Soil Microbial Community Composition and Tolerance to Contaminants in an Urban Brownfield Site

Abstract

Brownfields are unused sites that contain hazardous substances due to previous commercial or industrial use. The sites are inhospitable for many organisms, but some fungi and microbes can tolerate and thrive in the nutrient-depleted and contaminated soils. However, few studies have characterized the impacts of long-term contamination on soil microbiome composition and diversity at brownfields. This study focuses on an urban brownfield-a former rail yard in Los Angeles that is contaminated with heavy metals, volatile organic compounds, and petroleum-derived pollutants. We anticipate that heavy metals and organic pollutants will shape soil microbiome diversity and that several candidate fungi and bacteria will be tolerant to the contaminants. We sequence three gene markers (16S ribosomal RNA, 18S ribosomal RNA, and the fungal internal transcribed spacer (FITS)) in 55 soil samples collected at five depths to (1) profile the composition of the soil microbiome across depths; (2) determine the extent to which hazardous chemicals predict microbiome variation; and (3) identify microbial taxonomic groups that may metabolize these contaminants. Detected contaminants in the samples included heavy metals, petroleum hydrocarbons, polycyclic aromatic hydrocarbons, and volatile organic compounds. Bacterial, eukaryotic, and fungal communities all varied with depth and with concentrations of arsenic, chromium, cobalt, and lead. 18S rRNA microbiome richness and fungal richness were positively correlated with lead and cobalt levels, respectively. Furthermore, bacterial Paenibacillus and Iamia, eukaryotic Actinochloris, and fungal Alternaria were enriched in contaminated soils compared to uncontaminated soils and represent taxa of interest for future bioremediation research. Based on our results, we recommend incorporating DNA-based multi-marker microbial community profiling at multiple sites and depths in brownfield site assessment standard methods and restoration.

Keywords: 16S rRNA; 18S rRNA; Brownfields; Contamination; Heavy metals; ITS; Soil microbiome; eDNA.

Fig. 1

Study site and sampling design. Aerial view of the sampling location, known as the “Bowtie” property within Taylor Yard—a former rail yard in Los Angeles, CA, that is owned and managed by California State Parks. Soil samples were taken from below ground (at 0.1524 m, 1.524 m, 3.048 m, 4.572 m, and 6.096 m) at 12 ordinary sites, labeled B-01 to B-12. Samples from the surface (0.15 m) were collected by hand, and all other samples were collected using a direct-push drill rig in an acetate sleeve. Sampling locations with a red dot signify contaminants present at the surface as contamination at excess levels was only found at the surface

Fig. 2

Overview of the laboratory and computational workflow of this study. Briefly, soil samples were collected from five depths at the “Bowtie” site within Taylor Yard, a former rail yard in Los Angeles, CA, that is now owned and managed by California State Parks. After DNA extraction, two rounds of PCR amplified three targeted metabarcodes for microbes (16S rRNA marker, 18S rRNA marker, and fungal ITS marker). Sequences were processed in Anacapa and DADA2, generating a table of amplicon sequence variants counts for each sample. This ASV table along with sample metadata was uploaded into the R statistical program for statistical analyses and visualizations

Fig. 3

Community composition of the soil microbiome in a contaminated brownfield site. Stacked bar plots showing the relative frequency of sequences assigned to each microbial order across samples for A 16S rRNA profiles, B 18S rRNA profiles, and C fungal ITS profiles. Samples are grouped by depth (m), and each color represents a microbial order. Microbial abundance data was rarefied for these plots

Fig. 4

Soil microbiome structure varies with concentrations of heavy metals and hydrocarbons in a brownfield site. Canonical correspondence analysis (CCA) correlated soil microbiome structure at the sampled depths with concentrations of arsenic, cobalt, chromium, lead, and benzo(a)pyrene. Only samples from depths with known concentrations of contaminants (0.15 m, 1.52 m, and 6.10 m) are included. The two primary CCA axes are shown and samples are color-coded by depth for each marker. Arrows indicate the direction and magnitude of statistically significant relationships (at α = 0.05). See Table S6 for the detailed statistical output

Fig. 5

Microbial groups associated with contaminated surface soils represent potential sources for bioremediation efforts. Taxa enriched in contaminated vs. uncontaminated surface samples as determined by LEfSe for A 16S rRNA, B 18S rRNA, and C fungal ITS community profiles. Each dot represents a unique group and is color-coded by soil contamination (yes vs. no). Statistically significant genera (LDA > 3) are displayed and their most specific taxonomic classifications are on the x-axis. See methods for more details regarding the analysis