Based on above analysis, the application of lactic acid bacteria in saline-alkali soil significantly decreased the soil pH, and impacted the soil communities and structure.
Metagenomic analysis of the soil identified a total of 26 phyla. The phylum-level soil microbiota composition was highly similar to that reported from a study performed in Indian saline soil (Ahmed et al., 2018) and alkaline soil of the northern Loess Plateau in China (Zeng et al., 2016). Microbial diversity in different saline-alkali land varies greatly due to different formational causes and soil type (Bachran et al., 2019; Garbeva et al., 2004; Shao et al., 2019). The 26 phyla were distributed across one hundred families, with about one-third of unclassified phylotypes. Soil microbiota is known to be highly diverse, with a large number of unidentified or unexplored taxa, making it a challenge to conduct effective metagenomic studies (Daniel, 2005). Thus, large-scale metagenomic sequencing efforts would be necessary to resolve the complexity of the soil microbiome and to provide sufficient data to understand soil microbial community diversity and function (Mocali and Benedetti, 2010).
The soil microbiota diversity in LABC subgroup was significantly higher than other subgroups, suggesting differences in the post treatment responses between cucumber and tomato planted soils, and that the two plant types could have plant-specific modes and activities in carbon allocation to the soil, thus selecting different rhizosphere organisms (Ladygina and Hedlund, 2010). The increase in microbial diversity in LABC treatment might also be directly attributed to the decrease in soil pH, especially after pH drops to 8.7, that helped diversify the microbial community in the stringent alkalized soil environment (Lauber et al., 2009). The composition of soil bacterial community was mainly affected by the soil pH due to the relatively limited growth tolerances of most bacterial taxa (Rousk et al., 2010). Thus, the soil pH was proposed to be the dominant factor in determining the success in reconstructing the soil bacterial community, and both soil microbial composition and functionality were primarily determined by soil properties (Xun et al., 2015).
The soil microbiota of two vegetable crops, as well as subgroups of different treatments, showed subgroup-based clustering patterns in NMDS analyses of soil microbiota. Particularly, the LABC subgroup was enriched in Pseudomonadaceae and Hyphomicrobiaceae; and the Pseudomonadaceae subpopulation comprised the genera Pseudomonas, Azomonas, and Azorhizophilus. The nitrogen fixing bacteria, Pseudomonas, could produce organic acids, especially gluconic acid that could solubilize phosphate (Hayat et al., 2010), and act as root stimulators (Glick, 1995). On the other hand, the order Rhodocyclales was enriched in LABT subgroup. The order consists of a single family, Rhodocyclaceae, with physiologically diverse family members, including the nitrogen fixer Azovibrio and Azoarcus, as well as Thauera and Dechloromonas, degraders of aromatic compounds (Oren, 2014). The LAB treatment enriched different families of nitrogen fixing bacteria in cucumber and tomato.
No apparent difference was found in the diversity of functional metagenomes in soils treated with LAB or water, suggesting that the functions of complex soil ecosystem were redundant (Jia and Whalen, 2020); and it was likely that the soil consortia comprised taxonomically distinct microorganisms, which could encode the same energy-yielding metabolic functional potential and capacity (Louca et al., 2018). However, notably, the relative abundance of some microbial functional genes seemed to vary between plant types.
In the experiment with cucumber, LAB treatment increased the gene abundance of pathways related with environmental information processing, such as membrane transport, phosphonate metabolism, suggesting enhanced microbial activity and diminished environmental adaptation; such changes were probably related with the decrease in environmental stress. Membrane transport proteins responsible for proton-mediated efflux of monovalent cations were overrepresented in “drylands” and were found to be frequently linked to increased bacterial tolerance to alkaline or saline conditions (Delgado-Baquerizo et al., 2018). Moreover, the presence of lactic acid bacteria could dissolve phosphorus salts and enhance the rate of phosphorus salt utilization in plants (Zlotnikov et al., 2013). Additionally, LABC treatment also led to a decrease in at least one type of antibiotic biosynthesis-related genes, which might have been associated with the overall but mild shift in the microbiota composition after applying lactic acid bacteria (particularly Lactiplantibacillus plantarum ) that could produce bacteriocins to inhibit various bacteria (Olasupo, 1996).
