Effect of ammonium stress on phosphorus solubilization of a novel marine mangrove microorganism Bacillus aryabhattai NM1-A2 as revealed by integrated omics analysis

Background Phosphorus is one of the essential nutrients for plant growth. Phosphate-solubilizing microorganisms (PSMs) can alleviate available P deficiency and enhance plant growth in an eco-friendly way. Although ammonium toxicity is widespread, there is little understanding about the effect of ammonium stress on phosphorus solubilization (PS) of PSMs. Results In this study, seven PSMs were isolated from mangrove sediments. The soluble phosphate concentration in culture supernatant of Bacillus aryabhattai NM1-A2 reached a maximum of 196.96 mg/L at 250 mM (NH4)2SO4. Whole-genome analysis showed that B. aryabhattai NM1-A2 contained various genes related to ammonium transporter (amt), ammonium assimilation (i.e., gdhA, gltB, and gltD), organic acid synthesis (i.e., ackA, fdhD, and idh), and phosphate transport (i.e., pstB and pstS). Transcriptome data showed that the expression levels of amt, gltB, gltD, ackA and idh were downregulated, while gdhA and fdhD were upregulated. The inhibition of ammonium transporter and glutamine synthetase/glutamate synthase (GS/GOGAT) pathway contributed to reducing energy loss. For ammonium assimilation under ammonium stress, accompanied by protons efflux, the glutamate dehydrogenase pathway was the main approach. More 2-oxoglutarate (2-OG) was induced to provide abundant carbon skeletons. The downregulation of formate dehydrogenase and high glycolytic rate resulted in the accumulation of formic acid and acetic acid, which played key roles in PS under ammonium stress. Conclusions The accumulation of 2-OG and the inhibition of GS/GOGAT pathway played a key role in ammonium detoxification. The secretion of protons, formic acid and acetic acid was related to PS. Our work provides new insights into the PS mechanism, which will provide theoretical guidance for the application of PSMs. Supplementary Information The online version contains supplementary material available at 10.1186/s12864-023-09559-z.

Effect of ammonium stress on phosphorus solubilization of a novel marine mangrove microorganism Bacillus aryabhattai NM1-A2 as revealed by integrated omics analysis Zhaomei Lu 1,2,3 † , Sheng He 4 † , Muhammad Kashif 1,2 , Zufan Zhang 1 , Shuming Mo 1 , Guijiao Su 1 , Linfang Du 3* and Chengjian Jiang 1,2* Background Phosphorus (P) is essential to mangrove ecosystems, which is implicated as one of the most limiting nutrients for plant growth [1,2].Despite the high amount of total P in the soil, most of the P cannot be absorbed and utilized by plants [1,3,4].Chemical fertilizers have been widely used to satisfy the demand for P to promote plant productivity [4].However, the long-term application of chemical fertilizers has led to a series of problems limiting agricultural sustainable development, such as soil nutrient imbalance and water eutrophication [3,4].Consequently, it is urgent to find an eco-friendly and economically viable manner to maintain a continuous supply of P in plants.Phosphorus solubilization (PS) is a significant biochemical process in sediments, and plays important roles in alleviating P limitations of mangrove ecosystems [5].Several microorganisms can transform insoluble inorganic P into soluble forms available to plants, which are known as phosphate-solubilizing microorganisms (PSMs) [1,6].Furthermore, PSMs can not only release the accumulated P left by P fertilizer but also promote plant growth without environmental hazards [1,6].PSMs mainly include phosphate-solubilizing bacteria (PSB) and phosphate-solubilizing fungi (PSF).PSB include various strains from the genera Pseudomonas, Bacillus, Agrobacterium Azotobacter, Xanthomonas, and Enterobacter [7,8].PSF include strains of Penicillium, Mucor, Aspergillus, Pichia, and Chaetomium [8,9].These microorganisms can serve as biofertilizers that improve the agricultural yield [10].
The secretion of low molecular weight organic acids is considered to be the principal mechanism for mineral phosphate dissolution [3,11].These organic acids can liberate P either by lowering the pH or chelating the cations bound to phosphate [10].Gluconic acid produced by the direct oxidation of glucose is usually regarded as the main organic acid in the solubilization of mineral phosphate [3].Accumulating evidence suggested that intermediates in the tricarboxylic acid cycle (TCA cycle), such as succinic acid, malic acid, and citric acid, were also the main organic acids secreted during the process of PS [11].Other mechanisms such as the production of inorganic acids and chelating substances, are less effective than organic acids in releasing P [8].PS conducted by PSMs is a complicated process.The efficiency of PS is affected by many factors such as nutrition, pH, and oxygen concentration [12,13].Since PSMs are heterotrophic, it is generally accepted that carbon (C) and nitrogen (N) are two of the limiting factors for the growth of PSMs [12].It was reported that easily utilizable C source was deemed as a contributing factor to increase the PS efficiency of PSMs [14].Besides, ammonium (NH 4

