Introduction

Alfalfa (Medicago sativa L.) is perennial leguminous forage with high yields, a strong regenerative capacity, multiple harvests in one growing season, and high crude protein concentration. Alfalfa can adapt to different regional environments. It has played an important role in the adjustment of the animal husbandry structure in China (Brink et al. 2015). Alfalfa roots can reach more than ten meters long and absorb water from the deep soil (Sim et al. 2017). Alfalfa planting is of great significance in arid and semiarid areas (Gu et al. 2018). Xinjiang has a temperate continental arid climate with low rainfall, uneven seasonal rainfall distribution, and shortages of surface water and groundwater resources (Zhang et al. 2020). Alfalfa consumes large amounts of water, and artificial irrigation is necessary for alfalfa cultivation in arid areas. Therefore, water is one of the main factors limiting the development of alfalfa in Xinjiang. Some studies demonstrated that sufficient irrigation could increase the photosynthetic activity of alfalfa, resulting in increased alfalfa dry matter yield would increase. However, the water absorption capacity in the roots of alfalfa gradually decreased under drought stress, and the transpiration rate and photosynthetic rate decreased, which led to a decline in crop yield (Brookshire and Weaver 2015). Therefore, identifying the appropriate irrigation amount is the key to improving alfalfa production performance.

Phosphorus (P) is an essential element for plant growth and development, because it is involved in a wide range of physiological, biosynthetic, and metabolic processes (Lissbrant et al. 2009). It has been reported that the dry matter yield, nutrient quality and root growth of alfalfa are significantly affected by the soil P concentration (Mallarino and Rueber 2013). One study showed that water-P coupling can affect the level of P in the soil environment because of the unique properties of P in soil, such as its low solubility, low mobility, and high fixation by the soil matrix, and the recovery of applied P by crops in one growing season is often low (Pizzeghello et al. 2014). Most of the P remains in the soil in the form of poorly soluble P, which increases the concentration of P in the soil, limits the growth and development of alfalfa and causes P pollution in the soil (Thuynsma et al. 2014).

Low phosphorus use efficiency (PUE) seriously hinders further improvements in crop yield and nutritional quality. Phosphorus-solubilizing bacteria (PSB) can increase P availability to plants by releasing organic acids and phosphatases that enhance the solubility of various inorganic P forms in soil. Therefore, PSB may provide a good way to solve the problem of P limitation. Some studies have demonstrated that PSB play a crucial role in soil P solubilization and increase the bioavailability of soil P for plants (Shi et al. 2017). PSB can transform insoluble P to available phosphorus (AP) in the soil for plant absorption and utilization, and promote plant growth by increasing the plant P concentration (Heijden et al. 2008). P nutrient uptake by plants mainly occurs through the roots in contact with soil. Arbuscular mycorrhizal fungi (AMF) can form symbiotic relationships with more than 80% of terrestrial plants (Meena et al. 2018) and promote mineral nutrient uptake by host plants (especially P) and soil fertility. Compared with single AMF inoculation, simultaneous inoculation with two rhizotrophic bacteria can significantly improve the yield and P concentration in alfalfa, enhance the ability of plant roots to resist drought (Rodríguez-Caballero et al. 2017) and increase the effectiveness of the microorganisms in saving P and increasing crop production (Zhang et al. 2014).

At present, research on water and P mainly focuses on the performance of rice and other crops (Song et al. 2018), while there are relatively few studies on the effects on the production performance of alfalfa. In particular, there are few reports regarding the effects of simultaneous inoculation with PSB and AMF under water-P coupling conditions on the production performance and roots of alfalfa. Therefore, this study aimed to investigate the effects of water-P coupling on the simultaneous inoculation of PSB and AMF on the growth, nutrient quality and underground biomass of alfalfa to provide a theoretical basis for the development of a rational water and fertilizer management system as well as compound microbial fertilizers for alfalfa cultivation in the Xinjiang oasis region of China.

