Treatment impacts on temporal microbial community dynamics during phytostabilization of acid-generating mine tailings in semiarid regions

Abstract

Direct revegetation, or phytostabilization, is a containment strategy for contaminant metals associated with mine tailings in semiarid regions. The weathering of sulfide ore-derived tailings frequently drives acidification that inhibits plant establishment resulting in materials prone to wind and water dispersal. The specific objective of this study was to associate pyritic mine waste acidification, characterized through pore-water chemistry analysis, with dynamic changes in microbial community diversity and phylogenetic composition, and to evaluate the influence of different treatment strategies on the control of acidification dynamics. Samples were collected from a highly instrumented one-year mesocosm study that included the following treatments: 1) unamended tailings control; 2) tailings amended with 15% compost; and 3) the 15% compost-amended tailings planted with Atriplex lentiformis. Tailings samples were collected at 0, 3, 6 and 12months and pore water chemistry was monitored as an indicator of acidification and weathering processes. Results confirmed that the acidification process for pyritic mine tailings is associated with a temporal progression of bacterial and archaeal phylotypes from pH sensitive Thiobacillus and Thiomonas to communities dominated by Leptospirillum and Ferroplasma. Pore-water chemistry indicated that weathering rates were highest when Leptospirillum was most abundant. The planted treatment was most successful in disrupting the successional evolution of the Fe/S-oxidizing community. Plant establishment stimulated growth of plant-growth-promoting heterotrophic phylotypes and controlled the proliferation of lithoautotrophic Fe/S-oxidizers. The results suggest the potential for eco-engineering a microbial inoculum to stimulate plant establishment and inhibit proliferation of the most efficient Fe/S-oxidizing phylotypes.

Keywords: Iron oxidation; Microbial diversity; Mine reclamation; Plant growth promoting bacteria; Sulfur oxidation.

Figure 1

Analysis of the rarefaction of OTU alpha diversity and the taxonomic assignment of OTUs. OTU assignments are based on 97% sequence similarity. (A) rarefaction curves for all the samples with iTag reads normalized to 5,700 sequences. Dashed lines (lighter color) around curves correspond to standard error (95% confidence intervals). (B) the relative abundance of the 7 overall most abundant taxa (each taxon has multiple OTUs) for all the samples. These 7 taxa account for at least ~90% of all OTUs detected per samples.

Figure 2

Non-Metric Multidimensional Scaling (NMDS) ordination diagram of temporal variations in taxonomic community composition. The ordination is based on a Bray-Curtis distance matrix calculated from the relative abundance data of OTUs obtained from the iTag analysis. Each scatter point in the plot represents the structure of a microbial community in a particular treatment at a particular time point. The spatial separation between points approximates the similarity between their communities in terms of OTU composition. Stress: 0.0599.

Figure 3

Relative abundance (RA) of individual OTUs in replicate grab samples of starting material used for the experiment (t₀). Heat map is normalized within each OTU to the maximum RA (dark red) for that OTU at t₀. Analysis was limited to OTUs with cumulative t₀-RA of >3%. OTU taxonomic identifications are assigned with a minimum similarity of 0.9 at the designated taxonomic level; g_genus; f_family; o_order; and c_class. OTUs highlighted in gold represent known Fe/S-oxidizing organisms as defined in the text. The treatment designations represent grab samples of TO, tailings only (2 replicates), and TC, tailings plus 15% compost mixture used for tailings plus compost and quailbush treatments (3 replicates). Insufficient sequence reads were obtained from TO-t₀-1 for analysis.

Figure 4

Cladogram indicating the phylogenetic distribution of microbial lineages associated with the three mesocosm treatments generated using the Linear Discriminant Analysis (LDA) Effect Size (LEfSe) method. The 5 most abundant phyla were analyzed to the family level of taxonomic resolution; lineages with LDA values of 2 or higher as determined by LEfSe are displayed. Differences are represented by treatment color (red indicating QB, green TC, blue TO, and yellow non-significant). Each circle’s diameter is proportional to the taxon’s abundance. The strategy of multiclass analysis is non-strict (at least one class differential). Circles represent taxonomic ranks from domain to family inside to out. Labels are shown of the class, order and family levels. Fig. B shows scores for all the taxa with a LDA score ≥ 2.