Daihai Lake (112°10′–112°59′E, 40°11′–40°55′N) is located in Liangcheng County, Ulanqab City, Inner Mongolia Autonomous Region, China. It is surrounded by mountains and is a typical closed inland saltwater lake. The annual rainfall in the basin is 350–450 mm. The main methods by which water is supplied to the lake are precipitation, surface runoff, and underground recharge, which provide insufficient water replenishment, resulting in increased salinity and continuous deterioration of water quality in Daihai Lake. The water surface area of the lake has shrunk sharply. The water surface area of Daihai Lake dropped from 115.94 km² in 1989 to only 48.3 km² in 2020 (Pang et al., 2021; Zhang et al., 2021). Daihai Lake is the third largest inland lake in Inner Mongolia. Together with Hulun Lake and Lake Ulansuhai, these water bodies are collectively referred to as “One Lake, Two Seas” and have important roles in climate regulation, ecological restoration, and water conservation, and also form an important part of China’s northern ecological security barrier (Yang 2020; Pang et al., 2021; Zhang et al., 2021).

In this study, sediments in Daihai Lake were collected in August 2020 according to the lake pollution characteristics to investigate the microbial population structure. We used a grab bucket to collect surface sediment samples. Furthermore, three parallel samples were taken from each point, mixed thoroughly, and stored in a sterilized polyethylene ziplock bag to ensure a sterile environment. Five sampling points (DH1, DH2, DH3, DH4, and DH5) were deployed from different locations of the lake (Figure 1). The collected samples were divided into two parts. One part was stored in dry ice and sent to Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China) for metagenomics testing to analyze the microbial community characteristics and diversity. The other part of the sample was naturally air-dried, ground into a powder, and passed through a 100-mesh sieve for laboratory testing of various physicochemical indexes.

The physicochemical indexes of sediments included pH, electric conductivity (EC), salinity, total nitrogen (TN), total phosphorus (TP), ammonia nitrogen (NH₄ ⁺-N), and nitrate nitrogen (NO₃ ⁻-N). The specific detection methods and detection limits are shown in Table 1. All the experiments were performed in triplicate. Determination of pH: Water was used as extractant, the ratio of water to soil was 2.5:1, the temperature of the sample was controlled at (25 ± 1) °C, and the difference between the temperature of the sample and the temperature of the standard buffer solution should not exceed 2°C. The indicator electrode and the reference electrode (or pH composite electrode) were immersed in the soil suspension. The electrode probe was immersed in 1/3–2/3 of the vertical depth of the suspension under the liquid surface, and the sample was gently shaken. After the reading was stable, the pH value was recorded.

Determination of EC: Natural air-dried soil samples were taken, water was added in a ratio of 1:5 (m/V), and shocked at 20 ± 1°C. The electrical conductivity of the extract was measured at 25 ± 1°C. When two electrodes are inserted into the extracting solution, the resistance between the two electrodes can be measured. When the temperature is constant, the resistance value R is inversely proportional to the conductivity K, namely R = Q/K. When the conductivity cell constant Q is known, the conductivity can be obtained by measuring the resistance of the extracting solution.

Salinity determination: The ratio of water to soil was 5:1. The sample with dry weight of 6 g was placed in a 50 ml centrifuge tube, and 30 ml of ultrapure water was added. After shaking for 3 min, the supernatant was taken after centrifugation, filtered with a filter membrane with a pore size of 0.45 μm, and the salinity of the soil was determined by a salinometer.

Determination of TN: The principle is to take the sample 0.2000g–1.0000 g (about 1 mg nitrogen) into the Kjeldahl nitrogen digestion bottle, under the action of sodium thiosulfate, concentrated sulfuric acid, perchloric acid and catalyst, all converted into ammonium nitrogen by redox reaction. The ammonia distilled from the alkaline solution after digestion was absorbed by boric acid, and titrated with standard hydrochloric acid solution (c = 0.01 mol/L). The total nitrogen content in soil was calculated according to the amount of standard hydrochloric acid solution.

Determination of TP: The principle is that the phosphorus-containing minerals and organic phosphorus compounds in the sample are all converted into soluble orthophosphate by sodium hydroxide melting, and react with molybdenum antimony anti-colorant to form phosphomolybdate blue under acidic conditions. Take 10.0 ml in a 50 ml stopper colorimetric tube, then add two to three drops of 2,4-dinitrophenol indicator, and then adjust the pH value to about 4.4 with 1 + 1 sulfuric acid solution and 2 mol/L sodium hydroxide solution to make the solution yellowish, and then add 1.00 ml ascorbic acid and mix well. After 30 s, 2.0 ml molybdate solution was added and placed at 20–30°C for 15 min. The absorbance was measured at a wavelength of 700 nm.

