried out with Python (version 3.13), employing the “All-against-all” comparison strategy. Correlation analyses of environmental factors and interspecies networks were conducted using Python and the R package “stat” with Spearman’s correlation coefficient. Network visualization was performed using Gephi (version 0.10.1). Figures were generated using R-4.3.0 and cnsknowall (https://cnsknowall.com/).

2.9. Statistical Analysis and Data Availability

The experimental data are presented as mean ± standard deviation from three independent biological replicates. Statistical analyses were performed using R software (version 4.3.0). Two-way analysis of variance (ANOVA) was conducted to assess intergroup differences, with post-hoc Tukey’s HSD test and Duncan’s test applied for significance testing (significance threshold: p ≤ 0.05). Data visualization was performed using CNSKnowAll (https://cnsknowall.com/).The raw data generated in this study are publicly available and have been deposited in the Genomic Sequence Archive (GSA) of the National Genomics Data Center (NGDC) under accession number CRA021615. (https://ngdc.cncb.ac.cn/gsa/CRA021615 accessed on 18 August 2024). The utilization rate was calculated using the following formula:

ΔCs = (|Cf − Cck|)/Cck × 100%

where ΔCs represents the fluctuation intensity of nutrients under salt stress gradients (subscript s = 0.3, 0.5 denotes salt concentration). Ci indicates soil nutrient content in salt gradient plots (s = 0.3, 0.5). Cck represents soil nutrient content in normal paddy fields |Cf − Cck| denotes the absolute change in nutrient content (eliminating offset effects of positive/negative values). This metric highlights the magnitude of nutrient changes induced by salt stress, enabling cross-comparison of soil stability across different salinity gradients.

3. Results

3.1. Soil Nutrients

Long-term exposure to salt stress markedly influenced the distribution and utilization efficiency of soil nutrients. Compared with the control (CK), the utilization rates of soil organic matter (SOM), total nitrogen (TN), and alkaline hydrolyzable nitrogen (AHN) were significantly reduced under salinity levels of 0.3% and 0.5%. Specifically, at 0.3% salinity, the respective decreases in utilization rates were 20.37%, 3.56%, and 7.05%, while at 0.5% salinity, the reductions reached 24.09%, 2.94%, and 19.25%. (Table S5). Soils were classified as non-saline (Q_CK, 0.87 ± 0.25 dS/m), slightly saline (Q_03, 4.24 ± 0.29 dS/m), and moderately saline (Q_05, 8.11 ± 0.24 dS/m) based on their electrical conductivity measurements. Salt stress significantly increased soil electrical conductivity (EC) and the concentrations of key ions (Cl⁻, Na⁺, Ca²⁺, Mg²⁺), leading to a higher sodium adsorption ratio (SAR). (Table S6).

3.2. Soil Microbial Diversity and Community Composition

A total of 1,449,448 bacterial sequences and 1,436,673 fungal sequences were obtained from 18 soil samples. Following denoising and quality filtering, 572,048 high-quality bacterial sequences and 1,400,973 high-quality fungal sequences were retained for downstream analysis. These sequences corresponded to 57,864 bacterial amplicon sequence variants (ASVs) and 3149 fungal ASVs. Venn diagram analysis revealed marked differences in ASV distribution among samples, indicating that long-term exposure to saline–alkali conditions significantly altered the composition of soil microbial communities (Figure S1).

The α-diversity of soil microbial communities was assessed using the ACE, Chao1, and Sobs indices for community richness and the Shannon and Simpson indices for community diversity. Results indicated that, relative to the control (CK), microbial communities in saline–alkali soils exhibited significantly higher richness and diversity, reflecting strong microbial adaptation to long-term salt stress. Moreover, significant variations in ACE, Chao1, and Sobs indices were observed between pre- and post-planting samples across all treatment groups, suggesting that the rice cultivation cycle had a notable influence on microbial community richness (Figure 2 and Figure 3).

To assess β-diversity, principal coordinate analysis (PCoA) based on Euclidean distances was conducted. The analysis revealed clear separation among treatment groups, with distinct microbial community compositions corresponding to different salinity conditions. In contrast, replicates within each treatment clustered tightly, indicating strong intra-group consistency (Figure 4).

