The field experiment was conducted in Haofeng base, Jimo District, Qingdao, Shandong Province, China (36.38°N, 120.33°E). The acid field was grown with lettuce (pH 5.5–6.1) and the salinization field was grown with cucumber (pH 7.6–7.9, soluble salt 0.14–0.19%). The lettuce (Lactuca sativa L. planting density 3,300 trees per mu) was transplanted at the four-leave stage on 21st, August 2020, while the cucumber (Cucumis sativus L. planting density 3,500 trees per mu) was seeded on 22nd, June 2020. The quality and production of the plants were added up on the 9th, October 2020, and 20th, August 2020, respectively. The soil samples were also collected at different layers (0–20 and 20–40 cm) at the same time.

The completely randomized block design was applied with five treatments and three triplicates of each. Each block possessed an area of 30 m² (length of 6 m and width of 5 m). The specific application fertilization and strains of each treatment were summarized in Table 1. Regular fertilization was composed of 675 kg/ha. The microbial inoculants were used with a concentration of 600 L/ha. No fertilization in other periods.

Earthworm surveys were conducted using cubes of soil (length 1.0 m, width 2.0 m, and height 0.3 m) excavated manually in the field after harvest, all earthworms were counted.

The A. niger MJ1 was isolated from the from alkaline soil, which had been deposited in CCTCC (China Center for Type Culture Collection) on 5th, January 2015 (CCTCC No: M2015004), routinely solid cultivated with aeration at 30°C in the PDA medium (pH 7.0, Potato 200 g/L, Dextrose 20 g/L, and Agar 20 g/L). The wild-type strain P. stutzeri DSM4166 (deposited in DSMZ: Deutsche Sammlung von Mikroorganismen und Zellkulturen, No: DSM4166) and its derived strain CHA0-nif (deposited in CGMCC: China General Microbiological Culture Collection Center on 31st, July 2017, No: CGMCC 14476) used in this study were routinely cultivated with aeration at 30°C in the KB medium (pH 7.0, K₂HPO₄ 1.5 g/L, MgSO₄·7H₂O 1.5 g/L, peptone 20 g/L, glycerin 10 ml/L). At least 200 million colony-forming units per ml (CFU/ml) of P. stutzeri DSM4166 or CHA0-nif were added to the soil during the sowing of the crop’s seeds. While 5 million CFU/ml of A. niger MJ1 spore powder was applied to the soil.

The major physicochemical properties of collected soil samples, including pH, organic matter (OM), hydrolysis nitrogen (HN), total nitrogen (TN), available phosphorus (AP), available potassium (AK), soluble salt (SS), microbial biomass carbon (MBC), microbial biomass nitrogen (MBN), and microbial biomass phosphorus (MBP) were analyzed with regular methods according to previous reports.

The total weight of the cucumber and lettuce was evaluated after the harvest and converted to the yield per hectare. The weight and height of lettuce heads and the characteristics of the lettuce vanes (length and width) were measured to estimate the lettuce quality. The soluble protein, sugar, and nitrates in lettuce were analyzed according to reported methods (Aasfar et al., 2021). The content of vitamin C was assessed according to Li et al. with a UV-spectrophotometer (Sánchez-Rangel et al., 2013).

DNA extraction was carried out with the E.Z.N.A. soil kit (Omega Bio-Tek, United States) according to the manufacturer’s protocols. After evaluating the purity and concentration of isolated DNA with NanoDrop 2000 (Thermo Fisher, United States), the amplification process was conducted by PCR system. The primer sequences are as follows: 338F 5′-ACTCCTACGGGAGGCAGCAG-3′ and 806r 5′-GGACTACHVGGGTWTCTAAT-3′ for bacterial V3–V4 region of bacterial 16S rRNA gene, ITS1F 5′-CTTGGTCATTTAGAGGAAGTAA-3′ and ITS2R 5′-GCTGCGTTCTTCATCGATGC-3′ for fungal ITS hypervariable regions. The Illumina Miseq sequencing process was performed on an Illumina Miseq platform (Illumina, United States) by Novogene Bioinformatics Technology Co. (Beijing, China) according to the standard protocols. The quality control of raw data was conducted with Trimmomatic and merged by FLASH.