Materials and Methods

Experimental details and treatments

Experimental site: The pot experiment was conducted in 2018 at the experimental station of the Agricultural College of Shihezi University (44°18′ N, 86°03′ E), Xinjiang, China. The experimental site was located in a temperate continental climate zone that is dry and rainless. The diurnal temperature varied greatly; the mean annual temperature was 11.2–13.9ºC and the annual precipitation was 203.1–394.9 mm. The annual pan evaporation was approximately 1000–1500 mm. The test soil was collected from the experimental station of the Agricultural College of Shihezi University, Shihezi, China. The soil (0–20 cm layer) at the experimental site was a grey desert soil. The collected soil was air-dried and then passed through a 2 cm sieve to remove roots, stones and Table 1: Basic physical and chemical properties of the test soils

Table 2: Experimental design and implementation plan

Table 3: Amount of fertilizer application (mg kg^-1)

Note: P₀, P₁, P₂, and P₃ represent 0 mg kg^-1, 50 mg kg^-1, 100 mg kg^-1and 150 mg P kg^-1, respectively

other fine plants material in the soil and was brought back to the laboratory for the determination of physical and chemical properties. The specific physical and chemical properties of the soil are shown in Table 1.

Treatments: An orthogonal experimental design (L16(4³)) was adopted in this study. There were four levels of moisture concentration, P application and bacterial inoculation in the potted plants (without considering the interactions of various factors). The experimental treatments were the factorial combinations of the different treatment factors. The four-soil water holding capacities were 35% (W₁), 50% (W₂), 65% (W₃), and 80% (W₄). The four P rates were 0 mg kg^-1 (P₀), 50 mg kg^-1 (P₁), 100 mg kg^-1 (P₂) and 150 mg kg^-1 (P₃). The four inoculation treatments were no inoculation (J₀), B. megaterium inoculation (J₁), F. mosseae inoculation (J₂), and double inoculation (B. megaterium + F. mosseae) (J₃). Each treatment was repeated 6 times and the treatments were randomly arranged, as shown in Table 2. To ensure that the test was only affected by the phosphate fertilizer, based on the monoammonium phosphate (NH₄H₂PO₄) containing nitrogen fertilizer, the effect of the nitrogen fertilizer on the production of alfalfa was offset by adding urea (CN₂H₄O) to maintain the consistency of the test, as shown in Table 3.

In this experiment, the B. megaterium strain was purchased from the Agricultural Culture Collection of China (ACCC). B. megaterium can grow and form soluble P circles in NBRIP liquid medium with Ca₃(PO₄)₂ as the P source. F. mosseae was purchased from Qingdao Agricultural Mycorrhizal Research Institute of China. The inoculum of this fungus was a mixture of spores, hyphae, sand and root segments of its host plants. The density of spores was 25–35 g^-1. The alfalfa variety tested was WL354HQ.

The composition of the beef extract peptone liquid medium was: beef extract 5 g L^-1, peptone 10 g L^-1, NaCl 5 g L^-1, agar 30 g L^-1, pH 7.0. The composition of Hoagland's nutrient solution was: Ca(NO₃)₂ 945 mg L^-1, KNO₃ 607 mg L^-1, MgSO₄ 493 mg L^-1, iron salt solution 2.5 mg L^-1, trace element 5 mg L^-1, pH 6.0.

The strains of B. megaterium were rejuvenated and inoculated into beef extract peptone liquid medium for propagation. A plate with a colony count of 30–300 was used as the effective counting plate, and the colony count in the bacterial solution was approximately 10⁹ cfu mL^-1 for backup use. The soil was sterilized at high temperature and humidity and was heated at 121ºC and stored. Large seeds were selected, sterilized with 75% alcohol for 30 s, sterilized with 5% hypochlorite for 12 min, rinsed with sterile water many times and sown in the seedling tray. The seedling tray had a size of 72 holes plate^-1 and each hole had a diameter of 4 cm. One seed was planted per well. The seeding depth in the seedling pots was 1–2 cm. After the seeds were sown, the bacteria mentioned above were added B. megaterium was added in 10 mL volumes to the seedling tray, and F. mosseae was spread all around the seeds. On March 24, 2018, the seedling tray was placed in a constant temperature incubator to accelerate germination at 25℃. The culture conditions were as follows: 12 h of light (at 25℃), 12 h of darkness (at 20℃), 300 micromol m^-2 S^-1 light intensity, and 55% air humidity. Meanwhile, a black plastic basin of 24 cm × 16 cm × 19 cm (basin diameter × bottom diameter × height) was soaked in alcohol for 20 min and stored. On April 6, 10 seedlings with uniform growth were selected and placed in pot boxes. Five kilograms of sterile soil was added to each pot during transplantation, and the bacteria were added again (as above). Hoagland's solution without phosphoric acid was added every 10 days (100 mL per pot) for each treatment. The specific addition dates were March 24, March 30, April 9, April 19 and April 29, 2018. The application was stopped after adding phosphate fertilizer. The P fertilizer used in this study was monoammonium phosphate (P 52%), which has good water solubility. The P fertilizer was applied together with the irrigation water beginning at the branching period and after each cut. The specific fertilization dates were May 11 and July 4, 2018, and the distance between the pots was 20 cm. The water holding capacity of the pot soil was controlled by the weighing method at 35–80% at 10:00 every morning. Each treatment was repeated 6 times for a total of 96 pots. Each pot was surrounded by supports with a white plastic tarp on them. If it was rainy, the tarp was spread out to prevent the rain from influencing the pot experiment.