Determination of NH₄ ⁺-N: 10.0 ml of the sample was taken and put into a 100 ml stopper colorimetric tube, 40 ml of sodium nitroprusside-phenol chromogenic agent was added to it, fully mixed, and stood for 15 min, then 1.00 ml of sodium dichloroisocyanurate chromogenic agent was added to it, fully mixed, and stood for at least 5 h at 15–35°C. Based on the principle that under alkaline conditions, the ammonia ion in the potassium chloride extract reacts with phenol in the presence of hypochlorite ion to produce blue indophenol dye, it has the maximum absorption at 630 nm wavelength.

Determination of NO₃ ⁻-N: Take 1.00 ml of the sample to the reduction column, add 10 ml of ammonium chloride buffer solution, then open the piston, pass through the reduction column at a flow rate of 1 ml/min, and collect the eluent with a 50 ml stopper colorimetric tube. When the liquid level reaches the top of the cotton, 20 ml ammonium chloride buffer solution is added to collect all the effluent and remove the colorimetric tube. Finally, the reduction column was cleaned with 10 ml ammonium chloride buffer solution. Add 0.2 ml chromogenic agent to the above colorimetric tube, fully mixed, and stand at room temperature for 60 min to 90 min. The absorbance was measured at 543 nm wavelength with water as reference.

Sample DNA extraction was performed using the E. Z.N.A.^® Soil DNA Kit (Omega Bio-tek, United States). After DNA extraction, DNA concentration and purity were determined, and DNA integrity was examined using 1% agarose gel electrophoresis. DNA was fragmented by using a Covaris M220 ultrasonicator (Gene Corporation, China). Fragments of about 400 bp were screened, and PE libraries were constructed using the NEXTFLEX Rapid DNA-Seq Library Prep kit (PerkinElmer, United States). Major Bio-Pharm Technology Co., Ltd. (Shanghai, China) was commissioned to perform sequencing, statistics, and data quality control on the samples using the Illumina sequencing platform. After quality control, the optimized sequences (clean data) were single-spliced and assembled using the software MEGAHIT (Li et al., 2015) (https://github.com/voutcn/megahit, version 1.1.2), and contigs ≥300 bp were selected as the final assembly results among the spliced results. The open reading frames of contigs in the splicing results were predicted using Prodigal (Hyatt et al., 2010) and MetaGene (Noguchi et al., 2006) (http://metagene.cb.k.u-tokyo.ac.jp/), and genes with nucleic acid lengths greater than or equal to 100 bp were selected and translated into amino acid sequences. The amino acid sequences of the non-redundant gene sets were compared with the NR database (ftp://ftp.ncbi.nlm.nih.gov/blast/db/) using Diamond (Buchfink et al., 2015) (http://www.diamondsearch.org/index.php, version 0.8.35), and species annotations were obtained from the taxonomic information database corresponding to the NR database. The sum of gene abundances corresponding to the species was used to calculate the species abundance, which was used to obtain the microbial community composition in sediment samples and their diversity. Finally, the raw data sequences were deposited in the NCBI GenBank Sequence Read Archive (SRA) under accession numbers: PRJNA877385.

The raw data were preprocessed by Excel (2019) (Microsoft, United States) and drew with OriginPro 2021 (Origin Lab, United States) and ArcGIS Desktop 10.8 (Environment System Research Institute, United States). Data was checked for normality and skewed data was log transformed prior to analysis. Statistical comparisons of differences among different sampling sites in the three groups of repeated experiments were analyzed by One-way ANOVA (Analysis of variance) and post-hoc tests (Tukey’s HSD) using SPSS 26.0 software (IBM, United States). And SPSS 26.0 software (IBM, United States) was also used to perform Spearman correlation analysis to analyze the correlation between environmental factors and species as well as functional genes. The significance was set at p < 0.05. Redundancy analysis (RDA) was performed using Canoco 5.0 and Monte Carlo tests were used to analyze the significant differences between sediment parameters and microbial communities. Metabolic pathway map download from KEGG database.

