Long-term high-P fertilizer input decreased the total bacterial diversity but not phoD-harboring bacteria in wheat rhizosphere soil with available-P deficiency

Jinshan Liu,Qin Ma,Xiaoli Hui,Jinyi Ran,Qingxia Ma,Xingshu Wang,Zhaohui Wang
DOI: https://doi.org/10.1016/j.soilbio.2020.107918
IF: 8.546
2020-01-01
Soil Biology and Biochemistry
Abstract:Appling phosphorus (P) fertilizer to agroecosystems affects not only crop yield but also associated soil microbial communities. The bacterial phoD gene encodes alkaline phosphatase (ALP) and plays an important role in organic P decomposition in soils. However, the impacts of long-term P fertilization on the bacterial phoD gene community, the total bacterial community, and the relationships of these communities with soil properties are poorly understood in loess soils with available-P deficiency. In this study, the impact of mineral P fertilization on the soil bacterial community was assessed. The 16S rRNA and phoD genes were targeted in DNA extracted from wheat rhizosphere soils subjected to five P fertilization rates (0 (P0), 50 (P50), 100 (P100), 150 (P150) and 200 (P200) kg P2O5 ha−1 yr−1) applied annually for 14 years. Compared to the P0 treatment, the P fertilization treatments increased the soil organic C (SOC), microbial biomass C (MBC) and available P (AP), and in the high-P treatments (P150 and P200) the total P (TP) and organic P (OP) increased, while the ALP activity decreased. All P fertilization treatments reduced the total bacterial diversity (Shannon index). However, only P200 decreased the number of operational taxonomic units (OTUs) with the 16S rRNA gene, and no P fertilization treatments affected phoD-harboring bacterial OTUs or diversity when compared to those in the P0 treatment. Additionally, compared to P0, the P fertilization treatments changed the 16S rRNA and phoD gene bacterial community compositions, with increased relative abundances of 3 phyla and 7 genera and decreased abundances of 1 phylum and 2 genera for the 16S rRNA gene and increased abundances of 4 genera and decreased abundances of 2 phyla and 1 genus for the phoD gene. Microbial network analysis showed that the high-P treatments (P150 and P200) reduced the number of links in the microbial network at the genus level for the 16S rRNA gene. Principal coordinate analysis (PCoA) showed that P fertilization treatments shifted the total bacterial community structure, and redundancy analysis (RDA) revealed that soil dissolved organic C (DOC) and P (AP, OP, TP and ALP) levels were significantly related to the total bacterial community structure. In conclusion, this study demonstrated that long-term P fertilization significantly affected soil C and P as well as the total and phoD-harboring bacterial community compositions in wheat rhizosphere soils and that high P fertilizer application rates reduced total bacterial OTUs, diversity and the connections, which might affect soil biogeochemical cycles.
What problem does this paper attempt to address?