Alfalfa biomass

Taking each pot as a unit, 3 pots with the same growth of the 6 pots in each treatment were selected. The first crop was cut on June 30, 2018, and the second crop was cut on August 19, 2018. The alfalfa plants were cut (5 cm) with scissors and weighed, and the yield of fresh alfalfa forage was recorded. The roots of the alfalfa were rinsed and weighed, and the fresh weight was recorded. This was repeated 3 times. The absolute length of each taproot was measured. The aboveground and underground biomass samples were taken back to the laboratory. The sample were first oven-dried at 105°C for 30 min and then at 65°C to a constant mass. The aboveground biomass (g pot^-1) and underground biomass (g pot^-1) of alfalfa were calculated according to the following formula.

Aboveground biomass of alfalfa = Fresh yield of alfalfa × (1-moisture concentration)                         (1).

Underground biomass of alfalfa = Fresh root of alfalfa × (1-moisture concentration)                           (2).

Plant height

At the same time as the biomass measurement, 10 alfalfa plants with uniform growth were randomly selected in three pots. The vertical height of the alfalfa plants to the surface was measured by a steel tape, and the average height (cm) was calculated.

Stem diameter determination

At the same time as the plant height measurements, the stems of 10 alfalfa plants along the height of the plant were measured, the stem diameter at 5 cm from the ground was measured with a Vernier calliper, and the average values (mm) were determined.

Nutrient quality

Crude protein was determined by the semimicro Kjeldahl method. The neutral detergent fibre and acid detergent fibre were determined according to the procedures of Soest et al. (1991).

Phosphorus concentration

The fresh alfalfa grass and root samples were dried and crushed. The samples were placed in a 600°C Maofu furnace to burn to a white ash, and the ash was dissolved by hydrochloric acid. After filtration, the P concentration was determined using the molybdenum-antimony anti-spectrophotometric method (Fan et al. 2016). The soil in the pots was removed from the second cut, sieved through a 2 mm sieve and placed in a self-sealing plastic bag for the determination of total phosphorus (TP) and AP in the soil. TP was determined by the sulfuric acid-perchloric acid decoction molybdenum antimony colorimetric method and AP was determined by the NaHCO₃ extraction molybdenum antimony colorimetric method (Mehlich 1984).

Taproot length

After the soil was removed from the pots, 10 alfalfa roots from plants whose plant height and stem diameter had been measured were washed with water. The length of the main root was measured by straightening the main root with a steel tape measure and the average taproot length (cm) was determined.

Statistical analysis

Microsoft Excel 2010 was used for data processing, and all the plant data collected were statistically analysed in S.P.S.S. 20.0 using analysis of variance. The obtained results were tested withFisher’s least significant difference (Duncan’s) test with significance determined at the 5% level. The principal component analysis method was used to identify the best treatment. The principal component analysis method formula is as follows:

F_i= A_ijZ_ij, i=1，2，3…n.                     (3)

Where A is the eigenvector value and Z is the standardized value of the alfalfa index for each treatment.

The principal component synthesis model formula is as follows (Tang and Feng 2002):

F= F_iλ_i, i=1，2，3…n.                       (4)

Where λ_i represents the proportion of the variance contribution rate of the i-th principal component to the total extracted variance contribution rate.

Pearson's correlation analysis was used to analyse the correlation of each growth index of alfalfa to the treatments.