The Illumina sequencing was performed with a MiSeq PE250 sequencer (Illumina, United States). The USEARCH method (Edgar et al., 2011) was applied for the stitching, filtering, deduplication, and clustering of the data. With the help of the UPARSE algorithm, the sequences with similarity of over 97% were classified into the same OTU. Then, the sequences were classified and annotated to establish the taxonomic tree and phylogenetic tree. The alpha- and beta-diversity indexes were calculated with the rarefied OTUs. The alpha-diversity index includes Chao1, species, Shannon, Simpson, Faith, Pielou, and coverage, which were calculated by the R-package in QIIME2. For the further beta-diversity analysis, the principal coordinate analysis (PCoA) was conducted with four distance algorithms.

The obtained data were represented as mean value ± SD and analyzed by SPSS 26.0 software. The difference between groups was evaluated with one-way ANOVA followed by the Duncan post-hoc test. The microbial data were analyzed with the R software (Version 2.15.3) including the following parts. The alpha-diversity was estimated by the Ace and Shannon index and compared with the ANOVA. The beta-diversity was assessed through the principal coordinate analyses based on weighted and unweighted unifrac distance algorithms. The Lefse analysis was performed to identify differential microbial communities or species between different treatments. The correlation between soil physicochemical properties and microbial community was analyzed with RDA analysis. The function analysis was also performed to further predicted the potential functional role of identified microorganisms. PICRUSt is a bioinformatics tool that uses marker genes to predict the functional content of microorganism. In this study, this method was employed to predict the potential functions of each sample based on 16S rRNA sequencing data. We used the KEGG database and performed closed reference OTU picking using the sampled reads against a Greengenes reference taxonomy (Greengenes database Version). The 16S copy number was then normalized, molecular functions were predicted and final data were summarized into KEGG pathways. p < 0.05 indicates statistical significance.

Compared with regular fertilization, the inoculation of the studied strains dramatically influenced the OM, HN, TN, AP, AK, and SS contents in the barrier soil with various depths. The detailed soil physiochemical properties are summarized in Table 2.

Specifically, in the 0–20 cm of the salinized soil with cucumber, the MJ1 significantly improved the content of TN, AP, and AK. Meanwhile, the co-inoculation of CHA0-nif showed a significant promoted effect on TN in comparison with the co-inoculation of DSM4166. After reducing 30% of the regular fertilization, the inoculation of MJ1 with DSM4166 or CHA0-nif increased the soil content of TN, AP, and AK relative to treatment with regular fertilization, and no significant difference was observed between the DSM4166 and CHA0-nif. For the 20–40 cm soil, the contents of HN, TN, AP, and AK were found to elevate in the presence of MJ1 co-inoculated with DSM4166 and CHA0-nif, and the enhanced effect of CHA0-nif was significantly stronger than DSM4166 in the content of HN and AK. The TN in the soil with co-inoculation of MJ1 and DSM4166 or CHA0-nif was significantly higher than the regular fertilization, and the difference between the co-inoculations was insignificant. While MJ1 and DSM4166 could increase the HN content after reducing 30% of fertilization, the co-inoculation with DSM4166 or CHA0-nif after 30%-reducing fertilization showed no significant effects.

In the acid soil grown with lettuce, compared with the regular fertilization, the co-inoculation of MJ1 with CHA0-nif significantly improved OM, HN, TN, AP, and AK in the 0–20 cm of soil, while only the contents of HN and AP were elevated by MJ1 with DSM4166 and the promoted effect of CHA0-nif was stronger than that of DSM4166 in soil AP. After reducing 30% of fertilization, the combination of MJ1 and DSM4166 could improve soil SS, while MJ1 with CHA0-nif was found to improve soil AP. DSM4166 inoculating with MJ1 dramatically enhanced the HN and AP in 20–40 cm soil, while the soil treated with CHA0-nif with MJ1 possessed a higher content of HN, TN, and AK than DSM4166, p < 0.05. For the treatment with 30%-reducing fertilization reduction, the combination of DSM4166 or CHA0-nif with MJ1 could keep the soil content of TN, AP, and AK from reducing, and DSM4166 could improve HN and AK content.

