JIA-2019-11

2607 CHEN Xu et al. Journal of Integrative Agriculture 2019, 18(11): 2605–2618 fertilization and three replicates (60 m 2 each) (Song et al. 2007). In this study, the control plot without N fertilizer input was used. All crop residues in FL were removed following harvest for animal feed and cooking and heating fuel to simulate local practices. FL was plowed in autumn with a moldboard to a depth of 20 cm, disked in spring before planting and harrowed during the growing season. The basic soil properties are shown in Appendix A. The amount of available N was measured using alkaline diffusion (Bremner and Mulvaney 1982), extractable phosphorus (P) was measured using NaHCO 3 extraction (Olsen 1954), and extractable K was measured using NH 4 OAc (Jackson 1973). Soil pH was measured using deionized water (1:2.5). 2.2. Preparation of soil samples Ten soil cores (2.64 cm in diameter) were randomly collected from GL and BF to a depth of 10 cm with a probe sampler in August 2016. The cores from the same soil management were combined and placed in plastic bags. Cores were similarly collected from each of the three FL control replicates and combined into a single sample. The samples were transported to the laboratory and stored at 4°C the same day as collected. All samples were carefully handled to maintain their intact soil ped structure before being manually broken into small peds along natural cracks during the air-drying process. All visible organic debris werere moved by hand-picking before additional soil processing. Samples were separated into two treatments for incubation: intact aggregate distribution (IAD) and reduced aggregate distribution (RAD). The IAD samples consisted of intact aggregates, sieved to <6 mm. The RAD samples were crushed into smaller aggregates by hand-rolling with a glass cylinder (8.0 cm in diameter) to pass through a 1.0-mm sieve. Four classes of water-stable aggregates (WSA) were isolated using a wet-sieving procedure before and after aggregate disruption: large macroaggregates (>2 mm, WSA >2 mm ), small macroaggregates (0.25–2 mm, WSA 0.25–2 mm ), large microaggregates (0.053–0.25 mm, WSA 0.053–0.25 mm ), and small microaggregates (<0.053 mm, WSA <0.053 mm ) (Ding et al. 2014, 2015). 2.3. Incubation design The IAD and RAD soils were pre-incubated at 22°C for 7 days at 40% water-holding capacity (WHC) by uniformly distributing deionized water with a spray bottle. A 2×4×5 factorial experiment was established with two aggregate-size treatments, four substrate treatments and five destructive sampling dates. The pre-incubated soils (100 g dry weight) were weighed into 120-mL specimen cups and placed into 750-mL Mason jars, for a total of 288 jars. The substrate treatment soils received glucose, alanine and inorganic N as typical components of root exudates and fresh residue inputs and a kind of chemical fertilizer application, respectively. A control (CK) with no added substrates was included in the design. Each treatment had three replicates. Carbon substrate treatments received glucose or alanine at 0.4 g C kg –1 , and the inorganic-N treatment received (NH 4 ) 2 SO 4 at 2 µg N g –1 dry soil. Soil-moisture content was increased to 60% WHC using a syringe/needle, and the jars were incubated at 22°C. Soil-moisture content was weighed and checked weekly, and deionized water was added to maintain 60% WHC as needed. The soil was destructively sampled at 3, 15, 35 and 60 days of incubation for measuring β-glucosidase and N-acetyl-β-D-glucosaminidase activities and phospholipid fatty acid (PLFA) content. 2.4. Soil organiccarbonand total nitrogendetermination Total SOC content and total N were analysed with a VarioEL CHN Elemental Analyzer (Heraeus Elementar Vario EL, Hanau, Germany) in finely ground subsamples of the air- dried field samples. 2.5. PLFA analysis PLFAs were extracted from 4 g of freeze-dried soil sample to assess microbial community structure (Bossio et al. 1998). Briefly, lipids were extracted in a single-phase chloroform- methanol-citrate buffer (1:2:0.8). Phospholipids were separated from neutral lipids and glycolipids on silica solid- phase extraction columns (Supelco, Inc., Bellefonte, USA). The polar lipids were methylated, and PLFA methyl esters were analyzed using an Agilent 6890A (Agilent Tech. Co., Ltd., Santa Clara, USA) gas chromatograph (GC) equipped with an HP-5 capillary column (30 m×0.32 mm×0.25 mm) and a flame ionization detector. Nonadecanoic acid methyl ester (19:0, Sigma-Aldrich, St. Louis, USA) was added as an internal standard when the samples were dissolved in 150 μL of hexane before GC analysis. Super-purified nitrogen was used as the carrier gas at a flow rate of 0.8 mL min –1 . The Supelco 37 Component FAME Mix and bacterial acid methyl esters (Sigma-Aldrich) were used for peak identification and quantification. Twenty-seven PLFAs were consistently identified and used for data analysis. The fatty acids 14:0, a14:0, i14:0, 15:0, a15:0, 15:0DMA, i15:0, i15:1ω6c, 16:0, i16:0, 16:1ω7c, 17:0, a17:0, cy17:0ω7c, i17:0, 17:1ω8c, 18:0, 18:1ω5c, 18:1ω7c, cy19:0ω7c and 20:0, which are considered to be of bacterial origin (Frostegard and Baath 1996), were used as biomarkers of