3.1 Effect of slow-release nitrogen fertilizer application rate on maize yield
The yield of maize was significantly influenced by varying rates of slow-release nitrogen application. In this study, maize yield ranged from 11251.3 to 16665.2 kg ha− 1 (Fig. 2), with notable variations observed among different treatments. Comparatively, the incorporation of slow-release nitrogen fertilizer resulted in a significant increase in maize yield when compared to the control group (N0). Furthermore, it was found that maize yield initially increased and then decreased as the nitrogen application rate increased. Specifically, N8, N12, N16, N20 and N24 demonstrated respective increases of 30.72%、36.63%、45.98%、41.05%、39.41% when compared to CK (N0).
3.2 Effect of soil properties on slow-release nitrogen fertilizer application rate
Table 1
Effects of different nitrogen fertilizer applications on soil chemical and biological characteristics
Soil properties | N0 | N8 | N12 | N16 | N20 | N24 | F-value |
PH | 8.24 ± 0.04a | 8.14 ± 0.03b | 8.15 ± 0.01b | 8.13 ± 0.02b | 8.11 ± 0.02b | 8.04 ± 0.02c | 18.559*** |
SOC (g·kg− 1) | 11.52 ± 0.56c | 12.62 ± 0.82b | 13.07 ± 0.05b | 15.28 ± 0.15a | 13.4 ± 0.46b | 13.29 ± 0.62b | 22.216*** |
NH4+-N (mg·kg− 1) | 5.63 ± 0.17e | 5.82 ± 0.05de | 5.99 ± 0.07cd | 6.05 ± 0.07c | 6.36 ± 0.09b | 6.9 ± 0.13a | 67.989*** |
NO3−-N (mg·kg− 1) | 7.45 ± 0.06e | 9.13 ± 0.26d | 8.99 ± 0.04d | 9.66 ± 0.06c | 11.31 ± 0.02b | 26.16 ± 0.19a | 10014.25*** |
MBC (mg·kg− 1) | 173.61 ± 15.85d | 190.02 ± 5.97c | 218.86 ± 1.24b | 230.98 ± 6.31a | 195.01 ± 4.31c | 189.03 ± 2.24c | 30.74*** |
MBN (mg·kg− 1) | 69.58 ± 1.25d | 73.82 ± 1.12c | 75.39 ± 1.76bc | 80.14 ± 1.19a | 77.4 ± 1.04b | 74.81 ± 0.45c | 34.723*** |
Note: SOC: soil organic carbon; MBC: microbial biomass carbon; MBN: microbial biomass nitrogen; NO3−-N: nitrate nitrogen; NH4+-N: ammonium nitrogen.
The application of slow-release nitrogen fertilizer significantly altered the chemical and biological properties of the soil. This study revealed that fertilization led to a significant increase in soil NO3−-N and NH4+-N contents. Moreover, the contents of SOC, MBC, and MBN exhibited an initial increasing trend followed by a decrease with increasing nitrogen application rates, with the highest content observed in the N16 treatment, which was significantly higher than other treatments. Furthermore, nitrogen fertilizer application resulted in a reduction in soil pH, which decreased significantly as the amount of nitrogen applied increased (Table 1), with the lowest pH recorded in the N0 treatment. Additionally, compared to no fertilization, slow-release nitrogen fertilizer demonstrated its ability to enhance soil nutrient content, improve soil fertility and ultimately increase crop yield.