+
) was also the best N source for PS, among various nutrients used by microorganisms [9].

NH 4
+ is a fundamental substrate for nucleic acid and amino acid synthesis [15,16].It has been proved that NH 4 + played a positive role as a stress signal [16].For instance, NH 4 + nutrition enhanced salt stress resistance in citrus plants due to the induction of the antioxidant cellular machinery [17].Moreover, increased alkaline phosphatase activity led to the solubilization of P under high NH 4 + stress [18].However, excessive NH 4 + damages the growth of the living organism, which is widely known as NH 4 + toxicity [19].NH 4 + toxicity is a significant ecological and agricultural issue [15].Several hypotheses for the cause of toxicity have been proposed, including the futile transmembrane NH 4 + cycle, the lack of inorganic cations and organic acids, the impairment of hormone homeostasis, and the disorder of pH regulation [20].A previous study found that supplemented succinate effectively alleviated NH 4 + toxicity [20].Therefore, it was assumed that the production of organic acids in PS process performed a significant role in resisting NH 4 + toxicity.Until now, few studies have focused on the influence of NH 4 + stress on releasing precipitated inorganic P. In our current work, an efficient PSM Bacillus aryabhattai NM1-A2 was isolated from the marine mangrove sediment.The effects of different C sources, N sources, C/N, (NH 4 ) 2 SO 4 concentrations, NaCl concentrations and initial pH of solution on the solubilization of insoluble phosphate were investigated to determine its PS characteristics.Then, the effect of NH 4 + stress on PS was explored at the level of genome, transcriptome, and metabolite.Our work demonstrates the scientific basis for the application of PSMs, and it will promote environment-friendly agricultural development.

Isolation and identification of PSMs
Seven PSMs were selectively obtained from the mangrove sediments based on the obvious phosphate-solubilizing halos around colonies on SRSM solid medium (the enrichment medium invented by Sundara Rao and Sinha specifically for the isolation of PSMs) (Fig. 1).The bromocresol purple indicator on the SRSM medium changed from purple to yellow (Fig. 1), which was a sign of pH drop.Thus, the seven PSMs might dissolve mineral phosphate by secreting organic acids or expelling protons.
We performed the PCR amplifications using specific primers [21] (Additional file 1: Table S1) for 16 S rDNA.16 S rDNA sequences were subjected to agarose gel electrophoresis, and bright bands were detected at 1, 500 bp (Fig. 2a).The results showed that the seven PSMs were pure strains.The nucleotide sequences of 16 S rDNA were analyzed with the BLASTn search algorithm and aligned to their nearest neighbors (Table S2).A phylogenetic tree was created based on 16 S rDNA nucleotide sequences of the isolates and their nearest neighbors (Fig. 2b).It was found that NM1-A2 and HM3 were most closely related to B. aryabhattai.HM5 and HM6 were most closely related to B. licheniformis.HM2, HM4, and HM7 were most closely related to Metabacillus halosaccharovorans, B. haynesii, and B. velezensis, respectively.