Results

Aboveground biomass, plant height and stem diameter

The aboveground biomass, plant height and stem diameter of alfalfa were significantly higher in all treatments than in the CK treatment (P ≤ 0.05) and reached a maximum under the W₃P₂J₀ treatment (Table 4). The dry matter yield of alfalfa in the W₂, W₃ and W₄ treatments was significantly higher than that in the W₀ treatment (P ≤ 0.05). The aboveground biomass, plant height and stem diameter of alfalfa first increased and then decreased with increasing P application under the same water holding capacity and reached a maximum under the P₂ treatments in the first cut. The aboveground biomass, plant height and stem diameter of alfalfa were significantly different between the P₂ and P₃ and the P₀ and P₁ treatments under the W₂ and W₄ conditions (P ≤ 0.05), but there was no significant difference between the P₂ and P₃ and the P₀ and P₁ treatments (P ≥ 0.05). The aboveground biomass, plant height and stem diameter of the P₂ and P₃ treatments were significantly greater than those of the P₀ treatments under W₃ conditions (P ≤ 0.05). The aboveground biomass, plant height and stem diameter of alfalfa in the first cut were higher than those in the second cut.

Nutritional quality

The nutritional quality of alfalfa was determined by inoculating PSB and AMF under water-P coupling conditions (Table 5). The crude protein and P concentration of alfalfa first increased and then decreased with increasing P Table 4: Production performance of alfalfa under different treatments

Note: Different small letters within the same column indicate significant differences at the 0.05 level

Table 5: Nutrition quality of alfalfa under different treatments

Note: Different small letters within the same column indicate significant differences at the 0.05 level

application under the same water holding capacity and reached a maximum under the W₂P₂J₂ treatment. The crude protein in the W₂ treatments was significantly higher than that in the W₄ treatments (P ≤ 0.05). The P concentration of alfalfa in the P₂ treatment was significantly higher than those in the P₀, P₁ and P₃ treatments (P ≤ 0.05). The neutral detergent fibre and acid detergent fibre in the water, P and bacteria treatments were significantly lower than those in the CK treatment (P ≤ 0.05) and they increased first and then decreased with increasing P application under the W₂, W₃ and W₄ treatments. The crude protein in the first cut was lower than that in the second cut, and the P concentration was higher in the first cut than in highly than the second cut under the different treatments.

Underground biomass and soil phosphorus

The taproot length and underground biomass of alfalfa first increased and then decreased with increasing P application under the same water holding capacity and reached a maximum under the W₁P₂J₂, W₂P₂J₂ and W₄P₂J₁ treatments in the P₂ treatment, except for the W₂P₁J₃ treatment in the W₂ treatment (Table 6). The TP and AP increased gradually with increasing P application under the same water holding capacity and reached a maximum under the W₁P₃J₃, W₂P₃J₃, W₃P₃J₃ and W₄P₃J₀ treatments in the P₃ treatment. The TP and AP in the P₁, P₂ and P₃ treatments were significantly higher than those in the P₀ treatments (P ≤ 0.05).

The taproot length and AP of the water, P and bacteria treatments were significantly higher than those of the CK treatment (P ≤ 0.05), and the highest were W₃P₁J₃ and W₂P₃J₃, which increased 69.7 and 138.2%, respectively. The taproot length and underground biomass in the W₃ treatments were significantly higher than those in the W₁ treatment (P ≤ 0.05), and the underground biomass reached a maximum under the P₂ treatment. The taproot length and underground biomass in the J₁, J₂ and J₃ treatments were significantly higher than those in the J₀ treatment (P ≤ 0.05). The soil pH values in the J₁, J₂ and J₃ treatments were significantly less than those in the J₀ treatment (P ≤ 0.05).

Variance analysis

Table 6: Underground biomass of alfalfa and soil phosphorus concentration under different treatments

Note: Different small letters within the same column indicate significant differences at the 0.05 level.

Table 7: Variance analyses of the effects of water, phosphorus and bacteria on the indicators of alfalfa

Note: W: soil water holding capacity; P: phosphorus application; J: bacterium inoculated, P < 0.05 was significant; P < 0.01 was extremely significant.

The aboveground biomass, plant height, crude protein, neutral detergent fibre, acid detergent fibre, taproot length, and underground biomass were influenced by factors in the order of water > P > bacteria (Table 7). The effects of water, P and bacteria on the stem diameter and P concentration of alfalfa decreased in the order P > water > bacteria; the effects on pH value were bacteria > water > P, and the effects on soil TP and AP were in the order P > bacteria > water.