At the phylum level, the dominant bacterial taxa included Chloroflexi (28.83–42.41%), Proteobacteri, Desulfobacterota (2.47–9.22%), Firmicutes (3.89–14.96%), and Actinobacteriota (4.98–8.45%). Firmicutes were predominant in pre-planting samples, whereas Actinobacteriota were more abundant in Q_03, Q_05, and H_05 samples. Bacteroidota (3.64–4.80%) was notably dominant in Q_CK, H_CK, and H_03. For fungal communities, the predominant phyla included Ascomycota (22.59–55.05%), Basidiomycota (2.87–56.53%), unclassified_k Fungi (4.74–39.33%), and Fungi_phy_Incertae_sedis (3.30–16.08%). Additionally, Rozellomycota (3.39–4.69%) was abundant in Q_CK, H_CK, and H_03, while Mortierellomycota (2.95–4.69%) dominated in Q_05 and H_05 samples. These observations demonstrate that the saline–alkali environment significantly influences the taxonomic composition of both bacterial and fungal communities at the phylum level (Figure 5). Furthermore, Linear Discriminant Analysis Effect Size (LEfSe) identified microbial taxa with significant differences across multiple taxonomic levels, underscoring the substantial impact of prolonged salt stress on soil microbial community structure (Figures S3–S5).

3.3. Correlation-Based Network Analysis of Soil Microbial Communities

To elucidate the interactions among dominant microbial taxa under varying saline–alkali conditions, Spearman correlation analysis was conducted on the top 100 most abundant genera, with a threshold set at an absolute correlation coefficient ≥0.6 and p < 0.05. The resulting microbial co-occurrence networks were visualized using Gephi software (version 0.10.1). The bacterial community network consisted of three major modules, characterized by intricate intra-module associations and moderate inter-module linkages (Figure 6). This modular organization suggests that the core microbial network structure in saline–alkali soils possesses a high degree of stability and resilience under long-term stress conditions. In the Q_CK treatment, the core bacterial taxa primarily included members of Proteobacteria, Planctomycetota, Desulfobacterota, Chloroflexi, Firmicutes, and Cyanobacteria. The Q_03 group exhibited a similar core structure, with dominant taxa belonging to Proteobacteria, Planctomycetota, Chloroflexi, Firmicutes, Bacteroidota, and Actinobacteriota. In Q_05, the network expanded to incorporate Gemmatimonadota, alongside the aforementioned phyla. Post-planting soil samples showed shifts in core taxa composition: in H_CK, the core included Proteobacteria, Verrucomicrobiota, Planctomycetota, Desulfobacterota, Chloroflexi, Bacteroidota, Gemmatimonadota, and Firmicutes. H_03 was dominated by Proteobacteria, Chloroflexi, Actinobacteriota, Planctomycetota, Nitrospirota, Spirochaetota, and Bacteroidota, while H_05 featured Proteobacteria, Chloroflexi, Actinobacteriota, Planctomycetota, Gemmatimonadota, and Desulfobacterota as the primary core taxa. The fungal co-occurrence network comprised 5–6 distinct modules, each characterized by strong intra-module correlations and minimal inter-module connectivity (Figure 7).

3.4. Correlation Analysis Between Soil Microorganisms and Environmental Factors

The relationships between rhizospheric soil physicochemical properties and microbial community composition were examined using Canonical Correspondence Analysis (CCA) (Figure 8). For the bacterial community, the eigenvalues of the first two CCA axes (CCA1 and CCA2) were 9.73% and 7.17%, respectively, while for the fungal community, the corresponding eigenvalues were 13.96% and 9.39%. In the CCA biplots, the length of the arrows represents the strength of the influence of individual soil properties on microbial communities, while the cosine of the angle between arrows reflects the degree of correlation between those properties (smaller angles indicate stronger positive correlations). The analysis revealed that total potassium (TK) and available potassium (AK) were negatively correlated with ammonium nitrogen (NH₄⁺-N) and nitrate nitrogen (NO₃⁻-N). Soil organic matter (SOM) showed a strong association with bacterial communities in the Q_03 and H_05 treatments. For fungal communities, significant correlations were observed between TK, SOM, and microbial composition in the H_03, H_05, and Q_05 groups. To further explore the interactions between microbial taxa and environmental factors, correlation heatmaps were generated for the top 30 bacterial and fungal genera. Among bacteria, Pseudolabrys and Bellilinea exhibited strong positive correlations with SOM content, whereas norank_p__MBNT15 and norank_o__SBR1031 were significantly negatively correlated with NH₄⁺-N levels. Additionally, Bacillus showed a significant negative correlation with total phosphorus (TP), available phosphorus (AP), and AK content. In the fungal community, Limnoperdaceae_gen_Incertae_sedis, Emericellopsis, and Zopfiella were positively correlated with SOM, while unclassified_k__Fungi, unclassified_f__Nectriaceae, and Sesquicillium displayed significant negative correlations with TK. These findings demonstrate that rhizospheric soil physicochemical properties exert significant influences on the structure and function of soil microbial communities.