For the yield production of cucumber, all treatments showed significantly enhanced effect even the treatments with a 30% reducing regular fertilization, the treatment of MJ1 and CHA0-nif combination showed the maximum production of 5542.36 ± 14.60 kg/acre. For the content of protein and vitamin C in cucumber, MJ1 combined with DSM4166 or CHA0-nif showed a dramatically promoted effect in the presence of regular fertilization. Interestingly, after reducing 30% fertilization, the treatments of MJ1 and CHA0-nif combination showed the maximum contents of protein (0.737 ± 0.021%) and vitamin C (4.287 ± 0.032 mg/kg) in all treatments (Table 3).

The production of lettuce was increased by the combination of MJ1 and CHA0-nif under regular fertilization, and no significant difference was observed between different strain combinations. The quality of lettuce was assessed from two aspects, including growth indexes and physiological indexes. The lettuce that grew under the fertilization combined with MJ1 and CHA0-nif possessed the highest head weight of 0.269 ± 0.046 kg but was insignificantly different from other inoculated treatments. For the other growth indexes, no significant difference was observed in the lettuce vanes’ width, length, and the height of lettuce heads. Physiologically, the content of vitamin C, nitrate, soluble protein, and crude fiber of lettuce was significantly improved in the presence of MJ1 and CHA0-nif, while the combination of MJ1 with DSM4166 was only found to enhance the content of nitrate, which was insignificantly different from CHA0-nif (Table 3).

In salinized soil, the co-inoculation of MJ1 with DSM4166 or CHA0-nif significantly increased the MBC and MBN in 0–20 cm depth and the MBN in 20–40 cm in the presence of regular fertilization, and there was no significant difference in the effect between DSM4166 and CHA0-nif. In the presence of 70% of regular fertilization, the combination of MJ1 and DSM4166 or CHA0-nif markedly improved the MBN of 20–40 cm soil, and their effects on the MBC, MBN, and MBP in 0–20 cm soil were not obvious (Table 4).

According to the data, the co-inoculation of MJ1 with DSM4166 or CHA0-nif exerted an enhanced effect on the MBP level of 0–20 cm acid soil under regular fertilization. While only MJ1 inoculating with CHA0-nif was found to improve the level of MBC and MBN of 20–40 cm soil, the effect of DSM4166 was insignificant. After reducing regular fertilization, the combination of MJ1 with DSM4166 or CHA0-nif could protect the MBC, MBN, and MBP levels from declining (Table 4).

Among the critical alpha-diversity indexes, there were no significant changes were observed in the bacterial richness and diversity between different groups of salinized soil (Supplementary Figure S1A). For the fungal community, the co-inoculation of MJ1 with DSM4166 or CHA0-nif was found to reduce the number of observed species and Shannon indexes, but the differences were not significant (Supplementary Figure S1B). Although there was no significant effect of inoculated strains was observed on the microbial diversity and richness, an increasing number of earthworms was found in the cucumber-grown soil (Supplementary Figure S2).

While in the acid soil, in the treatments with regular fertilization, the co-inoculation of MJ1 and DSM4166 or CHA0-nif significantly improved the number of bacterial species and the Shannon indexes, indicating the improving bacterial richness and diversity. After reducing 30% of regular fertilization, the combination of MJ1 and CHA0-nif showed a markedly promoted effect on the bacterial richness and diversity, but the effect of DSM4166 was insignificant (Supplementary Figure S1C). For the fungal community, the inoculation of MJ1 with DSM4166 or CHA0-nif slightly elevated the observed species number and Shannon index, but the changes were not of statistical significance (Supplementary Figure S1D).

Based on two distance algorithms, the microbial community structure of soil with different treatments were evaluated (Figure 1). In the bacterial community of salinized soil, the combination of MJ1 with DSM4166 or CHA0-nif under 30%-reducing regular fertilization was found to significantly separate from the regular fertilization group in the results of weighted unifrac (total explanation rate of 56.69%) and unweighted unifrac (total explanation rate of 37.05%; Figure 1A). The fungal community structure of the soil with the MJ1-CHA0-nif combination was found to separate from the regular fertilization from the results of weighted unifrac (total explanation rate of 48.85%) and unweighted unifrac (total explanation rate of 22.86%; Figure 1A).