The soil water holding capacity had a highly significant effect on aboveground biomass, plant height, stem diameter, crude protein, neutral detergent fibre, acid detergent fibre, plant P concentration, taproot length, and underground biomass (P ≤ 0.01). The P application rate had a highly significant effect on aboveground biomass, plant height, stem diameter, crude protein, plant P concentration, taproot length, underground biomass, TP and AP (P ≤ 0.01), which had a highly significant effect on acid detergent fibre and taproot length (P ≤ 0.05). The P application rate had a highly significant effect on soil pH and soil TP (P ≤ 0.01) and had a significant effect on AP (P ≤ 0.01).

Pearson's correlation analysis

The plant height, stem diameter, taproot length and underground biomass were significantly positively correlated with the dry matter yield (P ≤ 0.05) (Table 8). The P concentration in alfalfa was positively correlated with dry matter yield (P ≤ 0.01) and neutral detergent fibre and acid detergent fibre were significantly negatively correlated with dry matter yield (P ≤ 0.01). Dry matter yield, plant height, stem diameter and plant P concentration, taproot and underground biomass were positively (P ≤ 0.01) or significantly negatively correlated with neutral detergent fibre and acid detergent fibre, respectively, while neutral detergent fibre was significantly positively correlated with acid detergent fibre. The stem diameter, P concentration and TP were significantly positively correlated with AP (P ≤ 0.01), and the other indexes were not significantly correlated with AP (P ≥ 0.01). The underground biomass was significantly negatively correlated with crude protein (P ≤ 0.01) and there were no significant correlations between underground biomass and the other indicators (P ≥ 0.05).

Principal component analysis

Since each treatment performed differently for the different indicators, it was not sufficient to evaluate the optimal treatment based on any single indicator. The aboveground biomass, plant height, crude protein, plant P concentration, stem diameter, main root length and underground biomass were positive indicators of plant performance, while neutral detergent fibre, acid detergent fibre, pH value and total P concentration were negative indicators (Table 9). The total accumulation rate was 86.18%. The comprehensive evaluation model was constructed as Y_T =0.561Y₁ + 0.203Y₂ + 0.097Y₃, where Y_T stands for the comprehensive score, and a larger Y Table 8: The correlation analysis of each index of alfalfa under different treatments

Note: *Significant correlation was found at the 0.05 level (bilateral), **significant correlation was found at the 0.01 level (bilateral).

Table 9: The principal component analysis of each index of alfalfa under different treatments

comprehensive value indicated a better growth performance in the treatment. The nutritional quality and underground biomass had the greatest impact. The top four treatments were W₃P₁J₃ > W₃P₂J₀ > W₃P₃J₁ > W₃P₀J₂.

Discussion

In this study, sufficient irrigation increased the aboveground biomass and plant height of alfalfa. The growth of alfalfa was inhibited when the soil water holding capacity was lower or higher than 65%, and the lower the soil moisture content was, the lower the aboveground biomass and plant height of alfalfa. When the moisture content in the soil is too low, drought stress occurs in the alfalfa plant. Biomass accumulation in plants occurs through photosynthesis, while drought stress inhibits photosynthesis and reduces plant biomass (Fan et al. 2016). On the other hand, when the soil moisture is low, alfalfa root growth is inhibited, which reduces the plant's ability to absorb water and nutrients. This prevents the products required by photosynthesis from being synthesized and thereby reduces the alfalfa biomass (Podlaski et al. 2017). In this study, the aboveground biomass and plant height of alfalfa at 80% soil water holding capacity were less than those at 65% soil water holding capacity under the same P application rate. The excess moisture in the pots could not spread into the surrounding soil, resulting in the alfalfa roots being immersed in water for a long time. The immersion of alfalfa roots in water obstructs aerobic respiration and enhances anaerobic respiration, which reduces the aboveground biomass of alfalfa (Zhang et al. 2020). However, the soil water holding capacity of the 80% treatment was higher than that of the 35% treatment. This is mainly because the potting box is made of plastic; in the sun from July to August, the temperature in the potting boxes is higher than the normal daily temperature. High temperatures increase water evaporation, which leads to a drier soil environment and forces the alfalfa roots to self-recover. It can be concluded that too high or too low of water content has a negative impact on the aboveground biomass of alfalfa and that a moderate moisture range should be used in the production of alfalfa.