3.5. Results of the Isolation, Screening, and Functional Identification of PGPM

Through multiple rounds of isolation and screening, a total of 67 candidate plant growth-promoting (PGP) strains were isolated from over 500 microbial colonies. Following taxonomic identification, strains exhibiting greater than 97% sequence similarity were considered redundant and subsequently excluded (Figure 9). This process yielded 21 unique strains with lower sequence similarity, from which 14 representative strains were selected for further functional analysis. These 14 strains were classified into four functional groups based on their growth-promoting traits: Group A (1-aminocyclopropane-1-carboxylate [ACC] deaminase producers), Group L (indole-3-acetic acid [IAA] producers), Group N (ammonia producers), and Group P (phosphorus-solubilizing strains) (Table S7).

3.6. Effects of Potential PGPM on the Growth and Physicochemical Properties of M58

Compared with the salt-stressed control group (CK_SALT), seedlings inoculated with experimental strains developed more extensive root systems and exhibited improved overall growth, although the effects of salt stress remained partially evident.

Inoculation with microbial strains significantly enhanced plant height, total dry weight, and fresh weight relative to CK_SALT. Notably, strains A2, N4, P2, and the synthetic community T_MIX were most effective in promoting plant height; N4, N_MIX, P2, and T_MIX significantly improved total dry weight, while N1, N4, P2, and T_MIX were most effective in increasing fresh weight (Figure 10).

To further explore physiological responses, representative strains from each functional group (A3, L2, N2, P3) and corresponding synthetic microbial communities (A_MIX, L_MIX, N_MIX, P_MIX, T_MIX) were selected for evaluation of physicochemical properties in leaves and roots. Compared with CK_SALT, inoculation with microbial treatments significantly increased leaf chlorophyll content while markedly reducing malondialdehyde (MDA) levels (Figure 11).

The antioxidant enzyme activities were significantly enhanced in all treatments compared with the control, including catalase (CAT), ascorbate peroxidase (APX), glutathione reductase (GR), and superoxide dismutase (SOD) (Figure 12).

A similar trend was observed in the roots, with significantly elevated antioxidant enzyme activities and reduced MDA concentrations. Furthermore, synthetic microbial communities consistently outperformed single-strain treatments in improving physicochemical indicators (Figure 13).

4. Discussion

Soils host highly diverse microbial communities that are essential for maintaining ecosystem functions and soil health [39]. As key components of the soil ecosystem, microorganisms not only drive the cycling and transformation of nutrients but also shape the physical, chemical, and biological properties of soil [40,41]. To sustain dynamic populations, these microorganisms interact with neighboring taxa, plants, animals, and abiotic factors, forming complex, stable, and predictable biological networks that support their survival as communities within soil habitats [42].

4.1. The Impact of Salinity on Soil Microorganisms and Their Adaptive Mechanisms

Our results showed that salt stress significantly increased soil electrical conductivity (EC) and the concentrations of key ions (Cl⁻, Na⁺, Ca²⁺, Mg²⁺), leading to a higher sodium adsorption ratio (SAR) (Table S6). This study analyzed soil nutrient levels and microbial community diversity in both conventional and long-term salinized rice fields. The results showed that prolonged salt stress significantly reduced the efficiency of soil nutrient utilization. Specifically, soil organic matter (SOM), total nitrogen (TN), and alkali-hydrolyzable nitrogen (AHN) markedly decreased under salt stress. These changes suggest that salinity may inhibit microbial metabolic activity, thereby disrupting nutrient cycling processes. These findings are consistent with previous research showing that high-salinity environments suppress microbial enzyme activity through osmotic stress and ion toxicity, ultimately reducing nutrient mineralization and utilization efficiency [43].