Determination of the soluble phosphate concentration in culture supernatant
The soluble phosphate concentration in culture supernatant of seven PSMs varied from 22.65 mg/L to 139.46 mg/L.Except for HM2 and HM7, the concentration of soluble phosphate decreased as time progressed.The soluble phosphate concentration in culture supernatant of B. aryabhattai NM1-A2 reached 139.46 mg/L at 48 h, which was much higher than those of other strains (Table 1).
The PS process of B. aryabhattai NM1-A2 on SRSM liquid medium containing different components was examined to better understand its PS characteristics.It was found that the solubilization of tricalcium phosphate (TCP) in the SRSM liquid medium was accompanied with the decrease of pH after 12 h (Fig. 3).The correlation analysis showed a significant negative correlation between the soluble phosphate concentration and pH in culture supernatant of B. aryabhattai NM1-A2 (Table 2).The results suggested that the secretion of organic acids or the release of protons might lead to a decrease in pH, which might be one of the PS mechanisms.The highest concentration of soluble phosphate reached 124.24 mg/L when B. aryabhattai NM1-A2 was cultivated with glucose as the sole C source, followed by sucrose, starch, fructose, and maltose (117, 73.07, 45.38 and 40.68 mg/L, p < 0.05) (Fig. 3a).As shown in Fig. 3b, the optimum N source for PS was (NH 4 ) 2 SO 4 (136.29 mg/L), followed by NH 4 Cl, CO(NH 2 ) 2 , KNO 3 (124.99,87.51, and 56.08 mg/L, p < 0.05).The soluble phosphate concentration in culture supernatant of B. aryabhattai NM1-A2 decreased rapidly with the increase in C/N ratio in the SRSM lipid medium (Fig. 3c).In addition, the efficiency of PS was affected by the concentration of (NH 4 ) 2 SO 4 .The highest soluble phosphate concentration of 196.96 mg/L was dissolved at 250 mM of (NH 4 ) 2 SO 4 concentration.in the range from 1 mM to 300 mM (Fig. 3d).And the production of soluble P could reach the highest when NaCl concentration was 2% in the range from 0 to 10% (Fig. 3e).Subsequently, the soluble phosphate concentration decreased continuously with the increase of NaCl concentration (Fig. 3e), which indicated that the process of PS was inhibited under salt stress.The change of pH value directly affected the survival and metabolism of microorganisms.As shown in Fig. 3f, B. aryabhattai NM1-A2 grew normally in the range of pH 5.0-10.0.The highest soluble phosphate concentration (142.79 mg/L) was detected when the initial pH value was 6.0, followed by 7.0, 5.0, 8.0, 9.0 and 10.0 (137.65,132.87, 126.504, 114.74, and 65.37 mg/L).

Effect of NH 4 + stress on the types and concentrations of organic acids
As we know, PS process of PSMs was associated with the release of low molecular weight organic acids [3,11].High performance liquid chromatography (HPLC) analysis showed the presence of four organic acids in culture supernatant, namely formic acid, malic acid, acetic acid and succinic acid, while lactic acid and citric acid were not detected (Fig. S1 and Table S3).Among the four organic acids, formic acid showed the highest concentration, followed by malic acid, acetic acid and succinic acid (Fig. 4).As shown in Fig. 4, the concentrations of formic acid and acetic acid in the high-NH 4 + group were 443.7 and 120.5 mg/L at 12 h, respectively.These concentrations were significantly higher than those in the low-NH 4 + group (160.1 and 57.3 mg/L at + group (20.8 mg/L, p < 0.05).However, there was no significant difference in the concentration of malic acid between the low-and high-NH 4 + group (Fig. 4).The increase of soluble phosphate concentration in culture supernatant of B. aryabhattai NM1-A2 could be potentially due to the production of organic acids.And the types and concentrations of secreted organic acids were influenced by (NH 4 ) 2 SO 4 concentration.