In this study, P application significantly affected the aboveground biomass, plant height, stem diameter and plant P concentration of alfalfa, but it was not true that “the more P, the better”. P application significantly increases the amount of chlorophyll in alfalfa leaves, increases the photosynthetic rate of alfalfa, promotes the growth of alfalfa plants, and increases the dry matter yield of alfalfa (Williams et al. 2018). However, excessive P application results in a decrease in dry matter quality (Fan et al. 2016). There is a certain threshold for the absorption of P by alfalfa plants. When P is below a certain threshold, additional P promotes alfalfa growth and development. When P level exceeds the maximum absorption of P by alfalfa, the dry matter yield of alfalfa plants decreases (Bai et al. 2013). Excess P has a negative impact on plant growth and development and can lead to early growth and premature senescence of alfalfa (Fan et al. 2016). Therefore, the reasonable application of P fertilizer can improve the alfalfa growth.

Bacterial inoculation effectively promoted alfalfa growth. Compared with the no-bacteria treatment, inoculation with B. megaterium improved the soil AP and the alfalfa biomass. Because the AP in soil increased, the alfalfa roots could immediately absorb and utilize it for the growth and development of roots and then transport the nutrients to the aboveground parts; this promoted an increase in the aboveground and underground biomass of plants (Luduena et al. 2018). Inoculation with F. mosseae formed a symbiont with the roots of alfalfa. Mycorrhizal hyphae can absorb water directly and increase the surface area of the roots. As a result, the use and absorbance efficiency of nutrients and water increases (Parniske 2008), which in turn increases the aboveground and underground biomass of alfalfa. The taproot length and underground biomass under the 50% water treatment were significantly higher than those under the 80% water treatment. This result indicated that inoculation can alleviate the restrictions on alfalfa root length and root biomass under mild drought conditions. Research has shown that inoculation with AMF can improve the growth environment of alfalfa by delaying the ageing of root nodules under drought stress (Kyriazopoulos et al. 2014). Therefore, the effect of drought stress on alfalfa roots can be alleviated under mild drought stress. The taproot length and underground biomass of alfalfa in the 50% water treatment were significantly greater than those in the 35% water treatment, and AMF and PSB could not completely offset the inhibition of drought on plants under severe water stress conditions (Rahimzadeh and Pirzad 2017).

P moves through plants in various forms, but its mobility in soil is poor. Alfalfa has the highest P use efficiency, especially in the absence of P or under suitable P conditions in soil. P transfer occurs earlier and more often under low-P stress. A series of changes will also occur in the transfer and distribution of P in alfalfa plants under P stress (Rodríguez et al. 2000). This is the reason why the P concentration of alfalfa still increased under water shortage conditions, and why the P concentration in the plants was related to the crude protein concentration. Therefore, the crude protein increases with the increase in the P concentration, and the nutritional quality of alfalfa is improved. The normal growth and development of alfalfa plants were hindered, the water concentration in the alfalfa plants decreased and the lignification degree increased under the severe water shortage conditions (35% water) (Zhang et al. 2016); as a result, the neutral detergent fibre and acid detergent fibre increased significantly. Suitable irrigation rate (65% water) provided an adequate water supply for the alfalfa plants, their growth and development were normal, and their neutral detergent fibre and acid detergent fibre decreased. With a further increase in the soil water holding capacity (80% water), the aboveground biomass of the alfalfa plants also increased. The fibre concentration was the highest in the stem, and the stem diameter increased significantly, which led to an increase in the neutral detergent fibre concentration and a decrease in the nutritional quality of the alfalfa.