3.3 Response of soil enzyme activities to slow-release nitrogen fertilizer application rate
Table 2
Response of soil enzyme activities to different nitrogen fertilizer applications
Soil enzyme activity | N0 | N8 | N12 | N16 | N20 | N24 | F-value |
S-NiR (umol·(d− 1·g− 1)) | 0.998 ± 0.11c | 10.301 ± 1.04d | 0.118 ± 0.01e | 997.35 ± 5.89d | 1768.87 ± 26.21d | 8.04 ± 0.02c | 36.92*** |
S-NR (ug·(d− 1·g− 1)) | 1.553 ± 0.05b | 12.607 ± 0.26c | 0.341 ± 0.82c | 1023.86 ± 5.97c | 2115.97 ± 0.03c | 13.29 ± 0.62b | 34.14*** |
ALPT (mg·(d− 1·g− 1)) | 2.095 ± 0.07a | 13.372 ± 0.04c | 0.415 ± 0.05b | 1050.82 ± 1.24b | 2344.54 ± 0.01b | 6.9 ± 0.13a | 131.26*** |
HR (ug·(d− 1·g− 1)) | 2.218 ± 0.07a | 15.518 ± 0.06a | 0.58 ± 0.15a | 1092.78 ± 6.31a | 2539.26 ± 0.02a | 26.16 ± 0.19a | 19.20*** |
UE (ug·(d− 1·g− 1)) | 1.66 ± 0.09b | 14.588 ± 0.02b | 0.302 ± 0.46c | 1043.95 ± 4.31bc | 2255.66 ± 0.02bc | 189.03 ± 2.24c | 22.714*** |
Note: lowercase letters indicate statistically significant differences (P < 0.05). Soil enzymes include soil nitrite reductase (S-NiR), soil nitrate reductase (S-NR), alkaline protease (ALPT), hydroxylamine reductase (HR), and urease (UE). |
The changes in soil enzyme activity were further determined under different levels of nitrogen application. The utilization of slow-release nitrogen fertilizer had an impact on soil enzyme activity. Nitrate reductase, nitrite reductase, alkaline protease, hydroxylamine reductase, and urease all increased with the rise in nitrogen application rate (Table 2), reaching their highest values at the N16 treatment. However, as the nitrogen application rate continued to increase beyond this point, enzyme activity began to decline. These findings suggested that an appropriate nitrogen supply could enhance the enzymatic activity of enzymes related to the nitrogen cycle and improve soil fertility and productivity.
3.4 Microbial diversity
Table 3
Response of α-diversity to increasing nitrogen fertilizer application rate
Type of microorganism | Nitrogen fertilizer application rate | Non-rhizosphere | Rhizosphere | Endophyte |
Chao index | Shannon index | Sob index | Chao index | Shannon index | Sob index | Chao index | Shannon index | Sob index |
Bacteria | N0 | 1241.15 a | 6.59 a | 1180 a | 880.45 b | 6.18 ab | 829.25 b | 482.27 b | 5.11 a | 452.75 ab |
N8 | 1321.95 a | 6.62 a | 1235 a | 1390.63 a | 6.71 a | 1293.25 a | 608.69 a | 5.18 a | 575 a |
N12 | 1180.94 a | 6.50 a | 1090.5 a | 1370.05 a | 6.65 a | 1269 a | 442.93 b | 4.81 b | 424.25 b |
N16 | 894.71 ab | 6.15 ab | 845.75 ab | 971.78 ab | 6.35 a | 923.25 ab | 608.80 a | 5.34 a | 575.75 a |
N20 | 1180.84 a | 6.50 a | 1095 ab | 1009.37 a | 6.23 a | 954.25 ab | 653.34 a | 5.27 a | 607.5 a |
N24 | 1077.39 ab | 6.45 a | 1026.25 ab | 1011.72 a | 6.32 a | 952.25 ab | 625.46 a | 5.18 a | 582 a |
Fungi | N0 | 392.46 a | 4.43 a | 392 a | 385.5 a | 4.42 a | 385.25 ab | 196.55 | 2.84 a | 196.5 a |
N8 | 390.16 a | 4.42 a | 389.75 a | 452.45 a | 4.03 a | 452.25 a | 206.06 | 2.84 a | 206 a |
N12 | 390.39 a | 4.09 a | 390 a | 398.02 b | 4.10 a | 397.75 a | 148.37 | 2.19 b | 148.25 b |
N16 | 358.71 b | 4.11 a | 358 b | 361.61 ab | 3.72 ab | 361 b | 171.84 | 2.89 a | 171.75 ab |
N20 | 352.28 b | 4.06 a | 352.25 b | 335.19 b | 3.93 a | 335 b | 169.64 | 2.78 a | 169.5 b |
N24 | 394.24 a | 4.21 a | 394 a | 362.34 b | 4.14 a | 362.25 b | 193.03 | 2.70 a | 192.75 a |
In this study, the analysis of poly α-diversity changes in microorganisms across different spatial locations and nitrogen application levels revealed that bacterial and fungal communities in the inner root space exhibited lower α-diversity compared to those in the rhizosphere soil and non-rhizosphere soil. Additionally, the α-like abundance of microorganisms varied with different root spaces, being highest in non-rhizosphere soil, followed by rhizosphere soil, and lowest in root space. Fertilization increased the richness (Chao index), evenness (Shannon index), and dominance (Sob index) of bacterial communities in both rhizosphere and root space. However, fertilization decreased these indices for bacterial communities in non-rhizosphere space. Furthermore, fertilization reduced the richness, evenness, and dominance of fungal communities across all three spatial locations (Table 3). These findings indicated that fertilization enhanced bacterial community diversity within the rhizosphere while reducing it within non-rhizosphere soil; however, it decreased fungal community diversity overall. The main effects analysis revealed that fungal diversity and bacterial diversity differed at significant levels among locations, did not reach significant levels under different nitrogen application treatments, and that location and fertilization interactions had a greater effect on bacterial diversity(Figure S1).