NH 4
+ was a preferred N source for most bacteria [2].The high-affinity NH 4 + transporters AMTs (encoded by amt) were used to absorb NH 4 + at low NH 4 + concentrations, while some low-affinity transporters were responsible for NH 4 + at high NH 4 + concentrations [22].The wholegenome of B. aryabhattai NM1-A2 (Table S4) contained genes coding for glutamine synthetase (GS, encoded by glnA), glutamate synthase (GOGAT, encoded by gltBD) and glutamate dehydrogenase (GDH, encoded by gdhA), which were key enzymes for NH 4 + assimilation [23].In addition, there were many types of PS-related genes, mainly including genes involved in organic acid synthesis and the phosphate regulatory system (Table S4).Inorganic phosphate (Pi) was mainly obtained through two independent uptake systems, inorganic phosphorus transporter (Pit) and phosphorus specific transporter (Pst) [24].The Pst system was a typical ABC transporter composed of PstS, PstC, PstA, PstB and PhoU [24,25].Pi was transported by the Pst system under P starvation [25,26].B. aryabhattai NM1-A2 also had histidine kinase PhoR, response regulator PhoB for induction of PHO regulon under P-limiting conditions [27].The main enzymes involved in TCA cycle and organic acid synthesis were found to be: isocitrate lyase, formate dehydrogenase, acetate kinase and isocitrate dehydrogenase, fumarate hydratase, and malate dehydrogenase, encoded by aceA, fdhD, ackA, idh, fumC and mdh, respectively.

Transcriptome sequencing of B. aryabhattai NM1-A2 and annotation analysis of DEGs
Transcriptome data results showed 2422 significantly differentially expressed genes (DEGs) between the low-and high-NH 4 + groups.A total of 739 significantly upregulated genes and 1683 significantly downregulated genes were detected (Fig. 5a and b).KEGG enrichment results showed that the DEGs were mainly enriched in carbohydrate metabolism, amino acid metabolism, metabolism of cofactors and vitamins, and energy metabolism (Fig. 5c).C and N metabolism played an important role in response to NH 4 + stress [22].Moreover, the numbers of upregulated or downregulated DEGs of the top 20 most enriched KEGG pathways were displayed (Fig. 5d).The genes related to pyruvate metabolism, oxidative phosphorylation, glycolysis, and TCA cycle were mostly upregulated.The GO enrichment analysis of DEGs between the low-and high-NH 4 + groups (Fig. 5e and f ) showed that the main enriched biological processes were cellular process, single organization and metabolic processes.The main enriched cellular components were cell and cell part.The main enriched functions were binding, catalytic activity, and transporter activity.These results indicated that catalytic and transport activities in B. aryabhattai NM1-A2 were significantly affected by high NH 4 + stress.
The transcriptome data showed that the expression level of amt was downregulated by 3.52 fold, which indicated that the NH 4 + transported by AMT was reduced (Table 3; Fig. 6).NH 4 + assimilation was regulated by two P II proteins (GlnB and GlnK) [28].GlnB and GlnK could be reversibly uridylylated by uridilyl-transferase/uridylylremovase (UT/UR), controlling the bifunctional enzyme adenylyl-transferase/adenylyl-removase (AT/AR), which could reversibly adenylylate (inactivate) or deadenylylate (activate) GS [28,29].GlnK was commonly cotranscribed with the NH 4 + transporter AMT and reversibly inhibited AMT's transport activity [29].The expression levels of glnB and glnK were downregulated by 2.779 and 2.323 folds, respectively (Table 3).AR enzymatic activity was promoted through the deuridylylation of GlnK-U and GlnB-U to rapidly inhibited the activity of GS.Then, GlnK binded to AMT to form a complex to inhibit NH 4 + uptake.
The GS-GOGAT cycle was the most crucial pathway for primary N assimilation [30,31].NH 4 + and glutamate (GLU) to form glutamine (GLN) under the catalysis of GS (GlnA) [29].The large subunit (GltB) and small subunit (GltD) of GOGAT complete the metabolic pathway of GLU regeneration from GLN and 2-oxoglutarate (2-OG) [29].The expression levels of gltB, and gltD were downregulated by 1.425, and 3.377 folds, respectively (Table 3), which revealed that the GS − GOGAT pathway was not the main approach for NH 4 + assimilation in B. aryabhattai NM1-A2.GDH has been reported to incorporate excess NH 4 + into GLU when exposed to NH 4 + stress [23].The expression level of gdhA was upregulated in the high-NH 4 + group compared with that in the low-NH 4 + group (Table 3).Therefore, the GDH pathway played a critical role in responding to NH 4 + stress (Fig. 6).The assimilation of NH 4 + was accompanied by the release of protons, which was one of the mechanisms of PS (Fig. 6).
The production of organic acids was the main mechanism for solubilizing insoluble P [9].The transcriptome data showed that the expression level of ackA was upregulated, wheras fdhD was downregulated (Table 3; Fig. 6).This might lead to increased concentrations of acetic acid and formic acid in high-NH 4 + group (Fig. 4).Therefore, formic acid and acetic acid played an important role in PS process of B. aryabhattai NM1-A2 under NH 4 + stress (Fig. 6).In addition, idh was also upregulated by 1.256 fold.More 2-OG was produced under NH 4 + stress, which provided sufficient C skeletons for the subsequent amino acid synthesis.The expression level of aceA was downregulated (Table 3), which might be the reason behind the decrease in succinic acid in the high-NH 4 + group (Fig. 4).fumC and mdh were upregulated in the high-NH 4 + group compared with those in the low-NH 4 + group (Table 3).The catabolism of malic acid was greater  than the synthesis, which led to the decrease in malic acid concentration under NH 4 + stress.The Pst system played an important role in transporting inorganic P under conditions of P limitation [27].PstS and PstB were the key proteins in Pst system [24].Genes related to the Pst system and PhoR were downregulated under NH 4 + stress, especially pstS and pstB (downregulated by 5.29 and 5.71 folds).As shown in the Fig. 3d, the content of soluble phosphate in the high-NH 4 + group (196.96mg/L) was significantly higher than that in the low-NH 4 + group (62.5 mg/L, p < 0.001).As a result, the Pst transporter was inhibited.The signal was transmitted from the Pst transporter to PhoR through PhoU, and inhibited the kinase function of PhoR, resulting in dephosphorylation of PhoB [27].When Pi was abundant, excess P was stored as polyphosphate (PolyP) [32].PolyP could be separated into two groups, pyrophosphates (two phosphate residues) and high molecular weight PolyP [33].Pyrophosphates (PPi) was a metabolic product of biosynthetic reactions, and pyrophosphatase (PPaX) catalyzed the hydrolysis of PPi [33].ppaX was upregulated in high-NH 4 + group, which could maintain cellular PPi homeostasis and provide energy for the PPi-generating biosynthetic reactions (Table 3; Fig. 6).