While the bacterial and fungal community structure of the acid soil showed a stronger response to the inoculated strains. Specifically, both the weighted unifrac (total explanation rate of 60.53%) and unweighted unifrac (total explanation rate of 20.25%) demonstrated the clear separation between the regular fertilization treatment and the other treatments with different co-inoculations. The combination of MJ1 with CHA0-nif under regular fertilization and the combination of MJ1 with DSM4166 in the presence of 70% regular fertilization was found to be more different from the regular fertilization (Figure 1B). Meanwhile, a clear separation was also observed in the fungal community between the fertilization group and the various combinations of MJ1 and DSM4166 or CHA0-nif through the weighted unifrac (total explanation rate of 59.90%) and unweighted unifrac (total explanation rate of 23.69%). MJ1 combined with CHA0-nif was found to mostly affect the fungal community structure under regular or reduced fertilization (Figure 1B).

In the acid soil, two bacterial and three fungal biomarkers were identified between different treatments. Moreover, in the presence of regular fertilization, five fungal biomarkers were screened under the combination of MJ1 with DSM4166 or CHA0-nif. After reducing 30% of fertilization, three bacterial biomarkers were screened (Supplementary Figures S3A, B). For different strain combinations, five bacterial biomarkers and six fungal biomarkers were identified in the presence of MJ1 with DSM4166, while seven fungal biomarkers were identified under the combination of MJ1 with CHA0-nif. No microbial biomarker was observed between DSM4166 and CHA0-nif in the presence of 70% fertilization with MJ1 (Supplementary Figures S4A, B).

In the salinized soil, one bacterial biomarker were screened between different treatments. Additionally, in the presence of regular fertilization, one differential bacterium was identified, while two differential bacteria were observed after a 30% reduction of the fertilization (Figures 2A, B). One fungal biomarker was identified between DSM4166 and CHA0-nif in the presence of regular fertilization with MJ1, and one bacterial biomarker was screened after reducing 30% fertilization (Supplementary Figures S5A, B).

The top 10 most abundant functions of enriched bacteria or fungi in the soil with different treatments were predicted and annotated. It was revealed that the functions of bacteria in both the salinized (Figure 3A) and acid (Figure 3B) soil were mainly associated with metabolism, genetic and environmental information processing, cellular processes, and a small part of human disease and organismal systems. The abundance of involved functions was relatively average between different treatments (Supplementary Figures S6A, B). The functions of the fungi community in both salinized (Figure 3C) and acid (Figure 3D) soil were mainly enriched as saprotrophs, especially the wood saprotroph-correlated function, such as plant and animal pathogen (Supplementary Figures S6C, D).

The PCA analysis on bacterial function enrichment of salinized soil showed that there was a clear separation of fertilization reduction treatments (MJ1 with DSM4166 or CHA0-nif) from the regular fertilization (Supplementary Figure S7A), while the difference in the fungal function was not obvious (Supplementary Figure S7B). Similarly in the acid soil, the bacterial functions of the reduction fertilization with MJ1 combined with DSM4166 or CHA0-nif were obviously separated from other groups (Supplementary Figure S7C). The fungal function under the regular fertilization with MJ1 combined with DSM4166 and the 30% reducing fertilization with MJ1 combined with CHA0-nif were separated from the other treatments (Supplementary Figure S7D).

In the salinized soil, the interpretation rates of RDA1 for the bacterial and fungal communities were 54.51 and 47.58%, respectively, while the interpretation rates of RDA2 were 26.24 and 25.39%, respectively (Figure 4A). pH, HN, TN, and OM were demonstrated as the major affected factors associated with the bacterial and fungal community. The total interpretation rates of RDA were 85.79 and 82.93% for the bacterial and fungal communities in lettuce-grown soil, respectively (Figure 4B). The most correlated factors were also revealed as pH, HN, TN, and OM.

