Biochar immobilized plant growth-promoting rhizobacteria enhanced the physicochemical properties, agronomic characters and microbial communities during lettuce seedling

Abstract

Spent mushroom substrate (SMS) is the by-products of mushroom production, which is mainly composed of disintegrated lignocellulosic biomass, mushroom mycelia and some minerals. The huge output and the lack of effective utilization methods make SMS becoming a serious environmental problem. In order to improve the application of SMS and SMS derived biochar (SBC), composted SMS (CSMS), SBC, combined plant growth-promoting rhizobacteria (PGPR, Bacillus subtilis BUABN-01 and Arthrobacter pascens BUAYN-122) and SBC immobilized PGPR (BCP) were applied in the lettuce seedling. Seven substrate treatments were used, including (1) CK, commercial control; (2) T1, CSMS based blank control; (3) T2, T1 with combined PGPR (9:1, v/v); (4) T3, T1 with SBC (19:1, v/v); (5) T4, T1 with SBC (9:1, v/v); (6) T5, T1 with BCP (19:1, v/v); (7) T6, T1 with BCP (9:1, v/v). The physicochemical properties of substrate, agronomic and physicochemical properties of lettuce and rhizospheric bacterial and fungal communities were investigated. The addition of SBC and BCP significantly (p < 0.05) improved the total nitrogen and available potassium content. The 5% (v/v) BCP addiction treatment (T5) represented the highest fresh weight of aboveground and underground, leave number, chlorophyll content and leaf anthocyanin content, and the lowest root malondialdehyde content. Moreover, high throughput sequencing revealed that the biochar immobilization enhanced the adaptability of PGPR. The addition of PGPR, SBC and BCP significantly enriched the unique bacterial biomarkers. The co-occurrence network analysis revealed that 5% BCP greatly increased the network complexity of rhizospheric microorganisms and improved the correlations of the two PGPR with other microorganisms. Furthermore, microbial functional prediction indicated that BCP enhanced the nutrient transport of rhizospheric microorganisms. This study showed the BCP can increase the agronomic properties of lettuce and improve the rhizospheric microbial community.

Keywords: biochar; lettuce seedling; microbial community; plant growth-promoting rhizobacteria; spent mushroom substrate.

Figure 1

Soil enzyme activities of substrate after the seedling experiment. SC, Sucrase; UE, Urease; CL, Cellulase; CAT, Catalase; ACPT, Acid protease; ALPT, Alkaline protease; ACP, Acid phosphatase; ALP, Alkaline phosphatase. CK, the commercial control; T1, CSMS based blank control; T2, T1 with 10% combined PGPR (v/v); T3, T1 with 5% SBC (v/v); T4, T1 with 10% SBC (v/v); T5, T1 with 5% BCP (v/v); T6, T1 with 10% BCP (v/v). All the data were expressed as mean ± SD (n = 3). Different lowercase letters represent significant differences by Turkey test at p < 0.05.

Figure 2

Agronomic and physicochemical properties of lettuce in different treatments. All the data were expressed as mean ± SD. Fresh weight, plant height, root length and leaves number were determined with nine replicates, and the others were determined with six replicates. Different lowercase letters represent significant differences by Turkey test at p < 0.05.

Figure 3

Taxa composition and LEfSe analysis in different treatments. Circos graphs at the phylum level and relative abundance at the genus level of bacteria (A,C) and fungi (B,D) in different treatments. The relative abundance below 1% were combined together and indicated as “others.” B-C-Paraburkholderia: Burkholderia-Caballeronia-Paraburkholderia. LEfSe analysis identified the significantly different abundant taxa (biomarkers) in different treatments (p < 0.05, LDA scores ≥ 3). The circles from the inside to outside indicate phylogenetic levels of bacteria (E) and fungi (F) from the domain to genus levels. Yellow nodes denote taxa with non-significance, and biomarker taxa are colored by their corresponding class color. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Figure 4

Network model visualizing the co-occurrence patterns of bacterial and fungal communities at the genus level (relative abundance > 0.5%) in different treatments. Each edge represents a significant strong relationship (r > 0.6, p < 0.05). The red and green edges depict positive and negative correlations, respectively. The nodes represent individual genera, and node size corresponds to their relative abundance. Larger modules with nodes >10 are denoted with different colors, and smaller modules are labeled in gray. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Figure 5

Function predicting and Mantel test analysis. (A) The predictive functional features of bacterial sequences mapped against KEGG database (level 2, p < 0.05, relative abundance >1%). (B) Significantly different relative abundances of bacterial functions between T2 and T5 treatments on the level 3 KEGG ortholog by Student’s t-test (p < 0.05, relative abundance >1%). (C) The predictive functional features of fungal sequences mapped against MetaCyc database (p < 0.05, relative abundance >1%). (D) Pairwise comparisons between agronomic and physicochemical properties are shown with a color gradient denoting Pearson’s correlation coefficient. The red and green color depict positive and negative correlations, respectively. AC, Leaf anthocyanin content; ACP, Acid phosphatase; ACPT, Acid protease; AGFW, Fresh weight of aboveground; CC, Leaf chlorophyll content; CL, Cellulase; EC, Electrical conductivity; LN, Leave number; MDA, Root malondialdehyde content; PH, Plant height; RA, Root activity; RL, Taproot length; TC, Total carbon; TN, Total nitrogen; TOM, Total organic matter; TP, Total phosphorus; UGFW, Fresh weight of underground. ^*0.01 ≤ p < 0.05, ^**0.001 ≤ p < 0.01, ^***p < 0.001. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)