By conducting a non-metric multidimensional scaling analysis (NMDS) on microbial β-diversity changes (Fig. 3), we compared differences in soil bacterial and fungal community structures between different treatments and spatial locations. ANOSIM results demonstrated significant variations among bacterial and fungal community structures across different spaces (P < 0.001). Notably, nitrogen application had a significant impact on bacteria and fungi community structures (P < 0.05), with the most pronounced effect observed within non-rhizosphere soil.
3.5 Species composition of microbial communities
The dominant bacteria in the rhizosphere and non-rhizosphere microorganisms under different nitrogen application treatments were Proteobacteria, Gemmatimonadota and Actinobacteria. The predominant phyla were Ascomycota, Basidiomycota and unclassified_k_Fungi (Fig. 4).
In the root microorganisms under different nitrogen application treatments, the dominant bacteria were Proteobacteria, Actinobacteria and Bacteroidota. The dominant phyla observed were Ascomycota, Olpidiomycota and Glomeromycota.
The abundance of Gemmatimonadota in rhizosphere and rhizosphere bacteria N8-N24 was higher compared to those without nitrogen (N0), while there was no significant difference observed in the non-rhizosphere samples. The abundance of Basidiomycota was higher in both the rhizosphere and non-rhizosphere samples with nitrogen application (N8-N24) compared to those without nitrogen (N0).
3.6 Analysis of the correlation between different spatial microorganisms and soil properties and yield
The community structure of soil bacteria and fungi was strongly influenced by the changes in soil properties resulting from different fertilization treatments. Spearman rank correlation analysis was employed to assess the impact of environmental factors on the composition of bacterial microbial communities in rhizosphere soil, non-rhizosphere soil, and endosphere of maize (Figure S2). Actinobacteriota, Myxococcota, Firmicutes, and Verrucomicrobiota significantly influenced the physical and chemical properties of rhizosphere soil. Proteobacteria, Acidobacteriota, Myxococcota, and Actinobacteriota were significantly affected by soil properties in non-rhizosphere soil while Proteobacteria, Fibrobacterota, and Bdellovibrionota were primarily impacted by pH in endosphere. These findings indicate that pH plays a significant role in shaping bacterial community structure. Linear regression analysis further revealed (Fig. 5) that Myxococcota had a positive effect on ALPT activity in rhizosphere soil whereas Actinobacteriota and Firmicutes had negative effects on ALPT activity as well as protein synthesis (Fig. 5A-5C). In non-rhizosphere soil, pH along with NH4+-N, NO3−-N, and UE exerted significant influences on changes observed within bacterial communities (Fig. 5E-5Q). Because of the accumulation of hydrogen ions in the soil and subsequent decrease in pH resulting from long-term fertilization, root growth was adversely affected, leading to an increased abundance of endophytic bacteria such as Actinobacteriota and Gemmatimonadota (Fig. 5R-5S).
Spearman rank correlation was employed to assess the impact of environmental factors on the microbial community composition in rhizosphere soil, non-rhizosphere soil, and endosphere of maize (Figure S2). Basidiomycota, Olpidiomycota, and soil significantly influenced the physical and chemical properties of rhizosphere soil. Glomeromycota and Mortierellomycota in non-rhizosphere soil had significant effects on the physical and chemical properties of the soil. Moreover, Olpidiomycota, an endosphere fungus, exhibited a significant correlation with the physical and chemical properties of the soil. Linear regression analysis further demonstrated these relationships (Fig. 6). In rhizosphere soil, Basidiomycota positively impacted MBC, MBN, ALPT activity, S-NR content as well as HR levels. The abundance of Olpidiomycota showed a positive correlation with nitrate and ammonium nitrogen contents (Fig. 6A-6G). In non-rhizosphere soil, Mortierellomycota positively affected ALPT activity while Glomeromycota had a positive effect on S-NiR activity but negatively influenced MBC levels along with UE, ALPT, and S-NR (Fig. 6H-6N). Furthermore, Olipidomycoata positively affected endosphere microbial biomass carbon (Fig. 6O-6P).