Using the CCA and RDA functions in the vegan package, we can calculate the r² and p values of each environmental factor’s impact on species distribution through the envfit function, and then use the selected environmental factors with significant impact to do CCA and RDA analysis. In the salinized groups, pH was positively correlated with the Bacillus and negatively related to the Caldicoprobacter, Sedimentibacter, Tissierella, Syntrophomonas, and Fastidiosipila genus. The Lactobacillus was positively correlated with OM, while TN and HN showed no significant correlation with abundant bacterial genus (Figure 5A). pH showed a significantly negative correlation with the fungal genus, including Cylindrocarpon, Lophotrichus, and Preussia. The OM-correlated genus involved Macroventuria, Lysurus, and an unidentified Mortierellales genus. The correlation of TN or HN with fungal genera was consistent, where the Pseudogymnoascus, Melanospora, Stagonosporopsis, Trichoderma, and Asepergillus genera were all positively correlated with TN and HN (Figure 5B).

In the acid soil, Pseudoxanthomonas, Arthrobacter, and Massilia genus were positively correlated with soil pH, while the Bryobacter genus showed a significant negative correlation. OM was found to be positively correlated with an unidentified Gemmatimonadaceae genus, and HN was negatively correlated with Gemmatimons genus, which was also negatively correlated with TN as well the Rhodanobacter, Flavisolibacter, and Pseudomonas (Figure 5C). For the fungal genus, pH was found to be positively correlated with the Dixhotomophthora and Plectosphaerella, while OM showed significant correlation with a variety of genera, such as Olpidium, ideriella, Trichocladium, and several unidentified genera. Both TN and HN were negatively correlated with the Pseudogymnoascus genus, and TN also showed a positive correlation with the Microidium genus (Figure 5D).

Due to the diversity of microbial function, the inoculated strains could ameliorate the soil environment and benefit the cycles of nutrients. For example, the inoculation of nitrogen-fixing microorganisms could fix the nitrogen elements in the air to meet the requirement of nitrogen and promote its uptake by plants (Reis and Teixeira, 2015; Dellagi et al., 2020; Aasfar et al., 2021). Phosphorus-solubilizing microorganisms could decompose a variety of organic acids, which can react with insoluble phosphates to convert them into available phosphorus (Alori et al., 2017). Consistently, although the inoculation of MJ1 with DSM4166 or CHA0-nif showed insignificant effect on soil pH, it dramatically elevated the contents of HN, TN, AP, and AK in shallow and deep soil layers.

Under regular fertilization, the combination of MJ1 with CHA0-nif was found to show a stronger enhancement than DSM4166. Additionally, after reducing 30% of regular fertilization the inoculated strains could keep or even elevated the nutrient contents in different soil layers, the difference in the effect of DSM4166 and CHA0-nif was not significant, indicating that the inoculation of MJ1 with DSM4166 or CHA0-nif could help reduce the application of regular fertilization and meet the requirement of carbon-neutral and environment friendly of the current status of agriculture. It is worthy to note that the effect of inoculated strains was distinct in different barrier soil. The combination of MJ1 with CHA0-nif was more effective to improve the content of nitrogen of salinized soil and the content of phosphorus in acid soil. Low availability of phosphorus is the typical barrier of acid soil due to the formation of precipitation with metal ions. Although there were no significant changes in pH of acid soil, the improved availability of soil nutrients also suggested the potential of inoculated strains in barrier soil remediation.

Microorganisms could produce a variety of metabolites, which could regulate plant growth and improve their ability to pest resistance. For example, it was reported that the inoculation of Azospirillum brasilense DSM-1843 could improve the content of chlorophyII and biomass, increase Fe in the cucumber leaves, and therefore promote the recovery of cucumbers from the Fe-deficiency (Pii et al., 2015). The application of microbiological fertilization significantly improved the content of nitrate and vitamin C in Lactuca sativa L. (Stojanovic et al., 2020). The production, protein content, and Vitamin C content of cucumber were promoted by the inoculated strains, meanwhile, the cucumber under 30%-reduction fertilization with MJ1 combining CHA0-nif showed the maximum contents of protein and Vitamin C. Therefore, this combination could benefit the target of “reducing fertilization and increasing production.” Although the effect of inoculated strains on lettuce production was not significant, the contents of Vitamin C, nitrate, soluble protein, and crude fiber of lettuce were dramatically improved by the combination of MJ1 with DSM4166 or CHA0-nif, especially the combination of MJ1 with CHA0-nif under regular or reducing fertilization. The increasing content of nitrate suggested that the fertilization can be reduced over 30% in the presence of inoculated strains, or the engineering bacteria can be established by knockout related regulator genes.