Analysis of microbial community dynamics in normal and long-term salinized paddy fields showed that prolonged salinization significantly altered microbial community composition while maintaining or even increasing microbial diversity, accompanied by enhanced structural complexity. This phenomenon can be attributed to microbial adaptive evolution mechanisms. Under high-salinity stress, microorganisms primarily employ two survival strategies: (i) selective membrane proteins regulate ion transport, promoting K⁺ uptake and Na⁺ expulsion to maintain protein structure and functional stability [44,45]; and (ii) accumulation of osmoprotectants such as glycine and betaine to maintain intracellular osmotic balance [46,47]. However, these adaptive strategies require substantial energy expenditure [48], leading to metabolic reorganization in microbial communities, favoring ATP generation over biosynthetic processes—consistent with the “genome streamlining” theory [49]. Furthermore, distinct shifts in microbial community structure were observed between pre- and post-planting periods. Rarefaction curve analysis demonstrated that the curves for both bacterial and fungal communities reached saturation with increasing sequencing depth (Figure S2), indicating that the sequencing effort was sufficient to capture the majority of microbial diversity present in the soil samples.

We observed that microbial communities developed more complex interaction networks through niche differentiation and stress-induced symbiosis (synergy between nitrogen-fixing bacteria and extracellular polysaccharide-producing bacteria), exhibiting higher functional redundancy and energy utilization efficiency. These results indicated that long-term salinization drives the “ecological domestication” of microbial communities [50], fostering resilient functional networks through intricate interspecies interactions and evolutionary adaptation to maintain soil ecosystem functions under stress [51].

Co-occurrence network analysis revealed differences in soil microbial community structure under varying salt stress conditions. Bacterial networks primarily consisted of three modules, whereas fungal networks comprised five to six modules. Co-occurrence network analysis helps elucidate microbial interactions and their potential ecological functions [52], with nodes in different modules likely performing distinct ecological roles [53]. Compared with low- or non-saline conditions, microbial networks under high-salinity stress exhibited higher complexity and connectivity. Specifically, bacterial networks under high salinity displayed higher average degrees (29–37) and connection numbers (1452–1832), indicating more complex community structures. Such highly connected microbial networks may enhance species interactions and niche competition [54], improving community stability and adaptation to saline environments. Additionally, fungal networks under high salinity showed higher modularity (0.894) and shorter average path lengths (1.006), suggesting faster responses to environmental disturbances [55].

In saline–alkali ecosystems, microbial community resilience and environmental adaptability are critical for supporting plant growth [56]. Our results demonstrate that microbial communities under different salt stress conditions may achieve these functions through distinct network structures. For example, in high-salinity bacterial communities, Proteobacteria and Firmicutes emerged as core taxa, potentially enhancing plant salt tolerance by regulating nutrient cycling and ion homeostasis. In fungal communities, the dominance of Ascomycota and Mucoromycota may be linked to their roles in organic matter decomposition and symbiotic relationships [57]. Notably, the proportion of unclassified fungi (unclassified k_Fungi) increased significantly under high salinity, possibly due to the lack of relevant reference sequences in existing databases [58,59]. This suggests the existence of numerous undescribed microbial taxa in saline–alkali environments, whose ecological functions warrant further investigation. Future studies should attempt to culture these unclassified microorganisms, assess their physiological traits, and elucidate their potential roles in saline–alkali ecosystems.

4.2. The Role of the Rhizosphere and Beneficial Microorganisms in Plant Adaptation to Salt Stress

Using four selective media (RIM-1 to RIM-4), this study successfully isolated over 500 colonies from the rhizosphere soil of rice subjected to long-term salt stress. After rigorous purification and functional screening, 16 candidate strains exhibiting significant plant growth-promoting (PGP) traits were obtained. These strains primarily belonged to Bacillus, Priestia, and Halobacillus, which exhibit strong adaptability to saline–alkali environments [60]. Functional characterization revealed that these strains possessed multiple PGP traits, including ACC deaminase production (A1–A4), indole-3-acetic acid (IAA) production (L1–L3), phosphate solubilization (P1–P3), and ammonia production (N1–N4). These traits are key mechanisms for alleviating plant salt stress [61]. Functional assays using various media showed that strains such as Bacillus altitudinis (P2) and Priestia megaterium (P3) exhibited strong phosphate-solubilizing abilities, consistent with previously reported traits of PGP microorganisms (PGPM) in saline–alkali soils [62]. Additionally, IAA-producing strains Shigella flexneri (L1) and Bacillus aerius (L2), as well as ACC deaminase-producing strains Shouchella miscanthi (A2) and Fictibacillus nanhaiensis (A3), further confirmed the presence of microorganisms capable of modulating plant hormone balance and ethylene metabolism in saline environments [63,64].