Discussion
P is an essential nutrient element for plant growth and development [1,2].PSMs can promote the availability and absorption of P in soil [8].The application of PSMs plays a very important role in the sustainable development of agriculture [1,6].
In this study, TCP was selected as the sole P source on SRSM medium because it represented the vast majority of insoluble P [34].Based on cultivation and observation, seven PSMs were screened from marine mangrove sediments (Fig. 1).According to 16 S rDNA sequence analysis, these strains were confirmed as members of Bacillus genus.Several studies have shown that Pseudomonas, Bacillus and Rhizobia were effective phosphate solubilizers [7,8].Bacillus spp.were preferred on account of lasting vitality [35].The soluble phosphate concentrations in culture supernatant of these strains were evaluated by molybdenum blue colorimetric method [21].Among them, the soluble phosphate concentration in culture supernatant of B. aryabhattai NM1-A2 reached a maximum of 139.46 mg/L at 48 h (Table 1).In addition, the soluble phosphate concentration in culture supernatant decreased as time progressed, which could probably be attributed to the utilization of P by bacteria [36].Bacillus aryabhattai was widely distributed in various habitats owing to its strong stress resistance, and had great potential in nitrogen fixation, PS and the production of indole-3-acetic acid [36][37][38].However, few studies have revealed the genomic potential of Bacillus aryabhattai on PS.
PSMs are heterotrophic microorganisms.The PS efficiency of PSMs depends on the provision of suitable C and N in a large extent [12].Scervino et al. reported that glucose and (NH 4 ) 2 SO 4 -based media showed the highest PS values [12], which was consistent with the results of this study (Fig. 3a and b).Glucose was the preferred C source for most bacterial growth [14].However, Li's report [6] showed that there was no correlation  between cell growth and soluble P production [6].Mineral phosphate could be solubilized by the direct oxidation of glucose to gluconic acid [8].This condition might explain the maximum PS ability with glucose as C source (Fig. 3a).Among all N sources tested, the optimum N source for PS was (NH 4 ) 2 SO 4 (Fig. 3b).Bacterial growth was greatly enhanced with the increasing concentration of (NH 4 ) 2 SO 4 , which led to the consumption of available P in the culture medium [6].Interestingly, the soluble phosphate concentration in culture supernatant of B. aryabhattai NM1-A2 was continuously improved with the increase of (NH 4 ) 2 SO 4 concentration and reached the highest (196.96mg/L) under 250 mM (NH 4 ) 2 SO 4 (Fig. 3d).NH 4 + was the preferred N source for most bacteria, but it was toxic when reaching a certain concentration [37].For example, Nitrobacter winogradskyi Nb-255 grew well under NH 4 + concentrations below 25 mM, but it was inhibited under NH 4 + concentrations higher than 35 mM [38].It was reported that transcript degradation or gene repression was part of the NH 4 + -responsive stimulon [39].Given its remarkable NH 4 + -tolerant survivability, the mechanism of PS and detoxification of B. aryabhattai NM1-A2 under NH 4 + stress was revealed by multi-omics analysis.
Ammonium exists predominantly in the ionic form (NH 4