The secretion and metabolites of microorganisms play critical roles in affecting the microbial community (Tyc et al., 2017). Yuan et al. (2017) reported that Bacillus amyloliquefaciens NJN-6 released volatile organic compounds that dramatically affected soil bacterial and fungal communities and decreased the alpha-diversity of soil microbial community. The exogenous strains have also been revealed to affect the microbial community. For instance, in a cucumber-planting soil, the myxobacterium Corallococcus sp. strain EGB could regulate soil microbial community and further control cucumber Fusarium wilt (Ye et al., 2020). The application of the Trichoderma hamatum strain MHT1134 in the soil with continuous cropping obstacles could alleviate the reducing microbial diversity and abundance and improve microbial community structure (Mao and Jiang, 2021).

The application of fertilization could also affect the soil microbial community, and the fertilization amount is a critical factor associated with its effect. A recent study demonstrated that the long-term application of chemical fertilization would dramatically destroy the soil microbial community, which resulted in reduced bacterial diversity and increased soil healthy problems (Xu et al., 2020). Herein, the inoculation of MJ1 with DSM4166 or CHA0-nif was found to improve soil microbial biomass C, N, and P, especially in the deeper soil layers. It was observed that the reduction of regular fertilization did not affect the bacterial and fungal diversity of soil grown with cucumber. The inoculation of MJ1 with DSM4166 or CHA0-nif significantly affected the bacterial community of acid soil, where the inoculated strains improved the richness and diversity of soil bacteria. Although the effects were not significant, the inoculation of MJ1 with DSM4166 or CHA0-nif could also slightly improve fungal diversity and richness of acid soil. Additionally, the protective effect of inoculated strains on soil ecology was revealed in the salinized soil, where an increasing number of earthworms was observed even reducing the application of fertilization. The inoculated strains could keep or improved the microbial diversity after reducing fertilization application.

The weighted and unweighted unifrac distances were employed in the present study to evaluate the microbial community structure of the soil with different treatments, which could represent the kinship and the abundance of different species (Lozupone and Knight, 2005). The effect of exogenous strains on soil microbial beta-diversity has been previously reported. For example, Rhizobium alamii was revealed to improve significantly modified the beta-diversity of water stress soil and improved the tolerance of rapeseed (Tulumello et al., 2021). The inoculation of Pigmentiphaga sp. D-2 dramatically affected the bacterial community structure of an acetamiprid-contaminated area and improved the activity of acetamiprid degradation-related bacteria (Yang et al., 2021). The inoculated trains dramatically changed the microbial community structure of salinized soil, especially in the presence of 30%-reducing fertilization. Similarly, in acid soil, the response of bacterial and fungal community structure to inoculated strains was stronger than the cucumber-grown soil. Both the weighted unifrac and unweighted unifrac results showed significant changes in the bacterial and fungal community structure of lettuce-grown soil, indicating that the inoculation of MJ1 with DSM4166 or CHA0-nif could affect the relationship and abundance of bacteria and fungi.

Interestingly, except for the effect of inoculated strains, it was found that the reduction of regular fertilization significantly influenced the structure of the microbial community in cucumber-grown and lettuce-grown soil. Previously, the application of nitrogen fertilization was illustrated to drive the beta-diversity of soil archaea, bacteria, and fungi, and affected the microbial abundance (Li et al., 2020). The bacterial and fungal beta diversity in the soil was also found to be influenced by the application of biofertilizer (Yang et al., 2022). Therefore, the changes in the beta-diversity of soil bacteria and fungi after reducing fertilization also indicated the response of soil bacteria and fungi to fertilization, and the differential microorganisms might help explain the mechanism.