There were differences in the sediment nutrient salt levels at different sampling points in the Daihai basin (Figure 2), which is consistent with the results of some studies that have shown that physicochemical factors such as pH, salinity, and N/P ratios in different environments can influence the distribution and composition of different microbial communities (Wu 2019; Peng et al., 2021; Zhang 2021). The pH of the entire lake ranged from 8.95 to 9.25 and was overall alkaline, with the highest alkalinity at point DH5 in the lake center. The salinity ranged from 14.11 to 20.32 g/kg, with DH4 having the highest salinity content and DH2 the lowest. The EC of the whole lake was in the range of 3490–6730 μs/cm, and the EC of DH3, DH4, and DH5 was significantly higher than that of DH1 and DH2. The TN content ranged from 1,632.18 to 2,975.64 mg/kg, with DH5 and DH2 being slightly higher than the other sampling sites, with significant differences between sites (p < 0.05). The TP levels ranged from 786.28 to 998.32 mg/kg, which were lower than the TN levels, and the DH5, DH2, and DH4 levels were significantly higher than the DH1 and DH3 levels (p < 0.01). The NH₄ ⁺-N and NO₃ ⁻-N levels were generally low in the sediment, both of which were highest in DH4 and lowest in DH3. The changing trend of the N/P ratio at each point was very obvious (p < 0.05). The ratio of DH4 was the lowest, and the ratio of DH5 was the highest. The N/P ratio of DH1, DH2, and DH3 was consistent. Overall, the present findings confirm there was some spatial variation in the physical and chemical indicators in the horizontal direction. The effect of pH on the sediments of the five sample sites in the Daihai basin was generally consistent, suggesting that the response to pH changes were available throughout the lake, and also reflected the important role of pH in the management of the water environment in the Daihai basin. The remaining physicochemical factors showed variability across geographic locations, and therefore the effect on microorganisms varied between sample sites in the Daihai sediments. A long-standing study on the diversity and structure of microbial communities in lakes in different regions found that environmental factors have an impact on microorganisms and lake ecology, with salinity being an important environmental factor in determining the distribution of lake microbial communities, with pH and the N/P ratio also playing an important role in influencing the characteristics of the lake microbial communities (Jackson and Vallaire 2009; Yang et al., 2020; Huang 2021).

Species annotation statistics of metagenome sequences indicated that the sediment microbial communities in Daihai Lake had a rich species composition. In terms of microbial composition, all samples were dominated by Bacteria, with relative abundances ranging from 62.5% to 76.4%, followed by Archaea, with relative abundances ranging from 7.0% to 10.0%, significantly lower than those of Bacteria (Figure 3A). During the research process, the microbial composition at the phylum and genus levels was regular and representative across samples. Therefore, the top 10 microorganisms at the phylum level in terms of relative abundance were selected to characterize the community composition of microorganisms involved in the process of phosphorus metabolism (Figure 3B). As shown in Figure 4, the main metabolic pathways involved in phosphorus metabolism were identified as the pentose phosphate pathway (ko00030), inositol phosphate metabolism pathway (ko00562), glycerophospholipid metabolism pathway (ko00564), phosphonate and phosphinate metabolism pathway (ko00440), and oxidative phosphorylation pathway (ko00190). The community composition of microorganisms involved in phosphorus metabolism in all of the samples at the phylum level consisted mainly of Proteobacteria, Actinobacteria, Bacteroidetes, Chloroflexi, Firmicutes, and Euryarchaeota, with the remaining bacterial phyla classified as “Others.” Proteobacteria was the most dominant phylum of the microbial community. The ko00030 and ko00562 pathways, which are the basic metabolic pathways for phosphate, also included Cyanobacteria, Nitrospirae, and Planctomycetes as the most dominant phyla in these pathways. The remaining pathways for the synthesis of cellular substances also had their own unique composition of species that provided most of the activity in that pathway, also referred to as “key species” for a given metabolic pathway (Banerjee et al., 2018). For example, ko00564, which is involved in lipid metabolism, also included Cyanobacteria, Nitrospirae, and Verrucomicrobia; ko00190, which is involved in energy metabolism, also included Spirochaetes and Planctomycetes; and ko00440, which is involved in other amino acid metabolism, also included Nitrospirae, Planctomycetes, and Nitrospinae. In summary, Proteobacteria, Chloroflexi, Firmicutes, Actinobacteria, and Bacteroidetes were the most important phyla in phosphorus metabolism, which is consistent with the results of other domestic and international studies (Cao et al., 2016; Yang et al., 2020; Zhou et al., 2020; Zhang et al., 2022; Zhang et al., 2021b). In general, the relative abundances of the dominant phyla in the phosphorus metabolic process were relatively uniform among the different sampling points, and only the ko00440 pathway showed a large spatial variation.