+
) at neutral pH, and the minor gaseous species (NH 3 ) can diffuse rapidly through the cell membrane.NH 4 + could be transported into the cell via AMT under low ammonium conditions [40].When bacteria encountered high concentration of NH 4 + , the passive diffusion of NH 3 can provide enough N for optimal cell growth [41].Transcriptome data showed that amt was downregulated, suggesting NH 4 + transported via AMT was inhibited (Table 3).NH 4 + assimilation was the main approach for NH 4 + detoxification [22].Under high-NH 4 + condition, excess NH 4 + could be assimilated into GLU mainly through GDH pathway, thereby reducing NH 4 + accumulation (Table 3) [41,42].Downregulating the GS/GOGAT pathway under NH 4 + stress was helpful to prevent the waste of energy (Table 3) [43].Moreover, research shows that NH 4 + assimilation by GS localized in the plastid rather than NH 4 + accumulation was a primary cause for toxicity [19].In this study, GS enzyme activity of B. aryabhattai NM1-A2 was inhibited in the regulation of GlnB and GlnK under NH 4 + stress, thereby preventing NH 4 + toxicity (Fig. 6) [29].Tian et al. reported that the sugar and starch contents of plants decreased under NH 4 + treatment because of the depletion of C skeletons used for excess NH 4 + assimilation [22].In the present study, the overall upregulated glycolysis, pyruvate metabolism and TCA cycle could provide abundant C skeletons for excess NH 4 + assimilation (Fig. 5d) [22].Among them, the upregulated expression of idh indicated that more 2-OG was induced under NH 4 + stress (Table 3) [44,45].It has been reported that supplemented 2-OG enhanced NH 4 + assimilation [20].That might be an important feature of B. aryabhattai NM1-A2 responded to NH 4 + toxicity.The assimilation of NH 4 + within microbial cells was accompanied by the release of protons resulting in the solubilization of P [3,46].
The secretion of organic acids was an important phosphate-solubilizing way in PSMs, which could be produced from cells to facilitate the solubilization of P by supplying both protons and metal complexing organic acid anions [34].Different PSMs would secrete the various types and concentrations of organic acids to solubilize P [3].The production of succinic acid was related to PS activity of Bacillus megaterium, while oxalic acid and citric acid were the main organic acids released by Pseudomonas sp [3].The secretion of organic acids was also influenced by the types and concentrations of P sources [47].Arvind et al. found that the generation of formic acid was limited to the dissolution of TCP, and oxalic acid production to the solubilization of Mussoorie rock phosphate, Udaipur rock phosphate, and North Carolina rock phosphate [48].In this study, transcriptome and HPLC analysis showed that formic acid and acetic acid played important roles in the improvement of available P fractions under NH 4 + stress, and the concentrations of formic acid and acetic acid in the high-NH 4 + group were generally higher, about 2.77 and 2.1 folds, respectively (Fig. 4; Table 3).It has been reported that acetate would accumulate when the glycolysis flux exceeded the catalytic capacity of the TCA cycle [49].Therefore, high glycolysis rate might present in B. aryabhattai NM1-A2, which was responsible for the accumulation of acetic acid under NH 4 + stress (Fig. 4).Furthermore, the PS process was affected by the properties of organic acids.Although formic acid and acetic acid were both mono-carboxylic acids, formic acid had a higher PS ability than acetic acid [50].Isocitrate lyase was a key enzyme of the glyoxylate cycle that was the bypassed pathway of the TCA cycle [51].As shown in Table 3, ackA was downregulated under NH 4 + stress, which suggested that the transfer of energy metabolism to the glyoxylate cycle was inhibited.This might lead to a decrease in succinic acid concentration in the high-NH 4 + group (Fig. 4).+ stress.This study demonstrated a scientific basis for the development of biofertilizers.

