Short-Term Rhizosphere Effect on Available Carbon Sources, Phenanthrene Degradation, and Active Microbiome in an Aged-Contaminated Industrial Soil

Abstract

Over the last decades, understanding of the effects of plants on soil microbiomes has greatly advanced. However, knowledge on the assembly of rhizospheric communities in aged-contaminated industrial soils is still limited, especially with regard to transcriptionally active microbiomes and their link to the quality or quantity of carbon sources. We compared the short-term (2-10 days) dynamics of bacterial communities and potential PAH-degrading bacteria in bare or ryegrass-planted aged-contaminated soil spiked with phenanthrene, put in relation with dissolved organic carbon (DOC) sources and polycyclic aromatic hydrocarbon (PAH) pollution. Both resident and active bacterial communities (analyzed from DNA and RNA, respectively) showed higher species richness and smaller dispersion between replicates in planted soils. Root development strongly favored the activity of Pseudomonadales within the first 2 days, and of members of Actinobacteria, Caulobacterales, Rhizobiales, and Xanthomonadales within 6-10 days. Plants slowed down the dissipation of phenanthrene, while root exudation provided a cocktail of labile substrates that might preferentially fuel microbial growth. Although the abundance of PAH-degrading genes increased in planted soil, their transcription level stayed similar to bare soil. In addition, network analysis revealed that plants induced an early shift in the identity of potential phenanthrene degraders, which might influence PAH dissipation on the long-term.

Keywords: DNA/RNA; PAH; bacterial diversity; plant root exudates; rhizosphere; ryegrass (Lolium).

Figure 1

Schematic summary of the microcosm set-up and subsequent sample analyses.

Figure 2

Changes in total dissolved organic carbon (DOC), organic acids, sugars, and phenanthrene concentration in soil extracts over time, in bare and bulk planted soils from day 0 (D0) to day 10 (D10). For each compound, maximum detected concentrations were set to 1 and measurements were scaled accordingly. Values are means of three independent replicate microcosms. P-values obtained from one-way ANOVAs testing for the overall effect of vegetation and the effect of time for each condition are given. Asterisks between two adjacent cells denote significantly different values between bare and bulk planted soil (t-test, P < 0.05).

Figure 3

Richness (A, number of OTUs; B, Chao1) and diversity (C, equitability; D, Shannon H′) indices in bare (orange), bulk planted (green), and rhizospheric soil (blue), based on DNA or cDNA libraries (plain and dotted line, respectively). The effect of sample type was tested separately for each time point and nucleic acid using ANOVA followed by post-hoc Tukey test, and letters denote group with significant difference (P < 0.05).

Figure 4

Dynamics of the taxonomic composition of resident and active bacterial communities in bare, bulk planted and rhizospheric soil. Values are mean of three independent microcosms. For comparison, data for day 0 in bare soil were also plotted for bulk planted soil.

Figure 5

Canonical Correspondance Analysis based on the distribution of abundant OTUs (>1% in at least one sample) in cDNA libraries. The model accounted for 61.6% of the total inertia, with a global P-value < 0.0001. Circles represent samples from bare (green palette) or bulk planted soil (orange palette). Environmental variables (red diamonds) include percentage of PHE degradation (PHEdeg%), rhizospheric soil fresh weight (RhizoSoil), total dissolved organic carbon (DOC), glucose (Gluc), mannose (Mann), trehalose (Tre), mannitol (Manni), fumarate (Fum), oxalate (Oxal), gluconate (Gnt), and formate (Form). Squares represent OTUs with their taxonomic affiliation, abbreviated as follows: Janthino., Janthinobacterium; Sphingom., Sphingomonas; Alcali., unclassified Alcaligenaceae; Magneto., Magnetospirillum; Phenylo., Phenlyobacterium; Oxalo., unclassified Oxalobacteraceae; Sphingob., Sphingobium; Geoderm., unclassified Geodermatophilaceae; Luteim., Luteiomas; Variov., Variovorax; Pseudo., Pseudomonas; B. sel., Bacillus selenatarsenatis; Achromo., Achromobacter; B. badius, Bacillus badius; Modesto., Modestobacter; Acidi., unclassified Acidimicrobiales; Xantho., unclassified Xanthomonadaceae; Kaisto., Kaistobacter; Nitro., Nitrospira; Gemm., Gemmatimonadetes. For clarity, only OTUs with contribution >1% to at least one axis were depicted.

Figure 6

Co-varying networks based on local similarity analysis of abundant OTUs in cDNA libraries (>1% in at least one sample) and environmental variables. (A) Sub-network from bare soil centered around phenanthrene degradation. (B–D) Sub-networks from bulk planted soil centered around phenanthrene degradation (B), mannose (C), and sucrose (D). Only associations with P < 0.01 were considered significant (Q < 0.005 and 0.012 for bare and bulk planted soil, respectively). Red and blue line edges represent positive and negative associations, respectively. OTUs are diamonds (with taxonomy) and environmental factors are circles.