Based on the results of metagenome species annotation, the bacterial community composition at the genus level was mapped for the top 15 genera belonging to Proteobacteria, Chloroflexi, Actinobacteria, Nitrospirae, Gemmatimonadetes, Verrucomicrobia, and Firmicutes in the five phosphorus metabolism-related pathways (Figure 3C). The results showed that the ko00030 and ko00562 pathways contributed the most to phosphorus metabolism and also had a consistent species composition, with unclassified Gammaproteobacteria, unclassified Deltaproteobacteria, Thiobacillus, and unclassified Bacteria as the dominant genera at each point in the pathway, while unclassified Anaerolineae and Caldilinea had significant spatial variability at different sample sites. For the metabolic pathways ko00564 and ko00190, the different sampling sites had a high similarity in species composition and the same dominant genera, such as Thiobacillus and unclassified Gammaproteobacteria; however, the variability in the spatial distribution according to sampling sites of the ko00190 pathway as a whole was more significant compared to the ko00440 pathway. In contrast, the ko00440 metabolic pathway was represented by low levels of different genera of bacteria at each sampling site and therefore generally did not contribute to the dominant bacteria. Overall, the most dominant genus in the entire phosphorus metabolic pathway was Thiobacillus, and unclassified Deltaproteobacteria, Caldilinea, and unclassified Anaerolineae also influenced the phosphorus metabolic process.

Phosphorus metabolic processes are mainly divided into two parts: phosphate metabolism and cellular material synthesis (Figure 4). Functional genes in the ko00030, ko00562, and ko00440 pathways involved in phosphate metabolism encode proteins involved in carbohydrate metabolism and amino acid metabolism, while the ko00190 and ko00564 pathways involved in cellular substance synthesis belong to the functional classification of energy metabolism and lipid metabolism. Among these, energy metabolism is central to microbial metabolism and therefore the ko00190 pathway played an important role in phosphorus metabolism. Studies have shown that there are multiple enzymes and genes involved in phosphorus metabolism. The functional genes predicted in the metagenome analysis were compared with the KEGG database, and the functional genes that were more closely involved in the process of phosphorus cycling in the Daihai sediments are shown in Table 2. Combined with Figure 3D, the abundance of functional genes related to phosphorus metabolism in the Daihai sediments, its functional genes control and synthesize phytase; alkaline phosphatase and acid phosphatase; polyphosphate kinase and exopolyphosphatase, which are involved in phytate hydrolysis; phosphate hydrolysis; polyphosphate synthesis; and exopolyphosphatase (Zhou et al., 2020).

Some responses between physicochemical indicators and dominant bacterial communities existed in the surface sediments of the Daihai basin (Figure 5), but the effects were not consistent across metabolic pathways and sample sites. With the help of Monte Carlo tests (Table 3), results demonstrate that EC was the main factor affecting bacterial activity in the related pathways of phosphorus metabolism, and related studies have shown that EC was also an important index to measure salinity in sediments in previous research. Therefore, under the strong influence of EC, we speculate that salinity affects the composition and structure of the phosphorus metabolic microbial community (McKinney et al., 2019; Joo H W et al., 2021; de Santana et al., 2022). Although salinity did not directly show the strong influence degree in the data results, salinity, as an important environmental factor index, is still a hot issue worthy of discussion under the positive influence of EC. Similarly, N/P ratio and pH also play different roles in the different metabolic pathways, and the contribution of pH is slightly greater than that of N/P ratio (Luo et al., 2020; Zhu et al., 2021; Zhang et al., 2022; Jackson et al., 2009). The results of the analysis showed that the content of the microbial community at the different sample sites showed spatial and temporal variability, and there was no clear similarity between the sites. Overall, EC, pH, salinity, and the N/P ratio at different sites had an influence on the phosphorus metabolic processes, but the ko00030 and ko00562 pathways had more significant effects based on their relative abundance. In the major pathways ko00030, ko00562, and ko00190, salinity all showed a positive correlation with Bacteroidetes, but negatively with Euryarchaea, where the ko00190 pathway also showed a positive correlation with Actinobacteria, Chloroflexi, and Spirochaetes. Under the influence of EC, the effect of pH and N/P ratio on each dominant phylum was less than that of salinity. And pH showed a positive correlation with Bacteroidetes in each pathway, while N/P ratio showed a negative correlation only in the two pathways ko00030 and ko00562 with the highest contribution. In each pathway, salinity was negatively correlated with N/P ratio uniformity, but positively correlated with pH in the ko00190 pathway. In the secondary pathways ko00564 and ko00440, salinity and pH were positively correlated with Bacteroidetes, and only N/P ratio was negatively correlated with Bacteroidetes in the ko00564 pathway. The effect of salinity on microbial community was greater than that of pH and N/P ratio. At this time, salinity was negatively correlated with N/P ratio and positively correlated with pH. According to the relative abundance figures of each pathway species, the metabolic function of ko00440 had the lowest response to the whole phosphorus metabolic process.