According to the Lefse analysis, several microbial biomarkers were identified in different treatments. Pyrinomonadaceae and Blastocatellia were enriched in the combination of MJ1 with DSM4166 in salinized soil. A previous study observed the relatively high abundance of these two genera, and Blastocatellia was demonstrated to be sensitive to soil pH, carbon, nitrogen, and drought (Wust et al., 2016; Ivanova et al., 2020; Huber et al., 2022). The enriched fungi were Motiereallaes and Mortierellomycetes, which were reported to be enriched in the shallow soil layers and negatively correlated with nitrogen in the soil (Zhou et al., 2021; Wang et al., 2022). For the different combinations of MJ1 with DSM4166 or CHA0-nif, Mortierellomycetes and plectosphaerellaceae were found to play vital roles in the inoculation of CHA0-nif and DSM4166, respectively, in the presence of regular fertilization. Interestingly, the identified biomarker microorganisms were significantly different between DSM4166 and CHA0-nif combined with MJ1 after reducing fertilization, indicating the different mechanism of the combinations of MJ1 with DSM4166 or CHA0-nif to meet the nutrients requirements of cucumber after reducing fertilization.

In the acid soil, Bacilli was identified as a major biomarker of different treatments, which plays a critical role in the combination of MJ1 with DSM4166 in the presence of regular fertilization or 30%-reducing fertilization. Bacilli have been widely applied as a nitrogen-fixing engineering bacteria and biopesticides due to its outstanding ability to fix nitrogen and antibacterial (Perez-Garcia et al., 2011). While Sordariales and Burkholderiales were enriched in the combination of MJ1 with DSM4166 under regular fertilization and 30%-reducing fertilization, respectively. Although few studies have reported the function of Sordariales and Burkholderiales in soil microbial communities, the function of enriched microorganisms was evaluated in the present study. Similar functions of enriched microorganisms were observed between salinized and acid soil, which were mainly associated with metabolism, genetic information processing, cellular processes, and human disease, and fungi were mainly related to the saprotroph process. The inoculated strains MJ1, DSM4166, and CHA0-nif were able to solubilize phosphorus and fix nitrogen. The different functional role of inoculated strains was considered the major reason for different enriched microorganism. Moreover, the regulatory effects of inoculate strains on correlated genes or procedures also contributed to the response of various functional microorganisms to the inoculations.

A recent investigation focused on the effect of fertilization patterns on soil microbial community revealed that the fertilization patterns could affect the nutrient status, which further affected the soil microbial biomass and microbial community structure (Li et al., 2022). In the present study, pH, OM, HN, and TN were identified to be the major factors correlated with the soil microbial community, which were found to strongly respond to the inoculated strains.

Although no significant changes were observed in the presence of different inoculated strains, pH was revealed to be the major effect factor in the salinized soil bacterial community, which correlated with several genera, including Bacillus. Bacillus was previously demonstrated to be alkaliphilic and possess the ability to solubilize phosphorus (Krulwich et al., 1999; Saeid et al., 2018). In terms of the fugal community, Aspergillus, Pseudogymnoascus, Melanospora, Stagonosporopsis, and Trichoderma were found to be more sensitive to soil properties, which were correlated with HN, TN, AP, AK, and SS. Aspergillus was also found to respond to soil biomass C, N, and P. These identified fungi have also been reported to play various functions in plant growth, including serving as pathogens and promoting crop production (Marin-Felix et al., 2018; Dou et al., 2019; Dong et al., 2021; Urbina et al., 2021). In the inoculated strains, MJ1 is a typical phosphorus-solubilizing bacteria, and the other strains DSM4166 and CHA0-nif were able to fix nitrogen, which improves the nutrient provision for the parasitic fungi. The acid soil properties-correlated bacterial genus was reported to play role in the rhizosphere, and the fungal genus was mainly involved in the plant diseases (Cho et al., 2017; Huq, 2019; Lee et al., 2019; Al-Sadi, 2021), which are consistent with the results of function prediction. Previously, Ren et al. (2021) tried to use organic fertilizer to replace chemical fertilizer and found that the effect of organic fertilizer on soil physicochemical properties influenced the microbial community and directly or indirectly affected the activity of soil enzymes. Therefore, it was speculated that the inoculated strains might affect the microbial community via regulating soil physicochemical properties.

This study focused on the growth and development of cucumber planting on salinized soil and lettuce on acid soil, which lacks negative controls. There have been some intolerant plants of salinized, acid, or other barrier soils identified in previous studies. Therefore, investigating the optimized fertilization or other management strategies would be beneficial and of urgent needs to improve the growth and development of sensitive plants on barrier soil.