The effects of AMF and PSB inoculation on the growth, nutritional quality and underground biomass of alfalfa were different under the different water-P coupling conditions. The evaluation of the optimal water, P and bacteria model through only one indicator does not fully explain the advantages and disadvantages of the different treatments. Principal component analysis can be used to evaluate the optimization of multiple indicators by synthesizing multiple indicators (Song et al. 2018). The four treatments that had the greatest influence on the production performance of alfalfa were W₃P₁J₃ > W₃P₂J₀ > W₃P₃J₁ > W₃P₀J₂. This indicated that the alfalfa performance was the highest when the soil water holding capacity was 65%, the P application rate was 50 mg·kg^-1, and AMF and PSB were inoculated simultaneously. This treatment effectively improved the aboveground biomass of alfalfa, dissolved more soil TP, promoted the absorption of AP by alfalfa plants, and improved the nutritional quality of alfalfa compared with the other treatments (Meena et al. 2018). PSB and AMF play more important functional roles under low P conditions, when PSB can dissolves more P. AMF uses the dissolved P from the PSB to infect roots and form mycorrhizae to improve the root absorption ability, thereby increasing the P concentration and biomass of alfalfa (Smith et al. 2004). Under high P conditions, the cells reach a supersaturated state because the P content in the soil is too high; the PSB themselves contain a large amount of P, which inhibits the functions of PSB and AMF (Rahimzadeh and Pirzad 2017). In addition, PSB and AMF improve drought resistance in plant roots under mild water stress, but AMF and PSB cannot completely offset the inhibitory effect of drought on plants under severe stress conditions (Shi et al. 2017). Therefore, only when suitable water and P coupling conditions were selected could the effects of AMF and PSB inoculation improve alfalfa production performance and nutritional quality as well as soil AP.

Conclusion

Compared with those under high P conditions, the effects of inoculation were more beneficial under low P conditions. Severe water stress (35% soil water holding capacity) seriously inhibited the growth of alfalfa. Simultaneous inoculation with B. megaterium (PSB) and F. mosseae (AMF) effectively alleviated the damage to alfalfa from mild water stress (50% soil water holding capacity), and the double inoculation effect was better than the single inoculation effect. When the soil water holding capacity was 65%, the P application rate was 50 mg kg^-1, and AMF and PSB were inoculated simultaneously, it effectively improved the aboveground biomass of alfalfa, dissolved more soil TP, promoted the absorption of AP by alfalfa plants, and improved the nutritional quality of alfalfa compared to the other treatments.

Acknowledgements

The research was supported by the National Natural Science Foundation of China (31660693), the Fok Ying Tung Education Foundation of China (171099), the China Postdoctoral Science Foundation (2018T111120, 2017M613252), the Youth Innovation Talent Cultivation Program of Shihezi University (CXRC201605) and the China Agriculture Research System (CARS-34).

Reference

Bai ZH, HG Li, XY Yang, BK Zhou, XJ Shi, BR Wang, DC Li, JB Shen, Q Chen, W Qin, O Oenema, FS Zhang (2013). The critical soil P levels for crop yield, soil fertility and environmental safety in different soil types. Plant Soil 372:27‒37

Brink GE, MA Sanderson, MD Casler (2015). Grass and legume effects on nutritive value of complex forage mixtures. Crop Sci 55:1329‒1337

Brookshire ENJ, T Weaver (2015). Long-term decline in grassland productivity driven by increasing dryness. Nat Commun 6; Article 7148

Fan JW, YL Du, BR Wang, NC Turner, T Wang, LK Abbott, K Stefanova, KHM Siddique, FM Li (2016). Forage yield, soil water depletion, shoot nitrogen and phosphorus uptake and concentration, of young and old stands of alfalfa in response to nitrogen and phosphorus fertilization in a semiarid environment. Field Crops Res 198:247‒257

Gu YJ, CL Han, JW Fan, XP Shi, M Kong, XY Shi, KHM Siddique, YY Zhao, FM Li (2018). Alfalfa forage yield, soil water and P availability in response to plastic film mulch and P fertilization in a semiarid environment. Field Crops Res 215:94‒103

Heijden MGAVD, RD Bardgett, NM Van Straalen (2008). The unseen majority: soil microbes as drivers of plant diversity and productivity in terrestrial ecosystems. Ecol Lett 11:296‒310

Kyriazopoulos AP, M Orfanoudakis, EM Abraham, ZMM Parissi, N Serafidou (2014). Effects of arbuscular mycorrhiza fungi on growth characteristics of Dactylis glomerata L. under drought stress conditions. Not Bot Hortic Agrobot 42:132‒137

Lissbrant S, S Stratton, SM Cunningham, SM Brouder, JJ Volenec (2009). Impact of long-term phosphorus and potassium fertilization on alfalfa nutritive value-yield relationships. Crop Sci 49:1116‒1124

Luduena LM, MS Anzuay, JG Angelini, M Mcintosh, A Becker, O Rupp, A Goesmann, J Blom, A Fabra, T Taurian (2018). Strain serratia spp. s119: A potential biofertilizer for peanut and maize and a model bacterium to study phosphate solubilization mechanisms. Appl Soil Ecol 126:107‒112