As an important part of the ecological environment, the structural characteristics of the dominant bacterial flora and the changes in physicochemical indicators are very closely related (Jackson et al., 2009; Yu et al., 2022). Figure 6 reveals the relationship between species and environmental factors during phosphorus metabolism in Daihai sediments at the genus level. In the ko00030 and ko00562 pathways, Thioalkalivibrio, Desulfococcus, and unclassified Dehalococcoidia were significantly correlated with salinity (p < 0.05), and unclassified Actinobacteria showed a significant negative correlation with the N/P. In the ko00190 pathway, N/P ratio had a significantly negative correlation with Candidatus Actinomarina and Ardenticatena, and had significantly positive correlation with Desulfococcus (p < 0.01), and in ko00564 pathway, Desulfococcus had significantly positive correlation with N/P ratio. At the same time, EC showed a significant negative correlation with Anaerolinea. The unclassified Actinobacteria in the two pathways were significantly negatively correlated with N/P ratio, and only Thioalkalivibrio was significantly negatively correlated with pH in the ko00190 pathway. In the ko00440 pathway with the lowest response to phosphorus metabolism, salinity played the greatest role, which was significantly negatively correlated with Pelobacter, while Hydrogenophaga and Pelobacter were significantly positively correlated with salinity and N/P ratio, respectively. Desulfococcus was significantly negatively correlated with salinity. Due to the significant effects of N/P ratio and salinity on genus level bacteria during the whole phosphorus metabolism process, it can be speculated that the two have an irreplaceable role in the lake ecosystem. In the related pathways of phosphorus metabolism, N/P ratio and salinity had obvious effects on genus level bacteria, and only pH was observed in the ko00190 pathway.

Therefore, the magnitude of the influence of environmental factors on the diversity of the bacterial community characteristics during phosphorus metabolism in Daihai sediments was ranked as salinity > N/P ratio > pH. Numerous studies have shown that salinity is the main factor controlling microbial communities in saline lakes, as the energy metabolic processes of microorganisms in saline environments are limited by the osmotic pressure of salinity (Xie 2012; Yang 2015; Yang et al., 2020; Yu et al., 2022). In Daihai sediments, salinity had a positive correlation with most bacterial communities, and changes in salinity led to changes in the microbial community composition as well as a loss of biodiversity, which affected the biochemical cycling of carbon, nitrogen, and phosphorus in the ecosystem, leading to an imbalance in the water ecological structure of Daihai Lake.

It is notable that functional genes annotated in sediment phosphorus metabolism may play an important role in the survival of microorganisms in lake bottom sediments (Zhou et al., 2020). For example, phytic acid is an organophosphate compound widely found in plant roots. Plant decay in the water column enters the sediment, and therefore the sediment also contains phytic acid, which is a “reservoir” of phosphorus. However, the phosphorus cannot be used directly by algae and instead must be converted into orthophosphate through a series of hydrolyses, thus generating an increase in bioavailable phosphorus in the water column (Feng et al., 2016; Feng et al., 2018; Zhou et al., 2020). The metagenome annotations results indicated the low levels of phytate hydrolases in the Daihai sediments, and reflected the serious ecological degradation in the Daihai basin. These problems are mainly due to the frequent occurrence of diseases that have led to fish species decline or even disappearance, and the continuous reduction of benthic organisms and submerged plants, resulting in an imbalance in the lake water ecosystem, which intrinsically operates at a low level year round. It is speculated that this may be related to the granulation of phosphorus and the morphological characteristics of phosphorus, which to a certain extent limits the N/P ratio of the lake and thus directly affects the survival of algae in Daihai Lake.