Samples and media
The sediment samples were isolated from the National Shankou Natural Reserve of Mangrove in Beihai, Guangxi, China (21°29′25.74′′N,109°45′49.43′′E)using the methods described in a previous study [52].

Screening for PSMs
The sediment (5.0 g) was mixed with LB medium (10 mL), followed by shaking for 1 h in a biochemical incubator at 37 °C, 200 rpm.Then serial dilutions (10 0 -10 − 3 ) were spread-plated onto LB medium plates.After incubation for 12 h at 37 °C, colonies were selected from the plates.Then, these isolates were purified by the streak plate method.Each isolate was added to the SRSM medium plates to verify its PS ability by the presence of a clear halo around colonies.

Identification of the isolated strains
All strains were cultured in 100 mL Erlenmeyer flasks containing 50 mL LB at 37 °C with continuous rocking on a shaker at 200 rpm for 12 h.Total genomic DNA was extracted using a TIANamp bacterial DNA kit (TIAN-GEN) according to the instructions of the manufacturer.The quality of the extracted DNA was measured by 1% agarose gel electrophoresis and analyzed using a Nano-Drop™ 2000 spectrophotometer [5] (Thermo Scientific, USA).The amplified products of 16 S rDNA were obtained with specific primers [21] (Table S1) under the following reaction conditions: 95 ℃, 5 min; [95 °C for 30 s; 54 °C for 30 s; and 72 °C for 1 min] × 35 cycles; and 72 °C, 7 min.The PCR products were verified by 1% agarose gel electrophoresis, and they were sequenced using the Sanger sequencing platform (BGI: https://www.bgi.com/).The nucleotide sequences of 16 S rDNA were compared with sequences in NCBI (http://www.ncbi.nlm.nih.gov/GenBank) to identify close phylogenetic relatives.And the 16 S rDNA phylogenetic tree was constructed by using the neighbor-joining method supported by 1,000 bootstrap replicates in MEGA 7.0 [54].

Determination of the soluble phosphate concentration in culture supernatant
The seven strains were cultured on SRSM medium for 48, 72, 96, and 120 h, and then centrifuged at 6, 000 rpm for 5 min.Then, the concentration of soluble phosphate in culture supernatant was determined by molybdenum blue colorimetric method [21].A separate SRSM medium inoculated with sterile double-distilled water served as the control.

Organic acids production of B. aryabhattai NM1-A2 under NH 4 + stress
The types and concentrations of organic acid in the supernatants of the low-and high-NH 4 + groups were analyzed using HPLC method [6].The following organic acids (i.e., formic acid, malic acid, lactic acid, acetic acid, citric acid, and succinic acid) were measured using an Alliance liquid chromatograph (Waters E2695, United States).The liquid chromatograph column was Poro-shell120 SB-Aq 4.6×250 mm, 4 μm (Agilent Technologies Inc., America), with a column temperature of 25 ℃.The detection wavelength was 210 nm.

Whole-genome sequencing and assembly of B. aryabhattai NM1-A2
The whole-genome sequencing and assembly were performed by PFOMIC Bioinformatics Company (Nanning, China) with the Illumina NovaSeq 6000 PE150, Oxford Nanopore Technologies PromethION [54].The whole genome data of B. aryabhattai NM1-A2 have been deposited in NCBI GeneBank with the accession number PRJNA760884.The MinKNOW software was used for filtering the low-quality reads.Then, the filtered reads were assembled by using the Unicycler (0.4.8) software.Moreover, Pilon software was used to obtain more accurate genome.Finally, DIAMOND was used to align the predicted protein sequence with the Kyoto Encyclopedia of Genes and Genomes (KEGG) [55].For the alignment of each protein sequence, the match with the highest score (default identity ≥ 30%) was used to select the functional annotation.