Phylogenetic analysis revealed significant divergence in the 16S rRNA gene sequences of these functional strains, suggesting they may belong to distinct phylogenetic lineages. This implies that saline–alkali environments may harbor unique microbial diversity [65]. Notably, strains such as Salipaludibacillus neizhouensis (N1) and Metabacillus indicus (N4) showed low similarity to known species, possibly representing novel taxa or possessing specialized adaptive mechanisms [66]. These findings provide new insights into the role of plant-microbe interactions in salt stress adaptation.

Previous studies have demonstrated the critical role of PGPM in enhancing plant stress resistance. For example, inoculation of Bacillus thuringiensis in sage and lavender increased K⁺ content and suppressed stomatal conductance, alleviating drought stress [67,68]. Consistent with these findings, our pot experiments systematically evaluated the effects of different functional PGPM on rice growth under salt stress. The results showed that PGPM inoculation significantly improved rice growth, with the synthetic microbial community (TMIX) exhibiting the most pronounced effects, notably increasing plant height and total dry weight. This aligns with prior studies highlighting the importance of microbial synergy in plant stress resistance [69,70,71].

Physiological assessments revealed that PGPM inoculation significantly enhanced chlorophyll content in rice leaves, particularly with IAA-producing strains (L2) and synthetic microbial community (LMIX). IAA helps mitigate oxidative stress in plants by promoting root growth and inducing the activity of antioxidant enzyme systems, thereby protecting the integrity of the chloroplast membrane and the photosynthetic apparatus. Under stress conditions, it provides a more stable and protected internal environment for the synthesis and stability of chlorophyll [72]. All inoculation treatments significantly reduced malondialdehyde (MDA) levels, indicating PGPM effectively mitigated membrane lipid peroxidation damage caused by salt stress [73,74]. Phosphate-solubilizing (P3) and ammonia-producing (N2) strains induced the highest antioxidant enzyme activities (CAT, SOD, APX, and GR) in rice, expanding on previous studies’ observations of PGPM-enhanced plant antioxidant capacity [75]. In root systems, the synthetic microbial community (TMIX) exhibited optimal performance in boosting antioxidant enzyme activity and reducing MDA content, likely due to synergistic mechanisms such as nutrient coordination and metabolic complementarity [76]. Moreover, different functional strains showed organ-specific effects: phosphate-solubilizing strains significantly enhanced root SOD activity, whereas IAA-producing strains improved leaf physiological indicators. This supports the “functional specificity” hypothesis, wherein distinct PGPM strains exert differential effects on plant organs and tissues [77].

In summary, long-term salt stress exerted significant effects on soil nutrient concentrations and shaped complex microbial network structures, enhancing community stability and functional diversity. These adaptive changes provide valuable microbial resources for saline–alkali soil remediation and crop production. This study successfully isolated multiple PGPM strains with potential applications from the rhizosphere of rice under prolonged salt stress, offering new microbial tools to enhance rice salinity tolerance through diverse PGP mechanisms.

5. Conclusions

This study focuses on paddy soils subjected to long-term salt stress, analyzing soil nutrients, properties, and the biodiversity of soil bacteria and fungi. The results indicate that prolonged salt stress significantly alters soil nutrient content, soil properties, and microbial community composition. To adapt to long-term salt stress environments, paddy soil microorganisms enhance the modularity and functional redundancy of their communities, forming a more resilient and functionally robust network system. This provides new insights into the adaptation mechanisms of paddy soil microorganisms to long-term salt stress environments. Through the isolation, screening, and functional analysis of rice root-associated microorganisms, it was found that rice roots can selectively recruit beneficial microorganisms under salt stress conditions to help the plant adapt to the saline environment. When partially isolated Plant Growth-Promoting Microorganisms (PGPM) were inoculated into rice plants under salt stress, either as single strains or as synthetic microbial communities, the synthetic communities demonstrated more significant improvements in rice growth, chlorophyll content, MDA levels, and antioxidant enzyme activities compared with single-strain inoculations. This suggests that individual microbial strains have limited effects, whereas a complex, interactive, and synergistic network of multiple strains better ensures their own survival and enhances plant salt tolerance. This study enriches the microbial resource library of saline–alkaline paddy soils and provides new strategies for leveraging PGPM to improve crop stress resistance and productivity.