Mallarino AP, D Rueber (2013). Alfalfa hay and soil-test phosphorus responses to long-term phosphorus fertilization strategies. Soil Sci Soc Amer J 69:1118‒1128

Meena RS, V Vijayakumar, GS Yadav, T Mitran (2018). Response and interaction of Bradyrhizobium japonicum and arbuscular mycorrhizal fungi in the soybean rhizosphere. Plant Growth Regul 84:207‒223

Mehlich A (1984). Mehlich 3 soil test extractant: A modification of mehlich 2 extractant. Commun Soil Sci Plant Anal 15:1409‒1416

Parniske M (2008). Arbuscular mycorrhiza: The mother of plant root endosymbioses. Nat Rev Microbiol 6:763‒775

Pizzeghello D, A Berti, S Nardi, F Morari (2014). Phosphorus-related properties in the profiles of three Italian soils after long-term mineral and manure applications. Agric Ecosyst Environ 189:216‒228

Podlaski S, S Pietkiewicz, DT Chołuj, T Horaczek, G Wisniewski, D Gozdowski, HM Kalaji (2017). The relationship between the soil water storage and water-use efficiency of seven energy crops. Photosynthetica 55:210‒218

Rahimzadeh S, A Pirzad (2017). Arbuscular mycorrhizal fungi and pseudomonas in reduce drought stress damage in flax (Linum usitatissimum L.): A field study. Mycorrhiza 27:1‒16

Rodríguez D, FH Andrade, J Goudriaan (2000). Does assimilate supply limit leaf expansion in wheat grown in the field under low phosphorus availability? Field Crops Res 67:227‒238

Rodríguez-Caballero G, F Caravaca, AJ Fernández-González, MM Alguacil, M Fernández-López, A Roldán (2017). Arbuscular mycorrhizal fungi inoculation mediated changes in rhizosphere bacterial community structure while promoting revegetation in a semiarid ecosystem. Sci Total Environ 584:838‒848

Shi XK, JJ Ma, LJ Liu (2017). Effects of phosphate-solubilizing bacteria application on soil phosphorus availability in coal mining subsidence area in Shanxi. J Plant Interact 12:137‒142

Sim RE, HE Brown, EI Teixeira, DJ Moot (2017). Soil water extraction patterns of lucerne grown on stony soils. Plant Soil 414:95‒112

Smith SE, FA Smith, I Jakobsen (2004). Functional diversity in arbuscular mycorrhizal (AM) symbioses: The contribution of the mycorrhizal P uptake pathway is not correlated with mycorrhizal responses in growth or total P uptake. New Phytol 162:511‒524

Soest PJV, JB Robertson, BA Lewis (1991) Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. J Dairy Sci 74:3583‒3597

Song T, FY Xu, W Yuan, YJ Zhang, TY Liu, MX Chen, QJ Hu, Y Tian, WF Xu, HJ Zhang (2018). Comparison on physiological adaptation and phosphorus use efficiency of upland rice and lowland rice under alternate wetting and drying irrigation. Plant Growth Regul 86:195‒210

Tang QY, MG Feng (2002). Practical Statistic Analysis and DPS data Processing System, pp:20‒50. Science Press, Beijing, China

Thuynsma R, A Valentine, A Kleinert (2014). Phosphorus deficiency affects the allocation of below-ground resources to combined cluster roots and nodules in Lupinus albus. J Plant Physiol 171:285‒291

Williams MR, KW King, EW Duncan, LA Pease, CJ Penn (2018). Fertilizer placement and tillage effects on phosphorus concentration in leachate from fine-textured soils. Soil Till Res 178:130‒138

Zhang L, J Fan, X Ding, X He, F Zhang, G Feng (2014). Hyphosphere interactions between an arbuscular mycorrhizal fungus and a phosphate solubilizing bacterium promote phytate mineralization in soil. Soil Biol Biochem 74:177‒183

Zhang QB, JY Liu, L Yu, WH Lu, CH Ma (2020). Effects of irrigation on growth traits, nutritional quality and seed characteristics of Medicago falcata var. romanica in an oasis. Intl J Agric Biol 23:391‒398

Zhang QB, L Yu, WH Lu, CH Ma, HX He (2016). Optimal irrigation regime improving yield and quality of alfalfa in year of sowing. Trans Chin Soc Agric Eng 32:16‒22