The key dominant bacterial phyla that affected yield were Proteobacteria, Acidobacteria and Myxomycota, whereas the dominant fungal phyla were Basidiomycota and Glomeromycota.
3.7 Microbial network analysis
The correlation networks of ASV were utilized to explore the combined patterns under different nitrogen application rates and spatial locations. It was observed that nitrogen application resulted in more intricate bacterial networks, while fungal networks exhibited the opposite trend (Figure S3; Figure S4).
The co-occurrence network of ASVs in the non-rhizosphere soil, rhizosphere soil, and endosphere bacterial and fungal communities revealed that the fungal network exhibited higher complexity and connectivity compared to the bacterial network(Figure 7). Within the bacterial network, both the average network distance and clustering coefficient were higher in the endosphere network than in the rhizosphere and non-rhizosphere soil networks. Conversely, within the fungal network, it was observed that the complexity of rhizosphere microorganism networks was greater (Table S1, Table S2).
Specifically, within bacterial networks, fertilization treatment demonstrated increased complexity and connectivity compared with control treatment along with relatively higher proximity centrality and degree centrality measurements (Table S3; Figure S3). Within fungal communities, fertilization enhanced aggregation as compared to control treatment (N0), thereby promoting greater complexity and stability within the microbial network.
Based on high proximity centrality and degree centrality, we identified key ASVs from the collinear network of bacteria at six different nitrogen application levels. In the N0, N16, N20, and N24 treatments, a total of 20 core bacteria ASVs were selected. These ASVs belonged to Gemmatimonadota, Actinobacteriota, Bacteroidota, Proteobacteria, Myxococcota and Acidobacteriota. The ASVs treated with N8 included additional taxa such as Firmicutes. Similarly, in the N12 treatment group, the selected ASVs belonged to Gemmatimonadota, Actinobacteriota, Bacteroidota, Proteobacteria, Myxococcota and Acidobacteriota.
For fungi networks, in the N0 and N16 treatments groups, a total of 20 core fungal ASVs were selected belonging to Olpididiomycota, Ascomycota and Mortierellomycota respectively; in the N12 and N24 treatments groups, twenty core fungi ASVs were selected, all belonging to Olpidiomycota, Ascomycota, Mortierellomycota, Basidiomycota and Glomeromycota, respectively; all the ASV treated with N8 belonged to Olpidiomycota, Ascomycota, Mortierellomycota and Basidiomycota, respectively; and all the ASVs treated with N20 belonged to Ascomycota, Mortierellomycota and Basidiomycota, respectively(Table S4).
The complexity of microbial networks was found to be influenced by varying levels of nitrogen application. Specifically, it was observed that these levels enhanced the complexity and stability of bacterial and fungal networks.
3.7 The direct and indirect effects of multiple soil multifunctionality drive
The results of structural equation modeling (Fig. 8) demonstrated a significant influence of nitrogen application levels on the variations in soil chemical characteristics, soil enzyme activity, and soil microbial biomass. Additionally, microbial biomass carbon and nitrogen exhibited a significant impact on enzyme activity changes. These results indicated that an appropriate amount of nitrogen fertilizer input could regulate microbial biomass carbon and nitrogen, and promote the production of enzymes, and bacteria had a more obvious regulation of enzyme activity (Fig. 8A). On this basis, the comprehensive effects of various factors could regulate the change of microbial diversity, especially the significant impact of bacterial diversity on corn yield, indicating that bacterial diversity may be an important potential factor in promoting the increase of yield (Fig. 8B).
These results provide a foundation for understanding how the amount of slow-release nitrogen fertilizers improved soil physicochemical characteristics and made changes in soil microbial diversity and richness, which in turn enhanced maize yield (Fig. 9).