In the experiment with tomato, LAB treatment increased signal transduction, including genes related with microbial interaction, two component system, and mineral absorption. Meanwhile, there was a decrease in genes related with carbohydrate metabolism pathway, such as carbon fixation in photosynthetic organisms.
Co-occurrence networks could examine the corelative changes of microbial community, showing LAB treatment increased the microbial network complexity and number of negative correlations of soil microbes. According to finding, soil erosion reduced the multi-functionality of soil, bacterial diversity, and inter-microbial network complexity (Qiu et al., 2021). The viable LAB-treated group also showed an increase in the degree of network subdominant nodes, and the enhanced networks of subdominant species might result in changes in the functional metagenome and metagenomic potential. Most previous studies focused on analyzing the dominant taxa, but a recent study found that relatively low-abundant taxa were indeed major drivers of multi-functionality (Chen et al., 2020). In fact, rare microbes could be the hidden backbone of microbial communities, and they have been increasingly recognized as drivers of key functions in terrestrial ecosystems and in host-associated microbiomes (Jousset et al., 2017). Thus, it was possible that LAB treatment shifted the soil microbial network, which in turn modulated the overall soil microenvironment, soil microbiota and its function.
Soil pH and soil organic matter are the key chemical parameters of soil quality that influence water relations, soil porosity, and gas exchange processes (Garg et al., 2019). Although our experimental plots were randomly arranged, it also was observed that LAB treatment could influence the soil liquid phase ratio. For arid soils, improving the soil moisture might enhance microbial survival in the microenvironment and the revival of dormant microbes. Such changes could have been reflected in the shift in the microbial interactive co-occurrence network, particularly in the LABC group, which showed enhanced inter-microbial interactions of subdominant bacteria.
Soil microbes also play key roles in facilitating soil nutrient cycling and absorption by crops. The levels of most soil nutrients did not show significant changes after LAB treatments, but slightly more total nitrogen was found in LABC subgroup than other subgroups. Such rise might have been related with enhanced growth of nitrogen fixing bacteria after treatment with viable LAB. The enrichment in nitrogen fixing bacteria subpopulation in LABC subgroup was consistently observed in LEfSe analysis.
RDA analysis revealed biomarker bacterial communities in different treatment were associated with soil physiochemical properties. Soil pH as an integrated index of soil conditions, that was negatively related to Rhodocyclaceae (belongs to β-Proteobacteria) and Pseudomonadaceae (belongs to γ-Proteobacteria), and these two OTUs were positively related to total nitrogen in LAB treated saline-alkaline soil. This was in agreement with other’s research that Proteobacteria typically increase with high N availability (Wang et al., 2021). Impact of pH on bacterial OTU is not always positive or negative in different circumstances, because decrease or increase in the abundance of OTUs is the effect of the sum of many environmental factors, both physical, chemical, and biological (Wolińska, 2019). In addition, soil pH showed no significant correlation with diversity, probably it’s means that was not simply linear correlation, because diversity was highest in soils with near neutral pH, while soils with the lowest levels of diversity were found in pH greater than 8 or less than 4.5 (Lauber et al., 2009). Besides, there are many possible reasons why soils harbor different levels of bacterial diversity, such as salinities, plant-carbon inputs, or moisture and temperature conditions (Fierer et al., 2012).
Although sterilized LAB preparation decreased the soil pH, it did not effectively modulate the soil microbiota compared with viable LAB treatment, suggesting not only the metabolites of lactic acid bacteria play an important role in improving soil quality, more importantly the living lactic acid bacteria as PPM take part in ecology regulation.
One shortcoming of this study was that the soil conductivity was not measured, and the effect of LAB treatment on soil salinity could not be determined. Moreover, cucumber and tomato were not planted at the same time point, and whether the mild differences in weather and soil environment caused by the time point gaps had exerted any impact on the experiment was uncertain.