The PHO, aphA, phoN, phoA, phoB, and phoD genes encoding phosphate hydrolases were predicted in the samples, but the relative abundances of genes encoding for phosphate hydrolysis functions were higher for alkaline phosphatases than for acid phosphatases in the Daihai sediments. The ppk gene encoding polyphosphate kinase and the PPX1 and ppx genes encoding epithelial polyphosphatase were predicted in all five sediment samples. The relative abundance of the ppk gene was higher than that of ppx, and the activity of polyphosphate kinase was slightly higher than that of epithelial polyphosphatase at different sampling sites. ppk and ppx encode for key enzymes in the respective synthesis and decomposition of polyphosphate, and their activities are directly related to the ability to perform phosphorus removal, anaerobic phosphorus release, and aerobic phosphate uptake (Peng et al., 2021; Xiao et al., 2016; Zhao et al., 2011; Zheng et al., 2013). According to the results of the KEGG database, the functional potential of genes for phosphorus cycling in different lake areas differed slightly, and reference to previous studies revealed that elemental nitrogen and phosphate were closely related to the taxonomic and functional changes of microorganisms (Luo et al., 2020; Nelson et al., 2016; Wu et al., 2022; Song et al., 2021). In the entire phosphorus metabolic process, only two pathways, ko00030 and ko00562, could be annotated nitrogen metabolism-relevant genes (Figure 3D). There were 18 types of genes common to both pathways, and the represented enzymes were all oxidoreductases. The ko00562 pathway had seven nitrogen metabolism genes specific to its pathway, also annotated as oxidoreductases. This further demonstrates that the biogeochemical cycling of sediment phosphorus is closely related to the nitrogen cycle. The results showed that the key functional genes of the nitrogen cycling process, including the nif genes of the nitrogen fixation pathway (nif D/H/K), the amo A/B/C genes of the nitrification pathway, the nir and nor genes of the denitrification pathway, and the nrfA gene of the anaerobic ammonia oxidation process, all showed significant spatial and temporal abundance differences in the Daihai sediments. Studies have shown that narG, nirS, and norB are the main functional genes closely associated with the sediment nitrogen cycle (Nelson et al., 2016; Kuypers et al., 2018; Wu et al., 2022; Song et al., 2021). narG encodes a nitrate reductase [EC: 1.7.5.1 1.7.99.4] that catalyzes the reduction of NO₃ ⁻ to NO₂ ⁻ by denitrification; nirS encodes a cytochrome-containing cd1 nitrite reductase [EC: 1.7.2.1 1.7.99.1] that catalyzes the reduction of NO₂ ⁻ to NO through denitrification; and the NO reductase [EC: 1.7.2.5] encoded by norB also catalyzes the reduction of NO to N₂O through denitrification. Thus, functional genes for denitrification in microorganisms play an important role in phosphorus metabolism. Similarly, among the nitrogen metabolism genes specific to the ko00562 pathway, only the nasA gene was present in high abundance, while the remaining nirA, pmoC-amoC, pmoA-amoA, pmoB-amoB, NR, and narB genes had very low abundance, further concluding that these six specific genes have minimal influence on the process of phosphorus metabolism.

Based on the main functional genes of nitrogen metabolism, narG, nirS, and norB, a Spearman correlation analysis was conducted between the three genes and related functional genes of phosphorus metabolism, including the KEGG functional orthologs K01083, K01093, K01078, K01077, K01113, K00937, K01514, and K01524, and the relationships between these were analyzed (Table 4). There was a very significant (p < 0.01) positive correlation between K01093 and norB and between K01113 and nirS, while there was a significant (p < 0.05) positive correlation between K01524, K01078, and K00937 and norB. The denitrifying genes nirS and norB had a strong influence on phosphorus metabolism, and norB had a specific influence on K01083, a phytic acid hydrolase. Additionally, norB played a role in the hydrolysis of phosphoric acid via an acid phosphatase gene (K01078) and the synthesis and decomposition of polyphosphate by K01524 and K00937, while nirS had a significant influence on the promotion of phosphoric acid hydrolysis via an alkaline phosphatase gene (K01113).