Transcriptome sequencing analysis of B. aryabhattai NM1-A2
We analyzed the transcriptome of B. aryabhattai NM1-A2 in the low-and high-NH 4 + groups to study the transcriptional expression changes between the N deficient and the NH 4 + -stressed environment.The RNA of B. aryabhattai NM1-A2 was extracted using the TRIzol method [56].Qualified libraries were sequenced on the Illumina Novaseq sequencing platform using the pairedend sequencing method (PE150).FastQC_v0.11.3 was used for quality control of the raw sequencing data [57].The filtered sequence was aligned with the rRNA database using Bowtie2 to remove the rRNA sequence.Qualified sequencing data were aligned to the B. aryabhattai NM1-A2 genome using Hisat2, and the count value and the Fragments Per Kilobase of transcript per Million mapped reads (FPKM) values of the genes were calculated using HTseq-count and AWK script.EdgeR was used to calculate the differentially expressed genes (DEGs) [p < 0.05 and |Log 2 (fold change) | > 1].Then, Gene Ontology (GO) and KEGG databases were used for functional enrichment of DEGs [55].

Verification experiment
RT-qPCR was used to validate the transcriptomic data.A two-step method was used for the RT-qPCR of key genes with specific primers (Table S2) [21].GAPDH was used as a reference gene for the normalization.The expression levels of selected genes were evaluated using the 2 −ΔΔCt method.

Statistical analyses
All experimental treatments were conducted in triplicate, and the results are expressed as the means ± standard error.Data were processed with SPSS 24 and Diamond software.The relevant query databases were KEGG and NCBI.The least significant difference test based on the analysis of variance model was used for multiple comparisons among treatments groups for soluble phosphate concentration or pH value.And two independent-sample t-test was used to determine differences in organic acid concentration between the low-and high-NH 4 + group.All figures were made with GraphPad Prism 8.0.2 (263) and Adobe Illustrator (2020).

Fig. 3
Fig. 3 (a) Effects of carbon sources, (b) nitrogen sources, (c) C/N, (d) (NH 4 ) 2 SO 4 concentrations, (e) NaCl concentrations, and (f) initial pH on the solubilization of tricalcium phosphate by B. aryabhattai NM1-A2.The least significant difference test based on the analysis of variance model was used to calculate statistical significance, as expressed in means ± standard error.And values significantly different from those of the control were represented by different lowercase letters

Fig. 4 Fig. 5
Fig. 4 Kinds and concentrations of organic acids in the supernatants of B. aryabhattai NM1-A2 in the low-and high-NH 4+ groups.Two independentsample t-test was used to determine differences in organic acid concentration between the low-and high-NH 4 + group.And values significantly different from those of the control were represented by asterisks, which " * " and " * * * " indicated p < 0.05 and p < 0.001, respectively, while "ns" indicated no significant difference

Fig. 6
Fig. 6 NH 4 + tolerance mechanism of B. aryabhattai NM1-A2."+" indicated that enzyme activity or substance content increases under NH 4 + stress."−" indicated that enzyme activity or substance content reduces under NH 4 + stress.Red letters indicated downregulation genes, and navy letters indicated upregulation genes.Purple letters indicated pathways, and orange ovals indicated transporters.Solid arrow corresponded to direct reaction, and dotted arrow corresponded to indirect reaction.The red line indicated that the process is inhibited.2-OG, TCP, Pi, polyP, Pst, AMT, GLU, GLN, and GS indicated 2-oxoglutarate, tricalcium phosphate, inorganic phosphorus, polyphosphate, phosphorus specific transporter, ammonium transporter, glutamate, glutamine, and glutamine synthetase, respectively

Table 1
The concentration of soluble phosphate in culture supernatant when strain was cultured in SRSM medium for 12 h

Table 2
Correlation coefficient between soluble phosphate concentration and pH while Bacillus aryabhattai NM1-A2 was cultured under different conditions